Abstract
Blood collection from minority populations improves the transfusion support of patients with sickle cell disease and thalassemia, but efforts are challenged by high deferral rates for hemoglobin (Hb) eligibility thresholds. This study sought to evaluate hemoglobin and iron status of a representative US female population to assess the suitability of 12.0 g/dL as minimum hemoglobin. Data were extracted from the National Health and Nutrition Examination Surveys (NHANES), 1999–2010. A national sample designed to reflect potential female blood donors (weight ≥110lbs, not pregnant, no infectious marker reactivity, and no blood donation in past year) aged 16–49 years was analyzed for Hb and serum ferritin (SF) measures by race/ethnicity(N=6937). Mean Hb and SF and the prevalence of iron deficiency ([ID] SF<12ng/mL and SF<26ng/mL) and low Hb (<12.5g/dL and <12.0g/dL) were estimated. Multivariable modified Poisson regression compared the prevalence for ID or low Hb at each cutoff by race/ethnicity. Mean SF values were higher and ID prevalence was lower in Non-Hispanic (NH) White (SF=45.3ng/mL, SF<12ng/mL=8.2%) than NH Black (SF=39.6ng/mL, SF<12ng/mL=14.2%) and Hispanic (SF=36.5ng/mL, SF<12ng/mL=12.7%) females. Compared to NH White females (13.7g/dL), mean Hb was lower in NH Black (12.6g/dL) and Hispanic females (13.4g/dL). The percentage with Hb<12.5g/dL was >4 times greater in NH Black (39.1%) and >2 times greater in Hispanic females (16.5%) compared to NH White (8.6%). Within 0.5g/dL incremental categories of Hb, NH Black had higher mean SF levels and lower prevalence of SF<12ng/mL or <26ng/mL compared to NH White and Hispanic females. At Hb of 12.0–12.4g/dL, NH Black females had better measures of iron status (SF=39.1ng/mL, %SF<12ng/mL=12.0%) than NH White (SF=33.6ng/mL, %SF<12ng/mL=15.8%) and Hispanic (SF=30.4ng/mL, %SF<12ng/mL=15.5%) females whose Hb was 12.5–12.9g/dL. Adjusting for age and Hb, the prevalence ratio for low SF was significantly lower in NH Black compared to NH White females at both SF<26ng/ml (adjusted prevalence ratio [aPR]=0.83, 95%CI=0.76–0.92) and SF<12ng/ml (aPR=0.66, 95%CI=0.52–0.83). NH Black females with Hb 12.0–12.4g/dL have better iron stores than NH White and Hispanic females whose Hb is 12.5–12.9g/dL. The distribution of Hb and iron may support the safe collection of blood for female donors below the current Hb eligibility requirement of 12.5g/dL.
Keywords: Blood donation, iron deficiency, hemoglobin, anemia, NHANES
INTRODUCTION
Patients with sickle cell disease and thalassemia have high rates of alloimmunization.[1, 2] Antigen-matching is critical for those who have alloantibodies or to reduce the risk of alloimmunization. In order to optimize matching, blood collection is needed from Black donors to meet the needs of Black patients. However, there is underrepresentation of Black individuals in the US blood donor pool.[3] Barriers to donation by Black individuals include medical conditions triggering ineligibility, levels of distrust, and onsite deferrals at the blood center.[4] For Black donors, as for other prospective US donors, the foremost reason for deferral is low hemoglobin (Hb). In both unadjusted[5, 6] and adjusted[7, 8] analyses, the risk of deferral for low hemoglobin in Black donors is approximately twice as high as in White donors, with deferral rates reaching as high as 29% in Black females in one report.[5] Given that donor deferral leads to lower return rates and diminished donor retention, the detrimental impact of deferrals compounds over time.
The current requirement for a minimum hemoglobin of 12.5 g/dL for female donors in the US is above the lower bound of normal recognized by the Centers for Disease Control and Prevention (CDC), 12.0 g/dL;[9] collection at a Hb of 12.0 g/dL would allow for inclusion of many deferred donors. A 2015 FDA Final Rule[10] and discussion at a November 2016 Blood Products Advisory Committee meeting[11] acknowledged 12.0 g/dL as physiologically normal, but concerns focused on the iron status of donors with low-normal hemoglobin. This study sought to describe hemoglobin and iron status of a representative US female population to assess the suitability of 12.0 g/dL as minimum hemoglobin acceptance criteria for some —or all— female blood donors.
METHODS
Study population
Data were sourced from the 1999–2010 National Health and Nutrition Examination Surveys (NHANES). NHANES is a cross-sectional survey conducted yearly by the National Center for Health Statistics (NCHS) of the CDC. NHANES uses a complex sampling design to produce nationally representative estimates of the non-institutionalized, civilian US population. Participation includes at-home interview, medical examination, and analysis of biological specimens.
During our study period 1999–2010, female participants 16–49 years-old reported their history of blood donation in the past year and had both hemoglobin and serum ferritin (SF) measured, as previously described.[12] Single-race non-Hispanic (NH) White, NH Black, and Hispanic female participants aged 16–49 with observed data on blood donation during the past year and Hb and SF measures were eligible for analysis. To approximate a population that would be eligible to donate blood, participants who were pregnant, <110 pounds, human immunodeficiency virus (HIV)-positive, hepatitis B virus (HBV) surface antigen positive, or hepatitis C virus (HCV) antibody-positive were excluded from the analysis.[13] Participants who identified as a NH Other race including multiracial persons were not included in the analytic sample due to insufficient sample sizes to perform analyses among this group. To assess serum ferritin and hemoglobin levels unaltered by recent blood donation, participants who reported donating blood within the past year (i.e., donors) were excluded from this analysis.
The Johns Hopkins University School of Medicine Institutional Review Board deemed this study of de-identified public data to be exempt from review.
Measures of serum ferritin and hemoglobin levels
Hb level was measured using two hematology analyzers during the study period. SF levels were assayed by three methods during the study period and were calibrated in accordance with NCHS standards.[14, 15] Details have been previously reported.[12]
Covariates
During the household interview, participants were asked to identify their racial and ethnic identity from a predefined list of categories. Additional sociodemographic data, history of cigarette use, dietary supplement use, and dietary iron were also collected during participant interviews, with computer-assisted technology. Participants who had consumed supplements within the last 30 days were asked to make available the name and also provide the supplement packaging. The NHANES Dietary Supplement Database, which contains product, ingredient, and blend information for >12,000 supplements, allowed for determination of iron content. Ascertainment of dietary iron was based on reported food and beverage consumption in the past 24 hours cross-referenced to a United States Department of Agriculture database; this database contains nutrient information for ~13,000 foods and beverages. Details have been previously reported.[12]
Data on current oral contraceptive use and menstruation status were obtained by personal or computer-assisted interview. Participants’ weight and height were measured to calculate their body mass index (BMI). Serum C-reactive protein (CRP) levels were measured by latex-enhanced nephelometry. Serum creatinine was measured by the Jaffe method and a subsample of each survey cycle was compared to “a gold standard reference method” to determine if adjustment was needed. The 1999–2000 and 2005–2006 serum creatinine values were standardized using the following equations:
1999–2000: YStandardized = 1.013 * XMeasured + 0.147
2005–2006: YStandardized = −0.016 + 0.978* XMeasured
HBV surface antigen was measured by solid-phase sandwich enzyme immunoassay. HCV antibody was measured by enzyme linked immunosorbent assay and confirmed by strip immunoblot assay. HIV antibody was measured by enzyme immunoassay; HIV testing was only performed on individuals ≥18 years old and participants under the age of 18 were assumed to be HIV-negative in this study.
