Intended for healthcare professionals

Clinical Review State of the Art Review

Depression in primary care: part 1—screening and diagnosis

BMJ 2019; 365 doi: (Published 08 April 2019) Cite this as: BMJ 2019;365:l794
  1. Erin K Ferenchick, assistant clinical professor of medicine1,
  2. Parashar Ramanuj, consultant psychiatrist, senior research fellow23,
  3. Harold Alan Pincus, professor and vice chair, co-director, senior scientist456
  1. 1Center for Family and Community Medicine, Columbia University Medical Center, New York, NY, USA
  2. 2Royal National Orthopaedic Hospital
  3. 3RAND Europe
  4. 4Department of Psychiatry, Columbia University, New York State Psychiatric Institute, New York, NY, USA
  5. 5Irving Institute for Clinical and Translational Research, Columbia University, New York, NY, USA
  6. 6RAND Corporation, Pittsburgh, PA, USA
  1. Correspondence to: E K Ferenchick ef2015{at}


Depression is a common and heterogeneous condition with a chronic and recurrent natural course that is frequently seen in the primary care setting. Primary care providers play a central role in managing depression and concurrent physical comorbidities, and they face challenges in diagnosing and treating the condition. In this two part series, we review the evidence available to help to guide primary care providers and practices to recognize and manage depression. In this first of two reviews, we outline an approach to screening and diagnosing depression in primary care that evaluates current evidence based guidelines and applies the recommendations to clinical practice. The second review presents an evidence based approach to the treatment of depression in primary care, detailing the recommended lifestyle, drug, and psychological interventions at the individual level. It also highlights strategies that are being adopted at an organizational level to manage depression more effectively in primary care.


Depression is an important health problem often seen in primary care. More than eight million doctors’ visits each year in the US are for depression, and more than half of these are in the primary care setting.1 Similarly, in the UK, more than a third of all visits with a general practitioner are estimated to involve a mental health component, with 90% of patients receiving treatment and care for their mental health solely in a primary care setting.2 Depression is a broad and heterogeneous condition. In this article, we provide an epidemiologic framework for depression in primary care and review the current evidence on how to most effectively screen for and diagnose major depressive disorder in the primary care setting, acknowledging that efforts to identify and treat depression are limited without adequate support systems in place.

Sources and selection criteria

To collect and review the evidence available on depression in primary care, we searched the PubMed and Cochrane databases for relevant articles published between 1 January 2007 and 30 September 2017 by using medical subject headings (MeSH) terms including “depressive disorder”, “depression”, and “major depressive disorder” in combination with “primary care” and “general practice”. We did individualized searches for specific interventions thought relevant through the review process, including the addition of “screening” and “diagnosis” in combination with the search terms noted above. We limited our search to the English language. We reviewed titles and abstracts, excluding publications not relevant to primary care or for which major depressive disorder was not the main focus of the study.

We then used a pragmatic prioritizing framework to determine which studies to include in the review. This incorporated the level of evidence, with systematic reviews and meta-analyses given priority and interventional studies prioritized over observational studies; the recency, with more recent studies prioritized over older studies; and the sample size, with larger studies prioritized over those with smaller sample sizes. Where competing priorities existed—for example, a more recent smaller interventional study with similar objectives to a larger but older randomized controlled trial (RCT)—we assigned priority by consensus between the authors on the basis of methodology, study population, aims of the study, and outcome measures.

In addition, we searched references of included studies for further relevant studies. If a study had been conducted before our search timeframe of the past 10 years, we considered it for inclusion by virtue of its size or methodology or owing to absence of other research conducted in the area. We made reference to older, but noteworthy, validation studies as appropriate.


The lifetime prevalence and course of depression vary across countries and regions,3 but the overall high prevalence and persistence of major depression globally confirm the growing worldwide importance of this disorder. The World Health Organization (WHO) estimates that more than 300 million people, or 4.4% of the world’s population, have depression, and the total number of people living with depression increased by 18.4% between 2005 and 2015.4 In 2015 in the US alone, 16.1 million adults reported at least one major depressive episode in the previous year, representing 6.7% of all US adults.5 Similarly, in England, an estimated 4-10% of people will experience depression in their lifetime.6

Depression is associated with a combination of genetic, environmental, biological, cultural, and psychological factors. Meta-analysis from six high quality family studies estimates the heritability of depression to be around 37% (95% confidence interval 31% to 42%).7 This is much lower than for other mental disorders,8 suggesting that most depression at a population level can be explained by environmental factors.

