Abstract
Rationale and Objective:
Anemia is common in chronic kidney disease (CKD) and, although associated with adverse outcomes, the available treatments are not ideal. We aimed to characterize the burden, risk factors for and risks associated with anemia by estimated glomerular filtration rate (eGFR) and hemoglobin level.
Study Design:
Cross-sectional and prospective cohort study.
Setting and Participants:
Outpatient data from 5,004,957 individuals across 57 healthcare centers in the US from 2016 to 2019, extracted from the Optum Labs Data Warehouse.
Exposures:
Severity of anemia, presence of low iron test results, eGFR.
Outcomes:
Incident end-stage kidney disease, cardiovascular disease, coronary heart disease, stroke, heart failure, death.
Analytical Approach:
The prevalences of anemia, low iron test results, vitamin B12 deficiency, and erythropoiesis-stimulating agent (ESA) use, stratified by sex and eGFR, were characterized. Polychotomous logistic regression was used to estimate adjusted odds ratios of different hemoglobin levels across eGFR. Cox proportional hazards regression was used to calculate adjusted hazard ratios for adverse outcomes across hemoglobin level.
Results:
The mean age was 54 years and 42% were male. Lower eGFR was very strongly associated with increased prevalence of anemia, even after adjustment. Although iron studies were checked infrequently in anemic patients, low iron test results were highly prevalent in those tested; 60.4% and 81.3% of men and women, respectively. Erythropoiesis-stimulating agent use was uncommon with a prevalence of use <4%. Lower hemoglobin was independently associated with increased risk of end-stage kidney disease, cardiovascular disease, coronary heart disease, stroke, heart failure, and death.
Limitations:
Reliance on ICD codes for medical diagnoses, death information obtained from claims data, observational study.
Conclusions:
Severe anemia was common and strongly associated with lower eGFR and multiple adverse outcomes. Low iron test results were highly prevalent in those tested despite iron studies being checked infrequently. ESA use in non-dialysis CKD patients was uncommon.
INDEX WORDS: Anemia, Chronic kidney disease, eGFR, Low iron test results, Iron deficiency, Erythropoiesis-stimulating agent, Erythropoietin, Adverse outcomes
Graphical Abstract
PLAIN LANGUAGE SUMMARY
Anemia, a common complication of chronic kidney disease, is associated with adverse outcomes. Unfortunately, the use of erythropoiesis-stimulating agents (ESAs) to target a normal hemoglobin level has been associated with increased cardiovascular risk. Our large study of over 5 million patients sought to describe the burden and risk factors associated with anemia by estimated glomerular filtration rate (eGFR) level. We found that severe anemia was common and strongly associated with lower eGFR and multiple adverse outcomes. ESA use was rare. Although iron studies were checked infrequently, low iron test results were common in those tested. This highlights the need for increased testing of iron studies in patients with anemia, as iron supplementation is an effective and low-risk intervention.
INTRODUCTION
Anemia, defined by the World Health Organization as a hemoglobin concentration <13g/dL in men and <12g/dL in women, is common among patients with chronic kidney disease (CKD)1–5. The prevalence of anemia increases as estimated glomerular filtration rate (eGFR) declines, likely through the development of erythropoietin deficiency and resistance and a heightened inflammatory state2–7. Characterizing the burden and risk factors associated with severe anemia, defined as a hemoglobin <10g/dL, is challenging since accurate estimation of the prevalence of both severe anemia and very low eGFR requires large study populations. People with anemia are at a greatly increased risk for adverse cardiovascular outcomes and death8–11. Although erythropoiesis-stimulating agents (ESAs) were a ground-breaking therapy that mitigated the need for blood transfusions, their use has been associated with an increased risk of cardiovascular disease. Multiple trials have shown ESA use targeting normal hemoglobin levels to be associated with an increased risk of myocardial infarction, stroke and death in CKD and ESKD (end-stage kidney disease) populations12–19. Treatment targets have thus been scaled back to minimize risk, and ESAs are generally not considered except in severe anemia20.
Previous studies addressing the burden of anemia by level of eGFR were based on older data and much smaller sample sizes2,3,21. Iron deficiency is a common cause of anemia and, although it is recommended that all CKD patients with anemia be screened and treated for iron deficiency20, it is unclear how often this is occurring in practice. There are also very limited data available on the prevalence of vitamin B12 deficiency, another easily treatable cause of anemia, in those with CKD. In addition, the relative frequency of ESA use in the non-dialysis CKD population is uncertain.
In our study of over 5.0 million patients receiving routine medical care in the US, we sought to describe the burden of anemia by level of eGFR. This is the largest study to date on this topic and is one of the only studies detailing the associations with severe anemia. We aimed to characterize the frequency of screening for low iron test results and vitamin B12 deficiency, the presence of both low iron test results and vitamin B12 deficiency by level of eGFR, and the receipt of ESAs in anemia. We also evaluated the risk of adverse outcomes associated with different levels of hemoglobin and eGFR, and whether these risks varied by the presence of low iron test results. We hypothesized that severe anemia is strongly associated with low eGFR, iron deficiency is common when looked for but iron studies are often not conducted.
METHODS
Our data were extracted from the Optum Labs Data Warehouse, which contains de-identified claims and electronic health record data from a wide range of healthcare centers across the US22. We included patients with an outpatient complete blood count and serum creatinine measured within 30 days of each other. In order to allow for sufficient follow-up for adverse outcomes, we required that the complete blood count be drawn within the 2016 calendar year. Baseline values were selected as the closest measurements of hemoglobin and creatinine. If there were multiple same day measurements of hemoglobin and creatinine, the earliest measurements were selected as the baseline. ESA use was ascertained as any documented use within 2016, the same calendar year as hemoglobin measurement. Any use of epoetin alfa, epoetin beta or darbepoetin alfa was included. Patients requiring dialysis and patients with cancer were excluded.
