Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Improving the precision of depression diagnosis in general practice: a cluster-randomized trial

Improving the precision of depression diagnosis in general practice: a cluster-randomized trial Background: Methods to enhance the accuracy of the depression diagnosis continues to be of relevance to clini‑ cians. The primary aim of this study was to compare the diagnostic precision of two different diagnostic strategies using the Mini International Neuropsychiatric Interview (MINI) as a reference standard. A secondary aim was to evalu‑ ate accordance between depression severity found via MINI and mean Major Depression Inventory (MDI) sum‑scores presented at referral. Methods: This study was a two‑armed, cluster ‑randomized superiority trial embedded in the Collabri trials inves‑ tigating collaborative care in Danish general practices. GPs performing case‑finding were instructed always to use MDI when suspecting depression. GPs performing usual clinical assessment were instructed to detect depression as they would normally do. According to guidelines, GPs would use MDI if they had a clinical suspicion, and patients responded positively to two or three core symptoms of depression. We compared the positive predictive value (PPV ) in the two groups. Results: Fifty‑ one GP clusters were randomized. In total, 244 participants were recruited in the case‑finding group from a total of 19 GP clusters, and 256 participants were recruited in the usual clinical assessment group from a total of 19 GP clusters. The PPV of the GP diagnosis, when based on case‑finding, was 0.83 (95% CI 0.78–0.88) and 0.93 (95% CI 0.89–0.96) when based on usual clinical assessment. The mean MDI sum‑scores for each depression severity group indicated higher scores than suggested cut‑ offs. Conclusions: In this trial, systematic use of MDI on clinical suspicion of depression did not improve the diagnostic precision compared with the usual clinical assessment of depression. Trial registration: The trial was retrospectively registered on 07/02/2016 at ClinicalTrials.gov. No. NCT02 678845. Keywords: Depression, Identification of depression, Primary Health Care, General practice *Correspondence: nadja.kehler.curth.01@regionh.dk Copenhagen Research Center for Mental Health, Mental Health Center Copenhagen, Gentofte Hospitalsvej 15, 2900 Hellerup, Denmark Full list of author information is available at the end of the article © The Author(s) 2021. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Brinck‑Claussen et al. BMC Fam Pract (2021) 22:88 Page 2 of 8 where the GP always uses a validation test on clinical sus- Background picion of depression, may be as good as high-risk screen- Depression is a common mental disorder in Denmark ing, but it is unclear if case-finding is better than usual [1], and according to the World Health Organization, clinical assessment, where a validation tool is used when the largest contributor to disability worldwide [2]. For the GP finds it appropriate. Therefore, a well-planned patients with depression, the first point of contact with RCT is needed to examine if case-finding is more effec - the health care system is usually in primary care, and tive in finding depression than usual clinical assessment. most patients with depression are treated in this setting without being referred. However, patients with mental Methods disorders often present with somatic symptoms [3] and Objectives social problems, and their depressive symptoms can The primary aim was to determine if case-finding with fluctuate and be mixed with, for example, anxiety symp - the mandatory use of MDI on clinical suspicion was toms, which can make the diagnostic process difficult more accurate than usual clinical assessment in identify- [4]. Research indicates that general practitioners (GPs) ing depression in Danish general practice. The hypoth - identify about 47% of depressed patients [5], and a meta- esis was that case-finding would be more accurate than analysis has found, based on data from a subsample of usual clinical assessment. Accuracy was estimated based 19 studies, that for every 100 unselected cases assessed on the positive predictive value (PPV) in the two groups for depression in primary care, there were more false using the MINI International Neuropsychiatric Interview positives than either identified or missed cases [5]. Early (MINI) [17] as a reference standard. A secondary aim was research failed to show that notification of depression to evaluate the accordance between depression severity status and education of GPs in identifying depression had found via MINI and mean MDI sum-scores. an effect on patient outcomes [6, 7]. Other studies have found early identification of depression as a predictive Design factor of better treatment outcomes [8]. On this back- The study was set up as a two-armed, cluster-randomized ground, enhancing the detection of depression in pri- clinical superiority trial with an intervention group (case- mary care continues to be of importance. Guidelines in finding) and a control group (usual clinical assessment). the US recommend routine screening for depression [9] It was nested within the Danish Collabri trials investi- in contrast to guidelines in the UK and Canada [10]. In gating collaborative care for depression and anxiety dis- Denmark, a depression screening tool is recommended orders in primary care [18–20] (see Fig.  1). The Collabri for use in high-risk groups [11]. However, this is not trials refer to the Collabri studies and Collabri Flex stud- supported by the literature [12–14]. A randomized con- ies. These studies investigated a complex intervention trolled trial (RCT) evaluating the effectiveness of screen - with a multi-professional approach to management and ing for major depressive disorder in high-risk groups in treatment of depression and anxiety. The intervention primary care found no difference in recognition rates in consisted of multiple components such as supervision by the screening group compared to the control group [12]. a psychiatrist and the introduction of a care manager to In a prospective cohort study investigating screening in collaborate with GPs in providing evidence-based care. high-risk groups, only 1% started treatment for major The control group in the Collabri studies received treat - depressive disorder as a result of screening [13]. ment as usual (TAU), whereas the GPs providing care for Systematic identification of patients with depression is the control group participants in the Collabri Flex studies an active ingredient in collaborative care [15], which is could also consult a team of mental health specialists. an effective way of managing depression in primary care [16]. Mandatory use of a diagnostic tool when suspect- Participants ing depression could, therefore, be an appropriate way of GPs in the Capital Region of Denmark (except the Island improving accurate diagnostics of depression in general of Bornholm) were invited to participate in the study. GPs practice. A Danish study investigating high-risk screen- were recruited from May 2014 to July 2015, and patients ing for depression compared with case-finding (use of a were recruited from November 2014 to July 2018. Clus- validation instrument on clinical suspicion of depres- ter-randomization was conducted using a centralized sion) found that screening in high-risk groups had lim- random computer-generated allocation sequence, car- ited effect in addition to case-finding, where the GP used ried out externally by the Research Centre for Prevention Major Depression Inventory (MDI) [14]. However, this and Health. Each cluster corresponded to a GP provider observational study had some limitations because the number. A provider number could include one or more GPs were free to perform either high-risk screening or GPs. First, cluster-randomization for collaborative care case-finding, which was not compared with usual clinical intervention was performed. This randomization was assessment. Based on the above literature, case-finding, Br inck‑Claussen et al. BMC Fam Pract (2021) 22:88 Page 3 of 8 Fig. 1 Flow chart performed in three rounds. The allocation ratio was 1:1 included pregnancy, a dementia diagnosis, and having in the first two rounds and 3:1 (control:intervention) in an unstable somatic condition as determined by the GP. the third, including four GPs. Clusters in each group Additionally, patients of GPs allocated to collaborative were randomized (allocation ratio 1:1) according to the care could not participate if they preferred treatment depression detection method, either case-finding or through the publicly subsidized psychologist program. usual clinical assessment. Many GPs from the Collabri For participants referred from the Collabri studies cur- studies also participated in the following Collabri Flex rent/past (within six