Advertisement

Childhood socioeconomic disadvantage and risk of physical multimorbidity in later life: The mediating role of depression

Published:September 23, 2022DOI:https://doi.org/10.1016/j.maturitas.2022.09.007

      Highlights

      • This is the first study to investigate whether depression mediates the association between childhood socioeconomic disadvantage and physical multimorbidity in adulthood.
      • Childhood socioeconomic disadvantage was associated with higher risks of depression and physical multimorbidity in later life, and females were more vulnerable to the adverse impact of childhood socioeconomic disadvantage on the risk of depression.
      • Depression mediated the association of childhood socioeconomic disadvantage with physical multimorbidity, especially among females.

      Abstract

      Objective

      To assess the impact of childhood socioeconomic disadvantage (SED) on the risks of depression and physical multimorbidity in later life and to explore whether depression mediates the association between childhood SED and physical multimorbidity.

      Methods

      Data on 8214 adults from the China Health and Retirement Longitudinal Study were analyzed. The mean (SD) age of the study population was 57.0 (8.0) years at baseline (2011) and 51.9 % were females. Multivariable logistic regressions were used to examine the associations of childhood SED (indexed by food insecurity, highest education level of parents, and self-perceived household financial situation, and scored 0–3) with later-life depression (scored ≥10 on the 10-item Center for Epidemiologic Studies Depression Scale) and physical multimorbidity (having two or more doctor-diagnosed chronic conditions) assessed in the 2018 follow-up survey. Mediation analysis was conducted in the overall sample and further stratified by sex to estimate the degree to which the association between childhood SED and physical multimorbidity could be explained by baseline depression.

      Results

      Participants with a childhood SED score of 3 (i.e., the most disadvantaged) had 2.63 (95 % confidence interval [CI]: 1.91–3.63) times and 2.08 (95 % CI: 1.56–2.77) times higher odds of depression and physical multimorbidity respectively compared with those who had a score of 0 (i.e., the least disadvantaged). Depression mediated 20 % of the association between childhood SED and physical multimorbidity (36 % in females and 5 % in males).

      Conclusions

      Childhood SED was associated with higher risks of depression and physical multimorbidity in later life, and the association of childhood SED with physical multimorbidity was mediated by depression, especially among females.

