Abstract
Objective: To assess the health-related quality of life (HRQL) and economic burden of chronic kidney disease (CKD) related anemia in non-dialysis patients in the United States (US) via literature review.
Methods: MEDLINE, EMBASE, PROQOLID, and Cochrane Library/Renal Group Resources were searched. Studies were appraised for patient populations, disease-specific versus generic HRQL assessments, and type and magnitude of health-related costs.
Results: The treatment costs for CKD patients with anemia compared to those without anemia were significantly higher and were blunted but persistent after controlling for comorbidities and confounders. Intervention with erythropoiesis stimulating agents (ESA) decreased anemia and avoided hospital admissions. Costs were higher when anemia was poorly controlled or untreated. HRQL burden was mainly due to physical limitations and difficulty in ability to perform activities of daily living. Significant positive correlations between increases in hemoglobin levels and HRQL measures were reported.
Conclusions: Although evidence is limited, the economic and HRQL burden of non-dialysis CKD-related anemia is substantial. Under-treatment of anemia may contribute to higher resource consumption and higher costs; however, patient co-morbidities, use of erythropoietin-stimulating agents, and overall management introduce potential confounds. The contribution of anemia to humanistic disease burden is due to a constellation of factors, including physical activity and functional status.
Introduction
Chronic kidney disease (CKD) is highly prevalent in the United States (US), affecting approximately 20 million AmericansCitation[1]. More than 80% of patients with impaired renal function eventually suffer from anemiaCitation[2],Citation[3]. The frequency of anemia in non-dialysis CKD patients range from 48%Citation[4] to 60%Citation[5] and the prevalence increases with disease severityCitation[4]. Furthermore, management of anemia is inadequate in this population: consequently, anemia remains a significant problem in the non-dialysis CKD populationCitation[3],Citation[6].
Anemia is more prevalent in certain population subgroups, especially among African-Americans, where the odds of anemia in CKD patients of all severity levels are three times greater than among white patientsCitation[7]. This trend persists in the non-dialysis CKD population, where the odds of having anemia are 1.6 to 2 times higher for African-Americans than for CaucasiansCitation[4]. Risk factors for developing CKD-related anemia include hypertension, cardiovascular disease, and diabetes – diseases that are on the rise due to increases in obesity and an aging population in the USCitation[7] and women appear to carry a higher riskCitation[3],Citation[5],Citation[8]. The link between diabetes and CKD-related anemia has also been established specifically within the non-dialysis CKD populationCitation[4],Citation[8].
The annual direct medical costs of managing CKD patients with anemia were estimated at approximately $78,209 per patient, over $20,000 more than for CKD patients without anemiaCitation[9]. Although, few studies have assessed the cost of anemia-related CKD specifically among non-dialysis patientsCitation[10], there is evidence that failure to treat anemia in the non-dialysis CKD population may result in even higher costs of careCitation[10], resulting in a growing concern for a population that often remains untreatedCitation[3].
In the US, management of anemia includes two erythropoietin-stimulating agents (ESAs), epoetin alfa (EPO) and darbepoetin alfa (DARB), which are used in addition to (or in lieu of) blood transfusions. The National Kidney Foundation has issued guidelines and recommendations for the treatment of CKD-related anemiaCitation[11]. These guidelines recommend ‘selection of the hemoglobin (Hb) target and selection of the Hb level at which ESA therapy is initiated in the individual patient should include consideration of potential benefits (including improvement in quality of life (QoL) and avoidance of transfusion) and potential harms (including the risk of life threatening adverse events).’ Thus, there is a clear need to understand the burden of disease in anemic CKD patients in order to evaluate the degree of benefits that can potentially be derived from treatment. To date, no review of the costs of anemia due to non-dialysis CKD has been completed. The purpose of this review is to summarize the primary literature on the economic and health-related quality of life (HRQL) burden of non-dialysis CKD-related anemia in the US.
Patients and methods
Separate searches were conducted for economic and HRQL burden of illness studies. Publications were identified by searching MEDLINE, EMBASE, PROQOLID, Cochrane Library, and The Cochrane Renal Group Resources. Databases were searched from January 1, 1996 to July 31, 2009. The search was limited to studies in adults (aged 18 years or older) within the US. Letters, literature reviews, commentaries, and editorials were excluded from the searches. The analysis used only primary data to avoid duplication of data in reviews, however, bibliographies of reviews were examined for primary data potentially missed in the search.
Search terms included anemia, chronic kidney failure, chronic renal insufficiency, chronic kidney disorder, chronic renal disease, chronic nephropathy, or chronic kidney disease.
Economic burden of CKD-related anemia search strategy
In combination with the search terms listed above, publications on the economic burden of anemia and CKD in the US were identified using the following terms: cost of illness, healthcare cost, costs and cost analysis, cost-benefit analysis, economic costs, burden, healthcare utilization, and resource use. Publications on inpatient, outpatient, and institutionalised or nursing home populations were included in the search.
HRQL burden of CKD-related anemia search strategy
Search terms for HRQL included the disease search terms listed above plus health-related quality of life, quality of life, functional status, health outcomes, patient satisfaction, patient-reported outcomes, self report, cognitive functioning, emotional functioning, and social functioning. Other searches for articles on measures of HRQL in CKD-related anemia were conducted by identifying articles containing information on questionnaires, surveys, instruments, scales, measures, assessments, indexes, diaries, or inventories.