months) medical/psychological studies, where they kept their detection allocation. We treatment for anxiety or depression and having a pending planned only to include participants from the Collabri disability pension application were also exclusion crite- studies; however, due to limited participant intake, the ria. The in-and exclusion criteria are updated from earlier recruitment strategy was changed to also include partici- descriptions [18]. pants from the Collabri Flex studies. The GPs who agreed to participate identified poten - Identification of depression tial participants with depression in their practice and In the usual clinical assessment group, GPs were referred them to the Collabri trials. This nested study instructed to diagnose depression as they would normally used data from the referral process and the eligibility do [18]. In accordance with National guidelines, GPs are interviews. Patients were included if the GP had diag- recommended to explore a patient’s condition further nosed depression, if they were 18 years or older, Danish- using the MDI or ICD-10 criteria if at least two core speaking, had given their written consent and did not symptoms of depression, according to ICD-10, are pre- meet any exclusion criteria. Only some of the exclusion sent [21]. In the case-finding group, GPs were instructed criteria from the Collabri and Collabri Flex studies [18, to systematically apply the MDI every time they sus- 20] pertained to this nested study. Exclusion criteria pected depression and to let the MDI result guide the Brinck‑Claussen et al. BMC Fam Pract (2021) 22:88 Page 4 of 8 Statistical methods diagnostics process. Usually, within a week from GP’s Initial sample size calculations showed that a minimum referral, a research assistant conducted the MINI inter- of 480 participants should be included [18]. Based on the view for DSM IV diagnoses [17] and asked ICD-10 study by Mitchell et al., the PPV for usual clinical assess- specific questions to validate the ICD-10 diagnosis of ment can be set at 45% [5]. A clinical possible and mean- depression. Depression severity was assessed using ICD- ingful increase is assessed to be 60%. Thus, we wanted to 10. Assessors were blinded for GP allocation and, in the detect a difference in PPV of 15%, with an alpha of 0.05, a Collabri studies, for the participant’s referral diagnosis. power of 0.8, a cluster-size of 10, and an ICC of 0.04. As the study was nested within the Collabri trials, GPs In the primary analyses, the PPV of GP depression could also refer patients with anxiety disorders who were diagnosis in the two detection groups was calculated by not eligible for this nested study. Assessors were trained constructing two-times-two tables. True positives were in the MINI interview and received ongoing support defined as the number referred by GPs with depression and supervision within the research team. If the MINI and found with depression via MINI interview. A true diagnosis differed from the GP diagnosis, a psychiatrist positive depression could be accompanied by another within the project group would be consulted, and the GP psychopathology found via MINI. However, depression, was contacted to determine the result. In case of discrep- as part of a bipolar disorder, was considered a false posi- ancy, the GP took the final decision. tive. We used STATA version Stata/SE 15.1 to calculate confidence intervals (command “bootstrap”) in the pri - The Major Depression Inventory (MDI) mary and secondary analyses and to perform summary The MDI consists of 12 items, each item scored on a six- statistics in the secondary analyses. point Likert scale. Used as a diagnostic tool, two out of three of the top three items corresponding to ICD-I0 depression diagnosis’s core symptoms must be present Results most of the time or all the time for two weeks. At least Fifty-one GP clusters from the Collabri trials were rand- two of the remaining seven items (two items have subi- omized according to detection method. Twenty-five were tems where only the item with the highest score counts) randomized to usual clinical assessment, and twenty-six corresponding to the accompanying symptoms of the were randomized to case-finding (Fig.  1). We included a ICD-I0 depression diagnosis must be present more than total of 500 participants who were referred by their GP half of the time the same period to suggest depression. with a depression diagnosis; 256 patients were included Hereafter the ICD-10 algorithms establish whether the in the usual clinical assessment group from a total of depression is mild, moderate, or severe [21]. The MDI 19 GP clusters (mean: 13.5, range: 1:24), and 244 were has been validated in an outpatient sample of 43 partici- included in the case-finding group from a total of 19 GP pants showing an acceptable sensitivity ranging from 0.86 clusters (mean 12.8, range 1:29). Of the 500 participants, to 0.92 and a specificity ranging from 0.82 to 0.86 [22], 347 were interviewed using the MINI via telephone and the tool has been used to determine the presence of as this was the mode of administration in the Collabri depression in clinical settings [23, 24]. studies, and 153 were interviewed using the MINI face- to-face as this was the mode of administration in the The MINI International Neuropsychiatric Interview (MINI) Collabri Flex studies. Participants in the randomization The MINI is a short and structured diagnostic interview groups were relatively similar regarding age, gender, and [17]. It has been validated in relation to the Structured severity of depression, according to MINI (Table 1). Clinical Interview for DSM-III-Revised Patients (SCID- The PPV of the depression diagnosis made by GPs in P) with a kappa-value for major depression of 0.83, a the case-finding group was 0.83 (95% CI 0.78–0.88) and sensitivity of 0.96, a specificity of 0.88, a PPV of 0.87, and 0.93 (95% CI 0.89–0.96) in the usual clinical assessment a negative predictive value (NPV) of 0.97 [17]; and in group (Table 2). The confidence intervals did not overlap. regards to the Composite International Diagnostic Inter- This indicates a significant difference in favor of the usual view (CIDI) for International Statistical Classification of clinical assessment group. Thus, findings do not sup - Disease (lCD) with a kappa-value for major depression of port the hypothesis that using MDI on clinical suspicion 0.73, a sensitivity of 0.94, a specificity of 0.79, a PPV of improves the precision of the depression diagnosis in pri- 0.82, and an NPV of 0.93 [25]. mary care, compared to usual clinical assessment, where The MINI was chosen as a reference standard in GPs, if they follow the guidelines, use the MDI or ICD- the present study. Initially, we investigated whether 10 criteria when they have ensured that at least two core it could be managed over the telephone. Details and symptoms of depression are present. In the usual clini- results of this validation are shown in Box  1 in Addi- cal assessment group, 84% of the GPs presented an MDI tional file 1. Br inck‑Claussen et al. BMC Fam Pract (2021) 22:88 Page 5 of 8 Table 1 Baseline characteristics Usual clinical assessment, n = 256 Case-finding, n = 244 Age mean, range 38 (18–81) 40 (18–75) Male gender, n (%) 96 (38) 88 (36) GP‑ diagnosis Depression as primary diagnosis, n (%) 231 (90) 218 (89) Depression as secondary diagnosis, n (%) 25 (10) 26 (11) MDI available, n (%) 215 (84) 214 (89) MDI mean sum, score 32 32 Primary diagnosis after MINI interview, n (%) Mild depression 30 (12) 24 (10) Moderate depression 101 (39) 82 (34) Severe depression 92 (36) 86 (35) Generalized anxiety, panic disorder or Social phobia 22 (9) 26 (11) Other diagnosis 4 (1) 11 (4) No diagnosis 7 (3) 15 (6) Total 256 (100) 244 (100) Abbreviations: GP General practitioner, MDI Major Depression Inventory, MINI Mini International Neuropsychiatric Interview sum-score at referral, corresponding to 89% of the GPs in assessment group followed current guidelines, which the case-finding group. suggest exploring diagnostic criteria in more detail if two Secondary exploratory analyses show that for the group core symptoms of depression were present, the higher assessed with a mild depression via MINI interview, the PPV could indicate that exploring diagnostic criteria only mean MDI sum-score was 26. The group found with when suspecting depression is insufficient. This would moderate depression had a mean MDI sum-score of 31, be a plausible explanation as the prevalence of depres and the group found with severe depression had a mean sion would be higher in a group of patients who the GP MDI sum-score of 36 (Table 3). For the group of partici - suspects have depression and that have at least two core pants found with no depression after MINI, the average symptoms, compared to patients who the GP only sus- MDI sum-score was 30. pects have depression. However, this can only be