This blog post originally appeared at Home Dialysis Central's website. Find more posts from this blog, Kidney Views, at http://www.homedialysis.org/news-and-research/blog.
Two or three years ago at the Annual Dialysis Conference, I was lucky enough to hear thought leader nephrologist Tom Parker III speak on U.S. dialysis measures and clinical practice guidelines. Since the non-profit Medical Education Institute developed the first set of DOQI guidelines in the mid 1990s before they were given over to the National Kidney Foundation—and I helped design the workgroup structure and staffed the Anemia work group—his spot-on points truly made me cringe.
Two leading causes of death on dialysis in the US are sudden cardiac death and infection, and there is no clinical practice guideline for either one, Dr. Parker noted in his talk.
The guidelines at that point focused on:
- Anemia (since Amgen was paying)
- Hemodialysis adequacy (since the RPA had recently done a guideline in this area, and politically it was imperative to include this)
- PD adequacy (to be parallel to HD adequacy)
- Bone disease (because we had recently done some work in this area)
At no time was there ever a conversation that started out, "what are the leading causes of preventable death in dialysis, and how can we address those?"
The truly appalling and terrifying thing is—13 years later, this conversation did finally occur (in the form of the well-publicized ESRD: State of the Art and Charting the Challenges for the Future conference held at Harvard in 2009, with Dr. Parker and a star-studded cast of U.S. nephrology experts. Boy, I wish I'd known about that conference at the time…)—but CMS is still stuck in the rut of the KDOQI and KDIGO guidelines, which have been solidified into stone, written into regulation, and still miss the point entirely about how best to improve dialysis outcomes. To me, this is a classic example of setting up a ladder against the wrong wall. We are measuring the wrong things, and missing the right ones.
In an article entitled Dialysis at a Crossroads: 50 Years Later,1 Dr. Parker, Dr. Ray Hakim, Dr. Allen Nissenson, Dr. Ted Steinman, and Dr. Richard Glassock recommend a sea change in our measures focus for dialysis, aligned with what really does matter for helping patients to live longer and better:
- Timely (not early) dialysis start
- Mindful management of the first 3-4 months of dialysis
- HD catheter avoidance
- Extracellular fluid volume control all the time (not just during treatments) Longer or more frequent sessions
- "Mind the left ventricle"
- Moderation in prescribed ESAs, iron, vitamin D, binders, and calcimimetic drugs
- Infection control
- Limiting only salt in the diet
So, how can we get to the RIGHT measures? The ones that really do make a difference to quality of life and survival? I have two thoughts:
- While each of the clinical performance measures (CPMs) individually has some merit, they are powerful predictors of survival when combined together! Rocco et al2 looked at adequacy (Kt/V≥1.2), anemia (hemoglobin ≥11.0 g/dL), fistula for vascular access, and albumin (≥4.0 g/dL or ≥3.7 g/dL bromcresol green or bromcresol purple lab methods, respectively) among 15,287 individuals on hemodialysis:
- 6% met none of the four CPM targets—and 12 month mortality was 29%
- 24% met one target—and 12 month mortality was 25%
- 39% met two targets—and 12 month mortality was 21%
- 24% met three targets—and 12 month mortality was 14%
- 7% met all four targets—and 12 month mortality was 7%
Dialysis clinics could change their lab "report cards" to share this information with consumers, so everyone can see at a glance how many measures each person is on target for, and act to improve the "score" and each patient's chance of survival.
Mendelssohn et al did a similar analysis of 6,664 individuals receiving care at DOPPS-enrolled clinics in seven countries.3 They looked at the facility level at what percentage of consumers met targets for Kt/V, hemoglobin, albumin, and catheter use, with similar findings, calling it the Practice Risk Score (PRS). For each 0.1% rise in the PRS, the relative risk of death increased by 5%, and these results were highly significant. A tool could be built to help clinics calculate their PRS—and this figure could be submitted to the National Quality Forum (NQF) as a clinical performance measure that would matter, and that CMS could adopt.
- Speaking of the NQF, I would like to call on the leadership of the Dialysis at a Crossroads conference to use their corporate resources to turn their suggested list into measures and submit them. Let's move from data and excellent rhetoric to action! Let's change the world for dialysis patients! It's not enough to do research and talk about it. We need to move it into practice.
A quote attributed (rightly or wrongly) to Albert Einstein says, "Not everything that can be counted counts, and not everything that counts can be counted." In the case of dialysis, we are still doing too much that doesn't count, and not enough that does. But, we can change that. And, we need to.
- Parker T, Hakim R, Nissenson AR, Steinman T, Glassock RJ. Dialysis at a crossroads: 50 years later. Clin J Am Soc Nephrol. 2011. 6:457-61
- Rocco MV, Frankenfield DL, Hopson SD, McClellan WM. Relationship between clinical performance measures and outcomes among patients receiving long-term hemodialysis. Annals Int Med. 2006;145:512-19
- Mendelssohn DC, Pisoni RL, Arrington CJ, Yeates KE, Leblanc M, Deziel C, Akiba T, Krishnan M, Fukuhara S, Lamiere N, Port FK, Wolfe RA. A practice-related risk score (PRS): a DOPPS-derived aggregate quality index for haemodialysis facilities. Nephrol Dial Transplant. 2008 Oct;23(10):3227-33