Purpose
Pancreatic cancer is a highly aggressive tumor primarily affecting older adults. Major risk factors include smoking and alcohol consumption. To investigate other potential contributors to pancreatic cancer risk, we studied postmenopausal women in the Women’s Health Initiative (WHI) who were never smokers and non-current drinkers.
Sample and setting
The present study included 40,205 women (aged 50–79 years at enrolment) who at baseline were never-smokers and non-current drinkers. Information on reproductive, health, and lifestyle factors was obtained at baseline. Diabetes and cardiovascular disease were reported at baseline and during the follow-up.
Procedures
Multivariable-adjusted Cox regression was used to estimate hazard ratios (HR) and 95% confidence intervals (CI) for pancreatic cancer risk.
Results
Over an average follow-up period of approximately 16.6 years, 336 incident cases of pancreatic cancer were ascertained. High body mass index [HR≥35vs ≤20kg/m2=2.33 (1.05-5.18)], large height-adjusted waist circumference [HR≥88vs<80cm=1.51 (1.16-1.97)], diabetes [HR=1.53 (1.18-1.98), and cardiovascular disease [HR=1.28 (1.01-1.63)] were associated with increased risk of pancreatic cancer. Higher quality diet [HRstandard-deviation-unit=0.89 (0.80-0.99)] and higher consumption of fruit and vegetables [HR ≥7vs ≤1.5serving/day=0.51 (0.30-0.89)] and of fiber [HR≥20vs<10g/day=0.58 (0.38-0.88)] were associated with reduced risk, while higher intake of red meat was associated with an increased risk [HR≥0.5vs<0.5serving/day=1.31 (1.03-1.68)]. No associations were observed for physical activity or reproductive/hormonal factors.
Conclusions and implications
This is the first prospective study to focus on identifying risk factors for pancreatic cancer among postmenopausal women who have neither smoked nor drunk alcohol. We found that excessive adiposity, diabetes, and several dietary factors were associated with altered pancreatic cancer risk. These results, while potentially pointing to preventive approaches, require replication.