Research report
Effects of social support during weekend leave on cortisol and depression ratings: a pilot study

https://doi.org/10.1016/S0165-0327(01)00414-1Get rights and content

Abstract

Background: The biological mechanism by which social support influences the course of a depressive episode may involve the stress response which is reflected and/or mediated by cortisol. The study took advantage of the weekend leave that inpatients receive towards the end of an admission to investigate the inter-relationship between social support, cortisol secretion, and the severity of depression. Method: For 23 inpatients with a major depressive episode (DSM IV) differences between ward and home in social support, depression ratings, and cortisol secretion were compared. The effect of hassles on cortisol secretion was also assessed. Results: An inverse linear relationship was found between changes in social support and depression ratings across the two settings. No relationship was found between changes in the other two sets of variables. Hassles resulted in increased cortisol secretion. Limitations: The small sample limits the analysis of hypotheses of interest. Findings are restricted to an inpatient tertiary referral sample. Conclusions: Weekend leave is an appropriate paradigm to study the effect of social influences on cortisol secretion, and the severity of depression. It is feasible for depressed inpatients to accurately collect timed saliva samples both on the ward and when at home, and for research workers to measure social support provided by a ward. The relationship between social support and depression has clinical implications in terms of interpreting mood changes following weekend leave. Hassles are associated with increased secretion of cortisol in depressed patients, which extends similar previous findings in normal subjects.

Introduction

The sensitivity of the hypothalamic–pituitary–adrenal axis to environmental effects suggests a mediating role in depression (Calloway and Dolan, 1989). Similar psychosocial factors affect both corticosteroid secretion in animals exposed to chronic stress and the course of depression in humans (Checkley, 1996). The protective effect of social support in reducing vulnerability to depression may act by alleviating the endocrine response to environmental stress including hassles (minor daily events; Nicolson, 1992).

The pulsatile pattern of cortisol secretion resulting in diurnal variation in levels, and sensitivity to stress, requires hourly cortisol measurements for accurate estimation of cortisol secretion. The acute stress response commences within a few minutes of a hassle, and lasts an hour. Salivary measurement provides an accurate indication of the free hormone (Kirschbaum and Hellhammer, 1994). Hourly measurement would reflect a hassle that has occurred in the previous hour.

Inpatient admission enables hospital and home to be compared in terms of depression, social support and cortisol levels. Weekend leave is used to prepare for discharge and to assess the robustness of improvements made. Across ward and home settings hypotheses were that differences in social support ratings will be negatively associated with (1) differences in depression ratings and (2) differences in cortisol levels. An additional hypothesis was that cortisol levels will be positively associated with the experience of hassles.

Section snippets

Method

Subjects were inpatients in the Affective Disorders Unit of the Bethlem and Maudsley Hospitals with DSM IV (American Psychiatric Association, 1994) diagnoses of major depressive episode.

The following measurements were taken on 2 days — (1) 1 weekday on the ward before the weekend leave and (2) 1 day during the weekend leave:

(i) The Montgomery Asberg Depression Rating Scale (MADRS; Montgomery and Asberg, 1979).

(ii) Salivary cortisol using the ‘salivette’ (Sarstedt, Leicester). Subjects made

Sociodemographic characteristics

The mean age was 49.1 years (S.D. 11.6; range 24–65). 14 (61%) of subjects were male; 18 (78%) were married or cohabiting; and 15 (65%) were employed.

Depression

The mean MADRS score on the ward was 16.9 (S.D. 8.8; range 4–38), and at home was 15.0 (S.D. 7.6; range 3–29).

Crisis support

Eleven subjects (48%) were able to elicit crisis support on the ward, and 13 (57%) at home. Seven subjects (30%) were able to elicit crisis support in both settings, 10 subjects (44%) in one, and six subjects (26%) in neither.

Cortisol

The

Feasibility of obtaining data

The characteristic pattern of the diurnal decline in cortisol and the high correlation between ward and home values demonstrate that depressed patients can accurately collect timed saliva samples both on the ward and on weekend leave. The study also shows that it is possible to measure social support provided by a ward setting.

Social support findings

There was a linear relationship between changes in depression scores across different levels of crisis support. The stratified analysis of the 10 subjects with crisis

References (14)

There are more references available in the full text version of this article.

Cited by (30)

  • Systematic review of therapeutic leave in inpatient mental health services

    2018, Archives of Psychiatric Nursing
    Citation Excerpt :

    The only study providing statistical data (Bolin et al., 1969) reported 177 suicides per 100,000 home leave episodes; deteriorating physical health, real, threatened or imagined loss, history of suicidal ideation or suicidal attempts, and short admission length increased risk. Based on assumptions that i) a depressive episode might be influenced by a stress response to changes in social support; and ii) the response might be mediated by, and/or reflected in, salivary cortisol levels, Sayal et al. (2002) studied 23 inpatients diagnosed with a major depressive episode on the ward and during weekend leave. Change in cortisol levels was not statistically significant and the study hypothesis was unsupported.

  • Social support and oxytocin interact to suppress cortisol and subjective responses to psychosocial stress

    2003, Biological Psychiatry
    Citation Excerpt :

    Socially supported subjects might appraise the anticipated acute stressor as less threatening and more controllable than unsupported subjects (Kirschbaum et al 1995). These results from laboratory research are in accordance with clinical studies indicating numerous health benefits of social support (Buchanan 1995; Erickson et al 1998; Evans and Steptoe 2001; Gallo et al 2000; Hays et al 2001; Macdonald et al 1998; Sayal et al 2002; Steptoe 2000; Uchino et al 1999; Uno et al 2002). This consistent finding speaks for the unique nature of positive social interaction that might modulate the activity of the HPA axis and the autonomic nervous system.

View all citing articles on Scopus
View full text