Dean McMillan

Optimal cut-off score for diagnosing depression with the Patient Health Questionnaire (PHQ-9): a meta-analysis

Background: The brief Patient Health Questionnaire (PHQ-9) is commonly used to screen for depression with 10 often recommended as the cut-off score. We summarized the psychometric properties of the PHQ-9 across a range of studies and cut-off scores to select the optimal cut-off for detecting depression. Methods: We searched Embase, MEDLINE and PsycINFO from 1999 to August 2010 for studies that reported the diagnostic accuracy of PHQ-9 to diagnose major depressive disorders. We calculated summary sensitivity, specificity, likelihood ratios and diagnostic odds ratios for detecting major depressive disorder at different cut-off scores and in different settings. We used random-effects bivariate meta-analysis at cutoff points between 7 and 15 to produce summary receiver operating characteristic curves. Results: We identified 18 validation studies (n = 7180) conducted in various clinical settings. Eleven studies provided details about the diagnostic properties of the questionnaire at more than one cut-off score (including 10), four studies reported a cut-off score of 10, and three studies reported cut-off scores other than 10. The pooled specificity results ranged from 0.73 (95% confidence interval [CI] 0.63–0.82) for a cut-off score of 7 to 0.96 (95% CI 0.94–0.97) for a cut-off score of 15. There was major variability in sensitivity for cut-off scores between 7 and 15. There were no substantial differences in the pooled sensitivity and specificity for a range of cut-off scores (8–11). Interpretation: The PHQ-9 was found to have acceptable diagnostic properties for detecting major depressive disorder for cut-off scores between 8 and 11. Authors of future validation studies should consistently report the outcomes for different cut-off scores.

Development of sexually abusive behaviour in sexually victimised males: a longitudinal study

BACKGROUND Sexual maltreatment is one of the most common forms of child abuse. To identify risk factors for sexually abusive behaviour by adults, we prospectively assessed childhood experiences and personal characteristics of male child victims who became abusers in later life. METHODS In a longitudinal study (7-19 years duration), we included 224 former male victims of sexual abuse. Risk factors contemporaneous with the abuse, and putative protective influences, were identified from social service and clinical records. Evidence of later criminal acts was obtained from a nationwide search of official records. FINDINGS Of the 224 former victims, 26 had subsequently committed sexual offences (victim-abusers), in almost all cases with children, mainly outside their families. Risk factors during childhood for later offending included material neglect (odds ratio 3.4, 95% CI 1.2-9.7), lack of supervision (3.0, 1.1-8.3), and sexual abuse by a female person (3.0, 1.1-8.7). Victim-abusers had more frequently witnessed serious intrafamilial violence (3.1, 1.0-10.0). Six (29%) of 21 victim-abusers on whom we had relevant data had been cruel to animals (7.9, 2.0-31.4). No single putative protective factor, nor a composite protective index, significantly reduced the risk of paedophilic behaviour. INTERPRETATION Most male victims of child sexual abuse do not become paedophiles, but particular experiences and patterns of childhood behaviour are associated with an increased risk of victims becoming abusers in later life. Our findings have implications for the design of selective interventions with a vulnerable subgroup of male victims, aimed at reducing the risk of paedophilic behaviour in later life.

Can we predict suicide and non-fatal self-harm with the Beck Hopelessness Scale: a meta-analysis

BACKGROUND Hopelessness is considered a pre-eminent risk factor for suicide and non-fatal self-harm. We aimed to quantify the ability of the Beck Hopelessness Scale (BHS) to predict these two outcomes. METHOD Medline, Embase, PsycINFO and Cinahl were searched to January 2006. We included cohort studies in which the BHS was applied and patients were followed-up to establish subsequent suicide or non-fatal self-harm. Four studies provided usable data on suicide, and six studies provided data on non-fatal self-harm. Summary sensitivity, specificity, likelihood ratios and diagnostic odds ratios (DORs) were calculated for each study. Random effects meta-analytic pooling across studies at the standard cut-off point (> or =9) was undertaken and summary receiver operating characteristic (ROC) curves constructed. RESULTS For suicide, pooled sensitivity was 0.80 [95% confidence interval (CI) 0.68-0.90], pooled specificity was 0.42 (95% CI 0.41-0.44), and the pooled DOR was 3.39 (95% CI 1.29-8.88). For non-fatal self-harm, pooled sensitivity was 0.78 (95% CI 0.74-0.82), pooled specificity was 0.42 (95% CI 0.38-0.45), and the pooled DOR was 2.27 (95% CI 1.53-3.37). CONCLUSION The standard cut-off point on the BHS identifies a high-risk group for potential suicide, but the magnitude of the risk is lower than previously reported estimates. The standard cut-off point is also capable of identifying those who are at risk of future self-harm, but the low specificity rate means it is unlikely to be of use in targeting treatment designed to lower the rate of repetition.