hypoth- esized because the approach and usage of MDI in the usual clinical assessment group could vary. Discussion A reason for the unexpected result could also be that In this study, case-finding compared to usual clinical GPs in the case-finding group did not fully implement assessment did not improve the precision of the depres- the detection method. GPs in the case-finding group sion diagnosis in primary care. If the usual clinical only enclosed an MDI sum-score for 89% of included cases and not for the expected 100%. Unfortunately, we cannot investigate the reason for the lacking MDI sum- Table 2 Positive predictive value for depression diagnosis based scores from the data available. Possibly GPs did not apply on case‑finding and standard detection the MDI in obvious cases, and greater application of MDI + – Total, n PPV could have given a higher PPV. Research has also shown Depression Depression (95%CI) that GPs could be reluctant to use scales because they feel (MINI), n (MINI), n that the scales do not fit into a fluid conversation, that the Depression 237 19 256 0.93 (0.89– results do not always correspond to their clinical impres diagnose by 0.96) sion, and could even be misleading [4]. GPs in the usual usual clinical assessment clinical assessment group enclosed an MDI sum-score Depression 203 41 244 0.83 (0.78– almost to the same extent (84%) as the GPs in the case- diagnose by 0.88) finding group. Still, the high usage of MDI in both groups case‑finding could not explain the higher PPV in the usual clinical Total 440 60 500 assessment group compared to the case-finding group. Abbreviations: MINI Mini International Neuropsychiatric Interview, PPV Positive Predictive Value Brinck‑Claussen et al. BMC Fam Pract (2021) 22:88 Page 6 of 8 Table 3 MDI sum‑score means according to depression severity Authors found that the MDI was a conservative measure of depression compared to the M-CIDI and a valid tool Depression degree No. of MDIs Mean SD 95% CI for for diagnosing depression when applied to persons who according to MINI mean were suspected to have depression [27]. It was a limita- No depression 46 30.2 7.9 27.9–32.5 tion that we used the MINI as standard reference and Mild depression 58 26.4 7.3 24.5–28.2 not SCID or CIDI. However, we used MINI because it is Moderat depression 164 31.5 6.9 30.4–32.5 less time-consuming and a widely used tool for validating Severe depression 159 36.0 5.7 35.1–36.9 symptoms and diagnoses. Performed MINI interviews Total 427 might not have diagnosed all patients with depression, Note: Two cases were excluded because of missing data on depression degree but if the research assistant’s diagnosis was inconsistent according to MINI with the GP’s diagnosis, the research assistant consulted Abbreviations: MDI Major Depression Inventory, MINI MINI International a psychiatrist in the Collabri group, who contacted the Neuropsychiatric Interview GP to agree on a result. This procedure would, however, Including cases without a diagnosis or other diagnosis not identify false-positive cases if the MINI would diag- nose patients with a depression that did not have depres- sion. The identification of participants relied on the GPs’ The awareness of participating in a study could have judgment, and other recruitment strategies such as iden- influenced the usual behavior of GPs’ in the usual clinical tification through records or waiting room screenings assessment group, but we cannot examine this from the could have identified a different sample of participants on data available. If GPs in the case-finding group applied which the MDI would have been applied. However, the MDI at a lower threshold of suspicion than normally, present strategy of GPs identifying participants is closer they would perhaps detect depression in a sample with to everyday clinical practice. We also cannot rule out a lower prevalence compared to GPs in the usual clini- the possibility that time between tests may have had an cal assessment group. Since PPV depends on prevalence, impact, as the symptoms may have fluctuated, or medical a reduction would also reduce the PPV. The baseline treatment could have been initiated. data did, however, not indicate important differences in We planned to estimate the sensitivity, specificity, and characteristics between groups. Perhaps GPs in the usual NPV of the GP’s depression diagnosis in the case-find - clinical assessment group had other advantages in their ing group. However, as GPs only referred few persons course of detecting depression, e.g., a better prior knowl- believed not to have depression after first suspecting edge of the patient, a different way of dealing with a dif - one, we consider these estimates as unreliable. Thus, less ferential diagnosis, or perhaps they applied MDI with a information about the case-finding method than first different timing along the diagnostic process. anticipated was gained. The mean MDI sum-scores for each depression sever - It can impact the external validity that some exclu- ity group according to MINI indicate higher scores than sion criteria from the Collabri trials also applied to otherwise suggested cut-offs of 21 for mild depression, this nested study. Moreover, the low number of gen- 26 for moderate depression, and 31 for severe depression eral practices participating (51 of 713 invited) is also a [26]. Thus, when using MINI as standard, initial MDI threat to the external validity, as participating GPs could sum-sores presented at referral might have overestimated represent those especially interested in collaborative depression severity in this study sample. However, fur- care or depression detection. Further, if both detection ther studies must be conducted to confirm this finding. groups perform well, it would be more difficult to detect a difference between groups. It is a limitation that we have no baseline information about GPs on factors such Strengths and limitations as sex, age, and years of practice, as these could have Strengths of this trial were the relatively large sample of had an impact on their clinical performance. Addition- participants, the centralized computer-based cluster- ally, there is a risk that the GP sample size is too small randomization, and the use of blinded assessors regard- to avoid significant baseline differences between GPs, ing the allocation. We consider it as a strength of the which could affect the detection practice and thereby, study that the MDI was used as an assessment tool. The the outcome. MDI is a measure already used in general practice and recommended in Danish guidelines [21]. In a study by Nielsen et  al., the validity of the MDI, compared against Comparison with existing literature the Munich-Composite International Diagnostic Inter- The results did not show large problems with false pos- view (M-CIDI), was investigated in a set up comparable itives in either of the studied groups. In comparison, to usual clinical practice in Danish general practice [27]. Br inck‑Claussen et al. BMC Fam Pract (2021) 22:88 Page 7 of 8 Authors’ contributions Mitchell et al. [5] found a PPV of 42.0% (39.6%-44.3%) LFE led trial development as the principal investigator. UØBC drafted the for the GP depression diagnosis across a sample of 19 manuscript’s first version, of which NKC finalized. UØBC, NKC, ASD, CH, ML, studies. In this meta-analysis, depression checklists JHM, MEL, CC, KSC, and MN contributed to the development of the study. Where statistical software was required, CH conducted the analyses. The or questionnaires for GPs  were,  for example, used  to authors approved the final version of the manuscript. assess the diagnosis. Golden standards were often Structural clinical interview for DSM (SCID), Com- Funding This study was funded by a grant from the Danish Ministry of Health to posite International diagnostic interview (CIDI), and strengthen the collaboration between general practice and Mental Health the Diagnostic interview scale (DIS). Christensen Services through shared care. et al. found a true positive rate of 60.5% for a group of Availability of data and materials GPs  using case-finding [14]. Still, in our study, 7% of Data is not available because of the General Data Protection Regulation. Upon patients in the usual clinical assessment group and 17% reasonable request data can be retrieved. of patients in the case-finding group were diagnosed with a depression that was not verified using MINI. Declarations Ethics approval and consent to participate This study was a part of the Collabri and Collabri Flex trials, which were Implications for research and/or practice approved by a scientific ethics committee in the Capital Region of Denmark. Our findings suggest that usual clinical assessment is Reference number H‑3–2013‑203 and H‑16034303. Participants provided informed consent. All methods were performed in accordance with relevant more precise than case-finding