Training alone doesn’t improve outcomes for depression in primary care


Depression is a major public health problem. We know that there are effective treatments, but getting these successfully implemented in the setting of primary care, where most people with depression are treated, remains a major challenge worldwide, not least in Latin America, where the authors of this review are based.

A key systematic review of the topic from Simon Gilbody and colleagues (Gilbody et al, 2003) concluded that ‘commonly used guidelines and educational strategies are unlikely to be effective’. However a more recent systematic review from (Sikorski et al, 2012 Tomlin, 2012) was more optimistic suggesting that training primary care staff and implementation of guidelines together may improve outcomes for new-onset depression. Both of these reviews are clear that reorganising the structure of care, such as by the introduction of collaborative care models (for example see Coventry et al, 2015; Cristea, 2014), is more likely to be effective than training.

A new review has now been published to evaluate all of the international evidence on healthcare team training programs aimed at improving the outcomes of patients with depression (Vöhringer et al, 2016).

Reorganising the structure of care by introducing collaborative care models has been shown to be more effective than training alone.

Reorganising the structure of care by introducing collaborative care models has been shown to be more effective than training alone.


Three databases were searched for articles in English or Spanish indexed up to November 20,2014 (with no time limit). Clinical trials, meta-analyses, or systematic reviews were included if they evaluated a training or educational program intended to improve the management of depression by primary health care teams, and assessed change in depressive symptoms, diagnosis or response rates, referral rates, patients’ satisfaction and/or quality of life, and the effectiveness of treatments. 


All 9 included studies were randomised controlled trials (RCTs), which included a total of 4,581 participants. Neither of the previous reviews cited above were included.

5 RCTs tested the effectiveness of more complex or multi-component interventions that included clinician training, plus some combination of depression-screening, patient education (provided by trained staff), antidepressant treatment, treatment coordination by a member of staff designated as a ‘case-manager’, and support tools (guides, algorithms) for decision making.

One also included therapist delivery of CBT.

(Note that these share many of the characteristics of collaborative care interventions for depression)

  • All of these studies were effective

The remaining four RCTs evaluated the effectiveness of specific clinician training programs alone.

  • Only 2 of these were effective- (one of which was of training health visitors rather than GPs)

Overall, 7 of the nine studies (78%) showed clinically positive results (a significant decrease in depression scales scores for the active group, when compared with controls.


The authors conclude:

stand-alone training programs are less effective than multi-component interventions.

And, given their particular perspective from Latin America they also note that:

In applying the evidence gathered from developed countries …these training programs must consider and address local conditions of mental health systems, and therefore multi-component interventions may be warranted.

However they also comment that not only was it impossible to isolate and measure the specific contribution that training per se made to the multicomponent interventions, there was also little information about health providers’ prior characteristics and attitudes. This further limits what can be learned in order to generalise the findings.

Finally they also draw attention to the importance of ‘task-shifting’, in these multicomponent interventions. ‘Task shifting’ in this setting occurs when health personnel without formal training in mental health perform tasks usually reserved for specialised personnel.

This research could not isolate and measure the specific contribution that training per se made to the multicomponent intervention.

This research could not isolate and measure the specific contribution that training per se made to the multicomponent intervention.

Strengths and limitations

The authors themselves acknowledge:

  • No priori study protocol
  • For economic reasons articles written in other languages than Spanish (native language of the authors) or English were excluded
  • The databases searched were restricted to PubMed, Embase and the Cochrane Library. PSYCHINFO and CINAHL are missing, which could lead to an exclusion of several studies
  • There was neither a structured critical appraisal of the included studies assessment, nor consideration of possible publication bias among the included studies.

From this author’s perspective, in addition to the above, a major weakness is inclusion of both studies of training and multicomponent interventions (most of which could be better described as collaborative care interventions) in the same review- this does not help to advance the literature. The absence of any reference to existing systematic reviews in this field is also puzzling.

The main strength of the paper is that is does consider what a primarily high-income country literature has to offer to development of training interventions in Low and Middle Income Countries (LMIC).

This 'systematic' review may not have included ALL of the published international evidence, because they didn't look as thoroughly as they could have done. 

This ‘systematic’ review may not have included ALL of the published international evidence, because they didn’t look as thoroughly as they could have done.


