Assessing the construct validity of three indicators of psychological distress in relation to perceived health and physical illness

Objectives The aim of this study is to compare three indicators of psychological distress (PD) on the strength of their association with subjective (or perceived) health and to analyse to what extent these associations will change after adjusting for physical illness measures and other possible confounding variables. Methods Data were used from a communitybased sample of adults (N=9,428). Psychological distress was measured using three different instruments: the Negative Affect Scale of Bradburn, a nervousness scale, and a self-reported depressive complaints. Physical illness was measured by seven specific chronic conditions, a co-morbidity index of 17 conditions and two disability measures. Subjective health was assessed by a single question. Ordinary least square and logistic regression as well as structural equation modelling were used to analyse the data. Results The relation between subjective health and PD is strongest in case nervousness and this, or negative affect, are used as indicators of PD. The measure of depressive complaints is less strongly, but still substantially, related to subjective health. After correction for physical illness variables, the change in strength of the association is slightest for depressive complaints and highest for nervousness. Only small differences between negative affect and nervousness were established. These measures, which were more contaminated by physical ill health than depressive complaints, have the strongest association with subjective health both before as well as after correction for physical illness components. Conclusion Negative affect and nervousness are reliable and valid indicators of PD, which can be used to predict subjective health. However, for this purpose, a correction for the confounding effects of physical illness variables will be necessary. The depressive complaints measure is not only less predictive of subjective health but also less contaminated by physical illness variables, making it a better indicator of PD if correction for physical illness variables is not possible.