The Relationship Between Intensity and Duration of Medical Services and Outcomes for Hospitalized Patients

Much recent research on the delivery of care in health care systems has concentrated on the issues of cost containment and assurance of quality care. The research reported here is part of a larger study examining the organizational and contextual determinants of service intensity and duration, costs, and quality of outcomes for hospitalized patients in a stratified random sample of short-term hospitals in the United States. The focus of this paper is on the relationship between the services received by patients in a hospital and their resultant outcomes, i.e., do hospitals which provide more services to their patients achieve better outcomes? The adequacy of previous research bearing on this relationship is briefly reviewed and the problems for defining these measures and for making comparisons among hospitals with differing case mixes of patients are described. Data for assessing the amount and mix of services and the outcomes of patients were obtained from an abstract of each patient's medical record for over 600,000 patients treated in 17 hospitals. Both for the intensity and duration of services and for outcome, an empirical standardization procedure is used to predict the level of services and outcome required by a patient, given his type of disease and physical status as indicated by various demographic and medical history attributes recorded on the abstract. This procedure allows comparisons of levels of services and outcomes for patients which take into account differences in patient populations. Thus the question being asked finally is whether services in excess of those typically provided to this type of patient produce outcomes better than expected. Using a composite measure of the intensity of specific medical services reflecting the mix, amount, and relative costliness of the services provided, out data clearly supported the expectation that hospitals providing higher than expected levels of specific services to their patients also had better outcomes than expected. When a measure of duration of services—the number of days stay—was used, hospitals which kept their patients longer than expected had worse outcomes than expected for their patients. However, when these results were examined for regional effects, strong evidence for regional variations in medical practice was found, especially for duration of services. Further, when the relationship between services and outcomes was examined for hospitals within regions, there was no longer any relationship between duration of services and outcomes. In contrast, even within regions, hospitals providing more specific services to patients also had better outcomes.