Authors: Yucong Bi, Liping Zheng, Leping Zhang, Longyang Han, Yang Liu, Xiaowei Zheng, Chongke Zhong
Categories: Research, Anemia, Depressive symptoms, Dynamic changes, Middle-aged and older Chinese adults
Source: Archives of Public Health
Previous studies have reported that anemia was associated with depression, but the association between changes in depressive symptoms and the risk of anemia was unclear. This study aimed to explore whether changes in depressive symptoms were associated with anemia among the middle-aged and elderly adults.
A total of 6112 patients aged 45 years and older from the China Health and Retirement Longitudinal Study (CHARLS) were included in this analysis. Elevated Depression Symptoms (EDS) was defined as the Center for Epidemiological Studies Depression Scale-10 score ≥ 10. Depression status was defined as no depressive symptom [no EDS at Wave 1 (2011–2012) and Wave 2 (2013–2014)], decreasing depressive symptoms (EDS at Wave 1, no EDS at Wave 2), increasing depressive symptoms (no EDS at Wave 1, EDS at Wave 2), persistent depressive symptoms (EDS at Wave 1 and Wave 2). Multivariable logistic regression analyses were conducted to estimate the relationships between depressive symptoms and the changes and risk of anemia.
During the follow-up of Wave 1 and Wave 3 (2015–2016), 906 participants (14.82%) developed anemia, the multivariable-adjusted odds ratio for the depressive symptom compared with the no depressive symptom was 1.24 (95% CI, 1.12–1.58) for anemia. From Wave 2 to Wave 3, there were 828 participants (14.62%) diagnosed with anemia. Compared to participants with no depressive symptom, those with persistent depressive symptoms during Wave 1 and Wave 2 had the significantly elevated risk of anemia (odds ratio 1.44, 95% CI 1.21–1.84).
The present study demonstrated that baseline depressive symptoms and changes in depressive symptoms were associated with increased risks of anemia.
Keywords: Depressive symptoms, Anemia, Dynamic changes, Middle-aged and older Chinese adults
Depressive symptom, a common illness, has become one of the most severe psychiatric disorders all over the world. The prevalence of depression is high in China, especially in middle-aged and elderly Chinese adults [1, 2]. The depressive disorders are common across the life course and are present in up to a third of older adults [3]. Numerous epidemiological studies have shown that depression is associated with an increased risk of cardiovascular disease and all-cause mortality [4–7]. Depressive symptom limits psychosocial functioning and diminishes quality of life, and is one of the leading causes of disease burden worldwide [8, 9].
People with depression may suffer from poor health behaviors, such as excessive alcohol consumption or malnutrition caused by an unhealthy diet, which may trigger a drop in hemoglobin levels [10]. Anemia is manifested by a decrease in hemoglobin levels that is more common in the elderly [11]. Previous studies have reported that anemia affects individual’s quality of life and physical function [12, 13]. Currently, most studies have assessed the relationship between anemia and risk of depression [14–17], only a few studies have evaluated whether depressive symptoms would increase the risk of subsequent anemia or decreased hemoglobin levels [10, 18–21]. The InCHIANTI study indicated that the risk of anemia progressively and significantly increased with increasing depression severity [18]. A recent cohort study suggested that depression symptoms seemed related to anemia in the middle-aged and elderly in China [22]. However, the Netherlands Study of Depression and Anxiety showed that there was no independent association between depressive disorders and hemoglobin levels or anemia status [21]. Up to now, evidence of longitudinal association of depression with risk of anemia or decreased hemoglobin levels is still lacking.
Furthermore, depressive symptoms were only measured at baseline in most studies, which could not capture within-person and inter-person variation over time. Therefore, single time assessment of depressive symptoms may be insufficient to reveal the complex predictive effects of depressive fluctuations. Since depressive disorders may vary considerably over the course of a lifetime [23, 24], it is needed to assess the dynamic changes of depressive symptoms and risk of anemia.
This study aimed to explore the relationships between changes in depressive symptoms and risk of anemia among middle-aged and older Chinese adults, using data from the China Longitudinal Study of Health and Retirement (CHARLS).
