Short-Term Impact of Anemia on Mortality

Abstract

Objective: To estimate the short-term (14 months) impact of anemia on mortality among Mexican older adults (OAs). Method: Longitudinal analyses using data from a quasi-experimental study in a non-contributory pension program in Mexico with a sample of 3,621 OAs aged 65 to 74 years. Data on health, nutrition, life conditions, and mortality were gathered at both baseline and follow-up. Logistic regression model was used to estimate the impact of anemia and hemoglobin quintiles on mortality. Results: Overall mortality rate was 2.1%. Both mild anemia and moderate/severe anemia increased mortality risk at 14 months (odds ratio [OR] = 2.04, 95% confidence interval [CI] = [1.1, 4.1]; and OR = 6, 95% CI = [2.1, 16.9], respectively). Discussion: In the short term, degree of severity of anemia is an independent predictor of mortality risk. Because anemia is a modifiable factor, further research is required to better understand this condition in terms of main causes, prevention, treatment, and impact on OAs’ survival.

Keywords

anemia mortality Mexican older adults

Introduction

Anemia is a highly prevalent condition among older adults (OAs), especially among males. Data from National Health and Nutrition Survey in Mexico showed that the prevalence of anemia in OAs was 18.35% in males, 17.35% in females, and reached 31% in OAs aged 80 years and above. Prevalence in rural areas was 20.64% in males and 15.6% in females (Gutiérrez et al., 2012).

There is growing evidence that lower hemoglobin (Hb) concentrations constitute a pathological condition in the elderly, and not a normal consequence of aging (Nissenson, Goodnough, & Dubois, 2003). Anemia in OAs has been linked to high morbidity (Patel, 2008), poor functional status (den Elzen et al., 2009), cognitive decline and dementia (Peters et al., 2008), increased risk of hospitalization, and mortality (Riva et al., 2009). In the same vein, higher Hb concentrations have been associated with higher risk of mortality. In a sample of American OAs, risk of mortality was tied to the lowest and highest quintiles of Hb concentration (Zakai et al., 2005). Similarly, mild anemia has been identified as an independent predictor of mortality among OAs in developed countries (Peters et al., 2008).

While former studies have estimated the prognosis of anemia in severely ill patients over a short period of time (Price, Mehra, Holmes, & Schrier, 2011), whether differences exist with regard to the severity of anemia prognosis of ambulatory OAs over a short period of time has not been widely documented. The objective of this study was to estimate the short-term (14 months) impact of anemia on mortality among ambulatory OAs residing in rural localities in Mexico. In this study, we used the term ambulatory older adults to highlight that OAs were able to carry out some basics activities of daily living (walking, eating, bathing, etc.) and not to be severely ill to be hospitalized or be in bed at time of interview.

Method

Study Design and Sampling

OAs in our sample were part of the impact evaluation study of Program “70 y más” conducted in Mexico during the years 2007-2009. Implemented nationally throughout Mexico, Program 70 y más was aimed at improving the living conditions among adults aged 70 years and older by boosting their social protection through policy mechanisms. Centered on two components, 70 y más pursues a twofold objective: (a) to raise the income of the elderly and (b) to improve the social protection of the elderly. Under the first program objective, elderly receive a direct unconditional cash transfer of 500 Mexican pesos (approximately US$40) every month, which can be collected every 2 months. Following is a brief description of the program impact evaluation design.

In 2007, the Ministry of Social Development established two criteria for selecting program beneficiaries: first, being ≥70 years old and, second, residing in a rural locality of ≤2,500 inhabitants.¹ Both were used in the impact evaluation study as the basis for setting up four study groups. Power analysis and sample size calculation determined a total of 6,000 OAs distributed in four study groups (1,500 per group). According to an empirical comparison group formation analysis, it was determined that OAs aged between 65 and 74 years would be included in the evaluation study, with an intervention group composed of OAs aged 70 to 74 years residing in rural areas. The remaining OAs were distributed among three control groups. The first included OAs, aged 70 to 74 years, living in communities slightly larger than rural localities (2,501-2,700 inhabitants); the second, OAs aged 65 to 69 years, living in rural localities; and the third, OAs aged 65 to 69 years, living in localities of 2,501-2,700 inhabitants. Impact evaluation study was conducted in 516 Mexican rural localities. The 70 y más Program description, methodological details of evaluation design, and the sampling procedures and results have been previously reported (Salinas-Rodríguez et al., 2013).