in this setting. However, guidelines and regulations. more research is needed to support this result. Identifi - cation of depression might be improved by integrating Consent for publication Not applicable. the case-finding approach into a stepped care model; however, this should be studied further. Optimally, Competing interests the MDI should be used for people with at least two Authors state that they have no competing interests. of three core symptoms of depression. Thus, it would Author details be relevant to examine the precision of the first three Copenhagen Research Center for Mental Health, Mental Health Center questions in the MDI used as a screening tool before Copenhagen, Gentofte Hospitalsvej 15, 2900 Hellerup, Denmark. Research Unit for General Practice, Institute of Public Health, Aarhus University, testing with the full MDI. Bartholins Allé 2, 8000 Aarhus C, Denmark. The Research Unit for General Practice and Section of General Practice, University of Copenhagen, Oester Farimagsgade 5, Postbox 2099, 1014 Copenhagen K, Denmark. Mental Health Center Frederiksberg, Nordre Fasanvej 57‑59, 2000 Frederiksberg, Denmark. Conclusions Stolpegaard Psychotherapy Center, Stolpegaardsvej 20, 2820 Gentofte, Den‑ In this study, routine clinical assessment outperformed 6 7 mark. General Practitioner in Copenhagen, Copenhagen, Denmark. Mental case-finding in the identification of depression in pri - Health Center North Zealand, Dyrehavevej 48, 3400 Hilleroed, Denmark. mary care. Further studies should be conducted to con- Received: 6 November 2020 Accepted: 5 April 2021 firm this finding. Abbreviations CIDI: Composite International Diagnostic Interview; GP: General Practitioner; References ICD: International Statistical Classification of Disease; MDI: Major Depression 1. Olsen LR, Mortensen EL, Bech P. Prevalence of major depression and Inventory; MINI: MINI International Neuropsychiatric Interview; NPV: Negative stress indicators in the Danish general population. Acta Psychiatr Scand. Predictive Value; PPV: Positive Predictive Value; RCT : Randomized Controlled 2004;109:96–103. Trial; SCID‑P: Structured Clinical Interview for DSM ‑III‑Revised, Patients. 2. World Health Organization. Depression and other common mental disorders: global health estimates. Licence: CC BY‑NC‑SA 3.0 IGO. Geneva; Supplementary Information 3. Kroenke K. Patients presenting with somatic complaints: Epidemiology, The online version contains supplementary material available at https:// doi. psychiatric co‑morbidity and management. Int J Methods Psychiatr Res. org/ 10. 1186/ s12875‑ 021‑ 01432‑w. 2003;12:34–43. 4. Davidsen AS, Fosgerau CF. What is depression? Psychiatrists’ and GPs’ experiences of diagnosis and the diagnostic process. Int J Qual Stud Additional file 1: Box 1. Information about the validation of the MINI Health Well‑being. 2014;9:1–10. interview conducted by telephone. 5. Mitchell AJ, Vaze A, Rao S. Clinical diagnosis of depression in primary care: a meta‑analysis. Lancet. 2009;374:609–19. 6. Dowrick C, Buchan I. Twelve month outcome of depression in gen‑ Acknowledgements eral practice: Does detection or disclosure make a difference? BMJ. We thank research assistants and care managers in the Collabri and Collabri 1995;311:1274. Flex studies for conducting MINI interviews. We also thank general practition‑ 7. Thompson C, Kinmonth AL, Stevens L, Peveler RC, Stevens A, Ostler KJ, ers and patients for their participation in the study. et al. Eec ff ts of a clinical‑practice guideline and practice ‑based education on detection and outcome of depression in primary care: Hampshire Depression Project randomised controlled trial. Lancet. 2000;355:185–91. Brinck‑Claussen et al. BMC Fam Pract (2021) 22:88 Page 8 of 8 8. Hung C‑I, Yu N‑ W, Liu C‑ Y, Wu K‑ Y, Yang C‑H. The impact of the duration 20. Curth NK, Brinck‑ Claussen U, Jørgensen KB, Rosendal S, Hjorthøj C, Nor‑ of an untreated episode on improvement of depression and somatic dentoft M, et al. Collaborative care vs consultation liaison for depression symptoms. Neuropsychiatr Dis Treat. 2015;11:2245–52. and anxiety disorders in general practice: study protocol for two rand‑ 9. Siu AL, Bibbins‑Domingo K, Grossman DC, Baumann LC, Davidson KW, omized controlled trials (the Danish Collabri Flex trials). Trials. 2019;20:607. Ebell M, et al. Screening for depression in adults: US preventive services 21. Dansk Selskab for Almen Medicin. Klinisk vejledning for almen praksis. task force recommendation statement. JAMA. 2016;315:380–7. Unipolar depression. Diagnostik og behandling. 2010. https:// www. dsam. 10. Thombs BD, Ziegelstein RC, Roseman M, Kloda LA, Ioannidis JPA. There dk/ files/9/ depre ssion_ med_ links. pdf . Accessed 20 Oct 2020. are no randomized controlled trials that support the United States 22. Bech P, Rasmussen NA, Olsen LR, Noerholm V, Abildgaard W. The sensitiv‑ Preventive Services Task Force Guideline on screening for depression in ity and specificity of the Major Depression Inventory, using the Present primary care: a systematic review. BMC Med. 2014;12:13. State Examination as the index of diagnostic validity. J Aec ff t Disord. 11. Sundhedsstyrelsen. Referenceprogram for unipolar depression hos vok‑ 2001;66:159–64. sne. 2007. https:// www. sst. dk/‑/ media/ Udgiv elser/ 2007/ Publ2 007/ PLAN/ 23. Madsen MT, Zahid JA, Hansen CH, Grummedal O, Hansen JR, Isbrand A, SfR/ SST_ Dep,‑d‑ ,rappo rt,‑d‑ pdf. ashx. Accessed 12 June 2020. et al. The effect of melatonin on depressive symptoms and anxiety in 12. Romera I, Montejo ÁL, Aragonés E, Arbesú JÁ, Iglesias‑ García C, López patients after acute coronary syndrome: the MEDACIS randomized clini‑ S, et al. Systematic depression screening in high‑risk patients attend‑ cal trial. J Psychiatr Res. 2019;119:84–94. https:// doi. org/ 10. 1016/j. jpsyc ing primary care: a pragmatic cluster‑randomized trial. BMC Psychiatry. hires. 2019. 09. 014. 2013;13:83. 24. Okkels N, Jensen LG, Skovshoved LC, Arendt R, Blicher AB, Vieta E, et al. 13. Baas KD, Wittkampf KA, Van Weert HC, Lucassen P, Huyser J, Van Den Lighting as an aid for recovery in hospitalized psychiatric patients: a ran‑ Hoogen H, et al. Screening for depression in high‑risk groups: Prospective domized controlled effectiveness trial. Nord J Psychiatry. 2020;74:105–14. cohort study in general practice. Br J Psychiatry. 2009;194:399–403.https:// doi. org/ 10. 1080/ 08039 488. 2019. 16764 65. 14. Christensen KS, Sokolowski I, Olesen F. Case‑finding and risk ‑ group 25. Lecrubier Y, Sheehan DV, Weiller E, Amorim P, Bonora I, Sheehan KH, et al. screening for depression in primary care. Scand J Prim Health Care. The Mini International Neuropsychiatric Interview (MINI). A short diag‑ 2011;29:80–4. nostic structured interview: reliability and validity according to the CIDI. 15. Gilbody S, House A, Sheldon T. Screening and case finding instruments Eur Psychiatry. 1997;12:224–31. for depression. Cochrane Database Syst Rev. 2005. 26. Bech P, Timmerby N, Martiny K, Lunde M, Soendergaard S. Psychometric 16. Archer J, Bower P, Gilbody S, Lovell K, Richards D, Gask L, et al. Collabo‑ evaluation of the Major Depression Inventory (MDI) as depression sever‑ rative care for depression and anxiety problems (Review). Cochrane ity scale using the LEAD (Longitudinal Expert Assessment of All Data) as Database Syst Rev. 2012. index of validity. BMC Psychiatry. 2015;15:190. 17. Sheehan D, Lecrubier Y, Harnett Sheehan K, Janavs J, Weiller E, Keskiner 27. Nielsen MG, Ørnbøl E, Bech P, Vestergaard M, Christensen KS. The A, et al. The validity of the Mini International Neuropsychiatric Inter‑ criterion validity of the web‑based major depression inventory when view (MINI) according to the SCID‑P and its reliability. Eur Psychiatry. used on clinical suspicion of depression in primary care. Clin Epidemiol. 1997;12:232–41. 2017;9:355–65. https:// doi. org/ 10. 2147/ CLEP. S1329 13. 18. Brinck‑ Claussen U, Curth NK, Davidsen AS, Mikkelsen JH, Lau ME, Lunds‑ teen M, et al. Collaborative care for depression in general practice: Study Publisher’s Note protocol for a randomised controlled trial. Trials. 2017;18:1–12. Springer Nature remains neutral with regard to jurisdictional claims in pub‑ 19. Curth NK, Brinck‑ Claussen UØ, Davidsen AS, Lau ME, Lundsteen M, Mik‑ lished maps and institutional affiliations. kelsen JH, et al. Collaborative care for panic disorder, generalised anxiety disorder and social phobia in general practice: study protocol for three cluster‑randomised, superiority trials. Trials. 2017;18:382. Re Read ady y to to submit y submit your our re researc search h ? Choose BMC and benefit fr ? Choose BMC and benefit from om: : fast, convenient online submission thorough peer review by experienced researchers in your field rapid publication on acceptance support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png BMC Family Practice Springer Journals

Loading next page...