Statistical analysis
Reported sample sizes are unweighted; however, all data were analyzed using svy commands in Stata/MP v15.1 (Statacorp LP, College Station, Texas) to account for the complex study design and incorporate NCHS-derived weights to generate nationally-representative estimates unless otherwise noted. The weights account for unequal selection probabilities, nonresponse, and adjustment to the civilian, noninstitutionalized US population. Survey weights were pooled across years as recommended by the NCHS.[16] Taylor series linearization was used to estimate design-adjusted standard errors (SE), and 95% confidence intervals (CI) for proportions were calculated using the Korn-Graubard method.
The analyses were restricted to 16–49 year-old females with no history of blood donation in the past year but who otherwise appeared eligible for blood donation. Multiple outcomes were examined in this study. Iron deficiency was assessed at two cutoffs: SF <12ng/mL and SF <26ng/mL. Hemoglobin was also examined at two cutoffs of operational importance for female blood donor eligibility: Hb <12.5g/dL and Hb <12.0g/dL, representing the current and potentially altered cutoffs, respectively.
All descriptive analyses were stratified by race/ethnicity. Specifically, the distributions of covariates were compared by race/ethnicity using descriptive statistics. Age-specific geometric mean SF, mean Hb, and iron deficiency and low Hb prevalence were estimated. Participant SF was further summarized within 0.5g/dL bins of Hb. Hemoglobin distributions were also investigated among females whose SF was >12ng/ml and thus were not iron deficient. The assessment of hemoglobin distributions was repeated using a ferritin cutoff of >26 ng/mL as a sensitivity analysis since erythropoiesis might be affected in subjects with ferritin between 12 and 26 ng/mL.[17]
A series of modified Poisson regression models were performed to further investigate differences in the prevalence of iron deficiency among NH Black and Hispanic females compared to NH White females. Model 1 estimated crude prevalence ratios. Model 2 adjusted for all covariates identified a priori as potential predictors of iron deficiency,[18–21] which included age group, family income-to-poverty ratio, birthplace, cigarette smoking history, dietary iron intake during the past 24 hours, iron-containing supplement use during the past 30 days, hormonal contraceptive use, menses during the past 12 months, BMI, C-reactive protein levels, and serum creatinine levels. Model 3 added continuous Hb to the covariates listed in Model 2. Finally, Model 4 examined the prevalence of iron deficiency by race/ethnicity after adjustment for age group and continuous Hb to represent the more limited data routinely available during blood donor eligibility screening.
Similar modified Poisson regression models were performed to investigate differences in the prevalence of hemoglobin below current (<12.5g/dL) or altered (12.0g/dL) eligibility thresholds by race/ethnicity. Model 1 estimated the crude prevalence of low hemoglobin by race/ethnicity. Model 2 included adjustment for the same covariates as the iron deficiency model described above, while Model 3 added adjustment for continuous log-transformed SF. Model 4 examined the prevalence of low Hb by race/ethnicity, adjusting for age group and continuous log-transformed SF.
Missing covariate data were handled by list-wise deletion of participants in regression models.
RESULTS
Study population
Of 9196 NH White, NH Black, and Hispanic female participants in NHANES 1999–2010 who were 16-to-49-years-old with data available on hemoglobin, serum ferritin, and blood donation history, 7408 met the inclusion criteria as potential candidates for blood donation (i.e., non-pregnant, negative infectious disease markers, weight ≥ 110 pounds). Among these, 471 participants reported a blood donation within the past year, representing approximately 9% of NH White participants and 5% of the NH Black and Hispanic participants. As the exact time of donation was unknown, this study focused exclusively on the 6937 females who did not report any blood donation in the past year, which included 2883 NH White participants, 1679 NH Black participants, and 2375 Hispanic participants (Table 1, Figure 1).
Table 1.
Characteristics of the study population stratified by race/ethnicity, females aged 16–49 years with no history of blood donation in the past year, National Health and Nutrition Examination Survey, 1999–2010 (n=6937).a
| Non-Hispanic White | Non-Hispanic Black | Hispanic | |
|---|---|---|---|
|
|
|
|
|
| N (%) | N (%) | N (%) | |
| Total | 2883 (100.0) | 1679 (100.0) | 2375 (100.0) |
| Age, years | |||
| 16–17 | 285(5.6) | 284(6.4) | 333 (5.9) |
| 18–19 | 208(3.8) | 253 (5.7) | 352 (6.0) |
| 20–29 | 694 (24.6) | 339 (25.7) | 548 (29.5) |
| 30–39 | 800 (29.2) | 372 (30.1) | 523 (31.3) |
| 40–49 | 896 (36.8) | 431 (32.0) | 619 (27.4) |
| Family income to poverty ratio | |||
| <1.0 | 490(12.2) | 491 (26.5) | 715 (27.9) |
| 1.0–3.0 | 934 (30.4) | 652 (39.7) | 1015(43.1) |
| >3.0 | 1344 (53.4) | 411 (26.7) | 409 (19.5) |
| Missing | 115(3.9) | 125(7.1) | 236 (9.5) |
| Educational attainment | |||
| Less than high school | 628 (15.7) | 614 (26.5) | 1213 (41.3) |
| High school/GED | 648 (22.8) | 358 (22.8) | 456 (20.5) |
| Some college or more | 1598 (61.4) | 698 (50.5) | 672 (37.3) |
| Missing | 9 (0.2) | 9 (0.2) | 34 (0.8) |
| Birthplace | |||
| US born | 2749 (95.2) | 1549 (91.8) | 1122 (44.2) |
| Foreign born | 134 (4.8) | 129(8.2) | 1252 (55.8) |
| Missing | - | 1 (0.0) | 1 (0.1) |
| Cigarette smoking history | |||
| Never | 1457 (53.0) | 1166 (69.6) | 1629 (69.9) |
| Ever | 1419 (46.8) | 507 (30.3) | 730 (29.7) |
| Missing | 7 (0.1) | 6(0.1) | 16 (0.4) |
| Dietary iron intake (mg)b | 2842 (13.0) | 1660(12.1) | 2327(12.5) |
| Dietary supplement use | |||
| None | 1234(39.2) | 1137 (63.6) | 1588(63.7) |
| Yes, without iron | 702 (25.4) | 201 (13.8) | 354(18.1) |
| Yes, with iron | 939 (35.2) | 333 (22.2) | 412 (17.1) |
| Missing | 8 (0.3) | 8 (0.4) | 21 (1.0) |
| Menses in past 12 months | |||
| None/Irregular | 585 (21.6) | 357 (22.6) | 471 (19.6) |
| At least one/Regular | 2293 (78.3) | 1321 (77.3) | 1899 (80.2) |
| Missing | 5 (0.1) | 1 (0.1) | 5 (0.2) |
| Hormonal contraceptive | |||
| No | 2345(81.3) | 1528 (91.2) | 2150(89.5) |
| Yes | 521 (18.2) | 144(8.3) | 216 (10.0) |
| Missing | 17 (0.5) | 7 (0.6) | 9 (0.5) |
| Body mass index (kg/m2) | |||
| Under-/Normal weight | 1363 (46.3) | 498 (24.3) | 828 (30.7) |
| Overweight | 665 (23.8) | 399 (24.6) | 737 (32.9) |
| Obese | 823 (28.9) | 766 (50.4) | 793 (35.6) |
| Missing | 32 (0.9) | 16 (0.7) | 17 (0.7) |
| CRP (mg/dL) | |||
| ≤0.06 | 796 (26.4) | 428 (20.6) | 522 (19.9) |
| 0.07–0.22 | 733 (25.7) | 355 (21.0) | 620 (25.3) |
| 0.23–0.51 | 708 (24.7) | 405 (25.5) | 649 (27.3) |
| >0.51 | 645 (23.1) | 490 (32.9) | 581 (27.4) |
| Missing | 1 (0.0) | 1 (0.1) | 3 (0.0) |
| Serum creatinine (mg/dL) | |||
| ≤0.65 | 475 (15.1) | 228 (13.7) | 1094 (43.6) |
| 0.66–0.70 | 711 (24.8) | 372 (21.3) | 619 (26.0) |
| 0.71–0.80 | 977 (34.1) | 549 (32.3) | 470 (21.3) |
| >0.80 | 707 (25.6) | 515 (32.0) | 181 (8.7) |
| Missing | 13 (0.5) | 15 (0.7) | 11 (0.4) |
Data are unweighted sample sizes and weighted column percentages (%) unless otherwise noted. Estimates generated from cell sizes <30 should be interpreted with caution.
