Chronic Kidney Disease, Anemia, and the Association Between Chronic Kidney Disease-Related Anemia and Activities of Daily Living in Older Nursing Home Residents
Article Outline
Objectives
The aim of this study was to evaluate the potential association between anemia associated with chronic kidney disease (CKD) and the ability for older nursing home (NH) residents with these conditions to perform basic activities of daily living (ADL).
Design
This is a multicenter, retrospective, observational study utilizing data abstracted from medical charts, laboratory records, and Minimum Data Set (MDS) records from the 1-year period before the date of data abstraction.
Setting
24 NHs in the United States.
Participants
Older (≥65 years of age) NH residents with CKD (not receiving dialysis) who were not comatose and were alive as of the day of data abstraction were included in this analysis.
Measurements
Resident demographics, hemoglobin, and estimated glomerular filtration rate records, and ADL items were abstracted from the medical charts, laboratory records, and MDS. The prevalence of CKD and anemia associated with CKD was evaluated from laboratory records. The relationship between CKD, anemia, and ADLs was assessed through a repeated resampling (bootstrap) analysis.
Results
CKD was present in 45.7% of the residents. Of these, 60.5% had anemia. Residents with CKD and anemia were more likely to require limited or extensive assistance in performing ADLs such as bed mobility, transfer, walk in room, walk in corridor, locomotion on unit, dressing, and toilet use. Conversely, a greater proportion of residents with CKD but without anemia required no or slight supervision in performing these ADLs.
Conclusion
CKD and anemia associated with CKD are common in older NH residents. The data suggest that older NH residents with CKD and anemia require greater assistance in performing ADLs than residents with CKD alone. These data support the importance of evaluating the older NH population for CKD and anemia. As with all retrospective analyses, study limitations must be considered in the interpretation of these results.
Keywords: Chronic kidney disease, anemia, quality of life, nursing home
Kidney function often declines with age1, 2 and, as a result, chronic kidney disease (CKD) is common in the older population. Anemia is a frequent complication in older patients with CKD.3, 4, 5 The presence of anemia in the setting of CKD has been shown to be associated with increased morbidity, mortality, and hospitalization rates and contributes to increased weakness, fatigue, and frailty.6, 7 Anemia in CKD patients can also have a negative impact on quality of life (QOL) and is associated with a decrease in physical activity.8
CKD-associated anemia is treatable. CKD patients treated with erythropoiesis-stimulating agents (ESAs) have improved cardiovascular performance, slowed CKD progression, and reduced mortality rates.9, 10, 11, 12, 13, 14 In addition, recent reports indicate that CKD patients not on dialysis who were receiving ESA therapy have improved QOL measures.15, 16, 17 However, studies evaluating the benefits of ESA treatment on QOL parameters specifically in elderly CKD patients are lacking.
To fully appreciate the significance of QOL changes when elderly CKD patients are treated with ESAs, it is important to recognize the impact that CKD-related anemia has on QOL in this population. One way of assessing QOL is through the evaluation of changes in functionality. In the nursing home (NH) environment, as well as in the community, functionality is often described in terms of activities of daily living, or ADL, such as walking, dressing, eating, toilet use, and taking care of personal hygiene. The inability of the elderly to perform basic ADL can affect their independence. Additionally, in NHs, increased dependency in carrying out ADL is one of the most important factors in predicting the cost of NH care.18
ADL assessment in NHs is captured on the Minimum Data Set (MDS) form. The MDS is a standardized, primary screening and assessment tool used to describe the physical, medical, psychological, and social functioning of NH residents.19 Completion of the MDS is mandatory; a full assessment is performed when a resident enters a NH and a less comprehensive assessment is done every 90 days. Although the extensive data captured on the MDS is primarily used by professionals from multiple disciplines for assessing, planning, and improving the quality of care in NHs, a number of studies have validated the MDS as a tool for research purposes.19, 20, 21
The purpose of this study was to evaluate how ADLs may be affected by the presence of anemia in older NH residents with CKD. The analysis presented uses the information abstracted from the MDS and medical charts of older NH residents.