Depression can occur at any age, although it often begins in the second or third decade of life.9 Prevalence rates vary by age, peaking in older adulthood, with an estimated prevalence of 7.5% among women and 5.5% among men aged 55-74 years.4 Population studies have consistently shown major depression to be about twice as common in women as in men.1011 Differing patterns of depression exist within racial and ethnic minorities and are attributable to nativity, socioeconomic status, and the interplay between other protective effects and risk factors.121314

Depression as a comorbidity

Depression has an important bidirectional relation with other chronic physical diseases. These conditions are often worse when depression is present, and depression and chronic conditions have a joint effect on functional disability.15 A large population based study among 245 404 people in 60 countries found that 9-23% of people with one or more chronic physical conditions experienced comorbid depression, compared with 3.2% (3.0% to 3.5%) of people who experienced depression in the absence of other physical conditions.16 A 2007 population based study using data on 30 801 adults from the National Health Interview Survey analyzed the prevalence and odds of major depression and the incremental effect of major depression on health resource utilization, lost productivity, and functional disability in people with common chronic medical disorders. Results included 12 month prevalence and age/sex adjusted odds of major depression by chronic conditions and were as follows: congestive heart failure 7.9% (odds ratio 1.96), hypertension 8.0% (2.00), diabetes 9.3% (1.96), coronary artery disease 9.3% (2.30), chronic obstructive pulmonary disease 15.4% (3.21), and end stage renal disease 17.0% (3.56).17 Similarly, neurological disorders also carry a high risk of depression. A 2018 study from the Neurological Institute of the Cleveland Clinic sought to estimate the prevalence of depression among patients with epilepsy, stroke, and multiple sclerosis. Analysis of 23 000 visits involving 7946 patients in neurology specialty clinics showed an overall point prevalence of depression of 29%. For stroke, epilepsy, and multiple sclerosis, prevalence of depression was 23% (21% to 25%), 33% (31% to 35%), and 29% (28% to 30%), respectively.18

In addition, prescription drugs taken for common chronic medical conditions may cause side effects that contribute to depression. This has been confirmed by multiple studies, including by a recent large cross sectional survey study with 26 192 participants. Adults who were concurrently using three or more drugs that had the potential to cause depression as an adverse effect were more likely to report concurrent depression (8.5%, 5.0% to 12.0%) than those not using any of them (4.5%, 4.0% to 5.0%).19

Personal and economic cost of depression

Depression is associated with increased morbidity and mortality and has a major effect on quality of life for patients and their families.2021 Depression has become the leading cause of disability worldwide,4 resulting in severe impairments that limit the ability to carry out activities of daily living. Depressive disorders are ranked globally as the single largest contributor to non-fatal health loss (7.5% of all years lived with disability).4 In the US, depression accounts for 3.7% of all disability adjusted life years and 8.3% of all years lived with disability.5 People who have depression miss work because of illness at a rate twice that of the general population.22 The financial burden of depression is substantial and rising. Although little research has been done to assess the overall economic burden of depression in Europe, the cost of depression was estimated at £9 billion in England in 2000.23 In the US, the incremental economic burden of people with major depression was $173.2 billion in 2005 and climbed to $210.5 billion in 2010, an increase of 21.5% over this period. A large portion of this increase was attributable to higher direct medical costs and presenteeism, defined as being present but not fully functional in the workplace.24