Data on dietary intake in the past 24 hours were expressed as weighted geometric means.
Figure 1.
Sample flow diagram of analytic population, National Health and Nutrition Examination Survey 1999–2010.
Indicators of socio-economic status suggested a less advantaged status for NH Black and Hispanic participants compared to NH Whites. The proportion of female participants with family income to poverty ratio < 1.0 was more than twice as large in NH Black (26.5%) and Hispanic (27.9%) females compared to NH White females (12.2%), and a higher proportion of NH White females reported attending some college or more compared to NH Black and Hispanic females. Average 24-hour dietary iron intake was comparable across race/ethnicity. In contrast, past-30-day use of dietary supplements containing iron was more frequently reported by NH White females (35.2%) than by NH Black (22.2%) or Hispanic (17.1%) females. Across all three race/ethnicity groups, markers for kidney disease (serum creatinine) and inflammation (C-reactive protein) were low; the cutoff denoting the highest quartile was well below recognized levels that may be associated with hematopoietic and iron status.
Serum ferritin and hemoglobin levels
Geometric mean SF values were higher in NH White females (SF=45.3 ng/mL) than NH Black (SF=39.6) and Hispanic (SF=36.5) females (Table 2). Prevalence of SF<12 ng/mL was lower in NH White females (8.2%) than NH Black (14.2%) and Hispanic (12.7%) females (Table 2). Within age groups, geometric mean SF tended to increase with age among all racial/ethnic groups. Prevalence of SF<26 ng/mL was lower in NH White females (22.6%) than NH Black (32.7%) or Hispanic (31.4%) females; after stratification on age, prevalence of SF <26ng/ml exceeded 25% in all age groups for NH Black and Hispanic females. NH White females had higher mean Hb (13.7 g/dL) than NH Black (12.6 g/dL) and Hispanic (13.4 g/dL) females with only modest variation by age (Table 2). The percentage with Hb<12.5 g/dL was more than 4 times greater in NH Black (39.1%) than NH White (8.6%) females and twice as great as Hispanic females (16.5%). Similar results were observed for the prevalence of Hb<12.0 g/dL, which was 4.1% in NH White, 23.0% in NH Black, and 8.8% in Hispanic females.
Table 2.
Serum ferritin and hemoglobin status of females with no history of past-year blood donation, stratified by race/ethnicity and age. National Health and Nutrition Examination Survey, 1999–2010 (n=6937).
| Na | Serum Ferritin (SF) |
Hemoglobin (Hb) |
|||||
|---|---|---|---|---|---|---|---|
| Geometric mean, ng/mL (95% CI)b | Prevalence SF <12.0ng/mL, % (95% CI)c | Prevalence SF <26.0ng/mL, % (95% CI)c | Mean, g/dL (95% CI)b | Prevalence Hb <12.5g/dL, % (95% CI)c | Prevalence Hb <12.0g/dL, % (95% CI)c | ||
| Non-Hispanic White | |||||||
| Overall | 2883 | 45.3 (43.7–47.0) | 8.2 (7.0–9.5) | 22.6 (21.0–24.4) | 13.7 (13.7–13.8) | 8.6 (7.5–9.8) | 4.1 (3.3–5.1) |
| Age, years | |||||||
| 16–17 | 285 | 32.7 (29.9–35.8) | 9.6 (5.6–15.0) | 34.7 (28.2–41.5) | 13.7 (13.5–13.9) | 7.4 (4.5–11.3) | 2.2 (0.6–5.7)d |
| 18–19 | 208 | 31.3 (28.0–34.9) | 12.0 (7.0–18.8) | 38.4 (31.2–46.0) | 13.5 (13.4–13.7) | 14.2 (8.6–21.5) | 6.9 (3.1–13.1)d |
| 20–29 | 694 | 41.9 (39.3–44.7) | 5.5 (3.7–7.8) | 22.5 (18.7–26.6) | 13.8(13.7–13.9) | 6.8 (4.4–9.8) | 2.5 (1.3–4.3) |
| 30–39 | 800 | 48.2 (44.9–51.7) | 7.9 (5.9–10.4) | 19.3 (16.2–22.7) | 13.7 (13.6–13.8) | 7.1 (5.3–9.3) | 3.5 (2.3–5.2) |
| 40–49 | 896 | 49.6 (46.1–53.3) | 9.6 (7.3–12.3) | 22.0 (18.6–25.6) | 13.7 (13.6–13.8) | 10.6 (8.7–12.9) | 5.7 (4.1–7.7) |
| Non-Hispanic Black | |||||||
| Overall | 1679 | 39.6 (37.2–42.3) | 14.2 (12.0–16.7) | 32.7 (30.3–35.2) | 12.6(12.6–12.7) | 39.1 (36.3–42.0) | 23.0 (21.0–25.2) |
| Age, years | |||||||
| 16–17 | 284 | 29.9 (26.7–33.5) | 13.7 (9.5–18.9) | 41.2 (34.7–47.9) | 12.5 (12.4–12.7) | 41.6 (35.1–48.3) | 24.8 (19.2–31.1) |
| 18–19 | 253 | 31.1 (27.4–35.3) | 17.8 (12.3–24.4) | 40.3 (34.0–46.9) | 12.7 (12.5–12.8) | 36.9 (28.7–45.7) | 21.4 (15.7–27.9) |
| 20–29 | 339 | 33.7 (30.6–37.2) | 13.2 (9.3–18.1) | 37.0 (31.6–42.5) | 12.7 (12.6–12.9) | 35.1 (29.8–40.8) | 18.5 (14.4–23.2) |
| 30–39 | 372 | 39.8 (35.8–44.2) | 13.6 (10.2–17.6) | 32.4 (27.5–37.5) | 12.6(12.5–12.7) | 40.3 (35.2–45.6) | 22.8 (18.1–28.0) |
| 40–49 | 431 | 49.7 (43.5–56.8) | 15.1 (11.5–19.3) | 26.5 (22.1–31.2) | 12.5 (12.4–12.7) | 41.1 (35.5–46.8) | 26.9 (22.6–31.5) |
| Hispanic | |||||||
| Overall | 2375 | 36.5 (34.2–38.9) | 12.7 (10.9–14.6) | 31.4 (28.7–34.2) | 13.4(13.2–13.5) | 16.5 (14.0–19.2) | 8.8 (7.1–10.8) |
| Age, years | |||||||
| 16–17 | 333 | 30.2 (25.9–35.3) | 14.1 (9.6–19.7) | 40.1 (31.2–49.5) | 13.4(13.2–13.6) | 19.8 (12.4–29.0) | 8.3 (4.8–13.1) |
| 18–19 | 352 | 30.4 (27.3–33.7) | 13.9 (9.9–18.8) | 41.8 (35.2–48.6) | 13.4(13.2–13.6) | 16.6 (11.8–22.5) | 8.5 (4.9–13.6) |
| 20–29 | 548 | 36.5 (32.8–40.6) | 9.0 (6.4–12.2) | 31.0 (25.5–36.9) | 13.5 (13.3–13.6) | 12.6 (9.0–16.9) | 5.2 (3.5–7.5) |
| 30–39 | 523 | 38.3 (34.6–42.4) | 12.2 (9.2–15.8) | 27.1 (22.3–32.3) | 13.3(13.2–13.4) | 16.6 (12.7–21.1) | 9.2 (6.6–12.3) |
| 40–49 | 619 | 37.4 (33.1–42.2) | 16.6 (12.9–20.8) | 32.7 (28.2–37.4) | 13.3(13.1–13.5) | 19.8 (15.9–24.1) | 12.6 (9.2–16.6) |
Unweighted sample sizes.
