Hide Sidebar

Anemia Associated with Higher Mortality Among Adults 85 and Over

Anemia Associated with Higher Mortality Among Adults 85 and Over

Critique date: Fri, 02/12/2010 - 10:23am
Institution: 
Duke University School of Medicine
   
Article Citation
Citation: 
den Elzen WPJ, Willems JM, Westendrop GJ, et al. Effect of anemia and comorbidity on functional status and mortality in old age: results from the Leiden 85-plus Study. CMAJ 2009; 181: 151-157.
   
Clinical Bottom Lines

In a very elderly population in northern Europe, anemia (either at baseline or new onset) was independently associated with an increased risk of death in the ensuing 5 years after controlling for a number of possible causal factors and diagnoses. Anemia was associated with only modest declines in ability. Specific causes of anemia and death were not examined in this study.

   
Methods
Type of Study: 
Prognosis
Study Design: 
Cohort study
Follow-up Period: 
Annual follow-up for 5 years or until death
Setting: 
Large population-based registry in Leiden, Netherlands.
Patient Population: 

562 (of 705 eligible) 85 year old residents of Leiden. 33.6% female, 64.9% < high school education, 51.3 % low income, 18.1% in long term care, 34.9% with ≥ 2 comorbid conditions, median MMSE 26, median GDS 2. Moderate disability (median 10/36 on ADLs, 18/36 IADLs by Groningen Activty Restriction Scale).

Significant Exclusions: 

No systematic exclusions. 143/705 not included due to death, refusal to participate or no blood sample.

Intervention/Exposure: 

Anemia (Hgb <12.0 g/dL for women and < 13.0 g/dL for men) based on baseline and annual blood samples. Anemia at baseline defined as prevalent; Newly anemic at follow-up defined as incident.

Outcome Measures: 

Primary: Mortality—obtained from municipal registry. Survival analysis used to examine association of prevalent and incident anemia with mortality. Secondary: Functional status as defined by ADLs and IADLs on Groningen scale, MMSE and GDS. Linear regression analysis used to examine association of prevalent and incident anemia with function. Both models controlled for the following variables: sex, level of education, income, residence in LTC, presence of ≥ 1 comorbid condition (including stroke, MI, severe cognitive impairment, DM, Parkinson’s disease, hip fracture, arthritis, COPD, and malignancy), creatinine clearance and C-reactive protein at baseline.

Participant Follow-up: 
Of 562 participants at baseline, number surviving or remaining in study by age: 86 yrs= 506, 87 yrs=438, 88 yrs=385, 89 yrs=327, 90 yrs=275. (80% of attrition due to deaths.)
   
Conclusion
Results: 

Baseline prevalence of anemia= 26.7% (150/562). (Men, former smokers and LTC residents more likely to be anemic.) Of those remaining, 24% (99/412) developed incident anemia. Results of survival analysis as follows:

Anemia  Unadjusted HR (95% CI)   Adjusted HR (95% CI)
Prevalent anemia 1.71 (1.40 to 2.10) 1.41 (1.13 to 1.76)
Incident anemia  2.16 (1.68 to 2.77) 2.08 (1.60 to 2.70)

In a subgroup analysis, more severe anemia was more strongly associated with mortality in a dose-response fashion.

With respect to functional decline, prevalent anemia was associated after adjustment with significant annual decline in IADLs (0.7, 95% CI 0.1 to 1.4, p=0.03) and incident anemia was associated with annual decline in ADLs (1.4, 95% CI 0.4 to 2.5; p<0.01).

Concerns Regarding Methodology, Applicability to Older Adults, etc.: 

The study was methodologically sound, enrolling the entire population of the city of Leiden, Netherlands. While the population was relatively healthy, rates of anemia were comparable to other very elderly cohorts. Given the limited demographic diversity of this group, its unclear if these results are generalizable. The utility of the study may have been enhanced by further characterization of type or cause of anemia and causes of death.

Funding Source and Role: 
Funded in part by the Dutch Ministry of Health, Welfare and Sports.
Created By: 
Mitchell T. Heflin, MD, MHS, Associate Professor of Medicine, Division of Geriatrics, Duke University.
This is a review of the validity of a single study; the ‘bottom lines’ do not reflect comparison with the rest of the literature on this subject.