VOL 50, NO 5, NOVEMBER 2007
EDITORIALS Imperial or Empirical Measures of Dialysis Quality? Related article, p. 774 So now the Emperor walked under his high canopy in the midst of the procession, through the streets of his capital; and all the people standing by, and those at the windows, cried out, “Oh! How beautiful are our Emperor’s new clothes! What a magnificent train there is to the mantle; and how gracefully the scarf hangs!” In short, no one would allow that he could not see these much-admired clothes; because, in doing so, he would have declared himself either a simpleton or unfit for his office. Certainly, none of the Emperor’s various suits, had ever made so great an impression, as these invisible ones. “But the Emperor has nothing at all on!” said a little child. “Listen to the voice of innocence!” exclaimed his father; and what the child had said was whispered from one to another. “But he has nothing at all on!” at last cried out all the people. The Emperor was vexed, for he knew that the people were right; but he thought the procession must go on now! And the lords of the bedchamber took greater pains than ever, to appear holding up a train, although, in reality, there was no train to hold. Hans Christian Andersen, “The Emperor’s New Clothes”1
T
he Centers for Medicare and Medicaid Services’ end-stage renal disease (ESRD) Clinical Performance Measures represented a response to the political imperative to show value for dollars spent treating kidney failure. In publishing selected results on the internet, the Centers for Medicare and Medicaid Services attempt to use market forces to enforce accountability for dialysis care.2 However, after reporting hemoglobin level and hemodialysis dose for more than a decade,3 dialysis facilities continue
to vary significantly in values on these parameters.4,5 Perhaps some facilities are good and some bad, and the savvy consumer of dialysis should check the web and avoid the latter. How, then, should we take the news that 2 accepted measures of dialysis quality fail to correlate strongly with one another? In this issue of AJKD, Spolter and colleagues6 report their study of hemodialysis patients in ESRD Network 5 (Maryland, Virginia, West Virginia, and the District of Columbia). They asked which patients had both an average hemoglobin of at least 11 g/dL (110 g/L) and an average urea reduction ratio of at least 65% during the last quarter of 2004. If at least 80% of a facility’s patients satisfied this hemoglobin criterion, the authors consider it to have met the hemoglobin standard. If at least 80% of patients satisfied this dialysis dose criterion, they consider it to have met the dose standard. As compared to facilities that did not meet the hemoglobin standard, those that did were 2.6 times more likely to meet the dose standard. However, the authors show that meeting the hemoglobin standard explained little inter-facility variability in meeting the dose standard, and conclude that different processes determine quality for each of these measures. Are we surprised? Physicians, whether working individually or as prompted by a protocol, can prescribe a higher erythropoietin dose whether or not they specify a larger dialyzer, Address correspondence to Dana C. Miskulin, MD, Division of Nephrology, Tufts-New England Medical Center, Box #391, Boston, MA 02111.E-mail:
[email protected] © 2007 by the National Kidney Foundation, Inc. 0272-6386/07/5005-0001$32.00/0 doi:10.1053/j.ajkd.2007.09.008
American Journal of Kidney Diseases, Vol 50, No 5 (November), 2007: pp 691-695
691
692
persuade the patient to stay for a longer treatment, or coordinate the creation of an arteriovenous fistula. The hypothesis that facilities satisfying one standard would be more likely to satisfy the other would imply either a strong biological association between the 2 outcomes, or that the results of anemia treatment and hemodialysis prescription reflect common underlying organizational characteristics. It is true that uremic patients are usually anemic, but at hemodialysis doses generally delivered in 2004, it would not seem plausible that many patients were anemic because of inadequate solute clearance. As for organizational characteristics, in 2004, practice guidelines had recommended urea reduction ratio of at least 65% for more than a decade,7,8 but hemoglobin of at least 11 g/dL (rather than 10 g/dL [100 g/L]) for only 3 years.9 One explanation for the discordance is that there had been considerably less time to raise hemoglobins to 11 g/dL than to raise urea reduction ratios to 65%. Increases in the number of patients with hemoglobin greater than or equal to 11 g/dL and urea reduction ratio greater than or equal to 65% over the past decade10 do suggest that dialysis facilities are learning quality management techniques. It is possible that units put all of their efforts into achieving one performance measure and not another, but differences in patient characteristics across units are also likely to play an important role. For instance, the sick patient with many comorbid illnesses and a small volume of urea distribution has no trouble achieving a urea reduction ratio of 65%, but his hemoglobin remains below 11 g/dL, despite large doses of erythropoietin and adequate iron stores. The large patient who refuses to lengthen his dialysis treatment, despite urging by the nephrologist and the staff, doesn’t quite achieve the dose target, but is not so uremic that he fails to achieve the hemoglobin target. It is well accepted that dialysis unit populations vary considerably by case-mix.11 In the average 60- to 70-patient unit on which the study reports, a shift of only a handful of patients from one group to another would be sufficient to alter the unit’s classification; we are not convinced that this change would represent signal rather than noise. In short, discordance in meeting hemoglobin and hemodialysis dose standards surprises us neither at the level of the patient nor of the facility.
