June 13, 2011
06/13/2011—In the contentious Washington debates over healthcare reform, there has been far too little attention paid to what the country desperately needs: evidence-based reforms that can improve healthcare access and outcomes at an affordable cost.
A major new research project proves that such a win-win-win solution is possible. And the study design could have come straight from our own HR field, since it is based on the introduction of a pay-for-performance (P4P) scheme for healthcare professionals that aligns their interests with those of patients.
The study was conducted by an old friend and colleague from my days at the RAND Corporation, Dr. John Peabody. John is now a Professor and Deputy Director of the Institute for Global Health at the University of California, San Francisco. With a grant from the National Institute for Child Health and Development, he and colleagues designed a very impressive piece of research: the Philippine Child Health and Policy Experiment.
Working closely with the Philippine Government and health system, they conducted a huge longitudinal demonstration project. It involved 30 hospitals across 11 provinces, covering about one-third of the country, with an average population size of 250,000. They divided these hospitals into 3 randomized matched groups:
Access (A) Intervention provided all indigent children under the age of 6 years and their families with free access to basic healthcare services;
Bonus (B) Intervention targeted physicians and other health professionals by introducing financial incentives (equivalent to 5% of an individual’s salary) for practitioners providing high-quality care on a range of measures, including both child health outcomes and patient satisfaction;
Control (C) group saw no change in the existing system.
Dr. Peabody and Dr. Orville Solon then gathered a comprehensive amount of data over a three-year period, including: structural measures (such as the availability of electricity and sanitation), overall cleanliness, qualified staff, organization of services, drugs, and equipment, as well as individual health data on children, including blood tests and subjective health assessments, which were done at the same time. They describe their method:
Two tracer conditions, diarrhea and pneumonia, identified a subset of patients to be followed longitudinally. Children screened in with these tracer conditions were followed home, post hospital discharge. At home, we did a detailed socioeconomic survey to investigate the social and economic impact of the illness on the family, including full estimations of direct and indirect costs, employment status and time lost from work. Importantly, we also measured illness rates in the indexed child (i.e., the one that was discharged) and care utilization rates for siblings, including preventive care practices and the schooling of the older children (and parents)… We also conducted a population-based community sample of households with children in Round 1 for which there were a total of 1,535 children randomly selected from the same catchment areas as the discharged patients and surveyed identically (e.g. same questionnaires, same blood tests, cognitive tests).
They followed these children and their families over the next two years, keeping an amazing 99.6% in the study. They analyzed the impact of the two interventions on four different child health outcomes: incidence of acute infection, anemia, presence of age-adjusted wasting, and general self-reported health (GSRH) measure. They found that both the Access and Bonus interventions resulted in significant reductions in wasting (9.2% for Bonus and 8.5% for Access compared to the control), and Bonus also a very meaningful 7.0% in the P4P sites compared to controls. They note that these health improvements, reported uniquely on a pediatric population, are higher than the 2- to 4-percentage point improvement in outcomes from P4P (beyond what was seen in controls) for cardiovascular disease, community-acquired pneumonia, and hip or knee replacement previously reported
Striking for policymakers is that the Bonus intervention was able to achieve similar results at a fraction of the cost: $13,000/province on average vs. about $80,000/province for providing universal access. The other fascinating result was that as a result of enhanced quality of public sector provision through the P4P program, there was “a 23 percent probability that quality will improve among private doctors” competing with the public system in the same community, even though the private doctors were not offered any incentives. The other key finding with direct implications for US healthcare reforms is that “when insurance benefits were increased, households did not enroll or necessarily use their insurance at first—at least not until it was clear that the quality of the services had also improved.” In other words, quality improvement was consistently found to be a driver of better health outcomes and thus, thinks Dr Peabody, “a new and highly leveraged opportunity to successfully extend health benefits to larger populations.”
It will be very interesting to see if these results are replicated when Dr. Peabody and his team conduct a similar controlled experiment in Oregon over the next few years.
For more details on this exciting project go to:
Quality Improvement Demonstration Study [http://www.qids.ph/ ]
Health Affairs [http://content.healthaffairs.org/content/30/4/773.full.html ]
1 Lindenauer et al., 2007, Glickman et al., 2007, and Grossbart, 2006.
2 (Quimbo et al., 2008).