Anemia, defined by the WHO as a hemoglobin (Hb) level of less than 12 g/dL in women and less than 13 g/dL in men,1 is a common problem among older adults in general. The Third National Health and Nutrition Examination Survey (NHANES III) found the prevalence of anemia (using WHO criteria) among community-dwelling older adults aged 65 years or older to be 11% and 10% among men and women, respectively, doubling to more than 20% among those older than 85 years.2 Prevalence rates of anemia among patients with cancer have been estimated to be much higher, with rates as high as 39% among older patients at enrollment in the European Cancer Anemia Survey3 and 100% in patients with cancer undergoing chemotherapy.4
The cause of chronic anemia in older adults is likely multifactorial. In the NHANES III study,2 nutritional deficiencies, particularly iron, vitamin B12, and folate deficiency, accounted for 35% of diagnosed cases of anemia. Anemia of chronic disease or chronic inflammation and unexplained anemia each accounted for approximately a third of diagnosed cases of anemia in the same study.2 Among older adults with cancer, the cause of chronic anemia extends to the effect of cancer per se and side effects of chemotherapy. It has also been hypothesized that aging may directly contribute to the development of anemia through age-related dysregulation of proinflammatory cytokines, such as interleukin-6 (IL-6).5
Regardless of the cause of anemia among older adults, uncorrected anemia has been found to be associated with poor functional status, specifically functional disability,6 and with increased mortality.7 However, the association between anemia and functional status among older adults with cancer has not been well characterized. Although a few studies in older adults with cancer have demonstrated an association between anemia and performance status8–10 (using either the ECOG performance status [PS] and Karnofsky performance scale [KPS] for assessment), very few studies11 have actually examined the association between anemia and activities of daily living (ADLs) among older adults with cancer. This is clinically relevant because ECOG PS and KFS may underestimate the extent of functional impairment among older adults, and self-reported ADLs and instrumental activities of daily living (IADLs) present a more comprehensive approach for assessing the functional status of older adults with cancer.12 It is in this context that we sought to examine the association between anemia and functional disability (the need for assistance with ≥1 IADL) among a cohort of older adults with cancer.
Patients and Methods
Study Design and Patient Population
This study is a cross-sectional analysis of data derived from a multicenter prospective study that identified clinical and biological predictors of chemotherapy toxicity among 500 patients aged 65 years and older with cancer. Details of the study methods have been described elsewhere.13 Briefly, between November 2006 and November 2009, study participants were recruited from ambulatory oncology clinics from 7 institutions. To be eligible, study participants had to be aged 65 years or older, diagnosed with stage I–IV cancer of any tumor type, and had to be scheduled to receive chemotherapy. The current secondary analysis was limited to patients with available data on serum Hb level and functional status. The study was approved by the respective Institutional Review Boards of each participating institution.
Study Procedures and Data Collection
Patients who provided informed consent completed a brief comprehensive geriatric assessment (CGA) before initiation of a chemotherapy regimen. Measures completed by the health care provider included the Karnofsky Physician-Rated Performance Rating Scale (MD-KPS)14 and the Blessed Orientation-Memory-Concentration (BOMC) test,15 a screening instrument for cognitive function. Measures that were self-completed by patients included the KPS Self-Reported Performance Rating Scale (SR-KPS)16 and the Older American Resources and Services (OARS) IADL scale.17 Patient self-completed measures included comorbidity (the physical health section of the OARS17), psychological state (the Hospital Anxiety and Depression Scale [HADS]18), social support (Medical Outcomes Study [MOS] Social Support Survey19), social activity (MOS Social Activity Subscale20), MOS SF-36 Physical Functioning Scale,21,22 nutritional status, and falls. In addition, sociodemographic data were captured using a self-reported questionnaire. Medical records were abstracted to obtain pretreatment Hb levels, and tumor and treatment characteristics.
Primary Outcome Variable: The primary outcome variable was functional disability before the first cycle of chemotherapy, defined as the need for assistance with at least one IADL.23,24 IADLs measure the need for assistance with shopping, using the telephone, managing medications, housekeeping, laundry, transportation, managing finances, and preparing meals.17,23 Scores range from 0 to 14, with lower scores denoting that assistance is needed with these tasks. For the purposes of the current analyses, scores were dichotomized as functional disability (Yes or No), with scores of less than the 14-point maximum denoting functional disability.
