Introduction
Depression is one of the most prevalent and disabling of psychological problems related to diminished quality of life and role functioning, medical morbidity and mortality (Spijker et al. Reference Songprakun and McCann2004; Üstün et al. Reference Üstün, Ayuso-Mateos, Chatterji, Mathers and Murray2004; Paykel et al. Reference Pampallona, Bollini, Tibaldi, Kupelnick and Munizza2012). There are over 350 million people experiencing depression annually [World Health Organisation (WHO), 2012]. In addition, depression is a significant public health concern across all regions of the world (Bromet et al. Reference Bromet, Andrade, Hwang, Sampson, Alonso, de Girolamo and Kessler2011). Within an EU context, research carried out by Wittchen et al. (Reference Wittchen, Jacobi, Rehm, Gustavsson, Svensson, Jönsson and Steinhausen2011) reported that ~38% of the EU population (164.8 million people) struggle with mental health difficulties each year. Major depression affects 30 million people in the EU, becoming the most burdensome disorder of all diseases. The existing evidence for the effectiveness of psychological treatments and/or antidepressant medication treatments for those diagnosed with moderate levels of depression is ambiguous and a review of such evidence is, therefore, warranted.
Based on severity of symptoms, depression can be reported and experienced at different levels of severity. The Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-V) classifies major depressive disorder into three distinct levels, namely mild, moderate, or severe, based on the intensity of depressive symptoms, symptom count, and degree of functional impairment (American Psychiatric Association, 2013). Studies investigating depression in the general population have reported that the prevalence of Major Depressive Disorder (MDD) for 12 months as 6.6%, affecting between 13.1 and 14.2 million US adults. Kessler et al. (Reference Hamilton2003) has shown that, of those with MDD, 10.4% were reported as ‘mild’, 38.6% ‘moderate’, and 51% ‘severe or very severe’ depression. Another study by Kessler et al. (Reference Kessler, Berglund, Demler, Robert, Koretz, Merikangas, Rush, Walters and Wang2005) reported that of 9282 respondents, 6.7% identified as depressed and within that group, 30.4% reported ‘severe’ depression, 50.1% ‘moderate’ depression, and 19.5% ‘mild’ depression. Epidemiological data depicting the severity of depression are largely restricted to populations from North America, and there is a need for data depicting the severity of depression from a representative European sample.
Many treatments have been recommended for depression including different medications, a range of psychological therapies, exercise regimes, and alternative remedies. One of the influencing factors in determining the efficacy of any or all of these treatments is the level of severity of the depressive symptoms [see National Institute of Care and Excellence (Reference Moher, Shamseer, Clarke, Ghersi, Liberati, Petticrew and StewartNICE), 2009].
Severity of depression as a significant factor in treatment
The severity of depression is a significant factor in examining treatment effectiveness. The effectiveness of antidepressant medication and placebo was investigated by Fournier et al. (2010), which concluded that the benefits of antidepressant medication over placebo were only substantial for patients with severe depression. Similar studies by Khan et al. (Reference Kessler, Chiu, Demler and Walters2002) and Kirsch et al. (Reference Khan, Faucett, Lichtenberg, Kirsch and Brown2008), identified that clinically meaningful improvements could only be achieved where initial Hamilton Depression Rating Scale (HDRS) scores exceeded 28 (where a score of 23 is indicative of very severe depression). Fournier et al. pointed out that there is a paucity of investigations of the true effects of antidepression medication as a treatment for patients with less than severe depression. Fournier et al. also commented that evidence concerning the effects of antidepressant medication in patients with mild and moderate MDD has been sparse. An important conclusion of these findings was that high levels of depression symptom severity was required for clinically meaningful drug/placebo differences to emerge. These findings are surprising, particularly, given the evidence that the majority of patients receiving antidepressant medication in clinical practice present with scores below this level of severity. Fournier et al. concluded that prescribers, policy makers, and consumers might not be aware that the efficacy of medications has been largely established on the basis of studies that have included only those individuals with more severe forms of depression.