Survey-weighted means and 95% confidence intervals (CI).
Survey-weighted proportions with Korn-Graubard 95% CI.
Prevalence estimate has a residual squared error >30% and/or fails to meet NCHS presentation guidelines and should be interpreted with caution.
Within 0.5 g/dL incremental groupings of Hb, NH Black females had higher mean SF levels and lower prevalence of SF<12 or <26 ng/mL compared to NH White and Hispanic females (Table 3). At 12.0–12.4 g/dL, NH Black females had better measures of iron stores (SF=39.1ng/mL, SF <12.0ng/ml=12.0%) than NH White (SF=33.6ng/mL, SF <12.0ng/ml =15.8%) or Hispanic (SF=30.4ng/mL, SF <12.0ng/ml =15.5%) females whose Hb was 12.5–12.9 g/dL.
Table 3.
Serum ferritin (SF) status by hemoglobin level among females with no history of past-year blood donation, stratified by race/ethnicity. National Health and Nutrition Examination Survey, 1999–2010 (n=6937).
| Hemoglobin (g/dL) |
||||||||
|---|---|---|---|---|---|---|---|---|
| <11.0 | 11.0–11.4 | 11.5–11.9 | 12.0–12.4 | 12.5–12.9 | 13.0–13.4 | 13.5–13.9 | ≥14.0 | |
| Non-Hispanic White | ||||||||
| Na | 37 | 27 | 53 | 129 | 321 | 457 | 658 | 1201 |
| Geometric mean SFb | 6.1 (3.3–11.2) | 12.9 (5.3–31.8) | 15.1 (10.8–21.2) | 26.2 (21.9–31.2) | 33.6 (29.9–37.7) | 43.0 (40.0–46.2) | 48.6 (45.7–51.6) | 58.8 (56.5–61.2) |
| Prevalence SF <12ng/mlc | 88.1 (42.8–99.9)d | 58.4 (13.2–94.5)d | 51.1 (34.4–67.5) | 21.0 (13.4–30.6) | 15.8 (11.3–21.3) | 8.3 (5.9–11.3) | 4.5 (2.9–6.6) | 1.2 (0.7–2.0) |
| Prevalence SF <26ng/mlc | 93.0 (54.5–100.0)d | 77.5 (26.1–99.2)d | 64.5 (45.5–80.6)d | 49.5 (39.0–60.0) | 34.3 (29.0–39.9) | 26.9 (22.7–31.3) | 18.4 (15.3–21.7) | 12.1 (10.3–14.2) |
| Non-Hispanic Black | ||||||||
| Na | 145 | 94 | 144 | 261 | 349 | 270 | 229 | 187 |
| Geometric mean SFb | 12.2 (9.5–15.7) | 20.3 (15.1–27.3) | 31.6 (27.1–36.8) | 39.1 (34.5–44.3) | 44.3 (38.8–50.6) | 47.1 (42.4–52.3) | 60.4 (52.7–69.2) | 67.8 (59.9–76.7) |
| Prevalence SF <12ng/mlc | 64.7 (54.9–73.7) | 36.3 (25.3–48.4) | 17.2 (11.2–24.8) | 12.0 (8.2–16.8) | 9.0 (5.5–13.7) | 3.2 (1.3–6.2)d | 2.8 (0.8–6.8)d | 2.1 (0.4–5.8)d |
| Prevalence SF <26ng/mlc | 77.9 (68.2–85.8) | 61.7 (46.5–75.3) | 45.6 (36.8–54.5) | 36.5 (30.3–43.1) | 29.0 (23.6–34.8) | 21.6 (16.2–27.8) | 14.0 (8.4–21.5) | 11.0 (6.7–16.7) |
| Hispanic | ||||||||
| Na | 92 | 55 | 76 | 203 | 365 | 421 | 509 | 654 |
| Geometric mean SFb | 5.7 (5.0–6.6) | 11.9 (7.9–17.8) | 19.8 (12.5–31.1) | 23.1 (20.0–26.7) | 30.4 (26.9–34.4) | 38.1 (34.3–42.2) | 44.4 (40.4–48.8) | 56.1 (51.7–60.9) |
| Prevalence SF <12ng/mlc | 90.6 (82.0–96.0) | 66.4 (45.2–83.7)d | 40.8 (25.4–57.7)d | 23.2 (17.4–30.0) | 15.5 (11.1–20.7) | 8.1 (5.0–12.3) | 3.5 (2.0–5.6) | 1.0 (0.4–2.2)d |
| Prevalence SF <26ng/mlc | 95.4 (88.3–98.8) | 83.6 (62.8–95.4) | 62.1 (40.8–80.5)d | 54.6 (46.2–62.8) | 39.6 (33.1–46.3) | 27.5 (22.0–33.7) | 24.0 (19.2–29.3) | 13.6 (9.8–18.1) |
Unweighted sample sizes.
Survey-weighted means and 95% confidence intervals (CI).
Survey-weighted proportions with Korn-Graubard 95% CI.
Residual squared error >30% and/or estimate fails to meet one or more criteria for NCHS presentation standards and should be interpreted with caution.
