Testimony Before the Institute of Medicine - November 20, 2002
Testimony of Dennis S. O'Leary, M.D., President, The Joint Commission on Accreditation of Healthcare Organizations
Before the "Work Environment for Nurses and Patient Safety Committee," Institute of Medicine
I am Dr. Dennis O'Leary, President of the Joint Commission on Accreditation of Healthcare Organizations. I want to thank the Committee for the opportunity to provide you with information on the Joint Commission's approach to addressing problems related to sufficient numbers of nursing and other staff in health care organizations. While some have portrayed this approach as an alternative to mandated staff-to-patient ratios, it could as easily be viewed as complementary to such ratios or even as a mechanism for evaluating, and perhaps validating, the use of uniform ratios.
The Joint Commission has indeed long had standards that require hospitals to establish organization-specific staff-to-patient ratios based upon the organization's assessment of patient care needs. The assessment usually involves consideration of numbers, types, and acuity of various patient subgroups. Hospital compliance with these standards has generally been determined by the consistency of its adherence to its staffing plan (i.e., its staff-to-patients ratios.)
This approach might be defensible in a pure, ideal, resource-flush world, but hospital care in this country has not been provided in such a world for a long time, if ever. In fact, hospitals have for the past two decades provided care in a progressively resource-constrained environment, and they have been paid exactly the same dollar, irrespective of the outcomes of that care. It is small wonder that some, perhaps many, hospitals found it tempting, or simply a practical necessity, to test and even push themselves beyond the limits of conventionally accepted staff-to-patient ratios. This trend did not pass unnoticed by nursing leaders and unions who were complaining bitterly; the IOM which issued a significant nursing report early in the decade; and the Joint Commission which was increasing by sensing the inadequacy of its own requirements. But there were no data upon which to hang these concerns, no smoking gun, no way to tease out nursing impacts on patient outcomes.
However, we have come to a new time, a time unhappily of severe staffing shortages, particularly of nurses. But we also have our smoking guns, our data from a growing number of excellent studies which link nurse staffing levels to patient outcomes. This panel is already aware of these linkages, so I need not elaborate, but the strength and depth of the reactions has been gratifying. For the Joint Commission's part, the breadth of the evidence in our Sentinel Event Database linking serious adverse events with insufficient staffing drove the nurse staffing issue to the top of our public policy agenda in 2001. As you heard yesterday, the Joint Commission convened a roundtable of experts, including such luminaries as Marilyn Chow who is on your Committee, to discuss and make recommendations on how to solve the crisis in the nurse workforce. Their recommendations are contained in our report, Health Care at the Crossroad: Strategies for Addressing the Evolving Nursing Crisis.
For the several years prior to convening the roundtable, the Joint Commission grappled unsuccessfully to reach agreement on a different, more effective approach to defining nursing requirements. Various screening and drill-down evaluation models, as well as staffing ratio proposals, were elaborated and rejected through this time. But perhaps by process of elimination, the Joint Commission eventually settled upon a simple concept -- Staffing Effectiveness -- and a time tested methodology - Continuous Quality Improvement (CQI) - for addressing this concept. Staffing Effectiveness is defined as the number, competency, and skill mix of staff in relation to the provision of needed services. CQI simply involves the selection and application of sensitive measures to identify potential patient care problems, the analysis of underlying factors that are contributing to the problem, and system re-design (or resource allocation) to resolve the problem. It, of course, is not all that simple, but by the year 2000, we knew that nursing-sensitive measures existed, and we certainly knew that we had a problem.
To help us settle upon the specific measures, we called on more than 100 experts who had experience in direct patient care, performance measurement, and health care management. They also represented various agencies and organization and professional disciplines. In the end, we selected 20 measures - nine related to clinical care, while eleven covered various dimensions of human resources (e.g., staff injuries, overtime wages.) With these indicators in hand, the Joint Commission created a simple CQI-based standards framework:
HR. 2.1 The organization uses data on clinical indicators in combination with human resource indicators to assess staffing effectiveness.
This standard and related standards which articulate expectations around data analysis, active intervention and demonstrated improvements, became effective for all accredited hospitals beginning July 2002. Specifically, hospitals are expected to select two clinical and two human resource measures and establish expected performance targets for each. Of the four measures, a hospital may select two defensible measures that relate to the unique care characteristics of that hospital; however, at least one clinical measure and one human resource measure must come from the Joint Commission's list. The measures are expected to encompass both direct caregivers (e.g., nurses, respiratory therapists) and indirect caregivers (e.g., pharmacists, dieticians.)