Independent Variable: The independent variable was anemia ascertained before the first cycle of chemotherapy, and defined according to the WHO criteria1 as a serum Hb concentration of less than 12 g/dL in women (Yes or No) and less than 13 g/dL in men (Yes or No).
Sociodemographic variables included age as a continuous variable; race (white, black, Asian, other); sex; marital status (single, married, separated/divorced, widowed); educational status (≤8th grade, 9–12th grade, >12th grade); living companion (spouse/partner/children, alone); and job status (employed, retired, homemaker, other). Tumor characteristics included cancer type (breast, gastrointestinal, gynecologic malignancies, lung, genitourinary, other) and stage (I, II, III, IV). Geriatric variables included comorbidity count (<3, ≥3); unintentional weight loss of 10% or greater in the past 6 months (Yes or No); social support measured with the MOS Social Support subscale, with scores dichotomized as less than 50 versus 50 or greater to denote lower versus higher social support, respectively; psychological state measured with the HADS, with scores dichotomized as less than 15 versus 15 or greater to denote absence versus presence of anxiety/depression, respectively; and cognitive status measured with the BOMC, with scores dichotomized as less than 11 versus 11 or greater denoting absence versus presence of cognitive impairment, respectively. Other measures of physical function included the MD-KPS, SR-KPS, and MOS SF-36 Physical Functioning Scale,with scores for all 3 instruments dichotomized as less than 70 versus 70 or greater, with higher scores denoting better function.
We conducted descriptive analysis to examine participants' baseline characteristics. We conducted bivariate analyses of all baseline variables, comparing patients who had anemia versus patients who did not have anemia. The chi-square test or Fisher exact test, when appropriate, was used to determine whether a statistically significant difference existed in the distribution of baseline characteristics between the groups.
We also examined the bivariate association between MD-KPS and MOS Physical Functioning scores and anemia status to demonstrate a consistent association of anemia with other measures of physical function aside from IADL scores. Furthermore, we categorized serum Hb levels into quartiles and examined the relation between Hb quartiles and IADL scores.
We developed our multivariable model by first identifying potential confounders of the association between anemia and functional disability. A variable was considered a potential confounder if it was significantly associated with both functional disability and anemia at a P value of less than .05 on bivariate analysis. We then developed a base logistic regression model with anemia as the independent variable and functional disability as the dependent variable. Potential confounders that changed the β coefficient for anemia in the base model by at least 10% were then retained in our final multivariate logistic regression model. We did not include other measures of physical function (MD-KPS and MOS Physical Functioning) in the models because of colinearity. All P values presented are 2-sided. All analyses were conducted using SAS 9.2 (SAS Institute Inc., Cary, NC).
Participants' Baseline Characteristics
Baseline characteristics are displayed in Table 1. Of 491 evaluable patients, the mean age of participants was 73.1 years (SD, 6.1), 85% were white, 56% were female, and most (62%) had more than a high school education. The most common types of cancer included lung (29%), and most (62%) had stage IV cancer. At the time of study enrollment, 44% of participants had 3 or more comorbidities. The prevalence of functional disability and anemia were 43% and 51%, respectively.
Participants' Baseline Characteristics According to Anemia Status
The bivariate association between anemia and baseline characteristics is displayed in Table 1. Compared with participants who did not have anemia before receiving the first cycle of chemotherapy, participants who had anemia were more likely to be older (mean age, 73.7 years [SD, 6.3] vs 72.4 years [SD, 5.9]; P=.02), black (10% vs 6%; P=.05), and male (53% vs 39%; P<.0001), and to have a gastrointestinal malignancy (34% vs 20%; P=.0002), have stage IV disease (70% vs 53%; P<.0001), have unintentional weight loss (61% vs 43%; P<.0001), score 15 or greater on the HADS (20% vs 11%; P=.02), and score 11 or greater on the BOMC (10% vs 2%; P=.0005).