      Keywords

      1. Introduction

      Depression and multiple chronic conditions (physical multimorbidity) are two major challenges confronting aging societies. Depression, one of the most prevalent psychiatric disorders among the older population, affects 23.6 % of Chinese older adults (≥55 years old) [
      • Li D.
      • et al.
      A meta-analysis of the prevalence of depressive symptoms in chinese older adults.
      ], causing higher risks of functional impairment, mortality, poor quality of life, and healthcare utilization [
      • Noel P.H.
      • et al.
      Depression and comorbid illness in elderly primary care patients: impact on multiple domains of health status and well-being.
      ]. Physical multimorbidity, the presence of two or more chronic physical conditions, afflicts >45.5 % of older individuals in low- and middle-income countries [
      • Vancampfort D.
      • Stubbs B.
      • Koyanagi A.
      Physical chronic conditions, multimorbidity and sedentary behavior amongst middle-aged and older adults in six low- and middle-income countries.
      ]. Around one in two Chinese people in their 50s suffers from physical multimorbidity, and the prevalence increases progressively with age [
      • Zhao Y.
      • et al.
      Physical multimorbidity, health service use, and catastrophic health expenditure by socioeconomic groups in China: an analysis of population-based panel data.
      ]. People suffering from physical multimorbidity have complex health and social care needs, thus posing a heavier burden on healthcare systems than single diseases [
      • Hopman P.
      • et al.
      Effectiveness of comprehensive care programs for patients with multiple chronic conditions or frailty: a systematic literature review.
      ,
      • Ho I.S.
      • et al.
      Examining variation in the measurement of multimorbidity in research: a systematic review of 566 studies.
      ]. Furthermore, the co-existence of depression and physical multimorbidity is pervasive, which could be partly explained by shared risk factors and might result in higher risks of disability and mortality [
      • Koyanagi A.
      • et al.
      Mortality in unipolar depression preceding and following chronic somatic diseases.
      ,
      • Quinones A.R.
      • et al.
      Prospective disability in different combinations of somatic and mental multimorbidity.
      ]. Thus, both depression and physical multimorbidity are clinical and public health priorities to cope with among the aging population.
      Integrating a life-course lens into aging research is critical. Accumulating evidence has shown that health disparities begin in early life. Children who experienced childhood socioeconomic disadvantage (SED) face higher risks of chronic diseases and mortality in adulthood [
      • Braveman P.
      • Barclay C.
      Health disparities beginning in childhood: a life-course perspective.
      ]. Childhood SED affects health status in later life both directly (independent of adulthood factors) and indirectly (mediated through adulthood factors) [
      • Cohen S.
      • et al.
      Childhood socioeconomic status and adult health.
      ,
      • Lynch J.
      • Smith G.D.
      A life course approach to chronic disease epidemiology.
      ]. Several previous studies have linked childhood SED with depression and multimorbidity [
      • Chen Y.
      • et al.
      Association between early life circumstances and depressive symptoms among chinese older adults: results from China health and retirement longitudinal study: early life circumstances and depression.
      ,
      • Henchoz Y.
      • et al.
      Childhood adversity: a gateway to multimorbidity in older age?.
      ,
      • Kim S.S.
      • et al.
      Association between childhood adversities and adulthood depressive symptoms in South Korea: results from a nationally representative longitudinal study.
      ,
      • Pavela G.
      • Latham K.
      Childhood conditions and multimorbidity among older adults.
      ,
      • Tucker-Seeley R.D.
      • et al.
      Lifecourse socioeconomic circumstances and multimorbidity among older adults.
      ,
      • Yang L.
      Childhood adversity and trajectories of multimorbidity in mid-late life: China health and longitudinal retirement study.
      ]. However, some studies included several SED factors in the regression models separately or dichotomized childhood SED by a cut-off value, without constructing a multidimensional composite indicator to reflect the cumulative effect of SED [
      • Pavela G.
      • Latham K.
      Childhood conditions and multimorbidity among older adults.
      ,
      • Tucker-Seeley R.D.
      • et al.
      Lifecourse socioeconomic circumstances and multimorbidity among older adults.
      ,
      • Yang L.
      Childhood adversity and trajectories of multimorbidity in mid-late life: China health and longitudinal retirement study.
      ,
      • Yazawa A.
      • et al.
      Early childhood adversity and late-life depressive symptoms: unpacking mediation and interaction by adult socioeconomic status.
      ]. Others did not separate childhood SED from other early life adverse experiences (e.g., domestic violence, child labour, and parental divorce) [
      • Chen Y.
      • et al.
      Association between early life circumstances and depressive symptoms among chinese older adults: results from China health and retirement longitudinal study: early life circumstances and depression.
      ,
      • Henchoz Y.
      • et al.
      Childhood adversity: a gateway to multimorbidity in older age?.
      ,
      • Kim S.S.
      • et al.
      Association between childhood adversities and adulthood depressive symptoms in South Korea: results from a nationally representative longitudinal study.
      ], thus failing to depict the separate effect of childhood SED on health conditions in later life. Furthermore, previous studies have found that depression is associated with the onset of some of the most common chronic diseases (e.g., diabetes, hypertension, ischemic heart disease, and arthritis) and increases the risk of multimorbidity [
      • Triolo F.
      • et al.
      The complex interplay between depression and multimorbidity in late life: risks and pathways.
      ,
      • Qiao Y.
      • et al.
      Bidirectional association between depression and multimorbidity in middle-aged and elderly Chinese adults: a longitudinal cohort study.
      ,
      • Ye B.
      • et al.
      Bidirectional association between physical multimorbidity and subclinical depression in chinese older adults: findings from a prospective cohort study.
      ,
      • Birk J.L.
      • et al.
      Depression and multimorbidity: considering temporal characteristics of the associations between depression and multiple chronic diseases.
      ]. However, it remains unknown whether depression mediates the association between childhood SED and physical multimorbidity. Thus, the objectives of this study are first to examine the impact of childhood SED on the risks of depression and physical multimorbidity in later life, and second to understand the mediating role of depression in the association between childhood SED and physical multimorbidity. Furthermore, given previous studies suggested that females were more vulnerable to depression and physical multimorbidity than males [
      • Salk R.H.
      • Hyde J.S.
      • Abramson L.Y.
      Gender differences in depression in representative national samples: meta-analyses of diagnoses and symptoms.
      ,
      • Alimohammadian M.
      • et al.
      Multimorbidity as an important issue among women: results of a gender difference investigation in a large population-based cross-sectional study in West Asia.
      ], and the adverse effect of childhood SED on several health outcomes (e.g., blood pressure, depression, body mass index) in later life was stronger among females [
      • Janicki-Deverts D.
      • et al.
      Sex differences in the association of childhood socioeconomic status with adult blood pressure change: the CARDIA study.
      ,
      • Csajbók Z.
      • et al.
      Sex differences in the association of childhood socioeconomic position and later-life depressive symptoms in Europe: the mediating effect of education.
      ,
      • Chapman B.P.
      • et al.
      Can the influence of childhood socioeconomic status on men’s and women’s adult body mass be explained by adult socioeconomic status or personality? Findings from a national sample.
      ], we further present our results stratified by sex.

