Antidepressants, social adversity and outcome of depression in general practice
Introduction
This paper is based on a secondary exploratory analysis of a randomised controlled trial addressing the effectiveness of anti-depressant treatment in general practice (Kendrick et al., 2009). Its aim is to consider such treatment in the light of psychosocial factors known to be of aetiological importance. Most social research concerning depression has focussed on onset (Brown and Harris, 1978, Kendler et al., 2002, Monroe, 1998). Evidence suggests that the majority are provoked by adverse social circumstances often involving humiliation, a sense of entrapment, or both (Brown et al., 1995, Kendler et al., 2003, Farmer and McGuffin, 2003). Furthermore the prevalence of depression differs markedly across populations, with this paralleling rates of adversity in these populations (Brown, 1998, Brown, 2002). But more relevant for questions raised by clinical trials are findings that psychosocial factors also play an important role in determining the course of an episode. For example, among London women with a child living at home with an onset of major depression a severe interpersonal difficulty at point of onset or a history of childhood maltreatment was associated with a fourfold increase in risk of the episode lasting at least 1 year: 43% (23/54) compared with 9% (4/47) of others with an onset (Brown and Moran, 1994; also Brown et al., 1994). This result has recently been replicated in a general population sample of women under 65 (Brown et al., 2008b). There is reason to believe that the experience of entrapment in an aversive situation is often involved in such chronic conditions, by serving to perpetuate hopelessness and a sense of a lack of a way forward. However, positive events can introduce a sense of hope and contribute to remission (Brown et al., 1992, Brown et al., 1988, Harris et al., 1999, Tennant et al., 1981). Such events may largely eliminate an ongoing severe difficulty (e.g. news of rehousing from an overcrowded flat), or be unrelated to a difficulty (e.g. a boyfriend of the same woman suggesting they get married).
When depression is recognised in general practice anti-depressant medication is often prescribed. While its effectiveness remains controversial (Parker, 2009, Moncrieff et al., 1998), current evidence suggests that its prescription is likely to be associated with a modest effectiveness. This is supported by the trial on which the present analysis is based of GP practices. This was an open pragmatic randomised controlled trial of treatments for depression seen in UK general practice, among 220 adult patients with a new episode of depression lasting at least eight weeks about whose treatment both GP and patient were in equipoise. The design required Hamilton Depression Rating Scale (HDRS) scores between 12 and 19, thus excluding any with what is commonly judged as a ‘severe’ condition. Supportive care (i.e. four consultations over a 12-week period) was compared with supportive care plus treatment with a selective serotonin reuptake inhibitor. Differences in favour of SSRI treatment were found in mean HDRS scores at 12 weeks adjusted for baseline HDRS (primary outcome) and the same at 26 weeks (secondary outcome) (Kendrick et al., 2009). The contribution of a placebo effect could not, given the pragmatic design, be ruled out. The significance of the data rests on its likely representativeness of antidepressant treatment in general practice and the fact its prescription on top of supportive care emerged as having a modest effect over that of supportive care alone in terms of symptoms, quality of life and cost effectiveness (Kendrick et al., 2009). Given these findings, the possible presence of a placebo effect does not detract from the relevance of exploring the contribution of social risk factors.
To our knowledge no study has systematically considered the role of antidepressants in the context of the kind of negative and positive social experiences outlined earlier. On the assumption they and SSRIs are make independent contributions, we examine whether in addition there is evidence of statistical interaction — that is whether the extent antidepressants increase chances of remission differs by social risk. The aversive and positive LEDS factors involve a confirmatory analysis of variables at baseline and follow up (that is both predating and postdating random allocation) that previous prospective research has shown directly relate to remission. The post allocation experience makes a critical contribution to chances of remission, especially via ‘fresh-start events’ and ‘difficulty reduction’. A separate exploratory analysis involving an unemployment-entrapment index is based on previously confirmed theory concerning the role of entrapment in the development of depression.
Section snippets
Basic design
Full details of the design are given in Kendrick et al. (2009). In brief GPs from three areas of England (Southampton, London and Liverpool) referred patients diagnosed with new episodes of depression and potentially in need of antidepressant medication. Inclusion criteria were age 18 and over, symptoms for at least 8 weeks, no antidepressant treatment within 12 months, baseline score 12 to 19 on the 17 item HDRS (Hamilton, 1967). Patients were visited at home after referral within a few days.
Demographic factors, and outcome and the LEDS index
Table 1 provides details of the baseline sample demographics. The arms were on the whole well balanced and we will deal with the exceptions in the course of the analysis.
Table 2 shows the development of the LEDS-index in terms of the aversive/positive ratings just outlined. The first two columns deal with aversive experience and the final column the result of reallocating some aversive to non-aversive because of positive experience in the follow-up period. This third column represents the final
Medication and aversive social experience
This exploratory analysis has taken as its point of departure the conclusion of the core paper concerning the trial (Kendrick et al., 2009) that SSRIs are associated with a statistically significant but modest effect when compared with supportive care alone. But among those in aversive contexts as defined by the Life Events and Difficulties Schedule (LEDS) less than a fifth in both arms had remitted 12-weeks after baseline (Table 3). This is consistent with a naturalistic study in primary care
Role of funding source
The study was funded by the National Institute for Health Research Health Technology Assessment programme, England. The opinions expressed in this paper are the authors' and do not represent the views of the Department of Health for England.
Conflict of interest
The authors declare the following competing interests: Tirril Harris and George Brown have published articles on the importance of social factors in determining the course and outcome of depression. Tony Kendrick has received fees for presenting at educational meetings, and/or research funding, from Lilly, Lundbeck, Servier, and Wyeth pharmaceuticals, and has also received HTA funding for research into psychological treatments. The remaining authors have declared no competing interests.
Acknowledgements
We are indebted to the other members of the THREAD study group, Morven Leese, Clare Flach and Mauricio Moreno for statistical and data management advice, and Richard Byng, Christopher Dowrick, Mark Gabbay, Paul McCrone, Richard Morriss, Michael Moore, Robert Peveler, and Andre Tylee for valuable comments on the draft. We would also like to thank the patients who gave up their time to answer our detailed personal questions.
References (47)
- et al.
Parental maltreatment and proximal risk factors using the Childhood Experience & Abuse (CECA) instrument: a life-course study of adult chronic depression - 5
J. Affect. Disord.
(2008) - et al.
Development of a retrospective interview measure of parental maltreatment using the childhood experience of care and abuse (CECA) instrument — a life-course study of adult chronic depression - 1
J. Affect. Disord.
(2007) - et al.
Parental maltreatment and adulthood cohabiting partnerships: a life-course study of adult chronic depression. - 4
J. Affect. Disord.
(2008) - et al.
Parental maltreatment and proximal risk factors using the Childhood Experience & Abuse (CECA) instrument: a life-course study of adult chronic depression - 5
J. Affect. Dis.
(2008) - et al.
Predictability of the one-year course of depression and generalised anxiety in primary care
Gen. Hosp. Psych.
(2002) - et al.
Depressive symptoms in relation to physical health and functioning in the elderly
Am. J. Epidemiol.
(1986) Genetic and population perspectives on life events and depression
Soc. Psychiatry Psychiatr. Epidemiol.
(1998)Social roles, context and evolution in the origins of depression
J. Health Soc. Behav.
(2002)- et al.
Social Origins of Depression
(1978) - et al.
Motherhood, employment and the development of depression: a replication of a finding?
Br. J. Psychiatry
(1990)