Differences by race in hemoglobin distributions remained evident after excluding females with SF <12 ng/mL (Figure 2). After removing participants with SF<12ng/ml, the lower 5th percentile for NH White, NH Black, and Hispanic females was 12.4g/dL, 11.2g/dL, and 12.1g/dL, respectively. In addition, the proportion of individuals with Hb <12.5 g/dL among NH White, NH Black, and Hispanic females was 5.4%, 32.5%, and 9.9% respectively. The sensitivity analysis excluding participants with SF <26 ng/mL did not substantively alter these findings. The lower 5th percentile for NH White, NH Black, and Hispanic females was 12.5g/dL, 11.4g/dL, and 12.2g/dL, respectively, and the proportion of individuals with Hb <12.5 g/dL among NH White, NH Black, and Hispanic females decreased to 4.1%, 28.1%, and 7.4% respectively.
Figure 2.
Distribution of hemoglobin values among females aged 16–49 with no recent history of blood donation and with serum ferritin >12ng/ml for non-Hispanic White (A), non-Hispanic Black (B), and Hispanic (C) individuals. Data represent weighted counts. National Health and Nutrition Examination Survey 1999–2010 (n=6,095).
A series of regression models evaluated the association between race/ethnicity and iron deficiency including control for different combinations of covariates (Table 4). The univariable model for race/ethnicity (Model 1) affirms the results shown in Table 2, with NH Black females having higher prevalence for SF <12 ng/mL (PR = 1.74, 95%CI 1.40–2.16) and for SF <26 ng/mL (PR = 1.44, 95%CI 1.30–1.61) than NH White females and Hispanic females having iron deficiency prevalence intermediate to that of NH White and NH Black females. Controlling for a number of sociodemographic and behavioral factors that might affect iron status did not markedly attenuate the associations by race/ethnicity (Model 2). With the addition of hemoglobin level in Model 3, the associations were reversed such that NH Black females have reduced prevalence of SF <12ng/mL (aPR=0.64, 95%CI=0.50–0.81) and SF <26 ng/mL (aPR=0.83, 95%CI=0.75–0.92) compared to NH White females. In Hispanic females, the prevalence of iron deficiency was no longer significantly different from those for NH White females after accounting for hemoglobin levels. Model 4, which adjusted only for age and hemoglobin, estimated similar prevalence ratios as those in Model 3.
Table 4.
Association of race/ethnicity with low serum ferritin (SF) among healthy 16-to-49-year-old females with no history of blood donation in the past year, National Health and Nutrition Examination Survey, 1999–2010.
| Prevalence Ratio (95% CI) | ||||
|---|---|---|---|---|
|
| ||||
| Model 1a | Model 2b | Model 3c | Model 4d | |
| SF <12ng/mL | ||||
| Race/Ethnicity | ||||
| Non-Hispanic White | Ref. | Ref. | Ref. | Ref. |
| Non-Hispanic Black | 1.74 (1.40–2.16) | 1.65 (1.33–2.04) | 0.64 (0.50–0.81) | 0.66 (0.52–0.83) |
| Hispanic | 1.55 (1.23–1.94) | 1.50 (1.15–1.95) | 1.09 (0.87–1.37) | 1.02 (0.82–1.26) |
| SF <26ng/mL | ||||
| Race/Ethnicity | ||||
| Non-Hispanic White | Ref. | Ref. | Ref. | Ref. |
| Non-Hispanic Black | 1.44 (1.30–1.61) | 1.36 (1.21–1.52) | 0.83 (0.75–0.92) | 0.83 (0.76–0.92) |
| Hispanic | 1.39 (1.24–1.55) | 1.22 (1.05– 1.42) | 1.06 (0.92–1.23) | 1.11 (0.999–1.23) |
Model 1 includes race/ethnicity only (n=6937).
Model 2 includes race/ethnicity and adjusts for age group, family income-to-poverty ratio, birthplace, cigarette smoking history, iron-containing supplement use during the past 30 days, dietary iron intake during the past 24 hours, menses during the past 12 months, hormonal contraceptive use, body mass index, C-reactive protein levels, and serum creatinine levels (n=6152).
Model 3 adjusts for hemoglobin (g/dL) and all factors included in Model 2 (n=6152).
Model 4 includes race/ethnicity and adjusts for age group and hemoglobin (g/dL) (n=6937).
Analogous modified Poisson regression models with Hb <12.5 g/dL and <12.0 g/dL (Table 5) as outcome measures similarly assess the relationship between race/ethnicity and Hb levels determining current or potentially future donor eligibility, respectively. The prevalence of Hb <12.5 g/dL was notably higher for NH Black (PR=4.54, 95%CI=4.00–5.16) and Hispanic females (PR=1.91, 95%CI=1.58–2.31) compared to NH White females, consistent with differences in prevalence shown in Table 2. Similar results were obtained for Hb <12.0 g/dL. In Model 2, adjustment for covariates hypothesized to influence iron status did not notably change the association between race/ethnicity and low Hb. Further adjustment for SF in Model 3 attenuated the relationship between race/ethnicity and low Hb, though low Hb remained significantly higher among NH Black and Hispanic females at both Hb <12.5g/dL and Hb <12.0g/dL. Model 4, which adjusted only for age group and serum ferritin, again indicated a higher prevalence of Hb <12.5g/dL (PR=3.84, 95%CI=3.38–4.36) and for Hb <12.0g/dL (PR=4.28, 95%CI=3.415.38) for NH Black females compared to NH White females. Similar results were seen among Hispanic females, again with an adjusted prevalence that are intermediate to those of NH White and NH Black females (Model 4 for Hb <12.0g/dl PR=1.64, 95%CI=1.31–2.05).
Table 5.
Association of race/ethnicity with low hemoglobin (Hb) among healthy 16-to-49-year-old females with no history of blood donation in the past year, National Health and Nutrition Examination Survey, 1999–2010.
| Prevalence Ratios (95% CI) | ||||
|---|---|---|---|---|
|
| ||||
| Model 1a | Model 2b | Model 3c | Model 4d | |
| Hb <12.5g/dL | ||||
| Race/Ethnicity | ||||
| Non-Hispanic White | Ref. | Ref. | Ref. | Ref. |
| Non-Hispanic Black | 4.54 (4.00–5.16) | 4.41 (3.81–5.11) | 3.71 (3.23–4.28) | 3.84 (3.38–4.36) |
| Hispanic | 1.91 (1.58–2.31) | 1.71 (1.33–2.19) | 1.45 (1.17–1.79) | 1.60 (1.37–1.86) |
| Hb <12g/dL | ||||
| Race/Ethnicity | ||||
| Non-Hispanic White | Ref. | Ref. | Ref. | Ref. |
| Non-Hispanic Black | 5.58 (4.44–7.01) | 5.08 (3.86–6.70) | 3.85 (2.96–5.00) | 4.28 (3.41–5.38) |
| Hispanic | 2.14 (1.62–2.83) | 2.00 (1.40–2.86) | 1.54 (1.14–2.08) | 1.64 (1.31–2.05) |
Model 1 includes race/ethnicity only (n=6937).
Model 2 includes race/ethnicity and adjusts for age group, family income-to-poverty ratio, birthplace, cigarette smoking history, iron-containing supplement use during the past 30 days, dietary iron intake during the past 24 hours, menses during the past 12 months, hormonal contraceptive use, body mass index, C-reactive protein levels, and serum creatinine levels (n=6152).
