SAGA Costs Comparison - February, 2001

You asked me to compare costs per person for SAGA medical recipients at the Hill Health Center (HHC) in New Haven vs. those costs for recipients statewide. The premise that there might be a difference is based on the fact that the HHC has had a system of case management in place for GA recipients for over a decade.

The Hill Health Center cared for 3002 SAGA consumers in FY 2000 and billed a total of $1,437,810 for physician services, clinic services, transportation, dental and lab (not X-ray, although the state's accounting system merges those two) during FY 2000. This comes to an annual cost of $479/person.

Statewide in June 2000, the state spent $1,242,080 on those same services (including X-ray) for a caseload of 22,563 (for the next month - July 2000). Annualizing that month's expenses comes to $660.06/person.

In this analysis, HHC clients cost 27% less than the average CT SAGA medical client did. If these savings could be extended to the rest of the population and other SAGA services, the state could realize a savings of $28,091,086 each year. This is more than double the $11,930,000 savings proposed by the Governor for FY 02 from service reductions.

If these savings are realistic, it is even possible that a statewide system of care management by community health centers could save enough money to allow an expansion of SAGA medical eligibility to higher income levels.

There are several sources of potential error in these calculations. Three come to mind:

1. The statewide service costs were based on only one month - June 2000. While total year costs are available, the only caseload numbers I could find were monthly. It is impossible to determine how many clients enter and leave the program each month to get a yearly, unduplicated count. However, the June 2000 costs are not based on services provided during that month (as we use a modified cash basis for accounting, not GAAP, but that is another story). That number represents bills paid by DSS during that month; the service dates probably vary over a number of months. This introduces some randomization into the analysis, which helps reduce the error.
2. The numbers are not rate-adjusted for health status. While it is possible that the HHC clients were healthier than the statewide average, the inverse is equally possible.
3. Only some services were included in this analysis, as those are the only services over which the HHC has control. While it is possible that the services selected are more sensitive to case management, it is unlikely. In other states with Primary Care Case Management, these services (outpatient) increased and in-patient hospitalization (the greatest part of SAGA medical by far) were reduced over fee-for-service levels.