Intended for healthcare professionals

Practice Uncertainties

Does depression screening in primary care improve mental health outcomes?

BMJ 2021; 374 doi: https://doi.org/10.1136/bmj.n1661 (Published 19 July 2021) Cite this as: BMJ 2021;374:n1661
  1. Brett D Thombs, professor1,
  2. Sarah Markham, visiting researcher2,
  3. Danielle B Rice, doctoral student1,
  4. Roy C Ziegelstein, professor3
  1. 1Lady Davis Institute for Medical Research, Jewish General Hospital and McGill University, Montréal, Québec, Canada
  2. 2King’s College London, UK
  3. 3Johns Hopkins University School of Medicine, Baltimore, MD, USA
  1. Correspondence to B D Thombs brett.thombs{at}mcgill.ca

What you need to know

  • International guidelines and practice differ regarding screening for depression; it is not currently recommended in the UK

  • Little high quality evidence is available from primary care settings on the benefits of depression screening in improving mental health outcomes for patients

  • Instead of routinely screening all patients in primary care, engage patients in discussions about their overall wellbeing, including mental health, and be alert to clinical cues that could suggest depression

Depression is usually identified when patients report symptoms or when clinicians recognise them through routine assessment of patient wellbeing. Screening can potentially increase rates of depression recognition. Depression screening involves administering a symptom questionnaire to all patients not known or not suspected of having depression. It is intended to identify symptomatic people who may not otherwise be recognised or seek treatment.12 A cut-off is used to classify positive and negative results, with further assessment of those with positive results, and, as appropriate, management. The Patient Health Questionnaire-9 (PHQ-9) is among the most used depression screening tools in primary care.3

In the UK, the National Institute for Health and Care Excellence (NICE) encourages general practitioners to be alert to possible depression, but not to screen routinely.4 The National Screening Committee recommends against screening.5 Depression screening in general practice was financially incentivised by the UK Quality and Outcomes Framework from 2006 to 2013 but was subsequently removed owing to disappointing results; almost 1000 patients had to be screened for each new depression diagnosis and almost 700 for each new antidepressant prescription.6 In North America, the Canadian Task Force on Preventive Health Care (CTFPHC) recommends against screening,7 whereas the US Preventive Services Task Force (USPSTF) recommends screening all primary care patients “with adequate systems in place to ensure accurate diagnosis, effective treatment, and appropriate follow-up.”8 This is described as, at a minimum, dedicated nursing staff to manage the screening process and protocols for referral to evidence based behavioural treatments. More intensively, it involves components such as dedicated staff training programmes, mental health specialists to conduct assessments, trained therapists, and co-payments for medications.8 Only 3% of US adult ambulatory care visits in 2015 included depression screening, however, even though it has been recommended by the USPSTF since 2009.9

Screening tool cut-off points are typically set to maximise combined sensitivity and specificity, but this does not cover important clinical considerations, such as minimising false positive screens, identifying patients with high symptom levels, or ruling out those who meet diagnostic criteria but have mild symptoms and may be less likely to benefit from treatment.10 Assuming 10% prevalence in a general practice setting with half of patients with depression already recognised without screening, screening 100 patients with the PHQ-9 tool (standard cut-off threshold ≥10) would result in 18/100 (18%) of patients screening positive with 14 (77%) of these false positives (fig 1).3

Fig 1
Fig 1

Screening results assuming 10% prevalence with half of depressed patients already recognised before screening, using a cut-off of 10 or greater on the Patient Health Questionnaire-9. Original calculator tool (http://www.depressionscreening100.com/phq/) is based on Negeri et al. Blue=true positive screens; red=false positive screens; grey=true negative screens; black=false negative screen

It is uncertain if depression screening in primary care, alone or combined with other mental health screens (eg, anxiety), would improve mental health outcomes via better identification and treatment in people with depression who would otherwise go unrecognised.

What is the evidence of uncertainty?

Successful depression screening would require patients without known depression to agree to be screened, identification of a significant number of new cases while limiting false positive screens, and effective treatment of newly identified cases. Trials of screening programmes must determine eligibility and randomly assign before screening, exclude patients already known to have depression or in depression treatment, and provide similar depression care options to patients in screened and unscreened trial arms to avoid conflating screening and management effects.111 A 2008 Cochrane review12 reported that interventions that included depression screening did not reduce depressive symptoms (five randomised trials; standardised mean difference=-0.02, 95% confidence interval -0.25 to 0.20). Only one trial included in the review13 randomly assigned participants not known to have depression to be screened or not screened and appropriately separated screening and treatment effects.