Model 3 adjusts for log-transformed serum ferritin (ng/mL) and all factors included in Model 2 (n=6152).
Model 4 includes race/ethnicity and adjusted for age group and log-transformed serum ferritin (ng/mL) (n=6937).
DISCUSSION
More than seven years after FDA’s updated eligibility rules raised the possibility of US blood centers’ collecting blood from female donors with Hb of 12.0 g/dL, no blood collection organization (BCO) has, to the authors’ knowledge, received agency approval. While public discussion at advisory meetings and the text of the rule itself acknowledge 12.0 g/dL as within the population distribution of normal,[10, 11] concerns have focused on the higher likelihood of iron deficiency in females with low-normal hemoglobin levels. Leveraging data from a large national complex survey and excluding individuals with identifiable triggers for donation ineligibility, this study assessed the iron status and hemoglobin levels of a reference population to make comparisons by race/ethnicity, across the full range of hemoglobin levels. We found that while lower hemoglobin values provide some insight as to the likelihood of iron deficiency, there is notable variability by race/ethnicity in distributions of Hb and SF, and a Hb eligibility cutoff of 12.5 g/dL may not be suitable for candidate female donors who identify as NH Black.
The finding that average hemoglobin levels are lower in NH Black than in NH White females has been recognized in earlier NHANES surveys,[9, 21] and it is consistent with a statistically adjusted deferral rate for low hemoglobin twice as large or more in NH Black than in NH White donors of both sexes.[7, 8, 18, 22] Data from NHANES-III (collected from 1988–1994), combined with a large health center dataset, informed recommendations in 2006 for lower limits of normal hemoglobin varying by race, sex, and age. After excluding subjects with markers of inflammation and renal disease, Beutler and Waalen recommended a lower limit of normal (5th percentile) of 12.2 g/dL for White women (both 20–49 and 50 years old or greater) and 11.5 g/dL for Black women of both age groups. A contemporaneous publication from the CDC also used NHANES-III as reference data to establish 12.0 g/dL as the boundary between normal and anemic for non-pregnant women age 15 years and older.[9] Similar to adjustments for altitude and smoking, the CDC noted the appropriateness of an adjustment for NH Black adults, commenting that the differences in Hb distributions were not accounted for by socioeconomic status nor biomarkers for iron. Despite a recommendation by the National Academy of Medicine for incorporation of an adjustment of 0.8 g/dL into reference intervals for Hb and anemia,[23] the CDC refrained from adopting race-specific criteria because the reason(s) for the different distributions remained unexplained. Ongoing efforts continue to examine the molecular basis and historical drivers of differences in iron hemostasis and red cell traits based on genetic ancestry.[24–29]
The NHANES data from 1999–2010 summarized here are consistent with the 1988–1994 NHANES data by showing lower mean Hb, higher prevalence of anemia, and lower SF among NH Black females than among NH White females. Our presentation of the joint distribution of Hb and SF by race/ethnicity (Table 3, Figure 2), together with multivariable regression analysis (Tables 4–5), nonetheless suggests reconsideration of Hb eligibility criteria required for NH Black females presenting to donate. The higher aggregate prevalence of SF <12 ng/mL in NH Black females is accounted for by the higher proportion of NH Black females with Hb levels at the lower end of the range. After accounting for Hb level (Table 4), NH Black females had a significantly lower risk for SF <12 ng/mL than NH White females. The net effect is that NH Black females with a Hb of 12.0–12.4 h/dL, which is below the current eligibility cutoff, have stronger indicators of iron status than do NH White or Hispanic females with a Hb value of 12.5–12.9 g/dL, which is within the current range of donation eligibility. At least one blood center, Hema-Quebec in Canada, has aligned its eligibility criteria for Black females with natural Hb distributions in order to better support the transfusion needs of the local Black population. Hema-Quebec accepts Black females with Hb as low as 11.5 g/dL, provides post-donation iron supplements sufficient to replenish the amount of iron lost in the donation, and conducts a post-donation ferritin test.[30] The stronger iron profile of NH Black females with Hb 12.0–12.4 g/dL as shown here compared to NH White or Hispanic females with Hb in the allowable range would suggest an acceptable level of safety for post-donation ferritin testing for at least the 16-to-49-year-old population examined here, and quite possibly for all age ranges.[8, 31]
Over the last 20 years, blood collectors, regulators and funding sponsors have invested considerable resources to characterize and to mitigate donation-associated iron depletion. Across multiple jurisdictions, findings have shown that iron depletion is common, particularly in females and in frequent donors, with large, identifiable sub-groups of donors having prevalence >50% for SF <26ng/mL and an appreciable share with SF <12ng/mL.[32, 33] In females, estimates from the US and Canada indicate that 20% or more of donations from female donors occurred with the donor’s SF <12 ng/mL.[8, 31, 34] Against that backdrop, the prevalence of SF <12ng/mL in this study’s female population with Hb 12.0–12.4 g/dL is reassuring if considering them as candidates for blood donation: 12.0% in Black females, 21.0% in White females, and 23.2% in Hispanic females. A post-donation ferritin test, with downstream donor management (e.g., support for iron supplementation, lengthened donation intervals),[35–38] arguably adds an appreciable margin of safety compared to current donor management practices.
Even as interested stakeholders have worked to reduce the prevalence of donation-associated iron depletion in recent years, it bears mention that studies to date find little support for harm resulting from donation by non-anemic blood donors with low SF. A recent randomized trial in the US assessed cognition in iron-deficient blood donors, finding no difference in cognitive performance of subjects randomized to blinded IV iron repletion vs those receiving placebo treatment.[39] A lack of a negative impact on cognitive function from blood donation were also reported by the INTERVAL trial in England and a study from Denmark.[40, 41] Investigators in Denmark and Canada assessed risk for low birth weight and other adverse outcomes in pregnancy following blood donation; these studies reported no serious outcomes associated with blood donation or iron deficiency.[42, 43] The INTERVAL trial did find increased incidence of tiredness and related symptoms in donors assigned to donate at shorter interdonation intervals, but outcomes of this nature were not serious nor unexpected.[40]
This study has some limitations. First, most of our analyses excluded NHANES subjects who reported donating blood within the prior year, given that recent donation would have altered (i.e., lowered) estimates of Hb and SF.[12] Given the “healthy donor effect,” or the recognition that blood donors on average are healthier than the general population,[44, 45] our population estimates may be mis-specified to some degree. We applied exclusion criteria based on several current donation eligibility criteria, but we did not have the ability to exclude all blood donation criteria as evaluated by the Donor Health Questionnaire. Hence, our hypothetical donor cohort may be less healthy than those who might choose to present for blood donation. Second, the data are from years 1999–2010; however, serum ferritin levels and hemoglobin values do not significantly change over time in a given population. Our analysis is limited to females up to age 49, and it excludes males. This limitation reflects restriction of ferritin testing in the 1999–2010 NHANES to populations considered at highest risk for iron deficiency, which includes children aged 1 to 5 years of both sexes and pre-menopausal females. In addition, this study relied on broad pre-coded categorizations of race/ethnicity that may have masked differences within each race/ethnicity group examined. Further, these data do not explore blood iron status in NH Asian, multiracial, or other racial/ethnic groups and thus the results described here cannot be applied to these populations. Given that iron stores and Hb levels of females tend to increase after childbearing years,[7, 8, 18, 31] population averages presented here may not be applicable to females of older age, though we would expect iron levels to be similar or possibly improved in older females. Finally, there are some limitations to use of ferritin as a biomarker for iron, given that as an acute-phase reactant it may be elevated due to infection, inflammation, malignancy and other conditions.[46]
In the current era of highly configurable blood collection organization computer systems and increasing personalization of medicine, applying our understanding of differences across donor sub-populations can help support the availability of blood products and the health needs of an underserved population.
