|
Effectiveness and cost of the inpatient treatment of posttraumatic stress disorder: comparison of three models of treatment |
Fontana A, Rosenheck R |
|
|
Record Status This is a critical abstract of an economic evaluation that meets the criteria for inclusion on NHS EED. Each abstract contains a brief summary of the methods, the results and conclusions followed by a detailed critical assessment on the reliability of the study and the conclusions drawn. Health technology Inpatient psychiatric therapy for the treatment of post-traumatic stress disorder (PTSD), in veterans of the Vietnam War.
Economic study type Cost-effectiveness analysis.
Study population Veterans of the Vietnam War, with PTSD.
Setting The practice setting was inpatient units. The economic study was carried out in Connecticut, USA.
Dates to which data relate Effectiveness and resource use data related to the period from November 1991 to June 1995. The price date was 1995.
Source of effectiveness data Effectiveness data were derived from a single study.
Link between effectiveness and cost data Costing was undertaken prospectively on the same patient sample as that used in the effectiveness analysis.
Study sample Veterans attending one of ten Department of Veterans Affairs (VA) sites, who were eligible and who consented, were invited to participate in the study. A total of 785 veterans were recruited, 333 from long-stay PTSD units, 222 from PTSD short-stay units and 230 from general psychiatric units. Of those invited to participate from the long-stay PTSD units, 16.1% refused; the corresponding figure for the short stay PTSD unit was 8.9% and for the general psychiatric units, 21.4%. It was not stated whether power calculations were used to determine sample size. All veterans were male, with an average age of 45.2 (+/- 3.2) years. Veterans had an average of 13 (+/- 2.0) years of education.74.3% were Caucasian and 15.6% were African American.38.5% were married.
Study design The study was a non-randomised trial with concurrent controls. Veterans were followed-up at 4-month intervals for 1 year after discharge. Loss to follow up was not reported, but successful data collection averaged 66.1% across the three follow-up intervals. An 'independent' evaluator assessed outcomes.
Analysis of effectiveness The analysis was based on intention to treat. The main health outcomes used in the analysis were nine psychometric measures of symptoms and social functioning. The measures included 2 PTSD scores, 2 psychiatric symptom scores, 2 addiction severity indices, 1 rating for violent behaviour/thoughts, 1 rating for social involvement and, lastly, the number of days worked over the previous month. Groups were shown not to be comparable in clinical symptoms at the time of analysis.
Effectiveness results Changes in outcomes from admission to the end of the follow-up year were significantly different across the models for four of the nine outcome measures: PTSD as measured by the Clinician-Administered PTSD Scale, general distress as measured by the Addiction Severity Index and the Brief Symptom Inventory, and alcohol abuse. Generally, veterans in the short-stay PTSD units and the general psychiatric units showed sustained improvement in these areas, while veterans in the long-stay PTSD units showed a return to the levels at the time of admission.
Clinical conclusions Improvements in the short-stay PTSD units and the general psychiatric units were greater than those observed in the long-stay PTSD units. This was observed both from admission to discharge and over the subsequent year, even after adjustments were made for differences between groups in veterans' characteristics at admission.
Measure of benefits used in the economic analysis Effectiveness estimates were not converted to a measure of health benefit.
Direct costs Costs were estimated from the perspective of the health system. Costs and quantities were not reported separately. Resource use data was derived from structured interviews with patients and checked against patient records and included inpatient and outpatient care. Prices were taken from a VA report, which included overheads and capital costs; for non-VA services, prices were inflated by a factor of 1.6. Costs were compared across the three programme types for 3 time periods, including the year from the time of admission.1995 prices were used.
Statistical analysis of costs The mean and standard deviation were reported for each cost, and p values were estimated for cost differences using a chi-squared test, based on a one-way analysis of variance (ANOVA).
Sensitivity analysis A sensitivity analysis was not performed.
Estimated benefits used in the economic analysis Estimated benefits were proxied by the effectiveness estimates.
Cost results The mean per-patient long-stay PTSD unit cost, for the year following admission, was $47,091 (+/- $20,953). The equivalent short-stay PTSD unit cost was $25,809 (+/- $19,614) and the general psychiatric unit cost was $30,676 (+/- $25,881). Analysis of variance showed statistically significant differences between the three groups, but pair-wise comparisons were not reported.
Synthesis of costs and benefits A formal synthesis of costs and benefits was not performed.
Authors' conclusions Long-stay specialised inpatient PTSD units are costly, relative to short-stay specialised evaluation and brief treatment PTSD units and general psychiatric units. Higher gains in symptom reduction or in social functioning were observed in the short-stay PTSD group, relative to the long-stay units. Systematic restructuring of Veterans Affairs inpatient PTSD treatment could result in delivery of effective services to a larger number of veterans.
CRD COMMENTARY - Selection of comparators The reason for the choice of the comparators is clear, as they represented current practices in the authors' settings. You, as a database user, should consider if this applies to your own setting.
Validity of estimate of measure of benefit Although a comprehensive set of outcome measures was used, the lack of comparability between the characteristics of the three groups limits the interpretability of the results. The authors stated that adjustments were made to compensate for this, but did not report which methods were used. The authors noted that although differences between the groups were statistically significant, they were 'more modest clinically'.
Validity of estimate of costs Although costs were analysed statistically, costs and quantities were not reported separately. This limits the generalisability of the results to other settings.
Other issues The authors' conclusions, that a move to short stay inpatient units is cost effective within the study setting, are not justified by the study findings.
Bibliographic details Fontana A, Rosenheck R. Effectiveness and cost of the inpatient treatment of posttraumatic stress disorder: comparison of three models of treatment. American Journal of Psychiatry 1997; 154(6): 758-765 Indexing Status Subject indexing assigned by NLM MeSH Attitude to Health; Delivery of Health Care /economics; Follow-Up Studies; Health Care Costs; Hospitalization /economics; Hospitals, Veterans /economics; Humans; Length of Stay; Male; Mental Disorders /psychology /therapy; Middle Aged; Psychiatric Department, Hospital /economics; Psychiatric Status Rating Scales; Severity of Illness Index; Social Environment; Stress Disorders, Post-Traumatic /economics /psychology /therapy; Treatment Outcome AccessionNumber 21997000799 Date bibliographic record published 30/04/1999 Date abstract record published 30/04/1999 |
|
|
|