Anemia was defined based on the hemoglobin concentration as <13g/dL in men and <12g/dL in women1. Hemoglobin <10 g/dL defined severe anemia20, and sex-specific analyses further categorized hemoglobin, measured in g/dL, as: <9, 9–10, 10–11, 11–12, 12–13, and 13+. CKD-EPI 2009 eGFR from creatinine was also divided into 7 categories, in mL/min; <15, 15–29, 30–44, 45–59, 60–74, 75–89 and 90+23. Vitamin B12 deficiency was defined as a serum vitamin B12 <200pg/mL24. Low iron test results were defined as a serum ferritin </= 100ng/mL and/or a serum transferrin saturation </= 20%25–28. The same ferritin cut offs were used for all patients, including those with a normal eGFR, in order to increase the sensitivity for capturing those with functional iron deficiency who may still benefit from iron supplementation. Serum ferritin is elevated in inflammatory states irrespective of iron stores29, and is notoriously inaccurate for diagnosing iron deficiency28,30. A mean corpuscular volume (MCV) between 80 and 100 fL was defined as normal31.
Follow-up data were obtained on patients up until December 31, 2019. Adverse outcomes ascertained included ESKD, cardiovascular disease, coronary heart disease, stroke, heart failure and death, as indicated by ICD codes. Cardiovascular disease was defined as either coronary heart disease, stroke or heart failure; whichever occurred first. The study was approved by the IRB of Johns Hopkins Bloomberg School of Public Health, approval number IRB00003324, as exempt research without identifiers, and was conducted in accordance with the provisions of the Declaration of Helsinki. Individual-level informed consent was not obtained for data secondary analysis of an established database with privacy protection.
Statistical Methods
Baseline characteristics were stratified by sex and pooled across centers. MCV, frequency of iron study and vitamin B12 level test utilization, and frequencies of low iron test results and vitamin B12 deficiency were tabulated in patients with and without anemia, and were stratified by sex and eGFR category. The prevalence of ESA use was stratified by sex, hemoglobin level, and eGFR category. The prevalence of anemia was stratified by sex, hemoglobin level, and eGFR category. Polychotomous logistic regression was used to estimate odds ratios and adjusted prevalence of different levels of hemoglobin across eGFR, stratified by sex. Cox proportional hazards regression was used to calculate adjusted hazard ratios for the outcomes of ESKD, cardiovascular disease, stroke, heart failure and death across different levels of hemoglobin, eGFR, and the presence and absence of low iron test results, also stratified by sex. The model was adjusted for age, race, eGFR, prevalent cardiovascular disease, hypertension, urine albumin-creatinine ratio, diabetes mellitus, smoking, and healthcare organization. Race was used as an adjustment variable since it is related to both CKD and anemia. Race was not a primary outcome or exposure in this manuscript. The race categories were assigned by the health systems providing data to the Optum Labs Data Warehouse and likely reflect self-reported race with missing values coded as other/unknown22. The Cox proportional hazard analysis included patients with anemia only to allow a uniform interpretation of testing for low iron test results. In men, individuals with a hemoglobin of 12–13g/dL were used as the reference, and those with a hemoglobin of 13+g/dL were excluded from the analysis. In women, individuals with a hemoglobin of 11–12g/dL were used as the reference, and those with a hemoglobin of 12+g/dL were excluded from the analysis. Censoring was defined by the date of last interaction with the health system.
Data Availability
The data underlying the results of this study are third party data owned by OptumLabs and contain sensitive patient information; therefore, the data is only available upon request. Interested researchers engaged in HIPAA compliant research may contact connected@optum.com for data access requests. The data use requires researchers to pay for rights to use and access the data.
RESULTS
Baseline Characteristics
Overall, 5,004,957 individuals across 57 healthcare organizations were included in our analysis, and the study population was 42% male (Table 1). The mean age was 54 years and 10.1% of individuals were Black. Mean hemoglobin was 14 g/dL, and mean eGFR was 87 mL/min/1.73m2. Only 8.3% of individuals had a urine albumin-creatinine ratio available, but 35.6% of individuals had a urine dipstick, of which over 19% had at least trace proteinuria. These characteristics were relatively similar among men and women. Overall, 14.9% of participants had diabetes mellitus, and 39.8% had hypertension.
Table 1:
Baseline Characteristics
| Overall | Men | Women | |
|---|---|---|---|
| N | 5,004,957 | 2,088,371 | 2,916,586 |
| Mean age (SD), years | 54 (17) | 54 (16) | 54 (18) |
| Black, % | 10.1 | 8.5 | 11.2 |
| Mean hemoglobin, g/dL (SD) | 14 (2) | 15 (2) | 13 (1) |
| Mean eGFR, mL/min/1.73m2 (SD) | 87 (23) | 86 (22) | 88 (24) |
| n, ACR (%) | 417,117 (8.3) | 198,020 (9.5) | 219,097 (7.5) |
| n, dipstick (%) | 1,781,586 (35.6) | 711,411 (34.1) | 1,070,175 (36.7) |
| Trace proteinuria, % | 7.1 | 6.9 | 7.3 |
| + proteinuria, % | 7.7 | 7.7 | 7.6 |
| ++ proteinuria, % | 3.5 | 3.8 | 3.4 |
| >++ proteinuria, % | 1.0 | 1.1 | 0.9 |
| Diabetes mellitus, % | 14.9 | 17.0 | 13.4 |
| Hypertension, % | 39.8 | 43.3 | 37.3 |
| Cardiovascular disease, % | 14.0 | 17.2 | 11.7 |
| Current smoker, % | 5.5 | 5.7 | 5.5 |
| Former smoker, % | 11.8 | 12.9 | 10.9 |
ACR, albumin-creatinine ratio; eGFR, estimated glomerular filtration rate; SD, standard deviation
Anemia Prevalence
The prevalence of severe anemia (hemoglobin <10 g/dL) in men was 1.3%, 3.1%, 7.5%, 17.4% and 29.7% across eGFR categories of 60–74, 45–59, 30–44, 15–29 and <15 mL/min/1.73m2, respectively. In women, the corresponding prevalence of severe anemia was 1.9%, 3.9%, 8.6%, 19.4% and 37.6% (Table S1). Figure 1 illustrates that lower eGFR was very strongly associated with increased prevalence of anemia. This was true for all categories of hemoglobin. A similar pattern was seen in men and women; however, anemia was more prevalent in women compared to men across all categories of eGFR. For example, a hemoglobin <9 g/dL was present in 18.8% of women and 15.2% of men with an eGFR <15 ml/min/1.73m2.