Methods
This was a multicenter, retrospective, observational study conducted in a sample of NHs in the United States with complete chart audits. Nursing homes were eligible to participate if they had electronic data storage, data transmission capabilities to a large pharmaceutical care provider for the 1-year period before data abstraction, and a resident population of 50 or greater. Following enumeration of all eligible NHs, sites were stratified to the Midwestern, Southern, Western, and Northeastern regions of the United States. An allocation ratio of 2:1 between the Midwestern and Southern regions relative to the Western and Northeastern regions was used to reflect the NH distribution reported by the 1999 National Nursing Home Survey.22 Facilities were then randomly selected from each stratum to participate in the study. Thirty NHs were identified to participate in this study; however, 24 gave their consent to participate (4 [17%] NHs in the Northeast, 8 [33%] NHs in the Midwest, 7 [29%] NHs in the South, and 5 [21%] in the West). For each resident in the analysis, medical chart records were merged with electronic Minimum Data Set (MDS; Minimum Data Sets 2.0, Centers for Medicare and Medicare Services) records and electronic prescription claim records.
This study was conducted according to the ethical principles identified in the Declaration of Helsinki statement, and the principles used were consistent with both the International Conference on Harmonization guidelines for good clinical practice and with applicable regulatory requirements. The study protocol was approved by an independent Institutional Review Board. HIPPA waiver of authorization of consent was granted as all patient-identifying parameters were removed from the data used in this study to protect the privacy of study participants.
Study Population and Data Collection
The study population consisted of all individuals 65 years or older who were active (not deceased as of the day of data abstraction), not comatose, and not receiving dialysis. Data for residents who met the inclusion criteria were abstracted from the medical chart and MDS records for the 1-year period before the chart abstraction date. Abstracted data included residents' demographics, laboratory hemoglobin (Hb) values, and serum creatinine (SCr) measurements; anemia status from the medical chart; and ADL items from the MDS.
CKD and anemia were assessed from SCr and Hb records, respectively. CKD was defined as an estimated glomerular filtration rate (eGFR) less than 60 mL/min/m2, corresponding to stages 1 (eGFR < 15 mL/min/m2), 2 (eGFR 15 to 29 mL/min/m2), and 3 (eGFR 30 to 59 mL/min/m2) CKD as defined by the National Kidney Foundation Kidney Disease Outcomes Quality Initiative (NKF K/DOQI) classification of kidney function.23 The first available laboratory SCr value recorded over the 1-year period was used to calculate the eGFR, as determined by the Modification of Diet and Renal Disease (MDRD) study equation.24 The prevalence of CKD was assessed from this eGFR. Anemia was defined as Hb less than 13 g/dL for men or Hb less than 12 g/dL for women as defined by the World Health Organization (WHO).25 All available Hb results were abstracted from the laboratory records for this study; however, the prevalence of anemia was determined based on the Hb value recorded after, but temporally closest to, the SCr record (almost 80% of Hb measurements were recorded on the same day as the SCr record). A resident was considered nonanemic if there was no indication of anemia in any of their laboratory or chart records throughout the 1-year period (Figure 1).
Fig. 1.
Glomerular filtration rate and anemia collection schema.
The association analysis was based on a subset of the resident population with CKD who met the inclusion criteria. Two study cohorts were identified; one group had a diagnosis of CKD and a confirmed indication of anemia (CKD anemic residents) and the other group had a diagnosis of CKD with no indication of anemia (CKD nonanemic residents). Confirmed indication of anemia was defined as a laboratory indication of anemia (based on the WHO definition as described above) at any time during the 1-year period and an anemia indication recorded in the chart within 15 days after the laboratory record (Figure 1).