Results
Search strategy for economic burden
The search strategy for publications on the economic burden of CKD-related anemia yielded an initial 335 articles, of which 86 were fully reviewed. A total of 249 publications were excluded because they did not meet the pre-defined inclusion criteria. Another eight publications were identified from searching the reference lists of the reviewed articles. Overall, nine articles were accepted for inclusion in this review and 85 articles were excluded (). The reasons for rejection were: not CKD population (21%), no economic data for patients with CKD and anemia (41%), dialysis population (20%), direct comparison of anemia therapies (16%), non-US study (1%), or review (1%). summarises the nine accepted studies for the economic burden of non-dialysis CKD-related anemia.
Figure 1. Flow chart for identification of studies in the systematic review: economics.
Table 1. Economic burden literature included.
The majority of included economic burden studies were retrospective analyses of medical records or chart reviews (89%). One survey study was also included. Approximately one-half (44%) of these articles emphasised the impact of treatment in patients with CKD and anemia. The earliest economic studies of CKD and anemia appeared shortly after the introduction of erythropoietin (EPO) into the market in the late 1980s. In the 1990s, most studies focused on patients receiving dialysis; it was not until recently that the focus shifted to non-dialysis patients.
Search strategy for HRQL burden
The search strategy for publications on the HRQL impact of CKD-related anemia yielded a total of 290 initial titles, of which 213 were excluded because they did not meet the pre-defined inclusion criteria. Another four publications were identified by searching the references of reviewed articles. In total, 81 papers were retrieved, and 16 articles were accepted for this review (). The major reason for article exclusion was lack of any information on HRQL or patient-reported outcomes (PROs); 31% of all rejected articles were not reviewed for this reason. The remaining articles were excluded due to a lack of information on US patient populations (12%), review articles (32%), no data on anemic CKD patients (5%), non-English language (9%), or lack of non-dialysis patient population (11%). summarises the papers included in this review of the HRQL burden.
Figure 2. Flow chart for identification of studies in the systematic review: quality of life.
Table 2. Health-related quality of life literature.
Of those papers included, six (38%) reported results from non-randomized clinical trials (nRCTs), and seven (44%) summarized the findings of randomized clinical trials (RCTs). Three papers (19%) reported results from observational/survey studies. A total of 31% of articles did not specify dialysis status or included a mixed (non-dialysis and dialysis) patient population.
Only one paper focused specifically on the HRQL burden of non-dialysis CKD-related anemiaCitation[12]. All other information was extracted from study results or baseline values of untreated patients. Regarding interventions for CKD-related anemia, 25% of papers focused on treatment with recombinant human ESAs. Two papers directly elicited utility measures from patients. Eight papers (50%) included an HRQL measure that was specific to CKD.
Economic burden
Economic impact of anemia in CKD patients
The selected nine publications were separated into two groups for discussion: general top-line analyses of overall costs of anemia in CKD, treated or not (four studies), and the impact of anemia treatment costs in non-dialysis patients (five studies).
Ershler et al. (2005)Citation[9] investigated the general economic burden of anemia in several chronic conditions, including CKD, using US administrative claims (1999–2001) and productivity (1997–2001) data. The authors modeled the annual per-patient cost burden for anemic and non-anemic CKD patients and reported unadjusted direct costs of $78,209 and $24,784 per patient, respectively (difference $53,425). After adjusting for baseline characteristics, such as age, sex, region of residency, insurance type, disease severity (based on International Classification of Diseases – ICD-9, Healthcare Common Procedure Coding System – HCPCS, or pharmacy code), Charlson comorbidity index (CCI, which defines 17 categories of comorbid conditions using ICD-9 diagnosis codes), and presence of more than one of the selected conditions, the direct cost differences between groups were calculated to be $20,529; however, direct costs for anemic patients remained higher. Unadjusted indirect costs were similar between anemic and non-anemic CKD patients: $2,948 and $2,992 per patient, respectively (difference −$44). Adjusting for baseline characteristics, such as age, sex, comorbidities, and disease severity did not significantly affect indirect cost differences, although the polarity of the cost difference was reversed (difference $64 with indirect costs for anemic patients being higher). The authors also examined the direct costs associated with the severity level of CKD disease, independent of anemia status. Patients with mild disease (defined as no hospitalizations) incurred significantly lower costs than those with moderate-to-severe disease (defined as >1 hospitalization) ($24,202 vs. $85,145; p < 0.001). The authors also estimated the yearly direct costs for patients with anemia and congestive heart failure ($72,078) and anemia with cancer ($60,447). In this sub-analysis, anemic patients with CKD had the highest average annual costCitation[9].
Lefebvre et al. (2006)Citation[10] demonstrated that untreated anemia in elderly patients with non-dialysis CKD was associated with a significant increase in medical costs based on medical and pharmacy claims, with an unadjusted incremental monthly cost of $1,089 ($2,529 vs. $1,439; p < 0.0001) and a cost ratio of 1.8:1 (p < 0.0001). This result was primarily driven by hospitalization costs, which were more than twice as high in untreated anemic patients than in non-anemic patients ($1,734 vs. $856 per month). Additionally, a subanalysis of patients with more moderate CKD (stage 3) showed even greater cost differences between untreated anemic patients and non-anemic patients across all cost categoriesCitation[10]. As demonstrated by Ershler et al. (2005)Citation[9], adjustments for comorbidities reduced the cost impact of untreated anemiaCitation[10]. Multivariate analyses demonstrated that lower Hb and glomerular filtration rate (GFR) levels were associated with higher medical costs; however, even after controlling for covariates, cost differences between untreated anemic and non-anemic patients remained significantCitation[10].
