Patient Management: Practice guidelines and how they can impact patient care8/11/2009 1:02:07 PM
The evolution of nephrology guidelines: Calling for a mid-course adjustment
Editor's Note
The days are long gone-and, for the most part, thankfully, for patients-when nephrologists and the dialysis care team experimented daily with protocols, hardware, and treatment regimens in a search for adequate dialysis. Today, the dialysis prescription is one that can be fine-tuned based on patient needs, and several modes of delivery-peritoneal dialysis, home hemodialysis, in-center care-are available. Yet caregivers still need guidance, and practice guidelines have helped to fill that void. Starting with the Dialysis Outcome Quality Initiative in 1997, guidelines cover a number of clinical indications and best practices in renal care. But are they being developed and used properly-or has their purpose been misinterpreted? Can they stay fresh by keeping up with new research and changes in treatment approaches?
These two viewpoints look at the current direction in to developing practice guidelines in renal care, focusing in particular on the GRADE system being used by the international Kidney Disease Improving Global Outcomes (KDIGO) guideline development group. Later this month, Kidney International will publish KDIGO's latest practice guideline for the diagnosis, evaluation, prevention, and treatment of mineral and bone disorder among chronic kidney disease patients using the GRADE system.
There is no doubt that in the 12 years since practice guidelines were introduced to nephrology, many improvements in quality of care have occurred. This is evident in analysis of surrogate quality indicators, but much less so in terms of mortality rates.1-3 At least in part, guidelines have been given credit for this progress.
Notwithstanding this, dialysis remains an imperfect technology, with compromised survival and reduced quality of life for patients. Current common modalities and drug treatments are suboptimal, there is too little high quality, pivotal research, and innovation is far too slow. In this context, recent trends in guideline development as specifically applied to nephrology are worrisome. In particular, the "cutting and pasting" of guideline creation methodology copied from other disciplines who produce much more clinical research, is problematic.
This article examines some of the key guideline milestones, and concludes with a call to adjust how nephrology goes about creating them in the future.
The past
For nephrology, the journey into the guideline development game began in 1997.4,5 It had been recognized that end-stage renal disease patients were dying prematurely, and that there was considerable variation in patient outcomes across both centers and countries. The National Kidney Foundation (NKF) sponsored the Dialysis Outcomes Quality Initiative (DOQI), whose goals were to improve patient survival, reduce patient morbidity, increase efficiency of care, and improve quality of life with the development of practice guidelines. Four areas were covered-anemia management, peritoneal dialysis adequacy, hemodialysis adequacy, and vascular access. Of course, the nephrology community broadly supported these ideals. Later, the Renal Physicians Association developed three separate practice guidelines on adequacy of hemodialysis, shared decision mak-ing in initiation of and withdrawal from dialysis, and on appropriate patient preparation for renal replacement therapy.
In 2003, the NKF (now through its Kidney Disease Outcome Quality Initiative, or KDOQI) published guidelines on bone and mineral metabolism.6 The document contained 111 recommendations (including specific treatment target ranges), but the guideline work group acknowledged that only about one-third of the guidelines were evidence-based. A critical reaction ensued, and concerns were raised that the guidelines were unduly influenced by industry. This catalyzed a profound shift in how more recent guidelines are developed in nephrology.
This shift was first manifested in the 2006 NKF-KDOQI anemia clinical practice guidelines and clinical practice recommendations, an update from the original 1997 guidelines.7 The term guideline was now to be reserved for when the evidence was robust enough. In contrast, a recommendation was based on expert opinion, when sufficient hard data to make them guidelines was lacking. Despite the fact that anemia management has more randomized controlled clinical trials (RCT) than any other topic within nephrology, there were only three evidence-based guidelines, but 33 clinical practice recommendations. Still, this approach offered clinicians broad guidance on anemia management, based on the available research.
The present
Internationally, it was recognized that while jurisdictions that create guidelines examine the same evidence base, discordant timing or methods were causing significant differences to coexist. Consequently, an international body, Kidney Disease Improving Global Outcomes (KDIGO) was created, with the intention of harmonizing guidelines.8
KDIGO took a stricter approach than KDOQI to applying the methodology of evidence-based medicine, and endorsed the GRADE rules, whereby randomized clinical trials were considered high quality and preferable for guideline development, but observational and basic science data usually would not be.9 This interaction of evidence-based medicine, guidelines, and cost effectiveness analysis that currently dominate nephrology are depicted in Figure 1.
