سلامت روان و مدل فرضیه استعداد استرس: نقش عملکرد روانی و کیفیت روابط در ترویج بهزیستن ذهنی در یک مطالعه وقایع زندگی

Abstract Negative life events are associated with poor wellbeing and mental health outcomes. Following a diathesis-stress model, we tested whether psychological functioning and quality of interpersonal relationships moderated the effect of life events on subjective wellbeing. This study comprised data from a young and middle-aged adult sample (n = 364) drawn from an Australian university-student population. Results indicated that life events were associated with negative but not positive wellbeing outcomes. Perceived impact of life events was a stronger predictor of wellbeing than was the number of life events. Psychological functioning and quality of interpersonal relationships were associated with both wellbeing dimensions but only quality of interpersonal relationships moderated the effect of life events on wellbeing. In conclusion, perceived impact of life events was more strongly related to wellbeing than number of life events. Interpersonal relationships moderate the effect of life events with those reporting higher levels of quality of interpersonal relationships reporting less decrement in negative affect following stressful life events.

. Introduction The association between life events and health is well established (Sarason, Sarason, Potter, & Antoni, 1985). Stressful life events are implicated in the aetiology of common mental disorders (Bebbington et al., 1988, Brown et al., 1993, Newman and Bland, 1994 and Spinhoven et al., 2011). Differences in how individuals respond and adapt to stressful life events can be accounted for by a number of psycho-social factors. In one longitudinal study (Whisman & Kwon, 1993), the impact of life stress on longitudinal change in dysphoria was moderated by self-esteem and mediated by change in hopelessness. Higher self-esteem and lower hopelessness were associated with better wellbeing outcomes. Similarly, decreased neuroticism and increased extraversion have been indicated as moderating the long-term course of depressive and anxiety symptomatology in a positive way (Spinhoven et al., 2011). Social and environmental factors can also moderate the association between stressful life events and mental health outcomes. Social support is consistently identified as buffering the effects of life events on wellbeing outcomes in clinical samples (Ames & Roitzsch, 2000) and the general population (Falcon, Todorova, & Tucker, 2009). In a recent Dutch study (van den Berg, Maas, Verheij, & Groenewegen, 2010), environment was a significant moderator of the degree to which participants were affected by stressful life events. The authors concluded that the amount of green space, within 3 km of residents’ homes, buffered against the negative health impact of stressful life events. Although one’s vulnerability to poor mental health outcomes is purported to be diathetic (Zubin & Spring, 1977), the level of risk in developing poor mental health outcomes is clearly associated with the availability of those psycho-social resources with which an individual may utilise and cope with the occurrence of negative stressful events. Given the role of individuals’ resources in moderating the effect of life events on well-being outcomes, we believe that there is a strong theoretical basis on which to focus the examination of life events on the appraised impact that an event may have. The diathesis-stress hypothesis (Coyne & Downey, 1991) proposes that personal dispositions and social context moderate the effect of stressful life events on health and well-being (see Fig. 1). When psychological and social resources which aid adjustment to life events are absent or limited, then individuals are vulnerable to an increased likelihood of reporting a decrement in a range of health outcomes. Diathesis-stress model: Psychological disposition and social context moderate ... Fig. 1. Diathesis-stress model: Psychological disposition and social context moderate the effects of stressful life events on mental and physical health. Figure options Typically, investigations of the effect of life events associate the occurrence of a life event, or the number of life events that occurred in a preceding period, with subsequent mental health outcomes. Evidence for the perceived degree of impact of life events remains relatively unexplored. That is, the association between perceived impact of life events on health and wellbeing is less clear. In a similar vein, Horowitz, Wilner, and Alvarez (1979) proposed the Impact of Event Scale (IES) as a method of describing subjective distress in relation to specific life events, determining the extent to which participants reported degrees of intrusive thinking and avoidance. However, most utilisation of the IES has been restricted to clinical samples, particularly in relation to posttraumatic stress disorder (Sundin & Horowitz, 2003). We propose that the deleterious effect of a life event is associated more with its degree of impact on one’s life than its occurrence alone. For example, the negative impact of job-loss may be less damaging on the individual who is in a financial position to deal with job-loss, or for the individual who had the foresight that job-loss was impending and had begun to take steps to find alternative employment. Similarly, for one individual, the end of a difficult acrimonious relationship may impact less negatively than for an individual whose perceived nurturing and fulfilling relationship ends unexpectedly. In this study, we amend a common measure of Significant Life Events to determine the degree of impact of a life event, such that when a life event has occurred, a participant describes the extent to which the event impacted on their life. Finally, we test the effects of stressful life events on individual wellbeing, following a model of wellbeing (Huppert et al., 2009), that combines psychological function and feeling. There is considerable evidence for the independence of related wellbeing constructs that are either affective or cognitive-behavioural in basis (Burns and Machin, 2009, Burns and Machin, 2010 and Gallagher et al., 2009), with stronger evidence for the role of psychological functioning in determining feeling components of wellbeing and mental health outcomes (Burns et al., 2011 and Burns and Machin, 2012). We posit that quality of social relations and psychological function moderate the effect of perceived impact of life event individual wellbeing. 1.1. Aims Our aims are: 1. To compare the association between number of life events and the perceived impact of life events on wellbeing; and 2. to examine whether components of psychological functioning and social relations moderate the association between perceived impact of life events and wellbeing.