Figure 1:
Prevalence of anemia by eGFR category, stratified by sex
Figure 2 shows that the prevalence of different categories of anemia remained very strongly associated with lower eGFR after adjustment for age, race and healthcare organization. The logit scale (log (p/1-p)) allows for visualization of the marked variation in anemia prevalence from 0.1% to >50% and demonstrates that adjusted prevalence increased more sharply with lower eGFR for the lower hemoglobin categories, such as <9 g/dL, as compared to the higher hemoglobin categories, such as 12–13 g/dL. A steeper slope indicates a higher odds ratio of prevalent anemia per unit change in eGFR. A similar pattern was seen in both men and women. The full model allows for calculation of predicted prevalence for a given eGFR, sex, age and race (coefficients are listed in Table S2). Older age and Black race were associated with higher prevalence of anemia. Extremes of higher eGFR were also associated with an increased prevalence of anemia.
Figure 2: Adjusted prevalence of different levels of hemoglobin across eGFR, stratified by sex.
Ordinate is on the logit scale to allow visualization across the full range of prevalence with slopes indicating the strength of association on the log odds ratio scale. Adjusted for age, race (black and non-black), healthcare organization.
Iron Studies, Vitamin B12 Levels and ESA Usage in Anemia
Table 2 illustrates that iron studies and vitamin B12 levels were checked infrequently in patients with anemia. Of all men with anemia, only 15.6% and 11.7% had iron studies and vitamin B12 levels available, respectively. The respective percentages in women were 19.6% and 13.9%. Among those with iron studies, low iron test results were highly prevalent; 60.4% (including 50.0% with low transferrin saturation and 49.5% with low ferritin) and 81.3% (including 69.9% with low transferrin saturation and 76.4% with low ferritin) of men and women with anemia, respectively. In both men and women with anemia, the prevalence of low iron test results decreased with lower eGFR. Low iron test results were more common in women than men. For example, in those with anemia and an eGFR of 90+ mL/min/1.73m2, 64.9% of men and 90.2% of women had low iron test results. By comparison, in those with anemia and an eGFR of <15 mL/min/1.73m2, only 40.1% of men and 50.5% of women had low iron test results. Vitamin B12 deficiency was rare across all categories of eGFR, with a total prevalence of 3.0% in men with anemia and 3.5% in women with anemia. Finally, the average MCV values were normal across all categories of eGFR, indicating that anemia was most frequently normocytic. The prevalence of low iron test results in those with normal kidney function (eGFR >60mL/min.1.73m2) using an alternate ferritin cut-off for the general population of </=30ng/mL30 is shown in Table S3. Table 3 depicts that even in severe anemia, defined as hemoglobin <10g/dL, ESA use was extremely uncommon across all categories of eGFR, with a prevalence of use <4%.
Table 2:
Frequency of iron study and vitamin B12 level utilization, mean MCV values, and prevalence of low iron test results and B12 deficiency across eGFR in patients with anemia*, stratified by sex. Low iron test results were defined as a serum ferritin </= 100ng/mL and/or a serum transferrin saturation </= 20%. Vitamin B12 deficiency was defined as a serum vitamin B12 <200pg/mL.
| N | Mean MCV (SD) | Iron study by ferritin and/or TSAT, n (%†) | Low iron test results by ferritin and/or TSAT, n (%††) | Iron study by TSAT, n (%†) | Low iron test results by TSAT, n (%§) | Iron study by ferritin, n (%†) | Low iron test results by ferritin, n (%‖) | Vitamin B12 study, n (%†) | Vitamin B12 deficiency, n (%¶) | ||
|---|---|---|---|---|---|---|---|---|---|---|---|
| eGFR, mL/min/1.73m 2 | |||||||||||
| Men | Total | 243,395 | 89 (8) | 37,899 (15.6) | 22,904 (60.4) | 30,131 (12.4) | 15,059 (50.0) | 31,200 (12.8) | 15,437 (49.5) | 28,366 (11.7) | 844 (3.0) |
| 90+ | 63,691 | 87 (9) | 9,243 (14.5) | 5,996 (64.9) | 7,020 (11.0) | 4,015 (57.2) | 7,757 (12.2) | 4,199 (54.1) | 6,818 (10.7) | 238 (3.5) | |
| 75–89 | 45,642 | 89 (9) | 6,441 (14.1) | 4,168 (64.7) | 4,999 (11.0) | 2,666 (53.3) | 5,297 (11.6) | 2,966 (56.0) | 5,563 (12.2) | 163 (2.9) | |
| 60–74 | 41,649 | 90 (8) | 6,154 (14.8) | 3,908 (63.5) | 4,868 (11.7) | 2,497 (51.3) | 4,937 (11.9) | 2,748 (55.7) | 5,276 (12.7) | 157 (3.0) | |
