Evidence review for vitamin B12 replacement

Evidence review for vitamin B12 replacement

Vitamin B12 deficiency in over 16s: diagnosis and management

Evidence review E

NICE Guideline, No. 239

London: National Institute for Health and Care Excellence (NICE); 2024 Mar.

ISBN-13: 978-1-4731-5733-0

1.1. Review question

What is the clinical and cost effectiveness of vitamin B12 replacement for vitamin B12 deficiency, including the dose, frequency and route of administration?

1.1.1. Introduction

Traditionally the management of vitamin B12 deficiency with vitamin B12 replacement therapy has been carried out to treat and reverse haematological manifestations and treat and prevent progression of neurological problems. The optimal route, dose and frequency of administration of vitamin B12 replacement therapy in different situations has never been clearly established and has often been guided by custom and historical practice.

This review seeks to determine the best way of treating vitamin B12 deficiency. The evidence will be stratified by the different causes of the deficiency because the required intervention is expected to differ depending on the cause.

1.1.2. Summary of the protocol

For full details see the review protocol in Appendix A.

Table 1

PICO characteristics of review question.

1.1.3. Methods and process

This evidence review was developed using the methods and process described in Developing NICE guidelines: the manual. Methods specific to this review question are described in the review protocol in appendix A and the methods document.

Declarations of interest were recorded according to NICE’s conflicts of interest policy.

1.1.4. Effectiveness evidence

1.1.4.1. Included studies

Seven randomised controlled trials¹^, ³^–⁵^, ⁸^, ¹⁰^, ¹⁶ and two observational studies²^, ¹⁵ were included in the review. These are summarised in Table 2 and Table 3 below. Evidence from these studies is summarised in the clinical evidence summary tables below (Table 4, Table 5, Table 6, Table 7, Table 8, Table 9, Table 10, Table 11, Table 12, Table 13, Table 14, Table 15, and Table 16).

The population characteristics in the included studies were varied, with ages ranging from 18 to >75 years. The definition of B12 deficiency also varied significantly between studies, with B12 concentrations ≤350pg/mL being applied as the highest cut-off for inclusion. Stratification based on vitamin B12 deficiency cause was intended, however in most studies there was not enough information to determine the cause, or the study contained a mixed population that meant this analysis was not possible.

No relevant clinical studies were identified for subcutaneous hydroxocobalamin / cyanocobalamin. Additionally, no relevant clinical studies were identified that used dietary advice, no treatment or changing drug treatment as a comparator.

A range of oral vitamin B12 doses were used in the identified studies, ranging from 100 – 1000mcg. For intramuscular administration the only dose identified was 1000mcg, although there was variation in the frequency of administration, ranging from daily to monthly injections, typically with a declining frequency as studies progressed.

See also the study selection flow chart in Appendix C, study evidence tables in Appendix D, forest plots in Appendix E and GRADE tables in Appendix F.

1.1.4.2. Excluded studies

Three Cochrane systematic reviews were excluded from this review.⁹^, ¹⁷^, ¹⁹ These three papers were assessed as full texts but were excluded due to not containing a population relevant to the present review protocol. The populations in these reviews were required to have haematological values that defined vitamin B12 deficiency, whereas the present review accepted any author defined B12 deficiency. Additionally, these reviews contained interventions that were not specified in the present review protocol, including unlicensed forms of vitamin B12 such as liquid solutions. References to the included studies in these Cochrane reviews were screened for eligibility against the review protocol for this review. One study met the inclusion criteria and was included.

See the excluded studies list in Appendix I.

1.1.5. Summary of studies included in the effectiveness evidence

Table 2

Summary of randomised controlled trials included in the evidence review.

Table 3

Summary of observational studies included in the evidence review.

See Appendix D for full evidence tables.

1.1.6. Summary of the effectiveness evidence

Table 4

Clinical evidence summary: oral cyanocobalamin vs intramuscular cyanocobalamin.

Table 5

Clinical evidence summary: oral cyanocobalamin vs placebo.

Table 6

Clinical evidence summary: oral cyanocobalamin vs intramuscular hydroxocobalamin.

Table 7

Clinical evidence summary: oral cyanocobalamin 100mcg vs 250mcg.

Table 8

Clinical evidence summary: oral cyanocobalamin 100mcg vs 500mcg.

Table 9

Clinical evidence summary: oral cyanocobalamin 100mcg vs 1000mcg.

Table 10

Clinical evidence summary: oral cyanocobalamin 250mcg vs 500mcg.

Table 11

Clinical evidence summary: oral cyanocobalamin 250mcg vs 1000mcg.

Table 12

Clinical evidence summary: oral cyanocobalamin 500mcg vs 1000mcg.

Table 13

Clinical evidence summary: oral cyanocobalamin vs intramuscular cyanocobalamin (observational studies).

Table 14

Clinical evidence summary: intramuscular hydroxocobalamin (loading dose) vs intramuscular hydroxocobalamin (no loading dose) (observational studies).

Table 15

Clinical evidence summary: intramuscular hydroxocobalamin (loading dose) vs no treatment (observational studies).

Table 16

Clinical evidence summary: intramuscular hydroxocobalamin (no loading dose) vs no treatment (observational studies).

See Appendix F for full GRADE tables

1.1.7. Economic evidence

1.1.7.1. Included studies

Three health economic studies with the relevant comparison were included in this review⁶^, ¹¹^, ¹⁸. These are summarised in the health economic evidence profile below (Table 17) and the health economic evidence tables in Appendix H.

1.1.7.2. Excluded studies

One economic study relating to this review question was identified but was excluded due to methodological limitations (Masucci, 2013). This is listed in Appendix I, with reasons for exclusion given.

See also the health economic study selection flow chart in Appendix G.

1.1.8. Summary of included economic evidence

Table 17

Health economic evidence profile: oral B12 vs intramuscular B12.

1.1.9. Unit costs

Relevant unit costs are provided below to aid consideration of cost effectiveness.

Table 18

Unit costs of B12 treatment.

The BNF states “cyanocobalamin injection is less suitable for prescribing”. Also, the BNF states that Cytamen (cyanocobalamin) injection cannot be prescribed in NHS primary care. As the unit cost and frequency are higher compared to hydroxocobalamin injection, in addition to the BNF guidance, only hydroxocobalamin injections are considered for the cost analysis below.

For oral vitamin B12, only costs of Orobalin are included as this is lower cost as well as having 20 times the strength of the generic 50mcg tablet. Orobalin is the only licensed 1mg oral tablet currently and hence this is why this medicine is selected for analysis as other versions of the 1mg tablet would be considered unlicensed.

Table 19 below shows the costs of B12 treatment for people who do not have pernicious anaemia. These treatment doses for parenteral treatment include a loading dose. According to the committee there are usually 6mg (6 injections) given within the first month, but this can vary depending on a person’s symptoms. After the loading dose 1mg (one injection) is given every two or three months. Both parenteral treatment schedules are captured in Table 19. For parenteral treatment the administration costs are included which comprise of consumable costs and a 10-minute appointment (see Table 18). The dose for Orobalin is assumed to be 1mg daily.

Table 19

Comparison of treatment costs Orobalin vs Parenteral treatment.

Table 20 below shows the costs of B12 treatment for people who have pernicious anaemia. The treatment schedule for parenteral treatment is the same for pernicious anaemia and other causes of B12 deficiency. According to the committee there are usually 6mg (6 injections) given within the first month, but this can vary depending on a person’s symptoms. After the loading dose 1mg (one injection) is given every two or three months. Both parenteral treatment schedules are captured in Table 19. For parenteral treatment the administration costs are included which comprise consumable costs and a 10-minute appointment (see Table 18). Therefore, it is assumed that the treatment cost for parenteral treatment is the same in pernicious anaemia and other causes of B12 deficiency.

For oral treatment the dose is 4mg a day until remission of symptoms, with the length of time until remission varying depending on the person. It was proposed by the committee that the initial high dose frequency may be needed for a month. After this initial dose then the dose is reduced to 1mg daily.

Table 20

Comparison of treatment costs Orobalin vs Parenteral treatment (dose for pernicious anaemia).

A sensitivity analysis was conducted where 10% of patients (same proportion as the included study by Houle 2014⁶) were administered their intramuscular treatment at home (assumed 50 minutes of nurse time). In this analysis, parenteral treatment was cost saving by 12 months for those with vitamin B12 deficiency and by 6 months for those with pernicious anaemia.

1.1.10. Evidence statements

One cost comparison analysis found that parenteral B12 treatment was less costly than oral B12 treatment for treating B12 deficiency at 2 years. However, from year 2, the annual cost of oral was lower compared to parenteral treatment due to the initial switching costs. This analysis was assessed as directly applicable with potentially serious limitations.
One cost comparison analysis found that oral B12 treatment was less costly than intramuscular B12 treatment for treating B12 deficiency.
One cost–utility analysis found that oral B12 treatment was less costly and equally effective (therefore dominant) compared to parenteral B12 treatment for treating B12 deficiency. This analysis was assessed as partially applicable with potentially serious limitations.

All three studies would have found parenteral B12 therapy to be cost saving had they used the current NHS prices of licensed treatments.

1.2. Review question

What is the clinical and cost effectiveness of self-administration compared with healthcare professional administration of parenteral vitamin B12 replacement for vitamin B12 deficiency?

1.2.1. Introduction

Most people receiving parenteral intramuscular injections of vitamin B12 replacement for deficiency are given their treatment in a primary care setting by a healthcare professional. This requires periodic attendance at a primary care location.

Self-administration of vitamin B12 by intramuscular injection is possible and currently undertaken by a minority of patients. However, the relative safety, efficacy, and cost-effectiveness of self- versus healthcare-administration has not been established. It is unclear if self-administration is suitable for all people with vitamin B12 deficiency.

This review seeks to determine whether self-administration of parenteral intramuscular vitamin B12 replacement is a clinically and cost-effective way of treating vitamin B12 deficiency.

1.2.2. Summary of the protocol

For full details see the review protocol in Appendix A.

Table 21

PICO characteristics of review question.

1.2.3. Methods and process

Declarations of interest were recorded according to NICE’s conflicts of interest policy.

1.2.4. Effectiveness evidence

No relevant clinical studies comparing self-administration with health professional administration of parenteral vitamin B12 replacement were identified.

See also the study selection flow chart in Appendix C.

1.2.5. Summary of studies included in the effectiveness evidence

No studies were included.

1.2.6. Summary of the effectiveness evidence

No evidence was identified.

1.2.7. Economic evidence

1.2.7.1. Included studies

No health economic studies were included.

1.2.7.2. Excluded studies

No relevant health economic studies were excluded due to assessment of limited applicability or methodological limitations.

See also the health economic study selection flow chart in Appendix G.

1.2.8. Unit costs

Relevant unit costs are provided below to aid consideration of cost effectiveness.

Table 22

Unit costs for self-administration.

Table 23 below shows the treatment cost differences for people who may be receiving prophylactic treatment. There would be no loading dose required. This would be relevant for consideration if switching people from HCP administered treatment to self-administered treatment.

Table 23

Comparison of treatment/prophylaxis costs self-administration vs HCP administration treatment (no loading doses).

Table 24 below shows the treatment cost differences for people who may be receiving prophylactic treatment. There would be a loading dose required and this would be relevant for anyone that needs a treatment dose whether the cause is pernicious anaemia or another cause of B12 deficiency.

Table 24

Comparison of treatment costs self-administration vs HCP administration treatment (loading doses for pernicious anaemia).

1.2.9. Evidence statements

No relevant economic evaluations were identified that compared self-administration with HCP administration for people requiring parenteral B12 treatment.

1.3. The committee’s discussion and interpretation of the evidence

The committee discussion of the review of self-administration of parenteral vitamin B12 replacement is included in the discussion of vitamin B12 replacement.

1.3.1. The outcomes that matter most

The committee considered quality of life, patient reported outcomes including symptom scores, haematological values, complications and adverse events, adherence to treatment and education/work absence to be the most important outcomes of vitamin B12 replacement. The possible complications and adverse events listed by the committee were mainly those related to the possible complications and adverse events if a deficiency is insufficiently treated, as there are few complications and adverse events associated with vitamin B12 treatment itself. All outcomes were considered equally important for decision making and were rated as critical.

The majority of the evidence for vitamin B12 replacement identified from randomised controlled trials was for haematological values, with very little evidence identified for patient reported outcomes. No evidence was identified for quality of life, adherence to treatment, or education/work absence. The committee considered that although haematological values provide useful objective biomarkers, the end purpose of B12 replacement is to improve the patients’ quality of life, so the RCT evidence was not considered sufficient upon which to base recommendations. Therefore, a search for observational evidence was carried out. However, the evidence identified from observational studies was for haematological values only.

No evidence was identified for self-administration of parenteral vitamin B12 compared with healthcare professional administration.

1.3.2. The quality of the evidence

Evidence ranged from moderate to very low-quality, with the majority of evidence being assessed as very low-quality. Indirectness was the main reason for downgrading, with almost all evidence being downgraded by one increment due to the inclusion of populations with mixed causes of vitamin B12 deficiency. The most common concern for risk of bias was due to the randomisation process used and deviations from the intended interventions, although concerns arose for multiple domains in most cases. Partly due to the small sample sizes recruited in the identified evidence, a large proportion of the evidence was also downgraded for imprecision.

The committee considered the follow up periods reported in most of the studies to be sufficient. However, they noted that improvements in outcomes may not be clinically important until a person has received three months of replacement, particularly for oral replacement. Therefore, confidence in evidence for the outcomes reported before three months was reduced, especially for the comparison of oral cyanocobalamin versus hydroxocobalamin, where outcomes were reported at one month.

No evidence was identified comparing subcutaneous administration of cyanocobalamin or hydroxocobalamin, or a combination of treatment modalities with any protocol comparators. Additionally, no evidence was identified for any comparison at any time point for quality of life, most of the patient reported outcome measures (fatigue, sleep, peripheral neuropathy, psychiatric symptoms, pain), specific complications and adverse events (mortality, bleeds, self-harm, nerve damage, frailty/falls, severe cognitive effects, postural hypotension), adherence to treatment or education/work absence.

Overall, the committee felt that the evidence identified was of limited use in the decision-making process. The small number of studies identified for each comparator was a particularly limiting factor, with no meta-analyses possible across the review. Compounding the difficulty in interpreting the results was the significant differences in populations and interventions used in the identified studies. Only one study included participants with a single cause of vitamin B12 deficiency, with all others including people with a number of different causes. Furthermore, studies utilised a number of different definitions of deficiency and doses, frequencies and durations of treatment. These factors made it difficult to compare across studies and to draw conclusions from the evidence as a whole.

1.3.3. Benefits and harms

Oral cyanocobalamin vs intramuscular cyanocobalamin

For the comparison between oral and intramuscular cyanocobalamin, low to very low quality RCT evidence showed a benefit of intramuscular cyanocobalamin for total B12 at both ≤3 months and >3 months, and holotranscobalamin at ≤3 months. This finding was supported by the observational evidence, where very low-quality evidence showed a benefit of intramuscular cyanocobalamin for total B12 at >3 months. Moderate to very low-quality evidence showed no clinically important difference for methylmalonic acid and homocysteine at both ≤3 months and >3 months, patient reported outcome measures (general health) and mean corpuscular volume at >3 months, and adverse events at ≤3 months.

Oral cyanocobalamin vs placebo

For the comparison between oral cyanocobalamin and placebo, very low-quality evidence showed a benefit of oral cyanocobalamin for total B12 at ≤3 months, with both 0.4 and 1 mg doses, total B12, holotranscobalamin and homocysteine at >3 months, methylmalonic acid at ≤3 months and homocysteine at ≤3 months with a 0.4 mg dose only. In contrast, very low quality showed a harm of oral cyanocobalamin for folate at >3 months. Moderate to very low-quality evidence showed no clinically important difference for patient reported outcomes (general health and cognition) at >3 months, homocysteine at ≤3 months with a 0.4 mg dose, haemoglobin and folate at >3 months.

Oral cyanocobalamin vs intramuscular hydroxocobalamin

For the comparison between oral cyanocobalamin and intramuscular hydroxocobalamin, moderate to low-quality evidence showed a benefit of intramuscular hydroxocobalamin for total B12, holotranscobalamin and homocysteine at ≤3 months. Moderate to very low-quality evidence showed no clinically important difference for methylmalonic acid and adverse events at one month. A large effect in favour of intramuscular hydroxocobalamin was seen for this comparison. This was noted by the committee and the results were interpreted with caution given the limitations of the evidence, including the relatively short follow up period.

Oral cyanocobalamin dose comparison

For the comparison between treatment with 100 mcg and 250 mcg of oral cyanocobalamin, very low-quality evidence showed a benefit of 250 mcg for holotranscobalamin at both ≤3 months and >3 months. Very low-quality evidence showed no clinically important difference between doses for total B12, homocysteine or methylmalonic acid at either ≤3 months or >3 months.

For the comparison between treatment with 100 mcg and 500 mcg of oral cyanocobalamin, low to very low-quality evidence showed a benefit of 500 mcg for total B12, holotranscobalamin, homocysteine and methylmalonic at both ≤3 months and >3 months.

For the comparison between treatment with 100 mcg and 1000 mcg of oral cyanocobalamin, low-quality evidence showed a benefit of 1000 mcg for total B12, holotranscobalamin and homocysteine at ≤3 months, and holotranscobalamin at >3 months. Very low-quality evidence showed no clinically important difference between interventions for homocysteine or methylmalonic acid at >3 months, and for methylmalonic acid at ≤3 months.

For the comparison between treatment with 250 mcg and 500 mcg of oral cyanocobalamin, very low-quality evidence showed a benefit of 500 mcg for total B12 at ≤3 months. Very low-quality evidence showed no clinically important difference between treatments for total B12 at ≤3 months, or for holotranscobalamin, homocysteine or methylmalonic acid at both ≤3 months and >3 months.

For the comparison between treatment with 250 mcg and 1000 mcg of oral cyanocobalamin, very low-quality evidence showed a benefit of 1000 mcg for total B12 at ≤3 months, and for holotranscobalamin at both ≤3 months and >3 months. Very low-quality evidence showed no clinically important difference between treatments for homocysteine or methylmalonic acid at both ≤3 months and >3 months.

For the comparison between treatment with 500 mcg and 1000 mcg of oral cyanocobalamin, very low-quality evidence showed a benefit of 1000 mcg for total B12 and holotranscobalamin at ≤3 months. Very low-quality evidence showed no clinically important difference between treatments for holotranscobalamin at >3 months, and homocysteine and methylmalonic acid at both ≤3 months and >3 months.

Intramuscular hydroxocobalamin with loading dose vs intramuscular hydroxocobalamin with no loading dose

For the comparison between IM hydroxocobalamin treatment regimes, very low-quality evidence showed a benefit of IM treatment with a loading dose for total B12 at ≤3 months. Very low-quality evidence showed no clinically important difference between treatments for methylmalonic acid at ≤3 months.

Intramuscular hydroxocobalamin vs no treatment

For the comparisons between IM hydroxocobalamin regimens and no treatment, low-quality evidence showed a benefit of IM hydroxocobalamin with a loading dose for total B12 and methylmalonic acid at ≤3 months. Low to very low-quality evidence showed a benefit of IM hydroxocobalamin with no loading dose compared to no treatment for total B12 and methylmalonic acid at ≤3 months.

