Selective Cutoff Reporting in Studies of Diagnostic Test Accuracy: A Comparison of Conventional and Individual-Patient-Data Meta-Analyses of the Patient Health Questionnaire-9 Depression Screening Tool.

In studies of diagnostic test accuracy, authors sometimes report results only for a range of cutoff points around data-driven "optimal" cutoffs. We assessed selective cutoff reporting in studies of the diagnostic accuracy of the Patient Health Questionnaire-9 (PHQ-9) depression screening tool. We compared conventional meta-analysis of published results only with individual-patient-data meta-analysis of results derived from all cutoff points, using data from 13 of 16 studies published during 2004-2009 that were included in a published conventional meta-analysis. For the "standard" PHQ-9 cutoff of 10, accuracy results had been published by 11 of the studies. For all other relevant cutoffs, 3-6 studies published accuracy results. For all cutoffs examined, specificity estimates in conventional and individual-patient-data meta-analyses were within 1% of each other. Sensitivity estimates were similar for the cutoff of 10 but differed by 5%-15% for other cutoffs. In samples where the PHQ-9 was poorly sensitive at the standard cutoff, authors tended to report results for lower cutoffs that yielded optimal results. When the PHQ-9 was highly sensitive, authors more often reported results for higher cutoffs. Consequently, in the conventional meta-analysis, sensitivity increased as cutoff severity increased across part of the cutoff range-an impossibility if all data are analyzed. In sum, selective reporting by primary study authors of only results from cutoffs that perform well in their study can bias accuracy estimates in meta-analyses of published results.

A comparison of bivariate, multivariate random-effects, and Poisson correlated gamma-frailty models to meta-analyze individual patient data of ordinal scale diagnostic tests.

Individual patient data (IPD) meta-analyses are increasingly common in the literature. In the context of estimating the diagnostic accuracy of ordinal or semi-continuous scale tests, sensitivity and specificity are often reported for a given threshold or a small set of thresholds, and a meta-analysis is conducted via a bivariate approach to account for their correlation. When IPD are available, sensitivity and specificity can be pooled for every possible threshold. Our objective was to compare the bivariate approach, which can be applied separately at every threshold, to two multivariate methods: the ordinal multivariate random-effects model and the Poisson correlated gamma-frailty model. Our comparison was empirical, using IPD from 13 studies that evaluated the diagnostic accuracy of the 9-item Patient Health Questionnaire depression screening tool, and included simulations. The empirical comparison showed that the implementation of the two multivariate methods is more laborious in terms of computational time and sensitivity to user-supplied values compared to the bivariate approach. Simulations showed that ignoring the within-study correlation of sensitivity and specificity across thresholds did not worsen inferences with the bivariate approach compared to the Poisson model. The ordinal approach was not suitable for simulations because the model was highly sensitive to user-supplied starting values. We tentatively recommend the bivariate approach rather than more complex multivariate methods for IPD diagnostic accuracy meta-analyses of ordinal scale tests, although the limited type of diagnostic data considered in the simulation study restricts the generalization of our findings.

Screening for depression in high-risk groups: prospective cohort study in general practice.

BACKGROUND: Currently only about half of the people who have major depressive disorder are detected during regular health care. Screening in high-risk groups might be a possible solution. AIMS: To evaluate the effectiveness of selective screening for major depressive disorder in three high-risk groups in primary care: people with mental health problems, people with unexplained somatic complaints and people who frequently attend their general practitioner. METHOD: Prospective cohort study among 2005 people in high-risk groups in three health centres in The Netherlands. RESULTS: Of the 2005 people identified, 1687 were invited for screening and of these 780 participated. Screening disclosed 71 people with major depressive disorder: 36 (50.7%) already received treatment, 14 (19.7%) refused treatment and 4 individuals did not show up for an appointment. As a final result of the screening, 17 individuals (1% of 1687) started treatment for major depressive disorder. CONCLUSIONS: Screening for depression in high-risk populations does not seem to be effective, mainly because of the low rates of treatment initiation, even if treatment is freely and easily accessible.