Meyer and Miskulin
Should we care? Does it matter that purported indices of quality fail to correlate strongly? That depends on our beliefs about the relationship of these indices to the underlying construct of the quality of dialysis care. If we think that hemoglobin concentration and hemodialysis dose are themselves intrinsically important, that they are what is important about dialysis, then the dialysis facility’s ability to produce, respectively, a certain hemoglobin level and a certain urea reduction ratio would represent high-quality care. Shoppers perusing the Dialysis Facility Compare website might be invited to rank facilities by hemoglobin concentration or by dose, depending on their preference, just as they might select a digital camera. They would understand that they were buying grams of hemoglobin or fractional reduction in blood urea nitrogen, but should have no illusion that these were indices of any other facility quality or qualities. On the other hand, if hemoglobin concentration and dialysis dose are not only important in and of themselves, but also because we postulate them to be markers of the general quality of a facility’s operations, then their discordance is problematic. Recent studies have shown dialysis patient survival to increase as the individual patient meets goals for an increasing number of quality indicators.12-14 It is not clear whether this effect represents the cumulative benefit of meeting individual goals or reflects the general quality of care in the unit, or whether patients who meet more goals are the healthiest and destined to survive longer. We do care about discordance among putative quality measures: we think that dialysis facilities can and do produce more than hemoglobin and smallmolecule clearance. These indices that flow so easily from the vascular access to the autoanalyzer and hence to providers’, payors’, and regulators’ databases may fail to capture important qualities of dialysis care.15 Do we know what to measure? Avoiding transfusion is important, and dialysis patients randomized to higher hemoglobin targets report higher health-related quality of life than those randomized to lower targets.16-19 These findings lend the indicator criterion validity. However, among dialysis patients maintained transfusion-free, no survival benefit to higher hemoglobin values has been demonstrated.20 Indeed, among hemodialysis patients with cardiovascular disease, normal-
Editorial
izing hemoglobin may shorten survival.16 There is reason to fear that parenteral iron, commonly used to enhance erythropoiesis, may promote oxidative stress.21 In combination with economic incentives, the designation as a quality measure of a hemoglobin threshold, not a range, has been associated with a dramatic shift to the right in the distribution of dialysis patient erythropoietin doses, hemoglobin values, and bills. Adjustment of drug prescriptions and administration of drugs to promote erythropoiesis has become a major focus in dialysis units over the past 2 decades,22 and consumes many hours of experienced dialysis nurses’ time (P. Havermann, written communication, September 2007). Anemia management is a perfectly appropriate subject for clinical practice guidelines, but practice guidelines do not necessarily make good performance measures.23 The proportion of patients in whom hemoglobin is at least 11 g/dL lacks construct validity as a measure of the quality of dialysis care. What about hemodialysis dose—surely this is a reliable index of dialysis quality? A urea reduction ratio exceeding 65% is certainly reassuring, if one can be sure that the treatment measured was representative of delivered treatments and that the postdialysis urea nitrogen sample was obtained properly. Sloppy practice—drawing the sample at full flow—will only make the unit look better. Conversely, a unit that appropriately accommodates patients by measuring residual kidney function and adjusting delivered urea clearance accordingly will be penalized if the required urea reduction ratio turns out to be less than 65%. It is also worth remembering the Choices for Healthy Outcomes in Caring for ESRD (CHOICE) study finding that 11% of hemodialysis patients would not be willing to prolong their treatments by an hour even if that change would make them live twice as long.24 For most patients, the treatment prolongation required to achieve a urea reduction ratio of 65% is much more modest. However, particularly for very sick and elderly patients, even a brief extension of hemodialysis treatment may be very burdensome. People make tradeoffs, and what looks best on the facility’s report may not always be what is best for the individual patient. Spolter et al illustrate the difficulty and complexity of measuring the quality of dialy-
