Home » Health »

Assessing key cost drivers associated with caring for chronic kidney disease patients


Our findings suggest that CKD patient costs are significantly affected by treatment type, namely dialysis, transplant, and non-dialysis dependent CKD; costs differed by region (highest in the West region of the U.S.); and average expenditure for patients with co-morbid conditions is significantly higher than those without co-morbid conditions. Also, patients with public insurance incurred higher costs than those with private insurance, a finding that adds to the mixed results in the literature on cost variation between publically and privately insured individuals [18, 19]. The finding that costs are different by geographic region is consistent with the body of knowledge on geographic variation in care [2022], and that topic has been well researched. Targeting decision-making units rather than geographic units has been recently discussed in the literature as a way to reduce variation and total cost in health care spending [23]. Because most medical decisions are made by individuals or small groups of individuals rather than by geographic units, findings ways to improve care coordination across medical specialties and use real-time data sharing to support communication and group decision making among multiple providers seems important for improving cost effective care for patients with CKD. The following discussion mainly focuses on the treatment of CKD patients on dialysis and on managing CKD patients with co-morbid conditions because of the relatively higher costs these patients incur.

Given that the total cost of patients on dialysis is significantly higher in comparison to NDD-CKD and transplant patients, CKD specialists and policy and decision-makers should focus attention on managing the expenditures of dialysis patients. It is not surprising that once technology (i.e. a dialysis machine) is introduced in caring for the sickest patients or the most severe of cases, health care costs escalate. In some conditions, where patient costs dramatically escalate, investment in technology and expenditures on invasive procedures for the “sickest of the sick”, it could be argued, does pay off [24]. An example of this is the use of Biventricular heart assist devices (BiVad) in the congestive heart failure population. For patients with the most severe illness, the BiVad offers a decrease in hospitalization rates, but at considerable upfront costs. In contrast, if one considers CKD and dialysis, there has only been a small incremental improvement in the mortality rate of dialysis patients. The prevalence of End Stage Renal Disease is about seven percent per annum and the longevity of the average dialysis patient is five years. These numbers are actually quite interesting, if one considers the CKD/dialysis mortality rate that amounts to a 20% drop off each year, which is worse than most forms of cancer and many other chronic diseases.

The prevalence of stage 5 CKD is increasing and costs are escalating with the steepest portion of the costs for a dialysis patient being incurred during the first 180 days of dialysis treatment. This suggests that one potentially impactful strategy for decreasing these costs is to ensure patients are dialysis-ready well in advance of actually starting a patient on dialysis. Likewise, starting prospective dialysis patients on an outpatient dialysis schedule (where costs are lower), or making use of urgent-start peritoneal dialysis programs [25], and avoiding traumatic scenarios, whereby dialysis patients require emergency care and consequently incur a minimum five-day hospital stay, could significantly lower total costs for patients with CKD.

Some efforts to reduce the costs associated with caring for patients with CKD are already underway. The Centers for Medicare and Medicaid Services (CMS) ESRD Seamless Care Organization is the first disease specific Accountable Care Organization designed by CMS to identify, test, and evaluate new ways to improve care for Medicare beneficiaries with ESRD. Likewise, the Large Dialysis Organizations are developing strategies to reduce costs incurred by CKD patients by identifying “high spenders”. These include reaching out to such patients to remind them of their dialysis appointments and helping them with other disease related activities, such as managing their medications, diet, and transportation needs to and from appointments. Additionally, some dialysis providers are developing methods to mine their databases to identify “high spenders” and build economic models to determine what CKD care is best. Future research should examine the efficacy of these efforts and investigate how to best scale up successful aspects of these efforts.

We found that total costs for patients with CKD are significantly affected by co-morbid conditions (such as hypertension, diabetes, and other heart disease) and their associated costs. Cardiovasular disease and diabetes mellitus are the two leading causes of CKD. Better management of hypertension and diabetes mellitus, common co-morbidities associated with CKD, would slow the progression of kidney disease and reduce healthcare expenditures. Awareness of potential side effects due to the medications or renal insufficiency could prevent unnecessary harm to patients and provide cost-containment. Active involvement of all healthcare team members can reduce progression of CKD and improve quality of life outcomes in CKD patients. Moreover, while it is not surprising that more medically complex CKD patients incur higher costs, it reinforces the importance of delivering coordinated patient-centered care that is attentive to the whole person including the management of polypharmacy, the use of multiple medications and/or the administration of more medications than are clinically indicated. Polypharmacy continues to increase in the U.S. and is a known risk factor for morbidity and mortality. Additionally, treatment plans for CKD patients that activate and educate patients, family members and other care providers to better manage co-morbid conditions and adhere to targets established by physicians should be at the forefront of cost reduction strategies for treating individuals with CKD.

