To determine the prospective associations among cardiorespiratory fitness (CRF), different measures of adiposity, and all-cause mortality in women.
A total of 11,335 women completed a comprehensive baseline examination between 1970 and 2005. Clinical measures included body mass index (BMI), waist circumference (WC), waist-to-height ratio (W/HT), waist-to-hip ratio (W/Hip), percent body fat (%BF), and CRF quantified as duration of a maximal exercise test. Participants were classified by CRF as low (lowest 20%), moderate (middle 40%), and high (highest 40%) as well as by standard clinical cut points for adiposity measures. Hazard ratios (HR) were computed using Cox regression analysis.
During a mean follow-up of 12.3 ± 8.2 yr, 292 deaths occurred. HR for all-cause mortality were 1.0, 0.60, and 0.54 for low, moderate, and high fit groups, respectively (P for trend <0.01). Adjusted death rates of overweight/obese women within each adiposity exposure were somewhat higher compared with normal-weight women and approached statistical significance for BMI, %BF, and W/HT (P = 0.08, P = 0.08, and P = 0.07, respectively). When grouped for joint analyses into categories of fit and unfit (upper 80% and lower 20% of CRF distribution, respectively), HR were significantly higher in unfit women within each stratum of BMI compared with fit-normal BMI women. Fit women with high %BF (HR = 1.0), high WC (HR = 0.9), and high W/HT (HR = 1.2) had no greater risk of death compared with fit-normal-weight women (referent).
Low CRF in women was a significant independent predictor of all-cause mortality. Higher CRF was associated with lower mortality within each category of each adiposity exposure. Using adiposity measures as predictors of all-cause mortality in women may be misleading unless CRF is also considered.
1The Cooper Institute, Dallas, TX; 2The Cooper Clinic, Dallas, TX; and 3Winston-Salem State University, Winston-Salem, NC
Address for correspondence: Stephen W. Farrell, Ph.D., The Cooper Institute, 12330 Preston Rd., Dallas TX 75230; E-mail: email@example.com.
Submitted for publication December 2009.
Accepted for publication March 2010.