Treatments for some conditions are likely to be expensive, some much less so. Together, the seven conditions examined account for about 46% of all spending. Some treatments for included conditions are arguably quite predictable, such as birth-related spending, while others might be considered more random, such as injuries and poisonings. More info We classify all health care claims according to the primary diagnosis attached to the claim.
Patterns of Spending
Table 1 provides an overview of the patterns of utilization across the service types identified. By and large utilization of different service types is relatively stable over the three years observed. The most notable change is the reduction in the share of people with birth related spending. In 1991, 25.7% of the sample had birth-related spending compared to 16.7% in 1993. Birth of a child may have initiated a period of eligibility for some of these women, accounting for the elevated rate in the first year.
Table 2 describes patterns of utilization and spending for the sample in 1993. Birth related spending has the highest expected level of spending of all the listed conditions ($653) or 19.2% of total spending. Most of the other conditions have expected spending levels of $126 to $250 or between 3% and 7% of total spending.
The sixth and seventh columns of Table 2 give an indication of the correlation of spending on a service with spending on other types of care, and of predictability, key elements of the formula for shadow prices (11). The sixth column reports the correlation between spending on each of our nine service categories and the sum of spending on all other services. In general, these correlations are quite low. None of the eight service-specific correlations exceed 0.20, with the exception of the “other” category. Gastrointestinal care, cancer care and treatment of injuries and poisonings have the highest correlation with all other types of spending.
Correlation with spending in the previous year for each category indicates persistence of spending. Persistent spending is probably more predictable. Several of the illness thought to be more chronic in character, hypertension, mental health/substance abuse and musculoskeletal conditions, display relatively high correlations in service-specific spending over time. Mental health spending has the highest year-to-year correlations.
Estimation of Components of the Ratio of Shadow Prices
Risk Adjusted Premiums: We first calculate the premium assuming that a single payment is made for all enrollees. This premium is based on the simple average level of spending across all enrollees and corresponds to a case with no risk adjustments. We next construct two sets of true “risk adjusted” premiums, one based on the Ambulatory Diagnosis Group (ADG) classification system (Weiner et al., 1996,) and one based on the DCG classification system (Ellis et al., 1996).10 In each case we adjusted the risk adjustment upward to make the marginal profit per enrollee positive on average, as it must be if plans are to be induced to compete for enrollees by service quality.