During onsite evaluations, Joint Commission surveyors specifically examine:
- The rationales for the indicators selected, and the setting of performance targets,
- The data actually gathered.
- The organization's analyses of the data (i.e., what do they think the data mean.)
- The actions taken on the bases of the analyses.
- Evaluation of the effectiveness of the actions taken and periodic reports on this activity to the organization leaders
The organizations are specifically challenged to analyze the measure results in relation to each other. This approach is based on experiences in American industry more than two decades ago which found that groupings of related measures produced much richer and understandable portrayals of performance than single isolated measures. Today, there are techniques such as scatter diagrams, star plots, and line graphs to support such analyses and portrayals which will lead organizations to valid conclusions about their performance.
Not all that is found through this CQI approach will necessarily relate to nurse staffing or other staffing insufficiencies. In some instances, other staff-related issues will be identified (e.g., staff competency), and some issues will be unrelated to staffing in any respect. This is the nature of using measurement as a screening tool. And, in fact, identification and resolution of other issues not related to staffing may also prove to be important to improving the safety and quality of care in a given organization. What the Joint Commission is doing, however, is tipping the scales toward identification of the staffing-related issues by selecting measures that are sensitive to such issues.
Similar standards are currently under development for long-term care, home care, ambulatory care, and behavioral health care settings. At the suggestion of our experts, the Joint Commission is also exploring the development of specialty-related screening indicators that might, for example, be applicable to pediatric or rehabilitation care. At the same time, it bears emphasis that the Joint Commission surveyors will continue to look at actual versus planned staffing in the organization as well as issues such as competency evaluation and staff training.
It is of interest that the new hospital Staffing Effectiveness standards were piloted in 43 hospitals during 2001. We found that most hospitals were already collecting significant amounts of data and in fact, some were gathering data on all of the Joint Commission priority measures. But only a few had undertaken any analyses of the data, and essentially none had examined the data in relation to each other. This may in part relate to the siloization of data collection, analyses and communications within health care organizations. But it is also a sobering commentary on the lack of sophistication about CQI tools and methods inside these organizations. We have a lot of work to do.
We are currently monitoring hospital experience with the new requirements. While the data are preliminary, 54% of hospitals are using hospital-wide indicators, 29% are collecting data on unit-specific indicators, and 17% of the collected indicators relate to special patient populations. Hospitals in this latter group tend to be specialty hospitals, such as rehabilitation or psychiatric hospitals. We have also developed questionnaires that we are asking our hospitals to complete in order to provide us experiential information about the indicator approach as well as specific data such as the type of staffing issues being identified. We plan to share this information as it becomes available.
It is finally fair to say that the Joint Commission's Staffing Effectiveness Standards are a work in progress. We do believe that the standards framework is on target, but there are real questions about some of the specifics:
- Is four the right number of screening indicators?
- Should organizations be given more or less choice in selecting indicators?
- Should the indicators to be used be standardized?
- Should data on standardized indicators be gathered? Be publicly reported? What impacts might this have on the measurement and improvement of Staffing Effectiveness?
And there are clearly more questions but the answers to today's staffing problems in hospitals and other types of health care organizations likely lie here in this realm.
We may now reasonably ask how the Joint Commission's nursing requirements - both new and old - relate to mandated staff-to-patient ratios. First, mandated ratios do constitute an initial finger-in-the-dike approach that at least establishes a patient safety floor. But this one-size-fits-all approach also has its limitations.
- It does not fully account for differences in experience levels and skill mixes of nurses.
- It does not adequately factor in differences in patient acuity.
- Ratios which are legislated do not address ancillary staffing levels which hospitals would in some cases actually be incented to adjust downward in order to afford the new nurse requirements.
- Mandated ratios do not make nurses materialize in an environment where there is, overall, a nursing shortage.
It is also theoretically possible that the imposition of mandatory ratios could lead to expenditures for which benefits are not actually realized (e.g., settings in which outcomes were already good or excellence using innovative staffing models.) Finally, mandated ratios do not guarantee excellence in care. Measurement, monitoring and improvement of care will be necessary no matter what the staffing ratios are.
All of this is to acknowledge that we have a daunting problem whose solutions must necessarily be multi-faceted. There is also an urgency to move forward. The Joint Commission interposes no objections to mandated staff-to-patient ratios, but we believe that final answers lie in requiring health care organizations to systematically pay attention to and improve the results of care day after day after day and to do what is necessary to achieve that goal. That's what the Joint Commission's Staffing Effectiveness Standards are all about.
Thank you for the opportunity to discuss this important issue with you.