Depression symptomatology and diagnosis: discordance between patients and physicians in primary care settings

1471-2296-9-1 1471-2296 Research article Depression symptomatology and diagnosis: discordance between patients and physicians in primary care settings Ani Chizobam cani@cdrewu.edu Bazargan Mohsen mobazarg@cdrewu.edu Hindman David dhindman@ladhs.org Bell Douglas dbell@mednet.ucla.edu Farooq A Muhammad drfarooqmd@yahoo.com Akhanjee Lutful luakhanjee@ladhs.org Yemofio Francis fyemofio@aol.com Baker Richard rickbaker4@aol.com Rodriguez Michael mrodriguez@ucla.edu

Family Medicine Department Charles R Drew University of Medicine and Science, 2594 Industryway, Lynwood, CA 90262, USA

Family Medicine Department, Charles R Drew University of Medicine and Science, 5850 South Main Street, Los Angeles, CA 90003, USA

Division of General Internal Medicine, David Geffen School of Medicine, UCLA, 911 Broxton Plaza, Los Angeles, CA 90024, USA

Internal Medicine Department, Charles R Drew University of Medicine and Science, 1807 East 120th Street, Los Angeles, CA 90059, USA

Research Centers in Minority Institutions Charles R Drew University of Medicine and Science, 5850 South Main Street, Los Angeles, CA 90003, USA

Family Medicine Department, David Geffen School of Medicine, UCLA. 10880 Wilshire Blvd. Suite 1800, Los Angeles, CA 90024, USA

BMC Family Practice 1471-2296 2008 9 1 1 http://www.biomedcentral.com/1471-2296/9/1 18173835 10.1186/1471-2296-9-1 04 5 2007 03 1 2008 03 1 2008 2008 Ani et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

Background

To examine the agreement between depression symptoms using an assessment tool (PHQ-9), and physician documentation of the same symptoms during a clinic visit, and then to examine how the presence of these symptoms affects depression diagnosis in primary care settings.

Methods

Interviewer administered surveys and medical record reviews. A total of 304 participants were recruited from 2321 participants screened for depression at two large urban primary care community settings.

Results

Of the 2321 participants screened for depression 304 were positive for depression and of these 75.3% (n = 229) were significantly depressed (PHQ-9 score ≥ 10). Of these, 31.0% were diagnosed by a physician with a depressive disorder. A total of 57.6% (n = 175) of study participants had both significant depression symptoms and functional impairment. Of these 37.7% were diagnosed by physicians as depressed. Cohen's Kappa analysis, used to determine the agreement between depression symptoms elicited using the PHQ-9 and physician documentation of these symptoms showed only slight agreement (0.001–0.101) for all depression symptoms using standard agreement rating scales. Further analysis showed that only suicidal ideation and hypersomnia or insomnia were associated with an increased likelihood of physician depression diagnosis (OR 5.41 P sig < .01 and (OR 2.02 P sig < .05 respectively). Other depression symptoms and chronic medical conditions had no affect on physician depression diagnosis.

Conclusion

Two-thirds of individuals with depression are undiagnosed in primary care settings. While functional impairment increases the rate of physician diagnosis of depression, the agreement between a structured assessment and physician elicited and or documented symptoms during a clinical encounter is very low. Suicidality, hypersomnia and insomnia are associated with an increase in the rate of depression diagnosis even when physician and self report of the symptom differ. Interventions that emphasize the use of routine structured screening of primary care patients might also improve the rate of diagnosis of depression in these settings. Further studies are needed to explore depression symptom assessment during physician patient encounter in primary care settings.

Background

Approximately 5–10% of patients attending primary care settings meet the full DSM-IV diagnostic criteria for major depression 123. Additionally, about 16% of a random sample of primary care patients were estimated to have sub-syndromal depression associated with some functional impairment 4. Numerous studies demonstrate that patients with untreated depression with co-occurring medical illnesses have higher morbidity and mortality than comparable patients who have their depression treated 56. Additionally, depression is associated with marked impairment in psychosocial function, reduced productivity, increased suicide attempts, and increased health care utilization 7.

Primary care settings have become the de facto settings for the treatment of many mental health conditions, and primary care providers are often the sole contacts for more than 50% of patients with mental illness 8910. Minority populations utilize outpatient specialty mental health services for psychiatric symptoms and disorders at much lower rates than non-Hispanic white persons, and are more likely to receive care in general medical settings without seeing a specialist 1112131415. Efforts aimed at increasing the appropriate diagnosis and treatment of depression in minority populations have, however, met with mixed success and depression still goes under-recognized and under-treated, especially in primary care settings 1617181920.