We used the same analytical sample for analyses in this work, but restricting to OAs who had all biological measurements (see next paragraph), and we used the information from the four evaluation groups only as a control variable in the statistical analysis.

Data Collection

Data on health, nutrition, and life conditions were gathered both at baseline (in 2007) and at intervals of 11 (in 2008) and 14 (in 2009) months. Information on OA death was collected in 2008 and 2009 for all participants in the study. This information was obtained from the household key informant if OA not lived alone and the proxy informant if OA lived alone. The follow-up periods were planned according to the evaluation design in the original study. Information included demographic and socioeconomic data collected during interviews by trained personnel from the National Institute of Public Health of Mexico. Eligible participants numbered 6,000 OAs at baseline, and capillary blood specimens were collected by finger puncture from a sample of 4,288 OAs to measure their Hb levels in portable Hemocue photometers. However, complete information on all variables and blood samples were obtained from 3,621 OAs (Figure 1).

Figure 1.

Population study and analytical sample.

Definition of Variables

Outcome

The OA death was registered only in those cases where the household key informant or proxy informant declared that elderly has died from any cause. The causes of death were registered and classified on the basis of death certificates from the clinics nearest the OA households.

Exposure variable

We explored three different ways of analyzing the impact of anemia on mortality. First, anemia was defined as a dummy variable equal to 1 where Hb concentration, adjusted by altitude as recommended by the World Health Organization (WHO; Cohen & Hass, 1999), was <120 g/L for females and <130 g/L for males (DeMaeyer et al., 1989; WHO, United Nations University, & United Nations Children’s Fund, 2001). Second, anemia was classified by degree of severity as follows: mild anemia, 10 g/dl ≤ Hb < 12 g/dl for females and 10 g/dl ≤ Hb < 13 g/dl for males; moderate anemia 7 ≤ Hb < 10 g/dl; and severe anemia Hb < 7 g/dl (DeMaeyer et al., 1989). And third, we estimated the effect of Hb quintiles on mortality. As distribution of Hb was different for males and females, we calculate the quintiles separately for both samples. First, we use the male sample, and we calculated their quintiles, and then we did the same for females. Then, we combined both in just one variable. This procedure allowed us to recognize the biological differences among men and women regarding the Hb distribution and incorporate this information in our analysis (Zakai et al., 2005).

Control variables

We included variables at different levels of aggregation. At the individual level, we considered indigenous condition (if OAs spoke some dialect or if they reported being members of some ethnic group), age, sex, paid job, literacy, risk factors (tobacco and alcohol consumption), physical activity (measured through a short version of the International Physical Activity Questionnaire [IPAQ]; Lee, Macfarlane, Lam, & Stewart, 2011), functional dependency (defined as difficult to perform any of the basic activities of daily living; Katz et al., 1983), presence of depressive symptoms assessed by the 15-item version of the Geriatric Depression Scale (Yesavage et al., 1982), anthropometric measurements (weight and height) and body mass index (BMI) computed on the basis of weight and height (Cole, Bellizzi, Flegal, & Dietz, 2000), and presence of the following self-reported chronic conditions: hypertension, type 2 diabetes, hypercholesterolemia, arthritis, chronic pulmonary disease, cardiac disease, cancer, and osteoporosis. At household level, we included total monthly household expenditure as an indicator of socioeconomic status. At locality level, we considered variables associated with access to health services such as the nurse, physician, and clinics density (defined as the number of nurses, physicians, or clinics by 10,000 inhabitants) as well as the travel time (minutes) to the nearest health units. Localities ≤2,500 inhabitants were considered as rural.

