Single item on positive affect is associated with 1-year survival in consecutive medical inpatients

Abstract

Objective

To determine the independent effects of positive and negative affect items on mortality in consecutive medical inpatients.

Methods

Consecutive general medical inpatients were asked to complete the Hospital Anxiety and Depression Scale (HADS) at admission. Prognostic indicators were obtained from patients' records and physicians' ratings. The study end point was mortality from all causes at 1 year.

Results

The baseline assessment was completed by 575 patients (87.7%). Survival data were available for 572 of these (86 deaths). HADS depression scores and several physical risk indicators predicted mortality. Independent effects could be observed for HADS item 1 (“Can enjoy things as much as before”) adjusted for physicians' ratings of prognosis, a principal diagnosis of hemato-oncological disease and Charlson comorbidity scores. In contrast, HADS depression items 2–7 (Model 1) as well as positive HADS depression scores did not contribute significantly to the prediction of mortality.

Conclusion

Our present results suggest that one single item on positive affect independently predicts 1-year survival in consecutively admitted medical inpatients. Interestingly, this item has a stronger association with survival status than the presence of depressed mood.

Introduction

Depressive mood has repeatedly been identified as a predictor of mortality in a variety of medical patient groups. In an earlier study, we could confirm an independent effect of depressed mood on mortality in a typical mixed patient population seen on general medical wards [1]. Like most other research in this area, our previous study focused on negative affect as a predictor of mortality, using an established measure of depressive mood, such as the Hospital Anxiety and Depression Scale (HADS) [2].

Depressive mood, however, is only part of the emotional dimensions of well-being. Measures for depressive mood, such as the HADS, include items that reflect negative affect (e.g., “I feel slowed down”) as well as items that reflect positive affect (e.g., “I can enjoy things”) that are then coded in reverse order and added to negative affect items in order to calculate the depression scale. However, psychological literature suggests that positive and negative affect are not simply bipolar opposites [3]. High positive affect may be understood as a state of high energy, full concentration and pleasurable engagement, whereas low negative affect may be seen as a state of calmness and serenity [4]. Therefore, positive and negative affect should be treated as relatively independent constructs.

A growing body of evidence indicates that positive affect has a stronger association with clinically relevant end points including premature mortality and adverse cardiovascular outcomes than negative affect [5], [6], [7], [8], [9], [10]. For example, Ostir et al. [11] demonstrated that the positive items on the Center for Epidemiologic Studies Depression Scales (CES-D) but not the negative affect items were predictive of better mobility and functional status and lower risk of mortality in a population-based sample of older participants. However, authors of a comprehensive review suggested that there is too little evidence regarding positive affect and long-term prognosis across the few survival studies to draw any conclusions [12]. Especially in populations with heterogeneous principal diagnoses the impact of positive affect on survival remains yet unclear.

Given the distinct properties of positive and negative affect, it is important to distinguish their potential influences on health outcomes and especially on mortality [4]. We therefore re-analysed the data of our previous study cited above and examined the question of how and to what extent the single items of the HADS reflecting positive or negative affect independently predict mortality in consecutively admitted medical inpatients.