 
/lp/springer-journals/improving-the-precision-of-depression-diagnosis-in-general-practice-a-BK8z1XkI7V
Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2021
eISSN
1471-2296
DOI
10.1186/s12875-021-01432-w
Publisher site
See Article on Publisher Site

Abstract

Background: Methods to enhance the accuracy of the depression diagnosis continues to be of relevance to clini‑ cians. The primary aim of this study was to compare the diagnostic precision of two different diagnostic strategies using the Mini International Neuropsychiatric Interview (MINI) as a reference standard. A secondary aim was to evalu‑ ate accordance between depression severity found via MINI and mean Major Depression Inventory (MDI) sum‑scores presented at referral. Methods: This study was a two‑armed, cluster ‑randomized superiority trial embedded in the Collabri trials inves‑ tigating collaborative care in Danish general practices. GPs performing case‑finding were instructed always to use MDI when suspecting depression. GPs performing usual clinical assessment were instructed to detect depression as they would normally do. According to guidelines, GPs would use MDI if they had a clinical suspicion, and patients responded positively to two or three core symptoms of depression. We compared the positive predictive value (PPV ) in the two groups. Results: Fifty‑ one GP clusters were randomized. In total, 244 participants were recruited in the case‑finding group from a total of 19 GP clusters, and 256 participants were recruited in the usual clinical assessment group from a total of 19 GP clusters. The PPV of the GP diagnosis, when based on case‑finding, was 0.83 (95% CI 0.78–0.88) and 0.93 (95% CI 0.89–0.96) when based on usual clinical assessment. The mean MDI sum‑scores for each depression severity group indicated higher scores than suggested cut‑ offs. Conclusions: In this trial, systematic use of MDI on clinical suspicion of depression did not improve the diagnostic precision compared with the usual clinical assessment of depression. Trial registration: The trial was retrospectively registered on 07/02/2016 at ClinicalTrials.gov. No. NCT02 678845. Keywords: Depression, Identification of depression, Primary Health Care, General practice *Correspondence: nadja.kehler.curth.01@regionh.dk Copenhagen Research Center for Mental Health, Mental Health Center Copenhagen, Gentofte Hospitalsvej 15, 2900 Hellerup, Denmark Full list of author information is available at the end of the article © The Author(s) 2021. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Brinck‑Claussen et al. BMC Fam Pract (2021) 22:88 Page 2 of 8 where the GP always uses a validation test on clinical sus- Background picion of depression, may be as good as high-risk screen- Depression is a common mental disorder in Denmark ing, but it is unclear if case-finding is better than usual [1], and according to the World Health Organization, clinical assessment, where a validation tool is used when the largest contributor to disability worldwide [2]. For the GP finds it appropriate. Therefore, a well-planned patients with depression, the first point of contact with RCT is needed to examine if case-finding is more effec - the health care system is usually in primary care, and tive in finding depression than usual clinical assessment. most patients with depression are treated in this setting without being referred. However, patients with mental Methods disorders often present with somatic symptoms [3] and Objectives social problems, and their depressive symptoms can The primary aim was to determine if case-finding with fluctuate and be mixed with, for example, anxiety symp - the mandatory use of MDI on clinical suspicion was toms, which can make the diagnostic process difficult more accurate than usual clinical assessment in identify- [4]. Research indicates that general practitioners (GPs) ing depression in Danish general practice. The hypoth - identify about 47% of depressed patients [5], and a meta- esis was that case-finding would be more accurate than analysis has found, based on data from a subsample of usual clinical assessment. Accuracy was estimated based 19 studies, that for every 100 unselected cases assessed on the positive predictive value (PPV) in the two groups for depression in primary care, there were more false using the MINI International Neuropsychiatric Interview positives than either identified or missed cases [5]. Early (MINI) [17] as a reference standard. A secondary aim was research failed to show that notification of depression to evaluate the accordance between depression severity status and education of GPs in identifying depression had found via MINI and mean MDI sum-scores. an effect on patient outcomes [6, 7]. Other studies have found early identification of depression as a predictive Design factor of better treatment outcomes [8]. On this back- The study was set up as a two-armed, cluster-randomized ground, enhancing the detection of depression in pri- clinical superiority trial with an intervention group (case- mary care continues to be of importance. Guidelines in finding) and a control group (usual clinical assessment). the US recommend routine screening for depression [9] It was nested within the Danish Collabri trials investi- in contrast to guidelines in the UK and Canada [10]. In gating collaborative care for depression and anxiety dis- Denmark, a depression screening tool is recommended orders in primary care [18–20] (see Fig.  1). The Collabri for use in high-risk groups [11]. However, this is not trials refer to the Collabri studies and Collabri Flex stud- supported by the literature [12–14]. A randomized con- ies. These studies investigated a complex intervention trolled trial (RCT) evaluating the effectiveness of screen - with a multi-professional approach to management and ing for major depressive disorder in high-risk groups in treatment of depression and anxiety. The intervention primary care found no difference in recognition rates in consisted of multiple components such as supervision by the screening group compared to the control group [12]. a psychiatrist and the introduction of a care manager to In a prospective cohort study investigating screening in collaborate with GPs in providing evidence-based care. high-risk groups, only 1% started treatment for major The control group in the Collabri studies received treat - depressive disorder as a result of screening [13]. ment as usual (TAU), whereas the GPs providing care for Systematic identification of patients with depression is the control group participants in the Collabri Flex studies an active ingredient in collaborative care [15], which is could also consult a team of mental health specialists. an effective way of managing depression in primary care [16]. Mandatory use of a diagnostic tool when suspect- Participants ing depression could, therefore, be an appropriate way of GPs in the Capital Region of Denmark (except the Island improving accurate diagnostics of depression in general of Bornholm) were invited to participate in the study. GPs practice. A Danish study investigating high-risk screen- were recruited from May 2014 to July 2015, and patients ing for depression compared with case-finding (use of a were recruited from November 2014 to July 2018. Clus- validation instrument on clinical suspicion of depres- ter-randomization was conducted using a centralized sion) found that screening in high-risk groups had lim- random computer-generated allocation sequence, car- ited effect in addition to case-finding, where the GP used ried out externally by the Research Centre for Prevention Major Depression Inventory (MDI) [14]. However, this and Health. Each cluster corresponded to a GP provider observational study had some limitations because the number. A provider number could include one or more GPs were free to perform either high-risk screening or GPs. First, cluster-randomization for collaborative care case-finding, which was not compared with usual clinical intervention was performed. This randomization was assessment. Based on the above literature, case-finding, Br inck‑Claussen et al. BMC Fam Pract (2021) 22:88 Page 3 of 8 Fig. 1 Flow chart performed in three rounds. The allocation ratio was 1:1 included pregnancy, a dementia diagnosis, and having in the first two rounds and 3:1 (control:intervention) in an unstable somatic condition as determined by the GP. the third, including four GPs. Clusters in each group Additionally, patients of GPs allocated to collaborative were randomized (allocation ratio 1:1) according to the care could not participate if they preferred treatment depression detection method, either case-finding or through the publicly subsidized psychologist program. usual clinical assessment. Many GPs from the Collabri For participants referred from the Collabri studies cur- studies also participated in the following Collabri Flex rent/past (within six