The trade-off between costs and quality of care in the treatment of psychosomatic patients with somatoform pain disorder.

BACKGROUND: The introduction of efficiency-oriented provider payment systems in inpatient mental healthcare in various Western countries may lead to the use of less healthcare resources in the treatment of patients. To avoid unintended effects on quality of care that may result from reductions in resource utilization, it is essential for decision and policy makers to know whether there is a trade-off between costs and quality of care. AIM OF THE STUDY: The aim of this study was to investigate and quantify the relationship between costs and outcomes in psychosomatic inpatients with somatoform pain disorder. METHODS: The inclusion criteria for patient selection (n = 101) were (i) a main diagnosis of somatoform pain disorder according to International Classification of Diseases-10 (ICD-10) [F45.4, F45.40, F45.41]; (ii) complete data on the mental component summary reflecting overall functioning of mental health (MCS-8) measured with the Short Form-8 Health Survey (SF-8) within 3 days of the admission and discharge dates; and (iii) treatment at Charité Universitaetsmedizin (Berlin, Germany) during the period January 2006-June 2010. The change in the MCS-8 score incurred over the treatment period was used as an indicator of quality of care. Treatment costs were calculated from the provider's perspective, mainly using bottom-up micro-costing. The year of valuation for cost calculation was 2008 (with no inflation adjustment); for costs provided by the accounting department for services consumed by the patient, the valuation year was based on the year of service provision. We hypothesized that the outcome 'change in MCS-8 score' was a function of the independent variable costs, patient characteristics, socio-demographic variables, pain-related variables, co-morbidities and subjective illness attribution, i.e. whether patients attributed the origin of pain mainly to a somatic cause or not. An interaction term between costs and illness attribution was included to control for the hypothesized differing effects of resource input or costs on the outcome variable conditional on patients' illness attribution. Hausman tests indicated that endogeneity was not present, thus, ordinary least squares regression (OLS) was conducted. We assessed whether the change in the MCS-8 score was clinically meaningful and perceptible by the patient, using the minimal clinical important difference (MCID). For Short Form Health Surveys, the MCID for changes in the mental component summary is typically around 3 points. RESULTS: We found a trade-off between costs and outcome for patients without or with only minor somatic illness attribution (77 % of the sample). This patient group improved 0.4 points in outcome after every 100 increase in total costs per case (F 1,77 = 13.836, t(77) = 3.72, p = 0.0004). For patients with mainly somatic illness beliefs (23 % of the sample), we did not find a trade-off between costs and outcome. CONCLUSION: For the majority of patients, we found a trade-off between costs and health outcome, thus, it seems advisable to carefully monitor outcome parameters when applying cost containment measures.

Introduction of DRG-based reimbursement in inpatient psychosomatics--an examination of cost homogeneity and cost predictors in the treatment of patients with eating disorders.

OBJECTIVE: Various western countries are focusing on the introduction of reimbursement based on diagnosis-related groups (DRG) in inpatient mental health. The aim of this study was to analyze if psychosomatic inpatients treated for eating disorders could be reimbursed by a common per diem rate. METHODS: Inclusion criteria for patient selection (n=256) were (1) a main diagnosis of anorexia nervosa (AN), bulimia nervosa (BN) or eating disorder-related obesity (OB), (2) minimum length of hospital stay of 2 days, (3) and treatment at Charité Universitaetsmedizin Berlin, Germany during the years 2006-2009. Cost calculation was executed from the hospital's perspective, mainly using micro-costing. Generalized linear models with Gamma error distribution and log link function were estimated with per diem costs as dependent variable, clinical and patient variables as well as treatment year as independent variables. RESULTS: Mean costs/case for AN amounted to 5,251€, 95% CI [4407-6095], for BN to 3,265€, 95% CI [2921-3610] and for OB to 3,722€, 95% CI [4407-6095]. Mean costs/day over all patients amounted to 208€, 95% CI [198-218]. The diagnosis AN predicted higher costs in comparison to OB (p=.0009). A co-morbid personality disorder (p=.0442), every one-unit increase in BMI in OB patients (p=.0256), every one-unit decrease in BMI in AN patients (p=.0002) and every additional life year in BN patients (p=.0455) predicted increased costs. CONCLUSION: We see a need for refinements to take into account considerable variations in treatment costs between patients with eating disorders due to diagnosis, BMI, co-morbid personality disorder and age.

Inpatient costs and predictors of costs in the psychosomatic treatment of anorexia nervosa.

OBJECTIVE: In German inpatient psychosomatics per diem lump sums will be introduced as reimbursement rates by 2013. It was the aim to calculate total inpatient costs per case for the psychosomatic treatment of patients with anorexia nervosa and to identify cost predictors. METHOD: The sample comprised of 127 inpatients. Cost calculation was executed from the hospital's perspective, mainly using microcosting. Medical records provided data on patient characteristics and individual resource use. Two generalized linear models with gamma distribution and log link function were estimated to determine cost predictors by means of demographic data, comorbidities, and body-mass-index at admission. RESULTS: Inpatient costs amounted to 4,647 €/6,831 US$ per case (standard deviation 3,714 €/5,460 US$).The admission BMI and "Disorders of Adult Personality and Behavior" were significant cost predictors (p < 0.05). DISCUSSION: The formation of patient groups within the diagnosis anorexia nervosa should be oriented towards the determined cost predictors.