693
sis care in their observations that “facilities underperforming in URR [urea reduction ratio] or Hgb [hemoglobin] tended to be nonprofit, hospital based, and have more registered nurses per patient. The facilities that were deficient in anemia management also had more dialysis stations per patient.” There are several points here. First, patients dialyzed in hospital-based units are sicker than those at free-standing facilities.25 Second, multivariable analyses of epidemiologic data both from 1990-1993 and from 1995-2000 showed forprofit dialysis facilities to have higher mortality than nonprofit dialysis facilities.26,27 Third, more nursing care rather than less is generally presumed to be good for patients. These considerations suggest that the hypothetical Dialysis Facility Compare shopper who chose her facility on the basis of hemodialysis dose and hemoglobin might not be well served. Caveat emptor! The fourth point is lexical: “underperforming” and “deficient” are impressive and evocative words. It is hard to avoid the inference that patients ought not patronize underperforming and deficient facilities, that the organizations operating them must scramble to improve their performance, and that state surveyors should single them out for special scrutiny. Of course, such inference is arrant nonsense. The discrepancies which Spolter and colleagues identify remind us that dialysis treatment comprises both industrial processes, which must be controlled and improved, and comprehensive medical care for individuals. Some people receiving dialysis treatment are comparatively fortunate: kidney failure has complicated their lives only physically. Many others bear immense emotional and social burdens. For all, we worry about the unintended consequences of a construct of quality which is best satisfied by reviewing laboratory data and implementing protocols, perhaps never listening to patients or laying hands on them.28 Meaningful assessment of the quality of hemodialysis care would require careful adjustment for the characteristics of the patient population.29,30 Our own candidate list of important quality indicators would include dose, cannulation technique, infection control, and perhaps immunization rates. One should assess the effec-
694
Meyer and Miskulin
tiveness of patient education about treatment options and the facility’s attention to end-of-life care. One should measure the efficiency with which new problems are addressed. One should assess the coordination of care across different venues and providers, the facility’s response to measurements of health-related quality of life, and patient satisfaction with care. Measuring all this would not be simple, quick, or inexpensive. It would require instrument development, and data collection would entail not only chart review, but observation by experienced observers. Doing it right would make dialysis care even more expensive than it is now, and real measurement of real quality may not be worth the expense. Good stout homespun, real quality measures for real people, will be harder to weave and will actually cost far more than the emperor’s fantasy robes. Unless and until we make that investment, we must refer the concerned consumer to the little child, the empiricist, and abandon our awkward, dignified pretense. Klemens B. Meyer, MD Dana C. Miskulin, MD, MS Tufts-New England Medical Center Boston, Massachusetts ACKNOWLEDGEMENTS Support: None. Financial Disclosure: Drs Meyer and Miskulin receive salary support from Dialysis Clinic Inc and grant support from the National Institutes of Health.
REFERENCES 1. Andersen HC: Andersen’s Fairy Tales. Available at: http:// www.gutenberg.org. Accessed September 18, 2007 2. Frederick PR, Maxey NL, Clauser SB, Sugarman JR: Developing dialysis facility-specific performance measures for public reporting. Health Care Financ Rev 23:37-50, 2002 3. McClellan WM, Frederick PR, Helgerson SD, Hayes RP, Ballard DJ, McMullan M: A data-driven approach to improving the care of in-center hemodialysis patients. Health Care Financing Rev 16:129-140, 1995 4. Fink JC, Blahut SA, Briglia AE, Gardner JF, Light PD: Effect of center-versus patient-specific factors on variations in dialysis adequacy. J Am Soc Nephrol 12:164-169, 2001 5. Fink JC, Hsu VD, Zhan M, et al: Center effects in anemia management of dialysis patients. J Am Soc Nephrol 18:646-653, 2007 6. Spolter YS, Seliger SL, Zhan M, Hsu VD, Walker LD, Fink JC: The relationship between dialysis performance measures: adequacy and anemia management. Am J Kidney Dis 50:774-781, 2007
