Cost efficiency of nursing homes: do five-star quality ratings matter?

Abstract

Nursing homes may respond to the pressure to reduce costs by reducing quality of care, so the two are related. This study examines the determinants of nursing home costs and cost efficiency, and investigates how various measures of nursing home care quality influence both of these. It applies a one-step stochastic frontier approach to a large panel of California nursing homes surveyed between 2009 and 2013. Quality is measured by three different ratings available on the Nursing Home Compare website: rating on quality measures, rating on the health inspection, and rating on staffing levels. Results show that the rating on quality measures, an outcome-based measure of quality, is inversely related to costs but unrelated to mean cost efficiency. In other words, a better rating on quality measures is associated with lower nursing home costs. The health inspection rating is not associated with either costs or mean cost efficiency. The rating for staffing levels, a structural measure of quality, is negatively associated with cost efficiency. These findings reveal that different measures of quality have different relationships with costs and cost efficiency. The findings suggest that better quality outcomes in nursing homes may be achievable with fewer resources and/or improved care procedures, which in turn should reduce nursing home costs.

Appendices

Appendix A: Measures of quality for long-stay and short-stay residents

Quality measures for long-stay residents are:

• 1. Percent of residents whose need for help with daily activities has increased.

• 2. Percent of high risk residents with pressure sores.

• 3. Percent of residents who had a bladder inserted and left in the bladder.

• 4. Percent of residents who were physically restrained.

• 5. Percent of residents with a urinary tract infection.

• 6. Percent of residents who self-report moderate to severe pain.

• 7. Percent of residents experiencing one or more falls with major injury.

Quality measures for short-stay residents are:

• 1. Percent of residents with pressure ulcers (sores) that are new and worsened.

• 2. Percent of residents who self-report moderate to severe pain.

Appendix B: Case-mix index calculation

Following Cohen and Spector [31], the case mix index of a nursing home is measured at the facility-level as the number of minutes of staff time required for the care of the average resident. More specifically, using weights based on the management minutes system developed by Thoms and Schlesinger, the case-mix index is calculated as:

$$ \mathrm{A}(20)+\mathrm{B}(18)+\mathrm{C}(30)+\mathrm{D}(30)+\mathrm{E}(20)+\mathrm{F}(48)+\mathrm{G}(90)+\mathrm{H}(90)+\mathrm{I}(45)+\mathrm{J}(32)+\mathrm{K}(20)+\mathrm{L}(50)+\mathrm{M}(36) $$

where

A:

percentage of patients needing full assistance bathing

B:

percentage of patients needing partial assistance bathing

C:

percentage of patients needing full assistance dressing

D:

percentage of patients needing partial assistance dressing

E:

percentage of patients catheterized

F:

percentage of patients incontinent

G:

percentage of patients needing parental feeding

H:

percentage of patients needing tube feeding

I:

percentage of needing assistance eating

J:

percentage of patients non-ambulatory

K:

percentage of patients with pressure sores

L:

percentage of patients receiving bowl/bladder retraining, and

M:

percentage of patients receiving special skin care.