Recently, Wish et al. (2009)Citation[13] examined the direct cost burden of CKD stratified by anemia status and by treatment status and observed that presence of anemia was associated with greater medical expenditure in patients with CKD. Their results indicated that CKD patients with anemia had higher inpatient, outpatient, and pharmacy costs, even after adjusting for underlying covariates (age; sex; geographic location; employment status; insurance type; Chronic Disease Score; baseline expenditure, use of antihypertensive, antidiabetic or antihyperlipidemic therapy; and presence of comorbidities such as hypertension, diabetes, glomerulonephritis, vascular kidney disease, polycystic kidney disease, or congenital kidney disease; p < 0.0001) compared to non-anemic CKD patients. The unadjusted total cost for the patients without anemia was $2,664 vs. $4,076 for the patients with anemia (p = 0.0033). The authors also assessed cost differences between treated versus untreated anemic CKD patients. Although outpatient and pharmacy costs were higher in treated patients, untreated patients incurred significantly higher inpatient costs (largely due to acute admissions). Unadjusted total costs for patients with treated anemia and CKD versus untreated anemia and CKD were $3,806 vs. $4,470, respectively (p = 0.0033). These results remained robust after adjustments for underlying covariatesCitation[13].
The increased costs related to anemia reported by Lefebvre et al. (2006)Citation[10] and Wish et al. (2009)Citation[13] were due to higher resource use, particularly hospitalizations possibly due to poorly controlled anemia and attempts made to manage itCitation[9]. Rasu et al. (2007)Citation[14] reported that 48% of visits to physicians by patients with CKD included anemia as a reason for the visit. Only 10% of all visits resulted in prescribing medications or transfusions for managing the anemiaCitation[14]. London et al. (2002)Citation[15] found that only a minority of non-dialysis CKD patients diagnosed with anemia received prescription iron preparations (6.8%), vitamin D (4.0%), or phosphate binders (7.7%). Only 10.5% received non-dialysis EPO, with or without other therapies, yet more than 40% were diagnosed with anemiaCitation[15]. These findings indicate that a sizeable proportion of patients with anemia and CKD were undertreated, which resulted in increased resource usage, and that treatment of anemia could be associated with reduced inpatient costsCitation[10],Citation[13].
Cost burden of treating anemia
In 1990, recombinant human erythropoietin (epoetin) was licensed in the US and Europe for the treatment of anemia associated with chronic renal failure, including patients not on dialysis. The most recent National Kidney Foundation Kidney Disease Outcomes Quality Initiative (NKF KDOQI) guidelines (2007) consider ESAs as critical components in the management of anemia of CKDCitation[11]. Trends among end-stage renal disease (ESRD) patients show that ESA use has steadily risen since 199516. Available data suggest that ESAs may have contributed to rising costs for treating anemia in CKD patients. It is undetermined whether if these treatment costs are offset by the fewer hospital admissions associated with ESA therapy. Recent data relating to costs due to increased risk of cardiovascular events could hinder cost offsets and benefits of the therapyCitation[17].
Two studies showed a decrease in direct costs for non-dialysis patients treated with EPOCitation[18],Citation[19]. Moyneur et al. (2008)Citation[19] used employer claims data (January 1998–January 2005) to compare costs between anemic CKD patients treated with EPO prior to dialysis initiation and anemic CKD patients who did not receive ESA therapy prior to dialysis initiation. The authors calculated mean per member per month (PMPM) inpatient costs, outpatient costs, prescription drug costs, sick leave costs, and disability costs. Direct costs (defined by the authors as medical care costs) in the case of EPO-treated patients were significantly lower after adjusting for comorbidities ($2,779 vs. $4,222).
Incremental cost savings were $1,443 for direct healthcare costs for treated compared to untreated patientsCitation[19]. In Maddux et al. (2007)Citation[18], retrospective claims from a large US health plan database (approximately 13 million persons) were reviewed for inpatient, outpatient, emergency department, and prescription experience. The use of ESA therapy (14.6% of the sample) in non-dialysis patients was associated with $411 lower costs (not significant; p = 0.1337) per patient per month (the difference in adjusted total per patient per month anemia-related costs). These results reflected reduced inpatient and emergency department visits and costs, lower inpatient mortality, and longer time to dialysisCitation[18]. The lower monthly inpatient and emergency room costs among treated CKD patients compared to untreated patients were $2,507 vs. $3,849 and $47 vs. $81, respectively. Monthly outpatient and laboratory costs, however, were higher in ESA-treated patients compared to untreated patients ($602 vs. $397 and $24 vs. $14, respectively)Citation[18].
Nonetheless, London et al. (2002)Citation[15] observed no overall adjusted cost differences between EPO- and non-EPO-treated anemic CKD patients. Using retrospective claims analysis based on a large managed-care database (January 1997 to December 1999), the authors examined claims for facility use (hospitalization, emergency room), professional services (procedures, tests, visits), and pharmacy claims. Inpatient hospitalization was the only cost category that demonstrated significantly higher costs among non-EPO patients (p = 0.0013) after adjusting for age, sex, and the Charlson Comorbidity IndexCitation[15]. Adjusted costs for professional (procedures, visits, and tests) and pharmacy services (non-EPO medications) were significantly lower among non-EPO patients (p = 0.0001 and p = 0.0011, respectively).
Two studies examined the indirect effects of ESA treatment. Papatheofanis et al. (2008)Citation[20] examined medical costs and productivity losses (measured in days worked per week) for non-dialysis ESA-treated patients. The study reported higher productivity (more days worked) and lower costs to employers (both direct medical costs and indirect costs due to absenteeism) after individuals began ESA treatmentCitation[20]. Importantly, this group served as its own control so that differences in productivity before and after the initiation of EPO treatment were tracked. Moyneur et al. (2008)Citation[19] observed lower indirect costs (sick leave and disability cost) among EPO-treated patients compared to untreated anemic CKD patients after adjusting for comorbidities ($544 PMPM vs. $872 PMPM). In comparison to untreated anemic patients, there were incremental indirect cost savings for EPO-treated patients of $32819.