Statistical Analysis

Bivariate comparisons between outcome variable and covariates were based on the severity levels of anemia and their distribution across Hb quintiles. A logistic regression model was used to estimate the impact of anemia and Hb quintiles on mortality. Three consecutive models were applied: Model 1 considered anemia as a dichotomous variable, Model 2 considered the three severity levels of anemia, and Model 3 considered the adjusted Hb quintiles. All three models were adjusted by the following covariables: just by sex and age, and then for the full set of previously described control variables. For all three models, we estimated all-cause mortality attributable to anemia (Kleinbaum, Kupper, & Morgenstern, 1982). All analyses were done using STATA v11 software. (StataCorp, 2009).

Ethical Review

The Research and Ethics Committees of the National Institute of Public Health of Mexico reviewed and approved the 70 y más Program Impact Evaluation Study. Participants received a detailed explanation of the procedures and signed an informed consent form prior to data collection.

Results

Baseline Characteristics

Overall prevalence of anemia reached 16% and was more prevalent in males (16.5%) than in females (14%; p = .034). Mean Hb concentrations were 147 ± 19.6 g/L in males and 137 ± 17.8 g/L in females. With regard to severity, mild anemia was recorded in the highest proportion of cases (12.6%), followed by moderate (2.15%) and severe (0.44%) anemia. Because of the very low prevalence of severe anemia, moderate and severe levels were collapsed into one category for further comparisons. In all, 2.1% of OAs with Hb information died during the study period, with 86.6% of these cases occurring by the 11-month and the rest by the 14-month follow-up visits.

Table 1 shows the population characteristics at baseline stratified by severity of anemia and Hb quintiles. The death rate was significantly higher in the anemic versus the non-anemic population (9.57% for severe/moderate and 3.93% for mild anemia vs. 1.92% for no anemia; p < .001). When stratified by Hb quintile, mortality rate was higher in Quintile 1 of Hb (4.35%) and lowest in Quintile 5 (1.22%). Main causes of death in this population were cardiovascular disease and stroke, followed by cancer and hepatic diseases (data not shown).

Table 1.

Baseline Characteristics of Older Adults by Anemia Status and Hb Quintiles.