Our findings suggest that reducing the number of in-person visits and the number of prescription drugs might also reduce total costs per CKD patient. Reducing the frequency of in-person visits could be accomplished by using computer technologies, such as secure messaging (email) or telehealth systems, for non-essential office visits. Many medical systems are already using secure messaging systems to share the results from routine medical tests with patients [26, 27]. Reducing costs associated with prescription drugs is perhaps more difficult to address. One path to lowering the number of prescription medications used in the CKD population could come from better management of co-morbid conditions, as noted above. Another way could be through better efforts to more systematically address polypharmacy in this patient population. Predicting which CKD patients will end up on dialysis is very difficult. Often, patients at stage 3 or stage 4 will never receive dialysis. For every five to seven CKD patients seen, only one will end up on dialysis. The others will either not progress and/or die (usually of cardiovascular disease). Efforts might be best spent focusing on the cardiovascular health of the CKD patient population with less costly solutions (e.g. statins, blood pressure control, nutrition and exercise counseling) as a way to reduce long-term costs associated with co-morbid conditions.

Our findings also suggest that better care and reduced total costs for caring for patients with CKD could come from applying approaches to care that are grounded in systems thinking [28]. The majority of the frameworks applied to understand health care delivery systems emphasize the linear relationships among system components, where all variables and their relative weights are known [29]. Acknowledging the presence of nonlinear relationships and unpredictability in CKD treatment could help the individuals and organizations more accurately conceptualize the impact of various components of CKD care. Along these lines, models of care such as Patient-Centered Medical Homes, Accountable Care Organizations and Home-Based Primary Care are a step in the right direction for achieving a more integrated and holistic view of the complex needs shared by high-cost, high need patients with chronic disease. Such a reconceptualization of CKD processes and care trajectories could provide decision makers and policy makers with new insights into observed gaps in care and for visualizing novel paths forward for implementing CKD treatments and interventions more effectively.

Limitations and future research

The above analyses and conclusions are relevant steps toward understanding and minimizing cost of care for CKD patients. However, there are enhancements not addressed in this paper that may be worthy of further research.

First, the models could be adapted to account for various co-morbid combinations (or interactions) and their subsequent impact on costs. For instance, high blood pressure and obesity are two variables that could jointly influence the type of treatment to which a patient is subjected, which in turn would have a differing impact on costs. Introducing interaction variables, however, would require a much larger dataset since one would have to include more independent variables. Second, it would be interesting to develop a time-series model for total costs that would allow one to project expenditures at least two or three years into the future. Kidney disease is progressive as are its attendant costs. Thus, patients would be better informed if at least some costs were predicted ahead of time. Such long-term prediction models, while useful, are non-trivial to develop since, when viewed as a time series, expenditure data are highly non-stationary. Third, assessing percentage changes to total costs by changing existing protocols is worthy of study. For instance, suppose the source-of-care provider days were increased by 5%. How would that increase impact the total cost to patients by different treatment types? This type of sensitivity analysis could help providers evaluate the tradeoffs between cost and quality of care decisions. Fourth, our results suggest that costs of treating CKD patients are lower for those with private as opposed to public insurance. Research assessing this relationship perhaps comparing costs pre and post the recent U.S. Affordable Care Act is needed. Fifth, in this paper we wanted to understand CKD costs before the roll-out of the ACA whose impact on CKD patient behavior and treatment started to alter in 2013. Given the dynamics of the current election cycle, this wait-and-see approach is better justified. Also, we wanted a clean, comprehensive data that reflected some important considerations such as: use of Erythropoietin stimulating agents (which was at its peak in 2006); growing momentum for payment policy reform in certain states; and moving toward bundled payments. Some of these criteria were corrupted in data starting roughly around 2012–2013. Nonetheless, it would be useful to revisit certain aspects of this research using new data. Sixth, it would be useful to investigate the CKD cost data across different countries, given the different health care plan choices available in different nations. Perhaps, one could adapt the U.S. model for CKD costs by borrowing strength from the analysis of such data from other countries: a meta-analysis study may be appropriate to address this research issue. Finally, insights from this study could be used to inform future research to understand and manage the cost-quality tradeoff for other complex and debilitating chronic conditions such as epilepsy, asthma, and heart disease.


Related Posts

  • No Related Posts