There are many factors that influence the recognition or treatment of depression in primary care settings. Some evidence suggests that depressed patients in primary care settings present often with vague somatic complaints rather than with overt complaints of depression 2122. In contrast, patients consulting specialty mental health providers generally must have sufficiently recognized and acknowledged their depression symptoms to seek and accept care within a specialty metal health context. Additionally, patients may avoid disclosing emotional distress to their physicians for fear of being labeled mentally ill, either because they believe their feelings are part of their medical illness or because they don't want a psychiatric diagnosis recorded in their medical record 2324.

Physician influence on health care delivery is an increasingly important aspect of research inquiry 25262728. Little is known about physician-patient interactions and how these affect depression diagnosis. Some evidence suggests primary care providers may have negative attitudes toward mental health problems and do not feel responsible for managing them, may lack time, consulting skills, supporting resources and may be deterred by the workload of long term treatment and monitoring. Additionally, undiagnosed depression may tend to be in patients with mild or non-functionally impaired states that further mask the emotional burden of the disease in these individuals.

While some studies have examined depression diagnosis in the context of the physician-patient encounters 293031, others have emphasized the importance of physician training 3233. In 2005 Tai-Seale et al examined primary care physician's assessment of elderly patients for depression and found that in only 14% of such visits was an assessment conducted28. Relatively few studies have systematically examined these relationships within the context of patient self-report, physician elicitation and the influence on the agreement between the provider and the patient on the diagnosis of depression. The primary premise for this study is that the high rate of undiagnosed depression observed in primary care settings is anteceded by discordance between patients' depression symptoms and the elicitation of these symptoms by physicians.

Study Objectives

The objective of this study was to examine the agreement between patient self-reported depression symptoms and physician documentation of these symptoms. We then examined how these self-reported symptoms predicted physician diagnosis of depression in primary care patients.

Methods

Study Setting

This study was conducted at two large urban outpatient primary care clinics staffed by 50 attending and resident physicians who treat primarily underserved African American and Hispanic patients. This study represents the practice patterns of all providers at both study sites.

Design

We conducted a prospective study using interviewer administered depression assessment surveys and post clinic visit patient medical record reviews over a 1 year period. Face-to-face interviewer administered depression assessment surveys were conducted with a systematically selected sample of patients. The last patient (most recent arrival) on the waiting list for the clinic was approached for an interview. Patients consenting to participate were then screened for depression using a two-item Patient Health Questionnaire-2 (PHQ-2) 34. Patients scoring 3 or greater were invited to participate in a more in-depth interview using the Patient Health Questionnaire-9 (PHQ-9) 3536. All interviews were conducted by bi-lingual staff using Spanish and English survey instruments. The patient's medical records were reviewed after the clinic visit to record physician documented depression symptoms, diagnosis, and depression care received.

Study protocols were developed, reviewed, and approved by the institutional review boards at both sites and all participants gave written informed consent.

Study Population

Participants were eligible to participate in the study if they screened positive for depression on assessment with the PHQ-2 34 (scored 3 or greater), had no previous diagnosis of depression, were 18 years or older, spoke English or Spanish, and consented to a review of their medical records. All participants had no previously documented diagnosis of depression in their medical record within the past 9 months prior to the interview.

Primary Measures

The primary study outcome measure was physician elicitation and documentation of self-reported depression symptoms. Medical record reviews were conducted to record depression symptoms documented in medical record, diagnosis, and care for depression provided during the clinic visit.

Secondary Measures

Depression Symptoms

Patient Health Questionnaire – 9 (PHQ-9) 3536. The PHQ-9 is a brief, 9-item, patient self-report depression assessment tool specifically developed for primary care settings. The PHQ-9 scores each of the 9 DSM-IV symptoms of depression through patients' self report of each symptom over a 2-week period as Not at all (score = 0), Several Days((score = 1), More than Half the Days (score = 2) and Nearly every day (score = 3), with possible total scores ranging from 0 to 27. The PHQ-9 has demonstrated acceptable reliability, validity, sensitivity, and specificity (PHQ-9 score ≥ 10 has a sensitivity of 88% and a specificity of 88% for major depression). For this study, four sub-measures of depression symptom self-report were constructed from the PHQ-9 responses based on a normal or higher threshold or cut off for responses to each question item.

Low symptom cutoff

PHQ-9 responses were dichotomized as symptom absent (0) or symptom present (1): PHQ-9 question items on which participants scored "0" (Not at all) were recorded as having no symptoms and scores of 1–3 were categorized as positive symptom presence.