	Severity of anemia				Hb quintilesHemoglobin rank (g/dl)
	No anemia	Mild anemia	Moderate and severe anemia		Q1 (4.5-12.7)	Q2 (12.71-13.8)	Q3 (13.81-14.7)	Q4 (14.71-15.7)	Q5 (15.7-19.91)
	%	%	%	p value^a	%	%	%	%	%	p value^a
Demographic characteristics
Mortality rate	1.92	3.93	9.57	<.001	4.35	3.06	1.47	1.93	1.22	<.001
Age (years)^b	69.3 ± 0.05	69.7 ± 0.13	69.7 ± 0.3	.026	69.8 ± 2.8	69.5 ± 2.9	69.2 ± 2.9	69.3 ± 2.9	69.4 ± 2.9	.003
Sex (females)	50.37	42.36	60.64	.055	—	—	—	—	—
Paid job	24.02	26.64	19.15	.563	24.42	26.50	24.60	22.65	23.00	<.179
Indigenous	37.02	43.62	51.69	<.001	44.84	43.06	35.71	35.75	32.32	<.001
Literacy	33.92	31.88	26.6	.16	29.17	33.66	33.33	34.39	36.54	<.006
Rural dwelling	63.12	57.42	50	.002	55.88	60.36	60.13	61.88	71.58	<.001
Total monthly household expenditure (Mexican pesos)^b	2,177.4 ± 56.5	2,152.1 ± 152.5	2,216.4 ± 355	.979	2197.1 ± 3329.9	2285.9 ± 3044.4	2044.7 ± 2754.6	2271.6 ± 3444.1	2227.7 ± 3200.7	.781
Functional status
Level of physical activity
Low	15.2	21.99	21.05		21.35	18.15	11.21	16.85	13.92	.001
Medium	26.7	19.62	25	<.001	21.51	26.46	28.40	27.72	24.40	.221
Medium–high	58.1	58.39	53.95	.003	57.14	55.38	60.39	55.44	61.68	.13
Functional dependency	22.79	24.73	32.98	.069	25.22	25.42	24.93	20.17	20.87	.005
Depressive symptoms	25.37	27.25	29.76	.268	27.12	26.51	26.65	25.57	22.95	.084
Body mass index^b	25.3 ± 0.08	24.1 ± 0.23	23.9 ± 0.4	<.001	24.3 ± 4.9	24.5 ± 4.5	25.1 ± 4.6	25.6 ± 4.6	26.2 ± 4.5	<.001
Comorbidity^c
Hypertension	25.42	23.63	30.85	.855	24.16	26.54	23.83	24.45	27.64	.338
Type 2 diabetes	9.02	11.16	13.83	.049	10.92	10.89	8.57	7.87	8.94	.044
Hypercholesterolemia	5.91	6.35	3.19	.866	5.53	5.87	5.62	5.80	6.64	.432
Arthritis	14.67	12.04	12.77	.125	11.79	16.34	16.33	12.29	14.50	.801
Chronic pulmonary disease	4.44	4.16	2.13	.472	3.64	5.17	4.55	3.45	4.88	.771
Cardiac disease	4.51	2.84	2.13	.056	2.33	4.61	5.09	4.56	4.47	.099
Cancer	0.88	0.88	0	.68	0.73	1.12	1.07	0.55	0.81	.705
Osteoporosis	11.63	10.94	17.02	.734	12.08	12.43	11.91	10.36	11.65	.44
Risk factors
Smoking
Had never smoked or smoked <100 cigarettes	74.16	70.68	82.98		74.82	73.60	74.97	73.34	73.04	.461
Had smoked 100 cigarettes, but had quit smoking	20.09	21.01	14.89	.997	18.63	21.51	19.68	19.48	21.00	.576
Smoked currently	5.75	8.32	2.13	.229	6.55	4.89	5.35	7.18	5.96	.674
Current alcohol consumption	10.7	9.61	6.38	.226	9.00	9.64	10.05	11.23	12.21	.026
Access to and coverage by health services
Travel time to the nearest first-level health unit (minutes)^b	16.5 ± 0.35	16.4 ± 1	15.4 ± 2.3	.856	14.2 ± 19.3	14.1 ± 18.4	14.8 ± 19.3	14.7 ± 19	17.5 ± 19.4	.019
Nurse and physician density of first-level health unit (for 10,000 inhabitants)^b	12.3 ± 0.2	11.1 ± 0.5	11.7 ± 1.1	.088	10.7 ± 9.7	12 ± 13.8	12.3 ± 11.5	12.2 ± 10.8	12.4 ± 13.4	.110
Clinic density of first-level health unit (for 10,000 inhabitants)^b	5.3 ± 0.09	4.8 ± 0.2	4.7 ± 0.36	.045	4.6 ± 4.1	4.9 ± 4.8	5.2 ± 5.2	5.3 ± 5.1	5.3 ± 4.4	.030
Distance to second- level health unit (minutes)^b	67.7 ± 0.8	66.3 ± 2.24	63.9 ± 5.2	.668	61.1 ± 45.2	64 ± 46.5	66 ± 48.1	65.1 ± 46.9	66.1 ± 48.4	.562
Nurse and physician density of second-level health unit (for 10,000 inhabitants)^b	6.7 ± 0.07	6.7 ± 0.19	6.73 ± 0.4	0.944	6.8 ± 4.1	6.6 ± 4.1	6.9 ± 4.2	6.8 ± 4.1	6.9 ± 4.6	.340
Clinic density of second-level health unit (for 10,000 inhabitants)^b	0.84 ± .007	0.86 ± 0.021	0.84 ± 0.04	.744	0.8 ± 0.5	0.8 ± 0.4	0.9 ± 0.4	0.8 ± 0.4	0.8 ± 0.4	.89

ANOVA test for continuous variables and proportion test for categorical variables.

Mean ± SE.

Comorbidity diseases previously diagnosed by physician.

