ReviewThe prospective long-term course of adult depression in general practice and the community. A systematic literature review
Introduction
Depression is today's most common mental disorder. Its lifetime prevalence ranges between 11% and 17% (Alonso et al., 2004, Bromet et al., 2011, Kessler et al., 2005) and it is associated with immense disability burden (Wittchen et al., 2011). According to recent research mood disorders currently rank as the most cost intensive disorders of the brain at least in Europe (Gustavsson et al., 2011). Depressive disorders occur within all age groups. Their average onset age lies at around 26 years and thereby considerably later than that of other mental disorders (Bromet et al., 2011, Kessler et al., 2005). Women are twice as frequently affected compared to men (Van de Velde et al., 2010).
The long-term course of depression has been studied in primary care and general population samples (Hardeveld et al., 2010, Licht-Strunk et al., 2007, van Weel-Baumgarten et al., 2000) as well as clinical populations (Hardeveld et al., 2010, Keller and Boland, 1998, Richards, 2011). If the naturalistic long-term course of mental disorders is to be examined, it is important to differentiate between patient or clinical and non-patient or population based samples. Most outcome research focused on the longitudinal course of depression within clinical samples, i.e. samples from secondary or tertiary care (van Weel-Baumgarten et al., 2000), an observation that was valid over ten years ago and still is today. The previous research suggested that patients who attend secondary or tertiary care have more severe depressive courses than subjects from the community (Hardeveld et al., 2010). As a consequence, the long-term course of depression has often been described as being highly recurrent encompassing considerable costs for the individual patient as well as society (Keller and Boland, 1998, Mueller et al., 1999). However, the findings of studies with clinical patients might suggest a too negative outlook as they only include subjects already attending secondary or tertiary care and therefore are prone to bias (Berkson, 1946). This is impressively confirmed by the comparison of two long-term studies: in the Collaborative Depression Study (Keller and Boland, 1998) that included in- and out-patients with major depression, 85% of patients experienced a recurrence over a time span of 15 years while in the Baltimore ECA follow-up, a community study, this percentage was only 35% over the same period of time (Eaton et al., 2008).
It has been stated that the best way to study the naturalistic long-term course of depression is using a prospective design and subjects from the general population, who experience their first depressive episode within the study period (Eaton et al., 2008). As depression often is a recurring illness the latter is important to gain a valid estimate for the frequency of single-episode depression. Furthermore, research showed that past episodes are an important predictor of future episodes (Eaton et al., 2008, Lee, 2003). However, only very few studies implemented such a design (Eaton et al., 2008, Mattisson et al., 2007), therefore information on this issue is especially scarce.
Earlier reviews on the long-term course of depression have been conducted. The aspects they targeted were however different from the research objectives in focus here. Hardeveld et al. (2010) presented data about depressive recurrences in the adult population (including clinical samples). Due to a different research objective, all in all only two studies with samples from primary care or community settings were included. The recurrence rate was 31% after one year in a primary care sample and 35% after 15 years in a population based cohort. Notable predictors of recurrence were clinical factors that lay within the disorder itself, especially the number of previous episodes and subclinical residual symptoms after a recovery. Demographic factors on the other hand that can be a risk-factor for the onset of depression did not seem predictive of a negative course (Hardeveld et al., 2010).
Licht-Strunk et al. (2007) reviewed data from 21 elderly cohorts recruited in general practice and the community that were followed-up over one to ten years. They found long-term persistence rates of depression that varied between 32% and 54%.
Finally, more than a decade ago van Weel-Baumgarten et al. (2000) investigated the long-term course of depression in the general population as well as primary care patients. The authors found eight studies with follow-up periods of at least five years. Recurrence rates ranged between 30% and 40%. The generalizability of these results is however limited as the review included four samples comprising limited age groups. Furthermore two studies had a retrospective design.