      2. Methods

      2.1 Study population

      This study used data from the China Health and Retirement Longitudinal Study (CHARLS), an ongoing longitudinal survey of a nationally representative sample of community-dwelling middle-aged adults from 28 provinces in China. The baseline survey was conducted between June 2011 and March 2012, and follow-up surveys were carried out in 2013, 2015, and 2018. All alive respondents in the first two waves were also invited to participate in the 2014 life history survey to gather their early life experiences (e.g., residential history, health history, childhood socioeconomic status, childhood neighborhood quality, and relationship with parents). The CHARLS was approved by the Ethical Review Committee at Peking University. Details about the recruitment strategy and design of the CHARLS have been previously published [
      • Zhao Y.
      • et al.
      Cohort profile: the China health and retirement longitudinal study (CHARLS).
      ].
      In this study, the analysis included 8217 subjects who were ≥45 years old at baseline, attended the 2014 life course survey and 2018 follow-up survey, and had no missing information on childhood socioeconomic status, depression, and physical multimorbidity (Fig. 1).

      2.2 Definition of childhood socioeconomic disadvantage

      Similar to previous studies [
      • Celeste R.K.
      • et al.
      Socioeconomic life course models and Oral health: a longitudinal analysis.
      ,
      • Laitinen T.T.
      • et al.
      Childhood socioeconomic disadvantage and risk of fatty liver in adulthood: the cardiovascular risk in young finns study.
      ,
      • Parks C.G.
      • et al.
      Childhood socioeconomic factors and perinatal characteristics influence development of rheumatoid arthritis in adulthood.
      ], we constructed childhood SED using three indicators to reflect both subjective and objective life conditions during childhood and each was coded as ‘0’ or ‘1’: (1) food insecurity was examined by asking ‘when you were a child before age 17, was there a time when your family did not have enough food to eat?’ (0 = no, 1 = yes); (2) highest education level of parents was examined by asking ‘what is the highest level of education your biological mother/father completed?’ (0 = junior high school or higher, 1 = elementary school or lower); (3) self-perceived household financial situation was examined by asking ‘when you were a child before age 17, compared to the average family in the same community/village at that time, how was your family's financial situation?’ (0 = better off than or same as others, 1 = worse off than others). We then calculated the sum score of these three indicators to denote the overall childhood SED, ranging from 0 (least disadvantaged) to 3 (most disadvantaged).

      2.3 Ascertainment of depression

      Depressive symptoms were measured using the 10-item Center for Epidemiologic Studies Depression (CESD-10) Scale [
      • Andresen E.M.
      • et al.
      Screening for depression in well older adults: evaluation of a short form of the CES-D (Center for Epidemiologic Studies Depression Scale).
      ]. The CESD-10 is a subset of the 20-item CESD Scale and has been widely used in general populations and patients with chronic illnesses. The CESD-10 uses a 4-point Likert scale (ranging from 0 to 3 for each item) with higher scores representing greater depressive symptoms. A total score ≥ 10 on the CESD-10 was used to denote depressive symptoms. This threshold has been proved to correlate well with the clinical diagnosis of major depression [
      • Andresen E.M.
      • et al.
      Screening for depression in well older adults: evaluation of a short form of the CES-D (Center for Epidemiologic Studies Depression Scale).
      ] and was validated in measuring depression in Chinese older populations [
      • Fu H.
      • Si L.
      • Guo R.
      What is the optimal cut-off point of the 10-item Center for Epidemiologic Studies Depression Scale for screening depression among Chinese individuals aged 45 and over? An exploration using latent profile analysis.
      ].

