Years ago, I presented an analysis at a “fairly-loaded” conference in DC, and by loaded I mean that the room was filled with “heavy hitters”, and by heavy hitters I mean, well let’s just say significant stakeholders from the provider community and CMS. The PointRight analysis that I presented addressed negative surveys and the many things that were out of the control of a nursing home provider but were highly correlated to them, like ZIP code, facility size and proprietary status. I was surprised by the knowing reaction of the audience when I said that ZIP code was the best predictor of survey experience. It seemed that everyone was well aware that the best strategy to improve survey performance was to move the facility!
Since then, CMS and our professional associations have made significant strides to improve the consistency around the survey and certification process. In fact, that was one of the stated goals of QIS survey. CMS works with regional offices, associations and providers to clarify regulations and policies. Has this made a difference? Let’s see.
Over the last two years, the total number of deficiencies per facility has declined. Bravo! However, significant geographic differences remain. Seriously, what is going on in Regions 8, 9 and 10? While in Q3 2012, the nation’s nursing homes had, on average, seven survey and complaint deficiencies, facilities in these regions had 8.47, 9.66 and 8.78 respectively. Drilling in a little further, we see that 49% of Idaho facilities received a G-level deficiency when nationally only 19% did. It seems Idaho is not just producing potatoes and a variety of precious and semi-precious stones. How are our QIS states doing? Florida, a state that is 100% QIS, had, in Q3, a total deficiency count of 6.22; below the national average. However, drilling deeper into their State Survey districts, we see a range from 4.20 to 7.66, the latter district having 19% of facilities with level-G deficiencies compared to 8% for the state.
No one denies that there is much more work to be done to modernize the survey and certification process and ensure fairness and consistency. As important as addressing the geographic consistency issue and perhaps even more so, is formally studying the influence that distinct resident populations have on survey outcomes. For example, facilities with the highest proportion of residents with mental illness (> 40%) have decidedly worse survey outcomes, while a high Medicare population (>19%) almost seems to have a buffering effect.
Now colleagues, the message I hope to leave you with is certainly not to abandon the survey process and those outcomes due to non-modifiable factors; that would be counter-productive, but to ensure that you are measuring your performance accurately. As you engage in quality improvement efforts, put together data-driven marketing materials, complete a HUD refinance application, or talk with families and other community members take care to make sure you’re using an appropriate comparative benchmark. If you operate in Massachusetts, you’ll always be a rock star compared to those facilities in Colorado. If you operate almost exclusively with Medicare you’ll likely always outperform the inner city home specializing in end of life care. However, for an authentic evaluation of your performance, choose like facilities in similar locations, ideally the same survey district, caring for similar residents.
Steven Littlehale is a gerontological clinical nurse specialist, and EVP and chief clinical officer at PointRight Inc.