Impact of Anemia on Mortality

Table 2 contains the results of the models used to estimate the impact of anemia on mortality. Under the first model, anemia exerted a significant impact on all-cause mortality (odds ratio [OR] = 2.56; 95% confidence interval [CI] = [1.4, 4.7]; Model 1). Both mild and moderate/severe anemia were associated with mortality (OR = 2.04 and 6.0, respectively; Model 2). In the model stratified by quintile of Hb and fully adjusted, the lowest quintile was strongly associated with mortality compared with the fourth quintile (OR = 2.93, Model 3). The all-cause mortality attributable to anemia was 19%, while for mild and moderate/severe anemia, it was 11% and 8%, respectively. For the lowest quintile of Hb, it was 24% in comparison with the fourth quintile.

Table 2.

Impact of anemia on mortality.

	Adjusted by age and sex		Fully adjusted^a		Attributable mortality
	OR	[95% CI]	OR	[95% CI]	Attributable mortality
Model 1	R² = 2.36		R² = 17.38
No anemia	1.00		1.00
Anemia	2.68	[1.8, 4.1]	2.56	[1.4, 4.7]	0.19
Model 2	R² = 2.53		R² = 17.78
No anemia	1.00		1.00
Mild anemia	2.29	[1.4, 3.7]	2.04	[1.1, 4.1]	0.11
Moderate and severe anemia	4.46	[2.2, 8.9]	6.00	[2.1, 16.9]	0.08
Model 3	R² = 3.51		R² = 18.88
Quintile 1	2.64	[1.4, 4.8]	2.93	[1.2, 7.2]	0.24
Quintile 2	1.60	[0.8, 3.1]	2.15	[0.8, 5.4]
Quintile 3	0.85	[0.4, 1.8]	0.96	[0.3, 2.8]
Quintile 4 (reference)	1	—	1.00	—
Quintile 5	0.81	[0.4, 1.7]	0.66	[0.2, 2.1]

Adjusted by all variables in Table 1. CI = confidence interval; OR = odds ratio.

Discussion

In this study, anemia was an independent predictor of mortality over 14 months follow-up in ambulatory rural older Mexican adults. Moderate and severe anemia had the strongest association with mortality in this short period. Moderate and severe anemia correlated more strongly with mortality under the fully adjusted model than under that adjusted by age and sex. This is one of the more strongest effects reported in the literature, and it is similar to that reported by den Elzen et al. (2009).

Our results are in line with those reported in the literature, even when they come from specific OAs inpatient populations (Kosiborod, Smith, Radford, Foody, & Krumholz, 2003; Landi et al., 2007; Patel, 2008; Patel et al., 2009; Riva et al., 2009); Landi et al. (2007) found that in very old and frail elderly subjects living in a nursing home, the higher levels of Hb were associated with better survival. In contrast, Zakai et al. (2005) found that lower and higher Hb concentrations proved to be independently associated with increased mortality. In this regard, further research is needed.

Severity of anemia prognosis in our sample of OAs is noteworthy. Prevalence of anemia in our study sample proved greater than in others involving Mexican Americans (11.5% in males and 9.3% in females) and Americans (11.0% in males and 10.2% in females; Guralnik, Eisenstaedt, Ferrucci, Klein, & Woodman, 2004), Brazilians (10.2%; Santos, Scazufca, Lotufo, Menezes, & Bensenor, 2013), and other studies from community dwellers (12%; Gaskell, Derry, Andrew Moore, & McQuay, 2008). It is possible that this high prevalence might be due to precarious life conditions in which these rural OAs have been living. Because of poverty, they have suffered historical gaps in terms of food insecurity, scarce food intake variety, and a null or bad medical attention. According to our findings, mild anemia was more prevalent than severe and moderate anemia; the latter entails a grimmer prognosis for OAs. Moderate and severe anemia had the strongest risk of mortality over a short period of time; if we take into account that the 80% of OAs who died in this study were dead at 11 months of follow-up.

