Can an outcomes-based quality improvement program reduce unnecessary hospitalizations among home care patients?

Home health care agencies participating in a national demonstration program used regular reports on their patients' outcomes to plan and make improvements in care. The hospitalization rate fell by 22 percent over three years among agencies nationally and by 26 percent over four years among New York State agencies.

Why is this important?

The goals of home health care include "assisting patients to become or remain sufficiently independent to stay in their home environment, avoiding institutional long-term care or acute care" to the degree possible (Shaughnessy et al. 2002b). Nationally, more than one-quarter (28%) of home health care patients are hospitalized, a rate that has not changed in the past few years (AHRQ 2005).

Although hospital admissions are sometimes planned or necessary to provide optimal care and assure patients' health, some hospitalizations represent adverse outcomes resulting from preventable events, such as falls or acute exacerbations of chronic conditions. Home health agencies (HHAs) might be able to reduce such adverse outcomes (and costs of care) through proactive needs assessment and coordination with the patient's physician to provide timely care interventions in the home.

Interventions

The federal government, in collaboration with New York State and the Robert Wood Johnson Foundation, developed the Outcome-Based Quality Improvement (OBQI) system to support continuous quality improvement in HHAs (Shaughnessy et al. 2002a). Using the Outcome and Assessment Information Set (OASIS), patient assessments are centrally collected and analyzed to produce annual reports comparing an agency's performance with a national reference for 41 risk-adjusted outcomes measures.

HHAs use these outcome-based quality measurement reports to:

• target outcomes for improvement,
• investigate care processes to determine problems,
• identify best practices to improve care,
• plan and implement actions to achieve those practices, and
• monitor effectiveness.

For example, one agency's plan to reduce unplanned hospitalizations included criteria to identify patients with unstable conditions or otherwise in need of follow-up care and a protocol for nurses to contact the patient's physician within 24 hours to schedule a follow-up call or visit (Richard et al. 2000).

Findings

Through participation in the National OBQI Demonstration, which included training and technical assistance to implement OBQI, 54 HHAs in 27 states reduced risk-adjusted hospitalization rates by 22 percent over three years.

The 19 HHAs participating in the New York State OBQI Demonstration achieved a similar reduction of 26 percent over four years. In contrast, hospitalization rates changed only very little for a random sample of non-OBQI Medicare patients in the same 27 states during the three-year national demonstration (Shaughnessy et al. 2002b).

Implications

Although use of OBQI is voluntary, the federal government has collaborated with state agencies to offer training on the OBQI system to HHAs nationwide. Medicare quality improvement organizations (QIOs) also are helping HHAs implement OBQI, with a focus on reducing hospitalizations.

Improvement Ideas and Resources

The study authors noted that physician involvement was an important factor in agency-level improvement and that most agencies needed to improve communication with physicians to achieve this effect.

For additional resources, visit the Web site of the Centers for Medicare and Medicaid Service's Medicare Quality Improvement Community.

Measure:

This comparative before-and-after study included 157,548 patients admitted over three years to 54 Outcome-Based Quality Improvement (OBQI) agencies participating in the national demonstration trial in 27 states and 105,917 patients admitted over four years to 19 OBQI agencies participating in the New York State demonstration trial. The trials occurred from 1995 to 2000. OASIS data on 41 outcome measures were collected for each patient within the OBQI program at the start of care and every 60 days until discharge, when final data were collected. For each pre/post comparison shown in the chart, the first year is risk-adjusted (using logistic regression) and the second year is an observed rate. Risk models were revised for each annual reporting period. Rates differ between comparison periods because of risk-adjustment and differences in sample sizes (a few agencies were excluded in some years because of data reporting issues).

The net decreases in rates between each year in the chart were statistically significant. The 22 percent relative decrease in hospitalization rate for the national demonstration (described in the narrative) reflects a risk-adjusted net decrease of 7.2 percentage points when Year 3 is compared with Year 1. The 26 percent relative decrease in the New York State demonstration reflects a risk-adjusted net decrease of 7.9 percentage points when Year 4 is compared with Year 1. A comparative analysis used Medicare claims data for a 5 percent random sample of home health patients in the same 27 states who were cared for by agencies that were not participating in the demonstration. This analysis found a decrease in the hospitalization rate of 0.4 percent from Year 1 to Year 2 and a 0.3 percent decrease from Year 2 to Year 3 for the comparison patients (Shaughnessy et al. 2002b).

Source:

Results are based on Outcome and Assessment Information Set (OASIS) patient assessments. The study was conducted by researchers at the University of Colorado Health Sciences Center (Shaughnessy et al. 2002b).

References:

* Indicates source of data used in the chart(s).AHRQ (Agency for Healthcare Research and Quality). 2005. National Healthcare Quality Report, 2005. AHRQ Publication No. 06-0018. Rockville, Md.: U.S. Department of Health and Human Services.

Richard, A. A., K. S. Crisler, and P. M. Stearns. 2000. Using OASIS for Outcome-Based Quality Improvement. Home Health Nurse 18 (4): 232–7.

Shaughnessy, P. W., K. S. Crisler, D. F. Hittle et al. 2002a. OASIS and Outcome-Based Quality Improvement in Home Health Care: Research and Demonstration Findings, Policy Implications, and Considerations for Future Change. Denver, Colo.: University of Colorado Health Science Center, Center for Health Services Research.

* Shaughnessy, P. W., D. F. Hittle, K. S. Crisler et al. 2002b. Improving Patient Outcomes of Home Health Care: Findings from Two Demonstration Trials of Outcome-Based Quality Improvement. Journal of the American Geriatrics Society 50 (8): 1354–64.