Interventions on frequent attenders in primary care. A systematic literature review.

OBJECTIVE: To analyse which interventions are effective in influencing morbidity, quality of life, and healthcare utilization of frequently attending patients (FAs) in primary care. METHODS: A systematic literature search was performed for articles describing interventions on FAs in primary care (Medline, Embase, and PsycINFO). Outcomes were morbidity, quality of life (QoL), and use of healthcare. Two independent assessors selected all randomized clinical trials (RCT) and assessed the quality of the selected RCTs. Results. Five primary care based RCTs were identified. Three RCTs used frequent attendance to select patients at risk of distress, major depression, and anxiety disorders. These RCTs applied psychological and psychiatric interventions and focused on undiagnosed psychiatric morbidity of FAs. Two of them found more depression-free days and a better QoL after treating major depressive disorder in FAs. No other RCT found any positive effect on morbidity or QoL. Two RCTs studied an intervention which focused on reducing frequent attendance. No intervention significantly lowered attendance. Due to the difference in study settings and the variation in methods of selecting patients, meta-analysis of the results was not possible. CONCLUSION: No study showed convincing evidence that an intervention improves QoL or morbidity of frequent attending primary care patients, although a small effect might be possible in a subgroup of depressed frequent attenders. No evidence was found that it is possible to influence healthcare utilization of FAs.

Diagnostic accuracy of the mood module of the Patient Health Questionnaire: a systematic review.

OBJECTIVE: The nine-item mood module of the Patient Health Questionnaire (PHQ-9) was developed to screen and to diagnose patients in primary care with depressive disorders. We systematically reviewed the psychometric literature on the PHQ-9 and performed a meta-analysis to ascertain its summary sensitivity and specificity. METHODS: EMBASE, PubMed and PsycINFO were used to search literature up to July 2006. Studies were included if (1) they investigated the diagnostic accuracy of the PHQ-9 and (2) the PHQ-9 had been compared with a reference test. The quality of the studies was appraised using the Quality Assessment of Diagnostic Accuracy Studies. We calculated sensitivity, specificity and confidence intervals for each included study. We used the random effects model to calculate the summary sensitivity and specificity. RESULTS: We found a sensitivity of 0.77 (0.71-0.84) and a specificity of 0.94 (0.90-0.97) for the PHQ-9. The positive predictive value in an unselected primary care population was 59%, which increased to 85-90% when the prior probability increased to 30-40%. CONCLUSION: In primary care, the PHQ-9 is a valid diagnostic tool if used in selected subgroups of patients with a high prevalence of depressive disorder.

Brief cognitive behavioral therapy compared to general practitioners care for depression in primary care: a randomized trial.

BACKGROUND: Depressive disorders are highly prevalent in primary care (PC) and are associated with considerable functional impairment and increased health care use. Research has shown that many patients prefer psychological treatments to pharmacotherapy, however, it remains unclear which treatment is most optimal for depressive patients in primary care. METHODS/DESIGN: A randomized, multi-centre trial involving two intervention groups: one receiving brief cognitive behavioral therapy and the other receiving general practitioner care. General practitioners from 109 General Practices in Nijmegen and Amsterdam (The Netherlands) will be asked to include patients aged between 18-70 years presenting with depressive symptomatology, who do not receive an active treatment for their depressive complaints. Patients will be telephonically assessed with the Structured Clinical Interview for DSM-IV Axis I Disorders (SCID-I) to ascertain study eligibility. Eligible patients will be randomized to one of two treatment conditions: either 8 sessions of cognitive behavioral therapy by a first line psychologist or general practitioner's care according to The Dutch College of General Practitioners Practice Guideline (NHG- standard). Baseline and follow-up assessments are scheduled at 0, 6, 12 and 52 weeks following the start of the intervention. Primary outcome will be measured with the Hamilton Depression Rating Scale-17 (HDRS-17) and the Patient Health Questionnaire-9 (PHQ-9). Outcomes will be analyzed on an intention to treat basis. TRIAL REGISTRATION: ISRCTN65811640.