Along with the problem of the high prevalence of anemia, our results show that the strength of association between anemia and mortality is stronger compared with what has been reported with other samples of OAs with health conditions similar to those of our sample. For example, den Elzen et al. reported a hazard ratio [HR] = 2.93 for mortality in Canadian OAs with severe anemia, while our result was OR = 6.0. In ambulatory OAs with anemia, Stump reported a HR = 1.44 for 1-year mortality (Stump, Callahan, & Hendrie, 2001), and our result for anemia was 2.56. Somehow, these data show the importance of anemia as a public health problem that must be addressed as soon as possible, particularly in populations with significant economic and social backwardness.

Identifying the main causes of anemia in this population group is crucial to reformulate strategies for reducing adverse prognoses in OAs by abating anemia (Guralnik et al., 2004; Santos et al., 2013). In Mexico, for example, 19.9% of OAs aged 65 and above suffer from iron-deficiency anemia (Mejía-Rodríguez, Shamah-Levy, Villalpando, García-Guerra, & Méndez-Gómez, 2013). Other causes of anemia, such as folate and vitamin B₁₂ deficiencies, anemia of chronic inflammation and renal insufficiency, remain unexplored and unquantified in this age group in Mexico. Among American OAs with anemia, 34.3% show nutritional deficiencies, 32.2% renal insufficiency or chronic inflammation, and 33.6% remained unexplained (Guralnik et al., 2004), and those with nutritional anemia show the highest mortality risk (risk ratio [RR] = 2.34, p < .001; Shavelle, MacKenzie, & Paculdo, 2012).

Correction of anemia in ambulatory OAs has not been documented to date. Anemia in critically ill patients is a common condition with a poor outcome, as opposed to anemia in apparently healthy persons (Shander, 2004). In patients with critical or chronic disease, treating anemia with intravenous iron and erythropoietin has obtained contradictory results (Cortés-Berdonces, García Martín, & León Sanz, 2012; Muller & Juffermans, 2010; Muñoz, Leal-Noval, Garcia-Erce, & Naveira, 2007; Tsakiris, 2000; Vincent et al., 2002). At public health level, alternative therapies and novel approaches should be evaluated to prevent and correct anemia in ambulatory OAs who are not severely ill. Anemia treatment guidelines should be formulated with specific emphasis on this population group.

There were some limitations in this study. First, our sample was constituted by rural OAs who may not be representative of urban OAs or Mexican OAs as a whole, thus precluding extrapolation of results. Second, as Hb was measured only at baseline, it was impracticable to identify, first, whether anemia in the study subjects receded or persisted until death, and, second, whether incident anemia had an impact on survival. Third, even though our results allow us to suggest that there is a causal association between anemia and mortality among OAs, it is needed to be cautious because we do not have a strong study design (i.e., experimental) to make an explicit causal connection between anemia and mortality.

However, one of the strengths of our study lies in the fact that it is the first to document the relationship between severity of anemia and mortality among ambulatory OAs in Mexico on a short-term basis. Given the dearth of information on the subject, this relationship has not been sufficiently documented among Latin American OAs, and cannot be compared with that of OAs in industrialized countries, due to cultural and socioeconomic disparities.

Finally, further research is required to identify which are the main causes of anemia among Mexican OAs, what is the most effective treatment (if any) for these causes—considering that anemia is a modifiable factor—and what is the impact of these causes on OA quality of life and survival. The ensuing data will serve as a guide for governments to establish programs that not only prevent the onset and curb the prevalence of anemia but also slash the risk of adverse events and reduce the economic burden of anemia’s health-related impact (Ershler, Chen, Reyes, & Dubois, 2005).

Footnotes

Authors’ Note

All authors made some contribution to every part of the project, participated in extensive planning and design meetings, and helped to interpret the findings. Vanessa De la Cruz-Góngora and Aarón Salinas-Rodriguez drafted the manuscript and conducted the data analysis. Salvador Villalpando helped interpret findings. Martha María Téllez-Rojo Solís, Aarón Salinas-Rodriguez, and Betty Manrique-Espinoza had overall responsibility for the project and design of study. All authors contributed to the revision of the manuscript and approved the final version. The sponsor had no role in the design and conduct of the study; in the collection, analysis, and interpretation of data; in the preparation of the manuscript; or in the review or approval of the manuscript.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This project has been funded in part by the Mexican Ministry of Social Development.