Conclusions and committee experiences

It was the experience of the committee that improvements in haematological values do not necessarily translate to improved quality of life for patients. Given the limited evidence identified for outcomes other than haematological values, the committee felt it was difficult to draw conclusions on the preferable treatment option using the clinical evidence alone.

The main comparison of interest for the committee was that made between oral and intramuscular treatments. There was a preference towards intramuscular treatment from the lay members of the committee, due to their own experiences of intramuscular treatment being more effective in managing their conditions. Clinicians on the committee had no strong preference based on their experience and expertise but considered that people may prefer to receive oral treatment where there is a choice due to the discomfort and inconvenience of receiving intramuscular injections.

However, the committee considered circumstances where intramuscular vitamin B12 replacement may be preferable to oral replacement to ensure that the treatment works quickly and prevent the long-term symptoms and complications. Examples of these circumstances include: if there are concerns about adherence to oral treatment, such as those with delirium or a health inequality that affects access to care such as homelessness. Homeless people may not be able to store medicines safely therefore there may be concern that people may not be adherent to treatment. It would be more practical for people to attend one appointment to receive parenteral treatment every two to three months than to be responsible for ordering medicine on a regular basis and keeping the medicine safe and taking the medicine daily. To reduce the risk of diversion parenteral treatment may be the preferred route of treatment. Intramuscular injections may also be more appropriate in hospitalised older people with complex co-morbidity requiring multidisciplinary management including interacting polypharmacy. For example, undernutrition, disease associated anorexia, dementia, decompensated frailty, where long-term adherence with oral replacement is deemed to be compromised, and those at risk of rapid deterioration which could have a major effect on quality of life, such as those with neurological or haematological conditions. The committee agreed that in these circumstances, intramuscular replacement should be considered over oral replacement, regardless of the cause of vitamin B12 deficiency.

The committee noted that giving people an estimated timeframe for when they can expect to see an improvement in their symptoms will empower them to return to their healthcare professional sooner if symptoms are not improving as expected. Therefore, the committee made a recommendation to give people information about how long it takes for treatment to take effect and when they are likely to see an improvement in their symptoms.

The committee agreed there was no reason why treatment should be stopped during pregnancy or breastfeeding if the person is already receiving vitamin B12 replacement. To stop treatment may lead to a return or increase in symptoms and put the person and their baby at risk. Therefore, the committee made a recommendation to continue treatment in this group.

Deficiency caused by malabsorption

In people with vitamin B12 deficiency due to autoimmune gastritis, the committee agreed that intramuscular vitamin B12 replacement is the appropriate treatment. This is because this group are unable to absorb adequate vitamin B12 from oral replacement or their diet, meaning intramuscular treatment is the only option and should be given as a life-long treatment. In most cases, replacement should be given with an initial loading dose to rapidly increase vitamin B12 levels and stop the progression of neurological damage. This is current practice and should remain this way. The committee noted the difficulties in diagnosing this condition and agreed that those who are strongly suspected of having autoimmune gastritis should be offered the same treatment as if they had a confirmed diagnosis. The committee agreed that this treatment should not be discontinued unless another reversible cause of vitamin B12 deficiency is identified, at which point treatment should be reviewed to see if intramuscular treatment is still the best option or whether they will need long-term treatment.

In people with vitamin B12 deficiency due to surgery such as major gastric resection, bypass (including bariatric procedures) or terminal ileal resection, the committee agreed that lifelong vitamin B12 replacement is usually required to prevent deficiency. This is because in all cases, ability to absorb enough vitamin B12 from the diet is irreversibly impaired. In the absence of any evidence that either treatment route is more effective than another in this population, the committee debated whether oral or intramuscular vitamin B12 replacement would be preferable. The committee considered that current practice in the UK is most commonly to offer intramuscular replacement, although experience in other countries suggests that oral 1000 mcg cyanocobalamin is at least as effective for some forms of surgery. Therefore, the committee decided to split recommendations in two. People who have had a total gastrectomy, or a complete terminal ileal resection need lifelong intramuscular injections because these types of surgery cause permanent malabsorption. People who have had a partial gastrectomy, partial terminal ileal resection or some forms of bariatric surgery may be able to absorb vitamin B12. In these cases the committee agreed that the preference would be for intramuscular replacement, as it is often difficult to know how much of an oral dose will be absorbed and by giving it intramuscularly, there is more certainty that the person has received enough of the vitamin. Therefore, they recommended intramuscular injections over oral vitamin B12 replacement.

In people with vitamin B12 deficiency due to malabsorption that is not caused by autoimmune gastritis, gastric or bariatric surgery, such as those with coeliac disease, no evidence was identified on the most effective route of administration. Intramuscular treatment was found to be more cost effective than oral replacement when used for 6 months or more. In the committee’s experience, injections could be the better option for these groups of people because it can be difficult to judge how much of an oral dose will be absorbed by the body, so injections will help ensure they are getting enough of the vitamin. However, the committee did not rule out the use of oral vitamin B12 replacement, because there was no evidence on its use in these groups of people and therefore there was nothing to suggest it would be ineffective. Therefore, the committee recommended that intramuscular is considered instead of oral vitamin B12 replacement. The treatment may be lifelong, or for as long as is necessary, depending on the cause. For example, people with coeliac disease may be able to manage their deficiency effectively through a gluten free diet. In these people, symptoms and medicines should be regularly reviewed and discontinuation of replacement considered if the cause is successfully reversed.

In all cases the committee noted that intramuscular injections are also cheaper when vitamin B12 replacement is needed for more than six months. The committee agreed that for all malabsorption conditions where oral replacement has been recommended, a dose of 1 mg per day should be prescribed as a minimum.

Medicine induced deficiency

The committee agreed that in people who have vitamin B12 deficiency caused by a medicine such as metformin, a H2 receptor antagonist, a proton pump inhibitor, or an antiseizure medicine, they should remain on vitamin B12 replacement as long as they are receiving the causative medicine. No evidence was identified on the most effective route of administration in people with medicine induced deficiency, therefore the committee recommended that either intramuscular or oral replacement should be offered, based on clinical judgement and patient preference. If possible, medicine that can cause vitamin B12 deficiency should be stopped or changed to an alternative medicine, although the committee noted that there are many cases where this would not be possible or preferable. The committee agreed that if the medicine is stopped and the person no longer has symptoms of vitamin B12 deficiency, the need for vitamin B12 replacement should be reviewed.

Recreational nitrous oxide induced deficiency

No evidence was identified on the most effective route of administration in people with nitrous oxide induced deficiency, therefore the committee recommended that either intramuscular or oral replacement should be offered, and this should depend on patient preference. Nitrous oxide affects B12 concentrations by inactivating either the vitamin B12 molecule or the enzyme methionine synthase and therefore can cause vitamin B12 deficiency. Its longer-term effects on the body are unknown. Where nitrous oxide is being used recreationally, the committee agreed that there is probably a lack of awareness of the possible adverse effects by people using it. Therefore, the committee agreed that advising the person to stop using the drug, and the reasons why, is important and made a recommendation to reflect this. The committee agreed that if the person stops using nitrous oxide recreationally and they no longer have symptoms of vitamin B12 deficiency, the need for vitamin B12 replacement should be reviewed.

Dietary vitamin B12 deficiency

The committee agreed that anybody with a suspected or confirmed vitamin B12 deficiency due to a lack of the vitamin in their diet should be given information on how to improve their intake of vitamin B12, regardless of any other replacement that may be offered in addition. As low dietary intake is a potentially reversible cause of deficiency, the committee considered that people should be given the opportunity to manage their deficiency though changes to their diet.

However, the healthcare professional may recommend that the person receives B12 replacement, depending on the clinical presentation and individual circumstances. The committee recommended that for those with confirmed deficiency, oral replacement should be considered. This is the licensed treatment for vitamin B12 deficiency.

The committee considered making recommendations regarding dose for oral vitamin B12 replacement. Licensed doses of cyanocobalamin vary from 50 mcg to 1000 mcg. The committee noted that most of the studies included in the evidence review used 1000 mcg, however there was no strong evidence comparing different doses. The committee noted that dose may depend on the cause of the deficiency and the clinical presentation. For example, 1000 mcg may be considered a more appropriate dose for a person with malabsorption that is not caused by autoimmune gastritis or surgery, to ensure that enough of the vitamin is absorbed. For a person with a dietary deficiency, 50-150 mcg may be suitable. The committee considered that there would be the opportunity to increase the dose to at least 1000 mcg during follow up appointments if a lower dose did not adequately improve symptoms and vitamin B12 concentrations. However, in the absence of evidence, the committee agreed that dose should be a clinical decision.

The committee also agreed that during pregnancy or breastfeeding, oral replacement at a dose of at least 1000 mcg should be considered. This was because the demand for vitamin B12 increases during pregnancy, so setting a minimum dose of 1000 mcg per day would ensure that enough vitamin B12 is absorbed to meet this demand.

The committee agreed that people who follow restrictive diets should not be assumed to be deficient as a result of their diet. Due to the difficulty in identifying cause, other causes should be considered throughout the treatment pathway. The committee agreed that whilst restrictive diets can cause vitamin B12 deficiency, people who follow them may be aware of this and ensure they consume adequate vitamin B12. Assuming the diet is the cause of the deficiency can lead to under-investigation of other potential causes, for example autoimmune gastritis, which can have serious long-term implications if left undiagnosed. Therefore, the committee made recommendations to raise awareness that diet may not be the cause or the only cause of deficiency and to have a full discussion with the person suspected of having a dietary deficiency regarding sources of vitamin B12 in their diet, use of supplements, and signs, symptoms or risk factors that could suggest another reason for the deficiency. If the discussion suggests that diet may not be linked to the deficiency, the committee agreed that this would raise suspicion of other possible causes such as autoimmune gastritis and further investigations should be considered. See the recommendations on identifying the cause of vitamin B12 deficiency.

The committee noted that some people purchase over-the-counter supplements online, from pharmacies or health food shops. Vitamin B12 injections are also available from some beauty clinics and over the counter in other countries and some people import these products for personal use. The committee considered that there is wide variation in the forms and doses of vitamin B12 used in purchasable oral vitamin B12 containing products and while some can effectively treat deficiency, others do not contain enough of the vitamin. Some products contain B12 analogues, which are not used by the body in the same way as the cyanocobalamin, methylcobalamin and adenosylcobalamin forms of vitamin B12. The committee agreed that this should be explained to the person to give them the best chance of selecting an effective supplement if that is what they choose to do.

The committee also agreed that intramuscular vitamin B12 replacement could be the best option for some people with confirmed deficiency if there are concerns about adherence to oral treatment. This would include hospitalised older adults with multimorbidity; those with delirium or cognitive impairment; or those who have difficulties accessing care due to health inequalities, such as homelessness. This would ensure they get the treatment they need, prevent the long-term symptoms and complications caused by deficiency, resulting in fewer hospitalisations in the future. Homeless people may not be able to store medicines safely therefore there may be concern that people may not be adherent to treatment. It would be more practical for people to attend one appointment to receive intramuscular injections every two to three months than to be responsible for ordering medicine on a regular basis and keeping the medicine safe and taking the medicine daily. Intramuscular replacement could also be considered for older people in hospital with complex comorbidities, who are often taking multiple medicines and whose care is usually overseen by a multidisciplinary team. This is because there may be issues with long-term adherence to oral replacement in this group of people, which can include those with undernutrition, dementia or decompensation linked to frailty. The committee also agreed intramuscular replacement may be preferable when a person is at risk of rapid deterioration that could have a major effect on their quality of life, because the treatment works quickly. This includes people with neurological or haematological conditions.

Unknown causes of vitamin B12 deficiency

The committee agreed that people presenting with vitamin B12 deficiency where the cause is unknown should be treated with oral replacement initially. The committee made this recommendation on the basis that if malabsorption conditions are not suspected, then oral vitamin B12 replacement should be suitable for correcting the deficiency. If symptoms do not improve, or worsen, then the treatment can be changed to intramuscular (see ongoing care and follow up recommendations) and further investigations can be made into the presence of autoimmune gastritis or other malabsorption issues as the possible cause of vitamin B12 deficiency (see identifying the cause recommendations). The committee agreed that this was a suitable approach as the inefficacy of oral vitamin B12 replacement is indicative that an underlying malabsorption condition may be present and further investigations into the underlying cause are more difficult after intramuscular injections have been received.

Self-administration of parenteral vitamin B12 replacement

The committee considered that during the COVID-19 pandemic, many people self-administered their vitamin B12 replacement due to the inability to receive their injections from clinicians as usual. The committee were also aware of patient surveys indicating a preference for the option to self-administer vitamin B12 injections, which the lay members on the committee agreed with. The committee discussed the potential benefits of self-administration for people with vitamin B12 deficiency. These included being able to receive their injections at times convenient for them, in environments where they feel comfortable, not having to rely on the availability of GP appointments to receive their treatment when it is due and saving time and costs to attend GP appointments. The committee also noted other health conditions for which self-administration of injectable treatments were common practice, including rheumatoid arthritis, fertility issues, rhesus disease and anaphylaxis.

The committee considered the possible safety issues associated with injections, such as anaphylaxis and bleeds. The committee also considered groups of people in whom self-administration may not be suitable due to reduced physical or mental capacity. The committee concluded that without data on the effectiveness or safety of self-administration, a recommendation for further research was the most appropriate action.

1.3.4. Cost effectiveness and resource use

Oral cyanocobalamin

In the BNF there are predominantly two strengths of cyanocobalamin tablets. These are the 50mcg and 1000mcg. There is a Drug Tariff price for each strength but, the one for the 1000mcg tablet is lower than the one for the 50mcg tablet. Therefore, it was considered preferable to use the 1000mcg cyanocobalamin assuming that it would be at least as effective as the lower strength tablet as well as cyanocobalamin being generally well tolerated and unlikely to cause harm.

Although there are cheaper 1000mcg cyanocobalamin preparations in the BNF, only Orobalin 1000mcg is licensed. Where there is a licensed preparation, alternative preparations should not be prescribed. The reimbursement to pharmacy contractors who dispense any standard release preparation of cyanocobalamin 1000mcg will be the same, therefore the Drug Tariff price was used in the cost analysis. For the licensing of Orobalin, the initial dose is 4000mcg daily until remission. However, there is uncertainty about how long the time taken to remission is and how to assess remission which may be evaluated by further B12 tests or assessment of a person’s symptoms. The experience of the committee is that for newly diagnosed people with B12 deficiency, when cyanocobalamin 1000mcg tablets are prescribed, the starting dose is one tablet a day rather than four tablets a day. This was the assumption within the cost comparison for people who have B12 deficiency.

For people with suspected dietary deficiency whose test results indicate they are B12 deficient, it was recommended that dietary advice be provided. Evidence was not available to suggest everyone with a suspected or confirmed dietary vitamin B12 deficiency should be offered oral vitamin B12 replacement and the committee agreed for a lot of people dietary advice alone may be enough to correct the deficiency. However recognised that some people would need additional treatment and recommended that oral vitamin B12 replacement should be considered for people who have a suspected or confirmed dietary deficiency. The committee agreed that if treatment was offered to all people with a confirmed dietary deficiency of vitamin B12, this would have a significant resource impact because this does not reflect usual clinical practice.

For vitamin B12 deficiency (and not pernicious anaemia), the BNF recommended dose is 50mcg to 150mcg daily. By switching patients from 100mcg (costs £0.86 per daily dose) or 150mcg (costs £1.29 per daily dose) to one 1000mcg tablet (costs £0.33 per daily dose) this would be cost saving – this would save approximately £75 - £173 annually per patient. The committee members did not think that there would be any adverse effects from the increase in dose and the savings may be significant due to many people on these doses.

Oral cyanocobalamin vs intramuscular hydroxocobalamin

Published cost effectiveness evidence

Three economic evaluations were identified for this review comparing oral and parenteral vitamin B12 treatment. The three published economic evaluations had conflicting results. All three studies deemed both treatments as equally effective. In two papers the oral treatment was a lower cost strategy, however in the other economic evaluation parenteral treatment was the lower cost.

In the economic evaluation that deemed parenteral treatment as the lower cost strategy, the study found that from year two of treatment oral cost would always be lower than parenteral treatment cost. This was due to the high costs of switching people from parenteral treatment to oral treatment in year one, which would not be relevant if people were started on oral treatment rather than parenteral treatment. If the study time horizon had been longer than two years or if the additional monitoring costs were not required, then oral treatment would have been the lower cost strategy compared to parenteral treatment. However, imporantly, none of the studies used current NHS prices.

Consideration of cost effectiveness

A simple cost comparison was presented to aid the committee with the consideration of cost effectiveness.

For people who are newly diagnosed with B12 deficiency, oral treatment with Orobalin 1000mcg daily dose is lower cost than parenteral treatment over short time horizons due to the initial loading dosing with parenteral treatment. After the initial loading dose, if the parenteral maintenance dose is hydroxocobalamin 1000mcg every three months, then parenteral treatment will be the lower cost by six months. If the maintenance parenteral dose is hydroxocobalamin 1000mcg every two months, then parenteral treatment will be the lower cost strategy within eight months. For people who are newly diagnosed with autoimmune gastritis (also known as pernicious anaemia), or any other B12 deficiency that requires long-term treatment, parenteral treatment is the lower cost strategy. The cost of parenteral treatment is considerably more expensive for people who cannot travel to their general practice. When home visits for 10% of patients were accounted for, it took up to one year for parenteral treatment to be cost saving.

Despite the lack of evidence identified relating to quality of life, the committee expressed the view that quality of life will be improved more quickly with parenteral treatment for people with symptomatic B12 deficiency. Timely parenteral treatment can also reduce hospitalisations due to the complications of B12 deficiency, which is expected to improve patient outcomes and QALY gains, and this, in turn, is likely to improve the cost effectiveness of parenteral treatment compared to oral treatment. However, the committee noted that some people may prefer oral administration for ease of treatment rather than booking an appointment for parenteral administration. This may contribute to treatment adherence and improve patient outcomes for oral administration of B12. Furthermore, in some places, shortages of practice nursing staff might make the use of oral treatment more practical even if it is more costly.

The committee agreed that for people with malabsorption or autoimmune gastritis (also known as pernicious anaemia), the parenteral route is preferred over oral. The committee noted that some people may have been switched to oral administration of vitamin B12 during the Covid-19 pandemic and may have not resumed parenteral treatment. Parenteral treatment for autoimmune gastritis would be the preferred cost-effective option over oral treatment.

Self-administration of parenteral treatment vs healthcare professional (HCP) administration

Published cost effectiveness evidence

There was no economic evidence identified for this question.

Consideration of cost effectiveness

A simple cost analysis using unit costs was presented to aid the committee with the consideration of cost effectiveness.

From these calculations, if a person needed three or more injections, then treatment via self-administration would be cost saving. This is due to the initial one-off cost of the consumables that would be required to be prescribed to enable self-administration.

The most significant cost for HCP administration is the cost of the healthcare professional’s time; the cost of consumables and the cost of the medicine is low in comparison. For newly diagnosed autoimmune gastritis patients or people that required a loading dose, the total costs of the consumables required for self-administration would be offset within the first month, making self-administration the least costly option.

For people that are already receiving parenteral treatment who are switched to self-administration, it would take six months for self-administration to be cheaper than healthcare professional administration if dose frequency is two months, whilst it would take one year for self-administration to be cheaper if the dose frequency was every three months. This would be under the assumption that no loading dose would be required.