Anemia and Functional Disability
Compared with participants who did not have anemia, participants who had anemia were more likely to
Baseline Characteristics by Anemia Status
Anemia and Other Measures of Physical Function
We undertook additional analyses to determine whether the association between anemia and other measures of physical function would be consistent with the association between anemia and functional disability (Table 1). Compared with participants who did not have anemia, those who had anemia had a lower meanscore on the MOS Physical Functioning subscale (62 vs 75; P<.001), MD-KPS (82 vs 88; P<.001), and SR-KPS (83 vs 87; P<.001). These results demonstrate a consistent and robust association between anemia and physical function, irrespective of whether physical function is patient self-reported or provider-rated.
The independent association between anemia and functional disability is displayed in Table 2. Baseline characteristics that had a significant bivariate association with both anemia and functional disability and also accounted for a 10% change in the crude odds ratio (OR) of the association between anemia and functional disability included sex, stage, and unintentional weight loss (data not
Association Between Anemia and Functional Disability
In this cohort of older patients from multiple academic centers with stage I–IV cancer, anemia as defined by the WHO criteria was highly prevalent. In addition, we found that anemia was independently associated with functional disability and that the independent association between anemia and functional disability was not explained by sex, cancer stage, or unintentional weight loss within the past 6 months. Older patients with anemia were more than twice as likely to have functional disability than those without anemia.
Our study results showing that anemia is independently associated with functional disability is consistent with results from studies conducted among patients without cancer. In a cross-sectional study of 1156 community-dwelling older adults aged 65 years and older, Penninx et al6 demonstrated an independent association between anemia (according to WHO criteria) and functional disability. However, results from prospective studies have been mixed. Although the longitudinal association between anemia and physical performance (hand grip and knee extensor strength) was demonstrated in the InCHIANTI6 population-based study of community-dwelling older adults and in the Established Population for Epidemiologic Studies of the Elderly (EPESE)7 study, the Leiden 85-plus prospective cohort study25 conducted among 562 community-dwelling older adults aged 85 years and older failed to show an association between anemia and functional decline. Among older adults with cancer, few studies have demonstrated an association between anemia and physical performance, and most of these studies have not specifically evaluated the association between anemia and functional status.8,10,11,26 In a cross-sectional study of 3,523 women with breast and gynecologic cancers, women with anemia were more likely to have poor performance status (ECOG PS of 3 or 4) compared with women without anemia.26 Our study results demonstrate that the association between anemia and physical performance among older adults with cancer extends to functional disability.
The underlying mechanism through which anemia may affect muscle strength, physical performance, and functional disability have not been well examined. However, one hypothesis is that anemia may result in diminished muscle oxygenation, leading to impaired muscle strength, poor physical performance, and, subsequently, to functional disability. Another hypothesis suggests that the association between anemia in older adults and poor functional outcomes may be partly explained by chronic subclinical inflammation. Aging is accompanied by increasing serum levels of inflammatory cytokines and acute-phase proteins, and chronic diseases that may all contribute to chronic subclinical inflammation.27,28 Evidence from cross-sectional studies has shown an inverse association between biomarkers of inflammation and sarcopenia (loss of skeletal muscle mass and strength), slower walking speed, and functional disability.29–32 Prospective studies have also shown that IL-6 and C-reactive protein (CRP) levels and other inflammatory markers independently predict incident mobility limitations and decreases in physical function in older adults.33,34 Additionally, with increasing frailty, levels of IL-6 and CRP increase.35 A plausible explanation for the association between inflammation and functional outcomes is the catabolic effect of inflammation on muscle. Several specific cytokines upregulate the inflammatory response,36 which then induces hypercatabolism, stimulates protein breakdown, and suppresses muscle synthesis, leading to sarcopenia, impaired physical performance, and functional disability. Although the direct causal link between anemia and chronic subclinical inflammation remains to be firmly established, older adults with anemia have been shown to have higher serum levels of inflammatory biomarkers, such as CRP and IL-6, than those without anemia.6 In light of this background, chronic subclinical inflammation, a condition commonly found among older anemic adults, might be a key biological mechanism contributing to the functional disability among older adults with anemia.
