In response to: Powell J.


I was excited to read Powell's response to our published research report Impact of hospital nursing care on 30-day mortality for acute medical patients. Here, I expand on and respond to some of Powell's comments.

How nurses are educated is evolving in Ontario, Canada as it is in many other places across the world. In Ontario, Registered Practical Nurse preparation is being modified and curricula for these programs are being enhanced and updated. Further, diploma preparation for Registered Nurses previously offered exclusively in community colleges for Registered Nurse entry to practice is no longer an option in the Province of Ontario. Baccalaureate university preparation is now the minimum entry to practice criteria for the Registered Nurse. Baccalaureate university preparation for Registered Nurses has been in existence for many decades in the Province of Ontario. This educational route is not new. However, as with all educational programmes, baccalaureate university curricula across all universities undergo regular review and revitalization to ensure their appropriateness and relevance. As well, improvements in the preparation for unlicensed assistive personnel (e.g. personal support workers) are also being advocated and undertaken in Ontario. Because this category of personnel is unlicensed, this preparation is optional. However, in Ontario's very large healthcare system, the impact of changes to nursing and unlicensed assistive personnel preparation across the various healthcare sectors (e.g. acute care hospitals, community healthcare and long-term care) may not be felt or known for a decade or more.

Though not stated by Powell but implied by some is that the revised educational preparation for Registered Practical Nurses may be considered equivalent to the previous educational preparation of diploma or college prepared Registered Nurses. I strongly recommend that caution be used in making such an assumption or conclusion.

Firstly, I am compelled to review and reaffirm the study findings relevant to Powell's comments. The findings are not only that a higher proportion of Registered Nurses caring for acute medical patients is associated with lower 30-day medical mortality but that a higher proportion of baccalaureate university-educated nurses is associated with lower 30-day mortality rates for acute medical patients. In fact, the impact of care provided to acute medical patients by baccalaureate university-educated Registered Nurses on lower mortality rates is even larger than the impact of the overall proportion of Registered Nurses caring for medical patients. A 10% increase in the overall proportion of Registered Nurses (currently a mix of both diploma/college prepared AND university-educated nurses) is associated with six fewer patient deaths for every 1000 acute medical patients discharged. Keeping all other variables constant, including the proportion of Registered Nurses providing care to medical patients, a 10% increase in the proportion of baccalaureate university-prepared Registered Nurses is associated with nine fewer deaths for every 1000 patients discharged. Regardless of one's political or ideological position, the evidence strongly suggests that delivery models of nursing care for acute medical patients should use the highest possible proportion of baccalaureate university-educated Registered Nurses.

Powell is concerned that readers may make sweeping generalizations about study findings. He asserts that extreme caution must be taken when applying findings related to nursing staff mix and educational preparation. He makes this conclusion based on above described changes planned and being implemented to educational preparation of Registered Practical Nurses in Ontario and the changes to entry to practice requirements for Ontario Registered Nurses. Powell claims that ‘this study represents conditions at a point in time that have become significantly altered, rendering research finding inappropriate for comparison to the newly emerging healthcare system’. He suggests that the nursing workforce of 2006–2007 is dramatically different than that of 2003. I suggest that this is not so. It may be a decade or more before we see any significant change in the nature and makeup of the Ontario nursing workforce, particularly within acute care hospitals, which was the focus of this study. I do concur with Powell's recommendation to use caution before applying findings from any one study. Research consumers must always consider the weight of evidence, the context and time in which the evidence was generated, as well as the rigour of methods used to generate the evidence before application should be considered.

Powell stated that the results ‘do not address the human resources challenges for nursing in today's healthcare systems and practice environments’. I agree. It was never our intent to limit our research and target our findings to only the realm of what is possible given shortages of nursing personnel, limited educational seats for nurse education, costs of educational preparation, and so on. The purpose of this research was to understand the nursing-related determinants of 30-day mortality for acute medical patients. Good research is about trying to come to know the truth, regardless of how palatable that truth may be to consumers.

One might ask, what is the best model of hospital nursing care delivery for acute medical patients? Based on these study findings and those of others cited in the study report, I recommend consideration of a new model of care delivery to be considered for nursing care of hospitalized acute medical patients. I refer to a care delivery model consisting primarily of baccalaureate university-educated nurses. Though this model may not be seen as favourable by all, because of cost challenges, human resource supply challenges and so on, evidence is mounting that such a model may be best for specific groups of acutely and complexly ill patients such as those in this study. This indeed would be a new model of care delivery that Powell and others should consider to minimize the lower mortality for acute medical patients.

Research is undertaken in a period of time. Our world is not static even though we try to come to know it through activities such as research that are almost always time bound. Decisions about whether or not potential research consumers choose to apply the findings of a particular study undertaken at one time to different settings at different times should not be taken lightly. This research was undertaken less than 4 years ago in year 2003. The degree of change in the Ontario healthcare system landscape or any other jurisdiction over these few years is still to be determined. Opinions about the nature and impacts of these changes remain simply opinions and impressions until formally investigated. Further research will indeed be required to identify the real nature of the changes in educational preparation of nurses and the impact on patient and healthcare system outcomes such as mortality. I agree that future research is required to understand the impact of changes made to the educational preparation for nurses. These impacts will not likely be known for years to come. Not mentioned by Powell are other on-going changes to our healthcare system such as changes in primary care, changes to funding formulae, changes in physician educational preparation, and so on.

Powell also suggested that these findings may be used to support a political or ideological position and detract from the potential to realize a new and improved healthcare delivery model. I can think of no better source of support for a political or ideological position than that of theory and related evidence developed through rigorous research.