Ten ADL items were abstracted from Section G of the MDS 2.0 electronic dataset. These included bed mobility, transfer, walking in room or corridor, locomotion, on or off unit, dressing, eating, toilet use, and personal hygiene (Table 1). In the MDS, these activities are scored on a 6-point scale ranging from 0 (total independence in carrying out an activity) to 4 (total dependence in carrying out an activity), and a score of 8 indicating that the specific item describing the ADL did not occur during the 7-day period. For the analysis, the MDS scores were re-categorized so that a score of 0 or 1 indicated no/slight supervision and a score of 2, 3, or 4 indicated limited/extensive assistance needed. A score of 8 indicated that the specific ADL item was not performed during a 7-day MDS collection period; residents with a score of 8 were not included in the bootstrap analysis described below. Aggregation of the ADL categories was necessary to maximize the accuracy of the ADL data. Previous analyses of the MDS ADL items indicate that the most accurate assessment differentiates between the broader MDS categories indicated by “independent” and “supervision” at one end of the range compared to “limited assistance,” “extensive assistance,” and “complete dependence” at the other end of the range.26
Table 1. Description of the 10 ADL Item Included From the MDS Form
| Item⁎ | Description |
|---|---|
| Bed mobility | A resident's ability to move and turn in bed |
| Transfer | A resident's ability to move between surfaces such as to/from bed, chair, wheelchair, and standing |
| Walk in room | A resident's ability to walk within his/her room |
| Walk in corridor | A resident's ability to walk in corridor or unit |
| Locomotion on unit | A resident's ability to move between locations in his/her room and corridor |
| Locomotion off unit | A resident's ability to move to/from off-unit locations or distant areas |
| Dressing | A resident's ability to put on/take off street clothing |
| Eating | A resident's ability to eat and drink |
| Toilet use | A resident's ability to use the toilet room (or commode, bedpan, urinal) and transfer on/off toilet |
| Personal hygiene | A resident's ability to maintain personal hygiene, including combing hair, brushing teeth, shaving, applying makeup, washing/drying face and hands (does not include bathing) |
⁎Items were scored on the basis of the most support provided during a 7-day activity observation period. Table modified from the Minimum Data Set (MDS) 2.0 form (form is accessible at http://www.cms.hhs.gov/NursingHomeQualityInits/Downloads/MDS20MDSAllForms.pdf). |
Analytic Approach
Descriptive statistics, including means, standard deviations (SDs), and maximum and minimum values were calculated and used to summarize continuous variables. Categorical variables were summarized using frequency counts and percentages. Chi-square test was done for statistical comparison between the age groups.
A repeated resampling procedure, the bootstrap method,27 was used to assess differences in ADL between CKD anemic residents and CKD nonanemic residents. The original dataset was randomly sampled by selecting and replacing an individual before selecting the next individual. Here, 1000 bootstrap replications were used for each of the 2 distinct populations in this study. This approach allows for statistical inference of data on hand (especially in small sample sizes), with an estimation of the mean and confidence intervals (CIs) for the study samples. For CKD anemic residents the bootstrap technique was applied to the MDS record with a date closest to the anemia indication. Bootstrap analysis was done using the ADL score from the one MDS record selected for each corresponding resident. For the CKD nonanemic residents, a randomly selected MDS was used for each bootstrap iteration. The ADL scores from the MDS were then averaged for all replications. SAS 9.1 (SAS Institute, Inc, Cary, NC) software was used for the bootstrap analysis.
Results
Prevalence of CKD and CKD-associated Anemia
A total of 2204 residents were identified in the NHs. A total of 1501 met the inclusion criteria and had records to estimate their GFR (Figure 2); 80.5% were White, 75.1% were female, and the mean (SD) age was 84.1 (8.1) years. A total of 45.7% (686/1501) of the residents had CKD (eGFR < 60 mL/min/m2) with mean eGFR of 43.3 (11.1) mL/min/1.73 m2 (Table 2). Nearly 48% (541/1126) female residents and 39% (145/375) of the male residents had CKD. Residents classified as having no CKD for purpose of this analysis (eGFR ≥ 60 mL/min/m2) had a mean (SD) eGFR of 85.5 (25.8) mL/min/1.73 m2. The prevalence of CKD increased with age, rising from 32.9% (69/210) in the 65 to 74 years age group, to 43.1% (222/515) in the 75 to 84 years of age group, and was highest with 50.9% (395/776) in the 85 years and older age group. The prevalence was statistically different among the 3 age groups. Of the residents with CKD, 60.5% (400/661) were anemic (Table 2). The prevalence of anemia in residents with CKD (51.0%; 400/784) was higher than the prevalence of anemia in residents without CKD (49.0%; 384/784). Unadjusted analysis showed that the prevalence and severity of anemia increased with decreasing eGFR (data not shown).
