Dr. Linda Aiken, Director of the Center for Health Outcomes and Policy Research (CHOPR) at the University of Pennsylvania School of Nursing, is the Principal Investigator for a project funded by the National Institute of Nursing Research entitled, “Panel Study of Effects of Changes in Nursing on Patient Outcomes.”
In describing the ongoing project, Dr. Aiken explained that more than 550 hospitals and hundreds of home care agencies and nursing homes in four states (PA, NJ, CA, and FL) were studied over time. In Pennsylvania, institutions were studied at three time points (1999, 2005-2006, 2015-2016). Detailed information about patient-to-nurse ratios, nursing skill mix, nurses’ qualifications, and the quality of the clinical work environment was collected through surveys from tens of thousands of nurses practicing in these organizations. These nurse resource data were linked with patient outcomes from multiple sources, including hospital discharge data from the Pennsylvania Health Care Cost Containment Council (PHC4).
The study has two more years to completion, but several articles have been written describing Dr. Aiken’s present findings. Dr. Aiken shared some study results that are pending scientific journal publication, “Patient outcomes are better in hospitals in which nurses care for fewer patients each, where a higher proportion of registered nurses (RNs) have bachelor’s educational qualifications, and where the clinical environment includes nurse engagement in decision-making and good relations between doctors and nurses and nurses and institutional management. Better nursing resources are associated with lower risk-adjusted mortality, fewer infections, fewer readmissions, and lower per patient expenditures and thus providing cost-savings to institutions.”
Dr. Aiken added that hospitals where nurse staffing, education, and work environments have improved have seen greater improvements in patient safety and quality of care when compared to hospitals where nursing resources have remained the same or declined. She highlighted a recent study that showed patients at highest risk of poor outcomes, such as patients with Alzheimer’s and related dementias, who undergo general, orthopedic, and vascular surgery, experience significantly lower mortality in hospitals with a higher proportion of nurses with bachelor’s educational qualifications.
Dr. Aiken explained how she used the PHC4 data in her analysis, “The goal is to reduce variation in patient outcomes across hospitals by providing hospital leaders and clinicians rigorous scientific information to inform resource allocation and clinical interventions to provide the best care to patients and good value.” Her study team used PHC4 de-identified individual level patient outcomes data from general acute hospitals in Pennsylvania. With the data, the team determined whether patient outcomes were significantly better at some hospitals than others and identified major features of the hospitals that contributed to better outcomes. Dr. Aiken added, “PHC4 is the only source of data to study the outcomes of hospitalized patients across all age groups and non-federal hospitals. PHC4 information is very valuable in the study of hospital performance because the detailed information on patients’ conditions at the time of admission enables researchers to take into account that some hospitals, such as teaching hospitals, are caring for sicker patients.”
In discussing the benefits of lower nurse to patient ratios, Dr. Aiken described a paper published in Health Services Research using PHC4 data. Results showed that in 2005-2006, after implementation of mandated ratios in California, nurses in Pennsylvania cared for, on average, more than one additional patient each than nurses in California hospitals. Estimating if Pennsylvania hospitals staffed at the levels mandated in California, mortality following general surgery in Pennsylvania hospitals might be reduced by 13%. Analysis of recent data from 2015-2016, shows that 12 years after the implementation of nurse ratios, California still has substantially better nurse staffing ratios than other states, and mortality rates are still lower than those in Pennsylvania and other states.
In addition to nurse-to-patient ratios, Dr. Aiken stated that level of education seems to play a part in lower mortality rates. She explained, “Each 10% increase in the proportion of hospital direct care RNs with a bachelor of science in nursing degree (BSN) education is associated with a 5-7% decline in risk-adjusted mortality following common surgeries.”
Dr. Aiken noted that most states focus on voluntary efforts to benchmark and improve performance and patient outcomes rather than mandating specific quality initiatives—often using data from resources such as PHC4. According to Aiken, about half of the states have some form of RN staffing-related initiatives, most commonly requiring a process to determine safe hospital staffing that involves nurse input. A few states have mandated public reporting of hospital patient-to-nurse staffing, and many states including Pennsylvania stipulate minimum nurse staffing in nursing homes. Considerable attention is being given to modernizing state practice, including in Pennsylvania, that would enable advanced practice nurses, such as nurse practitioners, to have full scope of practice commensurate with their advanced training and expertise.
Dr. Aiken added that PHC4 data enabled their research group to determine that hospitals employing a higher number of nurse practitioners per 100 beds have significantly better patient outcomes including lower risk-adjusted mortality and fewer readmissions. Aiken noted that many hospitals in Pennsylvania and nationally preferentially hire BSNs—a practice that stems from the pioneering research at Penn Nursing’s Center for Health Outcomes and Policy Research showing that patient outcomes are better in hospitals with RNs with BSN. This work, using PHC4 data, was published in the Journal of the American Medical Association (JAMA) and The Lancet. According to Aiken, New York State recently passed legislation mandating that nurses obtain a BSN within 10 years of entering practice in order to retain their RN licenses.
To understand whether the relationship between nurse resources and patient outcomes is uniquely American or a generic relationship present in differently organized, financed, and resourced health systems, Dr. Aiken extended her research to other countries. Results showed that the relationship held in all the countries studied, suggesting that using nursing as a soft target for budget reductions adversely affects patients’ wellbeing and ultimately does not save money as patients with complications are more expensive to treat. By studying countries that mandate bachelor’s education, like Norway and Chile, Dr. Aiken’s research team can estimate the likely impact a better educated nurse workforce would have on quality improvement and hospital care in the U.S., which has lagged behind many other countries in moving to bachelor’s education for nurses. Dr. Aiken added that other countries, such as England, have more fully developed community nurse services, which offer useful models to bring to the U.S. in efforts to reduce hospital readmission rates. Many other countries have more hospital-based physicians than the U.S. This international example has led to more hospitalists and intensivists in U.S. hospitals that some research suggests has improved nurse-physician collaboration with good results for improved patient outcomes.
When asked about the benefits of using PHC4 data in her studies, Dr. Aiken stated, “PHC4 has been essential to our program of research. PHC4 is an important resource for monitoring quality and safety of hospital care, and providing information that guides clinicians and hospital leaders in improving hospital performance. PHC4 is in the public’s interest.”
Return to the Case Studies page: VIEW ALL CASE STUDIES
