Anna Bartak

The Multiple Propensity Score as Control for Bias in the Comparison of More Than Two Treatment Arms An Introduction From a Case Study in Mental Health

Background and Objective:The propensity score method (PS) has proven to be an effective tool to reduce bias in nonrandomized studies, especially when the number of (potential) confounders is large and dimensionality problems arise. The PS method introduced by Rosenbaum and Rubin is described in detail for studies with 2 treatment options. Since in clinical practice we are often interested in the comparison of multiple interventions, there was a need to extend the PS method to multiple treatments. It has been shown that in theory a multiple PS method is possible. So far, its practical application is rare and a practical introduction lacking. Methods:A practical guideline to illustrate the use of the multiple PS method is provided with data from a mental health study. The multiple PS is estimated with a multinomial logistic regression analysis. The multiple PS is the probability of assignment to each treatment category. Subsequently, to estimate the treatment effects while controlling for initial differences, the multiple PSs, calculated for each treatment category, are included as extra predictors in the regression analysis. Results:With the multiple PS method, balance was achieved in all relevant pretreatment variables. The corrected estimated treatment effects were somewhat different from the results without control for initial differences. Conclusions:Our results indicate that the multiple PS method is a feasible method to adjust for observed pretreatment differences in nonrandomized studies where the number of pretreatment differences is large and multiple treatments are compared.

Effectiveness of different modalities of psychotherapeutic treatment for patients with cluster C personality disorders: results of a large prospective multicentre study.

Background: No previous studies have compared the effectiveness of different modalities of psychotherapeutic treatment, as defined by different settings and durations, for patients with cluster C personality disorders. The aim of this multicentre study was to compare the effectiveness of 5 treatment modalities for patients with cluster C personality disorders in terms of psychiatric symptoms, psychosocial functioning, and quality of life. The following treatment modalities were compared: long-term outpatient (more than 6 months), short-term day hospital (up to 6 months), long-term day hospital, short-term inpatient, and long-term inpatient psychotherapy. Methods: The study was conducted between March 2003 and June 2008 in 6 mental health care centres in the Netherlands, with a sample of 371 patients with a DSM-IV-TR axis-II cluster C diagnosis. Patients were assigned to 5 different modalities of psychotherapeutic treatment, and effectiveness was assessed at 12 months after baseline. An intention-to-treat analysis was conducted for psychiatric symptoms (Brief Symptom Inventory), psychosocial functioning (Outcome Questionnaire-45), and quality of life (EQ-5D), using multilevel statistical modelling. As the study was non-randomised, the propensity score method was used to control for initial differences. Results: Patients in all treatment groups had improved on all outcomes 12 months after baseline. Patients receiving short-term inpatient treatment showed more improvement than patients receiving other treatment modalities. Conclusions: Psychotherapeutic treatment, especially in the short-term inpatient modality, is an effective treatment for patients with cluster C personality disorders.

Effectiveness of outpatient, day hospital, and inpatient psychotherapeutic treatment for patients with Cluster B personality disorders

Background: For patients with cluster B personality disorders there is no consensus regarding the optimal treatment setting. The aim of this study was to compare the effectiveness of different psychotherapeutic settings for patients with cluster B personality disorders, i.e. outpatient, day hospital, and inpatient treatment. Methods: The study was conducted between March 2003 and June 2008 in 6 mental health care centres in the Netherlands, with a sample of 207 patients with a DSM-IV-TR axis II cluster B diagnosis. Patients were assigned to 3 different settings of psychotherapeutic treatment and effectiveness was assessed at 18 months after baseline. An intention-to-treat analysis was conducted for psychiatric symptoms (Brief Symptom Inventory), psychosocial functioning (Outcome Questionnaire-45), and quality of life (EQ-5D), using multilevel statistical modelling. As the study was non-randomised, the propensity score method was used to control for initial differences. Results: Patients in all 3 settings improved significantly in terms of psychiatric symptoms, social and interpersonal functioning, and quality of life 18 months after baseline. The inpatient group showed the largest improvements. The comparison of outpatient and inpatient treatment regarding psychiatric symptoms showed a marginally significant result (p = 0.057) in favour of inpatient treatment. Conclusions: Patients with cluster B personality disorders improved in all investigated treatment settings, with a trend towards larger improvements of psychiatric symptoms in the inpatient setting compared to the outpatient setting. Specialised inpatient treatment should be considered as a valuable treatment option for cluster B personality disorders, both in research and in clinical practice.