High symptom cutoff

For the high symptom cutoff measure we adjusted the threshold for positive symptom presence to include only response scores of 2–3 (More than Half the Days; score = 2) and Nearly every day; score = 3) while participant responses scoring 0 and 1 (Not at all; score = 0 and Several Days; score = 1 were categorized as symptom absent.

Functional Impairment

Impairment resulting from depression symptoms was measured using the additional question item on the PHQ-9 (not one of the nine) which asks participants responding positively to any of the 9 question items on the PHQ-9 how difficult these symptoms have made their social, vocational and interpersonal functioning. A 4 point scale response ranging from "no difficulty at all" to "extremely difficult" was then dichotomized with participants reporting "no difficulty" categorized as having" no impairment" and participants reporting having "somewhat," "very" and "extreme" difficulty respectively categorized as having an" impairment 36."

Depression Diagnosis

Depression diagnosis documented in participants medical records were also abstracted and recorded as "No depression diagnosis" or "Depression diagnosis."

Other Measures

Demographic characteristics of study participants were recorded. Other variables were chosen based on the literature 567, and their suggested association with depression. Co-occurring mental and medical conditions documented in participant medical records or self-reported by participants during the interviews occurring within the past 9 months were recorded.

Statistical Analysis

The relationship between patient's self reported depression symptoms and physician documentation of these symptoms was examined in several ways. First, the frequency and distribution of depression symptoms, demographics, and co-occurring medical conditions were determined. Next, the level of agreement between the four sub-measures of self-reported depression symptoms and physician documentation of these symptoms were compared using Kappa statistics. We utilized the following validated interpretation of Kappa values 37: <0 'poor', 0–0.20 'slight', 0.21–0.40 'fair', 0.41–0.60 'moderate', 0.61–0.80 'substantial', 0.81–1.00 'almost perfect' agreement. For both the low and high symptom cutoffs we also examined the independent relationship between self-reported individual depression symptoms (PHQ-9), demographic characteristics of participants, co-occurring medical conditions and depression severity (PHQ-9 scores) with the diagnosis of depression by physicians. Next, we constructed nine logistic regression models to calculate the association between the each of the self-reported symptoms (using the high cutoff) as the dependent variable and the likelihood of a depression diagnosis by physicians when controlling for demographic characteristics, co-occurring medical conditions and depression severity (PHQ-9 scores).

Results

Patient characteristics

A total of 2321 patients were screened for depression using the PHQ-2 34. Of these, 304 participants screened positive for depression and were enrolled into the study. The mean age for participants was 50.26 years and 62.8% and 24.7% were Latino and African American respectively. About 48.4% of participants had no form of health insurance (Table 1).

Table 1

Characteristics of Study Sample (n = 304)

Variables

Categories

Gender

Male

24.3

Female

230

75.7

Age

≤ 34 years

9.5

35–44 years

20.4

Mean age: 49.72 years

45–54 years

104

34.2

Mean age: Male: 48.05 years

55–64 years

28.0

Mean age: Female: 50.26 years

≥ 65 years

7.9

Ethnicity

Latino/Hispanic

191

62.8

Black/African American

24.7

Caucasian/White

6.6

Other

5.9

Marital Status

Single/Never married

101

33.2

Married/Living with partner

111

36.5

Separated/Divorced

22.7

Widowed

3.9

Employment Status

Employed full time or part time

38.0

Unemployed

29.4

Disabled

12.2

Retired

4.7

Homemaker

15.7

Highest Level of Educational

None

6.3

Less than high school or GED

152

50.0

High school or GED

19.7

College degree or more

19.4

Health Insurance Status

No health insurance coverage

144

48.4

Some source of health insurance

147

47.4

Depression Diagnosis

Depression PHQ-9 Score ≥ 10

229

75.3

Depression Diagnosis by Physician when PHQ-9 ≥ 10

31.0

Depression Diagnosis and Functional Impairment

Depression PHQ-9 Score ≥ 10 and Functional Impairment

175

57.6

Depression Diagnosis by Physician when PHQ-9 ≥ 10 and Functional Impairment

37.7

Clinical Findings

Depression Diagnosis: Physician vs. PHQ-9

75.3% (n = 229) of study participants were depressed by PHQ-9 criteria (cutoff of ≥ 10) 36. Of these 31.0% were diagnosed by physician as depressed. When study participants with functional impairment were selected, 57.6% (n = 175) of study participants were depressed by PHQ-9 criteria (cutoff of ≥ 10) 36, and 37.7% were diagnosed by physicians as depressed.