Highlights.
Black females have mean hemoglobin levels ≈1g/dL lower than White females
Adjusted prevalence of low hemoglobin is 4-fold greater in Black vs White females
Differing hemoglobin distributions by race are not explainable by iron status
Hemoglobin eligibility cutoff of 12.0 g/dL may be suitable for Black female donors
Financial support
This was supported in part from the National Institute of Allergy and Infectious Diseases (R01AI120938 [AART], R01AI128779 [AART] and T32AI102623 [EUP]), the National Institute on Drug Abuse (F31DA054849 [EUP]), the National Institute of Diabetes and Digestive and Kidney Diseases (R01DK131926 [AART]), and the American Red Cross Scientific Affairs.
Footnotes
Conflict of Interest
BRS serves on the advisory board of HemaStrat. AART has received consulting or speaking fees from Ashland Global.
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.
References
- 1.Fasano RM, Meyer EK, Branscomb J, White MS, Gibson RW, Eckman JR. Impact of Red Blood Cell Antigen Matching on Alloimmunization and Transfusion Complications in Patients with Sickle Cell Disease: A Systematic Review. Transfus Med Rev 2019; 33(1):12–23. 10.1016/j.tmrv.2018.07.003. [DOI] [PubMed] [Google Scholar]
- 2.Waldis SJ, Uter S, Kavitsky D, Flickinger C, Vege S, Friedman DF, et al. Rh alloimmunization in chronically transfused patients with thalassemia receiving RhD, C, E, and K matched transfusions. Blood Adv 2021; 5(3):737–744. 10.1182/bloodadvances.2020003732. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Patel EU, Bloch EM, Grabowski MK, Goel R, Lokhandwala PM, Brunker PAR, et al. Sociodemographic and behavioral characteristics associated with blood donation in the United States: a population-based study. Transfusion 2019; 59(9):2899–2907. 10.1111/trf.15415. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Shaz BH, Hillyer CD. Minority donation in the United States: challenges and needs. Curr Opin Hematol 2010; 17(6):544–549. 10.1097/MOH.0b013e32833e5ac7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Mast AE, Schlumpf KS, Wright DJ, Custer B, Spencer B, Murphy EL, et al. Demographic correlates of low hemoglobin deferral among prospective whole blood donors. Transfusion 2010; 50(8):1794–1802. 10.1111/j.1537-2995.2010.02649.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Custer B, Schlumpf K, Simon TL, Spencer BR, Wright DJ, Wilkinson SL, et al. Demographics of successful, unsuccessful and deferral visits at six blood centers over a 4-year period. Transfusion 2012; 52(4):712–721. 10.1111/j.1537-2995.2011.03353.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Cable RG, Glynn SA, Kiss JE, Mast AE, Steele WR, Murphy EL, et al. Iron deficiency in blood donors: the REDS-II Donor Iron Status Evaluation (RISE) study. Transfusion 2012; 52(4):702–711. 10.1111/j.1537-2995.2011.03401.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Spencer BR, Johnson B, Wright DJ, Kleinman S, Glynn SA, Cable RG, et al. Potential impact on blood availability and donor iron status of changes to donor hemoglobin cutoff and interdonation intervals. Transfusion 2016; 56(8):1994–2004. 10.1111/trf.13663. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Centers for Disease Control and Prevention. Recommendations to prevent and control iron deficiency in the United States. Centers for Disease Control and Prevention. MMWR Recomm Rep 1998; 47(Rr-3):1–29 [PubMed] [Google Scholar]
- 10.United States Food and Drug Administration. Requirements for Blood and Blood Components Intended for Transfusion or for Further Manufacturing Use. In. Edited by Services HaH, vol. 80. Washington D.C.: Federal Register; 2015: 29841–29906. [PubMed] [Google Scholar]
- 11.Food and Drug Administration C. Blood Products Advisory Committee Meeting Transcript November 17, 2016. In. Silver Spring, MD: Center for Biologics Evaluation and Research; 2016. [Google Scholar]
- 12.Patel EU, White JL, Bloch EM, Grabowski MK, Gehrie EA, Lokhandwala PM, et al. Association of blood donation with iron deficiency among adolescent and adult females in the United States: a nationally representative study. Transfusion 2019; 59(5):1723–1733. 10.1111/trf.15179. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Eligibility Criteria Alphabetical Listing [https://www.redcrossblood.org/donate-blood/how-todonate/eligibility-requirements/eligibility-criteria-alphabetical.html]
- 14.National Health and Nutrition Examination Survey. 2003–2004 Data documentation, Codebook, and Frequencies. Data File: FERTIN_F.xpt. Centers for disease Control and Prevention website. [https://wwwn.cdc.gov/Nchs/Nhanes/2003-2004/L06TFR_C.htm] [Google Scholar]
- 15.National Health and Nutrition Examination Survey. 2009–2010 Data documentation, Codebook, and Frequencies. Data File: L06TFR_C.xpt. [https://wwwn.cdc.gov/Nchs/Nhanes/20092010/FERTIN_F.htm]
- 16.Johnson CL, Paulose-Ram R, Ogden CL, Carroll MD, Kruszon-Moran D, Dohrmann SM, et al. National health and nutrition examination survey: analytic guidelines, 1999–2010. Vital Health Stat 2 2013(161):1–24 [PubMed] [Google Scholar]
- 17.Kiss JE, Brambilla D, Glynn SA, Mast AE, Spencer BR, Stone M, et al. Oral iron supplementation after blood donation: a randomized clinical trial. JAMA 2015; 313(6):575–583. 10.1001/jama.2015.119. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Cable RG, Glynn SA, Kiss JE, Mast AE, Steele WR, Murphy EL, et al. Iron deficiency in blood donors: analysis of enrollment data from the REDS-II Donor Iron Status Evaluation (RISE) study. Transfusion 2011; 51(3):511–522. 10.1111/j.1537-2995.2010.02865.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Rigas AS, Sørensen CJ, Pedersen OB, Petersen MS, Thørner LW, Kotzé S, et al. Predictors of iron levels in 14,737 Danish blood donors: results from the Danish Blood Donor Study. Transfusion 2014; 54(3 Pt 2):789–796. 10.1111/trf.12518. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Goldman M, Uzicanin S, Scalia V, O’Brien SF. Iron deficiency in Canadian blood donors. Transfusion 2014; 54(3 Pt 2):775–779. 10.1111/trf.12380. [DOI] [PubMed] [Google Scholar]
- 21.Beutler E, Waalen J. The definition of anemia: what is the lower limit of normal of the blood hemoglobin concentration? Blood 2006; 107(5):1747–1750. 10.1182/blood-2005-07-3046. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Browne A, Fisher SA, Masconi K, Smith G, Doree C, Chung R, et al. Donor Deferral Due to Low Hemoglobin-An Updated Systematic Review. Transfus Med Rev 2020; 34(1):10–22. 10.1016/j.tmrv.2019.10.002. [DOI] [PubMed] [Google Scholar]
