Clinical effectiveness of integrating depression care management into medicare home health: the Depression CAREPATH Randomized trial.

IMPORTANCE: Among older home health care patients, depression is highly prevalent, is often inadequately treated, and contributes to hospitalization and other poor outcomes. Feasible and effective interventions are needed to reduce this burden of depression. OBJECTIVE: To determine whether, among older Medicare Home Health recipients who screen positive for depression, patients of nurses receiving randomization to an intervention have greater improvement in depressive symptoms during 1 year than patients receiving enhanced usual care. DESIGN, SETTING, AND PARTICIPANTS: This cluster randomized effectiveness trial conducted at 6 home health care agencies nationwide assigned nurse teams to an intervention (12 teams) or to enhanced usual care (9 teams). Between January 13, 2009, and December 6, 2012, Medicare Home Health patients 65 years and older who screened positive for depression on routine nursing assessments were recruited, underwent assessment, and were followed up at 3, 6, and 12 months by research staff blinded to intervention status. Patients were interviewed at home and by telephone. Of 502 eligible patients, 306 enrolled in the study. INTERVENTIONS: The Depression Care for Patients at Home (Depression CAREPATH) trial requires nurses to manage depression at routine home visits by weekly symptom assessment, medication management, care coordination, education, and goal setting. Nurses' training totaled 7 hours (4 onsite and 3 via the web). Researchers telephoned intervention team supervisors every other week. MAIN OUTCOMES AND MEASURES: Depression severity, assessed by the 24-item Hamilton Scale for Depression (HAM-D). RESULTS: The 306 participants were predominantly female (69.6%), were racially/ethnically diverse (18.0% black and 16.0% Hispanic), and had a mean (SD) age of 76.5 (8.0) years. In the full sample, the intervention had no effect (P = .13 for intervention × time interaction). Adjusted HAM-D scores (Depression CAREPATH vs control) did not differ at 3 months (10.5 vs 11.4, P = .26) or at 6 months (9.3 vs 10.5, P = .12) but reached significance at 12 months (8.7 vs 10.6, P = .05). In the subsample with mild depression (HAM-D score, <10), the intervention had no effect (P = .90), and HAM-D scores did not differ at any follow-up points. Among 208 participants with a HAM-D score of 10 or higher, the Depression CAREPATH demonstrated effectiveness (P = .02), with lower HAM-D scores at 3 months (14.1 vs 16.1, P = .04), at 6 months (12.0 vs 14.7, P = .02), and at 12 months (11.8 vs 15.7, P = .005). CONCLUSION AND RELEVANCE: Home health care nurses can effectively integrate depression care management into routine practice. However, the clinical benefit seems to be limited to patients with moderate to severe depression. TRIAL REGISTRATION: clinicaltrials.gov Identifier: NCT01979302.

A novel application of the Intent to Attend assessment to reduce bias due to missing data in a randomized controlled clinical trial.

BACKGROUND: Missing data are unavoidable in most randomized controlled clinical trials, especially when measurements are taken repeatedly. If strong assumptions about the missing data are not accurate, crude statistical analyses are biased and can lead to false inferences. Furthermore, if we fail to measure all predictors of missing data, we may not be able to model the missing data process sufficiently. In longitudinal randomized trials, measuring a patient's intent to attend future study visits may help to address both of these problems. Leon et al. developed and included the Intent to Attend assessment in the Lithium Treatment - Moderate dose Use Study (LiTMUS), aiming to remove bias due to missing data from the primary study hypothesis. PURPOSE: The purpose of this study is to assess the performance of the Intent to Attend assessment with regard to its use in a sensitivity analysis of missing data. METHODS: We fit marginal models to assess whether a patient's self-rated intent predicted actual study adherence. We applied inverse probability of attrition weighting (IPAW) coupled with patient intent to assess whether there existed treatment group differences in response over time. We compared the IPAW results to those obtained using other methods. RESULTS: Patient-rated intent predicted missed study visits, even when adjusting for other predictors of missing data. On average, the hazard of retention increased by 19% for every one-point increase in intent. We also found that more severe mania, male gender, and a previously missed visit predicted subsequent absence. Although we found no difference in response between the randomized treatment groups, IPAW increased the estimated group difference over time. LIMITATIONS: LiTMUS was designed to limit missed study visits, which may have attenuated the effects of adjusting for missing data. Additionally, IPAW can be less efficient and less powerful than maximum likelihood or Bayesian estimators, given that the parametric model is well specified. CONCLUSIONS: In LiTMUS, the Intent to Attend assessment predicted missed study visits. This item was incorporated into our IPAW models and helped reduce bias due to informative missing data. This analysis should both encourage and facilitate future use of the Intent to Attend assessment along with IPAW to address missing data in a randomized trial.

