Incident disease outcomes were assessed using the same definitions as in the EPIC-Potsdam report.¹ Similarly, we used the same 4 healthy factors and methods to form an index ranging from 0 to 4, as detailed by Ford et al.¹ However, because our physical activity measures were not available in hours per week, we included a definition of engaging in physical activity as 3 times/wk or more. The covariates adjusted for in incidence analyses included age, sex, educational status (having/not having tertiary qualifications), and occupational status (employed or unemployed).

Results
Of the 3654 participants at baseline, 2639 had information on all 4 healthy factors. As in EPIC-Potsdam, participants who had diabetes (n = 208 [7.9%]), MI (n = 219 [8.3%]), stroke (n = 84 [10.9%]), and cancer (n = 189 [7.2%]) at baseline were excluded from analyses. Of the remaining participants, 50.0% had a healthy diet score, 82.5% had a body mass index lower than 30 (calculated as weight in kilograms divided by height in meters squared), 48.2% had never smoked, and 44.1% had participated in physical activity at least 3 times/wk. After adjusting for age, sex, and educational and occupational status, the risk of developing incident diabetes (P value for linear trend, < .001) and MI (P value for linear trend, < .001) decreased as the number of healthy lifestyle factors increased (Table), but this was not observed for incident stroke. Having 1 or more healthful factors compared with 0 factors did not lower the risk of incident MI and stroke. However, having all 4 healthy lifestyle factors had the greatest impact on incident diabetes—an 83% risk reduction.

View this table: [in this window] [in a new window] [as a PowerPoint slide]   Table. Adjusted Hazard Ratios (aHRs) and 95% Confidence Intervals (CIs) for Incident Diabetes, Myocardial Infarction, and Stroke by Number of Healthy Lifestyle Factors in the Blue Mountains Eye Study

Comment
In our older population, a slightly higher proportion (11.4%) of participants than in the EPIC-Potsdam study (9.1%) adhered to all 4 healthful factors.¹ The BMES participants with all 4 healthy lifestyle factors had an 83% reduced risk of incident diabetes, and a significant linear trend was observed for incident MI, reinforcing the findings by Ford et al.¹ In contrast, having all 4 healthy lifestyle factors among the BMES participants did not significantly reduce the risk of incident stroke. As recommended by Katz et al,³ refinement of dietary quality (eg, including fish consumption, lean vs fattier meats) may have strengthened observed associations between lifestyle and health outcomes in our study. Nevertheless, our findings in a population older than those in the EPIC-Potsdam study¹ concur with its take-home message, and that from other prospective studies,^4-6 that healthful dietary and lifestyle factors are beneficial in reducing the risk of developing chronic diseases, particularly diabetes.

AUTHOR INFORMATION
Correspondence: Dr Mitchell, Centre for Vision Research, University of Sydney, Westmead Hospital, Hawkesbury Road, Westmead, New South Wales, Australia 2145 (paul_mitchell{at}wmi.usyd.edu.au).

Author Contributions: Study concept and design: Gopinath and Mitchell. Acquisition of data: Mitchell. Analysis and interpretation of data: Gopinath, Rochtchina, Flood, and Mitchell. Drafting of the manuscript: Gopinath. Critical revision of the manuscript for important intellectual content: Gopinath, Rochtchina, Flood, and Mitchell. Statistical analysis: Rochtchina. Obtained funding: Mitchell. Administrative, technical, and material support: Flood and Mitchell. Study supervision: Gopinath and Mitchell.

Financial Disclosure: None reported.