Screening for depression in primary care

The challenges associated with correctly recognizing depression in primary care have been widely documented and represent a combination of patient, provider, and system related barriers. Importantly, the likelihood of depression being diagnosed correctly is reduced when patients present with somatic complaints.2526 Although some patients with depression may present with the classic symptoms of depressed mood, many patients will report nonspecific symptoms or only somatic complaints in primary care settings. These frequently include changes in appetite, lack of energy, sleep disturbance, general aches and pains, back problems, severe headaches, menstrual symptoms, digestive problems, abdominal pain, and sexual dysfunction. Older patients are less likely to report low mood and often present with only physical complaints or a deterioration in cognitive ability.27 Similarly, some ethnic minorities are more likely to present with nonspecific somatic symptoms.2829 If they mention it at all, patients will often wait until the end of the primary care consultation to share any concerns about depressed mood.30 Thus, primary care providers must consider depression in patients on the basis of a composite profile of common somatic complaints and depressed mood, as well as the presence of any known risk factors for depression,31 which are summarized in box 1.

Box 1

Risk factors for depression

  • Previous episode of depression

  • Age

  • Sex

  • History of other mental illness

  • History of substance use

  • Family history of depression or suicide

  • Chronic medical illness32

  • Unemployment33

  • Poor social support systems34

  • Recent stressful life event that includes loss35

  • Intimate partner’s violence36


Recognizing depression in the primary care setting, particularly in patients with multiple comorbidities, can be difficult. Thus, screening with self reported questionnaires has emerged as an approach to aid primary care providers in identifying patients who may have depression but who do not yet have a diagnosis.

What do guidelines recommend?

Clinical practice guidelines on screening for depression in primary care differ between countries, and general consensus has not been reached. This can present a challenge for primary care providers as they implement screening practices.

In the US, several recommendations support general screening for depression in primary care. The US Preventive Services Task Force (USPSTF) recommends screening for depression in the general adult population, including pregnant and postpartum women, when an adequate system is in place to ensure accurate diagnosis, effective treatment, and appropriate follow-up.37 Similarly, the American Academy of Family Physicians and American College of Preventive Medicine both advocate screening in the general adult population, whereas the Institute for Clinical Systems Improvement recommends that providers screen for depression if it is suspected on the basis of risk factors or presentation.38

By comparison, the UK National Screening Committee (UKNSC) does not recommend screening the general population, and the National Institute for Health and Care Excellence (NICE) guidelines in the UK suggest screening only in those patients with a previous history of depression or significant physical illness.39 The Canadian Task Force on Preventive Health Care (CTFPHC) had previously endorsed screening in primary care, but in 2013 it changed its recommendation to no longer recommend routine screening for depression in adults who are at average risk of depression or in subgroups of the population who may be at increased risk.40

A review published in 2017 compared the guidelines on screening for depression recommended by three leading organizations (CTFPHC, UKNSC, and USPSTF) for consistency and sources of divergence. The authors found that when recommendations diverged, the USPSTF expressed confidence in the benefits of general screening based on indirect evidence, posited potential harms as minimal, and did not consider cost or resource use. By contrast, they noted that the CTFPHC’s and UKNSC’s recommendations against general depression screening focused on the lack of direct evidence of benefit and expressed concerns about both resource use and possible harm to patients, including overdiagnosis and overtreatment.41

Current evidence on screening and how it applies to practice

Recommendations for or against screening should ideally be based on high quality RCTs to assess the effectiveness of the screening itself. A 2005 systematic review of 12 RCTs of the administration of case finding/screening instruments for depression in non-mental health settings found that the routine use of depression screening questionnaires had little effect on the detection of depression (relative risk 1.00, 95% confidence interval 0.89 to 1.13). The intervention rates for those providers who received the screening results were variable, and the overall impact of screening on the management of depression was characterized as minimal (relative risk 1.35, 0.98 to 1.85).42 Similarly, a more recent 2013 pragmatic, cluster randomized trial of 3737 people evaluated the effectiveness of general screening in everyday clinical practice on recognition of depression. No significant differences were seen between recognition rates for the screening intervention group versus the control group (58% v 48.1%; odds ratio 1.40, 0.73 to 2.68). However, suboptimal adherence to implementation of screening throughout the study period may have partially explained the results.43

