| OCR Text |
Show Excellence in Public Health (APEXPH), a community assessment model that is currently being used by several Utah local health departments (National Association of County Health Officials, 1991). In preparing this report, categories of disease were chosen according to the first listed diagnosis code. Diagnostic codes were grouped to be comparable with those used by the National Center for Health Statistics to facilitate comparisons. An example of data to be included in that report is included as Table 3 (see page 34). Monitoring Health Care Quality. Access, and Utilization Studies of hospital discharge rates were one of the earliest population-based methods used to examine health care utilization. A notable finding of that work was that hospitalization and procedure rates varied substantially among different geographic areas (Bunker, 1970; Wenriberg, 1973). While need for care is one determinant of hospitalization, the observed variation does not appear to be due to variation in need alone (Connell et al., 1984; Roos & Roos, 1982). Different patterns of variation in utilization have been described (Paul-Shaheen, Clark & Williams, 1987), including areas that exhibit high (or low) rates for most procedures or conditions (Paul-Shaheen, Clark & Williams, 1987; Wennberg et al., 1989; Perrin, Homer, Berwick et al., 1989; Fisher, Wennberg, Stukel & Sharp, 1994; Wennberg & Gittelsohn, 1982) and areas within which utilization is high for only one or a few conditions (Chassin, 1993). In a climate of concern about progressively increasing health care spending, this variation in utilization that seems not to be explained by need for care has been viewed as a potential source of improved quality and of reduced expenditures. Some evidence suggests that in areas with high rates, patients are being managed in a hospital setting who might have been managed as outpatients in other areas. (Connell et al., 1984) However, other studies have not demonstrated that a large proportion of admissions were equivocal or inappropriate in areas with high rates (Chassin, 1993; Chassin et al., 1986; Leaps, Park, Solomon et al., 1990). In some cases, high utilization in an area has been found to be the result of one or a few physicians (Wennberg & Gittelsohn, 1982), and providing the results of variation studies to physicians in communities with high rates has resulted in reduced utilization in those communities (Wennberg, Soule, Keller, Conway & Sheehan, 1994). Though the reason for much of the observed variation in hospitalization and procedure rates remains unexplained, that experience suggests that, used properly, such data have substantial potential for improving care and reducing costs. A preliminary analysis of variation in rates of selected procedures in Utah revealed variation that might be pursued by further studies to elucidate the causes. Examples of variation by local health district are shown in Figures 3 and 4 (see pages 36 and 37) for two conditions, prostatectomy and hysterectomy. Hospitalizations for Ambulatory Care Sensitive Conditions One example of using hospital discharge data to monitor utilization patterns and guide efforts to improve both quality and access has been ambulatory care sensitive conditions (ACS). An ACS condition is a disease or other condition, the hospitalization for which can be prevented or reduced in frequency if timely and effective outpatient care is provided. Hospitalization rates for ACS conditions are calculated and compared among geographic areas or among populations characterized by health insurance coverage (e.g. Medicaid versus private insurance). In general, ACS conditions have been identified based on clinical expert judgment. Published studies have supported the use of the selected ACS conditions by showing that rates of those conditions correlated with other markers of access to outpatient care, such as Medicaid coverage and socioeconomic status, and with survey respondents' perceived access to care.3 Examples of conditions for which hospitalization should be avoidable with good access to outpatient care include asthma, congestive heart failure, diabetes, and dehydration. It is important to note that such analyses do not suggest that all such hospitalizations are avoidable. Rather, such analyses are used to identify areas or populations that have high rates of such hospitalizations. Those results might indicate an area where high rates for several ACS conditions suggest that access to outpatient care in general needs improvement. Alternatively, an area with a high rate for just one condition might suggest the need for improvement in quality of care and access aimed specifically at that condition. ACS conditions analyses have been used by a number of cities and states (Carty, Celebi, Shields & Edwards, 1994; Krasner, 1994), and at least one has begun a community-wide effort to prevent asthma admissions based on the findings of an ACS analysis (Carty, Celebi, Shields & Edwards, 1994). However, rigorous evaluations of the utility or validity of such analyses have not been conducted. A full analysis of ACS conditions in Utah has not been completed, but a selection of ACS conditions has been included in the 1995 edition of Utah's Health: An Annual Review [See Tables 68 and 69, page 86]. Figure 5 (see page 28 Hospital Discharge Data |