The Hospital Anxiety and Depression Scale: a diagnostic meta-analysis of case-finding ability.

OBJECTIVE To quantify the accuracy of the Hospital Anxiety and Depression Scale (HADS) as a case-finding instrument for anxiety and depressive disorders. METHOD MEDLINE, PSYCHINFO, EMBASE, CINAHL, BNI, and AMED were searched from January 1983 to June 2006. Studies were included that administered the HADS, used a standardized psychiatric interview to establish a diagnosis of anxiety or depression, and provided sufficient data on sensitivity and specificity (N=41). Summary sensitivity, specificity, likelihood ratios, and diagnostic odds ratios were calculated for each study. Random effects meta-analytic pooling across studies at the recommended clinical (7/8) and research (10/11) cutoff points was undertaken and summary receiver operating characteristic curves constructed. RESULTS For major depressive disorders, a cut point of ≥8 gave a sensitivity of 0.82 (95% CI, 0.73-0.89) and a specificity of 0.74 (95% CI, 0.60-0.84) and a cut point ≥11 gave a sensitivity of 0.56 (95% CI, 0.40-0.71) and a specificity of 0.92 (95% CI, 0.79-0.97). CONCLUSIONS Many studies have shown that the HADS is a useful screening tool to identify emotional distress in nonpsychiatric patients. However, it does not appear to be superior to other screening instruments in terms of identifying specific mental disorders in physical health settings.

Psychometric Comparison of the PHQ-9 and BDI-II for Measuring Response during Treatment of Depression

The Patient Health Questionnaire-9 Item (PHQ-9) and Beck Depression Inventory–II (BDI-II) are frequently used measures of depression severity, but little is known about their relative psychometric properties. The authors assessed psychometric properties of both measures during treatment for depression. The PHQ-9 and BDI-II scores from 172 depressed participants in two randomized controlled trials of treatment for depression were assessed and combined. Tests of internal consistency (Cronbachs α), factor analyses, correlational analyses, estimates of clinically significant change, and effect sizes (Cohens d) were calculated after treatment and follow-up. Both scales demonstrated adequate internal consistency at pre- and posttreatment (PHQ-9 α = .74 and .81; BDI-II α = .87 and .90, respectively). Factor analysis failed to confirm the one-factor model previously reported for the PHQ-9, but two factors evidenced good fit for the BDI-II. Both scales converged more with each other than with the Sheehan Disability Scale at pre- but not at posttreatment. Responsiveness to change of PHQ-9 and BDI-II was similar at both posttreatment and follow-up. The consistency of agreement on indices of clinical significance was fair to moderate, but the BDI-II categorised a greater proportion of participants with severe depression than the PHQ-9. The BDI-II and PHQ-9 demonstrated adequate reliability, convergent/discriminant validity, and similar responsiveness to change. Differences were found in how they categorised severity. Pending the results of further studies, the attributes of the PHQ-9, of being shorter and based on the diagnostic criteria for depression, may indicate an advantage over the BDI-II.

Behavioural activation delivered by the non-specialist: phase II randomised controlled trial

BACKGROUND Behavioural activation appears as effective as cognitive-behaviour therapy (CBT) in the treatment of depression. If equally effective, then behavioural activation may be the preferred treatment option because it may be suitable for delivery by therapists with less training. This is the first randomised controlled trial to look at this possibility. AIMS To examine whether generic mental health workers can deliver effective behavioural activation as a step-three high-intensity intervention. METHOD A randomised controlled trial (ISRCTN27045243) comparing behavioural activation (n=24) with treatment as usual (n=23) in primary care. RESULTS Intention-to-treat analyses indicated a difference in favour of behavioural activation of -15.79 (95% CI -24.55 to -7.02) on the Beck Depression Inventory-II and Work and Social Adjustment Scale (mean difference -11.12, 95% CI -17.53 to -4.70). CONCLUSIONS Effective behavioural activation appears suitable for delivery by generic mental health professionals without previous experience as therapists. Large-scale trial comparisons with an active comparator (CBT) are needed.