Large prospective cohort studies have also illustrated attempts to screen without benefit. In particular, a 2009 study evaluated the effectiveness of screening on initiation of treatment among 2005 participants in three high risk groups in primary care in the Netherlands—patients with mental health problems, patients with unexplained somatic complaints, and patients who frequently consult their general practitioner. As a final result of the screening, only 17 patients (1% of the 1687 screened) started treatment for major depressive disorder. Screening for depression in high risk populations was thus deemed to be ineffective. The authors attributed this mainly to low rates of treatment initiation.44

The current evidence suggests that screening alone is not an effective strategy to improve the quality and outcomes of care, but we believe that screening in primary care carries important benefit when primary care practices can support accurate diagnosis, effective treatment, and appropriate follow-up. Systems based interventions that complement screening are necessary for the adequate treatment of depression, just as they are for other chronic diseases. A controlled implementation trial in 1924 participants in four non-metropolitan primary care networks compared collaborative care versus screening and follow-up for patients with diabetes and depressive symptoms. It found that patients had greater 12 month improvements in the Patient Health Questionnaire (PHQ) (mean 7.3 (SD 5.6) compared with 5.2 (5.7) in active controls; P=0.015). Recovery of depressive symptoms (that is, PHQ reduced by 50%) was also greater among intervention patients (61% v 44%; P=0.03). Compared with trial patients, non-screened controls had significantly less improvement at 12 months in the PHQ score (3.2 (SD 4.9)) and lower rates of recovery (24%) (P<0.05 for both).45

For primary care practices that wish to screen for depression, there is little evidence available on the optimal timing for screening, nor on the optimum screening interval. We recommend following the guidance of the USPSTF,37 which promotes a pragmatic approach in the absence of data, including screening all adults who have not been screened previously and using clinical judgment in consideration of risk factors, comorbid conditions, and life events to determine whether additional screening of patients at high risk is warranted.

Although recognizing depression may be a process that often takes more than a single consultation and emerges in a context of ongoing building of trust, validated instruments can help primary care providers to identify and routinely monitor patients with depression. A wide range of screening instruments with variability in style, length, validity, and feasibility for the primary care setting is available. A 2002 systematic review found that median sensitivity across 16 instruments for major depression was 85%, ranging from 50% to 97%, whereas median specificity was 74%, ranging from 51% to 98%.46 A more recent 2018 review examining the psychometric properties of 55 tools or adaptions used for depression screening found an even greater range of sensitivity and specificity values from 28% to 100% and from 43% to 100%, respectively.47

Screening tools

The general consensus is that the PHQ, including both the two item (PHQ-2) and nine item (PHQ-9) questionnaires, meets the criteria for a good screening tool in its validity, reliability, and brevity. Additionally, the PHQ is free and available in the public domain. The diagnostic validity of the PHQ-9 was established almost two decades ago,48 and evidence continues to support its use today. Findings from a meta-analysis of 17 validation studies showed the acceptability of the PHQ-9 in a range of settings, countries, and populations. Across 14 studies with 5026 participants, the PHQ-9 had a sensitivity of 80% (95% confidence interval 71% to 87%) and a specificity of 92% (88% to 95%). The positive likelihood ratio was 10.12 (6.52 to 15.67), and the negative likelihood ratio was 0.22 (0.15 to 0.32). The PHQ-2, however, was validated in only three studies and showed a wider variability in sensitivity.49

Additional studies have further attempted to examine the diagnostic properties of ultra-short screening instruments, including the PHQ-2, as a more practical and efficient option for overburdened and resource constrained primary care practices. A 2007 pooled analysis and meta-analysis of 22 studies evaluated the accuracy of one, two, three, and four item screening instruments for depression in primary care. Pooled analysis of single question tests showed an overall sensitivity of 32.0% and specificity of 97.0%. A one question test identified only three out of every 10 patients with depression in primary care. For two and three item tests, overall sensitivity on pooled analysis was 73.7% and specificity was 74.7%, with a positive predictive value of only 38.3% but a pooled negative predictive value of 93.0%. Ultra-short two or three question tests performed better, identifying eight out of 10 cases. However, the false positive rate was high, with only four out of 10 cases with a positive score actually diagnosed as having depression.50