Given that research suggests that a significant proportion of the population experiencing depression are identified as moderately depressed (e.g. Kessler et al. Reference Kessler, Berglund, Demler, Robert, Koretz, Merikangas, Rush, Walters and Wang2005), the present paper systematically reviews studies of treatment effectiveness for adults experiencing depression at this level of severity.
Current treatment guidelines for moderate depression
The NICE guidelines provide recommendations for professional standards of practice in relation to a range of psychological difficulties in the United Kingdom, and are based on the DSM-V [American Psychiatric Association (APA), 2013]. Reference Moher, Shamseer, Clarke, Ghersi, Liberati, Petticrew and StewartNICE (2009) guidelines state, ‘if you have moderate or severe depression, you should be offered both an antidepressant and a psychological treatment’. The present paper examines the evidence supporting this recommendation by carrying out a systematic review of all studies on the treatment of moderate depression over a 16-year period. To date, there is no evidence to indicate that treatment intervention for moderate depression is the same as that for either mild or severe depression, it is important, therefore, that evidence for the treatment efficacy of this specific level of depression severity is examined.
Levels of severity of depression in research designs
Studies of depression tend to examine an amalgamation of severity levels within their samples, such as ‘mild to moderate’, and ‘moderate to severe’ (e.g. van der Lem et al. Reference van der Lem, van der Wee, van Veen and Zitman2012). The criteria used do not identify moderate depression as a specific severity level to be examined. For example, the specific severity band of depression was not a consideration in Cuijpers et al. (Reference Cuijpers, Sijbrandij, Koole, Andersson, Beekamn and Reynolds2009) meta-analysis on treatment effectiveness for chronic depression. Similarly, Cuijpers et al. (2010 Reference Cuijpers, van Straten, Bohlmeijer, Hollon and Anderssona ) looked at the effects of psychotherapy for depression and used bands of ‘moderate to severe’, ‘severe’, and ‘mild to moderate depression’. No conclusions were made in relation to moderate depression. In another meta-analysis of 16 randomised-controlled trials (RCTs) examining the effects of psychotherapy on chronic depression, Cuijpers et al. (2010 Reference Cuijpers, van Straten, Schuurmans, van Oppen, Hollon and Anderssonb ) included studies where participants met diagnostic criteria for ‘a depressive disorder’ but likewise no reference was made to a specific categorisation to be examined. More recently, Cuijpers et al. (Reference Cuijpers, van Straten, van Oppen and Andersson2014) reported that combined treatments were found to be more effective than either psychotherapy or medication alone and the criteria for inclusion used was ‘depressive disorder’. In this meta-analysis, the effect size was reported to be influenced by the severity of depression. Individual depression scores, however, were not considered, but rather an average score was taken for the sample as a whole. Meta-analytic studies have been carried out to establish guidelines on the treatment for different levels of depression. For example, Fournier et al. (2010) conducted a meta-analysis of RCTs of antidepressants in the treatment of the different severity levels of depression conducted between 1980 and 2009. In this study, the sample chosen was grouped into ‘mild to moderate’, ‘severe’, and ‘very severe’ based on the HDRS scores offered by the APA. Notably, there were no known studies on ‘moderate’ depression reported.
Numerous systematic reviews have also been carried out in order to assist with clarifying treatment effectiveness for depression. Conclusions, however, refer to MDD rather than the different severity levels of depression (e.g. Khan et al. Reference Fournier, Derubeis, Hollon, Dimidjian, Amsterdam, Shelton and Fawcett2001; Zhou et al. Reference Zhou, Michael, Liu, Del Giovane, Qin, Cohen, Gentile and Xie2014; Linde et al. Reference Le, Perry and Stuart2015). To date, however, there are no known systematic reviews carried out on treatment effectiveness solely for moderate depression.