Co-occurring Medical conditions Physician Diagnosis of Depression: (Table 2)

Table 2

Co-occurring Medical Conditions (n = 229)

Variables

Categories

^aMedical Health Co-occurring

Hypertension

119

52.0

Hyperlipidemia

42.8

Diabetes

32.8

Chronic Pain

20.5

Obesity

23.1

Asthma

6.6

Hypothyroidism

10.5

GERD ^b

19.2

Mean number of co-occurring medical condition = 2.3 (std ± 1.5)

Osteoarthritis

16.6

Headaches

7.4

Anxiety Disorders

6.1

^aDocumented medical record diagnosis ^bGastro Esophageal Reflux Disease

1 (0.3%) participant had diagnosed personality disorder

87.8% (n = 201) of the study participants meeting the PHQ-9 criteria for depression had at least one co-occurring medical condition. Hypertension was the most common condition (52%, n = 119), followed by Hyperlipidemia (42.8%, n = 98), Diabetes (32.8%, n = 75), Obesity (23.1%, n = 53), Chronic pain (20.5%, n = 47), GERD (19.2%, n = 44), Osteoarthritis (16.6%, n = 38), and Headaches (7.4%, n = 17). The mean number of chronic medical conditions was 2.3 (std ± 1.5). About 6.5% (n = 15) of the participants had an identified co-occurring mental health condition, specifically, an anxiety disorder (6.1%, n = 14).

Agreement between patients' self-reported depression symptoms using the PHQ-9 and Physician documentation of depression symptoms (Table 3)

Table 3

Agreement Between Participants Self-Reported Depression Symptoms Using The PHQ-9 And Physician Documentation Of Depression Symptoms (n = 304).

Measure of Agreement [Kappa Coefficient] and Significance

PHQ-9 Low symptom cutoff (n)

PHQ-9 High symptom cutoff (n)

Physician symptom elicitation Frequency

Anhedonia

.002 (224)

.036 (175)

Depressed mood

.001 (228)

.029 (206)

Insomnia or hypersomnia

.018 (213)

.011 (185)

Feeling tired or having little energy?

.003 (224)

.017 (192)

Poor appetite or overeating?

.019 (192)

.047* (155)

Feelings of worthlessness or excessive or inappropriate guilt

.014 (182)

.030 (122)

Diminished ability to think or concentrate, or indecisiveness

-.011 (177)

-.032 (128)

Psychomotor agitation or retardation

-.009 (146)

.006 (77)

Recurrent thoughts of death, suicidal ideation without/with a specific plan or attempt

.027 (55)

.083 (21)

The agreement between depression symptoms and physician clinical assessment was between 0.001–0.101 (slight agreement). Exceptions were however observed with the agreement for the "Diminished ability to think, concentrate, or indecisiveness" symptom which showed negative correlations across depression symptom sub-measures (-.011, -.032, -.015 and -.043 respectively, indicating poor agreement). In addition, the psychomotor agitation or retardation symptom for both low symptoms cutoffs 1 and high symptom cutoff 2 also had negative kappa values (-.009 and -.014 respectively indicating poor agreement). Results from both study sites were similar and the coefficient results are indicative of practice at either study site.

Predictors of Depression Diagnosis by Physicians (Table 4)

Table 4

Independent and Adjusted association between, Physician Depression Diagnosis and PHQ-9 elicited Depression symptoms.

Variables

UOR (95% CI)

≠ AOR (95% CI)

Anhedonia

1.64 [0.92–2.94]

1.66 [0.90–3.05]

Depressed mood

1.44 [0.73–2.83]

1.57 [0.77–3.20]

Insomnia or hypersomnia

1.78 [0.10–3.19]

*2.02 [1.09–3.75]

Feeling tired or having little energy

1.57 [0.86–2.86]

1.57 [0.84–2.93]

Poor appetite or overeating?

*1.74 [1.02–3.00]

1.74 [0.99–3.05]

Feelings of worthlessness or excessive or inappropriate guilt

1.35 [0.80–2.29]

1.43 [0.83–2.47]

Diminished ability to think or concentrate, or indecisiveness

1.68 [0.99–2.85]

1.62 [0.94–2.80]

Psychomotor agitation or retardation

1.45 [0.81–2.61]

1.56 [0.85–2.86]

Recurrent thoughts of death, suicidal ideation without/with a specific plan or attempt

**5.47 [2.06–14.49]

**5.41 [1.92–15.25]

*Sig. at <.05 ** Sig. at <.01 † Sig. at <.001

≠: Each reported AOR represents the likelihood of depression diagnosis for each depression in nine logistic regression analysis models.

