he public is inundated with media cov- erage of changes in health care that could adversely affect their access to and the quality of health care services. Public opinion polls confirm that consumers’ trust in hospitals is eroding and interest in the qual- ity of health care is increasing.1, 2 Consumer concerns are visibly evident in the ongoing congressional debates on the Patient Bill of Rights, federal and state legislation providing protection against prema- ture hospital discharge of new mothers and infants, and legislation stipulating minimum nurse-to-patient ratios in California hospitals. T
More than 80% of the public polled in a recent survey wanted to know how to evaluate the quality of hospital care.2 Various lists of the “best” hospi- tals, such as that in U.S. News and World Report (“America’s Best Hospitals”) have been published, generating consumer interest and creating market- ing opportunities for hospitals.3 Likewise, the desig- nation “accreditation with commendation” from the Joint Commission on Accreditation of Healthcare Organizations (JCAHO) has been touted by hospitals as an indicator of high-quality hospital care. There is little hard evidence that “best” hospitals achieve better outcomes than other hospitals4, 5 or that JCAHO commendation trans- lates into better patient care, but consumers appear to take note of these appraisals, perhaps because of a lack of alternatives.
For close to two decades, nursing has had a potential vehicle for informing consumers about the quality of hospital nursing care. In the early 1980s, 41 hospitals were awarded “magnet hospi- tal” designation as a result of a national study con- ducted by the American Academy of Nursing. Two aims of the AAN initiative were to identify hospi- tals that were successful in attracting and retaining nurses and to determine the organizational fea- tures those hospitals had in common that might account for their success.6 The magnet hospitals had in common organizational features that pro- moted and sustained professional nursing practice, including flat organizational structures, unit-based decision-making processes, influential nurse exec- utives, and investments in the education and expertise of nurses. Unlike other “best” hospitals, the AAN magnet hospitals (hereafter referred to as “original magnet hospitals”) have been the subject of considerable research; subsets of the magnet hospitals were reexamined in 1986,7 1989,8 and 1991,9 and the results documented that the distin- guishing organizational features have endured.
The organizational attributes that attract nurses to magnet hos- pitals have also been found to be consistently and significantly associated with better patient out- comes than those of matched nonmagnet hospitals. Several recently published research synthesis papers reinforce the empirical evidence suggesting that magnet hospitals achieve better outcomes than comparable hospitals.10, 11 We conducted two earlier studies that reaffirmed the magnitude of the supe- rior outcomes in the magnet hospitals.
In the first study, we examined Medicare mortal- ity rates using 1988 data for 39 of the 41 original magnet hospitals (one hospital had closed and one was a Veterans Administration hospital not included in the Medicare data) by using a multi- variate matched sampling procedure that controlled for hospital characteristics that previous research had shown to be associated with mortality (such as ownership, teaching status, size, location, financial status, physician qualifications, technology index, and emergency admissions). The 39 magnet hospi- tals were matched with 195 comparison hospitals (five per magnet hospital) selected from all non- magnet U.S. hospitals with more than 100 Medicare discharges. Medicare mortality rates in magnet and comparison hospitals were compared using vari- ance components models, which pool information from each group of five matched hospitals and adjust for differences in patient composition, as measured by predicted mortality. After adjustment for differences in predicted mortality for Medicare patients, the magnet hospitals had a 4.6% lower mortality rate (p = 0.026), which translates to between 0.9 to 9.4 fewer deaths per 1,000 dis- charges (with 95% confidence).12
The second study of the magnet hospitals involved data from 1,205 consecutively admitted patients with AIDS and from 820 nurses on 40 units in a subset of 20 magnet hospitals.13 Patient outcomes were com- pared for patients with AIDS in magnet hospitals without dedicated AIDS units and in comparison hospitals with and without dedicated AIDS units. Patients with AIDS in scattered-bed units in magnet hospitals had lower odds of dying than did AIDS patients in any other setting—lower by 60%, for example, than patients in scattered-bed units in non- magnet hospitals.9 Other analyses associated with this study showed that magnet hospitals had sig- nificantly higher levels of patient satisfaction,14 signif- icantly lower rates of nurse burnout,15 and lower rates of needlestick injuries in nurses14 than did comparison
JONA • Vol. 39, No. 7/8 • July/August Supplement 2009
Copyright @ 2009 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.