There is uncertainty whether self-administration will be as equally effective as healthcare professional administered treatment due to the possible risk of injuries with self-administration which would also incur treatment costs. There is also concern that people starting parenteral treatment may be at risk of anaphylaxis and, if not undertaken under healthcare professional supervision, this may delay management and affect outcomes.

Recommendation

Self-administration of parenteral B12 treatment was not recommended due to a lack of clinical evidence and concerns regarding the safety of recommending self-administration.

Hydroxocobalamin is not licensed for subcutaneous administration, but it is for intramuscular. For self-administration, the subcutaneous route is considered easier compared to intramuscular, however this would be considered off-label.

Despite self-administration being likely to be cost saving versus healthcare professional administration there was a concern regarding the risk of injuries with intramuscular administration as well as the risk of anaphylaxis. The committee discussed offsetting the risk of anaphylaxis by having the first injection with a healthcare professional and the following could be self-administered. However, there would still be risks of injury. With this uncertainty the committee thought it would be prudent to continue usual practice and to investigate self-administration as a research recommendation.

1.3.5. Recommendations supported by this evidence review

This evidence review supports recommendations 1.5.1 to 1.5.17 and the research recommendations on what is the clinical and cost effectiveness of vitamin B12 replacement for vitamin B12 deficiency, including the dose, frequency and route of administration; and what is the clinical and cost effectiveness of self-administration of parenteral vitamin B12 replacement for deficiency compared with administration by a healthcare professional.

1.4. References

1.: Castelli MC, Friedman K, Sherry J, Brazzillo K, Genoble L, Bhargava P et al. Comparing the efficacy and tolerability of a new daily oral vitamin B12 formulation and intermittent intramuscular vitamin B12 in normalizing low cobalamin levels: a randomized, open-label, parallel-group study. Clinical Therapeutics. 2011; 33(3):358–371.e352 [PubMed: 21600388]

2.: Couderc AL, Camalet J, Schneider S, Turpin JM, Bereder I, Boulahssass R et al. Cobalamin deficiency in the elderly: aetiology and management: a study of 125 patients in a geriatric hospital. Journal of nutrition, health & aging. 2015; 19(2):234–239 [PubMed: 25651452]

3.: Dangour AD, Allen E, Clarke R, Elbourne D, Fletcher AE, Letley L et al. Effects of vitamin B-12 supplementation on neurologic and cognitive function in older people: a randomized controlled trial. American Journal of Clinical Nutrition. 2015; 102(3):639–647 [PMC free article: PMC4548176] [PubMed: 26135351]

4.: Dhonukshe-Rutten RA, van Zutphen M, de Groot LC, Eussen SJ, Blom HJ, van Staveren WA. Effect of supplementation with cobalamin carried either by a milk product or a capsule in mildly cobalamin-deficient elderly Dutch persons. American Journal of Clinical Nutrition. 2005; 82(3):568–574 [PubMed: 16155269]

5.: Eussen SJ, de Groot LC, Clarke R, Schneede J, Ueland PM, Hoefnagels WH et al. Oral cyanocobalamin supplementation in older people with vitamin B12 deficiency: a dose-finding trial. Archives of Internal Medicine. 2005; 165(10):1167–1172 [PubMed: 15911731]

6.: Houle SK, Kolber MR, Chuck AW. Should vitamin B12 tablets be included in more Canadian drug formularies? An economic model of the cost-saving potential from increased utilisation of oral versus intramuscular vitamin B12 maintenance therapy for Alberta seniors. BMJ Open. 2014; 4(5):e004501 [PMC free article: PMC4025453] [PubMed: 24793247]

7.: Jones K, Burns A. Unit costs of health and social care 2021. Canterbury. Personal Social Services Research Unit University of Kent, 2021. Available from: https://www.pssru.ac.uk/project-pages/unit-costs/unit-costs-of-health-and-social-care-2021/

8.: Kuzminski AM, Del Giacco EJ, Allen RH, Stabler SP, Lindenbaum J. Effective treatment of cobalamin deficiency with oral cobalamin. Blood. 1998; 92(4):1191–1198 [PubMed: 9694707]

9.: Malouf R, Areosa SA. Vitamin B12 for cognition. Cochrane Database of Systematic Reviews 2003, Issue 3. Art. No.: CD004394. DOI: 10.1002/14651858.cd004394. [PubMed: 12918012] [CrossRef]

10.: Metaxas C, Mathis D, Jeger C, Hersberger KE, Arnet I, Walter P. Early biomarker response and patient preferences to oral and intramuscular vitamin B12 substitution in primary care: a randomised parallel-group trial. Swiss Medical Weekly. 2017; 147:w14421 [PubMed: 28421567]

11.: Mnatzaganian G, Karnon J, Moss JR, Elshaug AG, Metz M, Frank OR et al. Informing disinvestment with limited evidence: cobalamin deficiency in the fatigued. International Journal of Technology Assessment in Health Care. 2015; 31(3):188–196 [PubMed: 26179277]

12.: National Institute for Health and Care Excellence. Developing NICE guidelines: the manual [updated January 2022]. London. National Institute for Health and Care Excellence, 2014. Available from: http://www.nice.org.uk/article/PMG20/chapter/1%20Introduction%20and%20overview

13.: NHS Business Services Authority. NHS electronic drug tariff December 2022. 2022. Available from: https://www.nhsbsa.nhs.uk/pharmacies-gp-practices-and-appliance-contractors/drug-tariff Last accessed: 09/12/2022.

14.: Organisation for Economic Co-operation and Development (OECD). Purchasing power parities (PPP). 2021. Available from: https://data.oecd.org/conversion/purchasing-power-parities-ppp.htm Last accessed: 01/06/2022.

15.: Smelt HJ, Pouwels S, Said M, Berghuis KA, Boer AK, Smulders JF. Comparison between different intramuscular vitamin B12 supplementation regimes: a retrospective matched cohort study. Obesity Surgery. 2016; 26(12):2873–2879 [PubMed: 27146501]

16.: Ubbink JB, Vermaak WJ, van der Merwe A, Becker PJ, Delport R, Potgieter HC. Vitamin requirements for the treatment of hyperhomocysteinemia in humans. Journal of Nutrition. 1994; 124(10):1927–1933 [PubMed: 7931701]

17.: Vidal-Alaball J, Butler CC, Cannings-John R, Goringe A, Hood K, McCaddon A et al. Oral vitamin B12 versus intramuscular vitamin B12 for vitamin B12 deficiency. Cochrane Database of Systematic Reviews 2005, Issue 3. Art. No.: CD004655. DOI: doi: 10.1002/14651858.CD004655.pub2. [PMC free article: PMC5112015] [PubMed: 16034940] [CrossRef]

18.: Vidal-Alaball J, Butler CC, Potter CC. Comparing costs of intramuscular and oral vitamin B12 administration in primary care: a cost-minimization analysis. The European journal of general practice. 2006; 12(4):169–173 [PubMed: 17127603]

19.: Wang H, Li L, Qin L, Song Y, Vidal?Alaball J, Liu T. Oral vitamin B12 versus intramuscular vitamin B12 for vitamin B12 deficiency. Cochrane Database of Systematic Reviews 2018, Issue 3. Art. No.: CD004655. DOI: 10.1002/14651858.cd004655.pub3. [PMC free article: PMC6494183] [PubMed: 29543316] [CrossRef]

Appendices

Appendix A. Review protocols

A.1. Vitamin B12 replacement (PDF, 173K)

A.2. Self-administration (PDF, 219K)

Appendix B. Literature search strategies

These literature search strategies were used for the following reviews:

What is the clinical and cost effectiveness of vitamin B12 replacement for vitamin B12 deficiency, including the dose, frequency and route of administration?
What is the clinical and cost effectiveness of self-administration compared with healthcare professional administration of parenteral vitamin B12 replacement for vitamin B12 deficiency?

The literature searches for these reviews are detailed below and complied with the methodology outlined in Developing NICE guidelines: the manual.¹²

For more information, please see the Methodology review published as part of the accompanying documents for this guideline.

B.1. Clinical search literature search strategy (PDF, 153K)

B.2. Health Economics literature search strategy (PDF, 130K)

Appendix C. Effectiveness evidence study selection

C.1. Vitamin B12 replacement (PDF, 115K)

C.2. Self-administration (PDF, 111K)

Appendix D. Effectiveness evidence

D.1. Vitamin B12 replacement

Download PDF (450K)

D.2. Self-administration

No evidence identified.

Appendix E. Forest plots

E.1. Vitamin B12 replacement

Download PDF (467K)

E.2. Self-administration

No forest plots.

Appendix F. GRADE tables

F.1. Vitamin B12 replacement

Download PDF (414K)

F.2. Self-administration

No GRADE tables.

Appendix G. Economic evidence study selection

Download PDF (175K)

Appendix H. Economic evidence tables

H.1. Vitamin B12 replacement

Download PDF (201K)

H.2. Self-administration

No economic evidence tables.

Appendix I. Excluded studies

I.1. Vitamin B12 replacement

Clinical studies

Table 27

Studies excluded from the clinical review.

Health Economic studies

Published health economic studies that met the inclusion criteria (relevant population, comparators, economic study design, published 2006 or later and not from non-OECD country or USA) but that were excluded following appraisal of applicability and methodological quality are listed below. See the health economic protocol for more details.

Table 28

Studies excluded from the health economic review.

I.2. Self-administration

Clinical studies

No excluded studies.

Health Economic studies

None

Appendix J. Research recommendations – full details

J.1. Research recommendation (PDF, 158K)

J.2. Research recommendation (PDF, 160K)

Final

Evidence reviews underpinning recommendations 1.5.1 to 1.5.17 and the recommendations for research in the NICE guideline

Developed by NICE

Disclaimer: The recommendations in this guideline represent the view of NICE, arrived at after careful consideration of the evidence available. When exercising their judgement, professionals are expected to take this guideline fully into account, alongside the individual needs, preferences and values of their patients or service users. The recommendations in this guideline are not mandatory and the guideline does not override the responsibility of healthcare professionals to make decisions appropriate to the circumstances of the individual patient, in consultation with the patient and/or their carer or guardian.

Local commissioners and/or providers have a responsibility to enable the guideline to be applied when individual health professionals and their patients or service users wish to use it. They should do so in the context of local and national priorities for funding and developing services, and in light of their duties to have due regard to the need to eliminate unlawful discrimination, to advance equality of opportunity and to reduce health inequalities. Nothing in this guideline should be interpreted in a way that would be inconsistent with compliance with those duties.

NICE guidelines cover health and care in England. Decisions on how they apply in other UK countries are made by ministers in the Welsh Government, Scottish Government, and Northern Ireland Executive. All NICE guidance is subject to regular review and may be updated or withdrawn.

Bookshelf ID: NBK603292PMID: 38691634

Table 1PICO characteristics of review question

Population	Inclusion: adults with diagnosed vitamin B12 deficiency Exclusion: metabolic disorders Stratified by: Cause (dietary/non-dietary/drug-induced)
Interventions	Hydroxocobalamin Intramuscular injection Subcutaneous injection Cyanocobalamin Oral tablets Intramuscular injection Subcutaneous injection Combinations and sequences of the interventions above Stratified by: Dosage Frequency Duration
Comparisons	Each other (any differences in drug, route of administration, dosage, frequency, or duration) Dietary advice (dietary cause) Placebo No treatment Changing drug treatment (drug-induced)
Outcomes	All outcomes are considered equally important for decision making and therefore have all been rated as critical: quality of life (such as EQ5D, SF36) patient-reported outcomes (PROM scores including some/all symptoms): fatigue sleep peripheral neuropathy cognition psychiatric symptoms pain haematological values complications and adverse events mortality bleeds self-harm nerve damage frailty/falls severe cognitive effects postural hypotension adherence to treatment education/work absence Time points: short-term (up to and including 3 months), long-term (over 3 months)
Study design	Randomised controlled trials Systematic reviews of RCTs Non-randomised studies if insufficient RCT evidence is identified

Table 2Summary of randomised controlled trials included in the evidence review

Study	Intervention and comparison	Population	Outcomes	Comments
Castelli 2011¹	Oral cyanocobalamin (1000mcg per day for 90 days) n=24 Vs Intramuscular cyanocobalamin (1000mcg injected on days 1, 3, 7, 10, 14, 21, 30, 60 and 90) n=26	Adults (18-60y) with serum cobalamin concentration ≤350 pg/mL and gastrointestinal abnormalities, history of prolonged proton pump inhibitor use or a vegan/vegetarian diet	≤3 months: Haematological values: Total B12 Holotranscobalamin Methylmalonic acid Homocysteine Adverse events	Location: USA
Dangour 2015³	Oral cyanocobalamin (1000mcg per day for 12 months) n=99 Vs Placebo n=102	Older adults (≥75y) with moderate B12 deficiency (≥107 - ≤210 pmol/L) and haemoglobin concentration ≥110 g/L (female) or ≥120 g/L (male)	>3 months: Patient reported outcomes cognition general health Haematological values: Total B12 Holotranscobalamin Homocysteine Haemoglobin	Location: England Excluded participants with pernicious anaemia and very low B12 concentrations (<107pmol/L)
Dhonukshe-Rutten 2005⁴	Oral cyanocobalamin (1000mcg per day for 12 weeks) n=19 Vs Placebo n=24	Older adults (≥65y) with mild cobalamin deficiency (serum concentrations 100 – 300 pmol/L) and elevated plasma MMA (≥0.30 pmol/L)	≤3 months: Haematological values: Total B12 Homocysteine Methylmalonic acid	Location: The Netherlands Excluded participants with self-reported anaemia
Eussen 2005⁵	Dose comparison study with daily oral cyanocobalamin for 16 weeks: 100mcg n=24 250mcg n=25 500mcg n=22 1000mcg n=25	Older adults (>70y) with serum B12 concentration between 100-300 pmol/L, plasma MMA ≥0.26 µmol/L and serum creatinine ≤120 µmol/L	≤3 and >3 months: Haematological values: Total B12 Holotranscobalamin Homocysteine Methylmalonic acid	Location: The Netherlands Excluded participants with self-reported anaemia and very low B12 concentrations (<100pmol/L)
Kuzminski 1998⁸	Oral cyanocobalamin (2000mcg per day for 4 months) n=18 Vs Intramuscular cyanocobalamin (1000mcg on days 1, 3, 7, 10, 14, 21, 30, 60 and 90) n=15	People with serum cobalamin concentration <160 pg/mL and elevation of serum MMA, homocysteine (compared to normal controls)	>3 months: Patient reported outcomes: general health Haematological values: Total B12 Methylmalonic acid Homocysteine Mean corpuscular volume	Location: USA
Metaxas 2017¹⁰	Oral cyanocobalamin (1000mcg daily for 28 days) n=19 Vs Intramuscular hydroxocobalamin (1000mcg weekly for 28 days) n=18	Adults (≥18y) with B12 concentration <200 pmol/L	≤3 months: Haematological values: Total B12 Holotranscobalamin Homocysteine Methylmalonic acid	Location: Switzerland
Ubbink 1994¹⁶	Oral cyanocobalamin (400mcg daily for 6 weeks) n=17 Vs Placebo n=17	White males aged 20-73 years with confirmed hyperhomocysteinemia (>16.3 µmol/L)	≤3 months: Haematological values: Total B12 Homocysteine	Location: South Africa

Table 3Summary of observational studies included in the evidence review

Study

Intervention and comparison

Population

Outcomes

Comments

Couderc 2015²

Oral cyanocobalamin - 5 possible treatment regimens (80% received regimen 2):

1000µg/day for 1 month
1000µg/day for 15 days, then 1000µg/10 days for 1 month, then 1000µg/month for 2 months
1000µg three times per week for a month
1000µg/day for 1 week
1000µg/week for 1 month

n=111

Intramuscular cyanocobalamin – 2 possible regimens:

1000µg/day for 1 week, then 1000µg/month
1000µg three times per week for a month then 1000µg/month

n=14

Geriatric patients attending a geriatric university hospital with confirmed cobalamin deficiency (<200 pg/mL or <160 pg/mL for those who had already received oral or IM cobalamin treatment)

Confounders:

Cause

Mixed: 12 with pernicious anaemia, 72 with food-cobalamin malabsorption, 15 with nutritional deficiency, 26 undetermined

Severity of deficiency

Inclusion cut-off: 200 pg/mL, mean population value: 144.7 pg/mL

Severity of symptoms

38 patients had clinical signs of asthenia, oedema on the legs and dyspnea, 6 had signs of neuropathy, 15 had confusion and 53 had dementia

Age/comorbidities/polypharmacy

Median age: 85.5 +/− 7 years, 7 patients had undergone gastric surgery, 5 had alcoholism, 9 had a blood disease, 16 were on long-term metformin for diabetes mellitus, 6 were on ranitidine and 41 were on long-term omeprazole therapy for gastric disease

Renal function

Not reported

Ethnicity

Not reported

>3 months:

Haematological values:

Total B12

Non-randomised trial

Location: France

Smelt 2016¹⁵

Intramuscular hydroxocobalamin with loading dose (six injections in total, one every two weeks for two months, then one injection after three months (each injection contained 1000µg)) n=21

Intramuscular hydroxocobalamin with no loading dose (three injections in total, one injection per month for three months (each injection contained 1000µg)) n=21

No treatment n=21

Adults who had undergone a sleeve gastrectomy or Roux-en-Y gastric bypass.

Groups were matched for age, gender, pre-operative BMI, current BMI, surgical procedure.