| 45–59 | 39,754 | 91 (8) | 6,067 (15.3) | 3,599 (59.3) | 4,827 (12.1) | 2,317 (48.0) | 4,905 (12.3) | 2,438 (49.7) | 5,028 (12.7) | 153 (3.0) | |
| 30–44 | 30,044 | 91 (7) | 5,270 (17.5) | 2,925 (55.5) | 4,318 (14.4) | 1,893 (43.8) | 4,284 (14.3) | 1,872 (43.7) | 3,528 (11.7) | 106 (3.0) | |
| 15–29 | 16,174 | 91 (7) | 3,706 (22.9) | 1,900 (51.3) | 3,209 (19.8) | 1,346 (41.9) | 3,130 (19.4) | 1,060 (33.9) | 1,741 (10.8) | >16 (>0.9) | |
| <15 | 6,411 | 92 (7) | 1,018 (15.8) | 408 (40.1) | 890 (13.8) | 325 (36.5) | 890 (13.8) | 154 (17.3) | 412 (6.4) | <11 (<2.7) | |
| eGFR, mL/min/1.73m 2 | |||||||||||
| Women | Total | 468,071 | 86 (9) | 91,737 (19.6) | 74,570 (81.3) | 71,298 (15.2) | 49,803 (69.9) | 74,756 (16.0) | 57,077 (76.4) | 64,939 (13.9) | 2,295 (3.5) |
| 90+ | 215,096 | 83 (9) | 42,055 (19.6) | 37,926 (90.2) | 31,835 (14.8) | 25,081 (78.8) | 34,359 (16.0) | 30,584 (89.0) | 27,911 (13.0) | 1,174 (4.2) | |
| 75–89 | 71,921 | 86 (9) | 13,626 (19.0) | 11,363 (83.4) | 10,484 (14.6) | 7,563 (72.1) | 11,074 (15.4) | 8,824 (79.7) | 10,613 (14.8) | 362 (3.4) | |
| 60–74 | 57,113 | 88 (8) | 10,480 (18.4) | 8,100 (77.3) | 8,140 (14.3) | 5,386 (66.2) | 8,399 (14.7) | 6,048 (72.0) | 8,583 (15.0) | 257 (3.0) | |
| 45–59 | 53,136 | 89 (8) | 9,812 (18.5) | 7,146 (72.8) | 7,705 (14.5) | 4,771 (61.9) | 7,894 (14.9) | 5,156 (65.3) | 8,057 (15.2) | 223 (2.8) | |
| 30–44 | 42,967 | 91 (7) | 8,851 (20.6) | 6,000 (67.8) | 7,184 (16.7) | 4,098 (57.0) | 7,176 (16.7) | 4,073 (56.8) | 6,367 (14.8) | 199 (3.1) | |
| 15–29 | 21,992 | 91 (7) | 5,589 (25.4) | 3,366 (60.2) | 4,785 (21.8) | 2,404 (50.2) | 4,704 (21.4) | 2,049 (43.6) | 2,885 (13.1) | >69 (>2.4) | |
| <15 | 5,846 | 92 (7) | 1,324 (22.7) | 669 (50.5) | 1,165 (19.9) | 500 (42.9) | 1,150 (19.7) | 343 (29.8) | 523 (9.0) | <11 (<2.1) | |
eGFR, estimated glomerular filtration rate; MCV, mean corpuscular volume; TSAT, transferrin saturation
Anemia was defined as Hb <13g/dL in men and Hb <12g/dL in women
Denominator = N
Denominator = Iron study by ferritin and/or TSAT, n
Denominator = Iron study by TSAT, n
Denominator = Iron study by ferritin, n
Denominator = Vitamin B12 level, n
Table 3:
Frequency of ESA use across eGFR, sex, and different levels of hemoglobin in anemia
| Percent ESA use (%) | ||||||
|---|---|---|---|---|---|---|
| Hemoglob in 10–11.9 g/dL in women or 10–12.9 g/dL in men | Hemoglob in <10 g/dL | Hemoglob in 9–10 g/dL | Hemoglob in <9 g/dL | Men with hemoglob in <10 g/dL | Women with hemoglob in <10 g/dL | |
| eGFR, mL/min/1.73 m 2 | ||||||
| Total | 0.12 | 0.80 | 0.72 | 0.91 | 1.31 | 0.60 |
| 60+ | 0.02 | 0.16 | 0.12 | 0.20 | 0.35 | 0.10 |
| 30–59 | 0.18 | 0.93 | 0.81 | 1.09 | 1.14 | 0.81 |
| <30 | 1.17 | 3.53 | 3.31 | 3.80 | 4.15 | 3.13 |
eGFR, estimated glomerular filtration rate, ESA: erythropoiesis-stimulating agent
Anemia and Risk of Adverse Outcomes
Table 4 shows that lower hemoglobin was associated with an increased risk of subsequent ESKD, cardiovascular disease, coronary heart disease, stroke, heart failure, and death after adjustment for age, race, eGFR, prevalent cardiovascular disease, hypertension, urine albumin-creatinine ratio, diabetes mellitus, smoking and healthcare organization. For example, the adjusted hazard ratio for death was 2.91 (95% CI 2.82–3.00) in men with hemoglobin <9g/dL as compared to men with hemoglobin 12–13g/dL. Interestingly, the presence of low iron test results in anemia was generally associated with a decreased risk of adverse outcomes as compared to the absence of low iron test results in those tested. For example, in women with anemia, the adjusted hazard ratio for end-stage kidney disease was 0.76 (95% CI 0.69, 0.83) in those with low iron test results as compared to those without low iron test results. Similarly, low iron test results were associated with a decreased risk of cardiovascular disease, coronary heart disease, stroke, heart failure, and death in women. In contrast, low iron test results were only associated with a decreased risk of end-stage kidney disease, stroke, and death in men with anemia, and was associated with a marginally increased risk of heart failure. The relative hazards associated with lower hemoglobin remained statistically significant across different eGFR strata with stronger hazards at higher eGFR (60+ ml/min/1.73m2) and weaker at lower eGFR, although absolute risks are higher at lower eGFR (Table S4). Event rates and average follow-up times are documented in Table S5.
