Small Primary Care Physician Practices Have Low Rates Of Preventable Hospital Admissions
By Lawrence P. Casalino, Michael F. Pesko, Andrew M. Ryan, Jayme L. Mendelsohn, Kennon R. Copeland, Patricia Pamela Ramsay, Xuming Sun, Diane R. Rittenhouse and Stephen M. Shortell
Health Affairs, August 13, 2014 (online)
The Affordable Care Act and initiatives by private health insurance companies are driving major changes in the ownership of physician practices, the incentives practices face to improve the care they provide, and the processes practices use to improve care. Many practices are consolidating into larger medical groups. Many others are shifting from physician ownership to hospital ownership. Practices are increasingly subjected to pay-for-performance and public reporting programs and are being encouraged to implement processes used in patient-centered medical homes.
Ambulatory care–sensitive admissions are defined by the Agency for Healthcare Research and Quality (AHRQ) as admissions for conditions such as congestive heart failure for which good primary care may prevent admission.
In our large national study of small and medium-size primary care–based practices, practices with 1–2 physicians had ambulatory care–sensitive admission rates that were 33 percent lower than those of the largest small practices (having 10–19 physicians). Practices with 3–9 physicians also had rates that were lower than the rates for the largest small practices, although slightly higher than the rates for practices with 1–2 physicians. These findings were unexpected, since small practices presumably have fewer resources to hire staff to help them implement systematic processes to improve the care they provide. Larger practices did have higher patient-centered medical home scores than the practices with 1–2 physicians (though not higher than those with 3–9 physicians) and so appear to use more such processes, but these higher scores were not associated with lower ambulatory care–sensitive admission rates in multivariate analyses.
It is possible that small practices have characteristics that are not easily measured but result in important outcomes, such as fewer ambulatory care–sensitive admissions. For example, there is evidence that patients in smaller practices are better able to get appointments when they want them and better able to reach their physician via telephone, compared to larger practices. It is also possible that physicians, patients, and staff know each other better in small practices, and that these closer connections result in fewer avoidable admissions.
We cannot fully exclude the possibility that the largest practices, which had a somewhat higher percentage of specialists, had patients who were sicker and, therefore, more likely to have an ambulatory care–sensitive admission. However, we controlled for the percentage of specialists in practices and for patients’ demographic characteristics and comorbidities, and we found that the smallest practices cared for a significantly higher percentage of dual-eligible patients and for patients with more comorbidities.
Physician-owned practices had lower ambulatory care–sensitive admission rates than hospital-owned practices in both bivariate and multivariate analyses—approximately 13 percent lower in multivariate analysis.
Hospital ownership would be expected to result in a lower ambulatory care–sensitive admission rate if hospitals provided additional resources to practices to hire staff and implement systematic processes to improve care. In fact, consistent with prior studies, we found that hospital-owned practices used more patient-centered medical home processes than physician-owned practices. But these practices nevertheless had higher ambulatory care–sensitive admission rates. Hospital acquisition of a practice might disrupt longstanding referral relationships between the practice’s physicians and specialists outside the practice and might lead to other changes that result in worse performance by the practice and higher ambulatory care–sensitive admission rates.
We did not find an association between the ambulatory care–sensitive admission rate and the use of patient-centered medical home processes or between that rate and pay-for-performance or public reporting incentives. Prior research has resulted in inconsistent findings regarding the relationship between patient-centered medical homes and physician practice performance and between incentives and physician practice performance.
Physicians in small practices have no negotiating leverage with health insurers, so insurers typically pay them much lower rates for their services than they pay to physicians who practice in larger groups or are employed by hospitals. This policy might be penny wise and pound foolish if it drives small practices out of existence and if further research confirms that small practices have lower ambulatory care–sensitive admission rates, and possibly lower overall costs for patients’ care, than larger groups.
Small practices have many obvious disadvantages. It would be a mistake to romanticize them. But it might be an even greater mistake to ignore them, and the lessons that might be learned from them, as larger and larger provider organizations clash to gain advantageous positions in the new world of payment and delivery system changes catalyzed by health care reform.
By Don McCanne, MD
It is believed that consolidation of the health care delivery system through the formation of larger groups of physicians and through hospital ownership of physician practices is anti-competitive and drives up health care spending, especially through non-competitive pricing. Nevertheless this consolidation is being encouraged under the assumption that closer integration of the health care delivery system will improve processes and outcomes, one rapidly expanding model being accountable care organizations. This important study casts doubt on this concept.
One important measure of the quality of care being provided is ambulatory care-sensitive admissions – admissions that can be prevented through good primary care. This study shows that small primary care practices had lower preventable admission rates than did larger practices. Further, although larger practices did have higher patient-centered medical home scores, the scores were not associated with lower ambulatory care–sensitive admission rates. Also, hospital-owned practices used more patient-centered medical home processes than physician-owned practices, yet these hospital-owned practices had higher ambulatory care–sensitive admission rates. Neither pay-for-performance nor public reporting incentives improved the rate of ambulatory care-sensitive admissions.
The policy and political communities are pushing innovations such as more closely integrated groups through consolidation and accountable care organizations, pay-for-performance, and patient-centered medical homes, when there is sparse evidence that these measures will improve quality or reduce costs. On the other hand, studies such as this demonstrate that traditional Marcus Welby, MD-type primary care practices serve us very well (as long as they do see more than one patient a week).
Patients have better access through a long standing relationship with a health care professional they know and trust and who knows and respects them, while receiving their care at a lower cost. Although this traditional model is now being threatened, a single payer system would revitalize it as long as it serves patients well.