Monthly Hospital Admission Data Used to Monitor New Hampshire Health Insurance

Monthly Hospital Admission Data

Hospitals submit certain data monthly to the New Hampshire Hospital Association. Among the elements submitted are a monthly count of inpatient admissions to self-pay patients and the total number of inpatient admissions during that month. The hospital association then prepares quarterly "Trending Reports" that include 89 different measures of hospital operations. None of the measures on these reports currently relates to the number or percentage of patients who are primarily "self-pay."

Instead of waiting for the UHDD to become available, the hospital association could calculate the percentage of admissions that are to self-pay patients each quarter and share that with the Endowment for Health and the NH Center for Public Policy Studies. The three organizations could then make regular quarterly public announcements about the new numbers and any implications for health care in New Hampshire.

Table 2 displays the 1999-2004 data from the NH Hospital Association and compares the quarterly admissions data to the quarterly discharge data. There will always be small differences because the number of admissions and discharges during any month or quarter will not be equal and whether a patient is recorded as self-pay may change during the person's hospital stay.

Table 2: Quarterly admissions and discharge data 1999-2004

It appears that the differences between the two data sets are minor. The more rapidly available admissions data from NHHA should be used to monitor any changes in the uninsured population.

When the bureau releases its full year of discharge data many months later, more detailed analysis could be performed on the characteristics of the uninsured, such as their age and sex, medical diagnoses, etc.

The percentages in Table 2 could be misinterpreted to be the actual percentage of the entire population that is uninsured. As pointed out above, this is not the case. Only the change in the percentage over time indicates a change in the insurance status of the population. In order to avoid this potential misinterpretation, the data can be normalized and presented as an index. The average percentage among hospital admissions in 2001 was 5.21 percent. Benchmarking the indicator to this percentage allows the measure in other time periods to be stated as a multiplier of this rate. We have set the benchmark period to 100.0 in Table 3.

The moving average smoothes out some of the seasonal variation and random statistical fluctuations.

Table 3: Index data