A new proposal for randomized start design to investigate disease-modifying therapies for Alzheimer disease.

BACKGROUND: The increasing prevalence of Alzheimer disease (AD) and lack of effective agents to attenuate progression have accelerated research and development of disease modifying (DM) therapies. The traditional parallel group design and single time point analysis used in the support of past AD drug approvals address symptomatic benefit over relatively short treatment durations. More recent trials investigating disease modification are by necessity longer in duration and require larger sample sizes. Nevertheless, trial design and analysis remain mostly unchanged and may not be adequate to meet the objective of demonstrating disease modification. Randomized start design (RSD) has been proposed as an option to study DM effects, but its application in AD trials may have been hampered by certain methodological challenges. PURPOSE: To address the methodological issues that have impeded more extensive use of RSD in AD trial and to encourage other researchers to develop novel design and analysis methodologies to better ascertain DM effects for the next generation of AD therapies, we propose a stepwise testing procedure to evaluate potential DM effects of novel AD therapies. METHODS: Alzheimer Disease Assessment Scale-Cognitive Subscale (ADAS-cog) is used for illustration. We propose to test three hypotheses in a stepwise sequence. The three tests pertain to treatment difference at two separate time points and a difference in the rate of change. Estimation is facilitated by the Mixed-effects Model for Repeated Measures approach. The required sample size is estimated using Monte Carlo simulations and by modeling ADAS-cog data from prior longitudinal AD studies. RESULTS: The greatest advantage of the RSD proposed in this article is its ability to critically address the question on a DM effect. The AD trial using the new approach would be longer (12-month placebo period plus 12-month delay-start period; total 24-month duration) and require more subjects (about 1000 subjects per arm for the non-inferiority margin chosen in the illustration). It would also require additional evaluations to estimate the rate of ADAS-cog change toward the end of the trial. LIMITATIONS: A regulatory claim of disease modification for any compound will likely require additional verification of a drug's effect on a validated biomarker of Alzheimer's pathology. CONCLUSIONS: Incorporation of the RSD in AD trials is feasible. With proper trial setup and statistical procedures, this design could support the detection of a disease-modifying effect. In our opinion, a two-phase RSD with a stepwise hypothesis testing procedure could be a reasonable option for future studies.

Integrating statistical and clinical research elements in intervention-related grant applications: summary from an NIMH workshop.

OBJECTIVE: The authors summarize points for consideration generated in a National Institute of Mental Health (NIMH) workshop convened to provide an opportunity for reviewers from different disciplines-specifically clinical researchers and statisticians-to discuss how their differing and complementary expertise can be well integrated in the review of intervention-related grant applications. METHODS: A 1-day workshop was convened in October, 2004. The workshop featured panel presentations on key topics followed by interactive discussion. This article summarizes the workshop and subsequent discussions, which centered on topics including weighting the statistics/data analysis elements of an application in the assessment of the application's overall merit; the level of statistical sophistication appropriate to different stages of research and for different funding mechanisms; some key considerations in the design and analysis portions of applications; appropriate statistical methods for addressing essential questions posed by an application; and the role of the statistician in the application's development, study conduct, and interpretation and dissemination of results. RESULTS: A number of key elements crucial to the construction and review of grant applications were identified. It was acknowledged that intervention-related studies unavoidably involve trade-offs. Reviewers are helped when applications acknowledge such trade-offs and provide good rationale for their choices. Clear linkage among the design, aims, hypotheses, and data analysis plan and avoidance of disconnections among these elements also strengthens applications. CONCLUSION: The authors identify multiple points to consider when constructing intervention-related grant applications. The points are presented here as questions and do not reflect institute policy or comprise a list of best practices, but rather represent points for consideration.

