Aligning incentives to improve patient outcomes

Brian Rifkin

Care of patients with chronic kidney disease (CKD) and end-stage kidney disease (ESKD) is complex, involving many decisions that are not exclusively based upon patient outcomes, writes Brian Rifkin (Hattiesburg Clinic, Hattiesburg, USA)  

Doctors are human (OK, Captain Obvious). We are subject to the same psychological pressures that drive people’s worst impulses. Did you think physicians were above biases and outside influences? That we are somehow “superhuman”, and always do what is in our patient’s best interest? 

There are examples in clinical decision making that make our faults quite obvious. Many of them are associated with a shift in clinical care directly related to payment models. In an ideal world, incentives (including monetary payments) would align both patient and physician interests. Let me give you an example from the not-too-distant past in the treatment of anaemia in patients on dialysis. 

Epoetin came to market in 1989, and cut transfusion rates by 50% one year after its introduction. I have been a private practice nephrologist in the USA since the early 2000s: back in those days, erythropoietin stimulating agents (ESAs) were already ubiquitous, and were a separately billable item (more ESA use meant larger payments). We knew from the 1998 Normal Hematocrit Trial1, that normalisation of haematocrits in dialysis patients was associated with worse cardiovascular outcomes. Additional studies2 looked at increased haemoglobins over 13g/dL in dialysis patients, though none of these were able to show improved mortality or cardiovascular outcomes.  

Unfortunately, the available evidence did little to slow the increasing doses and increasing average haemoglobins in dialysis patients. Then, in 2006, the Kidney Disease Outcomes Quality Initiative (KDOQI) updated their guidelines on anaemia. These guidelines were underwritten by Amgen, the manufacturer of epoetin alfa. The guidelines, in the face of the above data, actually increased the target haemoglobin from 11–12 to 11–13g/dL. The most amazing thing about the detour to haemoglobins of 13 and beyond is that we knew everything we needed to know from 1998’s Normal Hematocrit Trial. Much of the enthusiasm for high haematocrits was finally derailed by CHOIR, CREATE, and then TREAT3 (in patients with CKD not on dialysis).  

These studies convinced the most important people of all, the government payers for ESAs, to finally follow the data. On 1 January, 2011, the dialysis “bundle” was initiated. It brought about a new frugality and cost-consciousness in the use of medications and other resources in the management of patients with ESKD. Dialysis providers were challenged to control costs, while providing quality care.4 As a result of the bundle, medicines suddenly became a cost centre versus a profit centre, and, as in many other practices, our use of ESAs plummeted by 40–45% in the following months. Finally, in February 2016, Glenn Chertow and colleagues5 looked at the rates of cardiovascular outcomes and found that, in 2011 and 2012 (the years with the biggest fall in average haemoglobins), heart failure, stroke, and venous thrombosis were all lower than predicted. Taking our foot off the ESA accelerator appeared to be benefitting patients. The decreased use of ESAs ultimately occurred well after the clinical evidence suggested it should, and much closer, rather, to the change in reimbursement. Would this frugality in ESA use have occurred without the changes in payment incentives? 

Which brings us to the subject of value-based care (VBC) models in nephrology. This is the latest iteration of incentives, or penalties, that attempt to align physician payments with patient care. Many accountable care organisations (ACOs) have been using VBC guidelines for shared savings bonus payments. Our ACO monitors VBC compliance by examining the rates of recommended procedures completed for common disorders. This has led to more cancer screenings (cancer prevention), diabetic eye screenings (blindness prevention) and improved use of statins and beta-blockers in patients with cardiac disease (cardiovascular death prevention). This in turn helps to maintain “per patient” budgets at or below estimated costs by early detection or slowing chronic disease progression.  

Most physicians would agree that prevention is preferable to treatment in most diseases. So, why should nephrology care about VBC models? Despite our “best efforts”, currently one in seven people in the USA have CKD, and 40% of individuals nearing dialysis are not aware of their diagnosis of CKD. More than 50% of patients “crash” into dialysis without prior CKD care. While the annual cost of CKD for Medicare patients is US$90 billion (for 37 million patients), the cost of ESKD is US$37 billion (for roughly 500,000 patients).  