7. National Institutes of Health: Morbidity and mortality of dialysis. NIH Consensus Statement Online, November 1-3, 1993, 11:1-33 8. Renal Physicians Association: Clinical practice guideline on adequacy of hemodialysis. Clinical Policy, 1993 9. National Kidney Foundation: K/DOQI Clinical Practice Guidelines for Anemia of Chronic Kidney Disease: update 2000. Am J Kidney Dis 37:S182-238, 2001 (suppl 1) 10. Centers for Medicare & Medicaid Services: 2006 Annual Report: ESRD Clinical Performance Measures Project. Baltimore, MD, Centers for Medicare & Medicaid Services, Office of Clinical Standards & Quality, 2006 11. McClellan WM, Soucie J, Flanders W: Mortality in end-stage renal disease is associated with facility-to-facility differences in adequacy of hemodialysis. J Am Soc Nephrol 9:1940-1947, 1998 12. Rocco MV, Frankenfield DL, Hopson SD, McClellan WM: Relationship between clinical performance measures and outcomes among patients receiving long-term hemodialysis. Ann Intern Med 145:512-519, 2006 13. Tentori F, Hunt WC, Rohrscheib M, et al: Which targets in clinical practice guidelines are associated with improved survival in a large dialysis organization? J Am Soc Nephrol 18:2377-2384, 2007 14. Wolfe RA, Hulbert-Shearon TE, Ashby VB, Mahadevan S, Port FK: Improvements in dialysis patient mortality are associated with improvements in urea reduction ratio and hematocrit, 1999 to 2002. Am J Kidney Dis 45:127-135, 2005 15. Meyer KB, Kassirer JP: Squeezing more cost and care out of dialysis: our patients would pay the price. Am J Med 112:232-234, 2002 16. Besarab A, Bolton WK, Browne JK, et al: The effects of normal as compared with low hematocrit values in patients with cardiac disease who are receiving hemodialysis and epoetin. N Engl J Med 339:584-590, 1998 17. Beusterien KM, Nissenson AR, Port FK, Kelly M, Steinwald B, Ware JE, Jr.: The effects of recombinant human erythropoietin on functional health and well-being in chronic dialysis patients. J Am Soc Nephrol 7:763-773, 1996 18. Furuland H, Linde T, Ahlmen J, Christensson A, Strombom U, Danielson BG: A randomized controlled trial of haemoglobin normalization with epoetin alfa in predialysis and dialysis patients. Nephrol Dial Transplant 18: 353-361, 2003 19. Jones M, Ibels L, Schenkel B, Zagari M: Impact of epoetin alfa on clinical end points in patients with chronic renal failure: a meta-analysis. Kidney Int 65:757-767, 2004 20. National Kidney Foundation: KDOQI Clinical Practice Guidelines and Clinical Practice Recommendations for Anemia in Chronic Kidney Disease. Am J Kidney Dis 47:S11-145, 2006 (suppl 3) 21. Brewster UC: Intravenous iron therapy in end-stage renal disease. Semin Dial 19:285-290, 2006 22. Krishnan M, Adams E: Operationalizing anemia management: organizing the program and coordinating the team. Nephrol Nurs J 30:567-570, 2003 23. Walter LC, Davidowitz NP, Heineken PA, Covinsky KE: Pitfalls of converting practice guidelines into quality measures: lessons learned from a VA performance measure. JAMA 291:2466-2470, 2004
Editorial 24. Bass EB, Wills S, Fink NE et al: How strong are patients’ preferences in choices between dialysis modalities and doses? Am J Kidney Dis 44:695-705, 2004 25. Sankarasubbaiyan S, Holley JL: An analysis of the increased demands placed on dialysis health care team members by functionally dependent hemodialysis patients. Am J Kidney Dis 35:1061-1067, 2000 26. Garg PP, Frick KD, Diener-West M, Powe NR: Effect of the ownership of dialysis facilities on patients’ survival and referral for transplantation. N Engl J Med 341:1653-1660, 1999 27. Szczech LA, Klassen PS, Chua B, et al: Associations between CMS’s Clinical Performance Measures project
695 benchmarks, profit structure, and mortality in dialysis units. Kidney Int 69:2094-2100, 2006 28. Casalino LP: The unintended consequences of measuring quality on the quality of medical care. N Engl J Med 341:1147-1150, 1999 29. Hayward RA: Performance measurement in search of a path. N Engl J Med 356:951-953, 2007 30. Miskulin DC, Meyer KB, Athienites NV, et al: Comorbidity and other factors associated with modality selection in incident dialysis patients: the CHOICE Study. Choices for Healthy Outcomes in Caring for End-Stage Renal Disease. Am J Kidney Dis 39:324-336, 2002