Table 4:
Adjusted hazard ratios for ESKD, CVD, CHD, stroke, HF and death among patients with anemia* across different levels of hemoglobin and presence of low iron test results, stratified by sex
| Hazard Ratio (95% CI) | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Men | Women | |||||||||||
| Hemoglobin, g/dL | ESKD | CVD | CHD | Stroke | HF | Death | ESKD | CVD | CHD | Stroke | HF | Death |
| 12–13 | ref | ref | ref | ref | ref | ref | ||||||
| 11–12 | 1.23 (1.15, 1.32) | 1.29 (1.25, 1.32) | 1.25 (1.19, 1.31) | 1.16 (1.09, 1.23) | 1.33 (1.29, 1.37) | 1.43 (1.40, 1.47) | ref | ref | ref | ref | ref | ref |
| 10–11 | 1.42 (1.32, 1.52) | 1.58 (1.53, 1.62) | 1.46 (1.37, 1.55) | 1.30 (1.21, 1.39) | 1.64 (1.59, 1.70) | 1.93 (1.88, 1.98) | 1.26 (1.17, 1.34) | 1.30 (1.27, 1.33) | 1.26 (1.20, 1.32) | 1.17 (1.11, 1.23) | 1.33 (1.30, 1.37) | 1.45 (1.42, 1.48) |
| 9–10 | 1.50 (1.38, 1.64) | 1.82 (1.75, 1.88) | 1.60 (1.49, 1.72) | 1.53 (1.40, 1.67) | 1.92 (1.85, 2.00) | 2.34 (2.27, 2.42) | 1.45 (1.35, 1.57) | 1.64 (1.59, 1.69) | 1.52 (1.43, 1.62) | 1.41 (1.32, 1.50) | 1.70 (1.64, 1.75) | 1.96 (1.91, 2.01) |
| <9 | 1.44 (1.31, 1.58) | 2.09 (2.00, 2.17) | 1.62 (1.49, 1.76) | 1.59 (1.45, 1.75) | 2.21 (2.12, 2.31) | 2.91 (2.82, 3.00) | 1.55 (1.42, 1.68) | 1.93 (1.87, 2.00) | 1.74 (1.62, 1.87) | 1.57 (1.46, 1.70) | 1.99 (1.92, 2.06) | 2.45 (2.38, 2.52) |
| Low iron test results | ||||||||||||
| Absent | ref | ref | ref | ref | ref | ref | ref | ref | ref | ref | ref | ref |
| Present | 0.79 (0.71, 0.87) | 1.04 (0.99, 1.10) | 1.03 (0.94, 1.14) | 0.99 (0.88, 1.11) | 1.06 (1.01, 1.12) | 0.93 (0.89, 0.97) | 0.76 (0.69, 0.83) | 0.93 (0.89, 0.97) | 0.90 (0.82, 0.99) | 0.86 (0.77, 0.95) | 0.95 (0.90, 0.99) | 0.81 (0.77, 0.84) |
| Unknown | 0.66 (0.61, 0.70) | 1.12 (1.07, 1.16) | 1.13 (1.05, 1.22) | 1.17 (1.06, 1.28) | 1.10 (1.05, 1.15) | 1.11 (1.08, 1.15) | 0.65 (0.60, 0.70) | 1.02 (0.98, 1.07) | 1.01 (0.93, 1.09) | 1.03 (0.94, 1.13) | 1.02 (0.97, 1.06) | 1.07 (1.03, 1.11) |
CHD, coronary heart disease; CVD, cardiovascular disease; ESKD, end-stage kidney disease; HF; heart failure
Anemia was defined as Hb <13g/dL in men and Hb <12g/dL in women
Adjusted for age, race, eGFR, CVD, hypertension, urine albumin-creatinine ratio, diabetes mellitus, smoking, healthcare organization
DISCUSSION
Our study of more than 5 million individuals confirmed that the presence and severity of anemia increased markedly with lower eGFR even after adjustment for other patient characteristics. These findings are consistent with prior studies2–5. Iron studies and vitamin B12 levels were checked infrequently in patients with anemia across the full eGFR range. Vitamin B12 deficiency was rare at all eGFR levels, while low iron test results were very common in those tested, although less so among patients with lower eGFR. Given that iron deficiency is easily mitigated through oral or intravenous iron supplementation, a relatively low-risk intervention, these data are particularly important to note.
Treatment of severe anemia with ESAs was infrequent. Despite an indication for ESA use in chronic kidney disease at a hemoglobin of less than 10g/dL, less than 4% of all patients with severe anemia and eGFR <30 ml/min/1.73 m2 were prescribed an ESA. We suspect that providers are hesitant to prescribe ESAs in chronic kidney disease patients due to the known adverse thrombotic and cardiovascular concerns associated with their use and how these risks may relate to common underlying comorbid conditions, such as cardiovascular disease and malignancy. Furthermore, there are significant logistical concerns related to prescribing ESAs, including insurance coverage issues and their subcutaneous or intravenous route of administration. It is also important to note that ESAs are often prescribed on in intermittent basis in the non-dialysis CKD population and ESA use may have been underestimated as a result.
We also illustrated that increasing severity of anemia is strongly associated with an increased risk of ESKD, cardiovascular disease, coronary heart disease, stroke, heart failure and death. These findings are consistent with prior literature8–11, 32–36, and are likely explained by anemia as a marker for the presence of more severe systemic disease. All of our data were obtained after the FDA ESA product label change that was spurred by the results of randomized controlled trials showing no improvement in mortality with ESA use, particularly the TREAT trial17. Although it is now known that the treatment of anemia to a normal hemoglobin with ESAs does not improve outcomes, anemia is still an important marker for the risk of adverse events. Further, anemia treatment has demonstrated clear quality of life benefits in numerous studies, which may be an indication for treatment in select patients12,17,37–38. We also observed that the presence of low iron test results in anemia was associated with a decreased risk of many of the adverse outcomes. This is likely due to the fact that the common alternatives to iron deficiency anemia are anemia of chronic disease or anemia of chronic kidney disease, which may reflect a heightened inflammatory state. Although our study did not distinguish between functional and absolute iron deficiency states, a recent study by Awan et al. observed that functional iron deficiency anemia was associated with an increased risk of mortality in non-dialysis chronic kidney disease patients but absolute iron deficiency was not39. Finally, although the relative hazards of adverse outcomes associated with anemia tended to be higher at higher eGFR and lower at lower eGFR, the absolute risk of both anemia and future adverse outcomes were highest at lower eGFR. Our detailed description of the epidemiology of anemia can inform testing and application of old and new anemia therapies40,41.
Our study has many strengths. First, our sample size of over 5 million patients allows for detailed insight into the burden and risks of mild and severe anemia by sex across the full range of chronic kidney disease. Furthermore, the Optum Labs Data Warehouse provides broad and comprehensive de-identified data from a wide range of healthcare centers across the US22. The limitations of our study include reliance on ICD codes for medical diagnoses which can provide inaccurate outcomes measures. While death information in Optum Labs Data Warehouse is assembled from multiple sources, death information is incomplete. In addition, care at centers not included in the Optum Labs Data Warehouse may have been missed. Finally, due to the observational nature of our study we cannot rule out residual confounding. In particular the risk association between anemia and adverse outcomes is not thought to be causal, but rather due to anemia as a marker of comorbidity severity.
Overall, our study provides generalizable, precise estimates from a large clinical population on the full spectrum of anemia severity. We quantify the association of low eGFR with severe anemia in both men and women. We find that studies to detect iron deficiency are conducted in less than one in five patients with anemia regardless of eGFR level, suggesting a need for greater testing and potentially for iron supplementation. We document that the presence of anemia is consistently associated with ESKD, cardiovascular disease, coronary heart disease, stroke, heart failure, and death in both men and women independent of other risk factors. Future studies should investigate safe strategies to mitigate the risks associated with anemia in chronic kidney disease.