Confounders:

Cause

Surgical

Severity of deficiency

Baseline mean B12 ranged from 200-226.2 (range 140-300 pmol/L)

Severity of symptoms

Not reported

Age/comorbidities/polypharmacy

Mean age ranged from 39-44.7 years, the majority of participants were 2 years post operation, but ranged from 0 to >5 years, 25.4% were using high-dose weight loss supplements, 68.3% were using multivitamin supplements

Renal function

Patients with renal insufficiency were excluded

Ethnicity

Not reported

≤3 months:

Haematological values:

Total B12
Methylmalonic acid

Non-randomised trial

Location: The Netherlands

16 patients (25.4 %) were using high-dose weight loss surgery (WLS) supplements, 43 patients (68.3 %) were using over-the-counter multivitamin supplements, and 4 patients (6.3 %) did not use any supplements

Table 4Clinical evidence summary: oral cyanocobalamin vs intramuscular cyanocobalamin

Outcomes	№ of participants (studies) Follow-up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	Anticipated absolute effects		Comments
Outcomes	№ of participants (studies) Follow-up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	Risk with intramuscular cyanocobalamin	Risk difference with Oral cyanocobalamin	Comments
Patient reported outcomes at >3 months (general health: marked improvement or clearing of paraesthesias, ataxia or memory loss, final values, higher is better)	33 (1 RCT) follow-up: 4 months	⨁◯◯◯ Very low^a^,^b^,^c	RR 0.83 (0.25 to 2.78)	267 per 1,000	45 fewer per 1,000 (200 fewer to 475 more)	MID (precision) = 0.8-1.25 MID (clinical importance) = 100 per 1000
Haematological values at ≤3 months (B12, pg/mL, final values, higher is better)	48 (1 RCT) follow-up: 3 months	⨁◯◯◯ Very low^b^,^c	-	The mean B12 was 2057 pg/mL	MD 100 pg/mL lower (634.25 lower to 434.25 higher)	MID = 91.5 (mean baseline SD × 0.5)
Haematological values at >3 months (B12, pg/mL, final values, higher is better)	32 (1 RCT) follow-up: 4 months	⨁⨁◯◯ Low^b^,^d	-	The mean B12 was 325 pg/mL	MD 680 pg/mL higher (391.86 higher to 968.14 higher)	MID = 34.5 (mean baseline SD × 0.5)
Haematological values at ≤3 months (holoTC, pmol/L, final values, higher is better)	48 (1 RCT) follow-up: 3 months	⨁◯◯◯ Very low^b^,^c	-	The mean holoTC was 877 pmol/L	MD 231 pmol/L lower (469.06 lower to 7.06 higher)	MID = 6 (mean baseline SD × 0.5)
Haematological values at ≤3 months (MMA, nmol/L, final values, lower is better)	48 (1 RCT) follow-up: 3 months	⨁⨁⨁◯ Moderate^b	-	The mean MMA was 149 nmol/L	MD 25 nmol/L lower (56.18 lower to 6.18 higher)	MID = 66.25 (mean baseline SD × 0.5)
Haematological values at >3 months (MMA, nmol/L, final values, lower is better)	33 (1 RCT) follow-up: 4 months	⨁⨁◯◯ Low^b^,^d	-	The mean MMA was 265 nmol/L	MD 96 nmol/L lower (200.76 lower to 8.76 higher)	MID = 3492 (mean baseline SD × 0.5)
Haematological values at ≤3 months (Hcy, µmol/L, final values, lower is better)	48 (1 RCT) follow-up: 3 months	⨁⨁◯◯ Low^b^,^c	-	The mean Hcy was 12 µmol/L	MD 1 µmol/L lower (2.98 lower to 0.98 higher)	MID = 2 (mean baseline SD × 0.5)
Haematological values at >3 months (Hcy, nmol/L, final values, lower is better)	33 (1 RCT) follow-up: 4 months	⨁⨁◯◯ Low^b^,^d	-	The mean Hcy was 12.2 µmol/L	MD 1.6 µmol/L lower (4.5 lower to 1.3 higher)	MID = 17.76 (mean baseline SD × 0.5)
Haematological values at >3 months (MCV, fL, final values, lower is better)	33 (1 RCT) follow-up: 4 months	⨁◯◯◯ Very low^b^,^c^,^d	-	The mean MCV was 91 fL	MD 1 fL lower (5.8 lower to 3.8 higher)	MID = 5.75 (mean baseline SD × 0.5)
Adverse events at ≤3 months (general, final values, lower is better)	48 (1 RCT) follow-up: 3 months	⨁◯◯◯ Very low^b^,^c^,^e	RR 1.02 (0.63 to 1.65)	577 per 1,000	12 more per 1,000 (213 fewer to 375 more)	MID (precision) = 0.8-1.25 MID (clinical importance) = 100 per 1000
Quality of life at ≤3 and >3 months	No evidence	-	-	-	-
Patient reported outcomes at ≤3 months	No evidence	-	-	-	-
Adherence to treatment at ≤3 and >3 months	No evidence	-	-	-	-
Education/work absence at ≤3 and >3 months	No evidence	-	-	-	-

a: Downgraded by 2 increments due to bias arising from deviations from the intended intervention and measurement of the outcome

b: Downgraded due to population indirectness (mixed B12 deficiency cause)

c: Downgraded by 1 increment if confidence intervals overlapped one MID and 2 increments if confidence intervals overlapped both MIDs

d: Downgraded by 1 increment due to deviations from the intended intervention

e: Downgraded by 1 increment due to bias due to measurement of the outcome

f: MIDs for continuous outcomes were as follows: B12 at ≤3 months 91.5, B12 at >3 months 34.5, holoTC at ≤3 months 6, MMA at ≤3 months 66.25, MMA at >3 months 3492, Hcy at ≤3 months 2, Hcy at >3 months 17.76, MCV at >3 months 5.75

Table 5Clinical evidence summary: oral cyanocobalamin vs placebo

Outcomes	№ of participants (studies) Follow-up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	Anticipated absolute effects		Comments
Outcomes	№ of participants (studies) Follow-up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	Risk with placebo	Risk difference with Oral cyanocobalamin	Comments
Patient reported outcomes at >3 months (general health: 30-item General Health Questionnaire, scale range unknown, final values, polarity unknown)	168 (1 RCT) follow-up: 12 months	⨁⨁⨁◯ Moderate^a	-	The mean general health score was 2.7	MD 0.3 lower (1.69 lower to 1.09 higher)	MID = 2.35 (CG baseline SD × 0.5)
Patient reported outcomes at >3 months (cognition: California Verbal Learning Test (sum of correct words in first 3 trials), scale range 0-48, final values, higher is better)	184 (1 RCT) follow-up: 12 months	⨁⨁⨁◯ Moderate^a	-	The mean number of words recalled was 24.6	MD 0.7 lower (2.64 lower to 1.24 higher)	MID = 3 (CG baseline SD × 0.5)
Patient reported outcomes at >3 months (cognition: California Verbal Learning Test - words recalled at delayed recall, scale range 0-16, final values, higher is better)	149 (1 RCT) follow-up: 12 months	⨁⨁⨁◯ Moderate^a	-	The mean number of words recalled was 7.7	MD 0.2 lower (0.74 lower to 0.34 higher)	MID = 1.55 (CG baseline SD × 0.5)
Haematological values at ≤3 months (B12, pmol/L, final values, higher is better, cyanocobalamin dose: 0.4 mg)	34 (1 RCT) follow up: 6 weeks	⨁◯◯◯ Very low^a^,^b	-	The mean B12 was 210.6	MD 167.9 higher (71.06 higher to 264.74 higher)	MID = 38.25 (CG baseline SD × 0.5)
Haematological values at ≤3 months (B12, pmol/L, change scores, higher is better, cyanocobalamin dose: 1 mg)	43 (1 RCT) follow up: 12 weeks	⨁◯◯◯ Very low^a^,^c	-	The mean change in B12 was 1	MD 280 higher (217.08 higher to 342.92 higher)	MID = 18.5 (CG follow-up SD × 0.5)
Haematological values at >3 months (B12, pmol/L, final values, higher is better)	144 (1 RCT) follow-up: 12 months	⨁◯◯◯ Very low^a^,^c	-	The mean B12 was 235.4	MD 405.5 higher (356.6 higher to 454.4 higher)	MID = 30.35 (CG baseline SD × 0.5)
Haematological values at >3 months (holoTC, pmol/L, final values, higher is better)	141 (1 RCT) follow-up: 12 months	⨁◯◯◯ Very low^a^,^c	-	The mean holoTC was 54.2	MD 185.8 higher (147.28 higher to 224.32 higher)	MID = 8.75 (CG baseline SD × 0.5)
Haematological values at ≤3 months (Hcy, umol/L, final values, lower is better, cyanocobalamin dose: 0.4 mg)	34 (1 RCT) follow up: 6 weeks	⨁◯◯◯ Very low^a^,^b^,^e	-	The mean Hcy was 30.7	MD 4.7 lower (16.12 lower to 6.72 higher)	MID = 3.06 (CG follow-up SD × 0.5)
Haematological values at ≤3 months (Hcy, umol/L, change scores, lower is better, cyanocobalamin dose: 1 mg)	43 (1 RCT) follow up: 12 weeks	⨁◯◯◯ Very low^a^,^c^,^e	-	The mean change in Hcy was -0.1	MD 1.7 lower (8.65 lower to 5.25 higher)	MID = 12.1 (CG baseline SD × 0.5)
Haematological values at >3 months (Hcy, µmol/L, final values, lower is better)	143 (1 RCT) follow-up: 12 months	⨁◯◯◯ Very low^a^,^c^,^d	-	The mean Hcy was 17.4 µmol/L	MD 3.2 µmol/L lower (4.9 lower to 1.5 lower)	MID = 2.8 (CG baseline SD × 0.5)
Haematological values at ≤3 months (MMA, µmol/L, change scores, lower is better)	43 (1 RCT) follow-up: 12 weeks	⨁◯◯◯ Very low^a^,^b^,^d	-	The mean change in MMA was 0	MD 0.18 lower (1.73 lower to 1.37 higher)	MID = 0.13 (CG follow-up SD × 0.5)
Haematological values at >3 months (haemoglobin, g/L, final values, higher is better)	149 (1 RCT) follow-up: 12 months	⨁◯◯◯ Very low^a^,^c^,^d	-	The mean haemoglobin was 137.2 g/L	MD 2.8 g/L higher (0.97 lower to 6.57 higher)	MID = 6.4 (CG baseline SD × 0.5
Haematological values at ≤3 months (folate, nmol/L, final values, higher is better)	34 (1 RCT) follow-up: 6 weeks	⨁◯◯◯ Very low^b^,^d	-	The mean folate was 6.1 nmol/L	MD 2 nmol/L lower (4.73 lower to 0.73 higher)	MID = 1.8 (CG baseline SD × 0.5)
Haematological values at >3 months (folate, nmol/L, final values, higher is better)	143 (1 RCT) follow-up: 12 months	⨁◯◯◯ Very low^a^,^c	-	The mean folate was 20.4 nmol/L	MD 0.2 nmol/L lower (4.42 lower to 4.02 higher)	MID = 6.9 (CG baseline SD × 0.5)
Quality of life at ≤3 and >3 months	No evidence	-	-	-	-
Patient reported outcomes at ≤3 months	No evidence	-	-	-	-
Complications and adverse events at ≤3 and >3 months	No evidence	-	-	-	-
Adherence to treatment at ≤3 and >3 months	No evidence	-	-	-	-
Education/work absence at ≤3 and >3 months	No evidence	-	-	-	-

a: Downgraded due to population indirectness (mixed cause of B12 deficiency)

b: Downgraded by 2 increments due to bias arising from the randomisation process and missing outcome data

c: Downgraded by 2 increments due to missing outcome data

d: Downgraded by 1 increment if the confidence interval overlapped one MID or 2 increments if the confidence interval overlapped both MIDs

e: MIDs for continuous outcomes were as follows: B12 (0.4mg) at ≤3 months 38.25, B12 (1.0mg) at ≤3 months 18.5, B12 at >3 months 30.35, holoTC at >3 months 8.75, Hcy (0.4mg) at ≤3 months 3.06, Hcy (1.0mg) at ≤3 months 12.1, Hcy at >3 months 2.8, haemoglobin at >3 months 6.4, cognition at >3 months 3, general health at >3 months 2.35, folate at ≤3 months 1.8, folate at >3 months 6.9

Table 6Clinical evidence summary: oral cyanocobalamin vs intramuscular hydroxocobalamin

Outcomes	№ of participants (studies) Follow-up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	Anticipated absolute effects		Comments
Outcomes	№ of participants (studies) Follow-up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	Risk with intramuscular hydroxocobalamin	Risk difference with Oral cyanocobalamin	Comments
Haematological values at ≤3 months (B12, pmol/L, final values, higher is better)	37 (1 RCT) follow-up: 1-month	⨁⨁⨁◯ Moderate^a	-	The mean B12 was 2796 pmol/L	MD 2442 pmol/L lower (3854.08 lower to 1029.92 lower)	MID = 12.92 (mean baseline SD × 0.5)
Haematological values at ≤3 months (holoTC, pmol/L, final values, higher is better)	37 (1 RCT) follow-up: 1-month	⨁⨁⨁◯ Moderate^a	-	The mean holoTC was 1269 pmol/L	MD 1113 pmol/L lower (2196.83 lower to 29.17 lower)	MID = 10.95 (mean baseline SD × 0.5)
Haematological values at ≤3 months (Hcy, µmol/L, final values, lower is better)	37 (1 RCT) follow-up: 1-month	⨁⨁◯◯ Low^a^,^b	-	The mean Hcy was 13.8 µmol/L	MD 5.3 µmol/L higher (2.13 higher to 8.47 higher)	MID = 2.69 (mean baseline SD × 0.5)
Haematological values at ≤3 months (MMA, nmol/L, final values, lower is better)	37 (1 RCT) follow-up: 1-month	⨁⨁⨁◯ Moderate^a	-	The mean MMA was 168 nmol/L	MD 12 nmol/L higher (33.42 lower to 57.42 higher)	MID = 104.72 (mean baseline SD × 0.5)
Adverse events at ≤3 months (final values, lower is better)	37 (1 RCT) follow-up: 1-month	⨁◯◯◯ Very low^a^,^c^,^d	not estimable	0 per 1,000	0 fewer per 1,000 (0 fewer to 0 fewer)	MID (precision) = 0.8 – 1.25 MID (clinical importance) = 100 per 1000
Quality of life at ≤3 and >3 months	No evidence	-	-	-	-
Patient reported outcome measures at ≤3 and >3 months	No evidence	-	-	-	-
Haematological values at >3 months	No evidence	-	-	-	-
Complications and adverse events at >3 months	No evidence	-	-	-	-
Adherence to treatment at ≤3 and >3 months	No evidence	-	-	-	-
Education/work absence at ≤3 and >3 months	No evidence	-	-	-	-

a: Downgraded due to population indirectness (mixed B12 deficiency cause)

b: Downgraded by 1 increment if the confidence interval overlapped one MID and 2 increments if the confidence interval overlapped both MIDs

c: Downgraded by 1 increment due to bias arising from measurement of the outcome

d: Downgraded due to zero events and small sample size (no imprecision: sample size >350, serious imprecision: sample size >70 ≤350, very serious imprecision: sample size <70)

e: MIDs for continuous outcomes were as follows: B12 at ≤3 months 12.92, holoTC at ≤3 months 10.95, Hcy at ≤3 months 2.69, MMA at ≤3 months 104.72

Table 7Clinical evidence summary: oral cyanocobalamin 100mcg vs 250mcg

Outcomes	№ of participants (studies) Follow-up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	Anticipated absolute effects		Comments
Outcomes	№ of participants (studies) Follow-up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	Risk with 250mcg	Risk difference with Oral cyanocobalamin 100mcg	Comments
Haematological values at ≤3 months (B12, pmol/L, change scores, higher is better)	49 (1 RCT) follow-up: 8 weeks	⨁◯◯◯ Very low^a^,^b^,^c	-	The mean B12 at ≤3 months was 128 pmol/L	MD 20 pmol/L lower (64.58 lower to 24.58 higher)	MID = 40.24 (mean SD at follow-up × 0.5)
Haematological values at >3 months (B12, pmol/L, change scores, higher is better)	49 (1 RCT) follow-up: 16 weeks	⨁◯◯◯ Very low^a^,^b^,^c	-	The mean B12 at >3 months was 153 pmol/L	MD 24 pmol/L lower (88.87 lower to 40.87 higher)	MID = 59.75 (mean SD at follow-up × 0.5)
Haematological values at ≤3 months (holoTC, pmol/L, change scores, higher is better)	49 (1 RCT) follow-up: 8 weeks	⨁◯◯◯ Very low^a^,^b^,^c	-	The mean holoTC at ≤3 months was 40 pmol/L	MD 14 pmol/L lower (26.76 lower to 1.24 lower)	MID = 12.63 (mean SD at follow-up × 0.5)
Haematological values at >3 months (holoTC, pmol/L, change scores, higher is better)	49 (1 RCT) follow-up: 16 weeks	⨁◯◯◯ Very low^a^,^b^,^c	-	The mean holoTC at >3 months was 48 pmol/L	MD 20 pmol/L lower (34.82 lower to 5.18 lower)	MID = 13.28 (mean SD at follow-up × 0.5)
Haematological values at ≤3 months (Hcy, µmol/L, change scores, lower is better)	49 (1 RCT) follow-up: 8 weeks	⨁◯◯◯ Very low^a^,^b^,^c	-	The mean Hcy at ≤3 months was -1 µmol/L	MD 0.5 µmol/L higher (0.27 lower to 1.27 higher)	MID = 0.76 (mean SD at follow-up × 0.5)
Haematological values at >3 months (Hcy, µmol/L, change scores, lower is better)	49 (1 RCT) follow-up: 16 weeks	⨁◯◯◯ Very low^a^,^b^,^c	-	The mean Hcy at >3 months was -1.4 µmol/L	MD 0.7 µmol/L higher (0.51 lower to 1.91 higher)	MID = 1.11 (mean SD at follow-up × 0.5)
Haematological values at ≤3 months (MMA, µmol/L, change scores, lower is better)	49 (1 RCT) follow-up: 8 weeks	⨁◯◯◯ Very low^a^,^b^,^c	-	The mean MMA at ≤3 months was -0.13 µmol/L	MD 0.04 µmol/L higher (0.05 lower to 0.13 higher)	MID = 0.082 (mean SD at follow-up × 0.5)
Haematological values at >3 months (MMA, µmol/L, change scores, lower is better)	49 (1 RCT) follow-up: 16 weeks	⨁◯◯◯ Very low^a^,^b^,^c	-	The mean MMA at >3 months was -0.14 µmol/L	MD 0.06 µmol/L higher (0.04 lower to 0.16 higher)	MID = 0.083 (mean SD at follow-up × 0.5)
Quality of life at ≤3 and >3 months	No evidence	-	-	-	-
Patient reported outcomes at ≤3 and >3 months	No evidence	-	-	-	-
Complications and adverse events at ≤3 and >3 months	No evidence	-	-	-	-
Adherence to treatment at ≤3 and >3 months	No evidence	-	-	-	-
Education/work absence at ≤3 and >3 months	No evidence	-	-	-	-

a: Downgraded by 1 increment due to bias arising from the randomisation process

b: Downgraded by 1 increment if the confidence interval crossed one MID and by 2 increments if the confidence interval crossed two MIDs (0.8 and 1.25 for dichotomous outcomes; 0.5 * median of baseline SD of the intervention and control group for continuous outcomes)

c: Downgraded due to population indirectness (mixed B12 deficiency cause)

d: MIDs for continuous outcomes were as follows: B12 at ≤3 months 40.24, B12 at >3 months 59.75, holoTC at ≤3 months 12.63, holoTC at >3 months 13.28, Hcy at ≤3 months 0.76, Hcy at >3 months 1.11, MMA at ≤3 months 0.082, MMA at >3 months 0.083

