Does certification of staff nurses improve patient outcomes?
- Correspondence to Mary A Blegen
School of Nursing, University of California San Francisco, 2 Koret, San Francisco, CA 94143, USA;
Implications for practice and research
■ The study adds to the small evidence base that having more nurses with baccalaureate education and certification is associated with improved patient outcomes (30-day mortality and failure to rescue (death after complications)).
■ More studies are needed to determine the best types of certification and actual practice changes associated with certification.
Certification of staff nurses by accrediting bodies and specialty organisations' attempts to recognise those who have developed expertise in particular areas. While the term certification is also used for advanced nurse practitioners and clinical nurse specialists; this study and commentary only address certification of non-advanced practice nurses. In the USA, certification is offered in 11 areas (not advanced practice) by the American Nurses Credentialing Center (http://www.nursecredentialing.org/Certification.aspx). Further, certification in specialty areas is offered by at least 13 specialty groups (http://www.nursezone.com/Edu-ProfDevelopment/certification.aspx).
While there are numerous studies showing that nurses with certification report that they are more empowered, have more intrinsic value, have better collaborative relationships, and in their view, provide better care and have more satisfied patients, there has been little research examining whether they actually provide care that results in better patient outcomes.1 2 One previous study, controlling for registered nurse (RN) staffing and other hospital and nurse characteristics, found that as the proportion of RNs in intensive care units that were certified increased, patient falls decreased; but there was no effect on medication errors, skin breakdown and infection rates.3
Kendall-Gallagher et al published the results of a research project that examined the relationship between nurse education and certification and patient outcomes. The study used data sets compiled for previous research projects and included data from a survey questionnaire sent to a random sample of licensed nurses in four states, characteristics of the hospitals in which they worked from administrative data set, and outcomes of the patients calculated from the patient discharge data for surgical patients treated in those hospitals.
Results showed that hospitals with a larger percentage of nurses who had baccalaureate degrees and certification had lower 30-day mortality and failure to rescue rates for surgical patients. Every 10% increase in the percentage of baccalaureate nurses in hospitals was associated with a 6% decrease in the odds of patients dying (both 30-day mortality and failure to rescue). Every 10% increase in baccalaureate nurses with certification was associated with a 2% decrease in these odds. Years of experience in the hospital did not have a significant effect on patient outcomes.
The study was well conducted and used a large sample from four states in the USA. However, the analyses could not rule out other explanations for the findings. For example, the investigators did not rule out the possibility that hospitals that had more highly educated and certified nurses may have differed in other ways that promote better patient outcomes (such as higher nursing hours per patient day). As the authors mentioned, some factors known to affect patient outcomes were not controlled such as physician board certification and existing quality improvement initiatives. Other unknowns that may have affected the findings include the types of nursing certification and how long the nurses had been certified. For example, while this study assessed the mortality of adult surgical patients; the nurses' certifications could have been in paediatrics, obstetrics, oncology or informatics, which would not likely improve surgical patient care. Further, other nurse sensitive outcomes (such as hospital acquired infections, patient falls and pressure ulcers) were not analysed.
While the results of this study provide evidence that supports the contribution of nurse certification to lower surgical patient mortality, it is premature to conclude that certification will generally improve patient care. There are several issues in the field that need to be dealt with in future studies before the evidence base warrants strong conclusions. Studies including a wider selection of nurse sensitive outcomes are needed. There is no standardisation across the various sources of certification and no study to date has compared the effects of different requirements for certification. Therefore, we do not know whether all the nurses certified by the various bodies will provide better patient care. Another possibility that has not been studied is whether the process of obtaining certification changes the way a nurse practices or whether nurses who decide to pursue certification practiced differently to start with. Further work that differentiates types of certification and the actual changes in nursing practice is needed. While the results of this study are promising, there are many more research questions to answer.