The key message of this paper is not new. As noted above, we have known for several years now that simply training primary care workers, particularly general practitioners (GPs), is unlikely to have a significant impact on outcomes for their patients. A major barrier seems to be the barriers that GPs perceive in implementing changes within the confines of the health care systems within which they work (Gask et al, 2005). There is insufficient time (even less in LMIC) to make use of the new psychological skills that we do know that it is possible for them to acquire (Gask et al, 1998). There is also a lack of specialist support for help with management, advice about prescribing or access to psychological therapy.

As the authors of this paper note, there is also likely to be a wide variation in the pre-existing skills and attitudes of those attending a course. This has led researchers, with some success, in an RCT of training, to ‘assess the readiness’ of those willing to be trained by applying ‘Stages of Change’ theory (Prochaska et al, 1992); and thus tailoring the training accordingly (Shirazi et al, 2009). However clinical outcomes have not been assessed.

What we can now conclude however is that multicomponent interventions, specifically collaborative care models, do have an impact on outcomes (Archer et al, 2012). Questions however remain. In Chile, the LMIC where collaborative care has been most effectively implemented (Araya et al, 2003), this was a public health programme rather than one based in everyday practice. In England some studies (e.g. Coventry et al, 2014) have employed Psychological Wellbeing Practitioners from the Improving Access to Psychological Therapies programme as case managers and others, practice nurses (Buszewicz et al, 2016). However in many countries GPs work alone and there is no obvious person to ‘task shift’ and fill such a role. Finally, as the authors of this paper note, it is not clear what contribution training makes to multicomponent or collaborative care interventions. Some have not included any training component for the doctors, yet this risks them being potentially disengaged from the intervention.

Despite these findings, we will continue to train health professionals because they require the knowledge, attitudes and skills to help people with depression. We simply must be realistic about exactly what training alone can achieve for patients, without additional resources and assistance from others working alongside primary care professionals.

We must be realistic about exactly what training alone can achieve for patients with depression.

We must be realistic about exactly what training alone can achieve for patients with depression.

Conflicts of interest

One of my papers is included in this review.


Primary paper

Vöhringer PA, Castro A, Martínez P et al (2016) Healthcare team training programs aimed at improving depression management in primary care: A systematic review. Journal of Affective Disorders. 31;200:142-7. [PubMed abstract]

Other references

Araya R, Rojas G, Fritsch R. et al (2003) Treating depression in primary care in low-income women in Santiago, Chile: a randomised controlled trial. The Lancet. 361(9362):995-1000.

Archer J, Bower P, Gilbody S, Lovell K. et al (2012) Collaborative care for depression and anxiety problems. The Cochrane Library. 2012 Oct 17.

Buszewicz M, Griffin M, McMahon EM. Et al (2016) Practice nurse-led proactive care for chronic depression in primary care: a randomised controlled trial. The British Journal of Psychiatry. 208(4):374-80. [PubMed abstract]

Coventry P, Lovell K, Dickens, C. et al (2015) Integrated primary care for patients with mental and physical multimorbidity: cluster randomised controlled trial of collaborative care for patients with depression comorbid with diabetes or cardiovascular disease. bmj. 16;350:h638.

Cristea I. (2014) Collaborative care for depression: psychological interventions, alone or in combination with medication, offer additional benefits Mental Elf

Gilbody S, Whitty P, Grimshaw J. et al (2003). Educational and organizational interventions to improve the management of depression in primary care: a systematic review. Jama, 289(23), 3145-3151. [PubMed abstract]

Gask L, Dixon C, May C. et al (2005) Qualitative study of an educational intervention for GPs in the assessment and management of depression. Br J Gen Pract 155(520):854-9. [PubMed abstract]

Gask L, Usherwood T, Thompson H. et al (1998) Evaluation of a training package in the assessment and management of depression in primary care. Medical education 32(2):190-8.

Prochaska JO, DiClemente CC, Norcross JC. (1992) In search of how people change: applications to addictive behaviors. American psychologist 47(9):1102. [PubMed abstract]

Shirazi M, Parikh SV, Alaeddini F. et al (2009) Effects on knowledge and attitudes of using stages of change to train general practitioners on management of depression: a randomized controlled study. The Canadian Journal of Psychiatry. 54(10):693-700. [PubMed abstract]

Sikorski C, Luppa M, König HH. Et al (2012). Does GP training in depression care affect patient outcome?-A systematic review and meta-analysis. BMC health services research. Jan 10;12(1):1.

Tomlin A. (2012) GP training and guidelines implementation improves depression care, but training alone does not help, according to new systematic review Mental Elf

Photo credits


Share on Facebook Tweet this on Twitter Share on LinkedIn Share on Google+