Manic/hypomanic symptom burden and cardiovascular mortality in bipolar disorder.

OBJECTIVES: To compare the risk for cardiovascular mortality between bipolar I and bipolar II subtypes and determine correlates of cardiovascular mortality. Bipolar disorder conveys an increased risk of cardiovascular mortality. METHODS: Participants with major affective disorders were recruited for the National Institute of Mental Health Collaborative Depression Study and followed prospectively for up to 25 years. A total of 435 participants met the diagnostic criteria for bipolar I (n = 288) or bipolar II (n = 147) disorder based on Research Diagnostic Criteria at intake and measures of psychiatric symptoms during follow-up. Diagnostic subtypes were contrasted by cardiovascular mortality risk using Cox proportional hazards regression. Affective symptom burden (the proportion of time with clinically significant manic/hypomanic or depressive symptoms) and treatment exposure were additionally included in the models. RESULTS: Thirty-three participants died from cardiovascular causes. Participants with bipolar I disorder had more than double the cardiovascular mortality risk of those with bipolar II disorder, after controlling for age and gender (hazard ratio = 2.35, 95% Confidence Interval = 1.04-5.33; p = .04). The observed difference in cardiovascular mortality between these subtypes was at least partially confounded by the burden of clinically significant manic/hypomanic symptoms which predicted cardiovascular mortality independent of diagnosis, treatment exposure, age, gender, and cardiovascular risk factors at intake. Selective serotonin uptake inhibitors seemed protective although they were introduced late in follow-up. Depressive symptom burden was not related to cardiovascular mortality. CONCLUSIONS: Participants with bipolar I disorder may face a greater risk of cardiovascular mortality than those with bipolar II disorder. This difference in cardiovascular mortality risk may reflect manic/hypomanic symptom burden.

Implications of clinical trial design on sample size requirements.

The primary goal in designing a randomized controlled clinical trial (RCT) is to minimize bias in the estimate of treatment effect. Randomized group assignment, double-blinded assessments, and control or comparison groups reduce the risk of bias. The design must also provide sufficient statistical power to detect a clinically meaningful treatment effect and maintain a nominal level of type I error. An attempt to integrate neurocognitive science into an RCT poses additional challenges. Two particularly relevant aspects of such a design often receive insufficient attention in an RCT. Multiple outcomes inflate type I error, and an unreliable assessment process introduces bias and reduces statistical power. Here we describe how both unreliability and multiple outcomes can increase the study costs and duration and reduce the feasibility of the study. The objective of this article is to consider strategies that overcome the problems of unreliability and multiplicity.

A randomized trial of depression assessment intervention in home health care.

OBJECTIVES: To determine whether an educational intervention would improve depression assessment and appropriate referral. Secondary analyses tested whether referral led to depression improvement. DESIGN: Training in the Assessment of Depression (TRIAD) was a three-group, nurse-randomized trial. Researchers interviewed randomly selected patients at baseline and 8 weeks. SETTING: Three certified home healthcare agencies in Westchester County, New York. PARTICIPANTS: Fifty-three medical/surgical nurses were randomized within agency to three intervention groups: full, minimal, or control. Research contact with nurses' patients (aged >65; N=477) yielded 256 (53.7%) enrolled subjects, 84 (17.6%) ineligibles, and 120 (25.2%) refusals; 233 of the 256 (87.1%) enrolled patients completed follow-up interviews. INTERVENTION: Nurse training in clinically meaningful use of depression sections of Medicare's mandatory Outcome and Assessment Information Set (OASIS). MEASUREMENTS: Nurse-assessed mood or anhedonia (OASIS) versus research assessments using the Structured Clinical Interview for Axis I Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition Disorders (SCID); referrals for mental health evaluation (agency records), and depression severity (24-item Hamilton Depression Rating Scale; HDRS). RESULTS: Referral rates for patients with (SCID) depressed mood or anhedonia (n=75) varied according to nurse group: 50.0% full intervention, 18.5% minimal, 21.4% control (P=.047). Rates for nondepressed patients (n=180) did not differ (4.9%, 2.0%, 5.8%, respectively; P=.60). In patients with major or minor depression (n=37), referral was associated with symptom improvement. Change in HDRS was 5 points greater in referred patients than others (P=.04). Concordance between OASIS and SCID did not differ between intervention groups. CONCLUSION: TRIAD showed that training nurses to assess for depression using an approach developed in partnership with home healthcare agencies led to appropriate referral and care for depressed patients.