Table 8Clinical evidence summary: oral cyanocobalamin 100mcg vs 500mcg

Outcomes	№ of participants (studies) Follow-up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	Anticipated absolute effects		Comments
Outcomes	№ of participants (studies) Follow-up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	Risk with 500mcg	Risk difference with Oral cyanocobalamin 100mcg	Comments
Haematological values at ≤3 months (B12, pmol/L, change scores, higher is better)	46 (1 RCT) follow-up: 8 weeks	⨁◯◯◯ Very low^a^,^b^,^c	-	The mean B12 at ≤3 months was 182 pmol/L	MD 74 pmol/L lower (126.87 lower to 21.13 lower)	MID = 47.64 (mean SD at follow-up × 0.5)
Haematological values at >3 months (B12, pmol/L, change scores, higher is better)	45 (1 RCT) follow-up: 16 weeks	⨁◯◯◯ Very low^a^,^b^,^c	-	The mean B12 at >3 months was 212 pmol/L	MD 83 pmol/L lower (160.55 lower to 5.45 lower)	MID = 70.59 (mean SD at follow-up × 0.5)
Haematological values at ≤3 months (holoTC, pmol/L, change scores, higher is better)	46 (1 RCT) follow-up: 8 weeks	⨁◯◯◯ Very low^a^,^b^,^c	-	The mean holoTC at ≤3 months was 49 pmol/L	MD 23 pmol/L lower (35.71 lower to 10.29 lower)	MID = 11.92 (mean SD at follow-up × 0.5)
Haematological values at >3 months (holoTC, pmol/L, change scores, higher is better)	45 (1 RCT) follow-up: 16 weeks	⨁⨁◯◯ Low^a^,^c	-	The mean holoTC at >3 months was 60 pmol/L	MD 32 pmol/L lower (46.02 lower to 17.98 lower)	MID = 12.37 (mean SD at follow-up × 0.5)
Haematological values at ≤3 months (Hcy, µmol/L, change scores, lower is better)	46 (1 RCT) follow-up: 8 weeks	⨁◯◯◯ Very low^a^,^b^,^c	-	The mean Hcy at ≤3 months was -1.9 µmol/L	MD 1.4 µmol/L higher (0.47 higher to 2.33 higher)	MID = 0.88 (mean SD at follow-up × 0.5)
Haematological values at >3 months (Hcy, µmol/L, change scores, lower is better)	45 (1 RCT) follow-up: 16 weeks	⨁◯◯◯ Very low^a^,^b^,^c	-	The mean Hcy at >3 months was -2.4 µmol/L	MD 1.7 µmol/L higher (0.56 higher to 2.84 higher)	MID = 1.05 (mean SD at follow-up × 0.5)
Haematological values at ≤3 months (MMA, µmol/L, change scores, lower is better)	46 (1 RCT) follow-up: 8 weeks	⨁◯◯◯ Very low^a^,^b^,^c	-	The mean MMA at ≤3 months was -0.22 µmol/L	MD 0.13 µmol/L higher (0.02 lower to 0.28 higher)	MID = 0.12 (mean SD at follow-up × 0.5)
Haematological values at >3 months (MMA, µmol/L, change scores, lower is better)	45 (1 RCT) follow-up: 16 weeks	⨁◯◯◯ Very low^a^,^b^,^c	-	The mean MMA at >3 months was -0.23 µmol/L	MD 0.15 µmol/L higher (0 to 0.3 higher)	MID = 0.12 (mean SD at follow-up × 0.5)
Quality of life at ≤3 and >3 months	No evidence	-	-	-	-
Patient reported outcomes at ≤3 and >3 months	No evidence	-	-	-	-
Complications and adverse events at ≤3 and >3 months	No evidence	-	-	-	-
Adherence to treatment at ≤3 and >3 months	No evidence	-	-	-	-
Education/work absence at ≤3 and >3 months	No evidence	-	-	-	-

a: Downgraded by 1 increment due to bias arising from the randomisation process

b: Downgraded by 1 increment if the confidence interval crossed one MID and by 2 increments if the confidence interval crossed two MIDs (0.8 and 1.25 for dichotomous outcomes; 0.5 * median of baseline SD of the intervention and control group for continuous outcomes)

c: Downgraded due to population indirectness (mixed B12 deficiency cause)

d: MIDs for continuous outcomes were as follows: B12 at ≤3 months 47.64, B12 at >3 months 70.59, holoTC at ≤3 months 11.92, holoTC at >3 months 12.37, Hcy at ≤3 months 0.88, Hcy at >3 months 1.05, MMA at ≤3 months 0.12, MMA at >3 months 0.12

Table 9Clinical evidence summary: oral cyanocobalamin 100mcg vs 1000mcg

Outcomes	№ of participants (studies) Follow-up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	Anticipated absolute effects		Comments
Outcomes	№ of participants (studies) Follow-up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	Risk with 1000mcg	Risk difference with Oral cyanocobalamin 100mcg	Comments
Haematological values at ≤3 months (B12, pmol/L, change scores, higher is better)	49 (1 RCT) follow-up: 8 weeks	⨁⨁◯◯ Low^a^,^c	-	The mean B12 at ≤3 months was 248 pmol/L	MD 140 pmol/L lower (209.34 lower to 70.66 lower)	MID = 60.06 (mean SD at follow-up × 0.5)
Haematological values at ≤3 months (holoTC, pmol/L, change scores, higher is better)	49 (1 RCT) follow-up: 8 weeks	⨁⨁◯◯ Low^a^,^c	-	The mean holoTC at ≤3 months was 67 pmol/L	MD 41 pmol/L lower (55.23 lower to 26.77 lower)	MID = 13.59 (mean SD at follow-up × 0.5)
Haematological values at >3 months (holoTC, pmol/L, change scores, higher is better)	49 (1 RCT) follow-up: 16 weeks	⨁⨁◯◯ Low^a^,^c	-	The mean holoTC at >3 months was 73 pmol/L	MD 45 pmol/L lower (59.82 lower to 30.18 lower)	MID = 13.59 (mean SD at follow-up × 0.5)
Haematological values at ≤3 months (Hcy, µmol/L, change scores, lower is better)	49 (1 RCT) follow-up: 8 weeks	⨁⨁◯◯ Low^a^,^c	-	The mean Hcy at ≤3 months was -5.1 µmol/L	MD 4.6 µmol/L higher (1.3 lower to 10.5 higher)	MID = 4.27 (mean SD at follow-up × 0.5)
Haematological values at >3 months (Hcy, µmol/L, change scores, lower is better)	49 (1 RCT) follow-up: 16 weeks	⨁◯◯◯ Very low^a^,^b^,^c	-	The mean Hcy at >3 months was -5.7 µmol/L	MD 5 µmol/L higher (2.07 lower to 12.07 higher)	MID = 5.22 (mean SD at follow-up × 0.5)
Haematological values at ≤3 months (MMA, µmol/L, change scores, lower is better)	49 (1 RCT) follow-up: 8 weeks	⨁◯◯◯ Very low^a^,^b^,^c	-	The mean MMA at ≤3 months was -0.31 µmol/L	MD 0.22 µmol/L higher (0.14 lower to 0.58 higher)	MID = 0.27 (mean SD at follow-up × 0.5)
Haematological values at >3 months (MMA, µmol/L, change scores, lower is better)	49 (1 RCT) follow-up: 16 weeks	⨁◯◯◯ Very low^a^,^b^,^c	-	The mean MMA at >3 months was -0.34 µmol/L	MD 0.26 µmol/L higher (0.12 lower to 0.64 higher)	MID = 0.29 (mean SD at follow-up × 0.5)
Quality of life at ≤3 and >3 months	No evidence	-	-	-	-
Patient reported outcomes at ≤3 and >3 months	No evidence	-	-	-	-
Complications and adverse events at ≤3 and >3 months	No evidence	-	-	-	-
Adherence to treatment at ≤3 and >3 months	No evidence	-	-	-	-
Education/work absence at ≤3 and >3 months	No evidence	-	-	-	-

a: Downgraded by 1 increment due to bias arising from the randomisation process

b: Downgraded by 1 increment if the confidence interval crossed one MID and by 2 increments if the confidence interval crossed two MIDs (0.8 and 1.25 for dichotomous outcomes; 0.5 * median of baseline SD of the intervention and control group for continuous outcomes)

c: Downgraded due to population indirectness (mixed B12 deficiency cause)

d: MIDs for continuous outcomes were as follows: B12 at ≤3 months 60.06, holoTC at ≤3 months 13.59, holoTC at >3 months 13.59, Hcy at ≤3 months 4.27, Hcy at >3 months 5.22, MMA at ≤3 months 0.27, MMA at >3 months 0.29

Table 10Clinical evidence summary: oral cyanocobalamin 250mcg vs 500mcg

Outcomes	№ of participants (studies) Follow-up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	Anticipated absolute effects		Comments
Outcomes	№ of participants (studies) Follow-up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	Risk with 500mcg	Risk difference with Oral cyanocobalamin 250mcg	Comments
Haematological values at ≤3 months (B12, pmol/L, change scores, higher is better)	47 (1 RCT) follow-up: 8 weeks	⨁◯◯◯ Very low^a^,^b^,^c	-	The mean B12 at ≤3 months was 182 pmol/L	MD 54 pmol/L lower (102.22 lower to 5.78 lower)	MID = 42.26 (mean SD at follow-up × 0.5)
Haematological values at >3 months (B12, pmol/L, change scores, higher is better)	46 (1 RCT) follow-up: 16 weeks	⨁◯◯◯ Very low^a^,^b^,^c	-	The mean B12 at >3 months was 212 pmol/L	MD 59 pmol/L lower (129.62 lower to 11.62 higher)	MID = 62.22 (mean SD at follow-up × 0.5)
Haematological values at ≤3 months (holoTC, pmol/L, change scores, higher is better)	47 (1 RCT) follow-up: 8 weeks	⨁◯◯◯ Very low^a^,^b^,^c	-	The mean holoTC at ≤3 months was 49 pmol/L	MD 9 pmol/L lower (22.38 lower to 4.38 higher)	MID = 12.68 (mean SD at follow-up × 0.5)
Haematological values at >3 months (holoTC, pmol/L, change scores, higher is better)	46 (1 RCT) follow-up: 16 weeks	⨁◯◯◯ Very low^a^,^b^,^c	-	The mean holoTC at >3 months was 60 pmol/L	MD 12 pmol/L lower (27.39 lower to 3.39 higher)	MID = 13.77 (mean SD at follow-up × 0.5)
Haematological values at ≤3 months (Hcy, µmol/L, change scores, lower is better)	47 (1 RCT) follow-up: 8 weeks	⨁◯◯◯ Very low^a^,^b^,^c	-	The mean Hcy at ≤3 months was -1.9 µmol/L	MD 0.9 µmol/L higher (0.18 lower to 1.98 higher)	MID = 1.01 (mean SD at follow-up × 0.5)
Haematological values at >3 months (Hcy, µmol/L, change scores, lower is better)	46 (1 RCT) follow-up: 16 weeks	⨁◯◯◯ Very low^a^,^b^,^c	-	The mean Hcy at >3 months was -2.4 µmol/L	MD 1 µmol/L higher (0.27 lower to 2.27 higher)	MID = 1.17 (mean SD at follow-up × 0.5)
Haematological values at ≤3 months (MMA, µmol/L, change scores, lower is better)	47 (1 RCT) follow-up: 8 weeks	⨁◯◯◯ Very low^a^,^b^,^c	-	The mean MMA at ≤3 months was -0.22 µmol/L	MD 0.09 µmol/L higher (0.07 lower to 0.25 higher)	MID = 0.14 (mean SD at follow-up × 0.5)
Haematological values at >3 months (MMA, µmol/L, change scores, lower is better)	46 (1 RCT) follow-up: 16 weeks	⨁◯◯◯ Very low^a^,^b^,^c	-	The mean MMA at >3 months was -0.23 µmol/L	MD 0.09 µmol/L higher (0.08 lower to 0.26 higher)	MID = 0.14 (mean SD at follow-up × 0.5)
Quality of life at ≤3 and >3 months	No evidence	-	-	-	-
Patient reported outcomes at ≤3 and >3 months	No evidence	-	-	-	-
Complications and adverse events at ≤3 and >3 months	No evidence	-	-	-	-
Adherence to treatment at ≤3 and >3 months	No evidence	-	-	-	-
Education/work absence at ≤3 and >3 months	No evidence	-	-	-	-

a: Downgraded by 1 increment due to bias arising from the randomisation process

b: Downgraded by 1 increment if the confidence interval crossed one MID and by 2 increments if the confidence interval crossed two MIDs (0.8 and 1.25 for dichotomous outcomes; 0.5 * median of baseline SD of the intervention and control group for continuous outcomes)

c: Downgraded due to population indirectness (mixed B12 deficiency cause)

d: MIDs for continuous outcomes were as follows: B12 at ≤3 months 42.26, B12 at >3 months 62.22, holoTC at ≤3 months 12.68, holoTC at >3 months 13.77, Hcy at ≤3 months 1.01, Hcy at >3 months 1.17, MMA at ≤3 months 0.14, MMA at >3 months 0.14

Table 11Clinical evidence summary: oral cyanocobalamin 250mcg vs 1000mcg

Outcomes	№ of participants (studies) Follow-up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	Anticipated absolute effects		Comments
Outcomes	№ of participants (studies) Follow-up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	Risk with 1000mcg	Risk difference with Oral cyanocobalamin 250mcg	Comments
Haematological values at ≤3 months (B12, pmol/L, change scores, higher is better)	50 (1 RCT) follow-up: 8 weeks	⨁◯◯◯ Very low^a^,^b^,^c	-	The mean B12 at ≤3 months was 248 pmol/L	MD 120 pmol/L lower (185.86 lower to 54.14 lower)	MID = 55.72 (mean SD at follow-up × 0.5)
Haematological values at ≤3 months (holoTC, pmol/L, change scores, higher is better)	50 (1 RCT) follow-up: 8 weeks	⨁◯◯◯ Very low^a^,^b^,^c	-	The mean holoTC at ≤3 months was 67 pmol/L	MD 27 pmol/L lower (41.83 lower to 12.17 lower)	MID = 14.35 (mean SD at follow-up × 0.5)
Haematological values at >3 months (holoTC, pmol/L, change scores, higher is better)	50 (1 RCT) follow-up: 16 weeks	⨁◯◯◯ Very low^a^,^b^,^c	-	The mean holoTC at >3 months was 73 pmol/L	MD 25 pmol/L lower (41.12 lower to 8.88 lower)	MID = 14.99 (mean SD at follow-up × 0.5)
Haematological values at ≤3 months (Hcy, µmol/L, change scores, lower is better)	50 (1 RCT) follow-up: 8 weeks	⨁◯◯◯ Very low^a^,^b^,^c	-	The mean Hcy at ≤3 months was -5.1 µmol/L	MD 4.1 µmol/L higher (1.83 lower to 10.03 higher)	MID = 4.4 (mean SD at follow-up × 0.5)
Haematological values at >3 months (Hcy, µmol/L, change scores, lower is better)	50 (1 RCT) follow-up: 16 weeks	⨁◯◯◯ Very low^a^,^b^,^c	-	The mean Hcy at >3 months was -5.7 µmol/L	MD 4.3 µmol/L higher (2.79 lower to 11.39 higher)	MID = 5.33 (mean SD at follow-up × 0.5)
Haematological values at ≤3 months (MMA, µmol/L, change scores, lower is better)	50 (1 RCT) follow-up: 8 weeks	⨁◯◯◯ Very low^a^,^b^,^c	-	The mean MMA at ≤3 months was -0.31 µmol/L	MD 0.18 µmol/L higher (0.19 lower to 0.55 higher)	MID = 0.29 (mean SD at follow-up × 0.5)
Haematological values at >3 months (MMA, µmol/L, change scores, lower is better)	50 (1 RCT) follow-up: 16 weeks	⨁◯◯◯ Very low^a^,^b^,^c	-	The mean MMA at >3 months was -0.34 µmol/L	MD 0.2 µmol/L higher (0.19 lower to 0.59 higher)	MID = 0.31 (mean SD at follow-up × 0.5)
Quality of life at ≤3 and >3 months	No evidence	-	-	-	-
Patient reported outcomes at ≤3 and >3 months	No evidence	-	-	-	-
Complications and adverse events at ≤3 and >3 months	No evidence	-	-	-	-
Adherence to treatment at ≤3 and >3 months	No evidence	-	-	-	-
Education/work absence at ≤3 and >3 months	No evidence	-	-	-	-

a: Downgraded by 1 increment due to bias arising from the randomisation process

b: Downgraded by 1 increment if the confidence interval crossed one MID and by 2 increments if the confidence interval crossed two MIDs (0.8 and 1.25 for dichotomous outcomes; 0.5 * median of baseline SD of the intervention and control group for continuous outcomes)

c: Downgraded due to population indirectness (mixed B12 deficiency cause)

d: MIDs for continuous outcomes were as follows: B12 at ≤3 months 55.72, holoTC at ≤3 months 14.35, holoTC at >3 months 14.99, Hcy at ≤3 months 4.4, Hcy at >3 months 5.33, MMA at ≤3 months 0.29, MMA at >3 months 0.31

Table 12Clinical evidence summary: oral cyanocobalamin 500mcg vs 1000mcg

Outcomes	№ of participants (studies) Follow-up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	Anticipated absolute effects		Comments
Outcomes	№ of participants (studies) Follow-up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	Risk with 1000mcg	Risk difference with Oral cyanocobalamin 500mcg	Comments
Haematological values at ≤3 months (B12, pmol/L, change scores, higher is better)	47 (1 RCT) follow-up: 8 weeks	⨁◯◯◯ Very low^a^,^b^,^c	-	The mean B12 at ≤3 months was 248 pmol/L	MD 66 pmol/L lower (137.74 lower to 5.74 higher)	MID = 61.84 (mean SD at follow-up × 0.5)
Haematological values at ≤3 months (holoTC, pmol/L, change scores, higher is better)	47 (1 RCT) follow-up: 8 weeks	⨁◯◯◯ Very low^a^,^b^,^c	-	The mean holoTC at ≤3 months was 67 pmol/L	MD 18 pmol/L lower (32.79 lower to 3.21 lower)	MID = 13.63 (mean SD at follow-up × 0.5)
Haematological values at >3 months (holoTC, pmol/L, change scores, higher is better)	46 (1 RCT) follow-up: 16 weeks	⨁◯◯◯ Very low^a^,^b^,^c	-	The mean holoTC at >3 months was 73 pmol/L	MD 13 pmol/L lower (28.39 lower to 2.39 higher)	MID = 14.08 (mean SD at follow-up × 0.5)
Haematological values at ≤3 months (Hcy, µmol/L, change scores, lower is better)	47 (1 RCT) follow-up: 8 weeks	⨁◯◯◯ Very low^a^,^b^,^c	-	The mean Hcy at ≤3 months was -5.1 µmol/L	MD 3.2 µmol/L higher (2.75 lower to 9.15 higher)	MID = 4.52 (mean SD at follow-up × 0.5)
Haematological values at >3 months (Hcy, µmol/L, change scores, lower is better)	46 (1 RCT) follow-up: 16 weeks	⨁◯◯◯ Very low^a^,^b^,^c	-	The mean Hcy at >3 months was -5.7 µmol/L	MD 3.3 µmol/L higher (3.78 lower to 10.38 higher)	MID = 5.27 (mean SD at follow-up × 0.5)
Haematological values at ≤3 months (MMA, µmol/L, change scores, lower is better)	47 (1 RCT) follow-up: 8 weeks	⨁◯◯◯ Very low^a^,^b^,^c	-	The mean MMA at ≤3 months was -0.31 µmol/L	MD 0.09 µmol/L higher (0.3 lower to 0.48 higher)	MID = 0.33 (mean SD at follow-up × 0.5)
Haematological values at >3 months (MMA, µmol/L, change scores, lower is better)	46 (1 RCT) follow-up: 16 weeks	⨁◯◯◯ Very low^a^,^b^,^c	-	The mean MMA at >3 months was -0.34 µmol/L	MD 0.11 µmol/L higher (0.3 lower to 0.52 higher)	MID = 0.34 (mean SD at follow-up × 0.5)
Quality of life at ≤3 and >3 months	No evidence	-	-	-	-
Patient reported outcomes at ≤3 and >3 months	No evidence	-	-	-	-
Complications and adverse events at ≤3 and >3 months	No evidence	-	-	-	-
Adherence to treatment at ≤3 and >3 months	No evidence	-	-	-	-
Education/work absence at ≤3 and >3 months	No evidence	-	-	-	-

a: Downgraded by 1 increment due to bias arising from the randomisation process

b: Downgraded by 1 increment if the confidence interval crossed one MID and by 2 increments if the confidence interval crossed two MIDs (0.8 and 1.25 for dichotomous outcomes; 0.5 * median of baseline SD of the intervention and control group for continuous outcomes)

c: Downgraded due to population indirectness (mixed B12 deficiency cause)

d: MIDs for continuous outcomes were as follows: B12 at ≤3 months 61.84, holoTC at ≤3 months 13.63, holoTC at >3 months 14.08, Hcy at ≤3 months 4.52, Hcy at >3 months 5.27. MMA at ≤3 months 0.33, MMA at >3 months 0.34

