Hospital Organization and Productivity
October 4-5, 2013
Julia Adler-Milstein, University of Michigan; Kirstin Scott, Harvard University; and Ashish Jha, Harvard School of Public Health
The U.S. is in the midst of an ambitious effort to achieve nationwide adoption of electronic health records (EHRs) to drive improvements in the quality and efficiency of care. However, recent studies fail to find a consistent relationship between EHR adoption and improved hospital performance, suggesting that EHRs alone are not sufficient to drive improvements in care delivery. In this study, Adler-Milstein, Scott, and Jha examine whether the quality of hospital management modifies the relationship between EHR adoption and cost and quality outcomes for acute myocardial infarction (AMI). In their random sample of U.S. hospitals, the authors find that when hospitals are poorly managed, EHRs are associated with less efficiency. Further, when hospitals are well managed, having an EHR is associated with greater efficiency. The authors' findings suggest that EHR implementation may need to be coupled with effective management in order to drive value from these systems.
Elizabeth Munnich, University of Louisville, and Stephen Parente, University of Minnesota
Over the past 30 years, outpatient surgery has become an increasingly important part of medical care in the United States. Munnich and Parente examine ambulatory surgery centers (ASCs), which have been praised as a low-cost, convenient alternative to outpatient surgery in hospitals, and criticized for "cream skimming" profitable patients and procedures, which are important sources of revenue for general hospitals. For a national sample of Medicare patients that varies over time and controlling for physician fixed effects, the authors show that ASCs treat healthier patients than hospital outpatient departments dothe highest risk quartile of patients was half as likely to be treated in an ASC as those in the healthiest quartile. Controlling for patient characteristics, the authors find that ASCs perform procedures faster than hospital outpatient departments. Combined with the fact that ASCs receive lower reimbursements than hospitals, outpatient surgery is less costly in ASCs. To the extent that ASCs provide cheaper and faster care than hospitals, the authors then consider whether they do so at the expense of quality of care. Using the variation in ASC use generated by exogenous changes in Medicare payments, they find that treatment in an ASC is associated with better health outcomes, holding patient risk constant; in fact, high-risk patients treated in an ASC are less likely to be admitted to a hospital within seven days of an outpatient surgery, and less likely to visit an emergency room on the same day as an outpatient surgery. These results suggest that health policy planners may need to trade off the superior and cheaper treatment in ASCs against the subsidy outpatient surgeries provide to hospitals.
Caroline Carlin, Medical Research Institute, and Roger Feldman and Bryan Dowd, University of Minnesota
In recent years, U.S. hospitals have accelerated the trend toward acquisition of physician practices. If the hospital and physician practice had different levels of market power prior to the acquisition, the tied contracting that results when health plans must negotiate with the combined entity may have an impact on prices. In addition, when the hospital already owns a number of clinics, this new acquisition results in horizontal integration of clinic systems, possibly increasing its market power for physician services. However, there is a paucity of empirical literature documenting the impact of vertical integration of hospitals and physician practices on prices. Carlin, Feldman, and Dowd leverage changes in a large metropolitan area in which three multi-specialty clinic systems were acquired by hospital-owned integrated delivery systems (IDSs) at the end of 2007. The authors examine the impact of these hospital acquisitions on global measures of hospital and physician prices, and on unit prices for several high-frequency physician procedures. They find evidence of an increase in physician prices for health plan enrollees attributed to both the acquired clinic system and the acquiring IDS's legacy clinic system, supporting the hypothesis that horizontal integration between the newly acquired clinics and the previously owned legacy clinics increases market power. In addition, they find changes in hospital prices consistent with the impact of tied physician and hospital contracting in a differentiated market.
William Bertschinger, Bijan Borah, David Cook, Joseph Dearani, Elizabeth Habermann, Veronique Roger, Jeffrey Thompson, and Sue Visscher, Mayo Clinic, College of Medicine
The business model of the full-service U.S. hospital has been described as a "solution shop." In this model medical problems are assumed to be complex and unstructured thereby requiring "expert" physicians to make decisions on course of care and resource utilization. This model can contribute to unwarranted variation in care delivery, poorer outcomes and increased cost. Cook, Thompson, Habermann, Bertschinger, Roger, Visscher, Dearani, and Borah undertook a comprehensive redesign of the adult cardiac surgical care at Mayo Clinic, Rochester, Minnesota. They created a standardized practice, "focused factory" surgical model for the process of care in qualified patients. Key elements were patient stratification to the model, electronic protocols, and empowerment of bedside providers. Results before 2008 and after 2012 were compared using propensity score matching. The analysis demonstrated that a protocol-driven, focused-factory model lowered resource utilization and reduced cost in the operating rooms, intensive care units (ICUs), and on the "floor". ICU and hospital length of stay were reduced; the mean cost of hospital care was reduced by 15 percent. Variation in care process, resource utilization, and cost were all reduced. The model did not increase mortality or in-hospital complications, and 30-day clinical outcomes (pneumonia, sepsis and renal failure) were significantly improved. While the "focused factory" has been criticized, its implementation within a full-service hospital, in parallel with the traditional solution shop care model, may increase care value while preserving, (in a lesser clinical footprint), the strengths of the solution shop for complex care.