Only self-reported "suicidal ideations or thoughts", and insomnia or hypersomnia were associated with a statistically significant likelihood of depression diagnosis by physicians in participants meeting the PHQ-9 criteria for depression when we controlled for demographic characteristics, and the number of chronic medical conditions. Using the high symptom cutoff individuals with suicidal ideations or thoughts were 5 times more likely to be diagnosed with depression when compared to individuals without these symptoms. [Unadjusted OR = 5.47 (CI = 2.06–14.49) P < .01 and adjusted OR = 5.41 (CI = 1.92–15.26) P < .01]. In addition on adjustment for other independent variables, individuals with insomnia or hypersomnia were twice as likely to be diagnosed by physician with depression when compared to individuals without this symptom [OR = 2.02 (CI = 1.09–3.75) P < .05 The number of chronic medical conditions and demographic characteristics including age, gender, highest level of education and ethnicity had were not statistically significant independent predictors of physician diagnosis of depression in individuals meeting the PHQ-9 criteria for depression.

Discussion

Prevalence of Depression in Study Sample

Of a total of 2321 patients screened for depression using the PHQ-2 about 304 (13%) met the PHQ-2 screening criteria for a high likelihood of depression (sensitivity 83% and specificity 92%). About 75.3% (n = 229) of these enrolled study participants were depressed by PHQ-9 (>10) assessment criteria, representing about 9.8% of the total screened. The observed prevalence (9.8%) of depression in the sample is similar to most reported primary care setting estimates 123.

Physician Diagnosis of Depression

This study observed a similar trend to previous studies, suggesting that depression remains under-diagnosed in most primary health care settings 1617181920. The rate of depression diagnosis by primary care providers observed in the study sample was only 31.0% in study participants meeting the PHQ-9 criteria for depression. This finding similar to the reported diagnosis rate in a recent study by Liu et al 2006 38 though it remains lower than reported in other studies for primary care clinical settings where missed diagnosis is estimated to be approximately 50% 39404142. There is evidence that functional impairment is strongly associated with depression 43 and increased depression severity is associated with an increase in functional impairment 36. Additionally, the presence of functional impairment as a result of depression symptoms is a DSM-IV prerequisite for a depression diagnosis. When functional impairment was considered in addition to the PHQ-9 score as criteria for depression diagnosis, physician diagnosis rates increased to 37.7%. This finding agrees with the previous study that physician sensitivity to depression is increased when there is functional impairment 35. However, this increase (6.7%) still allows for a missed diagnosis of depression in over 60% of depressed individuals presenting to primary care providers.

In addition to the morbidity, cost, and mortality associated with depression, we were particularly interested in this patient population because of the chronic disease profile (Table 2). Knowing that the burden and clinical outcomes from management of many of these conditions is poorer with these co-occurring conditions 56, we are concerned that appropriate diagnosis of depression should occur during clinical encounters with patients confronting multiple medical disorders. While the balancing of competing demands has previously been cited as a reason for physicians not exploring depressive symptoms and diagnosis during a clinic visits 4445 the number of chronic medical conditions did not affect physician diagnosis of depression in this study (Table 4).

Agreement between self-reported and physician documentation of depression symptoms

Across all depression symptoms the prevalent pattern of low agreement between self reported (PHQ-9) and physician elicited symptoms can be interpreted in several ways (Table 3). Specifically they include: 1) Non-documentation of symptoms by physicians: most patients with depression in primary care settings do not necessarily relate their depression symptoms as a mental health condition 2122, The role of the physician during a clinical encounter is to elicit and relate these symptoms to a diagnosis of depression when it exists. Elicitation of symptoms of depression, however, should logically antecede an appropriate diagnosis; our study demonstrates that when self reported symptoms exists (PHQ-9), physician documentation of each individual symptom is remarkably low. It is possible to suggest that deficiency in a comprehensive assessment for depression might account for some of the observations as suggested by Tai-Seale et al 2005 28. 2) Non disclosure by patients of symptoms to physicians: Patients are sometimes unwilling to disclose emotional distress to their physicians for fear of being stigmatized2324 or are unwilling to accept a diagnosis of mental illness. While this study did not evaluate the interaction between providers and physicians during the clinic encounter, it is probable that participants disclosing depression symptoms to study interviewers will be likely to do the same with their physicians if asked. It is noteworthy to mention that the initiation and continuity of care for depression is not based on empirical biochemical markers. Current practice guidelines require empirical clinical assessment and frequent reassessments of depression symptoms as the basis for diagnosis and monitoring of treatment effectiveness. So, while physician elicitation of depression symptoms observed in this study might be suggestive of either the lack of symptom elicitation or documentation, either practice represents deficiencies in appropriate care.