Most physicians would also agree the real focus of nephrology should include slowing CKD stage 4/5 progression, patient education, optimising dialysis initiation and decreasing the total cost of care. Unfortunately, we USA-based private practice nephrologists have been functioning under a very perverse reimbursement system. Currently, there are more financial incentives for initiation of dialysis rather than for prolonging the course of CKD prior to initiation of dialysis. In addition, in contrast to the highly structured processes in place for dialysis care, there are few organised, well-funded programmes that target preventing or slowing progression of kidney disease among individuals with CKD. Nor are there such programmes promoting a smooth transition to kidney replacement therapy if necessary and desired by the patient, or encouraging kidney transplantation and home dialysis. So: would it not make more sense, given our lizard brain’s need for dopamine, to flip reimbursement on its head and incentivise pre-dialysis CKD care? Well, change is hard. 

There are still many obstacles to be negotiated before VBC models perform as intended. Slowing dialysis initiation is a clinical and economical goal, but the incentives for pre-dialysis care have not been well established, and its development in some cases may be unfavourable for providers. Incentives and protocols matter, as was shown with ESAs, as only about 40% of patients with CKD are on an angiotensin-converting-enzyme (ACE) inhibitor/angiotensin II receptor blockers (ARBs), a standard of CKD care.6 Thus, there is a need for direct VBC economic incentives to physicians for initiating evidence-based CKD treatments.  

The medical industry is moving toward VBC and, if doctors do not get involved, someone else will. Currently, population health programmes, such as Monogram Health, receive payments to decrease patient expenditures through at-home patient education. Unfortunately, there is sometimes poor alignment with the patient’s primary nephrologist, and these organisations often have a narrow focus based solely upon economic factors. There will have to be better coordination on behalf of the patient for education, treatment initiation and dialysis choices. Finally, success in VBC requires physician engagement and significant data reporting in order to get to incentives, both of which may carry significant upfront cost. In addition, most VBC programmes will also ask physicians to take on risk to receive performance incentives, making financial benefits murkier. 

VBC models have many moving parts, and input from multiple sources. Policy makers must consider performance-based adjustments based upon socioeconomic disparities, attempt to eliminate barriers to education services like co-pays, and determine ways to make VBC models viable for all size practices and population distributions. Close co-operation between policymakers, economists and nephrologists is needed to ensure a high quality of pre-dialysis care that has a meaningful benefit to patients. Perhaps we can learn from the mistakes of the past and move more quickly to programmes that benefit physicians for making better choices for our patients. VBC models are a tool that will require some further adjustments but will, in the future, hopefully align physician incentives with patient care goals.


1.       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. 1998 Aug 27;339(9):584-90. doi: 10.1056/NEJM199808273390903. PMID: 9718377. 

2.       Furuland H, Linde T, Ahlmén J, et al. A randomized controlled trial of haemoglobin normalization with epoetin alfa in pre-dialysis and dialysis patients. Nephrol Dial Transplant. 2003 Feb;18(2):353-61. doi: 10.1093/ndt/18.2.353. PMID: 12543892. 

3.       Landmark Trials in Anemia & Kidney Disease – Renal Fellow Network,, downloaded 4/2/23. 

4.       Charnow JA. How “Bundling” Changed Dialysis Care. Renal and Urology News. 3/2/17. 

5.       Chertow GM, Liu J, Monda KL, et al. Epoetin Alfa and Outcomes in Dialysis amid Regulatory and Payment Reform. J Am Soc Nephrol. 2016 Oct;27(10):3129-3138. doi: 10.1681/ASN.2015111232. Epub 2016 Feb 25. PMID: 26917691; PMCID: PMC5042674. 

6.       Murphy DP, Drawz PE, Foley RN. Trends in Angiotensin-Converting Enzyme Inhibitor and Angiotensin II Receptor Blocker Use among Those with Impaired Kidney Function in the United States. J Am Soc Nephrol. 2019 Jul;30(7):1314-1321. doi: 10.1681/ASN.2018100971. Epub 2019 Jun 5. PMID: 31167823; PMCID: PMC6622408.


Please enter your comment!
Please enter your name here