Based on a retrospective cohort design, Perkins et al. (2007)Citation[21] investigated the impact on hospital resource use of a dedicated nurse-managed anemia clinic in patients with pre-end-stage chronic kidney disease. Patients received algorithmic treatment with recombinant human erythropoietin and intravenous iron sucrose. The study used a historical control group based on chart data of patients who did not receive hemodialysis, peritoneal dialysis, recombinant human erythropoietin, or intravenous iron at any time. The study demonstrated that the nurse-managed plan reduced emergency room visits (relative risk: 0.18; p < 0.05) and hospitalization (relative risk: 0.20; p < 0.05) compared with controlsCitation[21]. The average length of hospital stay per year was lower for treated anemic CKD patients (6.8 days) compared with untreated anemic CKD patients (6.8 days vs. 9.5 days; p = 0.05).
Using a prospective, open-label, non-randomized design, Provenzano et al. (2004)Citation[22] assessed the impact of treatment with ESA on transfusions and HRQL in anemic-CKD patients. Once-weekly injection of erythropoietin reduced the burden of CKD-related anemia by decreasing the percentage of non-dialysis patients requiring transfusions (11.1% during the 6 months prior to study entry to 3.7% during the study)Citation[22].
HRQL burden
Measures of HRQL in patients with CKD-related anemia
Studies employed a variety of instruments to evaluate patient HRQL outcomes in CKD-related anemia. Both disease-specific and generic health status and QoL measures were included in this research. Disease-specific measures included the Kidney Disease Quality of Life (KDQOL), the KDQOL-Chronic Renal Insufficiency (KDQOL-CRI) questionnaires, and the Kidney Disease Questionnaire (KDQ). The KDQOL contains a generic questionnaire, the SF-36, and disease-specific items. In addition, one studyCitation[23] included the Functional Assessment of Cancer (FACT) anemia and fatigue measures. The anemia and fatigue modules of the FACT tool, originally developed for use in oncology patients, have been validated in other chronic disease populations but not in CKD patientsCitation[24]. No HRQL instrument specific to anemia in CKD patients was identified. The most commonly-used generic questionnaire was the SF-36, which was used in four studies; the brief version of this instrument, the SF-12, was included in two studies. Components of the Sickness Impact Profile (SIP) were also frequently employed. A 100-mm Linear Analogue Scale Assessment (LASA) of energy, ability to perform daily activities, and overall QoL was included in three studies. A variety of other instruments were used to measure study-specific constructs such as mental status, occupational functioning, cognitive functioning, and sexual dysfunction. Only two studies directly evaluated the burden of CKD-related anemia – Revicki (1992)Citation[25] and Muirhead et al. (1994)Citation[26] – both of which measured the utility of health states and treatments for anemia using standard gamble and preference elicitation techniquesCitation[25], or time trade-off (TTO) methodsCitation[26].
Generic measures of functioning and QoL in untreated patients with CKD-related anemia reveal significant decrements in physical functioning for this population. Researchers reported that the average baseline SF-36 and SF-12 subscale scores for Physical Function, Role Physical, and Vitality all were well below the 25th percentile for the US general adult populationCitation[23],Citation[27]. The Physical Component Summary (PCS) mean scores for these studies were similarly below the 25th percentile for the US healthy populationCitation[28],Citation[29]. Bodily pain and subscales for socioemotional function appeared to be somewhat less affected by CKD-related anemia, with average scores at or above the 50th–75th population percentilesCitation[23],Citation[27]. Comparisons of study SF-12 or SF-36 scores to the US healthy adult normative population are presented in .
Table 3. Average SF-36 and SF-12 Scores: patients with CKD-related anemia and normative US general population*.
A study comparing HRQL among non-dialysis patients with differing hemoglobin levels showed significant improvements on every domain of the SF-36, except emotional well-being, as Hb levels increased from <11 to ≥13 g/dLCitation[12]. The authors further reported that the association between Hb levels and QoL was independent of ESA use. These findings were similar to those reported by Drueke et al. (2006)Citation[30], which compared changes in HRQL for a group of CKD patients with normal Hb measurements (Group 1; 13–15 g/dL) to a group of CKD patients with subnormal hemoglobin measurements (Group 2; 10.5–11.5 g/dL). EPO treatment was initiated at randomization for patients in Group 1 but only after Hb levels fell below 10.5 g/dL for patients in Group 2. After one year, HRQL, as measured by the SF-36, was significantly better among patients in Group 1 compared to those in Group 2 with regard to general health (p = 0.003), mental health (p < 0.001), physical function (p < 0.001), physical role (p = 0.01), and social function (p = 0.006), indicating a relationship between higher Hb targets and better HRQLCitation[30]. Another study investigating the impact of ESA use in anemic CKD patients reported significant improvement in four of eight domains of the SF-36: physical functioning, role physical, vitality, and social functioningCitation[31].