In summary, the different bands of depression have been predominantly amalgamated so that moderate depression has been either considered in combination with mild depression or with severe depression, but not as a stand-alone band of depression. Past research has predominantly focussed on the efficacy of pharmacotherapy and psychotherapy treatments for depression across a wide range of symptom severity in individuals diagnosed with depression, without paying particular attention to the three main categories of depression severity stated in DSM-V. The present paper, therefore, addresses a gap in the literature by systematically reviewing the efficacy of treatments for those diagnosed with moderate levels of depression.
Method
Inclusion criteria: data sources and search strategy
The present review examines studies of the treatment effectiveness for moderate depression alone. The systematic review was written up according to the PRISMA standard (a protocol used to evaluate systematic reviews; Moher et al. Reference Linde, Sigterman, Kriston, Rucker, Jamil, Meissner and Schneider2015). In order to identify studies which included antidepressant pharmacotherapy and psychotherapy (psychological treatment), a search of EBSCO and SCOPUS was carried out that included the following databases: Medline, Cumulative Index to Nursing and Allied Health Literature, Psych Articles, Psych Info, Omnifile, Amed, Academic search complete, Social Sciences, UK and Ireland reference centre. The final database to be used was the Cochrane Database of Systematic Reviews. All searches were restricted to the period from January 2000 to January 2014 and included all keywords ordered as the following; ‘Moderate Depression’ and/or ‘Psychotherapy’, and/or ‘Medication’. Subject headings included were as follows: depression, depressive disorders, antidepressants, psychotherapy, mental health services, and RCTs. Search terms for moderate depression, treatment, psychotherapy, medication (antidepressants), and combination treatments were explored.
Primary studies, including RCTs and systematic reviews that investigated pharmacotherapy, psychotherapy alone, and pharmacotherapy and psychotherapy combined, with moderate depression alone, were retained. Studies where moderate depression was investigated in combination with mild and/or severe depression or other types of depression were therefore excluded from this review. In order to classify the definition of psychological treatment, our review followed the system of Rush & Thase (Reference Paykel, Andrade, Njenga and Phillips1999) who regard interpersonal, cognitive, behavioural, and psychodynamic therapies as psychological treatment. Although each of these therapeutic interventions hold very different conceptual backgrounds, the rationale for including them all is that each treatment is focussed on the reduction in symptoms of depression and the prevention of reoccurrence and relapse (Pampallona et al. 2004).
Study selection
Published and unpublished studies were eligible. Studies that included the treatment of moderate depression solely were selected. All the abstracts of the papers for inclusion in the review were screened and the full paper was obtained where there was insufficient information in the abstract.
Data extraction and quality assessment
The quality of the studies were assessed in terms of methodological strength and limitations, that is recruitment procedure, sample size, and sufficient reporting of primary outcomes. Specific to RCTs, random allocation sequences, concealment of allocation sequences, blinding, and reporting of proportions of patients lost to follow-up were also assessed. An ad hoc form was designed for data extraction including information such as setting, number of participants, diagnosis, sex, mean age, type of intervention, measures used, and antidepressant drug administered. A data extraction table was developed detailing each of the variables of interest for the present review. Using the extraction table as a template, two reviewers extracted relevant data from all of the articles. After completion of the initial data extraction, a second independent reviewer checked the accuracy of the extracted data. In the case of disagreements between two reviewers, a third reviewer examined the full article to make a determination about whether to include or exclude the article. Review authors carefully considered the potential limitations of the included studies. Methodological appraisal of each study was conducted according to PRISMA standard.