Our study has several limitations that deserve comment. First, the study design is cross-sectional, which precludes us from inferring that anemia causes functional disability. Longitudinal studies and randomized controlled trials conducted among older adults with cancer are warranted to address this question. Second, our study did not determine the level of Hb that was associated with functional disability. Prior work, completed by Chaves et al,37 suggests that mildly low and even low-normal Hb levels—levels considered normal by the current WHO definition—may not be benign, and could be associated with increased mortality. Third, all study participants were recruited from academic centers, and most were non-Hispanic white and had an associate degree, suggestive that our patient population had a higher socioeconomic status. This may limit the generalizability of our results. However, our results are consistent with existing data from older community-dwelling adults, and therefore our results can be deemed generalizable. Fourth, the heterogeneity of the patient population may have introduced unmeasured confounders into the study and may make it difficult to draw conclusions regarding specific tumor types. Last, our study did not collect any data on biomarkers of inflammation or on fatigue, and therefore could not account for the mediating role of chronic subclinical inflammation or fatigue on the association between anemia and functional disability in this patient population. Hardy and Studenski38 demonstrated in a study of older community-dwelling adults that fatigue was independently associated with functional deficits that persisted for years.
This study shows that anemia was highly prevalent and independently associated with functional disability among this cohort of older adults with cancer. Longitudinal studies among older adults with cancer would be helpful in establishing a causal link between anemia and functional disability. In addition, given that functional status is a key factor for treatment tolerance, randomized controlled trials are warranted to evaluate whether treating anemia can improve functional status and treatment tolerance among older adults with cancer. Although 30% of all anemia in older adults may be unexplained,2 and therefore is not readily corrected, and the use of erythropoietin in patients with cancer may be associated with increased mortality39,40 and is therefore not recommended by current guidelines, a significant proportion of anemia in older adults is caused by nutritional deficiencies2 and is readily modifiable. Modifiable causes of anemia offer an opportunity to evaluate whether treating anemia can improve functional status and treatment tolerance among older adults with cancer.
Dr. Mohile has disclosed that she is a consultant for Seattle Genetics, Inc. Dr. Gross has disclosed that he is an advisory board member for FAIR Health, Inc. Dr. Gajra has disclosed that he receives research support from Celgene Corporation and is on the advisory board for Bayer and Celgene Corporation. Dr. Hurria has disclosed that she receives research support from Celgene Corporation and GlaxoSmithKline, and is a consultant for GTx, Inc., Boehringer Ingelheim Pharmaceuticals, and Carevive Systems.
The remaining authors have disclosed that they have no financial interests, arrangements, affiliations, or commercial interests with the manufacturers of any products discussed in this article or their competitors. HJC is supported by NIH Grant 5U01AG034661.
This study was supported by K23 AG026749-01 (PI: Hurria) Paul Beeson Career Development Award in Aging Research), and American Society of Clinical Oncology—Association of Specialty Professors—Junior Development Award in Geriatric Oncology (PI: Hurria).
Abstract presentation: Owusu C, Tew W, Hardt M, et al. Anemia and functional disability in older adults with cancer. Presented at the 2012 ASCO Annual Meeting; June 1–5, 2012; Chicago, Illinois.
Nutritional Anaemias. Report of a WHO Scientific Group. Available at: http://apps.who.int/iris/bitstream/10665/40707/1/WHO_TRS_405.pdf. Accessed September 21 2015.
GuralnikJMEisenstaedtRSFerrucciL. Prevalence of anemia in persons 65 years and older in the United States: evidence for a high rate of unexplained anemia. Blood2004;104:2263–2268.
LudwigHVan BelleSBarrett-LeeP. The European Cancer Anaemia Survey (ECAS): a large, multinational, prospective survey defining the prevalence, incidence, and treatment of anaemia in cancer patients. Eur J Cancer2004;40:2293–2306.
PenninxBWPahorMCesariM. Anemia is associated with disability and decreased physical performance and muscle strength in the elderly. J Am Geriatr Soc2004;52:719–724.
PenninxBWPahorMWoodmanRC. Anemia in old age is associated with increased mortality and hospitalization. J Gerontol A Biol Sci Med Sci2006;61:474–479.
AoeKHirakiAMaedaT. Serum hemoglobin level determined at the first presentation is a poor prognostic indicator in patients with lung cancer. Intern Med2005;44:800–804.
BeerTMTangenCMBlandLB. Prognostic value of anemia in newly diagnosed metastatic prostate cancer: a multivariate analysis of southwest oncology group study 8894. J Urol2004;172(6 Pt 1):2213–2217.
LudwigHMuldurEEndlerGHublW. Prevalence of iron deficiency across different tumors and its association with poor performance status, disease status and anemia. Ann Oncol2013;24:1886–1892.