. Results Bivariate correlations indicate that overall, there is considerable variance in subjective wellbeing that remains unexplained (Table 2). Of particular emphasis, correlations between subjective wellbeing and perceived impact of life events indicates that the assessment of the impact was mostly unrelated to affect. One significant issue relates to the strong correlation (r = .65) between the superordinate EGPS factor and positive affect; this consistent finding ( Burns and Machin, 2009, Burns and Machin, 2010 and Burns and Machin, 2012) has been previously identified and suggests up to 42% shared variance between these constructs. That saying, it needs to be emphasised that these factors have been derived from factor analytical procedures that show that the items from these respective well-being variables can discriminate their parent factors despite the strong correlations between factors. Table 2. Correlations between psychological wellbeing, subjective wellbeing, number and perceived impact of life events. PA NA # of LE Impact of LE EGPS PR AU PA – −.24⁎⁎⁎ −.06 .07 .65⁎⁎⁎ .30⁎⁎⁎ .22⁎⁎⁎ NA – .21⁎⁎⁎ .32⁎⁎⁎ −.38⁎⁎⁎ −.44⁎⁎⁎ −.35⁎⁎⁎ # of LE – .04 −.19⁎⁎⁎ −.21⁎⁎⁎ −.01 Impact of LE – .14⁎⁎ −.13⁎ −.06 EGPS – .29⁎⁎⁎ .30⁎⁎ PR – .18⁎⁎ Note. PA: Positive affect; NA: negative affect; LE: life events; EGPS: super ordinate factor derived from 4 of the PWB scales; PR: positive relations with others; AU: Autonomy. ⁎ p < .05. ⁎⁎ p < .01. ⁎⁎⁎ p < .001. Table options Hierarchical regression analyses tested the effects of Significant Life Events and PWB on both positive and negative affect (Table 3). Overall, the reported number of life events was only weakly associated with negative affect and its effect was accounted for by the inclusion of the PWB variables. In contrast, the perceived impact of life events was consistently positively associated with negative affect; those reporting higher impact of life events also reported higher levels of negative affect. Main effects for all three PWB factors were reported in Model 3, with significant negative associations between PWB and negative affect. However, when adjusting for the interaction between PWB and both number of life events and perceived impact of life events, most of the main PWB effects were no longer significant. One exception was for Positive Relations which reported a significant negative association with negative affect. One higher order interaction on negative affect was reported; an interaction between perceived impact of life events and positive relations. Plotting this interaction indicated high positive relations as protective of increased negative affect with increased perceived impact of life events (Fig. 2). Both the reported number of life events and perceived impact of life events were consistently unrelated to positive affect. Associations between the three PWB measures and positive affect were mixed. Autonomy was consistently unrelated with positive affect. Increased positive relations were strongly associated with positive affect (Model 3a) but not in the model adjusted for interactions between PWB and impact of life events (Model 4a). In contrast, EGPS, the super-ordinate PWB factor was consistently positively related with positive affect in both unadjusted and adjusted models. No higher order interactions between PWB and life events were reported. Inspection of Goodness of Fit indices suggest that for negative affect, Model 4 is the better fitting model. This was confirmed with a LL test of difference between Model 4 and 3 (χ2 LR = 26.08 (df = 6); p < .001). For positive affect, fit indices indicated comparable fit between Models 3a and 4a which was confirmed with a LL test of difference between Model 4a and 3a (χ2 LR = 7.15 (df = 6); p = .307). Table 3. Results of two series of hierarchical regressions to test the effects of PWB, life events and the perceived impact of life events on subjective wellbeing. Negative affect Positive affect Mode1 