Quality Forum Update

 National Quality Forum “Nursing Home” Project Developments:  Steering Committee Conference Call December 10, 2003

At its meeting on September 30, 2003, the National Quality Forum Board of Directors endorsed 14 voluntary consensus standards for assessing nursing-home performance – standards that CMS had agreed to use in its Nursing Home Quality Initiative.  At that time, however, the NQF Board failed to act on two of the draft consensus standards, weight loss and staffing, and instead directed that these standards be reconsidered within NQF’s Consensus Development Process with assistance from the Steering Committee.  The Steering Committee met by telephone conference call to review those two measures on December 10, 2003; participating in the call were Steering Committee members Lee Partridge (co-chair), Judy Salerno (co-chair), Irene Fleshner, David Gifford, Rosalie Kane, William Kubat, Donna Lenhoff, Katie Maslow, Dale Thompson, Marvin Tooman, and Lisa Payne Simon; CMS staff Lisa Hines, Pam Cheetham, and Jean Scott; and NQF staff Reva Winkler, Elaine Power, and Larry Gorban.

Weight Loss

The CHSRA weight loss measure had been recommended by the Steering Committee but was not approved by the Provider/Health Plan Council in the first round of voting because it did not incorporate “obvious exclusions” of residents who are near end-of-life or on weight-loss programs.  The Steering Committee discussion raised the following main concerns about those exclusions:

• Some proportion of residents identified as end-of-life on the MDS are alive after 6 months
• MDS 2.0 does not have a data element that distinguishes weight loss from weight gain programs, but rather merely a data element for residents in “planned weight change programs.”  It is senseless to exclude residents who are really on weight-gain programs from this standard.  Moreover, residents on weight loss-programs are a very small number in the chronic care nursing-home resident population.

On the other hand, inducing an incentive for use of feeding tubes to prevent weight loss was discussed, along with a consideration for adding a prevalence of feeding tubes measure paired with the weight loss measure.  The Committee was not inclined to recommend new measures at this point, and CMS noted that the feeding tube measure was not validated in the Mega QI study.

After some deliberation, the Steering Committee participants reached consensus on the following:

·        It is essential that a weight loss measure be part of the set of measures.

·        An exclusion for hospice patients is acceptable.

·         Because MDS 2.0 cannot identify patients on weight loss programs for possible exclusion, an exclusion using the MDS planned weight change program data element as a proxy is not recommended.

·        The possible incentive to use feeding tubes should be monitored carefully.

 Staffing

The nurse staffing measures had been approved by all four Member Councils but not endorsed by the Board (even though NQF Members supported its endorsement) at that time, instead opting for additional consideration.  At the September 30, 2003, Board meeting, CMS expressed concern that the OSCAR staffing data are not of sufficiently high quality (even though it displays these data on its web site).  Board members wanted the Steering Committee to address those concerns, as well as to consider the two measures of nurse staffing for acute care hospitals that the Steering Committee of NQF’s “Nursing-Sensitive Measures” project recently recommended (available at http://www.qualityforum.org/nursing_sensitive_performance_measurement.html).  Steering Committee members also noted the release, since its original review of the measure set, of a new IOM report, Keeping Patients Safe:  Transforming the Work Environment of Nurses (November 4, 2003), which recommends including a measure of nurse staffing in federal and state report cards of quality in nursing homes.

At the request of Steering Committee members, Dr. Charlene Harrington, principal investigator for the staffing study from California, joined the call. Dr. Harrington summarized the methodology used to report staffing measures on the “California Nursing Home Search” web site sponsored by the California HealthCare Foundation, explaining that the data elements are “essentially identical” as CMS uses on the Nursing Home Compare web site and that the measures are well defined—i.e., have clear specifications; that the data are collected annually as part of cost reports, and cover the entire year (in contrast to OSCAR, which reports data from a 2-week period before a survey); and that California uses the RUGS Case-Mix Index to categorize homes into high, average, and low case-mix groups, and assigns different staffing targets for each RUGS group.  She also explained that about one-third of states do not use cost reports, and that the format of cost reports varies among states.

CMS representatives reported that CMS has just started a contract with Andy Kramer at the University of Colorado to develop new staffing measures, including the types of measures recommended by NQF.  Representatives present did not have information on pending CMS research on staffing data using payroll information.

Members of the Steering Committee participating on the call discussed the issues at length.  While the importance of including a staffing measure – or of using the ratio of hours per resident day of care by RNs, LPNs, and CNAs as the appropriate measure -- was not in question, data standardization and quality were.  There was consensus that the staffing information on Nursing Home Compare must continue, and NQF should not do anything that might encourage that information to be removed.  In general, the Committee also felt that improving the OSCAR system is preferable to using various data sources at the state level, but differed on whether to endorse standards based on OSCAR pending its improvement.  Possibilities for handling this situation included (1) NQF endorsing the inclusion of the staffing standard in the nursing home set, encouraging the immediate improvement of OSCAR, and allowing for the best available data source to be used until an improved, federal data source is available; (2) including the staffing standard without specifying a data set but encouraging immediate improvement of OSCAR; and (3) not including a staffing standard at all pending the improvement of OSCAR.   Notably, the first or second options were supported not only by NCCNHR, but also by Steering Committee members Lee Partridge of the National Partnership for Women & Families, Dave Gifford of the Rhode Island Quality Improvement Organization, David Zimmerman of the University of Wisconsin Center for Health Systems Research and Analysis, and Katie Maslow of the Alzheimer’s Association. 

At the conclusion of the call, NQF staff were asked to draft these options for the Steering Committee to respond to electronically, so that their opinions could be shared with the Board more formally.  Final Steering Committee responses to the options are due December 29, 2003; the NQF Board is expected to take the matter up again at its January, 2004, Board meeting, currently scheduled for January 29.