      2.4 Ascertainment of physical multimorbidity

      Chronic physical conditions were assessed by the question ‘have you been diagnosed with conditions listed below by a doctor?’ The total number of 12 conditions was calculated: hypertension, dyslipidemia, diabetes or high blood sugar, malignant tumor, chronic lung diseases (e.g., chronic bronchitis, emphysema), liver disease, heart problems (e.g., heart attack, coronary heart disease, angina, congestive heart failure), stroke, kidney disease, stomach or other digestive diseases, arthritis or rheumatism, and asthma. Physical multimorbidity was defined as having two or more chronic conditions [
      • Barnett K.
      • et al.
      Epidemiology of multimorbidity and implications for health care, research, and medical education: a cross-sectional study.
      ].

      2.5 Covariates

      The following factors assessed at baseline were candidate covariates: age, sex, education level (elementary school or lower versus junior high school or higher), household per capita consumption (classified into low, middle, and high based on tertiles), body mass index (subject's weight in kilograms divided by the square of height in meters), smoking status (never smoking versus current or former smoker), and alcohol consumption (never drinking versus current or former drinker).
      We constructed a directed acyclic graph (DAG) of the associations of childhood SED with depression and physical multimorbidity in later life to guide our analyses (Supplemental Fig. 1). According to this DAG, we only included age and sex as covariates in the regression models. Adulthood socioeconomic status, health-related behaviors, and body mass index were not adjusted because they are colliders and would induce bias if included as covariates [
      • VanderWeele T.J.
      • Robins J.M.
      Directed acyclic graphs, sufficient causes, and the properties of conditioning on a common effect.
      ].

      2.6 Statistical analysis

      Baseline characteristics were compared among subjects with different childhood SED scores and between those with and without depression or physical multimorbidity in the 2018 follow-up survey. Differences were detected by the student's t-test or analysis of variance for continuous variables and the χ2 test for categorical variables where appropriate.
      The associations of childhood SED with the prevalence of depression and physical multimorbidity in the 2018 follow-up survey were assessed by multivariable logistic regressions. Childhood SED was presented as individual indicators (food insecurity, highest education level of parents, and self-perceived household financial situation) as well as the total score. Age and sex were included as covariates. The total score was further decomposed into different combinations of the three indicators (number of combinations: 2 × 2 × 2 = 8) to examine whether these indicators were exchangeable. Next, to find out whether there is effect modification by sex, we added an interaction term (sex × childhood SED score) in the model and examined the associations of childhood SED with depression and physical multimorbidity stratified by sex.
      We then estimated the degree to which the association between childhood SED and physical multimorbidity could be explained by baseline status of depression. We conducted this analysis among subjects who were free from any chronic conditions at baseline to ensure childhood SED and the diagnosis of depression temporally preceded physical multimorbidity. We conducted mediation analysis using the CAUSALMED procedure in SAS [
      • Yung Y.-F.
      • Lamm M.
      • Zhang W.
      Causal mediation analysis with the CAUSALMED procedure.
      ]. This method can decompose the total effect (i.e., unadjusted for the mediator) of a variable into direct (i.e., the effect of a variable on outcome adjusted for the mediator) and indirect (mediated) effects. Using this procedure, we could also obtain the percentage of the main association explained by the mediator.
      Results were expressed as odds ratios (ORs) and their 95 % confidence intervals (CIs). All statistical analyses were performed using SAS (version 9.4). A P value <0.05 was considered to be statistically significant.

      3. Results

      Baseline characteristics according to childhood SED scores and status of depression or physical multimorbidity in the 2018 follow-up survey are shown in Table 1 and Supplemental Table 1. Participants with higher childhood SED scores (i.e., more disadvantaged) were more likely to be older, males, with lower education level and lower household per capita consumption, with lower body mass index, and current or former smoker or drinkers. Participants with depression or physical multimorbidity were more likely to be females, with lower education level, and non-smokers.
      Table 1Baseline characteristics by childhood socioeconomic disadvantage score (bold numbers indicate statistical significance with P value < 0.05).
      Overall sample,

      n (%)
      Childhood socioeconomic disadvantage scoreP value
      0 (N = 211),

      n (%)
      1 (N = 1787),

      n (%)
      2 (N = 3692),

      n (%)
      3 (N = 2527),

      n (%)
      Age, years, mean (SD)57.0 (8.0)52.4 (7.8)55.7 (9.0)57.5 (7.7)57.6 (7.6)<0.001
      Sex<0.001
       Male3952 (48.1)73 (34.6)747 (41.8)1859 (50.4)1273 (50.4)
       Female4265 (51.9)138 (65.4)1040 (58.2)1833 (49.7)1254 (49.6)
      Education level<0.001
       Elementary school or lower5214 (63.5)61 (28.9)926 (51.8)2395 (64.9)1832 (72.5)
       Junior high school or higher3003 (36.5)150 (71.1)861 (48.2)1297 (35.1)695 (27.5)
      Household per capita consumption<0.001
       Low2365 (33.1)30 (15.8)430 (27.8)1095 (34.3)810 (36.7)
       Middle2456 (34.4)60 (31.6)505 (32.6)1116 (35.0)775 (35.1)
       High2315 (32.4)100 (52.6)614 (39.6)979 (30.7)622 (28.2)
      Body mass index, kg/m2,