- 23.Institute of Medicine (US) Committee on the Prevention Detection and Management of Iron Deficiency Anemia Among U.S. Children and Women of Childbearing Age. Iron Deficiency Anemia: Recommended Guidelines for the Prevention, Detection, and Management Among U.S. Children and Women of Childbearing Age: National Academies Press; 1993. [PubMed] [Google Scholar]
- 24.Hu Y, Stilp AM, McHugh CP, Rao S, Jain D, Zheng X, et al. Whole-genome sequencing association analysis of quantitative red blood cell phenotypes: The NHLBI TOPMed program. Am J Hum Genet 2021; 108(5):874–893. 10.1016/j.ajhg.2021.04.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Hodonsky CJ, Baldassari AR, Bien SA, Raffield LM, Highland HM, Sitlani CM, et al. Ancestry-specific associations identified in genome-wide combined-phenotype study of red blood cell traits emphasize benefits of diversity in genomics. BMC Genomics 2020; 21(1):228. 10.1186/s12864-020-6626-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Bell S, Rigas AS, Magnusson MK, Ferkingstad E, Allara E, Bjornsdottir G, et al. A genome-wide meta-analysis yields 46 new loci associating with biomarkers of iron homeostasis. Commun Biol 2021; 4(1):156. 10.1038/s42003-020-01575-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Kang W, Barad A, Clark AG, Wang Y, Lin X, Gu Z, et al. Ethnic Differences in Iron Status. Adv Nutr 2021; 12(5):1838–1853. 10.1093/advances/nmab035. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Goheen MM, Campino S, Cerami C. The role of the red blood cell in host defence against falciparum malaria: an expanding repertoire of evolutionary alterations. Br J Haematol 2017; 179(4):543–556. 10.1111/bjh.14886. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Goheen MM, Wegmüller R, Bah A, Darboe B, Danso E, Affara M, et al. Anemia Offers Stronger Protection Than Sickle Cell Trait Against the Erythrocytic Stage of Falciparum Malaria and This Protection Is Reversed by Iron Supplementation. EBioMedicine 2016; 14:123–130. 10.1016/j.ebiom.2016.11.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Black women and blood donation. [https://www.hema-quebec.qc.ca/sang/savoir-plus/femmes-noires-et-don-de-sang/remplacement-pertes-fer.en.html]
- 31.Spencer BR, Guo Y, Cable RG, Kiss JE, Busch MP, Page GP, et al. Iron status and risk factors for iron depletion in a racially/ethnically diverse blood donor population. Transfusion 2019; 59(10):3146–3156. 10.1111/trf.15448. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Spencer BR, Mast AE. Iron status of blood donors. Curr Opin Hematol 2022. 10.1097/moh.0000000000000733. [DOI] [PMC free article] [PubMed]
- 33.Spencer BR, Simon TL. Chronic effects of blood and plasma transfusion. In: Rossi’s Principles of Transfusion Medicine. 6th edn. Edited by Simon TL, Gehrie EA, McCullough J, Roback J, Snyder E. Hoboken, NJ: John Wiley & Sons; 2022, pp 56–64. [Google Scholar]
- 34.Goldman M, Uzicanin S, Osmond L, Scalia V, O’Brien SF. A large national study of ferritin testing in Canadian blood donors. Transfusion 2017; 57(3):564–570. 10.1111/trf.13956. [DOI] [PubMed] [Google Scholar]
- 35.Gorlin J, Katz L, Elsmore D, Kirbach K, Erickson Y, Hove A, et al. Prevalence of blood donor iron deficiency and feasibility ferritin-based iron replacement: a blood collection agency-based study. Vox Sang 2016; 111(2):206–208. 10.1111/vox.12408. [DOI] [PubMed] [Google Scholar]
- 36.Kiss JE, Vassallo RR. How do we manage iron deficiency after blood donation? Br J Haematol 2018; 181(5):590–603. 10.1111/bjh.15136. [DOI] [PubMed] [Google Scholar]
- 37.Cable RG, Birch RJ, Spencer BR, Wright DJ, Bialkowski W, Kiss JE, et al. The operational implications of donor behaviors following enrollment in STRIDE (Strategies to Reduce Iron Deficiency in blood donors). Transfusion 2017; 57(10):2440–2448. 10.1111/trf.14226. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Spencer BR, Haynes JM, Notari EPt, Stramer SL. Prevalence, risk factors, and ferritin testing to mitigate iron depletion in male plateletpheresis donors. Transfusion 2020; 60(4):759–768. 10.1111/trf.15729. [DOI] [PubMed] [Google Scholar]
- 39.Hod EA, Brittenham GM, Bitan ZC, Feit Y, Gaelen JI, La Carpia F, et al. A randomized trial of blood donor iron repletion on red cell quality for transfusion and donor cognition and wellbeing. Blood 2022. 10.1182/blood.2022017288. [DOI] [PMC free article] [PubMed]
- 40.Di Angelantonio E, Thompson SG, Kaptoge S, Moore C, Walker M, Armitage J, et al. Efficiency and safety of varying the frequency of whole blood donation (INTERVAL): a randomised trial of 45 000 donors. Lancet 2017; 390(10110):2360–2371. 10.1016/s0140-6736(17)31928-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Rigas AS, Pedersen OB, Rostgaard K, Sorensen E, Erikstrup C, Hjalgrim H, et al. Frequent blood donation and offspring scholastic attainment: an assessment of long-term consequences of prenatal iron deficiency. Transfusion 2019; 59(5):1717–1722. 10.1111/trf.15193. [DOI] [PubMed] [Google Scholar]
- 42.Rigas AS, Pedersen OB, Sørensen E, Thørner LW, Larsen MH, Katz LM, et al. Frequent blood donation and offspring birth weight-a next-generation association? Transfusion 2019; 59(3):995–1001. 10.1111/trf.15072. [DOI] [PubMed] [Google Scholar]
- 43.Germain M, Delage G, Robillard P, Katz LM, Grégoire Y. The association between frequency of blood donation and the occurrence of low birthweight, preterm delivery, and stillbirth: a retrospective cohort study. Transfusion 2016; 56(11):2760–2767. 10.1111/trf.13762. [DOI] [PubMed] [Google Scholar]
- 44.Atsma F, Veldhuizen I, Verbeek A, de Kort W, de Vegt F. Healthy donor effect: its magnitude in health research among blood donors. Transfusion 2011; 51(8):1820–1828. 10.1111/j.1537-2995.2010.03055.x. [DOI] [PubMed] [Google Scholar]
- 45.van den Hurk K, Zalpuri S, Prinsze FJ, Merz EM, de Kort W. Associations of health status with subsequent blood donor behavior-An alternative perspective on the Healthy Donor Effect from Donor InSight. PLoS One 2017; 12(10):e0186662. 10.1371/journal.pone.0186662. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Fletcher A, Forbes A, Svenson N, Wayne Thomas D. Guideline for the laboratory diagnosis of iron deficiency in adults (excluding pregnancy) and children. Br J Haematol 2022; 196(3):523–529. 10.1111/bjh.17900. [DOI] [PubMed] [Google Scholar]