In sum, there is no systematic review about the naturalistic longitudinal course of depression in adult subjects from the community or primary care that includes studies with a prospective design and a follow-up duration of at least several years. Such data is needed to identify the percentages of subjects with a favorable prognosis that will not have to undergo extensive treatments as well as those subjects with recurrent or persistent courses for which more intensive treatment is indicated (Lee, 2003). Additionally, information about potential predictors of a favorable or unfavorable course may help identify different subgroups and develop suitable and better fitted therapeutic interventions. As van Weel-Baumgarten et al. (2000) we decided to include samples recruited in general practice or the community, in order to get an insight into the course of depression within samples not referred to secondary or tertiary care. Our research was conducted along the following questions:
- •
How do depressive disorders develop over a time span of at least three years within depressed adult subjects recruited in general practice or the community?
- •
Which methodologies were applied in terms of follow-up length, number of follow-up assessments as well as diagnostic and outcome criteria?
- •
Which factors have been found significant as predictors of the long-term course?
Section snippets
Identification of relevant literature
To identify relevant literature, electronic databases Medline and PsycINFO were searched up to August 22nd 2013 for English language articles by two independent researchers (CS, MH). We excluded articles that were editorials, comments and others that were not reviews or empirical studies. Manual searches of reference lists of included studies and several reviews on the topic (Fox, 2002, Hardeveld et al., 2010, Holzel et al., 2011, Richards, 2011, van Weel-Baumgarten et al., 2000) were performed.
Study characteristics
Due to heterogeneous methodology (i.e. outcome variables, follow-up lengths with differences of up to several decades, number of assessments) the included studies are reviewed narratively.
After screening according to the selection criteria described above 12 different cohorts (13 studies) with 4009 followed-up depressed subjects were included in the review. Table 2 provides an overview over the main characteristics of included studies.
Seven cohorts included participants from the general
Main findings
Although there has been quite extensive longitudinal research on the course of depressive disorders that has in turn been the focus of several review articles (Fox, 2002, Hardeveld et al., 2010, Licht-Strunk et al., 2007, Richards, 2011, van Weel-Baumgarten et al., 2000) no systematic review so far summarized the literature on what we know today about the prospective long-term course of this disorder from studies conducted in primary care or general practice. One reason for this finding might
Conclusion
In sum, studies investigating the longitudinal course of depression in samples from the community and primary care, found that one third to two thirds of depressed participants achieved a stable recovery, while 10–17% encountered a persistent and the remaining part an intermittent course. These rates indicate that the prognosis of depression in the community and primary care is better than in clinical settings (i.e. secondary and tertiary care), as has also been suggested earlier by Hardeveld
Role of funding source
This research was supported by grants from the Dr. Karl-Wilder-Stiftung. The sponsor had no involvement in the study design, in collection, analysis and interpretation of data, in writing the article, and in the decision to submit the article for publication.
Conflict of interest
No conflict declared.
Acknowledgment
The authors would like to thank the anonymous reviewers for their thoughtful comments and suggestions which helped strengthen the paper.
References (45)
- et al.
Cognitive and social functioning in recovery from depression: results from a population-based three-year follow-up
J. Affect. Disord.
(2006) - et al.
Psychological well-being and residual symptoms in remitted patients with panic disorder and agoraphobia
J. Affect. Disord.
(2001) - et al.
Cost of disorders of the brain in Europe 2010
Eur. Neuropsychopharmacol.
(2011) - et al.
Risk factors for chronic depression—a systematic review
J. Affect. Disord.
(2011) - et al.
Are child-, adolescent-, and adult-onset depression one and the same disorder?
Biol. Psychiatry
(2001) - et al.
Implications of failing to achieve successful long-term maintenance treatment of recurrent unipolar major depression
Biol. Psychiatry
(1998) - et al.
Sociodemographic and psychiatric determinants of attrition in the Netherlands Study of Depression and Anxiety (NESDA)
Compr. Psychiatry
(2012) - et al.
The three-year naturalistic course of major depressive disorder, dysthymic disorder and double depression
J. Affect. Disord.
(2009) Prevalence and clinical course of depression: a review
Clin. Psychol. Rev.
(2011)Recovery from depression
Psychiatr. Clin. North Am.
(2012)