Psychometric Comparison of the Generalized Anxiety Disorder Scale-7 and the Penn State Worry Questionnaire for Measuring Response during Treatment of Generalised Anxiety Disorder

The Penn State Worry Questionnaire (PSWQ) is a widely used measure of the worry characteristic of generalised anxiety disorder (GAD). The 7-item Generalized Anxiety Disorder Scale (GAD-7) is a new brief screening tool for GAD, which is being increasingly used in research and clinical practice. The present study sought to provide comparison data on the relative psychometric properties of these two scales. The data of 195 adults who met Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) criteria for GAD and who participated in two randomised treatment controlled trials were used. Factor analyses, internal consistency, correlational analyses, responsiveness to change, and agreement between the scales based on indentified clinical cutoffs were conducted. Factor analyses confirmed a one-factor structure for the GAD-7 and a three-factor structure involving two method factors for the PSWQ. Both the GAD-7 and the PSWQ demonstrated adequate internal consistency (Cronbachs alpha: .79–.91 and .86–.91, respectively), and moderate correlations (r = .51–.71) were observed between the scales across the treatment time points. The scales exhibited small correlations with the Sheehan Disability Scale at pretreatment (GAD-7 r = .38; PSWQ r = .26), but moderate correlations at posttreatment and follow-up (r = .59–.79). Agreement between the scales was limited using various clinical cutoffs identified within the literature. Both measures were sensitive to change, although the GAD-7 appeared to be more sensitive and may, therefore, confer some advantages in clinical work.

A diagnostic meta-analysis of the Patient Health Questionnaire-9 (PHQ-9) algorithm scoring method as a screen for depression

BACKGROUND The depression module of the Patient Health Questionnaire-9 (PHQ-9) is a widely used depression screening instrument in nonpsychiatric settings. The PHQ-9 can be scored using different methods, including an algorithm based on Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition criteria and a cut-off based on summed-item scores. The algorithm was the originally proposed scoring method to screen for depression. We summarized the diagnostic test accuracy of the PHQ-9 using the algorithm scoring method across a range of validation studies and compared the diagnostic properties of the PHQ-9 using the algorithm and summed scoring method at the proposed cut-off point of 10. METHODS We performed a systematic review of diagnostic accuracy studies of the PHQ-9 using the algorithm scoring method to detect major depressive disorder (MDD). We used meta-analytic methods to calculate summary sensitivity, specificity, likelihood ratios and diagnostic odds ratios for diagnosing MDD of the PHQ-9 using algorithm scoring method. In studies that reported both scoring methods (algorithm and summed-item scoring at proposed cut-off point of ≥10), we compared the diagnostic properties of the PHQ-9 using these methods. RESULTS We found 27 validation studies that validated the algorithm scoring method of the PHQ-9 in various settings. There was substantial heterogeneity across studies, which makes the pooled results difficult to interpret. In general, sensitivity was low whereas specificity was good. Thirteen studies reported the diagnostic properties of the PHQ-9 for both scoring methods. Pooled sensitivity for algorithm scoring method was lower while specificities were good for both scoring methods. Heterogeneity was consistently high; therefore, caution should be used when interpreting these results. INTERPRETATION This review shows that, if the algorithm scoring method is used, the PHQ-9 has a low sensitivity for detecting MDD. This could be due to the rating scale categories of the measure, higher specificity or other factors that warrant further research. The summed-item score method at proposed cut-off point of ≥10 has better diagnostic performance for screening purposes or where a high sensitivity is needed.