Studies using other generic measures of HRQL in untreated CKD-related anemia patients obtained similar results and showed the positive effect of treatment on HRQL. Average LASA scores (possible score range: 0–100, higher = better) before treatment for energy and activity level ranged from 38.1–42.9 in anemic CKD patients. Provenzano et al. (2005)Citation[32] showed that LASA scores were maintained or improved with ESA treatment in patients with stable Hb levels (≥11 g/dL) randomized to one of four ESA treatment regimens. Singh et al. (2006)Citation[27] demonstrated improvements in HRQL measures for patients given ESA and randomly assigned to achieve either a high (13.5 g/dL) or low (11.3 g/dL) Hb target level. Average LASA scores improved by 16.6 points for energy and 15.0 points for activity in the high Hb group and by 15.5 points for energy and by 13.3 points for activity in the low Hb groupCitation[27]. Similar findings were reported in Benz et al. (2007)Citation[31], in which there was a 17.0 point mean change for activity and a 20.6 point mean change for energy in patients receiving ESAs. Overall, perceived QoL scores were somewhat higher (mean score range: 46.1–47.7), but still under 5022,Citation[27],Citation[33], and could improve with ESA treatmentCitation[27]. SIP subscale scores (0–100; lower score = better health) varied by scale. Revicki (1992)Citation[25] and Revicki et al. (1995)Citation[34] observed moderately low mean scores for the home management SIP scale, which measures patient's ability to perform activities of daily living (average scores about 37.0). Patients reported less difficulty with cognitive functioning, as measured by the Alertness Behavior Scale (mean score range: 15.4–25.1), as well as social interactions (mean score range: 20.5–24.4).
Assessments by disease-specific measures of HRQL obtained similar results. Patients who completed the KDQ (overall score range: 4–35; item score range: 1–7, higher score = better functioning) reported moderate limitations due to physical symptoms (average score 3.0) and fatigue (average score 3.4–3.5). Mean scores for condition-specific depression, frustration, and limitations in relations with others were not significantly altered, with scores ranging from 4.1 to 4.622,Citation[33]. Singh et al. (2006)Citation[27] reported only the overall mean KDQ score (20.3), a result similar to the overall mean score of 19.7 reported by Lefebvre et al. (2006)Citation[33]. Provenzano et al. (2004)Citation[22] did not report the overall KDQ score for their patient sample. Several authors reported the results of anemic CKD patient HRQL based upon the KDQOL (score range: 0–100; higher = better functioning). The average burden of kidney disease, as measured by KDQOL items that focused on the time needed to take care of this condition, ranged from 52.4 to 67.3 (Hb < 11 g/dL) to 72.712,Citation[23],Citation[29]. Concomitant with their findings from the SF-36 questionnaire, Finkelstein et al. (2008)Citation[12] reported significant increases in four domains of the KDQOL as hemoglobin levels improved. Patients with kidney disease reported minimal effects on their quality of life by fluid and dietary restriction, an inability to travel, and a dependence on doctors, with average scores ranging from 75.1 to 86.223,Citation[29]. Similarly, the kidney disease symptom scale, which measures the degree to which the respondent is bothered by symptoms such as muscle cramps, pruritus, and anorexia, ranged from 73.5 to 78.123,Citation[29].
Another indication of disease burden is the measure of patient preference of utility for different health states. Two studies evaluated HRQL from this perspective. Muirhead et al. (1994)Citation[26] used the Time Trade-Off (TTO) technique to elicit patient utility of present status before, during, and after treatment of anemia in post-renal transplant patients, with scores ranging from 0 (indifference between life and death) to 1 (perfect health).
The average baseline score was 0.60 (SD 0.28), which indicated a moderate utility for current health status; there was no significant difference after 24 weeks of therapy with EPO. Revicki (1992)Citation[25] used both standard gamble and categorical rating scale methods of evaluating health state preference for a variety of attributes in non-dialysis patients, using a 0–100 scale (100 = best possible health state). Patients in this study also showed a moderate utility for their current, untreated health state at baseline, with standard gamble score averages of 63 (SD 0.04) and a mean categorical scale preference of 59 (SD 0.04). The health status value showed a high degree of association with the SIP Home Management subscale, indicating that activities of daily life play a major role in how anemic CKD patients perceive the burden of their condition.
Impact of treatment on HRQL in patients with CDK-related anemia
The objectives of the majority of reviewed HRQL studies centered around the evaluation of treatments for CKD-related anemia, not on the impact of anemia on patient's functioning. Thus, anemia-related QoL was largely inferred from the baseline HRQL levels in the study populations. While treatment of CKD-related anemia might be assumed to improve patient HRQL, this may be attenuated by adverse events due to anemia therapy.
Other than a recent studyCitation[17], published after completion of this review, few papers describing clinical studies of CKD-related anemia reported adverse events associated with treatment. Adverse events associated with ESA treatment in non-dialysis patients resulted from infections, renal or urinary tract disorders, and problems at the injection siteCitation[28]. Hypertension, headache, diarrhea, rash, edema, and cardiac problems were also associated with ESA treatment in this populationCitation[22],Citation[28],Citation[35]. Muirhead et al. (1994)Citation[26] reported hypertension, flu-like symptoms, vomiting, and diarrhea among both non-dialysis and dialysis patients. The incidence of these adverse events was low – 1.6% – for problems at the site of medication administrationCitation[28]. Moreover, one study reported an increased risk of cardiac events in mixed CKD populations with untreated anemiaCitation[30]. These factors must be balanced against the risks and benefits of treatment in this population.
Discussion
Although only limited information is available, the overall economic and quality-of-life burden of CKD-related anemia appears to be substantial. Studies estimate that the average costs related to anemia in patients with CKD can be as high as $78,209 per patient per year in direct costs and $2,948 per patient per year in indirect costs. However, after adjusting for baseline characteristics including disease severity, the direct cost estimates decreaseCitation[9]. Controlling for covariates, such as patient characteristics and/or comorbidity, has also been shown to reduce cost differences between anemic and non-anemic patientsCitation[9],Citation[10],Citation[13], although a number of studies reported significantly higher costs among anemic patients after making these adjustmentsCitation[10],Citation[13]. Hospitalization was observed to be the most influential cost category in a number of studiesCitation[10],Citation[13],Citation[19], as a driver for higher costs in anemic versus non-anemic patientsCitation[13], and untreated versus treated anemic patientsCitation[10],Citation[13],Citation[19]. One study suggested that anemia management may help to reduce the inpatient costs associated with anemia in the CKD populationCitation[13].
