Chronic psychiatric status and satisfaction with life

Abstract

The present study represents the first to administer the Satisfaction With Life Scale (SWLS) as part of a semi-structured interview to a large sample of psychiatric patients with severe mental illness. Psychometric appraisal of the SWLS demonstrated that figures on its internal structure were quite favourable. SWLS scores were shown to be predictably associated with intimate relationship status (with patients involved having higher scores than those not involved), age (+), education (−), positive (+) and negative (−) affect and self-esteem. The SWLS was clearly more strongly associated with another general measure of satisfaction with life (convergent validity) than with measures of affect, self-esteem, symptoms, invalidity or demographic background (divergent validity). Evidence in favour of the discriminatory power of the SWLS was found in the significantly lower scores of the psychiatric patients as compared to those of non-psychiatric controls. The SWLS represents a potentially valuable clinical assessment and research tool with samples comprising patients with severe, chronic mental illness.

Introduction

The increase in interest in the quality of life of the mentally ill has been attributed to at least two concerns. First, it reflects the rise of a more consumer-oriented approach to medical care, in which the patients’ own opinion of what is happening to them is taken as important, rather than patients being the objects of expert attention from professionals who themselves judge the effectiveness and relevance of what they do. Second, it expresses a more serious concern for the broad definition of health advanced by the World Health Organization, namely as a state of complete physical, mental and social well-being and not merely the absence of disease (cf. Orley, Saxena, & Herrman, 1998).

Well-being is highly important as its absence has been advanced to reflect a state that creates conditions of vulnerability to possible future adversities (Berenbaum, Raghavan, Le, Vernon & Gomez, 1999). In addition, the route to recovery has been hypothesized to lie not only in the reduction of symptoms/distress (negative affect), but also in the generation of positive affect and well-being (Parloff, Kelman & Frank, 1954, Ryff, 1989, Ryff & Singer, 1996, Fava et al., 1998).

Orley et al. (1998) pointed out that without definitions and measures of well-being, however, there can be little progress towards its inclusion as an objective in the creation of a more healthy society (UK700 Group, 1999), and well-being measures provide one step towards such a goal (e.g. Candilis et al., 1999, Evenson & Vieweg, 1998, Koran, Thienemann & Davenport, 1996, van Nieuwenhuizen et al., 1997, Rosenfield, 1992, Rössler et al., 1999, Sainfort, Becker & Diamond, 1996).

SWB may be defined as a broad category of phenomena that includes people’s emotional responses, positive (e.g. joy, elation, contentment, pride) and negative (e.g. anxiety, depression), domain satisfactions (e.g. family, work, leisure, health, finances, self, one’s in-group) and overall cognitive evaluations of life satisfaction (Diener, Suh, Lucas & Smith, 1999). In general, the life satisfaction component has received far less attention than the affective components (Diener et al., 1985, Pavot & Diener, 1993), a situation which should be remedied.

The Satisfaction With Life Scale (SWLS; Diener et al., 1985) is a brief, multi-item scale for the measurement of life satisfaction as a cognitive-judgemental process. The SWLS was designed around the idea that, rather than summing across clients’ satisfactions with specific domains, one must ask them for an overall judgement of their life in order to have an adequate index of the concept of life satisfaction. In their review of the SWLS, Pavot and Diener (1993) concluded on the basis of data collected in several different samples, among other things, that it reflects a one-dimensional, internally consistent measure. Moreover, the SWLS has shown sufficient sensitivity to be potentially valuable to detect change in life satisfaction during the course of clinical intervention. In addition, it has also shown good convergent validity with theoretically-related scales and with other types of assessment of SWB, and discriminant validity in relation to measures of negative affective states.

Unfortunately, however, despite the potential of the SWLS as a valuable clinical assessment and research tool in psychiatric samples, hardly any published studies have addressed its psychometric properties in such samples. This is peculiar considering the remarkably profound impact that emotional disturbance and a psychiatric state can have on the individual, including adverse effects of stigma on well-being (Link et al., 1997, Markowitz, 1998), homelessness and many negative financial, role-functioning, interpersonal, health, and life-span consequences (cf. Berenbaum et al., 1999). To fill this gap, the present study was aimed at examining the psychometric qualities of the SWLS in a large sample of testable patients with chronic severe mental illness (n=487).

Specifically, the aims of the present investigation were fivefold. First, it sought to determine the internal structure of the SWLS. Second, it attempted to assess the internal consistency of the SWLS and the homogeneity and reliability of its items. Third, it addressed the convergent validity of the SWLS in relation to theoretically related measures and, in a subsample of the patients, its divergent validity in relation to symptom and fuctional status dimensions. Fourth, it tested some demographic correlates of the SWLS. The demographic variables included sex, age, educational level and intimate relationship status.

The association between the SWLS and intimacy is of special interest. The positive relation between marriage and SWB has been consistently replicated in a multitude of studies, even when variables such as sex, age and income were controlled. Based on a meta-analysis, the average correlation between marital status and SWB was small, namely 0.14 (Diener et al., 1999). In psychiatric samples, however, this positive relationship may be counteracted by two facts: (1) significant assortative mating (i.e. marital resemblance) has been demonstrated for both well-being, psychological distress (Galbaud du Fort, Kovess, & Boivin, 1994) and major psychiatric disorders (Maes et al., 1998); and (2) in the young adult population, some forms of psychopathology, especially depression, are significantly negatively correlated with both spouses’ marital adjustment scores (cf. Gagnon, Hersen, Kabacoff, & van Hasselt, 1999). From (1) and (2) would follow that the disorders of both partners, through the stress that is inflicted on the marital dyad, are likely to reduce the pay off of the emotional dividends that marital intimacy, commitment and support provide for most people in the community (Argyle, 1999, Myers, 1999). Fifth, the present study was aimed at presenting descriptive and specialized norms for psychiatric patients. In order to determine the discriminatory power of the SWLS, the descriptive norms were also contrasted with equivalent data previously collected in different non-psychiatric samples from the Dutch population at large.