Hospital Use

Medical science and medical opinion narrowly constrain clinical decisions about some conditions. For example, the severity of the illness dictates that patients with hip fracture are almost always hospitalized. But in treating other conditions, physicians have a good deal of discretion; for example, not all patients who break their arms are hospitalized. In these cases, physicians differ in their propensity to treat patients either in or outside the hospital and in their inclination to use surgery or to treat the fracture with a cast. Differences in clinical decision-making such as these are the immediate source of a good deal of the variation in rates of service among hospital referral regions. Although the patterns of practice vary across regions, they are to a remarkable degree constant within a region from year to year.

For some conditions, the majority of variation is associated with per capita supply of hospital beds. The influence of the supply of hospital beds on clinical decision-making does not uniformly apply to all conditions. Because the incidence of hip fracture determines the rate of hospitalization for hip fracture victims, the local supply of hospital beds has little influence on the rate at which patients with broken hips are hospitalized. The local supply of hospital beds has a modest relationship with the discharge rates for surgical conditions. In the case of common medical conditions, however, the local supply of staffed hospital beds has a critical influence on the relative risk of hospitalization.

The counts of discharges (numerators) for medical conditions are determined from the Medicare Provider Analysis and Review (MedPAR) file. Discharges are classified as medical or surgical using the Medicare diagnosis-related group (DRG) system. Enrollee counts were obtained from the Medicare Denominator file. The Medicare enrollee population includes those alive and age 65 to age 99 on June 30 of the measurement year. Measures based on a count of fewer than 11 patients are not displayed for reasons of patient confidentiality. Rates with fewer than 26 expected events are also suppressed because of a lack of statistical precision. These cells are marked "na." The rates are adjusted for the age, sex and race of the underlying Medicare population using the indirect method.


© 2019 The Trustees of Dartmouth College | 35 Centerra Parkway, Lebanon, NH 03766 | (603) 653-0800

The Dartmouth Atlas of Health Care is based at The Dartmouth Institute for Health Policy and Clinical Practice and is supported by a coalition of funders led by the Robert Wood Johnson Foundation, including the WellPoint Foundation, the United Health Foundation, the California HealthCare Foundation, and the Charles H. Hood Foundation.