# Impact of the MHRA safety update on vitamin B12 testing and coding in metformin users: a retrospective primary care analysis

## TL;DR

A UK safety alert about metformin and vitamin B12 deficiency led to slightly more testing but no change in how doctors document the deficiency.

## Contribution

The study shows how a regulatory safety update influenced testing rates but not diagnostic coding practices in primary care.

## Key findings

- Metformin users had a statistically significant increase in vitamin B12 testing after the MHRA guidance.
- Diagnostic coding for vitamin B12 deficiency remained unchanged in both metformin users and the general population.
- General population testing rates also increased but to a lesser extent than in metformin users.

## Abstract

Metformin is the most commonly prescribed oral treatment for type 2 diabetes mellitus (T2DM) in the UK. Long-term therapy has been linked to vitamin B12 deficiency, a concern recognised for decades but not consistently addressed. In June 2022, the UK Medicines and Healthcare products Regulatory Agency (MHRA) classified low vitamin B12 levels as a common adverse effect of metformin and advised clinicians to consider periodic testing in at-risk patients. Translating such regulatory advice into practice can be challenging, and the extent to which the guidance has influenced testing and diagnostic coding for vitamin B12 deficiency remains unclear. This study evaluated trends in vitamin B12 testing and deficiency coding in metformin-treated patients compared with the general population before and after the 2022 MHRA Drug Safety Update.

A retrospective quantitative analysis was conducted using Read code data from 148,000 electronic medical records across three Primary Care Networks (PCNs) in the Southwest of England. Vitamin B12 testing and deficiency coding rates were compared in patients prescribed metformin and the general population across two periods: pre-guidance (2017–2021) and post-guidance (2022–2024). Welch’s t-tests were used to determine statistical significance, with p < 0.05 considered significant.

Among patients prescribed metformin, vitamin B12 testing rates rose from 34.5% (SD = 1.8) pre-guidance to 38.2% (SD = 0.4) post-guidance (p = 0.008). In the general population, testing rates increased from 12.2% to 14.7% (p = 0.009). However, coding for vitamin B12 deficiency remained unchanged at 0.25% in the metformin group and decreased from 0.072% to 0.060% in the general population, with no statistically significant difference. The post-guidance period included only two years of data, which limits the ability to assess longer-term or comparative trends between groups.

This study demonstrated that the release of the MHRA Drug Safety Update was associated with a modest but statistically significant increase in vitamin B12 testing among patients prescribed metformin, paralleled by a smaller rise in testing within the general population. However, diagnostic coding practices did not change, suggesting limited translation of safety alerts into structured documentation. Further research is warranted to explore barriers and evaluate interventions to improve monitoring and coding compliance in primary care.

The online version contains supplementary material available at 10.1186/s12875-025-03116-1.