• What else can charity care do?

    The conclusions reached by Douglas Almond, Janet Currie, and Emilia Simeonova are very interesting:

    This paper explores the consequences of the expiration of charity care requirements imposed on private hospitals by the Hill–Burton Act. We examine delivery care and the health of newborns using the universe of Florida births from 1989 to 2003 combined with hospital data from the American Hospital Association. […]

    Private hospitals whose [Hill-Burton charity care] obligations expired reduced the provision of maternity care, and shifted mothers to public hospitals. Perhaps surprisingly, mothers in public hospitals received additional services, but did not benefit from any measurable improvements in outcomes. […]

    [W]e find some suggestive evidence that the practice patterns adopted by private hospitals that served Hill–Burton patients spilled over onto the rest of their caseload. Thus, constraining private hospitals to serve the indigent might help to “bend the cost curve” by encouraging hospitals to keep costs low. On the other hand, it is clear that private providers steered clear of the highest risk charity patients, so that a public safety net was a key option for these patients.

    In other words, private hospitals reduce costs by reducing unnecessary care in response to charity care requirements. The practice pattern that gives rise to the cost savings spills over to all patients, with no adverse effects on health. So, what will happen when charity care and requirements to provide it are diminished?

    The authors only examined child birth and C-sections. They did address the problem of selection of lower risk patients into Hill-Burton obligated private hospitals with an instrumental variables approach. Obviously, their results depend on the validity of their instrument, the fraction of births in Hill-Burton obligated hospitals in the county. That’s clearly related to the likelihood of giving birth in such a hospital. I don’t immediately see how it could be related to an individual-level birth outcome. Hence, I am inclined to believe the instrument.

    Share
    Comments closed
     
    • The problem with all these studies that find “no adverse effects on health” is that the only effects on health we can measure tend to be pretty severe – in this case, death of infant before age 1, complications of labor, APGAR score, gestational age.

      Effects of less care at the margin may be minimal on these severe outcomes; the more so with our limited datasets and imperfect instruments. But care at the margin could still have high value.

      Perhaps women receiving less care (e.g. shorter maternal hospital stays) take longer to recover from child birth trauma (or develop more chronic associated issues), have more difficulty nursing, etc.. Their children might have a higher probability of jaundice, early nutrition deficiencies, etc..

      These are all margins that we plausibly care about, and none are measured in the standard datasets…

      I find it very difficult to demonstrate convincingly that there were “no adverse effects on health”