Cost and Outcome of Behavioural Activation versus Cognitive Behavioural Therapy for Depression (COBRA): a randomised, controlled, non-inferiority trial

Summary Background Depression is a common, debilitating, and costly disorder. Many patients request psychological therapy, but the best-evidenced therapy—cognitive behavioural therapy (CBT)—is complex and costly. A simpler therapy—behavioural activation (BA)—might be as effective and cheaper than is CBT. We aimed to establish the clinical efficacy and cost-effectiveness of BA compared with CBT for adults with depression. Methods In this randomised, controlled, non-inferiority trial, we recruited adults aged 18 years or older meeting Diagnostic and Statistical Manual of Mental Disorders IV criteria for major depressive disorder from primary care and psychological therapy services in Devon, Durham, and Leeds (UK). We excluded people who were receiving psychological therapy, were alcohol or drug dependent, were acutely suicidal or had attempted suicide in the previous 2 months, or were cognitively impaired, or who had bipolar disorder or psychosis or psychotic symptoms. We randomly assigned participants (1:1) remotely using computer-generated allocation (minimisation used; stratified by depression severity [Patient Health Questionnaire 9 (PHQ-9) score of <19 vs ≥19], antidepressant use, and recruitment site) to BA from junior mental health workers or CBT from psychological therapists. Randomisation done at the Peninsula Clinical Trials Unit was concealed from investigators. Treatment was given open label, but outcome assessors were masked. The primary outcome was depression symptoms according to the PHQ-9 at 12 months. We analysed all those who were randomly allocated and had complete data (modified intention to treat [mITT]) and also all those who were randomly allocated, had complete data, and received at least eight treatment sessions (per protocol [PP]). We analysed safety in the mITT population. The non-inferiority margin was 1·9 PHQ-9 points. This trial is registered with the ISCRTN registry, number ISRCTN27473954. Findings Between Sept 26, 2012, and April 3, 2014, we randomly allocated 221 (50%) participants to BA and 219 (50%) to CBT. 175 (79%) participants were assessable for the primary outcome in the mITT population in the BA group compared with 189 (86%) in the CBT group, whereas 135 (61%) were assessable in the PP population in the BA group compared with 151 (69%) in the CBT group. BA was non-inferior to CBT (mITT: CBT 8·4 PHQ-9 points [SD 7·5], BA 8·4 PHQ-9 points [7·0], mean difference 0·1 PHQ-9 points [95% CI −1·3 to 1·5], p=0·89; PP: CBT 7·9 PHQ-9 points [7·3]; BA 7·8 [6·5], mean difference 0·0 PHQ-9 points [–1·5 to 1·6], p=0·99). Two (1%) non-trial-related deaths (one [1%] multidrug toxicity in the BA group and one [1%] cancer in the CBT group) and 15 depression-related, but not treatment-related, serious adverse events (three in the BA group and 12 in the CBT group) occurred in three [2%] participants in the BA group (two [1%] patients who overdosed and one [1%] who self-harmed) and eight (4%) participants in the CBT group (seven [4%] who overdosed and one [1%] who self-harmed). Interpretation We found that BA, a simpler psychological treatment than CBT, can be delivered by junior mental health workers with less intensive and costly training, with no lesser effect than CBT. Effective psychological therapy for depression can be delivered without the need for costly and highly trained professionals. Funding National Institute for Health Research.

Defining successful treatment outcome in depression using the PHQ-9: A comparison of methods

BACKGROUND Although the PHQ-9 is widely used in primary care, little is known about its performance in quantifying improvement. The original validation study of the PHQ-9 defined clinically significant change as a post-treatment score of ≤9 combined with improvement of 50%, but it is unclear how this relates to other theoretically informed methods of defining successful outcome. We compared a range of definitions of clinically significant change (original definition, asymptomatic criterion, reliable and clinically significant change criteria a, b and c) in a clinical trial of a community-level depression intervention. METHOD Randomised Control Trial of collaborative care for depression. Levels of agreement were calculated between the standard definition, other definitions, and gold-standard diagnostic interview. RESULTS The standard definition showed good agreement (kappa>0.60) with the other definitions and had moderate, though acceptable, agreement with the diagnostic interview (kappa=0.58). The standard definition corresponded closely to reliable and clinically significant change criterion c, the recommended method of quantifying improvement when clinical and non-clinical distributions overlap. LIMITATIONS The absence of follow-up data meant that an asymptomatic criterion rather than remission or recovery criteria were used. CONCLUSION The close agreement between the standard definition and reliable and clinically significant change criterion c provides some support for the standard definition of improvement. However, it may be preferable to use a reliable change index rather than 50% improvement. Remission status, based on the asymptomatic range and a lower PHQ-9 score, may provide a useful additional category of clinical change.