months) medical/psychological studies, where they kept their detection allocation. We treatment for anxiety or depression and having a pending planned only to include participants from the Collabri disability pension application were also exclusion crite- studies; however, due to limited participant intake, the ria. The in-and exclusion criteria are updated from earlier recruitment strategy was changed to also include partici- descriptions [18]. pants from the Collabri Flex studies. The GPs who agreed to participate identified poten - Identification of depression tial participants with depression in their practice and In the usual clinical assessment group, GPs were referred them to the Collabri trials. This nested study instructed to diagnose depression as they would normally used data from the referral process and the eligibility do [18]. In accordance with National guidelines, GPs are interviews. Patients were included if the GP had diag- recommended to explore a patient’s condition further nosed depression, if they were 18 years or older, Danish- using the MDI or ICD-10 criteria if at least two core speaking, had given their written consent and did not symptoms of depression, according to ICD-10, are pre- meet any exclusion criteria. Only some of the exclusion sent [21]. In the case-finding group, GPs were instructed criteria from the Collabri and Collabri Flex studies [18, to systematically apply the MDI every time they sus- 20] pertained to this nested study. Exclusion criteria pected depression and to let the MDI result guide the Brinck‑Claussen et al. BMC Fam Pract (2021) 22:88 Page 4 of 8 Statistical methods diagnostics process. Usually, within a week from GP’s Initial sample size calculations showed that a minimum referral, a research assistant conducted the MINI inter- of 480 participants should be included [18]. Based on the view for DSM IV diagnoses [17] and asked ICD-10 study by Mitchell et al., the PPV for usual clinical assess- specific questions to validate the ICD-10 diagnosis of ment can be set at 45% [5]. A clinical possible and mean- depression. Depression severity was assessed using ICD- ingful increase is assessed to be 60%. Thus, we wanted to 10. Assessors were blinded for GP allocation and, in the detect a difference in PPV of 15%, with an alpha of 0.05, a Collabri studies, for the participant’s referral diagnosis. power of 0.8, a cluster-size of 10, and an ICC of 0.04. As the study was nested within the Collabri trials, GPs In the primary analyses, the PPV of GP depression could also refer patients with anxiety disorders who were diagnosis in the two detection groups was calculated by not eligible for this nested study. Assessors were trained constructing two-times-two tables. True positives were in the MINI interview and received ongoing support defined as the number referred by GPs with depression and supervision within the research team. If the MINI and found with depression via MINI interview. A true diagnosis differed from the GP diagnosis, a psychiatrist positive depression could be accompanied by another within the project group would be consulted, and the GP psychopathology found via MINI. However, depression, was contacted to determine the result. In case of discrep- as part of a bipolar disorder, was considered a false posi- ancy, the GP took the final decision. tive. We used STATA version Stata/SE 15.1 to calculate confidence intervals (command “bootstrap”) in the pri - The Major Depression Inventory (MDI) mary and secondary analyses and to perform summary The MDI consists of 12 items, each item scored on a six- statistics in the secondary analyses. point Likert scale. Used as a diagnostic tool, two out of three of the top three items corresponding to ICD-I0 depression diagnosis’s core symptoms must be present Results most of the time or all the time for two weeks. At least Fifty-one GP clusters from the Collabri trials were rand- two of the remaining seven items (two items have subi- omized according to detection method. Twenty-five were tems where only the item with the highest score counts) randomized to usual clinical assessment, and twenty-six corresponding to the accompanying symptoms of the were randomized to case-finding (Fig.  1). We included a ICD-I0 depression diagnosis must be present more than total of 500 participants who were referred by their GP half of the time the same period to suggest depression. with a depression diagnosis; 256 patients were included Hereafter the ICD-10 algorithms establish whether the in the usual clinical assessment group from a total of depression is mild, moderate, or severe [21]. The MDI 19 GP clusters (mean: 13.5, range: 1:24), and 244 were has been validated in an outpatient sample of 43 partici- included in the case-finding group from a total of 19 GP pants showing an acceptable sensitivity ranging from 0.86 clusters (mean 12.8, range 1:29). Of the 500 participants, to 0.92 and a specificity ranging from 0.82 to 0.86 [22], 347 were interviewed using the MINI via telephone and the tool has been used to determine the presence of as this was the mode of administration in the Collabri depression in clinical settings [23, 24]. studies, and 153 were interviewed using the MINI face- to-face as this was the mode of administration in the The MINI International Neuropsychiatric Interview (MINI) Collabri Flex studies. Participants in the randomization The MINI is a short and structured diagnostic interview groups were relatively similar regarding age, gender, and [17]. It has been validated in relation to the Structured severity of depression, according to MINI (Table 1). Clinical Interview for DSM-III-Revised Patients (SCID- The PPV of the depression diagnosis made by GPs in P) with a kappa-value for major depression of 0.83, a the case-finding group was 0.83 (95% CI 0.78–0.88) and sensitivity of 0.96, a specificity of 0.88, a PPV of 0.87, and 0.93 (95% CI 0.89–0.96) in the usual clinical assessment a negative predictive value (NPV) of 0.97 [17]; and in group (Table 2). The confidence intervals did not overlap. regards to the Composite International Diagnostic Inter- This indicates a significant difference in favor of the usual view (CIDI) for International Statistical Classification of clinical assessment group. Thus, findings do not sup - Disease (lCD) with a kappa-value for major depression of port the hypothesis that using MDI on clinical suspicion 0.73, a sensitivity of 0.94, a specificity of 0.79, a PPV of improves the precision of the depression diagnosis in pri- 0.82, and an NPV of 0.93 [25]. mary care, compared to usual clinical assessment, where The MINI was chosen as a reference standard in GPs, if they follow the guidelines, use the MDI or ICD- the present study. Initially, we investigated whether 10 criteria when they have ensured that at least two core it could be managed over the telephone. Details and symptoms of depression are present. In the usual clini- results of this validation are shown in Box  1 in Addi- cal assessment group, 84% of the GPs presented an MDI tional file 1. Br inck‑Claussen et al. BMC Fam Pract (2021) 22:88 Page 5 of 8 Table 1 Baseline characteristics Usual clinical assessment, n = 256 Case-finding, n = 244 Age mean, range 38 (18–81) 40 (18–75) Male gender, n (%) 96 (38) 88 (36) GP‑ diagnosis Depression as primary diagnosis, n (%) 231 (90) 218 (89) Depression as secondary diagnosis, n (%) 25 (10) 26 (11) MDI available, n (%) 215 (84) 214 (89) MDI mean sum, score 32 32 Primary diagnosis after MINI interview, n (%) Mild depression 30 (12) 24 (10) Moderate depression 101 (39) 82 (34) Severe depression 92 (36) 86 (35) Generalized anxiety, panic disorder or Social phobia 22 (9) 26 (11) Other diagnosis 4 (1) 11 (4) No diagnosis 7 (3) 15 (6) Total 256 (100) 244 (100) Abbreviations: GP General practitioner, MDI Major Depression Inventory, MINI Mini International Neuropsychiatric Interview sum-score at referral, corresponding to 89% of the GPs in assessment group followed current guidelines, which the case-finding group. suggest exploring diagnostic criteria in more detail if two Secondary exploratory analyses show that for the group core symptoms of depression were present, the higher assessed with a mild depression via MINI interview, the PPV could indicate that exploring diagnostic criteria only mean MDI sum-score was 26. The group found with when suspecting depression is insufficient. This would moderate depression had a mean MDI sum-score of 31, be a plausible explanation as the prevalence of depres and the group found with severe depression had a mean sion would be higher in a group of patients who the GP MDI sum-score of 36 (Table 3). For the group of partici - suspects have depression and that have at least two core pants found with no depression after MINI, the average symptoms, compared to patients who the GP only sus- MDI sum-score was 30. pects have depression. However, this can only be hypoth- esized because