A more recent large validation study of 2642 participants in primary care practices in Auckland, New Zealand, compared both the PHQ-2 and PHQ-9 with a standard reference interview. The PHQ-2 was found to have a sensitivity of 86% and 61% for threshold scores of ≥2 and ≥3 and specificity of 78% and 92%, respectively. The positive predictive value was 21% for a threshold score of ≥2 and 34% for ≥3. The negative predictive value was 1.2% and 2.7%, respectively. For the PHQ-9, sensitivity and specificity were 74% and 91%, respectively, with a score of ≥10.51

Two step screening

Primary care providers thus face the challenge of identifying the optimal screening approach for their practice. Many have chosen to adopt a two step screening process, first screening with the PHQ-2 and then confirming with the PHQ-9. A 2011 prospective cohort study of 1024 participants evaluated the two step screening recommendation from the American Heart Association for identifying depression in patients with cardiovascular disease to assess the accuracy and prognostic value. Findings showed that this screening method had high specificity (91%, 89% to 93%) but low sensitivity (52%, 46% to 59%) for diagnosing depression. Participants who screened positive on the stepped depression protocol had a 55% greater risk of events than did those who screened negative (age adjusted hazard ratio 1.55, 1.21 to 1.97; P<0.001).52 Results from a more recent study of 562 patients with heart failure who were assessed for depression by using the two step method, however, showed no clear advantages of a stepped method compared with the two item screen alone or the nine item screen alone for predicting adverse prognostic effects of depressive symptoms.53 Given the limited and conflicting evidence on the effectiveness of this stepped approach, additional research is needed to guide clinical practice.

“Help” question

The utility of adding of a “help” question to screening tools has also been explored. A cross sectional validation study of 1025 patients showed that adding a single question about desire for treatment (for example, “Is this something with which you would like help?”) resulted in similar sensitivities but improved both the diagnostic specificity and patient centeredness of depression screening.54 The authors had previously found about five false positive responses for every true positive response for a two item screening test55; however, in this study, the ratio changed from 4.3 to 1.5 when patients responded to either screening question plus the help question. A cohort study of 937 patients in a primary care setting in Switzerland has challenged these findings. The sensitivity and specificity of the two question method alone were 91.3% (81.4% to 96.4%) and 65.0% (61.2% to 68.6%), respectively. Adding the “help” question improved the specificity to 88.2% (85.4% to 90.5%) but decreased the sensitivity to 59.4% (47.0% to 70.9%).56 Although additional studies are needed to further understand the potential benefits of a “help” question as a screening approach, there is likely value in using it to enable ongoing discussions and treatment planning in the primary care setting.

With this in mind, it is important to underscore that screening instruments should be used only to enhance, not replace, the clinical interview. Likewise, they should not be used in isolation for diagnostic purposes. The primary care provider should obtain a complete history of current illness, previous medical history, family history, current drug treatment, and any psychosocial stressors and do a focused physical examination.

Diagnosis of depression in primary care

A positive screen for depressive symptoms should trigger an additional diagnostic assessment. Two classification systems are widely used—the Diagnostic and Statistical Manual of Mental Disorder, 5th edition (DSM-5), developed by the American Psychiatric Association, and the International Statistical Classification of Diseases and Related Health Problems, 11th revision (ICD-11), developed and recently updated by WHO.

To date, no research has compared the effectiveness of DSM-5 and ICD-11 in diagnosing depression. However, the coexistence of DSM-IV and ICD-10 more generally has been much discussed,5758 and one isolated study focused on the differences between the two systems specifically in diagnosing depression. A 2010 study using item response theory to evaluate the DSM-IV and ICD-10 criteria for depressive disorders in 353 people in Japan found slight differences in the severity of depressive disorders between the DSM-IV and ICD-10 diagnostic criteria. ICD-10 was found to be more sensitive to the mild range of the depression spectrum than DSM-IV. Although some variations in severity existed among respondents, most of the respondents diagnosed as having a severe or moderate depressive episode according to the ICD-10 criteria were also diagnosed as having a major depressive episode according to DSM-IV.59 In the absence of overwhelming evidence supporting one classification system over another, we recommend that either may be used for diagnosing depression in the primary care setting.