This study was derived from the CHARLS, which used a multistage clustering sample method to select participants in China [25]. A total of 17,708 participants from 10,257 households recruited from 28 provinces within China were included at baseline (2011–2012, Wave 1). CHARLS respondents are followed every 2 years, using a face-to-face computer-assisted personal interview. Three subsequent follow-ups were carried out in 2013–2014 (Wave 2) and 2015–2016 (Wave 3). The design details and main results have been described previously [26]. The inclusion criteria were as (1) aged ≥ 45 years; (2) reported 2 complete measurements (Wave 1, Wave 2) about depressive symptoms evaluated by the Center for Epidemiological Studies Depression Scale-10 (CESD-10); (3) reported 2 complete measurements (Wave1, Wave 3) about hemoglobin level. Participants without data about study outcome (anemia, Wave 3) were excluded. Finally, a total of 6112 participants were included in this analysis (Fig. 1).
Fig. 1 Flow chart of sample selection and exclusion criteria. Prospective analysis The association between depressive symptoms at Wave 1 (2011–2012) and anemia at Wave 3 (2015–2016). Prospective analysis The association between changes in depressive symptoms between Wave 1 (2011–2012) and Wave 2 (2013–2014) and anemia at Wave 3 (2015–2016)
The CHARLS study was approved by the institutional review board of Peking University. Written informed consent was obtained from all participants.
Data was collected through face-to-face interviews by trained personnel at each wave including depressive symptoms. The CESD-10 was administrated to measure depressive symptoms at baseline and each follow-up visit of the CHARLS [27], which has been proved to be a reliable and valid approach to detect depression in Chinese adults [28, 29]. The CESD-10 scale consists of 10 items, including depression and positive affected parts. The total score ranges from 0 to 30, depressive symptom was defined as a CESD score of ≥ 10 [30]. Among those with CESD score ≥ 10, the higher the score, the higher the degree of depressive symptoms. In this study, all participants were classified into 4 groups according to the first and second wave of depression. Depression status was defined as No depressive symptom [no Elevated Depression Symptoms (EDS) at Wave 1 and Wave 2], Decreasing depressive symptoms (EDS at Wave 1, no EDS at Wave 2), Increasing depressive symptoms (no EDS at Wave 1, EDS at Wave 2), Persistent depressive symptoms (EDS at Wave 1 and Wave 2).
At baseline, trained interviewers used a structured questionnaire to collect information on socio-demographic status and health-related factors. Sociodemographic variables included age, sex, educational level (Illiteracy, Primary school, Middle school or above), living place (rural or urban). Health-related factors included smoking status (ever smoking vs. never smoking), drinking status (ever drinking or never drinking), and 8 common co-morbidities (hypertension, dyslipidemia, diabetes mellitus, heart disease, stroke, psychiatric disease, liver disease, and asthma). Body mass index was defined as weight in kilograms divided by the square of height in meters. “Ever smoking” means that the respondent reported smoking at some point, and “never smoking” means that the respondent reported never having smoked. “Ever drinking” means that the respondent reports having had an alcoholic beverage in the past, and “never drinking” means that the respondent reported not having any alcoholic beverage in the past. Common co-morbidities were measured with the following “Have you been diagnosed with conditions listed below by a doctor?” The conditions included hypertension, dyslipidemia, diabetes mellitus, heart disease, stroke, psychiatric disease, liver disease, and asthma. Each condition was assessed by trained investigators separately.
The details on blood sampling and determination of biological parameters have been described previously [31]. Anemia was defined according to the World Health Organization (WHO) criteria as a hemoglobin (Hb) concentration < 12 g/dl (7.5 mmol/l) in women and < 13 g/dl (8.1 mmol/l) in men (Fig. 2).
Fig. 2 Timeline of exposure and follow-up assessment
The baseline characteristics among the four groups were compared. Continuous variables are expressed as mean ± standard deviation or median (interquartile range). Categorical variables are expressed as frequency (percentage). Generalized linear regression models were applied for trends across depression status for continuous variables, and chi-square trend tests were used for categorical variables. Binary logistic regression model was conducted to estimate the relationships between baseline depressive symptoms and changes in depressive symptoms with anemia, with the odds ratios (ORs) and 95% confidence intervals (CIs) were calculated. Multivariable logistic regression models adjusted for age, sex, living place, education level, smoking, drinking, body mass index, systolic blood pressure, antidepressant, Mini-Mental State Examination score, and medical history (hypertension, dyslipidemia, diabetes mellitus, heart disease, stroke, psychiatric disease, liver disease, and asthma).