Assessment of community functioning in people with schizophrenia and other severe mental illnesses: a white paper based on an NIMH-sponsored workshop.

People with schizophrenia frequently have significant problems in community functioning. Progress in developing effective interventions to ameliorate these problems has been slowed by the absence of reliable and valid measures that are suitable for use in clinical trials. The National Institute of Mental Health convened a workgroup in September 2005 to examine this issue and make recommendations to the field that would foster research in this area. This article reports on issues raised at the meeting. Many instruments have been developed to assess community functioning, but overall insufficient attention has been paid to psychometric issues and many instruments are not suitable for use in clinical trials. Consumer self-report, informant report, ratings by clinicians and trained raters, and behavioral assessment all can provide useful and valid information in some circumstances and may be practical for use in clinical trials. However, insufficient attention has been paid to when and how different forms of assessment and sources of information are useful or how to understand inconsistencies. A major limiting factor in development of reliable and valid instruments is failure to develop a suitable model of functioning and its primary mediators and moderators. Several examples that can guide thinking are presented. Finally, the field is limited by the absence of an objective gold standard of community functioning. Hence, outcomes must be evaluated in part by "clinical significance." This criterion is problematic because different observers and constituencies often have different opinions about what types of change are clinically important and how much change is significant.

Bias reduction in effectiveness analyses of longitudinal ordinal doses with a mixed-effects propensity adjustment.

A mixed-effects propensity adjustment is described that can reduce bias in longitudinal studies involving non-equivalent comparison groups. There are two stages in this data analytic strategy. First, a model of propensity for treatment intensity examines variables that distinguish among subjects who receive various ordered doses of treatment across time using mixed-effects ordinal logistic regression. Second, the effectiveness model examines multiple times until recurrence to compare the ordered doses using a mixed-effects grouped-time survival model. Effectiveness analyses are initially stratified by propensity quintile. Then the quintile-specific results are pooled, assuming that there is not a propensity x treatment interaction. A Monte Carlo simulation study compares bias reduction in fully specified propensity model relative to misspecified models. In addition, type I error rate and statistical power are examined. The approach is illustrated by applying it to a longitudinal, observational study of maintenance treatment of major depression.

Assessment of outcome in depression.

Valid and reliable methods of assessing outcome in depression are crucial to antidepressant efficacy studies but are also important for all studies that relate response to other variables. In this paper we review basic issues of reliability and validity associated with outcome measurement. We distinguish between scales or inventories designed for screening or diagnosis and those intended for outcome assessment. We note historical changes in patient selection (e.g. outpatients instead of inpatients, differentiating psychotic and non-psychotic) and how these changes affect commonly used scales such as the Hamilton Depression Rating Scale. We examine whether antidepressants with different mechanisms of action are best assessed with similar or different symptoms. And we review studies that have identified 'core' symptoms of depression and compare these findings to the magnitude of individual symptom change we found in five studies of selective serotonergic or nonadrenergic antidepressants.

Distinguishing bipolar major depression from unipolar major depression with the screening assessment of depression-polarity (SAD-P).