Although, anemia in CKD patients is widely recognized, evidence suggests that it is not always treated adequately. There are indications, based upon physician prescribingCitation[14],Citation[15], that anemia may be undertreated in non-dialysis patients despite documented benefits of treating anemia, which may reduce costs associated with resource consumption, such as hospital admissionsCitation[10],Citation[13]. Some studies have suggested that treatment with ESA may lower overall directCitation[18],Citation[19] and indirectCitation[19],Citation[20] costs in non-dialysis patients; however, other studies have shown no overall cost differences between ESA-treated and untreated anemic patientsCitation[15]. Notably, however, London et al. (2002) reported significantly higher hospitalization costs among untreated patientsCitation[15]. Pharmacy costs do appear to increase among ESA-treated patientsCitation[15],Citation[18],Citation[19]; however, it is uncertain whether these costs offset other costs, such as hospitalization.
Anemia also impacts the HRQL of patients with CKD and anemia. The HRQL burden associated with CKD-related anemia is mainly due to physical limitations and consequent problems in ability to perform activities of daily life. This is not surprising, considering the most common symptom of anemia is fatigueCitation[36]. Despite significant disability compared to healthy populations, there is little evidence that anemic CKD patients experience a large decline in their psychological health status or their emotional well-being. It is unknown whether this finding is due to a low level of life satisfaction among healthy individuals or to measurement insensitivity for psychosocial problems in patients with CKD-related anemia. These findings are supported by a 2009 review that examined the effects of ESAs on HRQL among anemic CKD patientsCitation[37]. This study emphasised improvement of physical symptoms (vitality, energy, and performance) with ESA use, while noting that social, mental, and emotional functioning show little improvementCitation[37]. A more recent review reported that ESA use in controlled trials improved energy and physical function in non-dialysis CKD patientsCitation[38].
The attribution of decrements in HRQL to anemia rather than the CKD itself is demonstrated by the almost universal finding of improvements in most functional status measures upon achievement of a normalized Hb or hematocrit and the lack of significant improvements in HRQL in patients treated with placebo. However, patient utility values for current health did not appear to change post-treatment in either study of patient health-state preference of utility, and Revicki (1992)Citation[25] found that health status explained only 27% of the variance in current health-state utilitiesCitation[26],Citation[34]. Thus, the contribution of HRQL to overall patient disease burden may be due not only to Hb levels, but also to additional factors such as functional status.
As in all literature reviews, the study has limitations. First, as this study is a summary of the available literature rather than a re-analysis of published data using quantitative methods, no direct comparisons of the data across studies can be made. The heterogeneity of included studies would likely have precluded any formal statistical analysis.
Second, most of the economic studies are retrospective analyses of claims data or chart reviews and subject to issues of missing data and limited to direct medical resource use. Correcting for co-morbidities is a challenge in these retrospective analyses and the failure to fully identify the co-morbidities may explain, at least in part, some of the observed differences, particularly within the economic data.
These issues are further complicated in studies that failed to adjust for disease severity. Ershler et al. (2005)Citation[9] observed a large stratification in direct costs when costs were analyzed by disease severity (mild vs. moderate/severe disease). Wish et al. (2009)Citation[13] and Lefebvre et al. (2006)Citation[10] report higher costs among patients who are anemic, but they fail to make adjustments for disease severity. Thus, it is unclear if the patients in these two studies incur higher costs because of their anemia or because they are more sick. Although the majority of studies point to hospitalization as the primary cost driver in anemia costs, it is worth noting that Ershler et al. (2005)Citation[9] reports almost five times as many hospitalizations per year among moderate-to-severe CKD patients as compared to patients with mild disease.
Third, it is unclear if patients receiving ESA treatment are receiving better medical management in general. This could explain the cost differences between those receiving ESA and those not receiving ESA. Furthermore, of the studies that examine the costs associated with treating anemia, none specifically accounted for the costs of adverse events associated with ESA treatment, and a recent publicationCitation[17] suggested that ESA may increase the risk of stroke in non-dialysis CKD patients. Thus, the impact of ESA use on costs, as well as long-term outcomes, is unclear. Additionally, only one study directly compared costs between treated anemic CKD and untreated anemic CKD patients, and more data are needed to understand whether cost offsets are achieved with treatmentCitation[13].
Fourth, since only one study focused entirely on the HRQL burden of diseaseCitation[12], most of the information reported here was inferred from baseline and placebo group data in clinical or observational studies. In addition, two studies included in the review are over 10 years old and may not reflect current ESA treatment strategiesCitation[25],Citation[26]. Thus, their findings should be interpreted with caution. The study by Gandra et al. (2010) specifically reviewed the impact of ESA on energy and physical functionCitation[38]. Published after the search cutoff date for the current review, it was not captured, but all of the US articles published between January 1996 and July 2009 included in that more restricted review were included in the current review. That study concluded that ESA improved energy and physical functioning based upon clinical trial findingsCitation[38].
Fifth, although the objective of this review was to examine the burden of illness in non-dialysis CKD patients, populations in the selected articles were often mixed (combination of non-dialysis and dialysis) or unspecified. Most studies failed to identify their study populations by NKF KDOQI stages. Additionally, definitions of anemia in the studies were based on different hemoglobin levels (Hb < 10 g/dL to Hb < 12 g/dL), or were defined according to ICD-9 codes, or were not specified, complicating comparisons across studies. Thus, while the present analysis attempts to focus on the economic burden and HRQL for a specific patient group, it is unclear if these results are purely representative of non-dialysis, anemic patients.