the approach and usage of MDI in the usual clinical assessment group could vary. Discussion A reason for the unexpected result could also be that In this study, case-finding compared to usual clinical GPs in the case-finding group did not fully implement assessment did not improve the precision of the depres- the detection method. GPs in the case-finding group sion diagnosis in primary care. If the usual clinical only enclosed an MDI sum-score for 89% of included cases and not for the expected 100%. Unfortunately, we cannot investigate the reason for the lacking MDI sum- Table 2 Positive predictive value for depression diagnosis based scores from the data available. Possibly GPs did not apply on case‑finding and standard detection the MDI in obvious cases, and greater application of MDI + – Total, n PPV could have given a higher PPV. Research has also shown Depression Depression (95%CI) that GPs could be reluctant to use scales because they feel (MINI), n (MINI), n that the scales do not fit into a fluid conversation, that the Depression 237 19 256 0.93 (0.89– results do not always correspond to their clinical impres diagnose by 0.96) sion, and could even be misleading [4]. GPs in the usual usual clinical assessment clinical assessment group enclosed an MDI sum-score Depression 203 41 244 0.83 (0.78– almost to the same extent (84%) as the GPs in the case- diagnose by 0.88) finding group. Still, the high usage of MDI in both groups case‑finding could not explain the higher PPV in the usual clinical Total 440 60 500 assessment group compared to the case-finding group. Abbreviations: MINI Mini International Neuropsychiatric Interview, PPV Positive Predictive Value Brinck‑Claussen et al. BMC Fam Pract (2021) 22:88 Page 6 of 8 Table 3 MDI sum‑score means according to depression severity Authors found that the MDI was a conservative measure of depression compared to the M-CIDI and a valid tool Depression degree No. of MDIs Mean SD 95% CI for for diagnosing depression when applied to persons who according to MINI mean were suspected to have depression [27]. It was a limita- No depression 46 30.2 7.9 27.9–32.5 tion that we used the MINI as standard reference and Mild depression 58 26.4 7.3 24.5–28.2 not SCID or CIDI. However, we used MINI because it is Moderat depression 164 31.5 6.9 30.4–32.5 less time-consuming and a widely used tool for validating Severe depression 159 36.0 5.7 35.1–36.9 symptoms and diagnoses. Performed MINI interviews Total 427 might not have diagnosed all patients with depression, Note: Two cases were excluded because of missing data on depression degree but if the research assistant’s diagnosis was inconsistent according to MINI with the GP’s diagnosis, the research assistant consulted Abbreviations: MDI Major Depression Inventory, MINI MINI International a psychiatrist in the Collabri group, who contacted the Neuropsychiatric Interview GP to agree on a result. This procedure would, however, Including cases without a diagnosis or other diagnosis not identify false-positive cases if the MINI would diag- nose patients with a depression that did not have depres- sion. The identification of participants relied on the GPs’ The awareness of participating in a study could have judgment, and other recruitment strategies such as iden- influenced the usual behavior of GPs’ in the usual clinical tification through records or waiting room screenings assessment group, but we cannot examine this from the could have identified a different sample of participants on data available. If GPs in the case-finding group applied which the MDI would have been applied. However, the MDI at a lower threshold of suspicion than normally, present strategy of GPs identifying participants is closer they would perhaps detect depression in a sample with to everyday clinical practice. We also cannot rule out a lower prevalence compared to GPs in the usual clini- the possibility that time between tests may have had an cal assessment group. Since PPV depends on prevalence, impact, as the symptoms may have fluctuated, or medical a reduction would also reduce the PPV. The baseline treatment could have been initiated. data did, however, not indicate important differences in We planned to estimate the sensitivity, specificity, and characteristics between groups. Perhaps GPs in the usual NPV of the GP’s depression diagnosis in the case-find - clinical assessment group had other advantages in their ing group. However, as GPs only referred few persons course of detecting depression, e.g., a better prior knowl- believed not to have depression after first suspecting edge of the patient, a different way of dealing with a dif - one, we consider these estimates as unreliable. Thus, less ferential diagnosis, or perhaps they applied MDI with a information about the case-finding method than first different timing along the diagnostic process. anticipated was gained. The mean MDI sum-scores for each depression sever - It can impact the external validity that some exclu- ity group according to MINI indicate higher scores than sion criteria from the Collabri trials also applied to otherwise suggested cut-offs of 21 for mild depression, this nested study. Moreover, the low number of gen- 26 for moderate depression, and 31 for severe depression eral practices participating (51 of 713 invited) is also a [26]. Thus, when using MINI as standard, initial MDI threat to the external validity, as participating GPs could sum-sores presented at referral might have overestimated represent those especially interested in collaborative depression severity in this study sample. However, fur- care or depression detection. Further, if both detection ther studies must be conducted to confirm this finding. groups perform well, it would be more difficult to detect a difference between groups. It is a limitation that we have no baseline information about GPs on factors such Strengths and limitations as sex, age, and years of practice, as these could have Strengths of this trial were the relatively large sample of had an impact on their clinical performance. Addition- participants, the centralized computer-based cluster- ally, there is a risk that the GP sample size is too small randomization, and the use of blinded assessors regard- to avoid significant baseline differences between GPs, ing the allocation. We consider it as a strength of the which could affect the detection practice and thereby, study that the MDI was used as an assessment tool. The the outcome. MDI is a measure already used in general practice and recommended in Danish guidelines [21]. In a study by Nielsen et  al., the validity of the MDI, compared against Comparison with existing literature the Munich-Composite International Diagnostic Inter- The results did not show large problems with false pos- view (M-CIDI), was investigated in a set up comparable itives in either of the studied groups. In comparison, to usual clinical practice in Danish general practice [27]. Br inck‑Claussen et al. BMC Fam Pract (2021) 22:88 Page 7 of 8 Authors’ contributions Mitchell et al. [5] found a PPV of 42.0% (39.6%-44.3%) LFE led trial development as the principal investigator. UØBC drafted the for the GP depression diagnosis across a sample of 19 manuscript’s first version, of which NKC finalized. UØBC, NKC, ASD, CH, ML, studies. In this meta-analysis, depression checklists JHM, MEL, CC, KSC, and MN contributed to the development of the study. Where statistical software was required, CH conducted the analyses. The or questionnaires for GPs  were,  for example, used  to authors approved the final version of the manuscript. assess the diagnosis. Golden standards were often Structural clinical interview for DSM (SCID), Com- Funding This study was funded by a grant from the Danish Ministry of Health to posite International diagnostic interview (CIDI), and strengthen the collaboration between general practice and Mental Health the Diagnostic interview scale (DIS). Christensen Services through shared care. et al. found a true positive rate of 60.5% for a group of Availability of data and materials GPs  using case-finding [14]. Still, in our study, 7% of Data is not available because of the General Data Protection Regulation. Upon patients in the usual clinical assessment group and 17% reasonable request data can be retrieved. of patients in the case-finding group were diagnosed with a depression that was not verified using MINI. Declarations Ethics approval and consent to participate This study was a part of the Collabri and Collabri Flex trials, which were Implications for research and/or practice approved by a scientific ethics committee in the Capital Region of Denmark. Our findings suggest that usual clinical assessment is Reference number H‑3–2013‑203 and H‑16034303. Participants provided informed consent. All methods were performed in accordance with relevant more precise than case-finding in this setting. However, guidelines