BACKGROUND: Patients with bipolar I or II major depression are often misdiagnosed with unipolar major depression. The goal of this study was to develop and validate a brief instrument to screen for bipolar disorder in patients actively ill with major depression. METHOD: The sample consisted of subjects who enrolled in the National Institute of Mental Health-Collaborative Program on the Psychobiology of Depression-Clinical Studies from 1978 to 1981 during an episode of major depression and included 91 subjects with bipolar I major depression, 52 with bipolar II major depression, and 338 with unipolar major depression diagnosed according to Research Diagnostic Criteria. Most of the subjects were inpatients at the time of enrollment, and subjects were prospectively followed for up to 20 years. In order to create, test, and cross-validate the screening instrument, a split-sample data analytic procedure was performed. This procedure yielded 3 groups of subjects: the bipolar I index sample, the bipolar I cross-validation sample, and the bipolar II cross-validation sample. Each group included subjects with bipolar major depression and subjects with unipolar major depression. Within the bipolar I index sample, subjects with bipolar I major depression at study intake were compared with subjects with unipolar major depression at study intake on a pool of 59 sociodemographic and clinical candidate variables. The 3 variables showing the greatest disparity between bipolar I subjects and unipolar subjects were selected for the screen, the Screening Assessment of Depression-Polarity (SAD-P). The operating characteristics of the SAD-P were then examined within the bipolar I index sample, bipolar I cross-validation sample, and bipolar II cross-validation sample. RESULTS: The items selected for the screening instrument were (1) presence of delusions during the current episode of major depression, (2) number of prior episodes of major depression, and (3) family history of major depression or mania. The screen identified bipolar major depression with a sensitivity of 0.82 in the bipolar I index sample, 0.72 in the bipolar I cross-validation sample, and 0.58 in the bipolar II cross-validation sample. With regard to misclassifying subjects with unipolar major depression, the screen provided a positive predictive value of 0.36 in the bipolar I index sample, 0.29 in the bipolar I cross-validation sample, and 0.27 in the bipolar II cross-validation sample. CONCLUSION: We suggest using the 3-item SAD-P as a preliminary screen for bipolar disorder in patients who present with an active episode of major depression.

Psychosocial disability in the course of bipolar I and II disorders: a prospective, comparative, longitudinal study.

CONTEXT: Evidence of psychosocial disability in bipolar disorder is based primarily on bipolar I disorder (BP-I) and does not relate disability to affective symptom severity and polarity or to bipolar II disorder (BP-II). OBJECTIVE: To provide detailed data on psychosocial disability in relation to symptom status during the long-term course of BP-I and BP-II. DESIGN: A naturalistic study with 20 years of prospective, systematic follow-up. SETTING: Inpatient and outpatient treatment facilities at 5 US academic centers. Patients One hundred fifty-eight patients with BP-I and 133 patients with BP-II who were followed up for a mean (SD) of 15 (4.8) years in the National Institute of Mental Health Collaborative Depression Study. MAIN OUTCOME MEASURES: The relationship, by random regression, between Range of Impaired Functioning Tool psychosocial impairment scores and affective symptom status in 1-month periods during the long-term course of illness from 6-month and yearly Longitudinal Interval Follow-up Evaluation interviews. RESULTS: Psychosocial impairment increases significantly with each increment in depressive symptom severity for BP-I and BP-II and with most increments in manic symptom severity for BP-I. Subsyndromal hypomanic symptoms are not disabling in BP-II, and they may even enhance functioning. Depressive symptoms are at least as disabling as manic or hypomanic symptoms at corresponding severity levels and, in some cases, significantly more so. At each level of depressive symptom severity, BP-I and BP-II are equally impairing. When asymptomatic, patients with bipolar disorder have good psychosocial functioning, although it is not as good as that of well controls. CONCLUSIONS: Psychosocial disability fluctuates in parallel with changes in affective symptom severity in BP-I and BP-II. Important findings for clinical management are the following: (1) depressive episodes and symptoms, which dominate the course of BP-I and BP-II, are equal to or more disabling than corresponding levels of manic or hypomanic symptoms; (2) subsyndromal depressive symptoms, but not subsyndromal manic or hypomanic symptoms, are associated with significant impairment; and (3) subsyndromal hypomanic symptoms appear to enhance functioning in BP-II.

A mixed-effects quintile-stratified propensity adjustment for effectiveness analyses of ordered categorical doses.

