In the wake of transformative changes to the dialysis reimbursement scheme and recommendations for clinical care, hemoglobin levels and epoetin doses have decreased accordingly. These early findings were reported by the Dialysis Outcomes and Practice Patterns Study (DOPPS) Practice Monitor in the American Journal of Kidney Diseases (AJKD; 2012;59:309-312) and in updated form on the DOPPS website (www.dopps.org/DPM).
“When you substantially restructure an incentive system, you do get changes, particularly with the key element that was targeted for payment reform; in this case, it would be anemia management,” said Daniel E. Weiner, MD, MS, Assistant Professor of Medicine at Tufts University School of Medicine and a nephrologist at Tufts Medical Center, in a phone interview. Dr. Weiner is coauthor of a commentary on the DOPPS observations that was published in the same issue of AJKD (312-314).
“I think the other factor is that it's very, very difficult to disentangle the changes you see here from changes that may be in response to clinical data that came out and changes that are in response to recent FDA [Food and Drug Administration] advice,” he said.
This is because several recent developments have addressed anemia management. For one thing, the End-Stage Renal Disease (ESRD) Prospective Payment System (PPS), which went into effect Jan. 1, 2011, incorporated into a bundled payment aspects of dialysis care that previously were billed separately, such as erythropoiesis-stimulating agents (ESAs), taking these drugs from a profit center into a cost center.
Then in June 2011, the FDA moved to more conservative dosing guidelines for ESA use, responding to clinical trial data that revealed increased risks for death and cardiovascular events with the pursuit of higher hemoglobin targets. Instead of recommending a hemoglobin target range of 10 g/dL to 12 g/dL in patients on dialysis, the agency advised that ESA therapy be initiated at a hemoglobin level less than 10 g/dL and reduced or interrupted when hemoglobin approaches or exceeds 11 g/dL.
The label change led to an adjustment of the Centers for Medicare & Medicaid Services ESRD Quality Incentive Program (QIP). While the first year of the QIP exacts payment penalties for average hemoglobin levels above 12 g/dL or below 10 g/dL, the lower hemoglobin measure will be removed for the program's second payment year, 2013.
“What's happened in this field, at least as far as anemia management, is the current clinical data and financial considerations are all pushing in the same direction—i.e., less erythropoietin [EPO] usage,” said Chaim Charytan, MD, Chief of the Renal Division and Director of Dialysis Units at New York Hospital Medical Center of Queens, and Clinical Professor of Medicine at Weill Cornell Medical College, commenting on the DOPPS Practice Monitor (DPM) findings in a phone interview.
Substantial Change with Revised Label
The DPM is administered by the nonprofit Arbor Research Collaborative for Health of Ann Arbor, MI, and supported by scientific research grants from Abbott, Amgen, Baxter, Genzyme, Kyowa Hakko Kirin, and Vifor Fresenius Medical Care Renal Pharma.
The data on the DPM website come from about 140 US hemodialysis units that are randomly selected and representative of hemodialysis practice in the country.
“In anemia it's very interesting that while there was modest decrease in hemoglobin as the bundle was implemented in January 2011, we've actually seen more substantial change in the few months since the FDA ESA label was revised, which was in June of 2011,” said Bruce M. Robinson, MD, MS, principal investigator of the DOPPS and Vice President for Clinical Research at Arbor Research Collaborative for Health, in a phone interview.
“In the two months after that —July and August—we saw quite a substantial drop in hemoglobin and also in EPO dose. EPO doses have fallen most in patients who were receiving quite high doses.”
Mean hemoglobin level declined by 0.12 g/dL between August 2010 and July 2011 and subsequently fell another 0.10 g/dL in August 2011, landing at 11.26 g/dL. The percentage of patients with hemoglobin levels greater than 12 g/dL went down from 31.4% to 28.0% through June 2011, dropping even further to 22.6% in August 2011. The proportion of patients with levels less than 10 g/dL, on the other hand, rose from 8.5% to 10.0%.
“One of the things that was actually really reassuring is that you're not seeing a significant rise in the number of people with very low hemoglobins based on these data,” Dr. Weiner said. The percentage of patients with hemoglobin levels less than 9 g/dL remained stable at 2.8% to 2.9%.
The mean prescribed epoetin dose went down by 15% between August 2010 and August 2011, with the greatest decrease seen over the last three months of that time period. The percentage of patients prescribed epoetin doses exceeding 50,000 units/wk dropped from 11.6% to 6.5% between August 2010 and August 2011.
“For right now the anemia changes seem to be clinically appropriate,” Dr. Weiner said. “Regardless of the financial incentives, the safety signals show that people who are getting very high doses of erythropoietin to try to achieve moderate-to-high hemoglobin levels appear to do worse, and the major change seems to be occurring with those people, so that seems to be very appropriate.
“Unfortunately, there's no data to be able to tell if there are more transfusions or other adverse effects that can be seen with this. DPM has no way to capture that, and, as best we can tell, we're not going to be able to know that at least for the foreseeable future.”
Increases in iron administration and serum ferritin levels also were observed by the DPM. The percentage of patients prescribed intravenous (IV) iron over one month climbed from 56.9% to 78.1% between August 2010 and June 2011, dropping to 71.0% in August 2011. The median serum ferritin level rose from 555 ng/mL in August 2010, to 598 ng/mL in April 2011, to 647 ng/mL in August 2011.
“If you use more iron, you can use less erythropoietin, so I think around the time that the bundle came into play, iron use went up, and again this is something else that may have happened without substantial payment reform, as there's been a real push from some circles to use more iron rather than using higher doses of erythropoietin to deal with erythropoietin resistance,” Dr. Weiner said.