To assess study quality, six quality rating criteria were selected in order to inform whether or not each study should be rated as high, medium, or low quality. These criteria were selected based on the researchers reading of what elements should be required to ensure high research quality (mainly based on a RCT design). The six criteria were as follows: (1) the study addressed an appropriate and clearly focussed question, (2) random assignment allocation, (3) participants and investigators were ‘blind’ about treatment allocation, (4) treatment and control groups were equivalent at baseline, (5) the only difference between groups was the treatment under investigation, and (6) all relevant outcomes were measured in a standard, valid and reliable way. A quality checklist table was developed based on these criteria. Two reviewers independently assigned a quality rating (‘yes’, ‘no’, ‘can’t say’) to each study; results were compared and differences discussed until agreement was obtained. The decision for low, medium, and high-quality ratings were as follows: (a) low: two criteria were present and four were not/can’t say, (b) medium: four criteria were present and two were not/can’t say, and (c) high: all six criteria were met.
Results
Data extraction
From the initial searches on depression, over 78 800 articles were identified investigating depression. Based on broad inclusion criteria, that is studies of the treatment effectiveness solely for moderate depression between January 2000 and January 2014, 278 were relevant for further screening. The identified abstracts lead to the exclusion of 264 papers due to the inclusion of mild and severe severity depression levels, and the inability to reach the set inclusion criteria, namely primary studies including RCTs and systematic reviews that investigated pharmacotherapy, psychotherapy, and combination treatments with moderate depression. For each of the 14 remaining studies meeting our inclusion criteria, available information regarding the sample (i.e. sample size, basic demographics, and recruitment setting), study design, intervention type, and any descriptive findings related to behavioural and mental health outcome variables (see Table 1) were extracted.
Table 1 Studies in the review that matched inclusion criteria
RCT, randomized-controlled trial; BDI, Beck Depression Inventory; HAMD-17, Hamilton Depression Scale; BDI-II, Beck Depression Inventory-II; CES-D, The Center for Epidemiological Studies Depression Scale; LTPP, long-term psychodynamic psychotherapy; WAIS-III, Wechsler Adult Intelligence Scale Version III; CBT-I, cognitive behavioural therapy for insomnia.
Details missing in Table 1 is due to such information being unattainable from the studies cited.
a Country of origin.
A total of 14 studies with a total of 1743 participants met the inclusion criteria. Five of the studies used the Hamilton Depression Scale (HAMD-17) to measure the levels of moderate depression in the sample. The Beck Depression Inventory (BDI) was also used as a measure in five of the studies reviewed, with the remaining studies using measures such as The Center for Epidemiologic Studies Depression Scale (CES-D) and the ICD-10 Guide for Depression Diagnosis. Six of the studies examined medical treatments (including herbal), four studies examined psychological treatments, and a further four studies examined combined treatments of both medical and psychological interventions.
Study quality
With regard to study quality, the methodological quality of the 14 studies included in this review was standard. Of the 14 studies, eight studies were rated as high quality, while six studies were rated as medium. For those studies rated as medium, all met at least four of the six quality criteria; three did not blind the investigators and participants to treatment allocation, while two studies did not contain sufficient information to permit assessment. Two of the studies rated as medium did not randomly assign participants to treatment conditions. A bias check was also carried out on all studies with no reference to bias mentioned in any study, with the exception of Bastos et al. (Reference Bastos, Guimarães and Trentini2013) who employed bias-adjusted κ statistic.
Medical interventions
Of the six studies which investigated the effectiveness of medical interventions to treat moderate depression, four investigated the effectiveness of herbal remedies including Hypericum extract (St. Johns Wort) and Jieyu Pill (Chinese medicine) compared to traditional antidepressants (Uebelhack et al. Reference Uebelhack, Gruenwald, Graubaum and Busch2004; Gastpar et al. Reference Fraudenstein, Jagger, Arthur and Donner-Banzhoff2005, Reference Gastpar, Singer and Zeller2006; Yeung et al. Reference Yeung, Chung, Ng, Yu, Ziea and Ng2014). Each of these studies found that the herbal remedy was as effective as antidepressants when treating moderate depression. One study examined the effectiveness of pharmacotherapy in treating moderate depression when compared with the combination of pharmacotherapy and an aerobic training programme (Cerda et al. Reference Cerda, Cervelló, Cocca and Viciana2011) and found a decrease of depression symptoms only when the aerobic training programme was included. One study solely investigated the effectiveness of antidepressants in the treatment of moderate depression when compared with a placebo (Klein et al. Reference Kirsch, Deacon, Huedo-Medina, Scoboria, Moore and Johnson2014). The results indicated no significant difference between treatment and placebo groups in cases of moderate depression.