Notes

References

Cohen

J. H.

Hass

J. D.

(1999). Hemoglobin correction factors for estimating the prevalence of iron deficiency anemia in pregnant women residing at high altitudes in Bolivia. Pan American Journal of Public Health, 6, 392-396.

Cole

T. J.

Bellizzi

M. C.

Flegal

K. M.

Dietz

W. H.

(2000). Establishing a standard definition for child overweight and obesity worldwide: International survey. British Medical Journal, 320, 1240-1243.

Cortés-Berdonces

García Martín

León Sanz

(2012). Anemia in critically ill and surgical patient: Treatment with intravenous iron. Nutrición Hospitalaria, 27, 7-12.

DeMaeyer

E. M.

Dallman

Gurney

J. M.

Hallberg

Sood

S. K.

Srikantia

S. G.

(1989). Preventing and controlling iron deficiency anaemia through primary health care: A guide for health administrators and programme managers. Geneva, Switzerland: World Health Organization.

den Elzen

W. P.

Willems

J. M.

Westendorp

R. G.

de Craen

A. J.

Assendelft

W. J.

Gussekloo

(2009). Effect of anemia and comorbidity on functional status and mortality in old age: Results from the Leiden 85-plus Study. Canadian Medical Association Journal, 181, 151-157.

Ershler

W. B.

Chen

Reyes

E. B.

Dubois

(2005). Economic burden of patients with anemia in selected diseases. Value Health, 8, 629-638.

Gaskell

Derry

Andrew Moore

McQuay

H. J.

(2008). Prevalence of anaemia in older persons: Systematic review. BMC Geriatrics, 8, Article 1.

Guralnik

J. M.

Eisenstaedt

R. S.

Ferrucci

Klein

H. G.

Woodman

R. C.

(2004). Prevalence of anemia in persons 65 years and older in the United States: Evidence for a high rate of unexplained anemia. Blood, 104, 2263-2268.

Gutiérrez

J. P.

Rivera-Dommarco

Shamah-Levy

Villalpando-Hernández

Franco

Cuevas-Nasu

. . . Hernández-Ávila

(2012). Encuesta Nacional de Salud y Nutrición 2012. Resultados Nacionales (National Health and Nutrition Survey 2012. National Results). Cuernavaca, México: Instituto Nacional de Salud Pública.

10.

Katz

Branch

L. G.

Branson

M. H.

Papsidero

J. A.

Beck

J. C.

Greer

D. S.

(1983). Active life expectancy. The New England Journal of Medicine, 309, 1218-1224.

11.

Kleinbaum

D. G.

Kupper

L. L.

Morgenstern

(1982). Epidemiologic research: Principles and quantitative methods. New York, NY: Wiley.

12.

Kosiborod

Smith

G. L.

Radford

M. J.

Foody

J. M.

Krumholz

H. M.

(2003). The prognostic importance of anemia in patients with heart failure. American Journal of Medicine, 114, 112-119.

13.

Landi

Russo

Danese

Liperoti

Barillaro

Bernabei

. . . Onder

(2007). Anemia status, hemoglobin concentration, and mortality in nursing home older residents. Journal of the American Medical Directors Association, 8, 322-327.

14.

Lee

P. H.

Macfarlane

D. J.

Lam

T. H.

Stewart

S. M.

(2011). Validity of the International Physical Activity Questionnaire Short Form (IPAQ-SF): A systematic review. The International Journal of Behavioral Nutrition and Physical Activity, 8, 115.

15.

Mejía-Rodríguez

Shamah-Levy

Villalpando

García-Guerra

Méndez-Gómez

H. I.

(2013). Iron, zinc, copper and magnesium deficiencies in Mexican adults from the National Health and Nutrition Survey 2006. Salud Pública de México, 55, 3, 275-284.

16.

Muller

M. C.

Juffermans

N. P.

(2010). Anemia and blood transfusion and outcome on the intensive care unit. Critical Care, 14, 438.

17.