MancusoAMigliorinoMDe SantisS. Correlation between anemia and functional/cognitive capacity in elderly lung cancer patients treated with chemotherapy. Ann Oncol2006;17:146–150.
RepettoLFratinoLAudisioRA. Comprehensive geriatric assessment adds information to Eastern Cooperative Oncology Group performance status in elderly cancer patients: an Italian Group for Geriatric Oncology Study. J Clin Oncol2002;20:494–502.
HurriaATogawaKMohileSG. Predicting chemotherapy toxicity in older adults with cancer: a prospective multicenter study. J Clin Oncol2011;29:3457–3465.
LoprinziCLLaurieJAWieandHS. Prospective evaluation of prognostic variables from patient-completed questionnaires. North Central Cancer Treatment Group. J Clin Oncol1994;12:601–607.
KatzmanRBrownTFuldP. Validation of a short Orientation-Memory-Concentration Test of cognitive impairment. Am J Psychiatry1983;140:734–739.
KarnofskyDBurchenalJ. The clinical evalaution of chemotherapeutic agents in cancer. In: MacleodCM ed. Evaluation of Chemotherapeutic Agents. New York, NY: Columbia University Press; 1948:191–205.
FillenbaumGGSmyerMA. The development, validity, and reliability of the OARS multidimensional functional assessment questionnaire. J Gerontol1981;36:428–434.
WareJEJrSherbourneCD. The MOS 36-item short-form health survey (SF-36). I. Conceptual framework and item selection. Med Care1992;30:473–483.
SeemanTEMerkinSSCrimminsEMKarlamanglaAS. Disability trends among older Americans: National Health and Nutrition Examination Surveys, 1988-1994 and 1999-2004. Am J Public Health2010;100:100–107.
SpillmanBC. Changes in elderly disability rates and the implications for health care utilization and cost. Milbank Q2004;82:157–194.
den ElzenWPWillemsJMWestendorpRG. Effect of anemia and comorbidity on functional status and mortality in old age: results from the Leiden 85-plus Study. CMAJ2009;181:151–157.
Barrett-LeePBokemeyerCGasconP. Management of cancer-related anemia in patients with breast or gynecologic cancer: new insights based on results from the European Cancer Anemia Survey. Oncologist2005;10:743–757.
CesariMPenninxBWPahorM. Inflammatory markers and physical performance in older persons: the InCHIANTI study. J Gerontol A Biol Sci Med Sci2004;59:242–248.
BrinkleyTELengXMillerME. Chronic inflammation is associated with low physical function in older adults across multiple comorbidities. J Gerontol A Biol Sci Med Sci2009;64:455–461.
VisserMPahorMTaaffeDR. Relationship of interleukin-6 and tumor necrosis factor-alpha with muscle mass and muscle strength in elderly men and women: the Health ABC Study. J Gerontol A Biol Sci Med Sci2002;57:M326–332.
PayetteHRoubenoffRJacquesPF. Insulin-like growth factor-1 and interleukin 6 predict sarcopenia in very old community-living men and women: the Framingham Heart Study. J Am Geriatr Soc2003;51:1237–1243.
CohenHJHarrisTPieperCF. Coagulation and activation of inflammatory pathways in the development of functional decline and mortality in the elderly. Am J Med2003;114:180–187.
PenninxBWKritchevskySBNewmanAB. Inflammatory markers and incident mobility limitation in the elderly. J Am Geriatr Soc2004;52:1105–1113.
BastardJPMaachiMLagathuC. Recent advances in the relationship between obesity, inflammation, and insulin resistance. Eur Cytokine Netw2006;17:4–12.
ChavesPHXueQLGuralnikJM. What constitutes normal hemoglobin concentration in community-dwelling disabled older women?J Am Geriatr Soc2004;52:1811–1816.
HenkeMLaszigRRubeC. Erythropoietin to treat head and neck cancer patients with anaemia undergoing radiotherapy: randomised, double-blind, placebo-controlled trial. Lancet2003;362:1255–1260.
WrightJRUngYCJulianJA. Randomized, double-blind, placebo-controlled trial of erythropoietin in non-small-cell lung cancer with disease-related anemia. J Clin Oncol2007;25:1027–1032.