[Patient shifting between inpatient care and ambulatory care - exploring methods to measure the relevance of patient shifting]

Abstract

Background: The financial dimension of long term changes in the medical division of labour between inpatient care and ambulatory care has yet to be systematically monitored. While this is of general interest for health systems research there is now an acute need for the development of reliable methods to measure the effects of shifts in care as part of physician payment reform in Germany. The Social Code Book V (§ 87a Sec 4 No 3) requires the collective contracting partners to determine risk adjusted payment targets for regional populations thereby also taking into account shifts between inpatient and ambulatory care.

Methods: Using predictive modelling patient groups are identified which meet the following two criteria in two consecutive years: 1) increases in actual cost exceeded expected cost in sector a while expected cost exceeded actual cost in sector b; 2) absolute number of cases increased in sector a and decreased in sector b. The model is based on the definition of a limited set of risk groups as defined by the risk adjustment scheme applied to German sickness funds. For our study these risk groups have been calibrated separately for each sector creating a common set of predictors. The second criterion focuses the approach on patient shifting as the most tangible effect of shifted care. In order to quantify the effect of patient shifting another predictive modelling approach is developed using the difference between expected and actual inpatient cases per risk group to estimate the resulting change in ambulatory case load. The cost of the additional case load per risk group is calculated for Germany based on population-based claims data (77 million patients).

Results: The criteria for patient shifting as defined above apply to 26 out of 95 risk groups. At the level of risk groups hardly any patient shifting into ambulatory care was detected. On average for each patient with the respective risk factors 0.6 additional cases in ambulatory care were estimated as result of reduced incidence of inpatient care. In total the additional cost associated with patient shifting from inpatient care to ambulatory care was estimated 424 million € (2007). This represents 1.5% of total spending on ambulatory care and underlines the importance of the issue to health services research. Roughly 80% of this amount is likely to be eligible to physician services relevant to morbidity adjusted targets under payment reform. Prior to implementation as a payment formula, however, the approach needs to be based on a comprehensive risk adjustment model and needs further refinement.