National Nursing Home Quality Data Release Generally Applauded

by Susan M. Pettey, JD, MPA

The national data on nursing home quality in all 50 states is now available to the public, following its November release by the Department of Health and Human Services (HHS). Nearly all sectors of the long-term care industry heralded its release; however, the General Accounting Office (GAO) released a report, prepared in October 2002, stating that national implementation was premature (http://www.gao.gov/new.items/d03187.pdf).

The GAO critique did not deter HHS Secretary Tommy G. Thompson from moving forward with the release, however. "This is a new approach to bringing about better quality care in our nation's nursing homes. Not only will consumers be better informed, but nursing homes themselves will be able to see more clearly what they must do to make the quality grade," the Secretary said.

The data release is part of the national Nursing Home Quality Initiative, announced in November 2001, to augment existing public data on nursing home quality and provide assistance to nursing homes in quality improvement.

"We know that nursing homes are just as interested in improving the high quality care they already give to their residents as we are," said Tom Scully, Administrator of the Centers for Medicare and Medicaid Services. "By making this information available to the nursing homes and consumers, we are seeing a collaborative effort to do even more to raise the bar on quality."

Quality in Context

The nursing home quality data shows how individual facilities compare with others nationally and in their region. Information about the quality comparisons may be accessed at www.cms.hhs.gov, under "Nursing Home Quality Initiative"; the actual data are at www.medicare.gov under "Nursing Home Compare" or by calling 1-800-Medicare.

As noted previously in Caring (see "Mixed Reaction to CMS' Release of Nursing Home Quality Data" and "Making (Up) the Grade: What You Need to Know About Nursing Home 'Report Cards'" in the August 2002 issue, pages 1 and 3, respectively), there are ten indicators, which are divided into two groups based on length of stay in the nursing facility.

There are six measures for long-stay residents. These include the percentage of residents with:

• Loss of ability in basic daily activities
• Infections
• Pain
• Pressure sores
• Pressure sores after additional facility-level risk adjustment
• Physical restraints

There are four measures for short-stay residents. These include the percentage of residents with:

• Delirium
• Delirium after additional facility-level risk adjustment
• Improvements in walking
• Pain

To launch the report cards, HHS placed advertisements in 71 newspapers nationally, showing a sampling of the quality data. The November release followed a pilot program, which began in April 2002, in the states of Colorado, Florida, Maryland, Ohio, Rhode Island, and Washington. To develop the quality indicators, a National Quality Forum (NQF) steering committee reviewed recommendations by a CMS contractor, Abt Associates, and made initial recommendations for quality measures for the pilot. That steering committee included providers, state government representatives, consumer advocates, and others.

In response to the evaluation of the pilot, CMS dropped one measurement on weight loss based on results of reliability testing, and made minor adjustments to the measures. The CMS deadline for a November roll-out did not allow NQF adequate time to develop and approve final recommendations. Nonetheless, CMS felt sufficiently comfortable with the accuracy and reliability of the data to proceed with the national release.

Risk Adjustment Crucial

Long-term care experts have long sought effective methods for adjusting nursing home data based on relevant risk factors in order to more accurately compare "apples to apples". Risk adjustment has proved to be crucial to interpreting nursing home data, based on a report to CMS by Abt that showed that risk adjustment could raise or lower facility scores by 40% or more.

In October 2001, Abt recommended that CMS adopt 39 risk-adjusted quality indicators. The NQF, a second contractor, reviewed Abt's recommendations, and identified 11 indicators to use in the pilot; but NQF did not adopt Abt's risk adjustment approach because research on it was still ongoing.

Both Abt and NQF subsequently concluded that adjusting for the type of residents admitted to the nursing home required further research.

In light of those developments, the CMS Quality Initiative adopted a very limited number of risk adjustments for either resident-level risk or facility-level risk. Three of the ten indicators, pressure sores, delirium, and walking, have been adjusted for facility-level differences, to reflect practices such as admitting frailer, sicker patients or specializing in a particular area of care that may account for a higher rate on a particular quality measure.