The use of propensity score methods in psychotherapy research. A practical application.

Background: Randomized controlled trials are considered the best scientific proof of effectiveness. There is increasing concern, though, about their feasibility in psychotherapy research. We discuss a quasi-experimental study design for situations in which a randomized controlled trial is not feasible. Here, as an alternative strategy, the propensity score (PS) method is used to correct for selection bias. Methods: We used data from a Dutch research project, SCEPTRE (Study on Cost-Effectiveness of Personality Disorder Treatment). The sample consisted of 749 psychotherapy patients with personality pathology. We tested whether the PS method was useful and applicable. We examined differences between 2 treatment groups (short vs. long treatment duration) in pretreatment characteristics before and after PS correction. This revealed the impact of the PS on outcome differences. Results: The PS offered statistical control over observed pretreatment differences between patients in a non-randomized study. Conclusions: When a randomized controlled trial is not possible, this quasi-experimental design using the PS could be a feasible alternative. Its advantages and limitations are discussed. Implemented carefully, this method is promising for future effectiveness research.

Strengthening the Status of Psychotherapy for Personality Disorders : An Integrated Perspective on Effects and Costs

Objective: Despite scientific evidence of effectiveness, psychotherapy for personality disorders is not yet fully deployed, nor is its reimbursement self-evident. Both clinicians and health care policy-makers increasingly rely on evidence-based medicine and health economics when determining a treatment of choice and reimbursement. This article aims to contribute to that understanding by applying these criteria on psychotherapy as a treatment for patients with personality disorder. Method: We have evaluated the available empirical evidence on effectiveness and cost-effectiveness, and integrated this with necessity of treatment as a moderating factor. Results: The effectiveness of psychotherapy for personality disorders is well documented with favourable randomized trial results, 2 metaanalyses, and a Cochrane review. However, the evidence does not yet fully live up to modern standards of evidence-based medicine and is mostly limited to borderline and avoidant personality disorders. Data on cost-effectiveness suggest that psychotherapy for personality disorders may lead to cost-savings. However, state-of-the-art cost-effectiveness data are still scarce. An encouraging factor is that the available studies indicate that patients with personality disorder experience a high burden of disease, stressing the necessity of treatment. Conclusions: When applying an integrated vision on outcome, psychotherapy can be considered not only an effective treatment for patients with personality disorder but also most likely a cost-effective and necessary intervention. However, more state-of-the-art research is required before clinicians and health care policy-makers can fully appreciate the benefits of psychotherapy for personality disorders. Considerable progress is possible if researchers focus their efforts on evidence-based medicine and cost-effectiveness research.

Patients with Cluster A personality Disorders in psychotherapy: an Effectiveness study

Background: While psychopharmacological studies are common in patients with cluster A personality disorders, the effects of psychotherapy have received little attention. The aim of this study is to explore whether psychotherapeutic treatment yields health gains for these patients. Methods: The study was conducted between March 2003 and June 2008 in 6 mental health care centres in the Netherlands, with a sample of 57 patients with a DSM-IV-TR axis II cluster A diagnosis. Patients were assigned to 3 settings of psychotherapeutic treatment (outpatient, day hospital, inpatient), and effectiveness was assessed at 18 months after baseline. An intention-to-treat analysis was conducted for psychiatric symptoms (Brief Symptom Inventory), psychosocial functioning (Outcome Questionnaire-45) and quality of life (EQ-5D), using multilevel statistical modelling. As the study was non-randomised, the propensity score method was used to control for initial differences. Results: Patients in the day hospital and inpatient group improved substantially in terms of psychiatric symptoms, social and interpersonal functioning, and quality of life. Patients in the outpatient group showed less improvement. Direct comparison of the improvement of psychiatric symptoms showed significant results in favour of day hospital (p = 0.046) and inpatient (p = 0.01) treatment, as compared to outpatient treatment. However, due to substantial baseline differences, this direct comparison should be judged carefully. Conclusions: Cluster A psychopathology is not a contraindication to benefit from psychotherapy. This is especially true for more intensive forms like inpatient and day hospital treatment. Future research should focus more on psychotherapeutic treatment to gain further insight into effective treatment options for this patient group.