Due to some demographic characteristics and medical histories may affect the association between changes in depressive symptoms and anemia, we further conducted subgroup analyses to assess whether the relationships were potentially modified by age, sex, educational level, living place, smoking, drinking, history of hypertension and diabetes. Twelve is another cut point for CESD score, to examine the robustness of the findings, we conducted a sensitivity analysis by defining depressive symptom as a CESD score ≥ 12 [32].Multiple imputation for missing covariate values was performed using the Markov chain Monte Carlo method. All P values were 2-tailed, and a significance level of 0.05 was used. Data analyses were performed using SAS, version 9.4 (SAS Institute Inc; Cary, North Carolina, USA).
A total of 6112 participants (2856 men and 3256 women) were included in the current study, and the mean age of participants was 58.47 ± 8.52 years. Baseline characteristics of participants according to changes in depressive symptoms are presented in Table 1. The baseline characteristics such as sex, living place, education level, history of hypertension, history of dyslipidemia, history of heart problems, history of stroke, history of psychosis, history of liver disease, history of asthma, smoking, drinking, body mass index and antidepressant medication were significantly different among the four groups (Table 1).
During the follow-up of Wave 1 and Wave 3, 906 participants patients (14.82%) developed anemia. In age- and sex-adjusted model, participants with depressive symptoms at Wave 1 had a higher risk of anemia at Wave 3, compared with those with no depressive symptom. After additional adjustment for living place, education level, smoking, drinking, body mass index, systolic blood pressure, antidepressant and medical history, the OR for participants with depressive symptoms was 1.24 (95% CI 1.12–1.58) for the risk of anemia.
From Wave 2 to Wave 3, there were 828 participants (14.62%) diagnosed with anemia (Table 2). Compared to those with no depressive symptom, multivariable logistic regression models showed that participants with persistent depressive symptoms during Wave 1 and Wave 2 had the significantly elevated risk of anemia (OR 1.44, 95% CI 1.21–1.84). In addition, the subgroup analyses revealed that age, sex, educational level, living place, smoking, drinking, history of hypertension and diabetes did not modify the relationship between depressive symptoms and anemia (P for interaction > 0.05; Table 3). Persistent depressive symptoms were associated with increased risk of anemia in most strata.
Furthermore, in our sensitivity analysis, when depressive symptom was defined as a CESD score ≥ 12, multivariable logistic regression models showed that participants with persistent depressive symptoms had the significantly elevated risk of anemia (OR 1.39, 95% CI 1.14–1.72, Table 4).
This study explored the relationship between changes in depressive symptoms and anemia based on CHARLS. First, baseline depressive symptoms were associated with an increased risk of anemia. Specifically, there was a 27% elevated risk of anemia in patients with depressive symptoms at Wave 1 (2011–2012). In addition, participants with persistent depressive symptoms during Wave 1 (2011–2012) and Wave 2 (2013–2014) had a 43% increased risk of incident anemia, compared with no depressive symptoms after adjusting for known potential confounders. These findings were consistent across different subgroups, and were further confirmed in the sensitivity analysis using different definition of depressive symptoms. Our study suggested that participants with depressive symptoms, especially for those with long-term depression had a higher risk of anemia in Chinese middle-aged and elderly adults.
Depressive symptoms are common in the elderly, depressed patients often have unhealthy diet and lifestyle, leading to nutritional deficiencies that increase the risk of anemia. Furthermore, it has been shown that depression shares some similar pathological characteristics with anemia, but previous studies have shown inconsistent results of the relationships between depressive symptoms and decreased hemoglobin levels or anemia [14, 20, 21]. For example, the Netherlands Study of Depression and Anxiety recruiting 2920 participants reported that there was no clear evidence for an association between depressive disorders and hemoglobin levels or anemia status [21]. However, a large scale cross-sectional study including 44,137 participants showed that depressed participants were significantly more likely to have anemia compared to non-depressed participants after adjustment for sociodemographic and health-related variables (OR 1.36, 95% CI 1.18–1.57) [20]. Additionally, in a prospective population-based study of participants with a mean age of 75 years found that depressive symptoms are associated with an increased risk of anemia [18]. Our study validated these findings that depressive symptoms were associated with increased risk of incidence of anemia.