Two indicators, pain and delirium, have been adjusted for resident-level differences--for example, residents with cognitive impairments who may not be able to clearly express levels of pain.

Other adjustments have been made to exclude certain groups of residents--such as those in a coma--from measurements that cover unexpected loss of function in daily activities. Data on residents with missing assessment information were also excluded.

GAO Critique

The GAO warned that national implementation was premature because the final evaluation of the pilot study and final recommendations by NQF would not be concluded until 2003. GAO was apprehensive that changes soon after implementation would cause confusion for both consumers and providers.

GAO also indicated that concerns remain about the underlying accuracy of the MDS data used to develop quality indicators. In addition, GAO feared that the public may be confused by the quality data that are posted at the Nursing Home Compare site, citing conflicts between the quality indicator data and survey data.

GAO further found that CMS is not prepared to respond to consumers' questions, either through its hotline operators or through its technical assistance contractors, the Quality Improvement Organizations (QIOs; formerly known as Peer Review Organizations). CMS is using 37 QIOs to provide information and consultation to nursing facilities to identify and resolve the root causes of quality problems. Based on the pilot project, the QIO assistance is likely to be popular with providers. Half of all the nursing homes in the pilot states requested QIO assistance.

Key Stakeholders React

Long-term care providers praised the data release. "We support this initiative," said William L. Minnix, Jr., DMin, President of the American Association of Homes and Services for the Aging (AAHSA). "We welcome CMS' development of a technical-assistance program to help nursing homes improve their quality of care, and we support the public disclosure of quality measures to help consumers make more informed decisions."

Mary K. Ousley, Chair of The American Health Care Association (AHCA), said the information release signaled "a new level of cooperation between government, health care providers, patients, consumers, and family members in the effort to improve the quality of care provided to our nation's current and prospective nursing home patients." She added that the Quality Initiative must be supported by adequate funding of care for Medicare and Medicaid patients.

Nevertheless, some stakeholders expressed concerns similar to those raised by GAO. AMDA's Executive Director, Lorraine Tarnove, noted that "The GAO report echoed concerns that AMDA has raised since the beginning of the project. We supported the goal of providing consumers with usable information about facility performance, but also recognized the limitations, which were covered in the August 2002 issue of Caring, and are also discussed elsewhere in this issue. Therefore, AMDA created guidance to facilitate consumer understanding of how the quality measures can be used, and what questions to ask about them when making decisions about placement or judgments about a facility's quality performance." (see www.amda.com/consumers).

Providers are also concerned that the risk adjustments are not sufficient to provide accurate comparisons, citing, for example, the lack of an adjustment for the multiple medical conditions that frail, elderly nursing home residents usually present with.

On balance though, providers seem to feel that it is worth living with imperfections in the data in order to test their potential to spur quality improvement.

The National Citizens Coalition on Nursing Home Reform (NCCNHR) also welcomed the data release. NCCNHR Executive Director Donna Lenhoff, who is also a member of Caring's Editorial Board, expressed the hope that the data will alert consumers to "the substandard conditions in many of America's nursing homes." However, Ms. Lenhoff faulted the lack of a staffing measure, noting the "clear link between staffing levels and the quality of care." NCCNHR also cautioned that the data should not supplant thorough annual inspections, prompt complaint investigations, and strict enforcement.

Both NCCNHR and AARP recommend that consumers make repeated visits to potential nursing facilities at different times of day, and talk with staff and residents about their perceptions of the quality of care.

Continuing Developments

The NQF will consider the validation report and take into account lessons learned from the pilot study. NFQ may also reconsider the use of several quality indicators that CMS did not adopt in the national roll-out--namely, depression, incontinence, catheterization, bedfast residents, weight loss, and rehospitalization.

However, NFQ reaches its recommendations based on consensus, which can result in a prolonged decision-making process. Current plans call for final NQF recommendations in 2003. CMS has pledged to revise the indicators based on the NQF recommendations.

Susan Pettey, JD, MPA, is former Director of Government Affairs for the American Medical Directors Association and a Contributing Writer to Caring.

back to top