Laurence Baker, Kate Bundorf, and Daniel Kessler, Stanford University and NBER
In markets for medical care, vertical integrationcontractual or ownership relationships between hospitals and physicianscan have opposing effects. Integration can reduce health spending and increase quality by improving communication across care settings, but it can also enhance providers' market power and facilitate the payment of kickbacks for inefficient referrals. Baker, Bundorf, and Kessler investigate the impact of integration with hospital claims from the Truven MarketScan Commercial Claims and Encounters database on the non-elderly privately insured from 20012007. The authors construct county-level indices of hospital prices, volumes, and spending, adjusted for differences in enrollees' age and gender. They measure hospital-physician integration by combining information on hospitals' relationships with physicians from the American Hospital Association with information from Medicare. They find that hospital ownership of physicians leads to higher hospital prices and spending. Although they find that contractual integration reduces the frequency of hospital admissions, these effects are relatively small. Taken together, their results provide a mixed, although somewhat negative, picture of vertical integration from the perspective of the privately insured.
Kate Ho, Columbia University and NBER, and Ariel Pakes, Harvard University and NBER
Ho and Pakes estimate a preference function which rationalizes hospital referrals for privately-insured birth episodes in California. The function varies across insurers and is additively separable in a hospital price paid by the insurer, the distance traveled, and plan and severity-specific hospital fixed effects (capturing various dimensions of hospital quality). The authors use an inequality estimator that allows for errors in price and detailed hospital-severity interactions, and obtain markedly different results from those from a logit. The inequality estimator indicates that insurers with more capitated physicians are more responsive to hospital prices. Capitated plans allow patients to be sent farther to utilize similar-quality lower-priced hospitals, but the trade-off between quality and costs does not vary with capitation rates.
David Meltzer, University of Chicago and NBER, and Greg Ruhnke, University of Chicago
Hospital costs are heavily influenced by the physicians who direct care within hospitals. Over the past 15 years, there has been a major shift in the United States from having a patient's hospital care directed by his primary care physician (PCP) to having it directed by a hospitalist physician who specializes in hospital care but generally does not know the patient. Meltzer and Ruhnke review economic evidence that the use of hospitalists increased not because they reduce cost or improve outcomes, but because PCPs no longer had enough patients in the hospital to care for them efficiently. The authors also review evidence suggesting that the loss of continuity in the doctor-patient relationship for these patients dramatically increases costs. Based on this data, the authors have designed and are now executing a 2,000-patient randomized trial with support from the Centers for Medicare and Medicaid Innovation to implement a new Comprehensive Care Physician Model in which physicians focus their practices on a small number of frequently hospitalized patients so that they can provide care for them both in clinic and in the hospital. The authors describe: the key design elements of this intervention and their connections to the underlying economic rationale of the model; the major challenges that have arisen in the early implementation of the model; and the measures that will be used to evaluate the model as it is implemented.
Nicholas Bloom, Stanford University and NBER; Raffaella Sadun, Harvard University and NBER; and John Van Reenen, London School of Economics and NBER
Bloom, Sadun, and Van Reenen collect comparable data on core management practices in 1,700 hospitals across eight countries (Canada, France, Germany, India, Italy, Sweden, the United Kingdom, and the United States). In all countries where they have data, the authors find management quality is strongly correlated with better financial and clinical outcomes, such as heart attack survival rates. Exploiting within-country variation, the authors show that hospitals with more clinically trained managers, that are larger, and that are non-government owned appear to have significantly higher management scores. Unlike other sectors such as manufacturing, almost half of the variation in management scores is between country rather than within country. The authors speculate that the higher management scores in leading countries (like the United States and the United Kingdom) are the result of relatively politically independent appointment of hospital leaders and stronger accountability mechanisms.