Although this review focuses on major depressive disorder, depression is a broad and heterogeneous condition. DSM-5 draws a distinction between a range of eight depressive conditions including disruptive mood dysregulation disorder, major depressive disorder, persistent depressive disorder (dysthymia), premenstrual dysphoric disorder, substance/medication induced depressive disorder, depressive disorder due to another medical condition, other specified depressive disorder, and unspecified depressive disorder. Similarly, the ICD-11 classification of depressive disorders includes single episode depressive disorder, recurrent depressive disorder, dysthymic disorder, mixed depressive and anxiety disorder, premenstrual dysphoric disorder, other specified depressive disorder, and unspecified depressive disorder. The common feature shared by these depressive disorders across both classification systems is the presence of sad or empty mood accompanied by somatic and cognitive changes that affect a person’s ability to function, but they are distinct in their duration, timing, and cause.

Formal diagnosis and assessment of severity

A formal diagnosis of major depressive disorder using the DSM-5 criteria requires at least one key symptom (low mood, loss of interest and pleasure, or loss of energy) to be present, whereas the ICD-11 criteria require depressed mood or diminished interest in activities to diagnose a depressive episode. In both, symptoms should be present for at least two weeks and each symptom should be present at sufficient severity for most of every day. DSM-5 requires at least five out of nine symptoms for a diagnosis of depression, whereas the updated ICD-11 classification system does not quantify the number of symptoms needed.

The severity of a patient’s depression should be primarily assessed on the basis of the degree of functional impairment, taking symptom severity into account, rather than being based solely on symptom count. Although this approach makes the grading of severity more subjective, highlighting the distinction is important as evidence based treatment is guided by severity. Both systems classify depressive episodes as mild, moderate, or severe on the basis of the number, type, and severity of symptoms present and degree of functional impairment. Figure 1 summarizes the two classification systems.

Fig 1
Fig 1

DSM-5 and ICD-11 classification systems

Alternate psychiatric diagnoses

Depressive symptoms that do not meet criteria for major depressive disorder in DSM-5 or a single episode or recurrent depressive disorder in ICD-11 may be from other depressive disorders. The classification criteria should be consulted directly for a full description of each depressive disorder to assist the primary care provider in making a distinction and, thus, an accurate diagnosis.

Bipolar disorder

Any patient who presents with symptoms of depression should also be evaluated for possible bipolar disorder. Patients with bipolar disorder are often misdiagnosed as having major depressive disorder, particularly at initial presentation in the primary care setting, as several studies have found that more than a third of these patients remain misdiagnosed for up to 10 years.6061 Although patients with bipolar disorder are more likely to present with low mood rather than mania or hypomania, it is important to elicit any history of abnormally elevated or irritable mood, with persistent increased energy and a noticeable change from baseline behavior (DSM-5).

A 2013 systematic review of seven cross sectional studies measuring prevalence of bipolar disorder in primary care patients with depression or other psychiatric complaints found, through structured interviews, that bipolar disorder likely occurs in 3.4-9.0% of primary care patients with depression, exposure to trauma, medically unexplained symptoms, or a psychiatric complaint. Interestingly, by comparison, 20.9-30.8% of patients had positive results in studies that used screening measures,62 suggesting a high false positive screening rate and limited role for screening for bipolar disorder in the primary care setting. However, as the misdiagnosis of bipolar disorder as major depressive disorder can lead to inappropriate treatment with antidepressants instead of a mood stabilizer, which may induce mania or rapid cycling,63 future research is needed on how to diagnose bipolar disorder accurately in primary care patients presenting with depressive symptoms.

