Results of the Feasibility Study of Estimating the Risk of Meal Claiming Error in the CACFP

Measuring erroneous meal claims in the Family Day Care Home (FDCH) component of the Child and Adult Care Food Program (CACFP) poses great difficulty. FDCH providers report meal service counts, by day, at the end of each month but do not submit independent objective documentation to confirm the accuracy of the reports.

Therefore, the Food and Nutrition Service’s 2006 Improper Payments Information Act (IPIA) plan for the CACFP proposed to test the feasibility of estimating the risk of errors. The Food and Nutrition Service (FNS) proposed to select a random sample of sponsoring organizations and, from each, use a random selection of the sponsor’s monitoring visits of FDCHs.¹ Using these data, FNS would compare the number of meals claimed with the number of children observed at the time of the visit.

Although the IPIA requires measurement of both overpayments and underpayments, this approach can only estimate the risk of overpayment errors. Specifically:

• If a FDCH provider claims more meals than the number of children observed during the monitoring visit, it suggests there is a risk of overpayment. The number of children observed is an imperfect yardstick – for example, a child could be dropped off at an FDCH after the observer left but before a meal service ended – but for the most part if many more meals are claimed than children observed we know that some of the discrepancy is overpayment.
• On the other hand, if a FDCH provider claims fewer meals than the number of children observed during the monitoring visit, it is not an underpayment since neither the government nor the sponsoring organization denied a claim for reimbursement. Fewer meal claims than children present represents irrational behavior on the part of the provider so we assumed that this would be a rare occurrence.²

The purpose of the pilot test was to determine the feasibility and accuracy of this approach to estimating the risk of overpayments in the FDCH component of the CACFP.

¹ We proposed a similar exercise using state monitoring visits. There were difficulties in the coding and cleaning of these data and the database is only half complete. We do not believe the results using state visits would differ from the ones reported here based on sponsor visits.
² In a small number of cases, a provider that serves more meals than the CACFP’s three meal per child per day limit might not claim a snack so that they could claim the higher reimbursement meals, such as lunch. This is neither an overpayment nor an underpayment (but is a bargain for the taxpayer). It also only makes sense to be seen serving a snack that is not claimed if they are in fact serving the other meals.