Supplementary Material
Table S1: Prevalence of anemia by eGFR category, stratified by sex
Table S2: Beta coefficients from the polychotomous logistic regression of the association of hemoglobin category prevalence with age, sex, race and healthcare organization
Table S3: Frequency of iron study utilization and prevalence of low iron test results in patients with anemia* and eGFR >60 mL/min/1.73m2, stratified by sex. Low iron test results were defined as a serum ferritin </=30ng/mL and/or a serum transferrin saturation <20%.
Table S4: Adjusted hazard ratios for ESKD, cardiovascular disease, coronary heart disease, stroke, heart failure and death among patients with anemia* across different levels of hemoglobin, eGFR, stratified by sex
Table S5: Sample size, number of events, average follow-up time and incidence rates of ESKD, CVD, CHD, stroke, HF and death among patients with anemia* across different levels of hemoglobin and the presence of low iron test results, stratified by sex.
Support:
This project was funded in part by the US National Kidney Foundation (NKF funders include Fibrogen) and NIDDK (R01 DK100446 and DK115534). The funders had no role in the study design, data analysis, interpretation, writing or decision to submit for publication. Dr. Farrington was supported by the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) of the National Institutes of Health (NIH) under award number T32DK007732. Dr. Grams is supported by award number K24HL155861.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Financial Disclosure: Dr. Coresh is an advisor to Healthy.io, a digital health company. Dr Farrington declares that she has no other relevant financial interests. The remaining authors declare that they have no relevant financial interests.
Disclaimer: The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Prior Presentation: American Society of Nephrology Kidney Week, October 22–25, 2020.
References
- (1).World Health Organization. Nutritional anaemias: Report of a WHO scientific group. Geneva, Switzerland: World Health Organization; 1968 [PubMed] [Google Scholar]
- (2).Astor BC, Muntner P, Levin A, et al. Association of kidney function with anemia: the Third National Health and Nutrition Examination Survey (1988–1994). Arch Intern Med. 2002. Jun 24;162(12):1401–8. doi: 10.1001/archinte.162.12.1401. [DOI] [PubMed] [Google Scholar]
- (3).Hsu CY, McCulloch CE, Curhan GC. Epidemiology of anemia associated with chronic renal insufficiency among adults in the United States: results from the Third National Health and Nutrition Examination Survey. J Am Soc Nephrol. 2002. Feb;13(2):504–510. doi: 10.1681/ASN.V132504. [DOI] [PubMed] [Google Scholar]
- (4).Inker LA, Grams ME, Levey AS, et al. ; CKD Prognosis Consortium. Relationship of Estimated GFR and Albuminuria to Concurrent Laboratory Abnormalities: An Individual Participant Data Meta-analysis in a Global Consortium. Am J Kidney Dis. 2019. Feb;73(2):206–217. doi: 10.1053/j.ajkd.2018.08.013. Epub 2018 Oct 19. [DOI] [PMC free article] [PubMed] [Google Scholar]
- (5).Stauffer ME, Fan T. Prevalence of anemia in chronic kidney disease in the United States. PLoS One. 2014. Jan 2;9(1):e84943. doi: 10.1371/journal.pone.0084943. [DOI] [PMC free article] [PubMed] [Google Scholar]
- (6).Babitt JL, Lin HY. Mechanisms of anemia in CKD. J Am Soc Nephrol. 2012. Oct;23(10):1631–4. doi: 10.1681/ASN.2011111078. Epub 2012 Aug 30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- (7).Musio F Revisiting the treatment of anemia in the setting of chronic kidney disease, hematologic malignancies, and cancer: perspectives with opinion and commentary. Expert Rev Hematol. 2020. Nov;13(11):1175–1188. doi: 10.1080/17474086.2020.1830371. Epub 2020 Nov 2. [DOI] [PubMed] [Google Scholar]
- (8).Astor BC, Coresh J, Heiss G, et al. Kidney function and anemia as risk factors for coronary heart disease and mortality: the Atherosclerosis Risk in Communities (ARIC) Study. Am Heart J. 2006. Feb;151(2):492–500. doi: 10.1016/j.ahj.2005.03.055. [DOI] [PubMed] [Google Scholar]
- (9).Hoshino J, Muenz D, Zee J, et al. ; CKDopps Investigators. Associations of Hemoglobin Levels With Health-Related Quality of Life, Physical Activity, and Clinical Outcomes in Persons With Stage 3–5 Nondialysis CKD. J Ren Nutr. 2020. Sep;30(5):404–414. doi: 10.1053/j.jrn.2019.11.003. Epub 2020 Jan 21. [DOI] [PubMed] [Google Scholar]
- (10).Sarnak MJ, Tighiouart H, Manjunath G, et al. Anemia as a risk factor for cardiovascular disease in The Atherosclerosis Risk in Communities (ARIC) study. J Am Coll Cardiol. 2002. Jul 3;40(1):27–33. doi: 10.1016/s0735-1097(02)01938-1. [DOI] [PubMed] [Google Scholar]
- (11).Weiner DE, Tighiouart H, Vlagopoulos PT, et al. Effects of anemia and left ventricular hypertrophy on cardiovascular disease in patients with chronic kidney disease. J Am Soc Nephrol. 2005. Jun;16(6):1803–10. doi: 10.1681/ASN.2004070597. Epub 2005 Apr 13. [DOI] [PubMed] [Google Scholar]
- (12).Drüeke TB, Locatelli F, Clyne N, et al. ; CREATE Investigators. Normalization of hemoglobin level in patients with chronic kidney disease and anemia. N Engl J Med. 2006. Nov 16;355(20):2071–84. doi: 10.1056/NEJMoa062276. [DOI] [PubMed] [Google Scholar]
