Quality health outcomes depend upon the Social Determinants of Health

On the heels of the publication by the Center for Medicare and Medicaid Services (CMS) of how much money Medicare paid to individual physicians (discussed on this blog in Medicare payments to doctors: the big issue is the underpayment for primary care, April 9, 2014), we have revelations of inequity in Federal payments to health providers. A panel of the National Quality Forum (NQF), convened by the Administration to look at this issue, has determined  that payments for “quality of care” to hospitals under the Affordable Care Act (ACA) tend to reward those hospitals caring for higher-income patients and penalize those who care for the poor (Robert Pear, “Health law’s pay policy is skewed, panel finds”, New York Times, April 27, 2014). This cannot be the way we want to go, and thankfully that is the conclusion reached by the panel. It is, however, the NQF that developed these 600 or so quality indicators, and has not recommended adjusting them for the socioeconomic status of the patients that a hospital cares for (although it does adjust for severity of illness).

What is happening is that the measures of quality of care are largely not measures of what is done for patients in the hospital, but how they do after – are they readmitted shortly after discharge, are the diseases for which they are being cared for under better control or not, do they get follow-up care. The fact is that for a variety of reasons including money, education, transportation, and competing demands, poor people do not do as well as better-off people, even controlling for the quality of care that they receive when they are hospitalized. A number of panel members comment on this in the Times article, including NQF president Christine Cassel, who says “Factors far outside the control of a doctor or hospital — patients’ income, housing, education, even race — can significantly affect patient health, health care and providers’ performance scores,” and panel member Steven H. Lipstein, CEO of BJC HealthCare in St. Louis who adds “The administration’s current policy on adjustments for socioeconomic status are quite inadvertently exacerbating disparities in access to medical care for poor people who live in isolated neighborhoods. I’m sure that’s not what President Obama intended with the Affordable Care Act.”

These comments are true, but the thrust of the NQF’s comments was the unfairness to the hospitals. This is important as far as it goes – it is outrageous to pay extra money for “quality of care” to hospitals that care for the privileged and penalize those that care for the underserved. Many of the members of the panel and other commenters quoted by Mr. Pear focus on academic teaching hospitals, which indeed care for a disproportionate share of poor people; however, public hospitals (in those areas where they exist) are even more affected. But what is more important is how this issue illustrates the power of what are called the “social determinants of health”, the situation that people live in before they access medical care, and after they are discharged, have on health outcomes.

Health advocate and policy expert Kip Sullivan is more pointed in his comments on Don McCanne’s “Quote of the Day” for April 28, 2014. “The notion that doctors and hospitals are screwing up and will behave if they are subjected to punishment and reward by third parties is not new. The Code of Hammurabi (1750 BC) subjected Mesopotamian doctors to a combination of reward (more shekels) and punishment (cutting off of doctors' hands)…But even Hammurabi didn’t recommend punishing the patients.”  If hospitals that care for poor people are effectively financially penalized for doing so, they will (at best) be further financially challenged in providing that care, and at worst will do their best to not care for the poor to the extent that they can.

Why would the NQF and the Department of Health and Human Services (HHS) take such a position, one that seems both unfair and even mean? One might be tempted to suggest that rewarding the “haves” and punishing the “have nots” is what is usually done by government policy, but we would hope that, given its rhetoric on health care – and the creation of the ACA in the first place – the Obama administration would not be guilty of such intent. We get some better idea from Kate Goodrich, the director of quality measurement programs at the federal Centers for Medicare and Medicaid Services, who is quoted in Mr. Pear’s article as saying “We do not want to hold hospitals to different standards of care simply because they treat a large number of low-socioeconomic-status patients. Our position has always been not to risk-adjust for socioeconomic status within our measures because of concern about masking disparities, and potentially rewarding providers who provide a lower level of care for minorities or poor patients.”

Now, this sounds almost noble, like a values-based response to critics such as those on the NQF’s panel. However, the clear and obvious flaw in such logic is that hospitals have the power to change the lives of these patients in such a way as to decrease their risk for poor outcomes to be equal to, or better, than, those of higher socioeconomic status. They don’t, and to the extent that they could do more work in the community to help this situation, it would cost more money, so it is absurd that they be financially penalized. Dollars spent by government on health care should first and foremost be required to be spent on health, not on making money for providers (doctors or hospitals) who can by virtue of their location (and possibly other strategies) avoid taking care of the neediest. Hospitals should be judged and reimbursed on the quality of care that they deliver, equitably and without prejudice with regard to socioeconomic status, but cannot reasonably be judged on outcomes which depend on factors far outside the control of those hospitals.

The real issue is that people who are poor have a lot more to contend with than the services delivered as health care. It is not uncommon for our hospital to be treating a person with a bone infection made worse by their diabetes who needs 6 weeks of IV antibiotics. This can be delivered by a home health care agency, and most insurance will pay for it. But it becomes a problem if the person does not have insurance. And is even more complicated when they do not have a home. These people stay in the hospital, at exorbitant cost, for the whole duration of treatment. But would our quality measures be better if we only cared for those with homes and insurance? Would the hospital make more? Of course, as Mr. Sullivan points out, while the hospitals lose financially, ultimately it is the patients who suffer.

The social determinants of health are well-portrayed in the “cliff analogy” developed by Dr. Camara Jones and her colleagues,[1] and discussed in my blog of September 12, 2010, “Social Determinants, Personal Responsibility, and Health System Outcomes”. The care given by hospitals occurs at the bottom of the cliff, after people have fallen, but their risk, both before arriving at the hospital and in returning home, is that they are living so close to the cliff face; their housing is poor, their neighborhoods are dangerous and polluted, their schools do not educate, and food is often scarce and not nutritious.  In their study comparing health costs in the US and Europe, Elizabeth Bradley and colleagues discovered that while the US spends far more on “health care”, if you add in basic social service spending, the difference decreases, but that the US spends most of its combined health-and-social-service spending on medical care.[2] (Discussed in a New York Times op-ed, “To fix health care, help the poor” by Bradley and Lauren Taylor, and in my blog “To improve health the US must spend more on social services”, November 18, 2011.)

It is understandable that, given the political climate in Washington and state capitals and the flak that they took for ACA, the Obama administration does not want to put major effort into addressing the social determinants of health by developing programs to meet the core needs of poor people in our country, to prevent them from getting sick, to give them access to meaningful post-hospital care, to have health workers in communities, punish polluters, decrease crime, and limit health risks. Understandable, but not OK. And in the meantime, on this narrower issue, it obviously requires adjusting for socioeconomic risks for hospitals caring for the poor when their quality incentive payments are calculated.

But sometime soon we are going to have to address the core problems.

---

[1] Jones CP, Jones CY, Perry GS, “Addressing the social determinants of children’s health: a cliff analogy”, Journal of Health Care for the Poor and Underserved, 2009Nov;20(4):supplement pp 1-12. DOI: 10.1353/hpu.0.0228. Slides available on line at http://www.csg.org/knowledgecenter/docs/health/CamaraJones.pdf.

[2] Bradley EH, Elkins BR, Herrin J, Elbel B.,Health and social services expenditures: associations with health outcomes, BMJ Qual Saf. 2011 Oct;20(10):826-31. Epub 2011 Mar 29