Sources and selection criteria

We reviewed systematic reviews conducted to support depression screening guidelines for adults in general practice or in other populations (women during pregnancy or postpartum, children and youth) by the United Kingdom National Screening Committee, the CTFPHC, and the USPSTF for randomised controlled trials (RCTs) that investigated the effects of depression screening interventions on health outcomes. We then searched for more recent trials via Medline, Medline Epub Ahead of Print, in-process and other non-indexed citations, PsycInfo, Embase, CINAHL, and the Cochrane Registry of Controlled Trials through March 12 2021. Search terms, which included “depression,” “depressive disorder,” “mass screening,” and “screen*” are available online (https://osf.io/ptqdk/). We searched for RCTs that compared outcomes among participants randomly assigned to screening versus no screening. To avoid conflating effects of screening and different treatment options, we limited our analysis to RCTs in which participants in both arms had access to similar options for depression management.111 We excluded trials that compared communication or management strategies among patients with positive depression screens or an established diagnosis of depression, because in practice decisions about screening need to occur before screening results are known.

We identified four additional, more recent, trials that have evaluated depression screening in specific patient groups such as postpartum women,14 patients with osteoarthritis,15 patients after an acute coronary syndrome,16 and post-deployment military personnel.17 These trials reported mixed results or found that mental health symptoms were unimproved among participants randomly assigned to screening; three trials found no differences in mental health symptoms or wellbeing between screened and unscreened participants,131617 One trial reported both results that showed no difference and results that favoured screening,14 and one trial reported results that showed no difference and results that were worse for screened participants.15Table 1 shows trial details.

Table 1

Randomised controlled trials of depression screening interventions

View this table:

We did not identify adequately powered, well conducted trials on the benefits and harms of depression screening in general practice patient populations. The diverse populations and screening approaches used in the trials we identified, along with small sample sizes and methodological limitations in some, result in uncertainty about whether routine screening would reduce depression in general practice.

Is ongoing research likely to provide relevant evidence?

We searched Clinicaltrials.gov, WHO International Clinical Trials Registry Platform, and ISRCTN for ongoing trials. We did not identify any ongoing depression screening trials in any setting that planned to assign people randomly not known to have depression to screening or no-screening conditions and that appropriately separated screening and management.

Research on screening tool accuracy and methods is under way. We are part of an international collaboration (https://www.depressd.ca/) that is aggregating large databases from primary studies on depression screening tool accuracy. One goal of the collaboration is to determine how clinicians might move away from a crude dichotomous screening approach and instead use individualised risk estimates based on actual screening tool scores and individual risk factors (eg, sex, age, or medical comorbidities). Such an approach could increase precision for individual patients. It could also be used to engage patients in shared decision making and to identify appropriate care options, as recommended by NICE better.19

What should we do in light of the uncertainty?

Instead of screening with symptom questionnaires, we encourage clinicians to engage patients in discussions about their overall wellbeing, including mental health.19 Recognise that depression may be a process that takes more than a single consultation to investigate. Be alert to clinical cues that could suggest depression, particularly among patients at risk because of factors such as family or personal history of mental health concerns, including problematic substance use, unexplained medical symptoms, or overly frequent use of medical services.4 These include both somatic cues, such as insomnia, anhedonia, or fatigue, and psychological cues, such as low mood or overly negative thinking. If mental health concerns are reported by a patient or are otherwise identified, provide education about depression and other common mental health conditions, including the different ways that symptoms may be experienced and, when appropriate, discuss different management options.4

As national guidelines differ, clinicians are expected to be aware of and adhere to local guidance regarding screening. Until further evidence becomes available, it will be important to make an informed decision regarding screening in primary care after considering the benefits and harms. Depression screening would require substantial resources. Busy general practitioners must evaluate or refer all patients who have positive screens.1711 Like other types of screening, it can also lead to overdiagnosis or misdiagnosis. Overdiagnosis occurs in depression when people with mild, transient symptoms are diagnosed and treated, but without benefit, because symptoms will subside without intervention. Misdiagnosis can occur if screening leads to some people being diagnosed and treated even though they do not meet diagnostic criteria, including people with symptoms resulting from another health condition.10

Away from the context of screening, depression symptom questionnaires are often used in general practice settings for other purposes. They can be useful for assessing and discussing symptoms among patients who may be unsure if they have depression and for monitoring treatment response among patients with a diagnosis of depression.20

Recommendations for further research

  • Objectives: To test whether different depression screening approaches, with standard or enhanced management options, improve mental health compared with

    • not screening but providing access to the same management options

    • healthcare provider education programmes which would seek to improve depression identification and management. Additionally, education programmes would ideally be tested against no-screening usual care.