Fig. 2.
Resident disposition.
Table 2. Resident Demographics
| All Residents | ADL Analysis Subset | ||||
|---|---|---|---|---|---|
| CKD⁎ | No CKD⁎ | Total | CKDa Anemic† | CKD⁎ Nonanemic | |
| n (%) | 686 (45.7) | 815(54.3) | 1501 | 69 | 104 |
| Gender, n (%) | |||||
| Female | 541(78.8) | 585(71.7) | 1126(75.1) | 48(69.6) | 94(90.4) |
| Male | 145(21.1) | 230(28.2) | 375(24.9) | 21(30.4) | 10(9.6) |
| Race/Ethnicity, n (%) | |||||
| White | 568(82.8) | 641(78.6) | 1209(80.5) | 50(72.5) | 93(89.4) |
| Hispanic | 77(11.2) | 116(14.2) | 193(12.9) | 6(8.7) | 6(5.8) |
| Black | 37(5.4) | 54(6.6) | 91(6.1) | 13(18.8) | 4(3.85) |
| Asian/Pacific Islander | 4(0.5) | 4(0.5) | 8(0.5) | 0(0) | 1(1) |
| Age, years | |||||
| Mean (SD) | 85.5(7.8) | 83.0(8.2) | 84.1(8.1) | 84.2(8.3) | 85.8(6.4) |
| Age groups, n (%) | |||||
| 65–74 years | 69(10.0) | 141(17.3) | 210(14.0) | 7(10.1) | 6(5.8) |
| 75–84 years | 222(32.4) | 293(36.0) | 515(34.3) | 28(40.6) | 39(37.5) |
| ≥ 85 year | 395(57.6) | 381(46.7) | 776(51.7) | 34(49.3) | 59(56.7) |
| GFR level, mL/min/1.73 m2 | |||||
| Mean (SD) | 43.3(11.1) | 85.5(25.8) | 66.2(29.3) | 39.6(13.0) | 47.1(7.7) |
| Hb assessment | |||||
| N | 661 | 773 | 1434 | — | — |
| Anemic,‡ n (%) | 400(58.3) | 384(47.1) | 784(52.2) | — | — |
| Mean (SD), g/dL | 11.7(1.7) | 12.2(1.6) | 11.9(1.7) | — | — |
⁎CKD was defined as an eGFR < 60 mL/min/1.73 m2 and for the purpose of this analysis, having no CKD was defined as an eGFR ≥ 60 mL.min/1.73 m2. |
†Anemic residents for the ADL analysis subset had a laboratory diagnosis indicating a Hb value of <13 g/dL for men and <12 g/dL for women (based on the WHO criteria) and a chart indication of anemia within 15 days of the laboratory diagnosis. |
‡Anemic residents had a laboratory diagnosis indicating a Hb value of <13 g/dL for men and <12 g/dL for women (based on the WHO criteria). |
Relationship Between CKD-associated Anemia and ADL
As stated, the associative analysis used a conservative approach to identify anemia in a subset of residents with CKD. A total of 69 had CKD and anemia, and 104 had CKD without anemia (total of 173 residents) from the 1501 residents included in the study. The demographics of the 173 residents included in this analysis is shown in Table 2. Overall, the distribution for the demographics of both the CKD anemic and CKD nonanemic groups is similar to the total resident population. The majority in both the CKD anemic and CKD nonanemic groups were female (69.6% and 90.4%, respectively) and white (72.5% and 89.4%, respectively). The mean (SD) age was 84.2 (8.3) and 85.8 (6.4) years for the CKD anemic and CKD nonanemic groups, respectively. CKD anemic residents had a lower eGFR (mean [SD] of 39.6 [13.0] mL/min/1.73 m2 versus 47.1 [17.7] mL/min/1.73 m2, respectively) compared with CKD nonanemic residents.
