Distribution-free Inference of Zero-inated Binomial Data for Longitudinal Studies.

Count reponses with structural zeros are very common in medical and psychosocial research, especially in alcohol and HIV research, and the zero-inflated poisson (ZIP) and zero-inflated negative binomial (ZINB) models are widely used for modeling such outcomes. However, as alcohol drinking outcomes such as days of drinkings are counts within a given period, their distributions are bounded above by an upper limit (total days in the period) and thus inherently follow a binomial or zero-inflated binomial (ZIB) distribution, rather than a Poisson or zero-inflated Poisson (ZIP) distribution, in the presence of structural zeros. In this paper, we develop a new semiparametric approach for modeling zero-inflated binomial (ZIB)-like count responses for cross-sectional as well as longitudinal data. We illustrate this approach with both simulated and real study data.

Distribution-free models for longitudinal count responses with overdispersion and structural zeros.

Overdispersion and structural zeros are two major manifestations of departure from the Poisson assumption when modeling count responses using Poisson log-linear regression. As noted in a large body of literature, ignoring such departures could yield bias and lead to wrong conclusions. Different approaches have been developed to tackle these two major problems. In this paper, we review available methods for dealing with overdispersion and structural zeros within a longitudinal data setting and propose a distribution-free modeling approach to address the limitations of these methods by utilizing a new class of functional response models. We illustrate our approach with both simulated and real study data.

"Early withdrawal from mental health treatment: Implications for psychotherapy practice": Correction to Barrett et al (2008).

Reports an error in "Early withdrawal from mental health treatment: Implications for psychotherapy practice" by Marna S. Barrett, Wee-Jhong Chua, Paul Crits-Christoph, Mary Beth Gibbons and Don Thompson (Psychotherapy: Theory, Research, Practice, Training, 2008[Jun], Vol 45[2], 247-267). The fourth author's name was mistakenly left out of the author byline and table of contents. The correct author listing for this article is presented in the erratum. (The following abstract of the original article appeared in record 2008-07317-011.) Despite more than 50 years of research on client attrition from therapy, obstacles to the delivery and success of treatments remain poorly understood, and effective methods to engage and retain clients in therapy are lacking. This article offers a review of the literature on attrition, highlighting the methodological challenges in effectively addressing the complex nature of this problem. Current interventions for reducing attrition are reviewed, and recommendations for implementing these interventions into psychotherapy practice are discussed. (PsycINFO Database Record (c) 2010 APA, all rights reserved).

Power analyses for correlations from clustered study designs.

Power analysis constitutes an important component of modern clinical trials and research studies. Although a variety of methods and software packages are available, almost all of them are focused on regression models, with little attention paid to correlation analysis. However, the latter is arguably a simpler and more appropriate approach for modelling concurrent events, especially in psychosocial research. In this paper, we discuss power and sample size estimation for correlation analysis arising from clustered study designs. Our approach is based on the asymptotic distribution of correlated Pearson-type estimates. Although this asymptotic distribution is easy to use in data analysis, the presence of a large number of parameters creates a major problem for power analysis due to the lack of real data to estimate them. By introducing a surrogacy-type assumption, we show that all nuisance parameters can be eliminated, making it possible to perform power analysis based only on the parameters of interest. Simulation results suggest that power and sample size estimates obtained under the proposed approach are robust to this assumption.

Power analyses for longitudinal trials and other clustered designs.

Existing methods for power and sample size estimation for longitudinal and other clustered study designs have limited applications. In this paper, we review and extend existing approaches to improve these limitations. In particular, we focus on power analysis for the two most popular approaches for clustered data analysis, the generalized estimating equations and the linear mixed-effects models. By basing the derivation of the power function on the asymptotic distribution of the model estimates, the proposed approach provides estimates of power that are consistent with the methods of inference for data analysis. The proposed methodology is illustrated with numerous examples that are motivated by real study designs.

Generalized covariance-adjusted canonical correlation analysis with application to psychiatry.

The lack of control over covariates in practice motivates the need for their adjustment when measuring the degree of association between two sets of variables, for which canonical correlation is traditionally used. In most studies however, there is also a lack of control over the attributes of responses for the sets of variables of interest. In particular, a portion of the response variable may be continuous and the other discrete. For such settings, the traditional partial canonical correlation approach is restrictive, since a covariate-adjustment for a set of continuous variables is assumed. By ignoring the assumption of continuous variates and proceeding with a partial canonical correlation analysis in the presence of continuous and discrete variates, results in canonical correlation estimates that are not consistent. In this paper we generalize the traditional partial canonical correlation approach to covariate-adjustment by allowing the response variables to contain continuous, as well as discrete, variates. The methodology is illustrated with a psychiatric application for examining which sleep variables relate to which depressive symptoms, as measured by commonly used constructs that presents with both continuous and discrete outcomes.