Suicidal ideations, Insomnia or Hypersomnia and Physician Depression Diagnosis

Suicide is the ninth leading cause of death in the US 46. About 40% of individuals who successfully commit suicide had some contact with their primary care physician within a month of committing suicide 47. While the risk for suicide in patients with mood disorders is about 15% 48, too often physicians are unaware of their patient's suicidal ideation 49. Two findings relating to suicidal ideations in this study were 1) There was a low agreement between self reported suicidal ideations even when more stringent criteria were utilized. Secondly, even when physicians did not elicit or document suicidal ideations, the presence of these symptoms (PHQ-9) predicted a five fold increase in the likelihood of a diagnosis of depression by physicians. The increased emphasis on the recognition by physicians of suicidality in primary care patients has become an often touted tool for preventing suicides. This study suggests that physician's index of suspicion for depression increases with the presence of suicidality (PHQ-9) even when this symptom is not detected directly. Several explanations might be responsible for this observation. First, the PHQ-9 incorporates both active and passive suicidal ideation whereas primary care physicians tend to document only active suicidal ideation. Secondly, some index of distress associated with suicidality might be present in these individuals to alert physicians even when suicidality isn't recognized directly. The results showing that insomnia or hyperinsomnia is associated with an increase in physician diagnosis of depression is additionally interesting even when agreement on this symptom is low. It is suggestive that sleep disturbances promote depression assessment and diagnosis even when physicians don't document the presence of this symptom. Further studies are needed to elucidate these relationships.

Conclusion

Two-thirds of individuals with depression are undiagnosed in primary care settings and the agreement between a structured assessment and physician elicited and or documented symptoms during a clinical encounter is very low. Interventions that emphasize the use of routine structured screening of primary care patients might also improve the rate of symptom recognition. Further studies are needed to explore depression symptom assessment during physician patient encounter in primary care settings. Limitations to the findings of this study are; the use of medical record reviews in determining physician elicitation of depression symptoms may not be representative of physician knowledge of depression symptoms, and may only reflect physician medical record keeping practices. In an attempt to address this issue, the study was conducted with over 50 attending and resident physicians. No provider was excluded from the study. Additionally, while it is assumed that patients who were willing to disclose depression symptoms to study staff would be equally likely to acknowledge these symptoms to their primary care providers, it is also possible the experience of disclosing these symptoms to the study staff immediately prior to their clinic visit was a sufficiently cathartic experience, resulting in decreased patient motivation to disclose these symptoms to their physician. In addition while physicians at both study sites have a retinue of language interpreters available for clinic encounters, language as a predictor of the observed study results was not examined by this study. Further studies to examine the influence of provider language barriers on appropriate diagnosis will be beneficial. Finally the study results may be atypical since they represent practice in only two primary care clinics.

Competing interests

The author(s) declare that they have no competing interests.

Authors' contributions

All authors have read and approved the final manuscript

CA: Corresponding Author

Data collection, methods and analysis

MB.

Data analysis

DH.

Data collection, background

Data analysis and methods

MAF.

Study design and conclusions.

LA.

Study design and conclusions.

FY.

Methods and conclusion

Data collection and methods

Data collection, and methods

Acknowledgements

Data Collection was funded by the Health Resources and Services Administration (HRSA), Health Communities Access Project (HCAP). Analysis and manuscript development were supported by the Agency for Healthcare Research and Quality (1R24-HS014022-01A1), National Center for Research resources (G12-RR03026), and the national center on Minority health and health Disparities (P20MD000148).

