Appropriateness of B12 Administration in a Real-World Population

Introduction

B12 deficiency can lead to multiple complications, including severe pancytopenia, spinal cord demyelination, and cognitive dysfunction. Randomized controlled trials have demonstrated oral B12 supplementation is as effective as parenteral B12 administration in improving hematologic abnormalities and raising B12 levels. Despite this evidence base, a high proportion of patients continue to receive intramuscular (IM) B12. We assessed the prevalence of inappropriate B12 supplementation in a large population-based cohort.

Methods

We undertook a population-based, retrospective cohort study with the Institute for Clinical and Evaluative Sciences (ICES) in Ontario, Canada. ICES houses health service records for 13 million Ontarians, including laboratory values from the Ontario Laboratories Information System (OLIS), which contains 92% of annual provincial test volumes; the Ontario Drug Benefit (ODB) database, which houses prescriptions, including IM B12, for all Ontarians aged 65 and above; and the Ontario Health Insurance Program (OHIP), which contains all billing claims for physician services (including intramuscular B12 injections).

We included all patients (≥ 65 years or above) that received an IM B12 injection between September 1, 2009 and September 30, 2015, identified from both ODB and OHIP databases. Inappropriate B12 supplementation was defined as (1) IM B12 administration following a documented normal B12 level (≥250 pmol/L); or (2) no documented B12 testing in the 12 months prior to supplementation. Appropriate IM B12 supplementation occurred in those with documented B12 deficiency (<250 pmol/L) or a diagnosis of inflammatory bowel disease (IBD) or malabsorptive disorders. Other outcomes we evaluated included patterns of B12 prescribing, frequency of B12 level retesting in patients receiving supplementation, and the overall prevalence of B12 deficiency. Logistic regression analysis was used to determine physician and patient variables that predicted inappropriate B12 supplementation.

Results

A total of 173,138 Ontarians received at least one IM B12 prescription during the study interval. Of these, 37,801 (21.8%) had a normal serum B12 level (≥ 250 pmol/L) in the preceding twelve months, and an additional 73,886 (42.7%) had no documented B12 test in the preceding 12 months, suggesting that these prescriptions were inappropriate. Only 61,451 (35.5%) of patients that received IM B12 had documented B12 deficiency (<250 pmol/L) in the preceding 12 months. Only 1.1% (n=1925) of all IM prescriptions were to patients with B12 deficiency and either IBD or a malabsorptive disorder. Each patient received an average of 4.01 prescriptions during the study period (SD: 5.77; median: 2; range:1-198) from an average of 1.29 physicians (SD: 0.72; median: 1; range:1-32). The average length of time between the initial and first repeat prescription was 309.6 days (SD: 28.21, median: 263; range: 0-2206). On average, those patients that were tested were re-tested an average of 3.2 times during the 6-year study period (SD: 2.81, median: 3; range: 1-295). The overall prevalence of B12 deficiency in all patients tested in Ontario during the study interval was 7.7% (<150 pmol/L; n=500,453) or 46.9% (<250 pmol/L; n=3,056,986).

Conclusion

The prevalence of inappropriate B12 supplementation is extensive. Nearly two-thirds (65.6%) of patients received an IM B12 injection with either a documented normal B12 level or without B12 documentation in the year prior to supplementation. In this era of Choosing Wisely, our findings suggest that initiatives are required to educate physicians and patients on appropriate indications for IM B12 prescriptions.