As a primary morbidity in the dialysis population, hospitalizations provide a focal point for provider efforts to improve patient care and outcomes. This chapter presents data on rates of hospitalization among patients treated in freestanding dialysis facilities, overall and by cause inferred from discharge diagnosis codes.
We begin by examining rates during the first year of dialysis. After very little change in admission rates among successive cohorts of incident patients between 1996 and 2005, rates began to decline among incident patients in 2006. The decline has been modest, at slightly more than 1.5 percent per year, but the trend is clear. Maps and tables in this section show striking regional variations in first-year hospitalization rates, with the highest rates in the Middle Atlantic and East North Central Census Divisions, and the lowest in the Mountain and Pacific Divisions. These differences are especially relevant to the interpretation of standardized hospitalization ratios, which are not adjusted for the location of the dialysis facility.
Data for individual states illustrate wide variation not only in hospitalization rates themselves, but also in changes in those rates with time. Since 2003, for example, the first-year admission rate has fallen 1.4 percent per year in the entire Middle Atlantic Division, but 2.1 percent per year in New Jersey. In the East North Central Division, Wisconsin has achieved the greatest decline, at 3.1 percent per year; other states in the division, however, have made far less progress, resulting in a decrease of only 0.3 percent per year for the entire division. First-year admission rates increased in seven states, including Indiana and Michigan; underlying causes of these deleterious changes should be investigated.
We next present parallel figures, maps, and tables for the prevalent population. Seasonality of risk is apparent in this setting, with month-by-month admission rates reaching their apex in the winter months and their nadir in the summer months. Such patterns are similar to those for common communicable diseases, such as influenza, and we investigate them further throughout this chapter.
While admission rates in the prevalent population have fallen across many areas of the country, the states in the East North Central Division have seen little change since 2004, which is a source of concern. Hospitalization rates have also tended to increase in nearby states, including Kentucky and West Virginia. The common geographical feature for all of these states is the Ohio River basin. This feature of the data raises complex questions about environmental pollution, persistent unemployment and poverty, and local capacity for health care (beyond dialysis providers). The extent to which these factors are addressed by risk adjustment schema has received very little attention during the development and promulgation of quality metrics. Admission rates continue to be lowest in the states of the Mountain and Pacific Census Divisions, followed by the West South Central Division, all areas with relatively large shares of Hispanic white patients.