- (13).Hayashi T, Maruyama S, Nangaku M, et al. ; PREDICT Investigators. Darbepoetin Alfa in Patients with Advanced CKD without Diabetes: Randomized, Controlled Trial. Clin J Am Soc Nephrol. 2020. May 7;15(5):608–615. doi: 10.2215/CJN.08900719. Epub 2020 Apr 3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- (14).Jing Z, Wei-jie Y, Nan Z, et al. Hemoglobin targets for chronic kidney disease patients with anemia: a systematic review and meta-analysis. PLoS One. 2012;7(8):e43655. doi: 10.1371/journal.pone.0043655. Epub 2012 Aug 30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- (15).Koulouridis I, Alfayez M, Trikalinos TA, et al. Dose of erythropoiesis-stimulating agents and adverse outcomes in CKD: a metaregression analysis. Am J Kidney Dis. 2013. Jan;61(1):44–56. doi: 10.1053/j.ajkd.2012.07.014. Epub 2012 Aug 22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- (16).Palmer SC, Navaneethan SD, Craig JC, et al. Meta-analysis: erythropoiesis-stimulating agents in patients with chronic kidney disease. Ann Intern Med. 2010. Jul 6;153(1):23–33. doi: 10.7326/0003-4819-153-1-201007060-00252. Epub 2010 Jun 24. [DOI] [PubMed] [Google Scholar]
- (17).Pfeffer MA, Burdmann EA, Chen CY, et al. ; TREAT Investigators. A trial of darbepoetin alfa in type 2 diabetes and chronic kidney disease. N Engl J Med. 2009. Nov 19;361(21):2019–32. doi: 10.1056/NEJMoa0907845. Epub 2009 Oct 30. [DOI] [PubMed] [Google Scholar]
- (18).Phrommintikul A, Haas SJ, Elsik M, et al. Mortality and target haemoglobin concentrations in anaemic patients with chronic kidney disease treated with erythropoietin: a meta-analysis. Lancet. 2007. Feb 3;369(9559):381–8. doi: 10.1016/S0140-6736(07)60194-9. [DOI] [PubMed] [Google Scholar]
- (19).Singh AK, Szczech L, Tang KL, et al. ; CHOIR Investigators. Correction of anemia with epoetin alfa in chronic kidney disease. N Engl J Med. 2006. Nov 16;355(20):2085–98. doi: 10.1056/NEJMoa065485. [DOI] [PubMed] [Google Scholar]
- (20).KDIGO 2012 Clinical Practice Guideline for the Evaluation and Management of Chronic Kidney Disease. Kidney Int Suppl 2013; 4:2. [DOI] [PubMed] [Google Scholar]
- (21).Le CH. The Prevalence of Anemia and Moderate-Severe Anemia in the US Population (NHANES 2003–2012). PLoS One. 2016. Nov 15;11(11):e0166635. doi: 10.1371/journal.pone.0166635. [DOI] [PMC free article] [PubMed] [Google Scholar]
- (22).OptumLabs. OptumLabs and OptumLabs Data Warehouse (OLDW) Descriptions and Citation. Minnetonka, MN: n.p. July 2020. PDF. Reproduced with permission from OptumLabs [Google Scholar]
- (23).Levey AS, Stevens LA, Schmid CH, et al. ; CKD-EPI (Chronic Kidney Disease Epidemiology Collaboration). A new equation to estimate glomerular filtration rate. Ann Intern Med. 2009. May 5;150(9):604–12. doi: 10.7326/0003-4819-150-9-200905050-00006. Erratum in: Ann Intern Med. 2011 Sep 20;155(6):408. [DOI] [PMC free article] [PubMed] [Google Scholar]
- (24).Antony AC. Megaloblastic anemias. In: Hematology: Basic principles and practice, 4th ed, Hoffman R, Benz EJ, Shattil SJ, et al. (Eds), Churchill Livingstone, New York: 2005. p.519. [Google Scholar]
- (25).Fernández-Rodríguez AM, Guindeo-Casasús MC, Molero-Labarta T, et al. Diagnosis of iron deficiency in chronic renal failure. Am J Kidney Dis. 1999. Sep;34(3):508–13. doi: 10.1016/s0272-6386(99)70079-x. [DOI] [PubMed] [Google Scholar]
- (26).Kalantar-Zadeh K, Höffken B, Wünsch H, et al. Diagnosis of iron deficiency anemia in renal failure patients during the post-erythropoietin era. Am J Kidney Dis. 1995. Aug;26(2):292–9. doi: 10.1016/0272-6386(95)90649-5. [DOI] [PubMed] [Google Scholar]
- (27).Stancu S, Stanciu A, Zugravu A, et al. Bone marrow iron, iron indices, and the response to intravenous iron in patients with non-dialysis-dependent CKD. Am J Kidney Dis. 2010. Apr;55(4):639–47. doi: 10.1053/j.ajkd.2009.10.043. Epub 2010 Jan 15. Erratum in: Am J Kidney Dis. 2010 Nov;5(11):2137–8. [DOI] [PubMed] [Google Scholar]
- (28).Thomas DW, Hinchliffe RF, Briggs C, et al. ; British Committee for Standards in Haematology. Guideline for the laboratory diagnosis of functional iron deficiency. Br J Haematol. 2013. Jun;161(5):639–48. doi: 10.1111/bjh.12311. Epub 2013 Apr 10. [DOI] [PubMed] [Google Scholar]
- (29).Tran TN, Eubanks SK, Schaffer KJ, Zhou CY, Linder MC. Secretion of ferritin by rat hepatoma cells and its regulation by inflammatory cytokines and iron. Blood. 1997. Dec 15;90(12):4979–86. [PubMed] [Google Scholar]
- (30).Short MW, Domagalski JE. Iron deficiency anemia: evaluation and management. Am Fam Physician. 2013. Jan 15;87(2):98–104. [PubMed] [Google Scholar]
- (31).Maner BS, Moosavi L. Mean Corpuscular Volume. [Updated 2021 Jul 10]. In: StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing; 2021. Jan-. Available from: https://www.ncbi.nlm.nih.gov/books/NBK545275/ [PubMed] [Google Scholar]
- (32).Astor BC, Arnett DK, Brown A, et al. Association of kidney function and hemoglobin with left ventricular morphology among African Americans: the Atherosclerosis Risk in Communities (ARIC) study. Am J Kidney Dis. 2004. May;43(5):836–45. doi: 10.1053/j.ajkd.2003.12.047. [DOI] [PubMed] [Google Scholar]
- (33).Chang JM, Chen SC, Huang JC, et al. Anemia and left ventricular hypertrophy with renal function decline and cardiovascular events in chronic kidney disease. Am J Med Sci. 2014. Mar;347(3):183–9. doi: 10.1097/MAJ.0b013e31827981be. [DOI] [PubMed] [Google Scholar]