Pretreatment expectations, interpersonal functioning, and symptoms in the prediction of the therapeutic alliance across supportive-expressive psychotherapy and cognitive therapy.

Therapeutic alliance has been a robust predictor of therapy outcome, yet little is known about which patient variables predict the development of an alliance between patient and therapist in time-limited manualized therapies. The authors evaluated pretreatment predictors of therapeutic alliance, controlling for symptom change before its assessment, using a large sample of patients treated with either supportive-expressive (SE) dynamic psychotherapy or cognitive therapy. They found that SE patients with greater pretreatment expectations of improvement formed better alliances with their therapist at Session 2, and expectations significantly predicted alliance at Session 10 for both treatment groups. Further, patients in the SE condition demonstrated a significant relation between positive expectations and growth in alliance. Women achieved better alliances at Session 10. Finally, hostile-dominant interpersonal problems significantly predicted poor alliance. Pretreatment symptom level was not significantly predictive of alliance.

Impact of psychosocial treatments on associated problems of cocaine-dependent patients.

A previous report from the National Institute on Drug Abuse Collaborative Cocaine Treatment Study (P. Crits-Christoph et al., 1999) found relatively superior cocaine and drug use outcomes for individual drug counseling plus group drug counseling compared with other treatments. Using data from that study, the authors examined the relative efficacy of 4 treatments for cocaine dependence on psychosocial and other addiction-associated problems. The 487 patients were randomly assigned to 6 months of treatment with cognitive therapy, supportive-expressive therapy, or individual drug counseling (each with additional group drug counseling), or to group drug counseling alone. Assessments were made at baseline and monthly for 6 months during the acute treatment phase, with follow-up visits at 9 and 12 months. No significant differences between treatments were found on measures of psychiatric symptoms, employment, medical, legal, family-social, interpersonal, or alcohol use problems. The authors concluded that the superiority of individual drug counseling in modifying cocaine use does not extend broadly to other addiction-associated problems.

Early improvement during manual-guided cognitive and dynamic psychotherapies predicts 16-week remission status.

This study examined the extent to which improvement from baseline to weeks 2, 3, and 4 on the Beck Depression Inventory and Beck Anxiety Inventory predict week 16 clinical remission for patients with major depressive disorder, generalized anxiety disorder, and/or obsessive-compulsive or avoidant personality disorders who were receiving manual-based psychotherapies. Logistic regression and receiver-operator characteristic analyses revealed relatively accurate identification of remitters and nonremitters based on improvement from baseline to sessions 2 to 4 in both original and cross-validation samples. Predictive success did not vary as a function of diagnosis, treatment type (cognitive or dynamic), or treatment status (short-term or long-term). The clinical implications of the results are discussed.

Therapeutic alliance as a predictor of outcome and retention in the National Institute on Drug Abuse Collaborative Cocaine Treatment Study.

The authors examined the relation between therapeutic alliance, retention, and outcome for 308 cocaine-dependent outpatients participating in the National Institute on Drug Abuse Collaborative Cocaine Treatment Study. High levels of alliance were observed in supportive-expressive therapy (SE), cognitive therapy (CT), and individual drug counseling (IDC), and alliance levels increased slightly but significantly from Session 2 to Session 5 in all groups. In contrast to other studies, alliance was not a significant predictor of drug outcome. However, alliance did predict patient retention differentially across the 3 treatments. In SE and IDC, either higher levels of alliance were associated with increased retention or no relationship between alliance and retention was found, depending on the time alliance was measured. In CT, higher levels of alliance were associated with decreased retention.

Self-help group attendance and participation among cocaine dependent patients.

The authors examined frequency and patterns of self-help group attendance and active participation over a 6-month period among 411 patients receiving treatment in the NIDA Collaborative Cocaine Treatment Study. Nearly two-thirds of patients attended at least one self-help group, and nearly all of these actively participated. Alcoholics Anonymous and Narcotics Anonymous meetings were attended most frequently. Statistical analyses included chi square, one-way analyses of variance, and cluster techniques. While patterns of attendance were relatively consistent over time, findings suggest that a treatment emphasizing the importance of self-help groups is likely to encourage more self-help group attendance and participation over time.

The role of therapist characteristics in training effects in cognitive, supportive-expressive, and drug counseling therapies for cocaine dependence.