Psychological interventions
Four studies focused on psychological interventions such as cognitive behavioural therapy (CBT). Results showed CBT to be effective in reducing moderate depressive systems when used as an individual treatment, compared with usual care and a control condition (Antoni et al. Reference Antoni, Lehman, Kilbourn, Boyers, Culver, Alferi, Yount, McGregor, Arena, Harris and Price2001; Le et al. Reference Klein, Lee, Brouillette, Sheehan, Walmsley, Wong, Conway, Hull, Cooper, Haidar and Vezina2011; Carter et al. Reference Carter, Dyer and Mikan2013). Results also indicated that CBT was effective in treating moderate depression when integrated with other forms of psychotherapy (Hamamci, Reference Goldberg and Williams2006).
Combined interventions
Of the 14 studies that were included in this review, two studies investigated antidepressants and psychological interventions combined as an effective treatment. Stötter et al. (Reference Spijker, de Graaf, Bijl, Beekman, Ormel and Nolen2013) investigated the effectiveness of an 8-week mindfulness intervention with antidepressants compared to antidepressants treatment only. Results showed that the use of mindfulness therapy demonstrated significant additional benefits in the treatment of moderate depression when combined with the use of antidepressants. Bastos et al. (Reference Bastos, Guimarães and Trentini2013) investigated the use of long-term psychodynamic therapy (LTPP) alone, pharmacotherapy alone and both combined. Results indicated that LTPP and pharmacotherapy combined were more effective in modifying specific areas of cognition than antidepressants alone or LTPP alone, in patients with moderate depression. Results of this systematic review also indicated that the use of self-help and biblio-therapy were effective in reducing the symptoms of moderate depression. Songprakun & McCann (2012 Reference Rush and Thasea ) demonstrated that the inclusion of a self-help manual for depression with usual care (including medication treatment) marginally reduced levels of moderate depression over time. In a further study, Songprakun & McCann (2012 Reference Songprakun and McCannb ) demonstrated the effectiveness of including a self-help manual with usual care in reducing psychological distress in individuals with moderate depression and therefore improving treatment outcomes.
Discussion
In the present systematic review, 14 studies were identified, that were carried out between 2000 and 2014, specifically examining the treatment of moderate depression. Results indicate that psychological therapy on its own has been shown to be effective with this level of severity in depression in a small number of studies. In the present review, there is no evidence to support the effectiveness of antidepressant medication on its own in the treatment of moderate depression. In addition, this review highlights the lack of well-designed studies examining the effective treatments for moderate depression. Only one of the studies under investigation was identified that specifically compared psychotherapy alone with medication alone (i.e. Bastos et al. Reference Bastos, Guimarães and Trentini2013, n=272) and one further study compared psychotherapy with usual care (Le et al. Reference Klein, Lee, Brouillette, Sheehan, Walmsley, Wong, Conway, Hull, Cooper, Haidar and Vezina2011, n=217). In both studies, psychological treatment out-performed the medication. Given the limited number of studies in this review, it is argued that this small body of research is not sufficient to inform clinical guidelines and recommendations.
From reviewing the published research, the current review draws attention to the use of classifications of depression such as ‘mild to moderate’ and ‘moderate to severe’ within research (e.g. van der Lem et al. Reference van der Lem, van der Wee, van Veen and Zitman2012) and suggests that such classifications, whilst at times clinically useful perhaps, may not be the most useful to inform research. DSM-V is the standard bearer for categorisation but does not categorise the severity of depression using ‘mild to moderate’ or ‘moderate to severe’ as bands of depression.
