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.

The study compared two management options, usual care and intervention, for psychiatric inpatients. The usual care strategy was administered only by psychiatric house staff who had monthly rotations and by permanently assigned psychiatric attending physicians of the unit. In the intervention strategy, care was administered after collaboration of an internist with the usual psychiatric team.

Treatment.

Cost-effectiveness analysis.

The study population comprised all patients aged 18 years and above who were admitted to the two inpatient psychiatric units at the authors' setting with a documented diagnosis necessitating hospitalisation. Patients who could not participate in randomisation without medical consultation were excluded from the study.

The setting was secondary care (i.e. a 562-bed hospital that is in collaboration with the University Of Vermont College Of Medicine). The economic analysis was carried out in the USA.

The effectiveness and resource use data were collected from March 2001 until January 2002. The cost data were derived from the authors' setting but the price year was not reported.

The effectiveness data were derived from a single study.

Although not explicitly stated, it seems that the costing has been carried out prospectively on the same sample of patients as that used in the effectiveness study.

The sample size was not determined in the planning phase of the study. Consecutive patients who were admitted to the inpatient psychiatric units of Fletcher Allen Health Care in Burlington, Vermont were selected for inclusion in the study. It was reported that of the 272 patients who were admitted to the psychiatric units, 12 required medical consultation and were excluded from the study while 121 refused to participate. A total of 139 patients were randomly assigned to the intervention group (n=55) or to the usual care group (n=84).

The analysis was based on a single-centre randomised controlled trial. The patients were randomly assigned to the two groups, but the methods of randomisation or blinding were not reported. The patients were followed up for 1 year after discharge. Although the authors stated that data were not available for all patients at the end of the follow-up period, they did not report reasons for losses to follow-up and the percentage of the overall sample they comprised.

It was not reported whether the analysis was conducted on an intention to treat basis. The primary proxy health outcome used was health maintenance. Health maintenance referred to services that were either completed during hospitalisation or were programmed after discharge. Such services included alcohol risk plan, tobacco risk plan, mammogram, Pap test, digital rectal examination and prostate specific antigen test, tetanus vaccine, flu vaccine, pneumonia vaccine and lipid screening. The authors reported that the patient groups participating in the study were comparable at baseline in terms of their demographic characteristics. However, there were statistically significant differences between participants and non-participants (due to exclusion or refusal) in age and co-morbidity.

The health maintenance outcomes were as follows.

The completion of services like tobacco risk plan, mammogram, Pap test, tetanus vaccine, flu vaccine and lipid screening were statistically significantly different between the two groups, (p<0.001), while there were no statistical differences for the alcohol risk plan and digital rectal examination and prostate specific antigen test. The health maintenance summary score was 3 (+/- 7) in the usual care group and 56 (+/- 34) in the intervention group (z = -8.88, d.f. 135, p<0.001). The coordination of care and the equivalent summary score did not differ significantly between the two groups.

The addition of an internist to the psychiatric team resulted in improved health maintenance outcomes.

The authors did not derive a summary measure of benefit in the economic analysis. In effect, a cost-consequences analysis was performed.

It is impossible to know which aspects of costs were included in the analysis since only hospitalisation summary costs were reported. The costs and the quantities were not reported separately. The costs were based on charges from the authors' setting. Discounting was not relevant as the costs were incurred during less than 2 years. The price year was not reported.

The authors reported mean values with standard deviations (SDs) for hospitalisation costs. Differences in costs between the two groups were analysed using the Wilcoxon rank sum test. All analyses were conducted using STATA 8.2 software.

The indirect costs were not included in the analysis.

US dollars ($).

No sensitivity analysis was conducted.

See the 'Effectiveness Results' section.

The total hospital costs (presented as mean +/- SD) were $8,527 (+/- 6,512) in the usual care group (n=84) and $8,558 (+/- 5,703) in the intervention group (n=55).

The difference in costs between the two groups was not statistically significant (z = -0.41, d.f. 137, p=0.68).

The costs and benefits were not combined.

"Adding an internist to an inpatient psychiatric team is an effective way of improving care for this traditionally underserved population without increasing cost."

The authors compared the collaboration between an internist and psychiatric team in the provision of preventive health services to psychiatric inpatients. However, they did not discuss the existence of alternative strategies for improving care among psychiatric patients. If there were any, which is likely, it makes this study only a partial analysis.

The analysis was based on a single-centre randomised controlled trial, which was appropriate given the study question. The study sample was representative of the study population and the patient groups were shown to be comparable at analysis. However, documented statistical differences between participants and non-participants were not taken into account in the final analysis, and potential biases and confounding factors were not addressed. Methods of randomisation and losses to follow-up were not reported, which represent a limitation to the internal validity of the study. Although not explicitly stated, it would appear that the analysis was conducted for treatment completers only. This can make the intervention look more or less effective than it actually is.

The authors did not derive a summary measure of benefit in the economic analysis. In effect, a cost-consequences analysis was performed.

Although not explicitly stated, it appears that the analysis has been conducted from the perspective of a hospital since only hospitalisation costs were included. Various costs related to preventive care (i.e. tests like mammogram) were not included in the analysis, thus the total costs of the intervention might have been underestimated. The costs and the quantities were not reported separately, although resource use in terms of length of hospital stay was recorded. However, this would not enable the analysis to be easily reworked for other settings. No sensitivity analysis on the costs or quantities was conducted, and this may limit the interpretation of the study findings. Charges were used to proxy costs; however, the same procedure was followed in both groups, making the results of the two groups comparable. The price year was not reported, which will not enable any future reflation exercises to be conducted.

Since the authors did not compare their findings with those from other studies, it is not known how far their results agree with other published results. However, this might have been due to a lack of published studies in the same area. The issue of generalisability of the results to other settings was addressed. The authors do not appear to have presented their results selectively. The study enrolled psychiatric inpatients but the authors seem to have generalised their results across all psychiatric patients.

The authors reported a number of limitations to their study. First, a high proportion of eligible patients refused to participate in the study. In addition, there were no robust data (i.e. visits and tests) on losses to follow-up. Second, various important health outcomes were not taken into consideration in the current study. For example, transfers to other units for more intensive or specialised care, or changes in the percentage of patients with medical problems who visited inpatient psychiatric units because of the presence of an internist.

The authors did not make explicit recommendations for changes in policy or practice or for further research. However, the discussion highlighted areas where more research-based information could be useful.

Supported by a Research Development grant from Fletcher Allen Health Care.