The role of therapist characteristics in therapy training was examined for 62 therapists in a multisite psychotherapy outcome study that included cognitive therapy (CT), supportive-expressive (SE) psychodynamic therapy, and individual drug counseling (IDC) for cocaine-dependent patients. Demographic variables and experience and competence ratings prior to training were correlated with measures of change in competence during the training phase. Higher competence ratings before training were associated with greater change in competence for SE and higher average competence for IDC. More years of experience were associated with greater change in competence for CT therapists, but more hours of pre-training supervision in the CT treatment modality were associated with less change.

Alliance predicts patients' outcome beyond in-treatment change in symptoms.

The authors examined the relations among therapeutic alliance, outcome, and early-in-treatment symptomatic improvement in a group of 86 patients with generalized anxiety disorders, chronic depression, or avoidant or obsessive-compulsive personality disorder who received supportive-expressive dynamic psychotherapy. Although alliance at Sessions 5 and 10, but not at Session 2, was associated with prior change in depression, alliance at all sessions significantly predicted subsequent change in depression when prior change in depression was partialed out. The results are discussed in terms of the causal role of the alliance in therapeutic outcome.

Clinicians' impact on the quality of substance use disorder treatment.

Clinicians' impact on substance use disorder treatment has been much less studied than therapy and patient variables. Yet, in this selective review of literature, a growing body of empirical work on clinicians' impact highlights several key issues that have relevance both to clinical practice and future research. These issues include clinicians' effect on treatment retention and outcome, professional characteristics, recovery status, adherence to protocols, counter-transference, alliance, personality, beliefs about treatment, and professional practice issues. Specific recommendations are offered to help improve the quality of care clinicians provide. In particular, it is suggested that greater accountability for clinicians' performance be balanced with increased support for their very difficult role. Methodological issues in studying clinicians are also addressed.

Alliance and technique in short-term dynamic therapy.

We review existing studies on the extent to which the alliance and techniques predict the outcome of short-term dynamic psychotherapy (STDP). Although the alliance has been found to relate to outcome in diverse psychotherapies, evidence for its role in STDP is more inconsistent. Studies of technique factors have provided support for the relation of competence with exploratory interventions to outcome. Less support for the role of specific transference interpretations was found. Little empirical support exists for the notion that the technical factors are most potent in the context of a positive alliance, although adequate studies of this interaction are scarce. Similarly, adequate studies that attempt to tease apart the extent to which technical factors influence the alliance, and vice versa, are also rare. Despite the methodological limitations of research in this area, the fact that there are some consistent findings relating techniques to outcome suggests that nonspecific relationship factors do not by themselves account for the changes found over the course of STDP. Moreover, such findings encourage additional efficacy and process research on STDP. Some suggestions for future research are presented.

Psychosocial treatments for cocaine dependence: National Institute on Drug Abuse Collaborative Cocaine Treatment Study.

BACKGROUND: This was a multicenter investigation examining the efficacy of 4 psychosocial treatments for cocaine-dependent patients. METHODS: Four hundred eighty-seven patients were randomly assigned to 1 of 4 manual-guided treatments: individual drug counseling plus group drug counseling (GDC), cognitive therapy plus GDC, supportive-expressive therapy plus GDC, or GDC alone. Treatment was intensive, including 36 possible individual sessions and 24 group sessions for 6 months. Patients were assessed monthly during active treatment and at 9 and 12 months after baseline. Primary outcome measures were the Addiction Severity Index-Drug Use Composite score and the number of days of cocaine use in the past month. RESULTS: Compared with the 2 psychotherapies and with GDC alone, individual drug counseling plus GDC showed the greatest improvement on the Addiction Severity Index-Drug Use Composite score. Individual group counseling plus GDC was also superior to the 2 psychotherapies on the number of days of cocaine use in the past month. Hypotheses regarding the superiority of psychotherapy to GDC for patients with greater psychiatric severity and the superiority of cognitive therapy plus GDC compared with supportive-expressive therapy plus GDC for patients with antisocial personality traits or external coping style were not confirmed. CONCLUSION: Compared with professional psychotherapy, a manual-guided combination of intensive individual drug counseling and GDC has promise for the treatment of cocaine dependence.

Quality of life: expanding the scope of clinical significance.

Clinical researchers have turned their attention to quality of life assessment as a means of broadening the evaluation of treatment outcomes. This article examines conceptual and methodological issues related to the use of quality of life measures in mental health. These include the lack of a good operational definition of the construct, the use of subjective versus objective quality of life indicators, and the nature of the relationship between symptoms and quality of life judgments. Of special concern is the ability of quality of life measures to detect treatment-related changes. The authors review the application of quality of life assessment across diverse patient groups and therapies and provide recommendations for developing comprehensive, psychometrically sophisticated quality of life measures.