Does mandating nursing home participation in quality reporting make a difference? Evidence from Massachusetts

Abstract

Background: Quality report cards have been shown to be effective in influencing patients' referrals and promoting quality improvement in some instances and not others. In this study, we investigate one of the mechanisms that may detract from their effectiveness: voluntary versus mandatory participation of nursing homes in public quality reporting.

Objectives: To answer 2 questions: (1) Were the nursing homes choosing not to participate low-quality performers relative to those who chose to participate? (2) Once participation became mandatory, did those that did not voluntarily participate initially, improve more than those that participated voluntarily?

Research design: Massachusetts published the Massachusetts Satisfaction Survey report card for nursing homes for the years 2005, 2007, and 2009. Nursing homes' participation was voluntary in 2005 and mandatory in 2007 and 2009. We performed a retrospective statistical analysis of the relationship between nursing homes' decision to participate in quality reporting and 12 quality outcomes: deficiency citations, staffing, and 8 survey domains.

Subjects: A total of 424 Massachusetts nursing homes.

Results: Sixty-seven percent of nursing homes participated in reporting voluntarily. Volunteer nursing homes had better quality for all measures (significant at the 0.05 level or trending toward significance at the 0.10 level for all but 2). Once reporting became mandatory, nonvolunteers improved more than volunteers in all but 2 staffing measures (trending toward significance at the 0.10 level in 5).

Conclusions: Report cards are more effective if nursing homes' participation is mandated. Nonmandatory reporting systems, as those implemented by some states and professional associations, lead to missed opportunities for quality improvements.

Figure 1

Differences in facility-reported staffing level and deficiencies between volunteer and non-volunteer nursing homes: Positive values indicate better quality for volunteer nursing homes Note: Each bar shows the difference between the QM for the volunteers and the QM for the non-volunteers, for each year ( ${\mathrm{\Delta }}_t={\mathit{\text{QM}}}_t^{\mathit{\text{Vol}}}-{\mathit{\text{QM}}}_t^{\mathit{\text{Non}}-\mathit{\text{Vol}}}$). A decrease in the height of the bar in 2009, Δ₂₀₀₉, compared with 2007, Δ₂₀₀₇, indicates that the non-volunteers have improved more than the volunteers by 2009. # The difference in the QM between the volunteers and the non-volunteers, Δ_t, was significant at the 0.05 level. * The gap between volunteers and non-volunteers, Δ₂₀₀₉–Δ₂₀₀₇, decreased from 2007 to 2009, trending towards significance (p<0.10).

Figure 2

Differences in satisfaction measures between volunteer and non-volunteer nursing homes: Positive values indicate better quality for volunteer nursing homes Note: Each bar shows the difference between the QM for the volunteers and the QM for the non-volunteers, for each year. ( ${\mathrm{\Delta }}_t={\mathit{\text{QM}}}_t^{\mathit{\text{Vol}}}-{\mathit{\text{QM}}}_t^{\mathit{\text{Non}}-\mathit{\text{Vol}}}$) A decrease in the height of the bar in 2009, Δ₂₀₀₉, compared with 2007, Δ₂₀₀₉, indicates that the non-volunteers have improved more than the volunteers by 2009. # The difference in the QM between the volunteers and the non-volunteers, Δ_t, was significant at the 0.05 level. & The difference in the QM between the volunteers and the non-volunteers, Δ_t, trended towards significance (p<0.10). * The gap between volunteers and non-volunteers, Δ₂₀₀₉–Δ₂₀₀₇ decreased from 2007 to 2009, trending towards significance (p<0.10).