The results of the bootstrap analysis showing the potential association between CKD-associated anemia and ADLs are shown in Fig. 3, Fig. 4. All residents included in this analysis required some level of assistance; however, the need for assistance was different between CKD anemic residents and CKD nonanemic residents. A greater percentage of CKD anemic residents than CKD nonanemic residents required limited/extensive assistance in performing most ADL (Figure 3). Conversely, a greater percentage of nonanemic CKD residents than anemic CKD residents required no/slight supervision in performing most ADL (Figure 4). The ADL that showed significant differences, as indicated by the nonoverlapping 95% CIs, between anemic CKD residents and nonanemic CKD residents were bed mobility, transfer, walk in room, walk in corridor, locomotion on unit, dressing, and toilet use. Locomotion off unit, eating, and personal hygiene did not show significant differences between anemic CKD residents and nonanemic CKD residents.
Fig. 3.
Activities of daily living results: Limited/extensive assistance needed.
Fig. 4.
Activities of daily living results: No/slight supervision needed.
Discussion
The results of this analysis indicate that the point prevalence of CKD and anemia in the presence of CKD in elderly NH residents in a nationally representative sample of NHs were 45.7% and 60.5%, respectively, suggesting that these conditions are common in this population. Additionally, the functionality data presented here suggest that CKD anemic residents may require significantly more assistance in carrying out basic daily activities such as moving, walking, dressing, and toilet use than CKD nonanemic residents.
CKD and CKD-associated anemia are prevalent in the elderly population. A recent report based on the Third National Health and Nutrition Examination Survey (NHANES) data (1999–2004) shows that 37.8% of noninstitutionalized US adults 70 years and older have moderately (eGFR 30 to 59 mL/min/1.73 m2, stage 3 CKD) or severely (eGFR 15 to 29 mL/min/m2, stage 4 CKD) decreased kidney function.28 In the NH population, 43.1% of residents 65 years or older have CKD (eGFR < 60 mL/min/1.73 m2).29 Similarly, a study that used a large national database found that nearly 50% of individuals with CKD (without end-stage renal disease) who were 67 years and older had anemia.5 In the NH population, 64.9% of residents with CKD were reported as having anemia.29 In the current report 45.7% had CKD and 60.5% of those who had CKD had anemia, which closely reflects the results by Robinson et al.29 Unlike the Robinson et al study, however, this study was specifically designed with the intention of having a sample of NHs that was distributed in the same fashion as those found in Jones et al22 (ie, an allocation ratio of 2:1 between the Midwestern and Southern regions relative to the Western and Northeastern regions). When applied to the national NH population reported by Jones et al,22 approximately 671,600 older (≥65 years) NH residents have CKD (eGFR < 60 mL/min/m2). Of these, 406, 300 have anemia. The prevalence of CKD (including perhaps the prevalence of CKD-associated anemia) reported in this study may even be an underestimation since 22.4% (439/1956) of the NH residents in this study who had medical charts and who met the inclusion criteria did not have SCr records to estimate their GFR. Additionally, an eGFR of 85.5 mL/min/m2, as seen for residents classified as having no CKD, is indicative of mildly decreased kidney function (defined by the NKF/KDOQI guidelines as a GFR of 60 to 89 mL/min/1.73 m2, stage 2 CKD), so many in this group may indeed have CKD.
The results of this analysis suggest that older CKD anemic NH residents may require significantly more assistance in carrying out basic daily activities such as moving, walking, dressing, and toilet use than CKD nonanemic residents. In general, the 173 elderly residents with CKD included in this analysis required some level of assistance to perform ADL as indicated in Section G of the MDS form. However, a greater need for assistance as seen in the CKD anemic older resident population, can translate to poorer QOL and loss in autonomy for these residents. In addition, this can exact a heavier toll on caregivers (in terms of time and work load), and require more resources, including increased costs in care.