Depression in Primary Care. Detection and Diagnosis U S Department of HHS Public Health Service. 1993 Volume 1 pp. 23 75 Diagnosis, treatment, comorbidity, and resource utilization of depressed patients in a general medical practice Luber MP Hollenberg JP Williams-Russo P DiDomenico TN Meyers BS Alexopoulos GS Charlson ME Int J Psychiatry Med 2000 30 1 1 13 10.2190/YTRY-E86M-G1VC-LC79 10900557 Clinical guidelines for establishing remission in patients with depression and anxiety Ballenger JC J Clin Psychiatry 1999 60 Suppl 22 29 34 10634353 Depressive disorders in primary care: prevalence, functional disability, and identification Williams JW Jr. Kerber CA Mulrow CD Medina A Aguilar C J Gen Intern Med 1995 10 1 7 12 10.1007/BF02599568 7699487 Relationship between depression and other medical illnesses Roose SP Glassman AH Seidman SN Jama 2001 286 14 1687 1690 10.1001/jama.286.14.1687 11594878 [Depression and associated organic pathologies, a still under-estimated comorbidity. Results of the DIALOGUE study] Consoli SM Presse Med 2003 32 1 10 21 12610390 Cost of Lost Productive Work Time Among US Workers With Depression Stewart WF Ricci JA Chee E Hahn SR Morganstein D 2003 289 23 3135 3144 12813119 The influence of depression on physician-patient interaction in primary care Callahan EJ Bertakis KD Azari R Robbins J Helms LJ Miller J Fam Med 1996 28 5 346 351 8735062 Family physician beliefs about cancer screening. Development of a survey instrument Montano DE Manders DB Phillips WR J Fam Pract 1990 30 3 313 319 2307945 The de facto US mental health services system: a public health perspective Regier DA Goldberg ID Taube CA 1978 35 6 685 693 306803 Ethnic populations in public mental health: services choice and level of use Hu TW Snowden LR Jerrell JM Nguyen TD Am J Public Health 1991 81 11 1429 1434 1405672 1951799 Utilization of health and mental health services by Los Angeles Mexican Americans and non-Hispanic whites Hough RL Landsverk JA Karno M Burnam MA Timbers DM Escobar JI Regier DA Arch Gen Psychiatry 1987 44 8 702 709 3632245 Filters on the pathway to mental health care, II. Sociodemographic factors Gallo JJ Marino S Ford D Anthony JC Psychol Med 1995 25 6 1149 1160 8637945 Mental Health Culture, Race, and Ethnicity—A Supplement to Mental Health: A Report of the Surgeon General. US Department of Health and Human Services PHS 2001 Characteristics of patients with major depression who received care in general medical and specialty mental health settings Cooper-Patrick L Crum RM Ford DE Med Care 1994 32 1 15 24 10.1097/00005650-199401000-00002 8277799 The epidemiology of major depressive disorder: results from the National Comorbidity Survey Replication (NCS-R) Kessler RC Berglund P Demler O Jin R Koretz D Merikangas KR Rush AJ Walters EE Wang PS Jama 2003 289 23 3095 3105 10.1001/jama.289.23.3095 12813115 Diagnostic patterns in Latino, African American, and European American psychiatric patients Minsky S Vega W Miskimen T Gara M Escobar J Arch Gen Psychiatry 2003 60 6 637 644 10.1001/archpsyc.60.6.637 12796227 Depressive symptoms and psychiatric distress in low income Asian and Latino primary care patients: prevalence and recognition Chung H Teresi J Guarnaccia P Meyers BS Holmes D Bobrowitz T Eimicke JP Ferran E Jr. Community Ment Health J 2003 39 1 33 46 10.1023/A:1021221806912 12650554 The Behavioral Model for Vulnerable Populations: application to medical care use and outcomes for homeless people Gelberg L Andersen RM Leake BD Health Serv Res 2000 34 6 1273 1302 1089079 10654830 Health care utilization among homeless adolescents and young adults Barkin SL Balkrishnan R Manuel J Andersen RM Gelberg L J Adolesc Health 2003 32 4 253 256 10.1016/S1054-139X(02)00474-3 12667728 Early detection and management of mental disorders Walters P TAWPA Chichester , John Wiley Maj M LIJSNSMOA 2005 Somatized depression in primary care attenders Aragones E Labad A Pinol JL Lucena C Alonso Y J Psychosom Res 2005 58 2 145 151 10.1016/j.jpsychores.2004.07.010 15820842 Awareness, diagnosis, and treatment of depression Goldman LS Nielsen NH Champion HC J Gen Intern Med 1999 14 9 569 580 1496741 10491249 10.1046/j.1525-1497.1999.03478.x The underrecognition and undertreatment of depression: what is the breadth and depth of the problem? Davidson JR Meltzer-Brody SE J Clin Psychiatry 1999 60 Suppl 7 4 9; discussion 10-1 10326869 Primary care physician perspective on management of patients with depression Ford DE J Manag Care Pharm 2006 12 2 Suppl 10 13 16623602 What drives referral from primary care physicians to mental health specialists? A randomized trial using actors portraying depressive symptoms Kravitz RL Franks P Feldman M Meredith LS Hinton L Franz C Duberstein P Epstein RM J Gen Intern Med 2006 21 6 584 589 1924631 16808740 