Muñoz

Leal-Noval

S. R.

Garcia-Erce

J. A.

Naveira

(2007). Prevalence and treatment of anemia in critically ill patients. Medicina Intensiva, 31, 388-398.

18.

Nissenson

A. R.

Goodnough

L. T.

Dubois

R. W.

(2003). Anemia: Not just an innocent bystander? Archives of Internal Medicine, 163, 1400-1404.

19.

Patel

K. V.

(2008). Epidemiology of anemia in older adults. Seminars in Hematology, 45, 210-217.

20.

Patel

K. V.

Longo

D. L.

Ershler

W. B.

Semba

R. D.

Ferrucci

. . . Guralnik

J. M.

(2009). Haemoglobin concentration and the risk of death in older adults: Differences by race/ethnicity in the NHANES III follow-up. British Journal of Haematology, 145, 514-523.

21.

Peters

Burch

Warner

Beckett

Poulter

Bulpitt

(2008). Haemoglobin, anaemia, dementia and cognitive decline in the elderly, a systematic review. BMC Geriatrics, 8, Article 18.

22.

Price

E. A.

Mehra

Holmes

T. H.

Schrier

S. L.

(2011). Anemia in older persons: Etiology and evaluation. Blood Cells, Molecules & Diseases, 46, 159-165.

23.

Riva

Tettamanti

Mosconi

Apolone

Gandini

Nobili

. . . Lucca

(2009). Association of mild anemia with hospitalization and mortality in the elderly: The Health and Anemia population-based study. Haematologica, 94, 22-28.

24.

Salinas-Rodríguez

Manrique-Espinoza

Moreno-Tamayo

K. M.

Torres-Pereda

De la Cruz-Góngora

Ángeles

. . . Téllez-Rojo Solis

M. M.

(2014). Impact evaluation of the non-contributory social pension program 70 y más in Mexico. Retrieved from http://www.3ieimpact.org/media/filer/2014/04/07/ie_5-mexico_social_pension_0304_top_3.pdf

25.

Santos

I. S.

Scazufca

Lotufo

P. A.

Menezes

P. R.

Bensenor

I. M.

(2013). Causes of recurrent or persistent anemia in older people from the results of the Sao Paulo Ageing & Health Study. Geriatrics & Gerontology International, 13, 204-208.

26.

Shander

(2004). Anemia in the critically ill. Critical Care Clinics, 20, 159-178.

27.

Shavelle

R. M.

MacKenzie

Paculdo

D. R.

(2012). Anemia and mortality in older persons: Does the type of anemia affect survival? International Journal of Hematology, 95, 248-256.

28.

StataCorp. (2009). Stata Statistical Software: Release 11. College Station, TX: StataCorp LP.

29.

Stump

T. E.

Callahan

C. M.

Hendrie

H. C.

(2001). Cognitive impairment and mortality in older primary care patients. Journal of the American Geriatrics Society, 49, 934-940.

30.

Tsakiris

(2000). Morbidity and mortality reduction associated with the use of erythropoietin. Nephron, 85(Suppl. 1), 2-8.

31.

Vincent

J. L.

Baron

J. F.

Reinhart

Gattinoni

Thijs

Webb

. . . Peres-Bota

(2002). Anemia and blood transfusion in critically ill patients. Journal of the American Medical Association, 288, 1499-1507.

32.

World Health Organization, United Nations University, & United Nations Children’s Fund. (2001). Iron deficiency anaemia, assessment, prevention and control: A guide for programme managers. Geneva, Switzerland: World Health Organization.

33.

Yesavage

J. A.

Brink

T. L.

Rose

T. L.

Lum

Huang

Adey

. . . Leirer

V. O.

(1982). Development and validation of a geriatric depression screening scale: A preliminary report. Journal of Psychiatry Research, 17, 37-49.

34.

Zakai

N. A.

Katz

Hirsch

Shlipak

M. G.

Chaves

P. H.

Newman

A. B.

. . . Cushman

M. A.

(2005). A prospective study of anemia status, hemoglobin concentration, and mortality in an elderly cohort: The Cardiovascular Health Study. Archives of Internal Medicine, 165, 2214-2220.