Table 13Clinical evidence summary: oral cyanocobalamin vs intramuscular cyanocobalamin (observational studies)

Outcomes	№ of participants (studies) Follow-up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	Anticipated absolute effects		Comments
Outcomes	№ of participants (studies) Follow-up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	Risk with intramuscular cyanocobalamin	Risk difference with Oral cyanocobalamin	Comments
Haematological values - B12 at >3 months (pg/mL, final values, higher is better)	125 (1 observational study) follow up: 3.5 months	⨁◯◯◯ Very low^a^,^b	-	The mean haematological values - B12 at >3 months (pg/mL, final values, higher is better) was 730.9 pg/mL	MD 451.5 pg/mL lower (824.97 lower to 78.03 lower)	MID = 21.03 (baseline population SD × 0.5)
Quality of life at ≤3 and >3 months	No evidence	-	-	-	-
Patient reported outcomes at ≤3 and >3 months	No evidence	-	-	-	-
Complications and adverse events at ≤3 and >3 months	No evidence	-	-	-	-
Adherence to treatment at ≤3 and >3 months	No evidence	-	-	-	-
Education/work absence at ≤3 and >3 months	No evidence	-	-	-	-

a: Downgraded by 2 increments due to risk of bias arising from confounding, selection of participants into the study, classification of the interventions, deviations from the intended interventions, missing data and measurement of the outcome

b: Downgraded by 2 increments due to mixed causes of B12 deficiency

c: MID for hematological values – B12 at >3 months = 21.03

Table 14Clinical evidence summary: intramuscular hydroxocobalamin (loading dose) vs intramuscular hydroxocobalamin (no loading dose) (observational studies)

Outcomes	№ of participants (studies) Follow-up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	Anticipated absolute effects		Comments
Outcomes	№ of participants (studies) Follow-up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	Risk with intramuscular hydroxocobalamin (no loading dose)	Risk difference with Intramuscular hydroxocobalamin (loading dose)	Comments
Haematological values - B12 at <3 months (pmol/L, final values, higher is better)	42 (1 observational study) follow up: 3 months	⨁◯◯◯ Very low^a^,^b	-	The mean haematological values - B12 at <3 months (pmol/L, final values, higher is better) was 332.9 pmol/L	MD 217.4 pmol/L higher (13.73 lower to 448.53 higher)	MID = 20.4 (baseline population SD × 0.5)
Haematological values - MMA at <3 months (nmol/L, final values, lower is better)	42 (1 observational study) follow up: 3 months	⨁◯◯◯ Very low^a^,^b	-	The mean haematological values - MMA at <3 months (nmol/L, final values, lower is better) was 281.7 nmol/L	MD 100.6 nmol/L lower (164.48 lower to 36.72 lower)	MID = 107.35 (baseline population SD × 0.5)
Quality of life at ≤3 and >3 months	No evidence	-	-	-	-
Patient reported outcomes at ≤3 and >3 months	No evidence	-	-	-	-
Complications and adverse events at ≤3 and >3 months	No evidence	-	-	-	-
Adherence to treatment at ≤3 and >3 months	No evidence	-	-	-	-
Education/work absence at ≤3 and >3 months	No evidence	-	-	-	-

a: Downgraded by 2 increments due to risk of bias arising from confounding and selection of participants into the study

b: Downgraded by 1 increment if the confidence interval crossed one MID and 2 increments if the confidence interval cross both MIDs

c: MIDs for hematological values – B12 at <3 months = 20.4 and for hematological values – MMA at <3 months = 107.35

Table 15Clinical evidence summary: intramuscular hydroxocobalamin (loading dose) vs no treatment (observational studies)

Outcomes	№ of participants (studies) Follow-up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	Anticipated absolute effects		Comments
Outcomes	№ of participants (studies) Follow-up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	Risk with no treatment	Risk difference with Intramuscular hydroxocobalamin (loading dose)	Comments
Haematological values - B12 at <3 months (pmol/L, final values, higher is better)	42 (1 observational study) follow up: 3 months	⨁⨁◯◯ Low^a	-	The mean haematological values - B12 at <3 months (pmol/L, final values, higher is better) was 211.4 pmol/L	MD 288.9 pmol/L higher (95 higher to 482.8 higher)	MID = 17.65 (population baseline SD × 0.5)
Haematological values - MMA at <3 months (nmol/L, final values, lower is better)	42 (1 observational study) follow up: 3 months	⨁⨁◯◯ Low^a	-	The mean haematological values - MMA at <3 months (nmol/L, final values, lower is better) was 514.3 nmol/L	MD 333.2 nmol/L lower (437.8 lower to 228.6 lower)	MID = 104.33 (population baseline SD × 0.5)
Quality of life at ≤3 and >3 months	No evidence	-	-	-	-
Patient reported outcomes at ≤3 and >3 months	No evidence	-	-	-	-
Complications and adverse events at ≤3 and >3 months	No evidence	-	-	-	-
Adherence to treatment at ≤3 and >3 months	No evidence	-	-	-	-
Education/work absence at ≤3 and >3 months	No evidence	-	-	-	-

a: Downgraded by 2 increments due to risk of bias arising from confounding and selection of participants into the study

b: MIDs for hematological values – B12 at <3 months = 17.65 and for hematological values – MMA at <3 months = 104.33

Table 16Clinical evidence summary: intramuscular hydroxocobalamin (no loading dose) vs no treatment (observational studies)

Outcomes	№ of participants (studies) Follow-up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	Anticipated absolute effects		Comments
Outcomes	№ of participants (studies) Follow-up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	Risk with no treatment	Risk difference with Intramuscular hydroxocobalamin (no loading dose)	Comments
Haematological values - B12 at <3 months (pmol/L, final values, higher is better)	42 (1 observational study) follow up: 3 months	⨁◯◯◯ Very low^a^,^b	-	The mean haematological values - B12 at <3 months (pmol/L, final values, higher is better) was 211.4 pmol/L	MD 121.5 pmol/L higher (6.33 lower to 249.33 higher)	MID = 20.05 (population baseline SD × 0.5)
Haematological values - MMA at <3 months (nmol/L, final values, lower is better)	42 (1 observational study) follow up: 3 months	⨁⨁◯◯ Low^a	-	The mean haematological values - MMA at <3 months (nmol/L, final values, lower is better) was 514.3 nmol/L	MD 232.6 nmol/L lower (348.78 lower to 116.42 lower)	MID = 81.03 (population baseline SD × 0.5)
Quality of life at ≤3 and >3 months	No evidence	-	-	-	-
Patient reported outcomes at ≤3 and >3 months	No evidence	-	-	-	-
Complications and adverse events at ≤3 and >3 months	No evidence	-	-	-	-
Adherence to treatment at ≤3 and >3 months	No evidence	-	-	-	-
Education/work absence at ≤3 and >3 months	No evidence	-	-	-	-

a: Downgraded by 2 increments due to risk of bias arising from confounding and selection of participants into the study

b: Downgraded by 1 increment if the confidence interval crossed one MID and 2 increments if the confidence interval crossed both MIDs

Table 17Health economic evidence profile: oral B12 vs intramuscular B12

Study

Applicability

Limitations

Other comments

Incremental cost

Incremental effects

Cost effectiveness

Uncertainty

Vidal-Alaball, 2006¹⁸

United Kingdom

Directly applicable^(a)

Potentially serious limitations^(b)

A cost-comparison analysis model based on RCT data assuming no difference in outcomes of treatment strategies.
Population: Patients currently receiving i.m. B12 treatment
Comparators: Continuing i.m. or converting to oral B12 treatment

Time horizon: 2 years

£49.12^(c)

N/A

Over the two-year time horizon of the study, i.m. treatment was the lowest cost strategy. After the first year which incorporated conversion costs, oral treatment was less costly than i.m. therefore with a longer time horizon oral treatment would be the lower cost strategy.

The variable ‘‘home visit’’ had the most significant impact on the results. The authors found that even if there were no home visits required for i.m. treatment, oral treatment from year two was always less costly than i.m.

The annual cost from year 2 for oral treatment £35.55 versus £55.99 for i.m. treatment. If home visits are required for the annual cost of i.m. rises to £99.99 Therefore from year 2 onwards the oral cost will be lower than i.m. treatment.

Mnatzaganian, 2015¹¹, Australia

Partially applicable^(d)

Potentially serious limitations^(e)

Decision tree model comparing five mutually exclusive diagnostic-therapeutic strategies using serum cobalamin to test.
Cost-utility analysis (QALYs)
Population: 18 years of age or older, presenting with fatigue
Comparators:
1. No test and no treatment
2. Serum test and treat with i.m.
3. Serum test and treat with oral supplement
4. Do not test and treat all with i.m.
5. Do not test and treat all with oral
Time horizon: 3 months.

3-2: −£23^(f)

5-4: −£94^(f)

QALYs

3-2: 0

5-4: 0

N/A

Oral treatment was less costly in both cases compared to i.m. when patients undergo a test or if no test is conducted. There were no differences in QALYs thus oral treatment is the lower cost option compared to i.m. treatment.

The strategy “do not test but treat all with oral supplements” had a probability of being cost effective (£20/£30K threshold): 100%

A pa was performed which included sensitivity and specificity of the diagnostic test, cost, and utility input parameter values. Further deterministic sensitivity analysis was done with varying the prevalence levels of cobalamin deficiency at 1%, 5% 10% and 20%. The results remained robust in all analyses.

Houle 2014⁶ ([Canada])

Partially applicable^(g)

Potentially serious limitations^(h)

Cost-comparison analysis
Based on 3 RCTs assuming equivalence in clinical effectiveness for interventions
Population: >65 years, receiving intramuscular B12 treatment
Intervention: B12 tablet (oral) 1000mcg, one tablet once daily
Comparators: B12 10ml vial for injection, 1ml/1000mcg administered once per month

Time horizon: 5 years

£-279⁽ⁱ⁾

N/A

Sensitivity analysis showed that proportion of people assumed to have an additional physician visit in the i.m. arm had a significant impact on the cost savings.

: Abbreviations: i.m. = intramuscular, pa = probabilistic sensitivity analysis, QALY= quality-adjusted life years, RCT= randomised controlled trial.

(a): The clinical evidence indicated no difference in effectiveness between oral and i.m. vitamin B12 treatment therefore a cost-comparison economic evaluation was used to compare costs between the interventions. This study has a focus on primary care in the UK, with local costings.

(b): This is not a cost-utility evaluation, the study incorporates costs of converting therapy which may not be directly applicable for example the increased blood test monitoring which doesn’t routinely take place since the onset of COVID-19 whereby patients on parenteral have been converted to oral treatment with no increased initial monitoring. There is no account of whether there may be any outcome differences or the direct impact on patients i.e., side effects/acceptability.

(c): 2006 UK pounds. Cost components incorporated: Medication cost, laboratory cost, GP/nursing time, conversion costs from i.m. to oral (assuming three additional physician visits and additional blood testing), syringe and needle (for i.m.)

(d): The UK tariff was not used for EQ-5D; The utility scores were derived using hypothetical state scenarios which may not be comparable to one obtained by using patient data. Study is Australia based.

(e): This study focuses on diagnosis and intervention together using serum cobalamin testing with oral and i.m. treatment. Fatigue is only one symptom which may be related to B12 deficiency, so this study does not capture all potential B12 deficient patients. Only the first consultation was considered; all screening tests that could have been ordered—other than serum cobalamin—were not included in this economic evaluation. The cost of misdiagnosis/referral to specialists was not included which could have a significant impact of the cost-effectiveness. There is uncertainty regarding the baseline prevalence of B12 deficiency. Risk of recurrence of deficiency or symptoms after three months were not explored.

(f): 2013 US dollars converted to UK pounds using purchasing power parities. Cost components incorporated: GP-patient consultation fee, serum cobalamin test, Patient Episode Initiation (specimen) fees, medication costs, service costs for i.m. injections.

(g): Canadian perspective and cost comparison analysis rather than cost-utility analysis. B12 ingredient is not explicitly stated, if the intramuscular version is cyanocobalamin, this is not used in NHS primary care. Limited short term clinical evidence is used to estimate that the interventions are identical in terms of clinical effectiveness.

(h): The price of oral vitamin B12 treatment is much lower than the current NHS drug tariff. For the cost of the oral treatment, this is assumed to be funded 70% by the patient therefore savings. The study incorporates costs of converting therapy which may not be directly applicable for example the increased blood test monitoring which doesn’t routinely take place in the NHS. There is no account of whether there may be any outcome differences or the direct impact on patients i.e. side effects/acceptability.

(i): 2012 Canada Canadian Dollars converted to UK pounds¹⁴ Cost components incorporated: Medication cost, dispensing fee of oral, fee for intramuscular administration (in pharmacy and in physician office), blood sampling cost (staff time), cost for laboratory test and laboratory processing cost, physician consultation cos

Table 18Unit costs of B12 treatment

Resource	Unit costs (per tablet/vial)	Source
Oral - cyanocobalamin 50mcg	£0.43	NHS electronic drug tariff¹³
Oral - Orobalin 1mg	£0.33	NHS electronic drug tariff¹³
Parenteral – cyanocobalamin 1mg/ml	£2.90	NHS electronic drug tariff¹³
Parenteral – hydroxocobalamin	£0.89	NHS electronic drug tariff¹³
HCP administration appointment (10 mins) (for parenteral treatment only)	£7.00	Unit costs of health and social care 2021⁷
Injection consumables (2ml syringe, 21g 25mm drawing needle, 25g 35mm needle to inject, 1 litre sharps box) (for parenteral treatment only)	£0.15	NHS electronic drug tariff¹³, Committee members’ advice

Table 19Comparison of treatment costs Orobalin vs Parenteral treatment

Timeline	Orobalin 1mg daily	Parenteral hydroxocobalamin – 6mg in first month then 1mg per two months	Parenteral hydroxocobalamin – 6mg in first month then 1mg per three months
3 months	£30	£56	£48
6 months	£60	£64	£56
1 year	£119	£88	£72
2 years	£233	£134	£103

Table 20Comparison of treatment costs Orobalin vs Parenteral treatment (dose for pernicious anaemia)

Timeline	Orobalin 4mg daily in first month then 1mg daily	Parenteral hydroxocobalamin – 6mg in first month then 1mg per two months	Parenteral hydroxocobalamin – 6mg in first month then 1mg per three months
3 months	£60	£56	£48
6 months	£90	£64	£56
1 year	£149	£88	£72
2 years	£263	£134	£103

Table 21PICO characteristics of review question

Population	Inclusion: adults with diagnosed vitamin B12 deficiency Exclusion: metabolic disorders Stratify by: physical/mental barriers to self-administration (people with barriers and people without barriers)
Interventions	Self-administration (including family/carer administration) of parenteral vitamin B12 replacement Hydroxocobalamin intramuscular injection Hydroxocobalamin subcutaneous injection Cyanocobalamin intramuscular injection Cyanocobalamin subcutaneous injection
Comparisons	Healthcare professional administration of parenteral vitamin B12 replacement Hydroxocobalamin intramuscular injection Hydroxocobalamin subcutaneous injection Cyanocobalamin intramuscular injection Cyanocobalamin subcutaneous injection
Outcomes	All outcomes are considered equally important for decision making and therefore have all been rated as critical: quality of life (such as EQ5D, SF36) patient-reported outcomes (PROM scores including some/all symptoms): fatigue sleep peripheral neuropathy cognition psychiatric symptoms pain haematological values complications and adverse events mortality bleeds self-harm nerve damage frailty/falls severe cognitive effects postural hypotension adherence to treatment education/work absence Time point: short-term (up to 3 months) and long-term (over 3 months)
Study design	Randomised controlled trials Systematic reviews of RCTs Non-randomised comparative studies if no RCTs are identified (priority will be given to inclusion of non-randomised comparative studies that have controlled/adjusted for confounding factors. If insufficient evidence is identified from studies that have controlled/adjusted for confounding factors, non-randomised comparative studies that have not controlled/adjusted for confounding factors will be considered)

Table 22Unit costs for self-administration

Resource	Unit costs (per tablet/vial)	Source
Parenteral – cyanocobalamin 1mg/ml	£2.90	NHS electronic drug tariff¹³
Parenteral – hydroxocobalamin	£0.89	NHS electronic drug tariff¹³
HCP administration appointment (10 mins)	£7.00	Unit costs of health and social care 2021⁷
HCP teaching self-administration appointment (15 mins)	£10.50	Unit costs of health and social care 2021⁷
Injection consumables (2ml syringe, 21g 25mm drawing needle, 25g 35mm needle to inject, 1 litre sharps box)	£0.15 per HCP administration £14.75 for self-administration (one-off cost which will provide consumables for up to 100 injections)	NHS electronic drug tariff¹³, Committee members’ advice

Table 23Comparison of treatment/prophylaxis costs self-administration vs HCP administration treatment (no loading doses)

Timeline	Frequency 1mg injection every 2 months		Frequency 1mg injection every 3 months
	Self-administration	HCP	Self-administration	HCP
3 months	£27	£16	£26	£8
6 months	£28	£24	£27	£16
1 year	£31	£48	£29	£32
2 years	£36	£93	£32	£62

Table 24Comparison of treatment costs self-administration vs HCP administration treatment (loading doses for pernicious anaemia)

Timeline	Frequency 6mg in first month then 1mg per two months		Frequency 6mg in first month then 1mg per three months
	Self-administration	HCP	Self-administration	HCP
1 month	£31	£48	£31	£48
3 months	£31	£56	£31	£48
6 months	£32	£64	£31	£56
1 year	£35	£88	£33	£72
2 years	£40	£134	£37	£103