1 Mode1 2 Mode1 3 Mode1 4 Mode1 1a Mode1 2a Mode1 3a Mode1 4a R2 = .062 R2 = .157 R2 = .416 R2 = .457 R2 = .072 R2 = .074 R2 = .470 R2 = .481 B (SE a) B (SE a) B (SE a) B (SE a) B (SE a) B (SE a) B (SE a) B (SE a) Intercept 2.15 (.10)⁎⁎ 2.19 (.09)⁎⁎ 2.24 (.09)⁎⁎ 2.26 (.09)⁎⁎ 3.42 (.11)⁎⁎ 3.43 (.10)⁎⁎ 3.37 (.09)⁎⁎ 3.39 (.10)⁎⁎ # of LE 0.07 (.02)⁎⁎ .06 (.01)⁎⁎ .03 (.01) .02 (.01) −.02 (.02) −.02 (.02) .02 (.01) .02 (.01) Impact of LE .26 (.03)⁎⁎ .24 (.03)⁎⁎ .26 (.03)⁎⁎ .04 (.04) .01 (.03) .01 (.04) EGPS −.23 (.04)⁎⁎ −.18 (.08) .53 (.04)⁎⁎ .61 (.08)⁎⁎ PR −.22 (.04)⁎⁎ −.21 (.07)⁎ .11 (.03)⁎⁎ .12 (.08)⁎⁎ AU −.16 (.04)⁎⁎ −.01 (.07) −.01 (.04) −.00 (.07) # of events⁎ EGPS −.01 (.01) −.01 (.01) # of events⁎ PR .00 (.01) .00 (.01) # of events⁎ AU −.03 (.01) .00 (.01) Impact of LE⁎ EGPS −.04 (.04) .06 (.04) Impact of LE⁎ PR −.14 (.04)⁎ −.04 (.04) Impact of LE⁎ AU .00 (.04) −.06 (.04) Model fit indices AIC 871.24 834.01 706.46 692.38 886.54 887.68 690.44 695.29 BIC 902.42 869.08 753.22 762.53 917.72 922.75 737.21 765.44 LL −427.62 −408.01 −341.229 −328.191 435.27 −436.84 −333.22 −329.65 Note. LE: Life events; EGPS: super ordinate factor derived from 4 of the PWB scales; PR: positive relations with others; AU: Autonomy; AIC: Akaike Information Criteria; BIC: Bayesian Information Criteria; LL: Log-Likelihood. a Standard errors derived from Bootstrapped sample of 200. Estimates residualised for gender, age, education level currently studying for (Certificate, Diploma, Bachelor, Post-Graduate Diploma, Masters and Doctorate), English as first language, study load (full-time, part-time), mode of instruction (on-campus, distance, combination), and living location (hall of residence, rental property, parental home, own home). ⁎ p < .01. ⁎⁎ p < .001. Table options Positive relations with others moderates the effect of perceived impact of life ... Fig. 2. Positive relations with others moderates the effect of perceived impact of life events on negative affect. Figure options Finally, we supplemented our analyses to determine whether perceived impact of life events mediated the effects of psychological functioning and quality of interpersonal relationships on subjective wellbeing. Model 5 tested whether the perceived impact of life events mediated the effects of psychological function and quality of interpersonal relationships on subjective wellbeing. Results (Table 4) indicated that only a small amount of variance (9%) in perceived impact of life events was explained. Also, comparison of model fit with fit for Models 4 and 4a indicated that the mediation model (Model 5) reported poorer model fit. Along with the bivariate correlations, these results suggest that the perceived impacts of life events are mostly independent of the socio-demographic, psychological functioning and quality of interpersonal relationships variables. Table 4. Results of structural equation model to test the mediation effect of perceived impact of life events on subjective wellbeing. Model 5 Positive affect Negative affect Impact of life events R2 = .476 R2 = .474 R2 = .090 B SEa B SEa B SEa # of LE .02 .01 .02 .01 .01 .02 Impact of LE .02 .04 .26∗∗ .03 EGPS .59∗∗ .08 −.17⁎ .08 .17⁎⁎ .06 PR .14⁎ .07 −.17⁎ .07 −.16⁎⁎ .05 AU .01 .07 .01 .07 −.08 .05 Model fit indices AIC 8793.50 BIC 8826.11 LL −4351.75 Note. LE: Life Events; EGPS: super ordinate factor derived from 4 of the PWB scales; PR: positive relations with others; AU: Autonomy; AIC: Akaike Information Criteria; BIC: Bayesian Information Criteria; LL: Log-Likelihood. a Standard errors derived from Bootstrapped sample of 200. Estimates residualised for gender, age, education level currently studying for (Certificate, Diploma, Bachelor, Post-Graduate Diploma, Masters and Doctorate), English as first language, study load (full-time, part-time), mode of instruction (on-campus, distance, combination), and living location (hall of residence, rental property, parental home, own home). ⁎ p < .01. ⁎⁎ p < .001.