Insights from a general practice service evaluation supporting a lower carbohydrate diet in patients with type 2 diabetes mellitus and prediabetes: a secondary analysis of routine clinic data including HbA1c, weight and prescribing over 6 years.
Level 4 - case-series / case-control
Uncontrolled retrospective cohort / clinical service evaluation of routine practice data
PubMed 33521540 · doi:10.1136/bmjnph-2020-000072
What was done
A retrospective service evaluation analyzed routine clinical audit data collected between 2013 and 2019 in a single general practice in England (Norwood GP surgery, list size 9,800). Advice on following a lower-carbohydrate diet—focusing on reducing sugar, carbohydrates, and high-glycaemic-index foods—was routinely offered to patients with newly diagnosed or existing type 2 diabetes (T2D) and prediabetes using individual and group consultations. Interested patients were tracked to compare baseline with latest follow-up measures over a mean duration of 23 months.
What was found
A total of 128 patients with T2D (27% of the practice T2D register) and 71 patients with prediabetes chose to follow the diet. For patients with T2D, median weight decreased from 99.7 kg (IQR 86.2–109.3) to 91.4 kg (IQR 79.0–101.1, p < 0.001), and median HbA1c decreased from 65.5 mmol/mol (IQR 55–82) to 48.0 mmol/mol (IQR 43–55, p < 0.001). Drug-free T2D remission occurred in 46% of participants. For patients with prediabetes, median HbA1c fell from 44 mmol/mol (IQR 43–45) to 39 mmol/mol (IQR 38–41, p < 0.001), with 93% attaining a normal HbA1c. Practice diabetes drug prescribing expenditure was £50,885/year less than the regional average.
Why it matters
This evaluation demonstrates that lower-carbohydrate dietary advice can be successfully implemented in real-world primary care, with substantial improvements in glycemic control and reductions in medication costs among engaged patients.
Limits
The study is an uncontrolled, single-centre service evaluation without a parallel comparator group, preventing causal inference. There is substantial self-selection bias because only 27% of patients with T2D chose to adopt the diet. Dietary adherence, exact macronutrient intakes, and lifestyle confounders were not strictly controlled or measured.
Cited by
- supports A 150-gram serving of boiled white rice has a glycemic load equivalent to 10 four-gram teaspoons of sugar.
- context Dr. David Unwin has achieved drug-free type 2 diabetes remission in 157 patients in his clinical practice.