Table 27Studies excluded from the clinical review

Study	Code [Reason]
Anand, S., Thomas, S., Jayachandra, M. et al. (2019) Effects of maternal B12 supplementation on neurophysiological outcomes in children: a study protocol for an extended follow-up from a placebo randomised control trial in Bangalore, India. BMJ Open 9(2): e024426 [PMC free article: PMC6377540] [PubMed: 30782904]	- study protocol
Andres, E., Affenberger, S., Vinzio, S. et al. (2005) Food-cobalamin malabsorption in elderly patients: clinical manifestations and treatment. American Journal of Medicine 118(10): 1154–9 [PubMed: 16194648]	- Data not reported in an extractable format
Andres, E., Affenberger, S., Zimmer, J. et al. (2006) Current hematological findings in cobalamin deficiency. A study of 201 consecutive patients with documented cobalamin deficiency. Clinical and Laboratory Haematology 28(1): 50–56 [PubMed: 16430460]	- Data not reported in an extractable format
Andres, E., Dali-Youcef, N., Vogel, T. et al. (2009) Oral cobalamin (vitamin B(12)) treatment. An update. International Journal of Laboratory Hematology 31(1): 1–8 [PubMed: 19032377]	- Systematic review used as source of primary studies
Andres, E.; Fothergill, H.; Mecili, M. (2010) Efficacy of oral cobalamin (vitamin B12) therapy. Expert Opinion on Pharmacotherapy 11(2): 249–56 [PubMed: 20088746]	- Systematic review used as source of primary studies
Andres, E., Kaltenbach, G., Noblet-Dick, M. et al. (2006) Hematological response to short-term oral cyanocobalamin therapy for the treatment of cobalamin deficiencies in elderly patients. Journal of Nutrition, Health & Aging 10(1): 3–6 [PubMed: 16453051]	- Full text paper not available
Andres, E., Kurtz, J. E., Perrin, A. E. et al. (2001) Oral cobalamin therapy for the treatment of patients with food-cobalamin malabsorption. American Journal of Medicine 111(2): 126–9 [PubMed: 11498066]	- Study design not relevant to this review protocol
Andres, E.; Noel, E.; Kaltenbach, G. (2004) P.o. versus i.m. cobalamin treatment in megaloblastic anemia. Clinical therapeutics 26(6): 936authorreply937 [PubMed: 15262464]	- Not a peer-reviewed publication
Andres, E., Perrin, A. E., Demangeat, C. et al. (2003) The syndrome of food-cobalamin malabsorption revisited in a department of internal medicine. A monocentric cohort study of 80 patients. European Journal of Internal Medicine 14(4): 221–226 [PubMed: 12919836]	- Comparator in study does not match that specified in this review protocol
Andres, E., Vidal-Alaball, J., Federici, L. et al. (2007) Clinical aspects of cobalamin deficiency in elderly patients. Epidemiology, causes, clinical manifestations, and treatment with special focus on oral cobalamin therapy. European Journal of Internal Medicine 18(6): 456–62 [PubMed: 17822656]	- Systematic review used as source of primary studies
Andres, E., Vogel, T., Federici, L. et al. (2008) Update on oral cyanocobalamin (vitamin B12) treatment in elderly patients. Drugs & Aging 25(11): 927–32 [PubMed: 18947260]	- Systematic review used as source of primary studies
Andres, E., Zulfiqar, A. A., Serraj, K. et al. (2018) Systematic Review and Pragmatic Clinical Approach to Oral and Nasal Vitamin B12 (Cobalamin) Treatment in Patients with Vitamin B12 Deficiency Related to Gastrointestinal Disorders. Journal of Clinical Medicine 7(10): 26 [PMC free article: PMC6210286] [PubMed: 30261596]	- Systematic review used as source of primary studies
Andres, E.; Zulfiqar, A. A.; Vogel, T. (2020) State of the art review: oral and nasal vitamin B12 therapy in the elderly. Qjm 113(1): 5–15 [PubMed: 30796433]	- No additional studies not identified in search included
Andrès, E., Loukili, N. H., Noel, E. et al. (2005) Effects of oral crystalline cyanocobalamin 1000 ug/d in the treatment of pernicious anemia: an open-label, prospective study in ten patients. Current Therapeutic Research 66(1): 13–22 [PMC free article: PMC3964566] [PubMed: 24672108]	- Comparator in study does not match that specified in this review protocol
Anonymous (2006) Oral as good as intramuscular vitamin B12 replacement. Journal of Family Practice 55(11): 940 [PubMed: 17115493]	- Duplicate reference
Anonymous (2006) Oral as good as IM vitamin B12 replacement. South African Family Practice 48(9): 12	- Systematic review used as source of primary studies
Arendt, J. F. H., Horvath-Puho, E., Sorensen, H. T. et al. (2021) Plasma Vitamin B12 Levels, High-Dose Vitamin B12 Treatment, and Risk of Dementia. Journal of Alzheimer’s Disease 79(4): 1601–1612 [PMC free article: PMC7990402] [PubMed: 33459639]	- No outcomes relevant to this protocol
Arican, Pinar, Bozkurt, Oznur, Cavusoglu, Dilek et al. (2020) Various neurological symptoms with vitamin B12 deficiency and posttreatment evaluation. Journal of Pediatric Neurosciences 15(4): 365–369 [PMC free article: PMC8078635] [PubMed: 33936299]	- Study design not relevant to this review protocol
Baer, Janine and Peter, Mary St (2011) Vitamin b12 assessment and intervention in younger adult women. Journal for Nurse Practitioners 7(2): 117–122	- Full text paper not available
Bahadir, A.; Reis, P. G.; Erduran, E. (2014) Oral vitamin B12 treatment is effective for children with nutritional vitamin B12 deficiency. Journal of Paediatrics and Child Health 50(9): 721–725 [PubMed: 24944005]	- Study design not relevant to this review protocol
Bastrup-Madsen, P. and Paulsen, L. (1955) Oral treatment of megaloblastic anaemia with small amounts of vitamin B12 and intrinsic factor. Acta haematologica 13(4): 193–206 [PubMed: 14375708]	- Study design not relevant to this review protocol
Berlin, R., Berlin, H., Brante, G. et al. (1978) Vitamin B12 body stores during oral and parenteral treatment of pernicious anaemia. Acta Medica Scandinavica 204(1–2): 81–84 [PubMed: 685735]	- Population not relevant to this review protocol
Bhowmik, B., Siddiquee, T., Mdala, I. et al. (2021) Vitamin D3 and B12 supplementation in pregnancy. Diabetes Research & Clinical Practice 174: 108728 [PubMed: 33662489]	- Population not relevant to this review protocol
Bial, A. K. (2006) Review: Limited evidence from 2 randomised controlled trials suggests that oral and intramuscular vitamin B12 have similar effectiveness for vitamin B12 deficiency. Evidence Based Medicine 11(1): 9 [PubMed: 17213051]	- Systematic review used as source of primary studies
Bjorke-Monsen, A. L., Torsvik, I., Saetran, H. et al. (2008) Common metabolic profile in infants indicating impaired cobalamin status responds to cobalamin supplementation. Pediatrics 122(1): 83–91 [PubMed: 18595990]	- Population not relevant to this review protocol
Bjorkegren, K. and Svardsudd, K. (2004) A population-based intervention study on elevated serum levels of methylmalonic acid and total homocysteine in elderly people: Results after 36 months of follow-up. Journal of Internal Medicine 256(5): 446–452 [PubMed: 15485481]	- Population not relevant to this review protocol
Bjorkegren, K. and Svardsudd, K. (1999) Elevated serum levels of methylmalonic acid and homocysteine in elderly people. A population-based intervention study. Journal of Internal Medicine 246(3): 317–324 [PubMed: 10476000]	- Not a peer-reviewed publication
Blacher, J., Czernichow, S., Raphael, M. et al. (2007) Very low oral doses of vitamin B-12 increase serum concentrations in elderly subjects with food-bound vitamin B-12 malabsorption. Journal of Nutrition 137(2): 373–8 [PubMed: 17237314]	- Study does not contain an intervention relevant to this review protocol
Boddy, K., King, P., Mervyn, L. et al. (1968) Retention of cyanocobalamin, hydroxocobalamin, and coenzyme B12 after parenteral administration. Lancet 2(7570): 710–2 [PubMed: 4175091]	- Population not relevant to this review protocol
Bolaman, Z., Kadikoylu, G., Yukselen, V. et al. (2003) Oral versus intramuscular cobalamin treatment in megaloblastic anemia: a single-center, prospective, randomized, open-label study. Clinical Therapeutics 25(12): 3124–34 [PubMed: 14749150]	- Study does not contain an intervention relevant to this review protocol
Bronstrup, A., Hages, M., Prinz-Langenohl, R. et al. (1998) Effects of folic acid and combinations of folic acid and vitamin B-12 on plasma homocysteine concentrations in healthy, young women. American Journal of Clinical Nutrition 68(5): 1104–10 [PubMed: 9808229]	- Population not relevant to this review protocol
Buccianti, G., Bamonti Catena, F., Patrosso, C. et al. (2001) Reduction of the homocysteine plasma concentration by intravenously administered folinic acid and vitamin B12 in uraemic patients on maintenance haemodialysis. American Journal of Nephrology 21(4): 294–299 [PubMed: 11509801]	- Study does not contain an intervention relevant to this review protocol
Butler, C. C., Vidal-Alaball, J., Cannings-John, R. et al. (2006) Oral vitamin B12 versus intramuscular vitamin B12 for vitamin B12 deficiency: a systematic review of randomized controlled trials. Family Practice 23(3): 279–85 [PubMed: 16585128]	- Systematic review used as source of primary studies
Cameron, D. G.; Townsend, S. R.; English, A. (1954) Pernicious anaemia II: maintenance treatment with crystalline vitamin B12. Canadian medical association journal 70(4): 398–400 [PMC free article: PMC1825750] [PubMed: 13150272]	- Population not relevant to this review protocol
Castelli, M. C., Wong, D. F., Friedman, K. et al. (2011) Pharmacokinetics of oral cyanocobalamin formulated with sodium N-[8-(2-hydroxybenzoyl)amino]caprylate (SNAC): an open-label, randomized, single-dose, parallel-group study in healthy male subjects. Clinical Therapeutics 33(7): 934–45 [PubMed: 21722960]	- Population not relevant to this review protocol
Chalmers, J. N. and Hall, Z. M. (1954) Treatment of pernicious anaemia with oral vitamin B12 without known source of intrinsic factor. British medical journal 1(4872): 1179–1181 [PMC free article: PMC2085111] [PubMed: 13160431]	- Study design not relevant to this review protocol
Chalmers, J. N. and Shinton, N. K. (1965) Comparison of hydroxocobalamin and cyanocobalamin in the treatment of pernicious anaemia. Lancet 2(7426): 1305–8 [PubMed: 4165301]	- Study design not relevant to this review protocol
Chan, C. Q.; Low, L. L.; Lee, K. H. (2016) Oral Vitamin B12 Replacement for the Treatment of Pernicious Anemia. Frontiers in Medicine 3: 38 [PMC free article: PMC4993789] [PubMed: 27602354]	- No additional studies not identified in search included
Chan, J. C., Liu, H. S., Kho, B. C. et al. (2008) Longitudinal study of Chinese patients with pernicious anaemia. Postgraduate Medical Journal 84(998): 644–50 [PubMed: 19201940]	- Comparator in study does not match that specified in this review protocol
Chandelia, S., Chandra, J., Narayan, S. et al. (2012) Addition of cobalamin to iron and folic acid improves hemoglobin rise in nutritional anemia. Indian Journal of Pediatrics 79(12): 1592–6 [PubMed: 22415494]	- Study does not contain an intervention relevant to this review protocol
Dahele, A. and Ghosh, S. (2001) Vitamin B12 deficiency in untreated celiac disease. American Journal of Gastroenterology 96(3): 745–50 [PubMed: 11280545]	- Population not relevant to this review protocol
Dangour, A. D., Allen, E., Clarke, R. et al. (2011) A randomised controlled trial investigating the effect of vitamin B12 supplementation on neurological function in healthy older people: the Older People and Enhanced Neurological function (OPEN) study protocol [ISRCTN54195799]. Nutrition Journal 10: 22 [PMC free article: PMC3062585] [PubMed: 21396086]	- study protocol
Dawson, D. W.; Lewis, M. J.; Wadsworth, L. D. (1975) Changes in erythroblast morphology as an index of response to cyanocobalamin in patients with megaloblastic anaemia. British Journal of Haematology 31(1): 77–85 [PubMed: 764853]	- Study design not relevant to this review protocol
De La Fourniere, F., Ferry, M., Cnockaert, X. et al. (1997) Vitamin B12 deficiency and dementia: a multicenter epidemiologic and therapeutic study. Preliminary therapeutic trial. Semaine des hopitaux 73(56): 133–140	- Study not reported in English
De, L. F. F., Ferry, M., Cnockaert, X. et al. (1997) Vitamin B12 deficiency and dementia: a multicenter epidemiologic and therapeutic study. Preliminary therapeutic trial. DEFICIENCE EN VITAMINE B12 ET ETAT DEMENTIEL: ETUDE EPIDEMIOLOGIQUE MULTICENTRIQUE ET THERAPEUTIQUE. ESSAI PRELIMINAIRE. Semaine des hopitaux 73(56): 133–140	- Duplicate reference
Devi, S., Mukhopadhyay, A., Dwarkanath, P. et al. (2017) Combined Vitamin B-12 and Balanced Protein-Energy Supplementation Affect Homocysteine Remethylation in the Methionine Cycle in Pregnant South Indian Women of Low Vitamin B-12 Status. Journal of Nutrition 147(6): 1094–1103 [PubMed: 28446631]	- Data not reported in an extractable format
Didangelos, T., Karlafti, E., Kotzakioulafi, E. et al. (2021) Vitamin B12 Supplementation in Diabetic Neuropathy: A 1-Year, Randomized, Double-Blind, Placebo-Controlled Trial. Nutrients 13(2): 27 [PMC free article: PMC7912007] [PubMed: 33513879]	- Study does not contain an intervention relevant to this review protocol
Duggan, C., Srinivasan, K., Thomas, T. et al. (2014) Vitamin B-12 supplementation during pregnancy and early lactation increases maternal, breast milk, and infant measures of vitamin B-12 status. Journal of Nutrition 144(5): 758–64 [PMC free article: PMC3985831] [PubMed: 24598885]	- Population not relevant to this review protocol
Favrat, B., Vaucher, P., Herzig, L. et al. (2011) Oral vitamin B12 for patients suspected of subtle cobalamin deficiency: a multicentre pragmatic randomised controlled trial. BMC Family Practice 12: 2 [PMC free article: PMC3033340] [PubMed: 21232119]	- Population not relevant to this review protocol
Finkelstein, J. L., Kurpad, A. V., Thomas, T. et al. (2017) Vitamin B12 status in pregnant women and their infants in South India. European Journal of Clinical Nutrition 71(9): 1046–1053 [PMC free article: PMC8141370] [PubMed: 28402324]	- Population not relevant to this review protocol
Glass, G. B.; Skeggs, H. R.; Lee, D. H. (1966) Hydroxocobalamin. V. Prolonged maintenance of high vitamin B12 blood levels following a short course of hydroxocobalamin injections. Blood 27(2): 234–41 [PubMed: 5322750]	- Study design not relevant to this review protocol
Gomollon, F., Gargallo, C. J., Munoz, J. F. et al. (2017) Oral cyanocobalamin is effective in the treatment of vitamin B12 deficiency in crohn’s disease. Nutrients 9(3nopagination) [PMC free article: PMC5372971] [PubMed: 28335526]	- Study design not relevant to this review protocol
Greibe, E., Mahalle, N., Bhide, V. et al. (2019) Effect of 8-week oral supplementation with 3-microg cyano-B12 or hydroxo-B12 in a vitamin B12-deficient population. European Journal of Nutrition 58(1): 261–270 [PMC free article: PMC6424936] [PubMed: 29209773]	- Study design not relevant to this review protocol
Greibe, E., Mahalle, N., Bhide, V. et al. (2018) Increase in circulating holotranscobalamin after oral administration of cyanocobalamin or hydroxocobalamin in healthy adults with low and normal cobalamin status. European Journal of Nutrition 57(8): 2847–2855 [PMC free article: PMC6267412] [PubMed: 29038891]	- Data not reported in an extractable format
Health Quality, Ontario (2013) Vitamin B12 and cognitive function: an evidence-based analysis. Ontario Health Technology Assessment Series 13(23): 1–45 [PMC free article: PMC3874776] [PubMed: 24379897]	- Systematic review used as source of primary studies
Hemsted, E. H. and Mills, J. (1958) Vitamin B12 in pernicious anaemia: intramuscular or oral. Lancet 2(7060): 1302–1303 [PubMed: 13612203]	- Study design not relevant to this review protocol
Hill, M. H., Flatley, J. E., Barker, M. E. et al. (2013) A vitamin B-12 supplement of 500 mug/d for eight weeks does not normalize urinary methylmalonic acid or other biomarkers of vitamin B-12 status in elderly people with moderately poor vitamin B-12 status. Journal of Nutrition 143(2): 142–7 [PubMed: 23236022]	- No outcomes relevant to this protocol
Hoffer, L. J., Djahangirian, O., Bourgouin, P. E. et al. (2005) Comparative effects of hydroxocobalamin and cyanocobalamin on plasma homocysteine concentrations in end-stage renal disease. Metabolism: Clinical & Experimental 54(10): 1362–7 [PubMed: 16154437]	- Population not relevant to this review protocol
Hoffer, L. J., Saboohi, F., Golden, M. et al. (2005) Cobalamin dose regimen for maximum homocysteine reduction in end-stage renal disease. Metabolism: Clinical & Experimental 54(6): 835–40 [PubMed: 15931623]	- Population not relevant to this review protocol
Hughes, D., Elwood, P. C., Shinton, N. K. et al. (1970) Clinical trial of the effect of vitamin B12 in elderly subjects with low serum B12 levels. British Medical Journal 1(5707): 458–60 [PMC free article: PMC1700490] [PubMed: 4911774]	- No outcomes relevant to this protocol
Hvas, A. M.; Ellegaard, J.; Nexo, E. (2001) Vitamin B12 treatment normalizes metabolic markers but has limited clinical effect: a randomized placebo-controlled study. Clinical Chemistry 47(8): 1396–404 [PubMed: 11468228]	- Data not reported in an extractable format
Hvas, A. M., Juul, S., Lauritzen, L. et al. (2004) No effect of vitamin B-12 treatment on cognitive function and depression: a randomized placebo controlled study. Journal of Affective Disorders 81(3): 269–73 [PubMed: 15337331]	- Data not reported in an extractable format
Hvas, A. M., Juul, S., Nexo, E. et al. (2003) Vitamin B-12 treatment has limited effect on health-related quality of life among individuals with elevated plasma methylmalonic acid: a randomized placebo-controlled study. Journal of Internal Medicine 253(2): 146–52 [PubMed: 12542554]	- Data not reported in an extractable format
Hvas, A. M., Morkbak, A. L., Hardlei, T. F. et al. (2011) The vitamin B12 absorption test, CobaSorb, identifies patients not requiring vitamin B12 injection therapy. Scandinavian Journal of Clinical & Laboratory Investigation 71(5): 432–8 [PubMed: 21623649]	- Population not relevant to this review protocol
Kaltenbach, G., Andres, E., Barnier-Figue, G. et al. (2005) Dose of oral cobalamin therapy curing within one week low vitamin B12 levels in elderly patients. Presse medicale 34(5): 358–362 [PubMed: 15859569]	- Full text paper not available
Keche, Y. N.; Yegnanarayan, R.; Bhat, S. (2016) Effect of vitamin B12 supplementation on glycemic control in poorly controlled hyperhomocysteinemic type 2 diabetic patients. Bangladesh Journal of Pharmacology 11(1): 50–53	- No outcomes relevant to this protocol
Killander, A. and Werner, I. (1968) Maintenance therapy of pernicious anaemia. Comparison of cyanocobalamin, hydroxocobalamin and a cyanocobalamin-zinc-tannate complex. Acta haematologica 40(6): 305–314 [PubMed: 4978697]	- Study design not relevant to this review protocol
