Design: Retrospective matched cohort, difference-in-differences
Population: Polyclinic patients with diabetes, hypertension or hyperlipidaemia in Singapore
Primary outcome: Clinical parameters, screening adherence, utilisation and cost over 5 years
Effect: Foot screening +16%, eye screening +6%; net saving SGD 2186 per patient

An Annals of Family Medicine study evaluated a 'teamlet' model introduced in Singapore polyclinics in 2014, in which patients with diabetes, hypertension or hyperlipidaemia were looked after by a small team of two family physicians, a nurse and a care coordinator. Patients enrolled in 2014–2016 were compared with matched usual-care patients over five years using difference-in-differences analysis of national data.

Clinical gains were small: HbA1c fell by 0.1% in some subgroups, systolic pressure by 0.65 mm Hg and LDL by 0.03 mmol/L. But foot screening rose by 16% and eye screening by 6%. Hospital admissions, specialist visits and emergency attendances fell, polyclinic costs rose, and the net saving was about SGD 2186 per patient.

The design is observational and Singapore's integrated system differs from most Indian primary care. The underlying idea — the same small team seeing the same patients — is transferable to busy clinics, even without the infrastructure.

Continuity with a small, named team improved annual diabetes foot and eye checks.
A nurse or coordinator who tracks overdue checks can raise screening without more doctor time.
Do not expect large changes in HbA1c or blood pressure from reorganisation alone.
Fewer admissions drove most of the cost savings.

Why it matters

Continuity of care can be built into high-volume clinics by structure rather than by individual effort.

The statistics, in plain English

Difference-in-differences compares how outcomes changed over time in the programme group against how they changed in matched controls, which removes some but not all bias from comparing different groups.