OBJECTIVE: To identify factors predicting suboptimal glycaemic control in rural adults during the initial five years post-type 2 diabetes diagnosis. DESIGN: Retrospective medical record audit. Quantitative study. SETTING: Rural community-based primary health service, South Gippsland, Victoria, Australia. PARTICIPANTS: Two hundred and seventy-two de-identified medical records randomly selected from the type 2 diabetes outpatient database. MAIN OUTCOME MEASURES: Demographic, biochemical, anthropometric, pharmacological, co-morbidity and lifestyle data during the first five years post-diabetes diagnosis were retrospectively collected. Univariate analysis was performed to identify variables associated with poor diabetes control (HbA1c ≥ 7%). RESULTS: Independent predictors of poor glycaemic control in this rural cohort were elevated fasting glucose at diagnosis (odds ratio (OR) 1.97, 95% confidence interval (CI) 1.31-2.97, P < 0.001), weight gain during the initial 2.5 years of diabetes (OR 1.33, 95% CI 1.11-1.59, P < 0.01), excessive body weight at diagnosis (OR 1.07, 95% CI 1.03-1.12, P < 0.001) and younger age at diagnosis (OR 0.94, 95% CI 0.88-1.00, P < 0.05). These variables combined explained 48% of the variation in HbA1c. Gender, body mass index, waist circumference and lifestyle factors at diagnosis were not significant predictors of diabetes control. CONCLUSIONS: Young-middle-aged adults (≤58 years) with elevated fasting glucose (≥9.0 mmol L(-1) ) and excessive body weight (≥93.1 kg) at type 2 diabetes diagnosis and those unable to lose weight early in the course of the disease are more likely to experience a rapid deterioration in glucose control. Rural clinicians should target these individuals for aggressive diabetes management from the time of diagnosis.