“I don't think anybody knows what the best strategy is here. There's some data from the DRIVE [Dialysis Patients' Response to IV Iron with Elevated Ferritin] study and the DRIVE-II study that this is probably a safe strategy, but those studies were very small and relatively short-term, so it's difficult to draw long-term conclusions.”
Rise in PTH
While no clear changes in serum phosphorus or calcium have been observed to date, the mean parathyroid hormone (PTH) level went up 26% between August 2010 and April 2011, from 339 pg/mL to 427 pg/mL, and median PTH increased 29% during the same time period, from 246 pg/mL to 318 pg/mL. Between April 2011 and August 2011, PTH levels stayed steady or decreased slightly.
“I think there are two things going on here,” said Jay Wish, MD, Professor of Medicine at Case Western Reserve University School of Medicine and Medical Director of the Dialysis Program at University Hospitals Case Medical Center, commenting on the DPM findings in a phone interview.
“Number one is that bundling is going to constrain active vitamin D use in dialysis facilities because it's part of the bundle—it's a cost center now as opposed to being a profit center.”
The other factor is the adoption of the Kidney Disease: Improving Global Outcomes (KDIGO) Clinical Practice Guidelines for the Diagnosis, Evaluation, Prevention, and Treatment of Chronic Kidney Disease–Mineral and Bone Disorder, which suggested maintaining intact PTH levels in the range of approximately two to nine times the upper limit of normal for the assay, a departure from the previous National Kidney Foundation Kidney Disease Outcomes Quality Initiative (KDOQI) target PTH range of 150 pg/mL to 300 pg/mL.
“I think a lot of dialysis facilities were already shifting away from the KDOQI target range of 150-300 and were starting to adopt the KDIGO target range of 100-500,” Dr. Wish said. “I think if the KDIGO hadn't superseded the KDOQI and everybody was still adhering to that 150-300 target PTH range, we probably wouldn't have seen the same rise in PTHs that we're seeing.”
Large vs Small Facilities
The DOPPS Practice Monitor examined these trends by dialysis facility size—large- and medium-chain versus small-chain and independent facilities.
“The trends in hemoglobin through August were really quite similar in the larger and small DO [dialysis organization] facilities,” Dr. Robinson said.
“EPO doses tend to be higher in the larger DOs, which is consistent with prior publications over time, and, secondly, there's been a clearer trend towards lowering those doses in the larger DOs than in the small organizations. Doses started higher in the larger DOs, and they've actually fallen more than in the small DOs, although on the website you can see that despite that fall they're still higher than in the smaller DOs.”
The mean prescribed epoetin dose has stayed about 35% lower in small-chain and independent facilities versus large- and medium-chain facilities. The percentage of patients prescribed epoetin doses exceeding 50,000 units/wk decreased from 12.9% to 6.7% in large- and medium-chain facilities and remained stable at about 2% in small-chain and independent facilities.
Trends in black and nonblack patients also were analyzed.
“The GAO [Government Accountability Office] suggested that there is a need to monitor in particular trends in care in higher-cost populations, and they highlighted black patients as well as patients with Medicaid or dual coverage, and they highlighted those populations because they have higher costs due to generally higher EPO requirements,” Dr. Robinson said.
“At a high level we've not seen substantial differences by race. That's a reassuring point.”
Also reassuring is that there were no changes observed in single-pool Kt/V, dialysis session length, serum albumin levels, and use of cardiovascular medications.
“Many practices have remained stable despite a dramatic change in reimbursement for services,” Dr. Robinson said. “I think that is very good news for the community.”
Rise in Hospitalizations
In terms of clinical outcomes, preliminary data showed an increase in 30-day annualized hospitalization rate of about 1.8% per month between August 2010 and August 2011.
“I think it's too soon to be able to interpret what that means and whether or not that's going to be the trend that continues,” Dr. Wish said. “If we could say that that hospitalization rate is due to increased requirement for transfusions, then obviously we could say, aha, it's the lowering of the hemoglobins and the more conservative ESA use, but not knowing what the diagnoses are that are leading to that increased hospitalization rate, it's hard to say.
“I think it'll be very, very important for researchers to start gathering the data in terms of discharge diagnoses to figure out what is driving that and to see whether or not that was in fact due to a change in the way patients were treated under bundling or whether there was a secular trend already occurring even before bundling began.”
‘Vital to Follow Up’
At this early stage after the implementation of the bundle, the QIP, and the updated ESA label, it is critical to continue to follow these trends.
“In my opinion, most of what we're seeing here are the things I would have anticipated would happen as a result of the economic forces and the quality incentive payment that goes with bundling,” Dr. Wish said.
“The question is not did it happen, but to what extent did it happen, how great were the changes, and are those changes compromising patient well-being, and at this point I can't say that there's anything here that compromises patient well-being, with the exception of, what is the consequence of the low hemoglobin levels on transfusions?”
Dr. Robinson also noted the importance of longer-term data.
“I think it's going to be vital to follow up and see what happens in the coming months,” he said. “We have data from just two months after the label change, so it will be of great interest to find out what's gone on in subsequent months.
“In particular, we will update our DPM in April with data through December of 2011, so from my standpoint it will be of great importance to know whether hemoglobin levels are falling further and, in particular, whether the proportion of patients with very low levels is changing at all.”
Dr. Weiner praised the role of the DOPPS Practice Monitor in providing such information.
“I think the DPM is a really impressive initiative and, as of right now, it's far and away the best source for information as to what the nationwide effects are of what's a really significant payment reform system,” he said.
“I think it's really incumbent on the government to formalize other systems of monitoring so that we can have a better idea of things like transfusion requirements and other effects that aren't necessarily going to be captured within data that's just acquired through dialysis units. These are things for the future.”