  • Design: Clustered pragmatic trials with general practices randomly assigned to screening, non-screening usual care, or healthcare provider and patient education trial arms.

  • Population: All adults in general practice settings without a current diagnosis of depression and not receiving treatment for depression. In addition, screening that targets patients with risk factors (eg, social disadvantage, long term unemployment) may be considered.

  • Interventions:

    • Option 1 (dichotomous screening). Positive and negative results determined using an a priori defined cut-off point. Participants with positive screens are assessed for depression and, if appropriate, receive depression treatment. Treatment may be limited to treatments available in usual care or may include enhanced depression care with staff assistance to ensure accurate diagnosis, guideline consistent treatment, and follow-up.

    • Option 2 (risk based screening). Risk levels are determined by a model using actual screening tool scores and patient characteristics with several intervention options available (eg, watchful waiting, low intensity management option, or high intensity management option) based on risk and shared decision making.

    • Option 3 (education). Depression identification and management education is provided to healthcare providers to attempt to improve identification, communication with patients, and management.

  • Comparison:

    • Option 1 (screening or education compared with no-screening usual care). Participants in comparison trial arm are not screened for depression. Participants identified as possibly depressed via self-report or unassisted recognition by a healthcare professional are assessed for depression, and, if appropriate, receive depression treatment. Management options should be the same as in the intervention arm.

    • Option 2 (screening compared with education). Head-to-head comparison of screening (dichotomous or risk based) and education.

  • Outcome: The effect of depression screening on the severity of depressive symptoms, number of depression cases, suicidal thinking and attempts, and quality of life.

Education into practice

  • How do you ensure that patients know you are able to help them if they are depressed and want to communicate their mental health concerns to you?

  • How would you discuss patients’ wellbeing with them and integrate questions about their mood and experiences that will allow you to evaluate if you should further assess for depression?

  • What local referral resources do you have for patients who would benefit from additional assessment or mental health treatment, and are they accessible to patients with limited resources?

What patients need to know

  • As many as 1 in 10 patients in general practice settings may have depression, and this may be as high as 1 in 5 for patients with some chronic medical conditions

  • Most mild depression symptoms go away quickly without medical attention, but this is not always the case; symptoms that are ongoing and serious enough to affect the ability to enjoy social interactions or take care of home or work responsibilities usually require treatment

  • Using a questionnaire to screen for depression may not improve mental health outcomes compared with clinicians talking to patients about their experiences and concerns to determine if they may be depressed

  • Effective treatments for depression are available. If you are experiencing symptoms that might be related to depression, such as sad mood, difficulty enjoying activities that you normally like, feelings of worthlessness or guilt, fatigue or lack of energy, or changes in your sleep patterns, it is important to discuss these with your healthcare provider

  • Your healthcare provider can discuss with you your symptoms and can help you decide if you would like to undergo treatment, which usually involves taking medication or engaging in psychological therapy. They can also discuss advantages and disadvantages of options, and help you to determine your preferences

How patients were involved in the creation of this article

One of our authors, Sarah Markham, is a patient adviser and a member of the BMJ’s International Patient Panel. She provided input on the article content, including on the need to ensure that patients are informed about the purpose of and evidence on depression screening, possible harms from screening, and the need for education of patients and healthcare providers on depression diagnosis and management. In addition, a patient reviewer kindly reviewed this paper for The BMJ and made similar recommendations regarding the importance of patient education and physician training. We are grateful for their input.

Footnotes

  • This is one of a series of occasional articles that highlight areas of practice where management lacks convincing supporting evidence. The series adviser is Nai Ming Lai, clinical editor. You can read more about how to prepare and submit an Education article on our Instructions for Authors pages: https://www.bmj.com/about-bmj/resources-authors/article-types

  • Competing Interests: We have read and understood the BMJ policy on declaration of interests. SM, DBR, and RCZ have no relevant interests to declare. BDT is chair of the Canadian Task Force on Preventive Health Care, which develops guidelines on prevention in primary care, including on depression screening; he does not serve as a voting Task Force member on depression screening guidelines.

  • Contributors: All authors were responsible for the conception and planning of the content of the article. BDT and DBR were responsible for the database searches. BDT drafted the initial version of the manuscript. All authors provided critical revisions and approved the final manuscript. BDT is guarantor.

  • Funding: BDT was supported by a Tier 1 Canada Research Chair. DBR was supported by a Vanier Canada Graduate Scholarship. No specific funding was sought for this work, and no funders had any role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

References

View Abstract