Effectiveness of psychotherapy in personality disorders not otherwise specified: a comparison of different treatment modalities

OBJECTIVE Although personality disorder not otherwise specified (PDNOS) is highly prevalent and associated with a high burden of disease, only a few treatment studies in this patient group exist. This study is the first to investigate the effectiveness of different modalities of psychotherapy in patients with PDNOS, i.e., short-term (up to 6 months) and long-term (more than 6 months) outpatient, day hospital, and inpatient psychotherapy. METHOD A total of 205 patients with PDNOS were assigned to one of six treatment modalities. Effectiveness was assessed over 60 months after baseline. The primary outcome measure was symptom severity, and the secondary outcome measures included psychosocial functioning and quality of life. The study design was quasi-experimental, and the multiple propensity score was used to control for initial differences between treatment groups. RESULTS All treatment modalities showed positive outcomes, especially in terms of improvements of symptom severity and social role functioning. At 12-month follow-up, after adjustment for initial differences between the treatment groups, short-term outpatient psychotherapy and short-term inpatient psychotherapy showed most improvement and generally outperformed the other modalities concerning symptom severity. At 60 months after baseline, effectiveness remained but observed differences between modalities mostly diminished. CONCLUSION Patients with PDNOS benefit from psychotherapy both at short-term and long-term follow-up. Short-term outpatient psychotherapy and short-term inpatient psychotherapy seem to be superior to the other treatment modalities at 12-month follow-up. At 60-month follow-up, treatments showed mostly comparable effectiveness. KEY PRACTITIONER MESSAGES The effectiveness of different modalities of psychotherapy in patients with PDNOS (i.e., short-term vs long-term; outpatient versus day hospital versus inpatient psychotherapy) has not yet been compared. Different modalities of psychotherapy are effective for patients with PDNOS, and positive effects remain after 5 years. In patients with PDNOS short-term (less than 6 months) outpatient psychotherapy and short-term inpatient psychotherapy seem to be superior to the four other treatment modalities at 12-month follow-up. At 60-month follow-up, treatments showed mostly comparable effectiveness.

Memorable Experiences in Therapeutic Assessment: Inviting the Patient's Perspective Following a Pretreatment Randomized Controlled Trial

ABSTRACT Accumulating evidence documents the efficacy of Therapeutic Assessment (TA) in terms of symptom reduction and other outcomes, but only minimal data speak to the patients perspective of what is memorable, or potentially important, about this intervention. In line with the humanistic and phenomenological philosophy of TA, we solicited patient input by asking personality disorder (PD) patients who participated in a recent randomized controlled trial (De Saeger et al., 2014) about their experiences. We report on 10 PD patients who were administered semistructured interviews designed to assess an in-depth perspective of undergoing TA. Our methodological approach can be described as phenomenological and integrative, approximating guidelines provided by the Consensual Qualitative Research paradigm (Hill, 2012). Four core content domains emerged from the transcribed and coded interview protocols: (a) relationship aspects, (b) new insight into personal dynamics, (c) sense of empowerment, and (d) validation of self. Novel experiences were mostly of a relational nature, and pertained to feeling of being treated like an equal and essential partner in a highly individualized venture. Research and clinical implications of these patient reports of TA participation are discussed.

Estimating Subgroup Effects Using the Propensity Score Method A Practical Application in Outcomes Research

Objective:Our aim was to demonstrate the feasibility of the univariate and generalized propensity score (PS) method in subgroup analysis of outcomes research. Methods:First, to estimate subgroup effects, we tested the performance of 2 different PS methods, using Monte Carlo simulations: (1) the univariate PS with additional adjustment on the subgroup; and (2) the generalized PS, estimated by crossing the treatment options with a subgroup variable. The subgroup effects were estimated in a linear regression model using the 2 PS adjustments. We further explored whether the subgroup variable should be included in the univariate PS. Second, the 2 methods were compared using data from a large effectiveness study on psychotherapy in personality disorders. Using these data we tested the differences between short-term and long-term treatment, with the severity of patients’ problems defining the subgroups of interest. Results:The Monte Carlo simulations showed minor differences between both PS methods, with the bias and mean squared error overall marginally lower for the generalized PS. When considering the univariate PS, the subgroup variable can be excluded from the PS estimation and only adjusted for in the outcome equation. When applied to the psychotherapy data, the univariate and generalized PS estimations gave similar results. Conclusion:The results support the use of the generalized PS as a feasible method, compared with the univariate PS, to find certain subgroup effects in nonrandomized outcomes research.