and regulations. more research is needed to support this result. Identifi - cation of depression might be improved by integrating Consent for publication Not applicable. the case-finding approach into a stepped care model; however, this should be studied further. Optimally, Competing interests the MDI should be used for people with at least two Authors state that they have no competing interests. of three core symptoms of depression. Thus, it would Author details be relevant to examine the precision of the first three Copenhagen Research Center for Mental Health, Mental Health Center questions in the MDI used as a screening tool before Copenhagen, Gentofte Hospitalsvej 15, 2900 Hellerup, Denmark. Research Unit for General Practice, Institute of Public Health, Aarhus University, testing with the full MDI. Bartholins Allé 2, 8000 Aarhus C, Denmark. The Research Unit for General Practice and Section of General Practice, University of Copenhagen, Oester Farimagsgade 5, Postbox 2099, 1014 Copenhagen K, Denmark. Mental Health Center Frederiksberg, Nordre Fasanvej 57‑59, 2000 Frederiksberg, Denmark. Conclusions Stolpegaard Psychotherapy Center, Stolpegaardsvej 20, 2820 Gentofte, Den‑ In this study, routine clinical assessment outperformed 6 7 mark. General Practitioner in Copenhagen, Copenhagen, Denmark. Mental case-finding in the identification of depression in pri - Health Center North Zealand, Dyrehavevej 48, 3400 Hilleroed, Denmark. mary care. Further studies should be conducted to con- Received: 6 November 2020 Accepted: 5 April 2021 firm this finding. Abbreviations CIDI: Composite International Diagnostic Interview; GP: General Practitioner; References ICD: International Statistical Classification of Disease; MDI: Major Depression 1. Olsen LR, Mortensen EL, Bech P. Prevalence of major depression and Inventory; MINI: MINI International Neuropsychiatric Interview; NPV: Negative stress indicators in the Danish general population. Acta Psychiatr Scand. Predictive Value; PPV: Positive Predictive Value; RCT : Randomized Controlled 2004;109:96–103. Trial; SCID‑P: Structured Clinical Interview for DSM ‑III‑Revised, Patients. 2. World Health Organization. Depression and other common mental disorders: global health estimates. Licence: CC BY‑NC‑SA 3.0 IGO. Geneva; Supplementary Information 3. Kroenke K. Patients presenting with somatic complaints: Epidemiology, The online version contains supplementary material available at https:// doi. psychiatric co‑morbidity and management. Int J Methods Psychiatr Res. org/ 10. 1186/ s12875‑ 021‑ 01432‑w. 2003;12:34–43. 4. Davidsen AS, Fosgerau CF. What is depression? Psychiatrists’ and GPs’ experiences of diagnosis and the diagnostic process. Int J Qual Stud Additional file 1: Box 1. Information about the validation of the MINI Health Well‑being. 2014;9:1–10. interview conducted by telephone. 5. Mitchell AJ, Vaze A, Rao S. Clinical diagnosis of depression in primary care: a meta‑analysis. Lancet. 2009;374:609–19. 6. Dowrick C, Buchan I. Twelve month outcome of depression in gen‑ Acknowledgements eral practice: Does detection or disclosure make a difference? BMJ. We thank research assistants and care managers in the Collabri and Collabri 1995;311:1274. Flex studies for conducting MINI interviews. We also thank general practition‑ 7. Thompson C, Kinmonth AL, Stevens L, Peveler RC, Stevens A, Ostler KJ, ers and patients for their participation in the study. et al. Eec ff ts of a clinical‑practice guideline and practice ‑based education on detection and outcome of depression in primary care: Hampshire Depression Project randomised controlled trial. Lancet. 2000;355:185–91. Brinck‑Claussen et al. BMC Fam Pract (2021) 22:88 Page 8 of 8 8. Hung C‑I, Yu N‑ W, Liu C‑ Y, Wu K‑ Y, Yang C‑H. The impact of the duration 20. Curth NK, Brinck‑ Claussen U, Jørgensen KB, Rosendal S, Hjorthøj C, Nor‑ of an untreated episode on improvement of depression and somatic dentoft M, et al. Collaborative care vs consultation liaison for depression symptoms. Neuropsychiatr Dis Treat. 2015;11:2245–52. and anxiety disorders in general practice: study protocol for two rand‑ 9. Siu AL, Bibbins‑Domingo K, Grossman DC, Baumann LC, Davidson KW, omized controlled trials (the Danish Collabri Flex trials). Trials. 2019;20:607. Ebell M, et al. Screening for depression in adults: US preventive services 21. Dansk Selskab for Almen Medicin. Klinisk vejledning for almen praksis. task force recommendation statement. JAMA. 2016;315:380–7. Unipolar depression. Diagnostik og behandling. 2010. https:// www. dsam. 10. Thombs BD, Ziegelstein RC, Roseman M, Kloda LA, Ioannidis JPA. There dk/ files/9/ depre ssion_ med_ links. pdf . Accessed 20 Oct 2020. are no randomized controlled trials that support the United States 22. Bech P, Rasmussen NA, Olsen LR, Noerholm V, Abildgaard W. The sensitiv‑ Preventive Services Task Force Guideline on screening for depression in ity and specificity of the Major Depression Inventory, using the Present primary care: a systematic review. BMC Med. 2014;12:13. State Examination as the index of diagnostic validity. J Aec ff t Disord. 11. Sundhedsstyrelsen. Referenceprogram for unipolar depression hos vok‑ 2001;66:159–64. sne. 2007. https:// www. sst. dk/‑/ media/ Udgiv elser/ 2007/ Publ2 007/ PLAN/ 23. Madsen MT, Zahid JA, Hansen CH, Grummedal O, Hansen JR, Isbrand A, SfR/ SST_ Dep,‑d‑ ,rappo rt,‑d‑ pdf. ashx. Accessed 12 June 2020. et al. The effect of melatonin on depressive symptoms and anxiety in 12. Romera I, Montejo ÁL, Aragonés E, Arbesú JÁ, Iglesias‑ García C, López patients after acute coronary syndrome: the MEDACIS randomized clini‑ S, et al. Systematic depression screening in high‑risk patients attend‑ cal trial. J Psychiatr Res. 2019;119:84–94. https:// doi. org/ 10. 1016/j. jpsyc ing primary care: a pragmatic cluster‑randomized trial. BMC Psychiatry. hires. 2019. 09. 014. 2013;13:83. 24. Okkels N, Jensen LG, Skovshoved LC, Arendt R, Blicher AB, Vieta E, et al. 13. Baas KD, Wittkampf KA, Van Weert HC, Lucassen P, Huyser J, Van Den Lighting as an aid for recovery in hospitalized psychiatric patients: a ran‑ Hoogen H, et al. Screening for depression in high‑risk groups: Prospective domized controlled effectiveness trial. Nord J Psychiatry. 2020;74:105–14. cohort study in general practice. Br J Psychiatry. 2009;194:399–403.https:// doi. org/ 10. 1080/ 08039 488. 2019. 16764 65. 14. Christensen KS, Sokolowski I, Olesen F. Case‑finding and risk ‑ group 25. Lecrubier Y, Sheehan DV, Weiller E, Amorim P, Bonora I, Sheehan KH, et al. screening for depression in primary care. Scand J Prim Health Care. The Mini International Neuropsychiatric Interview (MINI). A short diag‑ 2011;29:80–4. nostic structured interview: reliability and validity according to the CIDI. 15. Gilbody S, House A, Sheldon T. Screening and case finding instruments Eur Psychiatry. 1997;12:224–31. for depression. Cochrane Database Syst Rev. 2005. 26. Bech P, Timmerby N, Martiny K, Lunde M, Soendergaard S. Psychometric 16. Archer J, Bower P, Gilbody S, Lovell K, Richards D, Gask L, et al. Collabo‑ evaluation of the Major Depression Inventory (MDI) as depression sever‑ rative care for depression and anxiety problems (Review). Cochrane ity scale using the LEAD (Longitudinal Expert Assessment of All Data) as Database Syst Rev. 2012. index of validity. BMC Psychiatry. 2015;15:190. 17. Sheehan D, Lecrubier Y, Harnett Sheehan K, Janavs J, Weiller E, Keskiner 27. Nielsen MG, Ørnbøl E, Bech P, Vestergaard M, Christensen KS. The A, et al. The validity of the Mini International Neuropsychiatric Inter‑ criterion validity of the web‑based major depression inventory when view (MINI) according to the SCID‑P and its reliability. Eur Psychiatry. used on clinical suspicion of depression in primary care. Clin Epidemiol. 1997;12:232–41. 2017;9:355–65. https:// doi. org/ 10. 2147/ CLEP. S1329 13. 18. Brinck‑ Claussen U, Curth NK, Davidsen AS, Mikkelsen JH, Lau ME, Lunds‑ teen M, et al. Collaborative care for depression in general practice: Study Publisher’s Note protocol for a randomised controlled trial. Trials. 2017;18:1–12. Springer Nature remains neutral with regard to jurisdictional claims in pub‑ 19. Curth NK, Brinck‑ Claussen UØ, Davidsen AS, Lau ME, Lundsteen M, Mik‑ lished maps and institutional affiliations. kelsen JH, et al. Collaborative care for panic disorder, generalised anxiety disorder and social phobia in general practice: study protocol for three cluster‑randomised, superiority trials. Trials. 2017;18:382. Re Read ady y to to submit y submit your our re researc search h ? Choose BMC and benefit fr ? Choose BMC and benefit from om: : fast, convenient online submission thorough peer review by experienced researchers in your field rapid publication on acceptance support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions

Journal

BMC Family PracticeSpringer Journals

Published: May 7, 2021

There are no references for this article.