The relationship between CKD-anemia and ADL presented in this report relied on the identification of 2 distinct nonoverlapping populations: a group that had anemia and a group that did not have anemia. A rigorous methodology was therefore applied to identify anemia in residents with CKD. The presence of anemia in a resident was confirmed using information from 2 sources: the chart record indication and the laboratory diagnosis. Individuals were considered nonanemic if there were no indication of anemia in any of their records. This rigorous methodology resulted in a small sample size (69 CKD anemic residents and 104 CKD nonanemic residents), however, it significantly improved the confidence of selecting the appropriate subjects for the bootstrap analysis. Increasing the 15-day window between laboratory and chart records to 30 days did not significantly increase the number of CKD anemic residents who could be included in the analysis.
The CKD anemic cohort in this study in general had worsened kidney function (mean eGFR of 39.6 mL/min/1.73 m2) as compared with CKD nonanemic residents (mean of 47.1 mL/min/1.73 m2). It is therefore possible that worsening kidney disease may contribute to the increased dependency seen in this analysis. However, others have shown that Hb levels in CKD patients, rather than the degree of kidney disease, correlated more strongly with QOL indicators.30 This observation is supported by studies that used ESAs to treat anemia in CKD patients; CKD patients treated for their anemia had improved QOL indicators.17, 31
This study has several limitations. This was a retrospective study where the analyses were limited by data that were available. Missing SCr and Hb values or gaps in records disqualified most residents for the purposes of this analysis. Secondly, the ADL parameters which were used as performance and outcome variables were recorded by different caregivers and are open to subjective interpretation. Third, a cause-and-effect relationship between CKD, anemia, and physical function assessments could not be implied conclusively, in part because the data lacked temporal relationship; SCr, Hb, and MDS records were collected at different times. Fourth, given that this was an exploratory analysis, the conclusions described are based on limited power to detect differences in ADL in a small sample size. Even so, the bootstrap methodology is effective in allowing for inference using small sample sizes. An important limitation is that residual confounding effects (the effects of age, gender, severity of disease, dementia, stroke, depression, malnutrition, frailty, and other diseases or conditions) can also have an effect on ADLs. The limitations caused by sample size and some of these confounding effects, however, may partly be overcome by using larger studies controlling for confounders and comorbidities. Even with the limitations stated above, this study provides insight into the possible effect CKD-associated anemia has on ADL and encourages further research to assess this relationship and understand its impact on elderly NH residents. The exploratory analysis presented here does not imply that treatment with ESAs reverses this association, although this is a plausible possibility that begs to be explored.
Conclusion
This analysis corroborates earlier findings that CKD and anemia in the presence of CKD are common among older residents in NHs. The study further suggests that there is a possible relationship between anemia in the setting of CKD and ADLs in older NH residents. The effects were characterized by the increased assistance to carry out basic functions, such as walking, dressing, and toilet use, by CKD anemic versus CKD nonanemic elderly NH residents.