Comorbid psychiatric conditions

Patients with depression may also have comorbid psychiatric conditions, particularly anxiety disorders and substance use disorders (SUD), and the presence of one disorder significantly increases the likelihood that another disorder may also be present.64 In the STAR*D trial, 53.2% of 2876 patients had anxious depression, defined as having a baseline 17 item Hamilton Rating Scale for Depression (HAM-D) anxiety/somatization factor score of 7 or higher. Remission was significantly less likely and took longer to occur in these patients.65 Overall, patients with depression and anxiety are likely have more chronic and recurrent forms of illness.6667 Unrecognized comorbid depression and anxiety has also been found to be associated with an increased rate of both psychiatric hospital admission and suicide attempts.6869

Identifying patients with comorbid anxiety can be challenging for primary care providers, particularly as patients often present with somatic complaints rather than classic psychiatric symptoms. Although a review of the evidence supporting the diagnosis and treatment of anxiety is beyond the scope of this series, we recommend using a screening instrument such as the Generalized Anxiety Disorder (GAD) scale, which has been validated for use in the primary care setting for patients presenting with anxiety or anxiety related symptoms. Findings suggest that the GAD-2, as well as the GAD-7, has an excellent negative predictive value, although only one half of patients with a positive screen actually have generalized anxiety disorder or panic disorder.70 The primary care provider must therefore follow up with a more in-depth diagnostic interview to confirm any diagnosis.

While SUD and major depressive disorder are distinct clinical entities, they are often seen together, with a high prevalence of comorbidity ranging from 8.6% to 25%.71 A 2009 meta-analysis of 74 studies of depression and substance use among people with alcohol use disorders found a positive association of depression with concurrent alcohol use and impairment. Although the effect sizes were small, 60.5% of patients with above average levels of depressive symptoms showed above average levels of current alcohol use and impairment, compared with 39.5% of those with below average levels of depressive symptoms.72

In the STAR*D trial, the demographics and clinical features were compared between those with and without concurrent SUD in 2541 outpatients with depression. Compared with those without SUD, patients with concurrent SUD were more likely to be younger, male, divorced or never married, and at greater current suicide risk and to have an earlier age of onset of depression, greater depressive symptomatology, more previous suicide attempts, more frequent concurrent anxiety disorders, and greater functional impairment (P=0.048 to <0.001).73 Potential SUD should be identified by asking a few screening questions that can be easily integrated into the primary care consultation. Although a comprehensive discussion of SUD in primary care is beyond the scope of this review, the SBIRT (screening, brief intervention, referral, and treatment) model, an evidence based approach to identifying, reducing, and preventing substance misuse, can be effectively implemented in the primary care setting.74

Alternate physical diagnoses

Medical conditions that mimic depression should also be excluded at initial presentation. As no single test, examination, or procedure to easily and effectively differentiate the cause of the presenting symptomatology exists, this can be challenging.75 Four guiding questions have been proposed to aid in establishing a cause and effect relation between psychiatric symptoms and physical findings76:

  1. Is the presentation of the psychiatric symptom atypical?

  2. Is the medical condition or substance use temporally related to the psychiatric symptom?

  3. Are the psychiatric symptoms better explained by a primary psychiatric disorder?

  4. Is the psychiatric presentation a direct consequence of a medical illness or substance use?

As part of a comprehensive assessment, the primary care provider should also evaluate onset and duration of symptoms, changes in drug treatment patterns, and any changes in the patient’s baseline psychosocial functionality. The diagnostic approach to ruling out medical conditions mimicking depression requires critical thinking and clinical diligence.77

Although the aim of this review is not to fully explore the differential diagnosis of all possible physical diagnoses, a few of those commonly encountered in the primary care setting are noted here. Neurological disorders such as Parkinson’s disease, multiple sclerosis, and dementia, in particular, often have symptoms that overlap with those of major depression.78 Patients with cognitive impairment may present with low mood, and those with major depressive disorder may have poor concentration with impaired executive functioning,79 making the clinical picture complex. Further assessment with a neurological examination and cognitive testing should be guided by clinical suspicion. Several of the somatic symptoms of depression, such as fatigue or weight loss, also have the potential to be confused with common medical conditions seen in the primary care setting, such as thyroid disorder or anemia. The clinical interview and physical examination should guide any further diagnostic investigation, including laboratory or other diagnostic studies.