Observational studies can be used to evaluate treatment effectiveness among patients with a broader range of illness severity than typically seen in randomized controlled clinical trials. However, there are several difficulties with observational evaluations including non-equivalent comparison groups, treatment doses and durations that vary widely, and, in longitudinal studies, multiple courses of treatment per subject. A mixed-effects approach to the propensity adjustment for non-equivalent comparison groups is described that can account for each of these perturbations. The strategy involves two stages. First, characteristics that distinguish among subjects who receive various levels of treatment are examined in a model of propensity for treatment intensity using mixed-effects ordinal logistic regression. Second, the propensity-stratified effectiveness of ordered categorical doses is compared in a mixed-effects grouped time survival model of time until recovery. The model is applied in a longitudinal, observational study of antidepressant effectiveness. Then a Monte Carlo simulation study indicates that the strategy has acceptable type I error rates and minimal bias in the estimates of treatment effectiveness. Statistical power exceeds 0.90 for an odds ratio of 1.5 with N = 250 and 500, and is acceptable for an odds ratio of 2.0 with N = 100. Nevertheless, with N = 100, the models that had high intraclass correlation coefficients had greater tendency towards non-convergence. This approach is a useful strategy for observational studies of treatment effectiveness. It is capable of adjusting for selection bias, incorporating multiple observations per subject, and comparing effectiveness of ordinal doses.

Comparison of statistical methods for analysis of clustered binary observations.

When correlated observations are obtained in a randomized controlled trial, the assumption of independence among observations within cluster likely will not hold because the observations share the same cluster (e.g. clinic, physician, or subject). Further, the outcome measurements of interest are often binary. The objective of this paper is to compare the performance of four statistical methods for analysis of clustered binary observations: namely (1) full likelihood method; (2) penalized quasi-likelihood method; (3) generalized estimating equation method; (4) fixed-effects logistic regression method. The first three methods take correlations into account in inferential processes whereas the last method does not. Type I error rate, power, bias, and standard error are compared across the four statistical methods through computer simulations under varying effect sizes, intraclass correlation coefficients, number of clusters, and number of observations per cluster, including large numbers 20 and 100 of observations per cluster. The results show that the performance of the full likelihood and the penalized quasi-likelihood methods is superior for analysis of clustered binary observations, and is not necessarily inferior to that of the fixed-effects logistic regression fit even when within-cluster correlations are zero.

Income distribution and risk of fatal drug overdose in New York City neighborhoods.

Accidental drug overdose is a substantial cause of mortality for drug users. Neighborhood-level factors, such as income distribution, may be important determinants of overdose death independent of individual-level factors. We used data from the Office of the Chief Medical Examiner to identify all cases of accidental deaths in New York City (NYC) in 1996 and individual-level covariates. We used 1990 US Census data to calculate the neighborhood-level income distribution. This multi-level case-control study included 725 accidental overdose deaths (cases) and 453 accidental deaths due to other causes (controls) in 59 neighborhoods in NYC. Overdose deaths were more likely in neighborhoods with higher levels of drug use and with more unequal income distribution. In multi-level models, income maldistribution was significantly associated with risk of overdose independent of individual-level variables (age, race, and sex) and neighborhood-level variables (income, drug use, and racial composition). The odds of death due to drug overdose were 1.63-1.88 in neighborhoods in the least equitable decile compared with neighborhoods in the most equitable decile. Disinvestment in social and economic resources in unequal neighborhoods may explain this association. Public health interventions related to overdose risk should pay particular attention to highly unequal neighborhoods.

Assessment and treatment of depression in disability claimants: a cost-benefit simulation study.

The economic burden of depression includes direct costs of treatment, as well as absenteeism and reduced productivity. In this study, we consider the costs and benefits of an intervention to assess and treat depressive symptoms in long-term disability claimants with nonpsychiatric medical illnesses. Cost-benefit simulations were conducted using data from a study sample of long-term disability claimants (N = 1229) and estimates of both the costs of treatment of depressive symptoms and the savings in claims payments for those who return to work as a result of treatment. We show that the savings that stem from returning a very few claimants to work can offset the assumed cost of a comprehensive program for the assessment of depressive symptoms in all claimants and intensive treatment of those with depressive symptoms. The economic and public health benefits both point toward the value of such an intervention for disability claimants.