The extreme application of the GRADE rules is apparent in the KDIGO draft guidelines for Chronic Kidney Disease Mineral and Bone Disorder (CKD MBD), released in 2008. Methodology included examining only RCTs with N > 50 and duration > 6 months. Although observational data might have been considered, none was assessed to be statistically strong enough to allow for inclusion.
Based on the limited RCT data considered adequate for consideration, drafting these guidelines using the GRADE system approach meant there were no specific recommendations, no targets for treatment, and no advice concerning drug selection. As part of the public review process, KDIGO guideline committee members reviewed a significant number of comments on the draft, including from this author, and preliminary reports indicate that small changes have been made to the final guidelines that improve its utility. But the strict rules of the GRADE approach, which may be the standard for future guidelines development, has some possible negative consequences. Let it be stressed that while many of the consequences are no doubt unintended by the evidence-based medicine approach, they are not unforeseeable.
1) Vague guidelines will cause increased variability of care, and worse performance relative to previous guideline targets derived from observational trials.
2) Justified by vague guidelines, expensive but promising new drugs (for example in mineral metabolism: sevelamer, lanthanum, and cinacalcet) that are now covered by many payers and insurers all over the world, may be removed from formularies. This may discourage applications for newly developed but more expensive medications and treatments (for example, daily home hemodialysis), now and in the future. Strict cost effectiveness considerations built into the guideline process allows payers and insurers to shift the blame for limited access onto the providers who create and endorse the guidelines.
3) Industry will look at the GRADE system as a sign that nephrology is an increasingly difficult discipline, and will choose to do research and development in easier markets. Less renal research, (not more, as the KDIGO authors intend) would be the result.
4) Local or national guidelines may attempt to counter the philosophy embodied by KDIGO, but KDIGO will trump local efforts.
So, how and why have we lost our way?10 I see three explanations.
First, nephrology has become dominated by a fanatical adherence to the rules of strictly applied EBM. In order to avoid any suggestion of industry influence, we have taken a pure but rather nave path. In the United States, there are legitimate concerns about payers using practice guidelines to tie outcomes with pay for performance measures. If guidelines become a stepping-off point for formulating these measures, then EBM plays a bigger role. But to withhold results from observational studies that can help shape our clinical approach to dialysis care is slighting the patients we serve. As a medical community, nephrology needs to recognize that the goals of a guideline process are broader than the short-term creation of a scholarly and scientific document. Guidelines must improve the quality of care and reduce variability in treatment approaches, while simultaneously encouraging research and innovation so that in 10 or 20 years, care for the renal patient will be markedly better than it is now. The arena we are playing in is not a purely scientific one, as we might like it to be. It is also a social, political, and economic arena that requires thoughtful and proactive short-term, medium-term, and long-term strategies.
Second, nephrology must accept the rather severe limitations of data that exist in our discipline. Few RCTs are done,11 and even when they are done, the results are not always positive,12, 13 and sometimes can be ambiguous.14, 15 While observational studies are often not as reliable a design as RCTs, the fact is that nephrology has an abundance of them. The academic literature suggests that well-done observational studies are usually valid, and that even several RCTs on the same subject may yield conflicting results.16, 17 To focus too narrowly on RCTs alone will produce scholarly but vague documents that fail to guide clinical decision-making.
Third, nephrology must adjust its approach to guideline development in order to advance the discipline for current and future generations of patients. This adjustment means stepping back to a process and output similar to what was used to create the 2006 KDOQI anemia guidelines. When considering the data, it is critical that after agreeing to a weighting scheme based on the hierarchy of study design and validity of execution, that all available data be considered. This must include RCTs, observational studies, and basic science. Next, we should create guidelines when the evidence is sufficient, and recommendations when the evidence is not. A vague guideline that does not provide genuine guidance for clinicians is not helpful in the short term and may be harmful. If the evidence and opinion is too weak to even apply the term "clinical practice recommendation," then call them position statements, or consensus statements. These must include specific targets that empower continuous quality improvement initiatives. Finally, professional societies that develop guidelines must consider their advocacy duty first and foremost, and should specifically not consider cost effectiveness. An adjusted approach to data, targets, and advocacy is recommended as shown in Figure 2.