Depression and suicide risk

Once the diagnosis of depression has been made, primary care providers should assess patients for risk of suicide. Depression is a major risk factor for both attempted and completed suicide,80 and a history of self harm attempts, in combination with a history of well developed suicide plans, place the patient at a greater eventual risk of completing a suicide attempt.81 Almost half of adults who complete suicide have had contact with primary care services in the month before death82; so primary care providers have a critical role to play in suicide prevention.

A 2017 systematic review of 21 studies assessed the risk of bias and diagnostic accuracy for suicide and suicide attempt. A meta-analysis of five instruments across the studies showed that none of the instruments reached the predetermined benchmarks (80% sensitivity and 50% specificity) for the suicide outcome or suicide attempt outcome. With few exceptions, low figures were observed for the positive predictive value for the suicide outcome (1-13%).83 This review, however, did not include the diagnostic accuracy of the Suicide Assessment Scale (SUAS) or the Columbia Suicide Severity Rating Scale (C-SSRS). The latter has a strong psychometric evidence base,84 and it has been validated for use in the primary care setting with color coded risk stratification for easy triage. The decision to intervene, however, must ultimately be based on the primary care provider’s clinical assessment of risk.85 In questioning a patient perceived to be at risk, primary care providers should specifically ask about suicide with a focus on suicidal thoughts, plans for suicide, and intent. They should assess the level of risk and define the level of care needed. If any uncertainty exists about either of these levels, consultation with a psychiatrist should be considered.86

Emerging diagnostic approaches

The use of validated screening and assessment rating tools and diagnostic criteria to complement the routine clinical interview is the current expected standard for primary care providers to identify and diagnose depression and track response to treatment. Attempts to standardize measurement based practice have gained significant support, although variability and inaccuracies still exist. Researchers are exploring new technologies that could provide more accurate diagnosis of depression, including genomic, proteomic, and metabolic profiling to identify biomarkers for major depressive disorder.87 Although the applicability of such biological techniques in the primary care setting remains to be seen, other groups have been developing tools for collecting and analyzing passively collected behavioral and cognitive data (for example, through smartphones) as a form of “digital phenotyping.”88


Primary care providers face many challenges as they respond to the growing health needs and psychosocial complexity of their patients. Increasingly, they play a central role in the screening, diagnosis, and treatment of depression. Depression in primary care thus must be managed like any other chronic disease.89Table 1 shows our recommendations. Systems based interventions must enable primary care practices to both screen and accurately diagnose depression in a setting where effective treatment and appropriate follow-up are available.

Table 1

Summary of recommendations

View this table:

Questions for future research

  1. As a research community, how can we unravel the biological and behavioral heterogeneity of depression to improve the diagnosis and treatment of depression in primary care?

  2. What primary prevention interventions are effective in the primary care setting to reduce the incidence of depression?

  3. What is the most effective approach to stratifying the risk of suicide in patients who are depressed and present with suicidal ideation in the primary care setting?

  4. What can be done as a society to reduce the stigma of depression and improve the perceptions of people with depression within the general population?

Patient involvement

We sought feedback from several different sources in compiling and modifying our reviews. Members of the Wellbeing Network Hounslow Community who live with depression in west London gave us feedback on the written manuscripts. Members of the Ealing Primary Care Mental Health Services Service Users Forum also shared insights from their care being transferred from specialist to primary care services in England. Finally, we sought individual and personal insight from a retired GP who has lived with paraplegia from a spinal cord injury for many years as well as with major depressive disorder, which is treatment resistant. The depression preceded the spinal cord injury and is unrelated to it. Their feedback is provided in detail in our second review.91


  • Clinical review, doi: 10.1136/bmj.l835
  • Series explanation: State of the Art Reviews are commissioned on the basis of their relevance to academics and specialists in the US and internationally. For this reason they are written predominantly by US authors

  • Contributors: All three authors made substantial contributions to the conception or design of the work and the acquisition, analysis, or interpretation of data for the work and to drafting the work or revising it critically for important intellectual content; had final approval of the version to be published; and agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

  • Competing interests: We have read and understood the BMJ policy on declaration of interests and declare the following interests: none.

  • Provenance and peer review: Commissioned; externally peer reviewed.