References
- . Longitudinal studies on the rate of decline in renal function with age. J Am Geriatr Soc. 1985;33:278–285
- The effect of age on creatinine clearance in men: A cross-sectional and longitudinal study. J Gerontol. 1976;31:155–163
- Association of kidney function with anemia: the Third National Health and Nutrition Examination Survey (1988–1994). Arch Intern Med. 2002;162:1401–1408
- The prevalence of anemia in patients with chronic kidney disease. Curr Med Res Opin. 2004;20:1501–1510
- . Anemia and cardiovascular disease, hospitalization, end stage renal disease, and death in older patients with chronic kidney disease. Int Urol Nephrol. 2005;37:395–402
- . Predictors of hospitalization and death among pre-dialysis patients: a retrospective cohort study. Nephrol Dial Transplant. 2000;15:650–658
- Haemoglobin at time of referral prior to dialysis predicts survival: An association of haemoglobin with long-term outcomes. Nephrol Dial Transplant. 2006;21:370–377
- . Quality of life in predialytic uremic patients. Qual Life Res. 1996;5:109–116
- . Treating anemia early in renal failure patients slows the decline of renal function: A randomized controlled trial. Kidney Int. 2004;66:753–760
- Cardiovascular effect of normalizing the hematocrit level during erythropoietin therapy in predialysis patients with chronic renal failure. Am J Kidney Dis. 2000;35:250–256
- Beneficial influence of recombinant human erythropoietin therapy on the rate of progression of chronic renal failure in predialysis patients. Nephrol Dial Transplant. 2001;16:307–312
- Canadian randomized trial of hemoglobin maintenance to prevent or delay left ventricular mass growth in patients with CKD. Am J Kidney Dis. 2005;46:799–811
- Correction of anemia with erythropoietin in chronic kidney disease (stage 3 or 4): Effects on cardiac performance. Cardiovasc Drugs Ther. 2008;22:37–44
- Cardiovascular effects of recombinant human erythropoietin in predialysis patients. Am J Kidney Dis. 1997;29:541–548
- 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
- . Association of anemia correction with health related quality of life in patients not on dialysis. Curr Med Res Opin. 2007;23:2997–3008
- 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
- . Trends in nursing home expenses, 1987 and 1996. Health Care Financ Rev. 2003;25:99–114
- Reliability estimates for the Minimum Data Set for nursing home resident assessment and care screening (MDS). Gerontologist. 1995;35:172–178
- . A comprehensive clinical assessment tool to inform policy and practice: Applications of the minimum data set. Med Care. 2004;42:III50–III59
- Validity and responsiveness of the Minimum Data Set. J Am Geriatr Soc. 1999;47:1000–1004
- . The National Nursing Home Survey: 1999 summary. Vital Health Stat. 2002;13:1–116
- . K/DOQI Clinical Practice Guidelines for chronic kidney disease: Evaluation, classification, and stratification. Am J Kidney Dis. 2002;39:S1–S266
- . Prediction equations to estimate glomerular filtration rate: an update. Curr Opin Nephrol Hypertens. 2001;10:785–792
- . Nutritional anemia: Report of a WHO scientific group. Technical Report Series, No. 405. Geneva, Switzerland: WHO; 1968;
- . Evaluating the accuracy of minimum data set bed-mobility ratings against independent performance assessments: Systematic error and directions for improvement. Gerontologist. 2005;45:731–738
- . An Introduction to the Bootstrap. New York: Chapman & Hall; 1993;
- Prevalence of chronic kidney disease in the United States. JAMA. 2007;298:2038–2047
- Prevalence of anemia in the nursing home: Contribution of chronic kidney disease. J Am Geriatr Soc. 2007;55:1566–1570
- Quality of life in chronic kidney disease (CKD): A cross-sectional analysis in the Renal Research Institute-CKD study. Am J Kidney Dis. 2005;45:658–666
- Chronic anemia and fatigue in elderly patients: Results of a randomized, double-blind, placebo-controlled, crossover exploratory study with epoetin alfa. J Am Geriatr Soc. 2007;55:1557–1565
This study was supported by Amgen Inc, Thousand Oaks, CA. John Schnelle, PhD, has served as a consultant for Amgen Inc. Dan Osterweil, MD, has served as a consultant and on the speaker bureau for Amgen Inc and Novartis, as a consultant for SCAN Health Plan, and on the speaker bureau for Forest Laboratories. Denise Globe, PhD, Angela Sciarra, PharmD, and Arie Barlev, PharmD are employees of Amgen Inc. Paul Audhya, MD, was previously employed at Amgen Inc. We would like to thank Yeshi Mikyas, PhD, (Amgen Inc) for providing assistance in writing this manuscript.
PII: S1525-8610(08)00294-6
doi:10.1016/j.jamda.2008.08.012
© 2009 American Medical Directors Association. Published by Elsevier Inc. All rights reserved.