The EBM fanatics or purists may consider this pragmatic adjustment to be flawed and to risk errors that may result in harm.18 This may be true occasionally, but vague recommendations that provide no guidance will cause harm too. If one accepts the evolutionary analogy that runs through this commentary, then recall that species evolve by both positive and negative selection. Inherently, medical knowledge evolves in much the same iterative way.
What ought to unite both pragmatists and purists is the understanding that to improve CKD and ESRD care over a 10 or 20 year timeline will require a higher quantity and quality of clinical research, and especially RCTs, than we are able to produce in 2009. While it would be ideal if the majority of funding for this effort came from peer reviewed sources, this seems unlikely. Therefore, industry support for research will remain essential. Is nephrology better served by scholarly but vague guidelines that marginalize our industry allies, or would a more liberal, pragmatic, and ultimately more industry friendly approach lead us further and faster? Can we achieve earlier access to promising new drugs and other therapies, and still acquire the evidence required to fully assess their benefits and risks? These fundamental questions deserve discussion at this crossroads in the evolution of the discipline of nephrology.
References
1. U.S. Renal Data System, USRDS 2007 Annual Data Report: Atlas of Chronic Kidney Disease and End-Stage Renal Disease in the United States, Bethesda, Md., National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases, 2007
2. Centers for Medicare & Medicaid Services, 2006 Annual Report: End Stage Renal Disease Clinical Performance Measures Project, Baltimore, Md., Department of Health and Human Services, Office of Clinical Standards & Quality, 2007
3. Tentori F, et al. Mortality risk for dialysis patients with different levels of serum calcium, phosphorus, and PTH: The Dialysis Outcomes and Practice Patterns Study (DOPPS). Am J Kidney Dis 2008; 52: 519-530
4. Levin N, Eknoyan G, Pipp M, Steinberg E. National Kidney Foundation: Dialysis Outcome Quality Initiative-development of methodology for clinical practice guidelines. Nephrology, Dialysis, Transplantation 1997; 12: 2060-2063
5. NKF-DOQI clinical practice guidelines for the treatment of anemia of chronic renal failure. National Kidney Foundation-Dialysis Outcomes Quality Initiative. AmJKidney Dis 1997; 30: S192-S240.
6. K/DOQI clinical practice guidelines for bone metabolism and disease in chronic kidney disease. Am J Kidney Dis 2003; 42: S1-201
7. KDOQI Clinical Practice Guidelines and Clinical Practice Recommendations for Anemia in Chronic Kidney Disease. Am J Kidney Dis 2006; 47: S11-145
8. Neumann ME. KDIGO takes clinical practice guidelines to a global level. Nephrol News Issues 2006; 20: 40-41, 43, 47
9. Uhlig K, et al. Grading evidence and recommendations for clinical practice guidelines in nephrology. A position statement from Kidney Disease: Improving Global Outcomes (KDIGO). Kidney Int 2006; 70: 2058-2065
10. Mendelssohn DC. Musings on guidelines and evidence: a pragmatic and nephrocentric view. Perit Dial Int 2007; 27: 31-34
11. Strippoli GF, Craig JC, Schena FP. The number, quality, and coverage of randomized controlled trials in nephrology. Journal of the American Society of Nephrology 2004; 15: 411-419
12. Besarab A, 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 1998; 339: 584-590
13. Singh AK, et al. Correction of anemia with epoetin alfa in chronic kidney disease. N Eng J Med 2006; 355: 2085-2098
14. Suki WN, et al. Effects of sevelamer and calcium-based phosphate binders on mortality in hemodialysis patients. Kidney Int 2007; 72: 1130-1137
15. Wanner C, Krane V, Marz W, Olschewski M, et al. Atorvastatin in patients with type 2 diabetes mellitus undergoing hemodialysis. N Engl J Med 2005; 353: 238-248
16. Concato J, Shah N, Horwitz RI. Randomized, controlled trials, observational studies, and the hierarchy of research designs. N Engl J Med 2000; 342: 1887-1892
17. Benson K, Hartz AJ. A comparison of observational studies and randomized, controlled trials. N Engl J Med 2000; 342: 1878-1886
18. Suri RS. Musings on guidelines and evidence: an opposing view. Perit Dial Int 2007; 27: 35-38; discussion 38-41
Dr. Mendelssohn is head of the division of nephrology and medical director of dialysis at Humber River Regional Hospital in Ontario, Canada. He is also an NN&I Editorial Advisory Board member.
http://www.nephronline.com/features.asp?F_ID=458
I Hate Dialysis Message Board
Author