Finally, recent evidence has emerged since the current review was undertaken that calls into question the benefits of ESA use to achieve a target hemoglobin level of 13 g/dL in patients with type 2 diabetes, CKD, and anemia. The report suggested that ESAs do not reduce the risk of death, or cardiovascular or renal events compared to placeboCitation[17]. This study also reported that ESAs showed significant improvement on only one of the PRO measures tested relative to placebo (the FACT-Fatigue score) and did not demonstrate benefits in domains of energy and physical functioningCitation[17]. Although this report was based on a different population than other studies examined in the current review, it raises concerns about whether the benefits of ESA use in anemic CKD patients outweigh the potential risks. The HRQL findings of this review must be considered in light of this recent evidence. Several recent reports raise major concerns regarding the use of ESAs because of serious adverse events related to therapeutic goals of higher Hb concentrationsCitation[17],Citation[27],Citation[30],Citation[39] and could, therefore, be the subject of re-evaluating the objectives and dosages of ESA therapy in non-dialysis CKD patients with anemia. While the present study does not attempt to address the therapeutic benefit controversy surrounding ESAs, the review of literature indicates that treatment of anemia in this sub-population of CKD patients may reduce treatment costs and help improve several quality of life indicators, particularly energy and physical functionCitation[38]. This has also been demonstrated in large clinical trials that used ESA therapy for non-dialysis patients with CKD-related anemiaCitation[17],Citation[27],Citation[30],Citation[38].
Despite these drawbacks, the current review identifies the published attempts to quantify the economic burden of anemia in non-dialysis CKD patients and the impact of anemia on HRQL. The review examines the impact of anemia and extends beyond the recent HRQL review focusing on the treatment impact of ESA on anemiaCitation[38].
When conducting future economic studies, the authors recommend that all types of resource use (direct as well as indirect) should be included to provide a more complete economic estimate. Moreover, control for confounds, such as disease severity, comorbidities, and the level of medical management are required to isolate true cost differences between patient groups. Ideally, prospective trials, are warranted to understand more fully the economic impact of anemia in CKD patients.
No anemia-specific HRQL instrument was found or validated for patients with non-dialysis CKD-related anemia. It is worthwhile to consider the development of such an instrument to gain understanding of the difference in HRQL between patients with CKD and patients with CKD and anemia.
Conclusion
When viewed in the context of the above considerations, this review offers insight into the available economic and HRQL disease burden literature of CKD-related anemia in non-dialysis patients and recommends approaches to future research to correct the drawbacks in many of the identified studies. The economic and HRQL burden of non-dialysis CKD-related anemia seems substantial; however, as evidence is limited, more information about the factors that contribute to overall patient disease burden is needed.
Transparency
Declaration of funding: This study was supported and reviewed by Centocor Ortho Biotech Services, LLC.
Declaration of financial/other relationships: F.E.vN., J.G., and R.B. have disclosed that they are employed by United BioSource, a company that received funding from Centocor Ortho Biotech to conduct this study. F.O.F. has disclosed that he has no relevant financial relationships.JW. has disclosed that he is on the speakers’ bureau for Centocor Ortho Biotech and Amgen.
References
- Departments of Labor, Health and Human Services, and Education. Senate Report 109–103. 109th Congress, 1st Session 2006.
- Melnikova I. Anaemia therapies. Nat Rev Drug Discov 2006; 5: 627–628
- McClellan WM, Jurkovitz C, Abramson J. The epidemiology and control of anaemia among pre-ESRD patients with chronic kidney disease. EurJ Clin Invest 2005; 35(Suppl 3)58–65
- McClellan W, Aronoff SL, Bolton WK, et al. The prevalence of anemia in patients with chronic kidney disease. Curr Med Res Opin 2004; 20: 1501–1510
- Jungers PY, Robino C, Choukroun G, et al. Incidence of anaemia, and use of epoetin therapy in pre-dialysis patients: a prospective study in 403 patients. Nephrol Dial Transplant 2002; 17: 1621–1627
- Obrador GT, Pereira BJ. Anaemia of chronic kidney disease: an under-recognized and under-treated problem. Nephrol Dial Transplant 2002; 17(Suppl 11)44–46
- McFarlane SI, Chen SC, Whaley-Connell AT, et al. Prevalence and associations of anemia of CKD: Kidney Early Evaluation Program (KEEP) and National Health and Nutrition Examination Survey (NHANES) 1999-2004. Am J Kidney Dis 2008; 51(4 Suppl 2)S46–55
- Lorber DL, Provenzano R, McClellan W. Prevalence and treatment of anemia with once-weekly epoetin alfa in patients with diabetes and chronic kidney disease. Endocr Pract 2006; 12: 506–513
- Ershler WB, Chen K, Reyes EB, et al. Economic burden of patients with anemia in selected diseases. Value Health 2005; 8: 629–638
- Lefebvre P, Duh MS, Buteau S, et al. Medical costs of untreated anemia in elderly patients with predialysis chronic kidney disease. J Am Soc Nephrol 2006; 17: 3497-3502
- KDOQI Clinical Practice Guideline and Clinical Practice Recommendations for anemia in chronic kidney disease: 2007 update of hemoglobin target. Am J Kidney Dis 2007;50:471–530.