Previous studies on depressive symptoms and anemia only measured depressive symptoms once at baseline, failing to take into account the effects of depressive symptoms change [18, 20]. It is reported that the course of depressive symptoms varies across individuals, with some may experience remission or relapse of depression, while others may have long-term chronic depression. Different patterns of depressive symptoms changes may have different risk of anemia. Our study investigated the relationship between changes in depressive symptoms and the risk of anemia, and the results showed that participants with persistent depressive symptoms during Wave 1 and Wave 2 had the 1.43-fold risk of anemia compared with those in the no depressive symptom.
The potential mechanism underlying the association between depressive symptoms and risk of anemia remains to be fully elucidated, while several potential biological mechanisms have been proposed. First, patients with depressed mood often have unhealthy eating practices that can lead to deficiencies in vitamins (such as vitamin B12, folic acid) and minerals (such as iron and zinc), unhealthy dietary intake may also cause or increase activation of inflammatory response systems, such as the inflammatory marker C-reactive protein or interleukin-6, which can lead to anemia [33]. In addition, depressed patients may have increased sympathetic tone, which can affect erythrocyte and bone marrow production by catecholamine modulation [34]. Furthermore, anemia and depressive symptoms are associated with many chronic conditions, and underlying chronic conditions such as cancer, heart failure and diabetes may modulate the observed relationships [35, 36]. However, the association of changes in depressive symptoms and anemia remained significant after adjusting for heart disease and diabetes, further studies are still needed to clarify the potential mechanisms.
In this study, we investigated the prospective associations of baseline depressive symptoms, as well as the changes in depressive symptoms with the risk of anemia, which may provide more predictive evidence of depressive symptoms on anemia. Furthermore, our study was based on a large nationally representative cohort study with a high response rate cohort of CHARLS, potential confounders were collected and controlled in the multivariable models. Nevertheless, there are also some limitations. First, this study obtained the information of the study subject through a self-report questionnaire, which would cause information bias. Second, the CHARLS study was exclusively a Chinese population, and the findings from our study might not be generalizable to other populations. Third, data about the types of anemia was not collected in this study, which limited us to examine the effects of changes in depressive symptoms on different types of anemia. Although we controlled for a range of covariates in our analyses, the influence of unknown factors cannot be ruled out in this study. Finally, CESD-10 score was divided into categorical variables and a lot of information may lost in these analyses. Some of the CESD-10 data in this study were unavoidably missing, and a potential underestimation of the association between depressive symptoms and anemia may exist due to the exclusion of missing data.
This nationally representative longitudinal study provided evidence of the association between depressive symptoms and anemia, and further demonstrated that changes in depressive symptoms were associated with increased risks of anemia among the middle-aged and older Chinese adults. Further experimental and clinical research are needed to verify our findings and clarify the potential mechanisms.
This analysis uses data or information from the Harmonized China Health and Retirement Longitudinal Study (CHARLS) dataset and Codebook, version C as of April 2018 developed by the Gateway to Global Aging Data. The development of the Harmonized CHARLS was funded by the National Institute on Ageing (R01 AG030153, RC2 AG036619, R03 AG043052). For more information, please refer to www.g2aging.org.
CKZ and XWZ contributed to the conception and design of the study; YCB LPZ, LPZ and LYH, contributed to the acquisition of data; YCB, LPZ, LYH, YL and CKZ contributed to the analysis of data, preparation of the figures and the drafting of the text. All authors read and approved the final manuscript.
This study was supported by the National Natural Science Foundation of China (grant No: 82273706), Interdisciplinary Basic Frontier Innovation Program of Suzhou Medical College of Soochow University (YXY2302013), the Project of MOE Key Laboratory of Geriatric Diseases and Immunology (No. JYN202406), and a project of the Undergraduate Training Program for Innovation and Entrepreneurship, Soochow University (grant No: 202310285072).
No datasets were generated or analysed during the current study.
Not applicable.
Not applicable.
The authors declare no competing interests.
Xiaowei Zheng, Email: zxw19921212@163.com.
Chongke Zhong, Email: ckzhong@suda.edu.cn.
No datasets were generated or analysed during the current study.