The adherence to clear guidelines for the differential diagnosis of mild, moderate, and severe depression may be necessary in order to more accurately inform treatment guidelines. Large-scale epidemiological studies, such as Wittchen et al. (Reference Wittchen, Jacobi, Rehm, Gustavsson, Svensson, Jönsson and Steinhausen2011), which refer to the gross figures of depression in Europe may in future consider including prevalence rates for the categories of mild, moderate, and severe depression. Gross figures for depression can lead to misinterpretation, due to the varying intervention approaches that may be indicated for the differing levels of severity (see Reference Moher, Shamseer, Clarke, Ghersi, Liberati, Petticrew and StewartNICE, 2009). This review found that the majority of research investigating depression, typically distinguishes between levels of depression using identification tools such as the HDRS (Hamilton, Reference Hamamci1960), BDI (Beck et al. Reference Beck, Ward and Mendelson1961), and the General Health Questionnaire (Goldberg & Williams, Reference Gastpar, Singer and Zeller1991). Identification tools such as these cannot be the sole measure used to identify a patient’s severity or type of depression and the resulting treatment recommendations incorporated. It is also important to take into account the degree of functional impairment and/or disability associated with the possible depression, the duration of the episode, and the clients’ own preferences, motivations, intentions, and likelihood of engaging successfully with a treatment modality. In this context, further research needs to conduct studies using diagnostic interviews as a gold standard to identify those with moderate levels of depression. Rather than seeing major depressive disorder as an all-or-nothing condition, it may be more practical and realistic to view it as occurring across a continuum of severity, with DSM categories of mild, moderate, and severe being considered as useful markers/bands or cutoff points for research purposes.
A search of studies from 2000 to 2014 was used in the present study. This resulted in a small number of studies being identified for further review. The decision to only included ‘moderate depression’ in our search terms in this review was deliberate in order to focus solely on available treatment evidence for moderate depression. Had we adopted a broader approach in our systematic review of studies, incorporating studies on depression other than specifically on moderate depression, it is likely that we would have generated a far larger number of studies for inclusion. This, however, would have defeated the purpose of the study, which was to do a specific review of moderate depression only, resulting in a small number of studies meeting the inclusion criteria. A limitation of this period of research is that it does not include the most recent of studies (i.e. 2014–2017). We would not expect changes in our conclusions from a more up to date review, however, as the results are representative of the historical paucity of research specifically related to ‘moderate’ depression. Nonetheless, future research studies may follow up on this issue now that it has been highlighted here.
The clinical implication of these findings is that current treatment guidelines for moderate depression, that is ‘both an antidepressant and a psychological treatment’, may need to be revisited. The treatment of moderate depression needs to be approached with more caution, as this research suggests that there is no evidence for a combined approach for the treatment of moderate depression specifically, as recommended in current guidelines, for example NICE.
In conclusion, given that depression is one of the biggest health challenges the world faces at present (The Economist, Reference Tao2014), we advocate for more research to examine the effectiveness of treatment for different levels/bands of depression severity. Further research may lead to different recommendations for treatment by symptom severity. The present study highlights the urgent need for comprehensive data to be collected on the incidence of and effective treatments, specifically, for moderately depressed people so as to further inform public policy, mental health care strategies, and service delivery. There is currently no funded research that the authors are aware of that is collecting data of this nature in Europe and the present paper hopefully contributes toward encouraging such research to be carried out in future.
Acknowledgements
None.
Financial Support
This research received no specific grant from any funding agency, commercial, or not-for-profit sectors.
Conflicts of Interest
Dr Declan Aherne, Dr Amanda Fitzgerald, Dr Cian Aherne, Dr Noelle Fitzgerald, Meghan Slattery, and Neal Whelan have no conflicts of interest to disclose.
Ethical Standards
The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional committee on human experimentation with the Helsinki Declaration of 1975, as revised in 2008. The current study is a review paper that does not involve human experimentation. Appropriate ethical standards were upheld throughout whereby research standards were held to the highest quality.