- Finkelstein FO, Story K, Firanek C, et al. Health-related quality of life and hemoglobin levels in chronic kidney disease patients. ClinJ Am Soc Nephrol 2009; 4: 33–38
- Wish J, Schulman K, Law A, et al. Healthcare expenditure and resource utilization in patients with anaemia and chronic kidney disease: a retrospective claims database analysis. Kidney Blood Press Res 2009; 32: 110–118
- Rasu RS, Manley HJ, Crawford T, et al. Undertreatment of anemia in patients with chronic kidney disease in the United States: analysis of national outpatient survey data. Clin Ther 2007; 29: 1524–1534
- London R, Solis A, Goldberg GA, et al. Health care resource utilization and the impact of anemia management in patients with chronic kidney disease. AmJ Kidney Dis 2002; 40: 539–548
- Kimmel PL, Greer JW, Milam RA, et al. Trends in erythropoietin therapy in the U.S. dialysis population: 1995 to 1998. Semin Nephrol 2000; 20: 335–344
- Pfeffer MA, Burdmann EA, Chen CY, et al. A trial of darbepoetin alfa in type 2 diabetes and chronic kidney disease. N Engl J Med 2009; 361: 2019–2032
- Maddux FW, Shetty S, del Aguila MA, et al. Effect of erythropoiesis-stimulating agents on healthcare utilization, costs, and outcomes in chronic kidney disease. Ann Pharmacother 2007; 41: 1761–1769
- Moyneur E, Bookhart BK, Mody SH, et al. The economic impact of pre-dialysis epoetin alpha on health care and work loss costs in chronic kidney disease: an employer's perspective. Dis Manag 2008; 11: 49–58
- Papatheofanis F, Bookhart BK, Muser E, et al. An examination of productivity and resource utilization associated with epoetin alfa treatment in employees with predialysis chronic kidney disease. J Occup Environ Med 2008; 50: 584–589
- Perkins R, Olson S, et al. Impact of an anemia clinic on emergency room visits and hospitalizations in patients with anemia of CKD pre-dialysis. Nephrology Nursing 2007; 34: 167–182
- Provenzano R, Garcia-Mayol L, Suchinda P, et al. Once-weekly epoetin alfa for treating the anemia of chronic kidney disease. Clin Nephrol 2004; 61: 392–405
- Alexander M, Kewalramani R, Agodoa I, et al. Association of anemia correction with health related quality of life in patients not on dialysis. Curr Med Res Opin 2007; 23: 2997–3008
- Cella D, Yount S, Sorensen M, et al. Validation of the functional assessment of chronic illness therapy fatigue scale relative to other instrumentation in patients with rheumatoid arthritis. J Rheumatol 2005; 32: 811–819
- Revicki DA. Relationship between health utility and psychometric health status measures. Med Care 1992; 30(5 Suppl)MS274–282
- Muirhead N, Cattran DC, Zaltzman J, et al. Safety and efficacy of recombinant human erythropoietin in correcting the anemia of patients with chronic renal allograft dysfunction. J Am Soc Nephrol 1994; 5: 1216–1222
- Singh AK, Szczech L, Tang KL, et al. Correction of anemia with epoetin alfa in chronic kidney disease. N Engl J Med 2006; 355: 2085–2098
- Abu-Alfa AK, Sloan L, Charytan C, et al. The association of darbepoetin alfa with hemoglobin and health-related quality of life in patients with chronic kidney disease not receiving dialysis. Curr Med Res Opin 2008; 24: 1091–1100
- Agarwal R, Rizkala AR, Bastani B, et al. A randomized controlled trial of oral versus intravenous iron in chronic kidney disease. Am J Nephrol 2006; 26: 445–454
- Drueke TB, Locatelli F, Clyne N, et al. Normalization of hemoglobin level in patients with chronic kidney disease and anemia. N Engl J Med 2006; 355: 2071–2084
- Benz R, Schmidt R, Kelly K, et al. Epoetin alfa once every 2 weeks is effective for initiation of treatment of anemia of chronic kidney disease. Clin J Am Soc Nephrol 2007; 2: 215–221
- Provenzano R, Bhaduri S, Singh AK. PROMPT Study Group. Extended epoetin alfa dosing as maintenance treatment for the anemia of chronic kidney disease: the PROMPT study. Clin Nephrol 2005; 64: 113–123
- Lefebvre P, Vekeman F, Sarokhan B, et al. Relationship between hemoglobin level and quality of life in anemic patients with chronic kidney disease receiving epoetin alfa. Curr Med Res Opin 2006; 22: 1929–1937
- Revicki DA, Brown R, et al. Health-related quality of life associated with recombinant human erythropoietin therapy for predialysis chronic renal disease patients. Am J Kidney Dis 1995; 25: 548–554
- Kleinman KS, Schweitzer SU, Perdue ST, et al. The use of recombinant human erythropoietin in the correction of anemia in predialysis patients and its effect on renal function: a double-blind, placebo-controlled trial. Am J Kidney Dis 1989; 14: 486–495
- Sobrero A, Puglisi F, Guglielmi A, et al. Fatigue: a main component of anemia symptomatology. Semin Oncol 2001; 28(2 Suppl 8)15–18
- Leaf DE, Goldfarb DS. Interpretation and review of health-related quality of life data in CKD patients receiving treatment for anemia. Kidney Int 2009; 75: 15–24
- Gandra SR, Finkelstein FO, Bennett AV, et al. Impact of erythropoiesis-stimulating agents on energy and physical function in nondialysis CKD patients with anemia: a systematic review. AmJ Kidney Dis 2010; 55: 519–534
- Unger EF, Thompson AM, Blank MJ, et al. Erythropoiesis-stimulating agents – time for a reevaluation. N Engl J Med 2010; 362: 189–192
- Odden MC, Whooley MA, Shlipak MG. Association of chronic kidney disease and anaemia with physical capacity: the heart and soul study. JAm Soc Nephrol 2004; 15: 2908–2915
- Lindgren C. New survey provides insight into CKD, exhaustion, anaemia. Nephrol News Issues 2003; 17: 29–31
- Ware JE, Kosinski M, Dewey JE. How to score Version Two of the SF-36 Health Survey. Lincoln, RI: Quality Metric, Incorporated, 2000.