- (34).Go AS, Yang J, Tan TC, et al. ; Kaiser Permanente Northern California CKD Outcomes Study. Contemporary rates and predictors of fast progression of chronic kidney disease in adults with and without diabetes mellitus. BMC Nephrol. 2018. Jun 22;19(1):146. doi: 10.1186/s12882-018-0942-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- (35).Mohanram A, Zhang Z, Shahinfar S, et al. Anemia and end-stage renal disease in patients with type 2 diabetes and nephropathy. Kidney Int. 2004. Sep;66(3):1131–8. doi: 10.1111/j.1523-1755.2004.00863.x. [DOI] [PubMed] [Google Scholar]
- (36).Yi SW, Moon SJ, Yi JJ. Low-normal hemoglobin levels and anemia are associated with increased risk of end-stage renal disease in general populations: A prospective cohort study. PLoS One. 2019. Apr 25;14(4):e0215920. doi: 10.1371/journal.pone.0215920. [DOI] [PMC free article] [PubMed] [Google Scholar]
- (37).Furuland H, Linde T, Ahlmén J, Christensson A, Strömbom U, Danielson BG. A randomized controlled trial of haemoglobin normalization with epoetin alfa in pre-dialysis and dialysis patients. Nephrol Dial Transplant. 2003. Feb;18(2):353–61. doi: 10.1093/ndt/18.2.353. [DOI] [PubMed] [Google Scholar]
- (38).Rossert J, Levin A, Roger SD, Hörl WH, Fouqueray B, Gassmann-Mayer C, Frei D, McClellan WM. Effect of early correction of anemia on the progression of CKD. Am J Kidney Dis. 2006. May;47(5):738–50. doi: 10.1053/j.ajkd.2006.02.170. [DOI] [PubMed] [Google Scholar]
- (39).Awan AA, Walther CP, Richardson PA, et al. Prevalence, correlates and outcomes of absolute and functional iron deficiency anemia in nondialysis-dependent chronic kidney disease. Nephrol Dial Transplant. 2021. Jan 1;36(1):129–136. doi: 10.1093/ndt/gfz192. [DOI] [PubMed] [Google Scholar]
- (40).Fishbane S Anemia management: a historical perspective. Kidney Int Suppl. 2021. Apr;11(1):3–7. doi: 10.1016/j.kisu.2020.12.004 [DOI] [Google Scholar]
- (41).Raichoudhury R, Spinowitz BS. Treatment of anemia in difficult-to-manage patients with chronic kidney disease. Kidney Int Suppl (2011). 2021. Apr;11(1):26–34. doi: 10.1016/j.kisu.2020.12.006. Epub 2021 Mar 18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- (42).Gansevoort RT, Correa-Rotter R, Hemmelgarn BR, et al. Chronic kidney disease and cardiovascular risk: epidemiology, mechanisms, and prevention. Lancet. 2013. Jul 27;382(9889):339–52. doi: 10.1016/S0140-6736(13)60595-4. Epub 2013 May 31. [DOI] [PubMed] [Google Scholar]
- (43).Batchelor EK, Kapitsinou P, Pergola PE, et al. Iron Deficiency in Chronic Kidney Disease: Updates on Pathophysiology, Diagnosis, and Treatment. J Am Soc Nephrol. 2020. Mar;31(3):456–468. doi: 10.1681/ASN.2019020213. Epub 2020 Feb 10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- (44).Collins AJ, Li S, Peter WS, et al. Death, hospitalization, and economic associations among incident hemodialysis patients with hematocrit values of 36 to 39%. J Am Soc Nephrol. 2001. Nov;12(11):2465–2473. doi: 10.1681/ASN.V12112465. [DOI] [PubMed] [Google Scholar]
- (45).Khan S, Pereira BJ. Hematocrit level associated mortality in hemodialysis patients, by Ma JZ, Ebben J, Xia H, Collins AJ. J Am Soc Nephrol 10:610–619, 1999. Semin Dial. 2000 Mar-Apr;13(2):112–3. doi: 10.1046/j.1525-139x.2000.00036.x. [DOI] [PubMed] [Google Scholar]
- (46).Kottgen A, Russell SD, Loehr LR, et al. Reduced kidney function as a risk factor for incident heart failure: the atherosclerosis risk in communities (ARIC) study. J Am Soc Nephrol. 2007. Apr;18(4):1307–15. doi: 10.1681/ASN.2006101159. Epub 2007 Mar 7. [DOI] [PubMed] [Google Scholar]
- (47).McClellan WM, Flanders WD, Langston RD, et al. Anemia and renal insufficiency are independent risk factors for death among patients with congestive heart failure admitted to community hospitals: a population-based study. J Am Soc Nephrol. 2002. Jul;13(7):1928–36. doi: 10.1097/01.asn.0000018409.45834.fa. [DOI] [PubMed] [Google Scholar]
- (48).Parfrey PS, Foley RN. The clinical epidemiology of cardiac disease in chronic renal failure. J Am Soc Nephrol. 1999. Jul;10(7):1606–15. doi: 10.1681/ASN.V1071606. [DOI] [PubMed] [Google Scholar]
- (49).Xia H, Ebben J, Ma JZ, et al. Hematocrit levels and hospitalization risks in hemodialysis patients. J Am Soc Nephrol. 1999. Jun;10(6):1309–16. doi: 10.1681/ASN.V1061309. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Table S1: Prevalence of anemia by eGFR category, stratified by sex
Table S2: Beta coefficients from the polychotomous logistic regression of the association of hemoglobin category prevalence with age, sex, race and healthcare organization
Table S3: Frequency of iron study utilization and prevalence of low iron test results in patients with anemia* and eGFR >60 mL/min/1.73m2, stratified by sex. Low iron test results were defined as a serum ferritin </=30ng/mL and/or a serum transferrin saturation <20%.
Table S4: Adjusted hazard ratios for ESKD, cardiovascular disease, coronary heart disease, stroke, heart failure and death among patients with anemia* across different levels of hemoglobin, eGFR, stratified by sex
Table S5: Sample size, number of events, average follow-up time and incidence rates of ESKD, CVD, CHD, stroke, HF and death among patients with anemia* across different levels of hemoglobin and the presence of low iron test results, stratified by sex.
Data Availability Statement
The data underlying the results of this study are third party data owned by OptumLabs and contain sensitive patient information; therefore, the data is only available upon request. Interested researchers engaged in HIPAA compliant research may contact connected@optum.com for data access requests. The data use requires researchers to pay for rights to use and access the data.