Improving Performance in Late-Life Care

I’ve now been involved in hospice and palliative care for 25 years, on the management rather than clinical side. During that time, I’ve seen how the principles of palliative care can improve care for those with advanced or serious illnesses. While hospice has been the organized vehicle through which much of this care has been applied, and there’s no disputing that improvements in late-life care have been made over the past couple of decades, hospice does have its linmitations. And so, I’ve spent much of my time studying organizational and business models that can successfully "distribute"  palliative care principles across a larger population.

Part of that study has involved benchmarking on an organizational level, that is, using benchmarking principles to improve the performance of individual hospices. As a hospice executive, this exercise certainly has value.

There’s another measure of performance improvement. The firm with which I’m associated assists palliative care organizations in crafting medical staff development plans. These are staffing plans that are aligned with the organization’s objectives and the community’s needs. We’ve determined that one of the best ways to identify a community’s needs for palliative care is to use available data about the community’s practices around palliative care. The best source , we’ve found, of such information, is the Dartmouth Medical Atlas. The Dartmouth Medical Atlas began in 1993 out of Dartmouth University as a study of utilization of health care resources, generally inpatient resources, in geographic markets across the US. The study uses Medicare claims data. The Dartmouth research group has been updating and expanding its studies. Recently, a study was conducted of Medicare beneficiaries who died (decedents) between 2001 and 2005. This study analyzed use of resources (inpatient, outpatient, home, and long-term care facility), in the final two years of life. The results begin to answer the question of how we care for those with advanced illness.

The researchers concluded, among other things, that there is enormous variation (across regions, and across hospitals within those regions) in how people are cared for near the end of life. When one considers that 75% of Medicare resources are spent to care for those in the last two years of life,  I would be understating to say there is great interest among Medicare officials and other payers on how care is provided. In fact, one of the calls to action the researchers recommended is that pay-for-performance programs should reward those practices that improve system-wide efficiency in providing care for those with chronic illness.

The goal of performance improvement as it applies to healthcare systems is to reduce variability in the way care is provided. And a key factor in closing the gap between best practice and common practice is the ability of health care providers and their organizations to rapidly spread innovations and new ideas. I'm interested to learn how palliative care specialists are "spreading the science".

Closing the Performance Gap in Palliative Care by Reducing Clinical Variation

The Dartmouth Medical Atlas gives us a compelling portrait of the variations in practice in late-life care. It also offers insights into the palliative performance improvement opportunities in communities.


How wide are the variations? Let’s take a look at Charlottesville, Virginia, an area of about 100,000 residents, with an academic medical center and a community hospital, a large not-for –profit hospice, and plenty of smaller hospices. According to the DAI Palliative Performance Profile (drawn from the Dartmouth Medical Atlas), Charlottesville earned a B grade for its late-life care practices. Respectable, for sure, but short of exemplary. What’s the difference between respectable performance and that of the top 10% performers? Consider the following: in Charlottesville, 32% of Medicare decedents died in a hospital. That’s better than the state average, yet 25% higher than residents of a community in the top 10%. In other words, a Charlottesville resident is 25% more likely than a resident of a high-performing community to die in a hospital than at home. That same Charlottesville resident is 40% more likely to have spent 7 days or more in a hospital during the last six months of life than someone in a top-performing community. Considerable room for improvement, I think you'd agree, and yet Charlottesville's performance is better than most communities across the nation.

Tightly integrated delivery systems have proven to be the most effective in reducing clinical variation. Because of their structure, these delivery systems are adept at disseminating evidence-based practices. It is time that hospice and palliative care leaders accelerate collaborative efforts to create networks with greater potential to reduce clinical variation in late-life care. What better opportunity for HPM specialists to take the lead?