Kim, H. I., Hyung, W. J., Song, K. J. et al. (2011) Oral vitamin B12 replacement: an effective treatment for vitamin B12 deficiency after total gastrectomy in gastric cancer patients. Annals of Surgical Oncology 18(13): 3711–7 [PubMed: 21556950]	- Study does not contain an intervention relevant to this review protocol
Kumaran, K., Yajnik, P., Lubree, H. et al. (2017) The Pune Rural Intervention in Young Adolescents (PRIYA) study: design and methods of a randomised controlled trial. BMC Nutrition 3: 41 [PMC free article: PMC7050839] [PubMed: 32153821]	- study protocol
Kvestad, I., Taneja, S., Kumar, T. et al. (2015) Vitamin B12 and Folic Acid Improve Gross Motor and Problem-Solving Skills in Young North Indian Children: a Randomized Placebo-Controlled Trial. PloS one 10(6): e0129915 [PMC free article: PMC4476750] [PubMed: 26098427]	- Population not relevant to this review protocol
Kwok, T., Chook, P., Qiao, M. et al. (2012) Vitamin B-12 supplementation improves arterial function in vegetarians with subnormal vitamin B-12 status. Journal of nutrition, health & aging 16(6): 569–573 [PubMed: 22659999]	- Study design not relevant to this review protocol
Kwok, T., Lee, J., Ma, R. C. et al. (2017) A randomized placebo controlled trial of vitamin B12 supplementation to prevent cognitive decline in older diabetic people with borderline low serum vitamin B12. Clinical Nutrition 36(6): 1509–1515 [PubMed: 27823800]	- Study does not contain an intervention relevant to this review protocol
Kwok, T., Tang, C., Woo, J. et al. (1998) Randomized trial of the effect of supplementation on the cognitive function of older people with subnormal cobalamin levels. International Journal of Geriatric Psychiatry 13(9): 611–6 [PubMed: 9777425]	- No outcomes relevant to this protocol
Lane, L. A. and Rojas-Fernandez, C. (2002) Treatment of vitamin b(12)-deficiency anemia: oral versus parenteral therapy. Annals of Pharmacotherapy 36(78): 1268–72 [PubMed: 12086562]	- Systematic review used as source of primary studies
Lee, M. K., Wong, P. K., Kung, K. et al. (2011) Review of vitamin B12 deficiency management in a family medicine clinic. Hong Kong Practitioner 33(2): 64–71	- Comparator in study does not match that specified in this review protocol
Limvorapitak, W. and Auewarakul, C. U. (2016) Clinical Features and Outcomes of Patients with Non-Iron Nutritional Deficiency Anemia in an In-Patient Setting at Siriraj Hospital: A 10-Year Retrospective Study. Journal of the Medical Association of Thailand 99(6): 637–44 [PubMed: 29900722]	- No outcomes relevant to this protocol
Lowenstein, L., Brunton, L., Shapiro, L. et al. (1957) Maintenance therapy of pernicious anaemia with oral administration of intrinsic factor and vitamin B12. Canadian medical association journal 77(10): 923–930 [PMC free article: PMC1824183] [PubMed: 13479845]	- Study does not contain an intervention relevant to this review protocol
Lubis, Z., Hardinsyah, Hidayat, S. et al. (2018) Effect of vitamin B12 supplement on serum levels of vitamin B12 and memorizing ability on preschool children. International Journal of Pharmaceutical Sciences and Research 9(10): 4436–4440	- Population not relevant to this review protocol
Mahawar, K. K., Reid, A., Graham, Y. et al. (2018) Oral Vitamin B12 Supplementation After Roux-en-Y Gastric Bypass: a Systematic Review. Obesity Surgery 28(7): 1916–1923 [PubMed: 29318504]	- Population not relevant to this review protocol
Malouf, R and Areosa, Sastre A (2003) Vitamin B12 for cognition. Cochrane Database of Systematic Reviews [PubMed: 12918012]	- Population not relevant to this review protocol
Markun, S., Gravestock, I., Jager, L. et al. (2021) Effects of Vitamin B12 Supplementation on Cognitive Function, Depressive Symptoms, and Fatigue: A Systematic Review, Meta-Analysis, and Meta-Regression. Nutrients 13(3): 12 [PMC free article: PMC8000524] [PubMed: 33809274]	- Population not relevant to this review protocol
Mearns, G. J., Koziol-McLain, J., Obolonkin, V. et al. (2014) Preventing vitamin B12 deficiency in South Asian women of childbearing age: a randomised controlled trial comparing an oral vitamin B12 supplement with B12 dietary advice. European Journal of Clinical Nutrition 68(8): 870–5 [PubMed: 24736677]	- Population not relevant to this review protocol
Miles, L. M., Allen, E., Clarke, R. et al. (2017) Impact of baseline vitamin B12 status on the effect of vitamin B12 supplementation on neurologic function in older people: secondary analysis of data from the OPEN randomised controlled trial. European journal of clinical nutrition 71(10): 1166–1172 [PubMed: 28225050]	- Secondary publication of an included study that does not provide any additional relevant information
Mulder, H. and Snelder, H. A. A. (1997) Vitamin B12 replacement and its effects on bone mass and bone markers in patients with osteoporosis associated with pernicious anaemia. Clinical Drug Investigation 14(5): 434–437	- Comparator in study does not match that specified in this review protocol
Nagpal, J., Mathur, M. R., Rawat, S. et al. (2020) Efficacy of maternal B12 supplementation in vegetarian women for improving infant neurodevelopment: protocol for the MATCOBIND multicentre, double-blind, randomised controlled trial. BMJ Open 10(5): e034987 [PMC free article: PMC7252986] [PubMed: 32457078]	- study protocol
Namikawa, Tsutomu, Maeda, Masahiro, Yokota, Keiichiro et al. (2021) Enteral Vitamin B12 Supplementation Is Effective for Improving Anemia in Patients Who Underwent Total Gastrectomy. Oncology 99(4): 225–233 [PubMed: 33601391]	- Study design not relevant to this review protocol
Orhan Kilic, B., Kilic, S., Sahin Eroglu, E. et al. (2021) Sublingual methylcobalamin treatment is as effective as intramuscular and peroral cyanocobalamin in children age 0-3 years. Hematology (United Kingdom) 26(1): 1013–1017 [PubMed: 34871525]	- Study does not contain an intervention relevant to this review protocol
Ozen, S.; Ozer, M. A.; Akdemir, M. O. (2017) Vitamin B12 deficiency evaluation and treatment in severe dry eye disease with neuropathic ocular pain. Graefes Archive for Clinical & Experimental Ophthalmology 255(6): 1173–1177 [PubMed: 28299439]	- Population not relevant to this review protocol
Parry-Strong, A., Langdana, F., Haeusler, S. et al. (2016) Sublingual vitamin B12 compared to intramuscular injection in patients with type 2 diabetes treated with metformin: a randomised trial. New Zealand Medical Journal 129(1436): 67–75 [PubMed: 27355231]	- Study does not contain an intervention relevant to this review protocol
Rakshitha; Rao, G. M.; Saritha Kamath, U. (2020) Effect of vitamin b12 supplementation on neurologic and cognitive functions in older peoplea review. Biomedicine (India) 40(3): 264–266	- Systematic review used as source of primary studies
Ramos, Rafael Jacques, Mottin, Cláudio Corá, Alves, Leticia Biscaino et al. (2021) Vitamin B12 supplementation orally and intramuscularly in people with obesity undergoing gastric bypass. Obesity Research & Clinical Practice 15(2): 177–179 [PubMed: 33622624]	- Study design not relevant to this review protocol
Rhode, B. M., Tamim, H., Gilfix, B. M. et al. (1995) Treatment of vitamin B12 deficiency after gastric surgery for severe obesity. Obesity Surgery 5(2): 154–158 [PubMed: 10733805]	- Study design not relevant to this review protocol
Sanz-Cuesta, T., Escortell-Mayor, E., Cura-Gonzalez, I. et al. (2020) Oral versus intramuscular administration of vitamin B12 for vitamin B12 deficiency in primary care: a pragmatic, randomised, non-inferiority clinical trial (OB12). BMJ Open 10(8): e033687 [PMC free article: PMC7440823] [PubMed: 32819927]	- Study does not contain an intervention relevant to this review protocol
Sanz-Cuesta, T., Gonzalez-Escobar, P., Riesgo-Fuertes, R. et al. (2012) Oral versus intramuscular administration of vitamin B12 for the treatment of patients with vitamin B12 deficiency: a pragmatic, randomised, multicentre, non-inferiority clinical trial undertaken in the primary healthcare setting (Project OB12). BMC Public Health 12: 394 [PMC free article: PMC3403849] [PubMed: 22650964]	- study protocol
Saraswathy, A. R., Dutta, A., Simon, E. G. et al. (2012) Randomized open label trial comparing efficacy of oral versus intramuscular vitamin b12 supplementation for treatment of vitamin b12 deficiency. Gastroenterology 1(5supplement1): 216	- Conference abstract
Schijns, W., Homan, J., van der Meer, L. et al. (2018) Efficacy of oral compared with intramuscular vitamin B-12 supplementation after Roux-en-Y gastric bypass: a randomized controlled trial. American Journal of Clinical Nutrition 108(1): 6–12 [PubMed: 29931179]	- Study does not contain an intervention relevant to this review protocol
Schwartz, M.; Lous, P.; Meulengracht, E. (1958) Absorption of vitamin B12 in pernicious anaemia; defective absorption induced by prolonged oral treatment. Lancet 2(7058): 1200–1204 [PubMed: 13612163]	- Study design not relevant to this review protocol
Seal, E. C., Metz, J., Flicker, L. et al. (2002) A randomized, double-blind, placebo-controlled study of oral vitamin B12 supplementation in older patients with subnormal or borderline serum vitamin B12 concentrations. Journal of the American Geriatrics Society 50(1): 146–51 [PubMed: 12028259]	- Study does not contain an intervention relevant to this review protocol
Seynabou, F., Fatou Samba Diago, N., Oulimata Diop, D. et al. (2016) Biermer anemia: Hematologic characteristics of 66 patients in a Clinical Hematology Unit at Senegal. Medecine et Sante Tropicales 26(4): 402–407 [PubMed: 28073728]	- Population not relevant to this review protocol Not all participants were B12 deficient
Sezer, R. G., Akoglu, H. A., Bozaykut, A. et al. (2018) Comparison of the efficacy of parenteral and oral treatment for nutritional vitamin B12 deficiency in children. Hematology (Amsterdam, Netherlands): 1–5 [PubMed: 29577819]	- Study design not relevant to this review protocol
Shahab-Ferdows, S., Anaya-Loyola, M. A., Vergara-Castaneda, H. et al. (2012) Vitamin B-12 supplementation of rural Mexican women changes biochemical vitamin B-12 status indicators but does not affect hematology or a bone turnover marker. Journal of Nutrition 142(10): 1881–7 [PubMed: 22915298]	- Population not relevant to this review protocol
Sil, A., Kumar, H., Mondal, R. D. et al. (2018) A randomized, open labeled study comparing the serum levels of cobalamin after three doses of 500 mcg vs. a single dose methylcobalamin of 1500 mcg in patients with peripheral neuropathy. The Korean journal of pain 31(3): 183–190 [PMC free article: PMC6037815] [PubMed: 30013732]	- Population not relevant to this review protocol
Singh, C., Kawatra, R., Gupta, J. et al. (2016) Therapeutic role of Vitamin B12 in patients of chronic tinnitus: A pilot study. Noise & Health 18(81): 93–7 [PMC free article: PMC4918681] [PubMed: 26960786]	- Population not relevant to this review protocol
Smelt, H. J.; Pouwels, S.; Smulders, J. F. (2017) Different Supplementation Regimes to Treat Perioperative Vitamin B12 Deficiencies in Bariatric Surgery: a Systematic Review. Obesity Surgery 27(1): 254–262 [PubMed: 27838841]	- Systematic review used as source of primary studies
Smelt, H. J., Smulders, J. F., Said, M. et al. (2016) Improving Bariatric Patient Aftercare Outcome by Improved Detection of a Functional Vitamin B12 Deficiency. Obesity Surgery 26(7): 1500–4 [PubMed: 26530713]	- Population not relevant to this review protocol
Solomon, L. R. (2016) Vitamin B12-responsive neuropathies: A case series. Nutritional Neuroscience 19(4): 162–8 [PubMed: 25710280]	- Study design not relevant to this review protocol No comparison between B12 administration routes
Srinivasan, K., Thomas, S., Anand, S. et al. (2020) Vitamin B-12 Supplementation during Pregnancy and Early Lactation Does Not Affect Neurophysiologic Outcomes in Children Aged 6 Years. Journal of Nutrition 150(7): 1951–1957 [PMC free article: PMC7330478] [PubMed: 32470975]	- Population not relevant to this review protocol
Srinivasan, K., Thomas, T., Kapanee, A. R. et al. (2017) Effects of maternal vitamin B12 supplementation on early infant neurocognitive outcomes: a randomized controlled clinical trial. Maternal & Child Nutrition 13(2): 04 [PMC free article: PMC6090548] [PubMed: 27356547]	- Population not relevant to this review protocol
Stabler, S. P., Allen, R. H., Dolce, E. T. et al. (2006) Elevated serum S-adenosylhomocysteine in cobalamin-deficient elderly and response to treatment. American Journal of Clinical Nutrition 84(6): 1422–9 [PubMed: 17158426]	- Study design not relevant to this review protocol
Strand, T. A., Ulak, M., Chandyo, R. K. et al. (2017) The effect of vitamin B12 supplementation in Nepalese infants on growth and development: study protocol for a randomized controlled trial. Trials 18(1nopagination) [PMC free article: PMC5399862] [PubMed: 28431557]	- study protocol
Syed, E. U.; Wasay, M.; Awan, S. (2013) Vitamin B12 supplementation in treating major depressive disorder: a randomized controlled trial. The Open Neurology Journal 7: 44–8 [PMC free article: PMC3856388] [PubMed: 24339839]	- Population not relevant to this review protocol
Sánchez, H., Albala, C., Lera, L. et al. (2011) Comparison of two modes of vitamin B12 supplementation on neuroconduction and cognitive function among older people living in Santiago, Chile: a cluster randomized controlled trial. a study protocol. Nutrition journal 10: 100 [PMC free article: PMC3195703] [PubMed: 21952034]	- study protocol
Tayebi, A., Biniaz, V., Savari, S. et al. (2016) Effect of Vitamin B12 supplementation on serum homocysteine in patients undergoing hemodialysis: A randomized controlled trial. Saudi Journal of Kidney Diseases & Transplantation 27(2): 256–62 [PubMed: 26997378]	- Study does not contain an intervention relevant to this review protocol
Thompson, R. B.; Ashby, D. W.; Armstrong, E. (1962) Long-term trial of oral vitamin B12 in pernicious anaemia. Lancet 2(7256): 577–9 [PubMed: 13920928]	- Study design not relevant to this review protocol
Torsvik, I. K., Ueland, P. M., Markestad, T. et al. (2015) Motor development related to duration of exclusive breastfeeding, B vitamin status and B12 supplementation in infants with a birth weight between 2000-3000 g, results from a randomized intervention trial. BMC Pediatrics 15: 218 [PMC free article: PMC4683944] [PubMed: 26678525]	- Population not relevant to this review protocol
Torsvik, I. K., Ueland, P. M., Markestad, T. et al. (2015) Motor development related to duration of exclusive breastfeeding, B vitamin status and B12 supplementation in infants with a birth weight between 2000-3000 g, results from a randomized intervention trials. BMC Pediatrics 15(1) [PMC free article: PMC4683944] [PubMed: 26678525]	- Duplicate reference
Torsvik, I., Ueland, P. M., Markestad, T. et al. (2013) Cobalamin supplementation improves motor development and regurgitations in infants: results from a randomized intervention study. American Journal of Clinical Nutrition 98(5): 1233–40 [PubMed: 24025626]	- Population not relevant to this review protocol
van Asselt, D. Z., Pasman, J. W., van Lier, H. J. et al. (2001) Cobalamin supplementation improves cognitive and cerebral function in older, cobalamin-deficient persons. Journals of Gerontology Series A-Biological Sciences & Medical Sciences 56(12): M775–9 [PubMed: 11723153]	- Study design not relevant to this review protocol
van Dyck, C. H., Lyness, J. M., Rohrbaugh, R. M. et al. (2009) Cognitive and psychiatric effects of vitamin B12 replacement in dementia with low serum B12 levels: a nursing home study. International Psychogeriatrics 21(1): 138–47 [PMC free article: PMC3523212] [PubMed: 18925978]	- Study design not relevant to this review protocol
Vidal-Alaball, J., Butler, C. C., Cannings-John, R. et al. (2005) Oral vitamin B12 versus intramuscular vitamin B12 for vitamin B12 deficiency. Cochrane Database of Systematic Reviews: cd004655 [PMC free article: PMC5112015] [PubMed: 16034940]	- Systematic review used as source of primary studies
Wang, H, Li, L, Qin, Ll et al. (2018) Oral vitamin B12 versus intramuscular vitamin B12 for vitamin B12 deficiency. Cochrane Database of Systematic Reviews [PMC free article: PMC6494183] [PubMed: 29543316]	- Population not relevant to this review protocol
Williams, J. A.; Baume, P. E.; Meynell, M. J. (1966) Partial gastrectomy: the value of permanent vitamin-B-12 therapy. Lancet 1(7433): 342–4 [PubMed: 4159478]	- Study design not relevant to this review protocol
Withey, J. L.; Jones, J. H.; Kilpatrick, G. S. (1963) Long-term trial of oral treatment of pernicious anaemia with vitamin-B12-peptide. British Medical Journal 1(5345): 1583–5 [PMC free article: PMC2124254] [PubMed: 14001379]	- Study design not relevant to this review protocol
Yajnik, C. S., Behere, R. V., Bhat, D. S. et al. (2019) A physiological dose of oral Vitamin B-12 improves hematological, biochemical-metabolic indices and peripheral nerve function in B-12 deficient Indian adolescent women. PLoS ONE 14(10) [PMC free article: PMC6786546] [PubMed: 31600243]	- Comparator in study does not match that specified in this review protocol
Yajnik, C. S., Lubree, H. G., Thuse, N. V. et al. (2007) Oral vitamin B12 supplementation reduces plasma total homocysteine concentration in women in India. Asia Pacific Journal of Clinical Nutrition 16(1): 103–9 [PubMed: 17215186]	- Population not relevant to this review protocol
Zaqqa, M. Q., Ismail, Y. A., Khatib, A. M. et al. (2005) Effect of oral mecobalamin treatment on chest pain in patients with cobalamin deficiency and no evidence of coronary artery disease. A randomized, placebo-controlled trial. Saudi Medical Journal 26(7): 1144–5 [PubMed: 16047075]	- Study does not contain an intervention relevant to this review protocol
Zec, M., Roje, D., Matovinovic, M. et al. (2020) Vitamin B12 Supplementation in Addition to Folic Acid and Iron Improves Hematological and Biochemical Markers in Pregnancy: A Randomized Controlled Trial. Journal of Medicinal Food 23(10): 1054–1059 [PubMed: 32302504]	- Population not relevant to this review protocol

Table 28Studies excluded from the health economic review

Reference	Reason for exclusion
Masucci L, Goeree R. (2013) Vitamin B12 intramuscular injections versus oral supplements: a budget impact analysis. Ontario Health Technology Assessment Series. Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3874775/pdf [PMC free article: PMC3874775] [PubMed: 24379898]	Excluded as rated very serious limitations due to omitting the cost of oral B12 treatment whilst only reporting the cost of parenteral treatment only. Also partially applicable, reasons include: Canadian setting may not reflect current NHS context.