Protein intake and incident frailty in the Women's Health Initiative observational study

Abstract

Objectives: To evaluate the association between protein intake and incident frailty.

Design: Prospective cohort study.

Setting: Subset of the Women's Health Initiative Observational Study conducted at 40 clinical centers.

Participants: Twenty-four thousand four hundred seventeen women aged 65 to 79 who were free of frailty at baseline with plausible self-reported energy intakes (600-5,000 kcal/day) according to the Food Frequency Questionnaire (FFQ).

Measurements: Baseline protein intake was estimated from the FFQ. Calibrated estimates of energy and protein intake were corrected for measurement error using regression calibration equations estimated from objective measures of total energy expenditure (doubly labeled water) and dietary protein (24-hour urinary nitrogen). After 3 years of follow-up, frailty was defined as having at least three of the following components: low physical function (measured using the Rand-36 questionnaire), exhaustion, low physical activity, and unintended weight loss. Multinomial logistic regression models estimated associations for uncalibrated and calibrated protein intake.

Results: Of the 24,417 eligible women, 3,298 (13.5%) developed frailty over 3 years. After adjustment for confounders, a 20% increase in uncalibrated protein intake (%kcal) was associated with a 12% (95% confidence interval (CI)=8-16%) lower risk of frailty, and a 20% increase in calibrated protein intake was associated with a 32% (95% CI=23-50%) lower risk of frailty.

Conclusion: Higher protein consumption, as a fraction of energy, is associated with a strong, independent, dose-responsive lower risk of incident frailty in older women. Using uncalibrated measures underestimated the strength of the association. Incorporating more protein into the diet may be an intervention target for frailty prevention.

Figure 1

Risk* of frailty compared to lowest quintile (Q) of protein intake (%kcal): uncalibrated versus calibrated protein**. *Adjusted for age, ethnicity, body mass index, income, education, having a current health care provider, smoking, alcohol, general health status, history of comorbid conditions (hip fracture, emphysema, treated diabetes, hypertension, arthritis, cancer), history of hormone therapy use, number of falls, whether participant lives alone, disabled defined by at least 1 activity of daily living affected, depressive symptoms, log-transformed energy intake (uncalibrated energy for uncalibrated protein models and calibrated energy for calibrated protein models) **Uncalibrated values represent food frequency questionnaire (FFQ) estimates and calibrated values represent estimates derived from linear regression equations developed on the basis of FFQ nutrient measures and participant characteristics.