10.1111/j.1525-1497.2006.00411.x Roles and practices of general practitioners and psychiatrists in management of depression in the community Tardieu S Bottero A Blin P Bohbot M Goni S Gerard A Gasquet I BMC Fam Pract 2006 7 5 1388221 16445855 10.1186/1471-2296-7-5 Understanding primary care physicians' propensity to assess elderly patients for depression using interaction and survey data Tai-Seale M Bramson R Drukker D Hurwicz ML Ory M Tai-Seale T Street R Jr. Cook MA Med Care 2005 43 12 1217 1224 10.1097/01.mlr.0000185734.00564.c1 16299433 Discordance between self-reported symptom severity and psychosocial functioning ratings in depressed outpatients: implications for how remission from depression should be defined Zimmerman M McGlinchey JB Posternak MA Friedman M Boerescu D Attiullah N Psychiatry Res 2006 141 2 185 191 10.1016/j.psychres.2005.05.016 16499976 Unwritten rules of talking to doctors about depression: integrating qualitative and quantitative methods Wittink MN Barg FK Gallo JJ Ann Fam Med 2006 4 4 302 309 1522158 16868233 10.1370/afm.558 The level of recognition of physical symptoms in patients with a major depression episode in the outpatient psychiatric practice in Puerto Rico: an observational study Tamayo JM Roman K Fumero JJ Rivas M BMC Psychiatry 2005 5 28 1183226 15967039 10.1186/1471-244X-5-28 Effectiveness of an educational strategy to improve family physicians' detection and management of depression: a randomized controlled trial Worrall G Angel J Chaulk P Clarke C Robbins M 1999 161 1 37 40 1232647 10420862 Improving the recognition and management of depression: is there a role for physician education? Gerrity MS Cole SA Dietrich AJ Barrett JE J Fam Pract 1999 48 12 949 957 10628575 The Patient Health Questionnaire-2: validity of a two-item depression screener Kroenke K Spitzer RL Williams JB Med Care 2003 41 11 1284 1292 10.1097/01.MLR.0000093487.78664.3C 14583691 Validation and utility of a self-report version of PRIME-MD: the PHQ primary care study. Primary Care Evaluation of Mental Disorders. Patient Health Questionnaire Spitzer RL Kroenke K Williams JB Jama 1999 282 18 1737 1744 10.1001/jama.282.18.1737 10568646 The PHQ-9: validity of a brief depression severity measure Kroenke K Spitzer RL Williams JB J Gen Intern Med 2001 16 9 606 613 1495268 11556941 10.1046/j.1525-1497.2001.016009606.x The measurement of observer agreement for categorical data Landis JR Koch GG Biometrics 1977 33 1 159 174 10.2307/2529310 843571 Depression diagnosis and antidepressant treatment among depressed VA primary care patients Liu CF Campbell DG Chaney EF Li YF McDonell M Fihn SD Adm Policy Ment Health 2006 33 3 331 341 10.1007/s10488-006-0043-5 16755394 Medical costs attributed to depression among patients with a history of high medical expenses in a health maintenance organization Henk HJ Katzelnick DJ Kobak KA Greist JH Jefferson JW Arch Gen Psychiatry 1996 53 10 899 904 8857866 Are Anxiety and Depression Addressed in Primary Care Patients With Chronic Obstructive Pulmonary Disease? A Chart Review Roundy K Cully JA Stanley MA Veazey C Souchek J Wray NP Kunik ME Prim Care Companion J Clin Psychiatry 2005 7 5 213 218 1257405 16308576 Is this patient clinically depressed? Williams JW Jr. Noel PH Cordes JA Ramirez G Pignone M Jama 2002 287 9 1160 1170 10.1001/jama.287.9.1160 11879114 Nondetection of depression by primary care physicians reconsidered Coyne JC Schwenk TL Fechner-Bates S Gen Hosp Psychiatry 1995 17 1 3 12 10.1016/0163-8343(94)00056-J 7737492 Socioeconomic burden of subsyndromal depressive symptoms and major depression in a sample of the general population Judd LL Paulus MP Wells KB Rapaport MH 1996 153 11 1411 1417 8890673 Barriers to initiating depression treatment in primary care practice Nutting PA Rost K Dickinson M Werner JJ Dickinson P Smith JL Gallovic B J Gen Intern Med 2002 17 2 103 111 1495010 11841525 10.1046/j.1525-1497.2002.10128.x The role of competing demands in the treatment provided primary care patients with major depression Rost K Nutting P Smith J Coyne JC Cooper-Patrick L Rubenstein L Arch Fam Med 2000 9 2 150 154 10.1001/archfami.9.2.150 10693732 Report of Final Mortality Statistics, 1995 Anderson RN KKD Murphy SL Hyattsville, Md , National Center for Health Statistics 1995 23 33 Contact With Mental Health and Primary Care Providers Before Suicide: A Review of the Evidence Luoma JB Martin CE Pearson JL 2002 159 6 909 916 12042175 Assessment of suicide risk Hawton K Br J Psychiatry 1987 150 145 153 3307975 Identification of suicide risk factors using epidemiologic studies Moscicki EK Psychiatr Clin North Am 1997 20 3 499 517 10.1016/S0193-953X(05)70327-0 9323310

Pre-publication history

The pre-publication history for this paper can be accessed here:

http://www.biomedcentral.com/1471-2296/9/1/prepub