The relationship between subjective and objective measures of health: A series of systematic reviews and analyses of data from national surveys

Abstract

The goal of this research was to consolidate and expand on current knowledge about the use of objective and subjective measures to assess the health of Canadians. Specifically, this dissertation examined the relationship between self-report and direct measures of health to determine the extent of discrepancies between the two methods. This was accomplished through a series of five research papers, including three systematic reviews that examined the relationship between self-reported and directly measured estimates of height, weight and Body Mass Index (BMI), hypertension and smoking. An in-depth examination of the discrepancies in reported obesity followed. It included two additional papers, one examining the feasibility of establishing correction factors to increase the accuracy of reported estimates of obesity and a second examining the trends in the reporting bias over time and between two different population groups (Canada and the U.S.). Results clearly documented a bias in self-reported estimates of height, weight, BMI, hypertension and smoking; reported estimates of height were consistently overestimated while reported weight, BMI, hypertension and smoking were consistently underestimated in both men and women. Statistical adjustment of reported values generated estimates of height, weight and obesity that were more accurate than those generated by self-report alone. Historical analysis with Canadian and American datasets indicated that the bias in reported obesity is increasing, implying that the generalizability of the correction equations may be limited and that in Canada, monitoring trends in obesity based solely on self-reported height and weight may produce inaccurate estimates because of the diverging discrepancy between self-reported and measured data. This work has made an important contribution to understanding the bias in current surveillance practices that are grounded in the collection of self-reported data and presents a series of recommendations to improve population health surveillance in Canada.