      mean (SD)
      23.7 (3.9)24.3 (3.7)24.0 (4.1)23.7 (3.9)23.5 (3.8)<0.001
       Smoking status<0.001
      Never smoking4976 (60.6)147 (69.7)1188 (66.5)2186 (59.2)1455 (57.6)
      Current/former smoker3241 (39.4)64 (30.3)599 (33.5)1506 (40.8)1072 (42.4)
       Alcohol consumption<0.001
      Never drinking4961 (60.4)134 (63.5)1156 (64.7)2179 (59.0)1492 (59.0)
      Current/former drinker3255 (39.6)77 (36.5)631 (35.3)1512 (41.0)1035 (41.0)
      In terms of individual childhood SED indicators, food insecurity and poor self-perceived household financial situation (i.e., worse off than others) were significantly associated with higher risks of depression and physical multimorbidity. Participants with highest childhood SED score had 2.63 (95 % CI: 1.91–3.63) times and 2.08 (95 % CI: 1.56–2.77) times higher odds of depression and physical multimorbidity respectively compared with those who had lowest score, after adjusting for age and sex. A 1-point increase in childhood SED score was associated with 1.42-fold (95 % CI: 1.34–1.51) and 1.26-fold (95 % CI: 1.19–1.34) increased odds of depression and physical multimorbidity respectively (Table 2). We then explored the impact of different combinations of the three indicators. For subjects who scored a “1” on childhood SED, poor self-perceived household financial situation and food insecurity had relatively greater associations with depression and physical multimorbidity, respectively. For those who scored a “2” on childhood SED, the combination of food insecurity and worse self-perceived household financial situation had relatively greater associations with depression and physical multimorbidity (Supplemental Fig. 2).
      Table 2Associations of childhood socioeconomic disadvantage with the prevalence of depression and physical multimorbidity in the 2018 follow-up survey (bold numbers indicate statistical significance with P value < 0.05).
      Depression cases,

      n (%)
      Odds ratio (95 % CI) for depression
      Adjusted for age and sex.
      Physical multimorbidity cases,

      n (%)
      Odds ratio (95 % CI) for physical multimorbidity
      Adjusted for age and sex.
      Highest education level of parents
       Junior high school or higher239 (36.1)Ref404 (61.0)Ref
       Elementary school or lower2923 (38.7)1.15 (0.97–1.36)5089 (67.4)1.21 (1.02–1.42)
      Food insecurity
       No713 (31.5)Ref1377 (60.9)Ref
       Yes2449 (41.1)1.61 (1.45–1.79)4116 (69.1)1.37 (1.24–1.52)
      Self-perceived household financial situation
       Better off than or same as others1711 (34.4)Ref3205 (64.4)Ref
       Worse off than others1451 (44.7)1.59 (1.45–1.74)2288 (70.6)1.33 (1.21–1.47)
      Childhood socioeconomic disadvantage score
       056 (26.5)Ref109 (51.7)Ref
       1546 (30.6)1.27 (0.92–1.76)1076 (60.2)1.31 (0.98–1.75)
       21403 (38.0)1.89 (1.38–2.60)2507 (67.9)1.77 (1.33–2.35)
       31157 (45.8)2.63 (1.91–3.63)1801 (71.3)2.08 (1.56–2.77)
      1-point increase in childhood socioeconomic disadvantage score1.42 (1.34–1.51)1.26 (1.19–1.34)
      a Adjusted for age and sex.
      In subgroup analysis (Fig. 2.1, Fig. 2.2), compared with subjects with lowest childhood SED score, the OR for depression among subjects with highest score was 1.50 (95 % CI: 0.88–2.54) in males and 3.46 (95 % CI: 2.33–5.15) in females (P-value for interaction term = 0.094); the OR for physical multimorbidity among subjects with highest score was 1.87 (95 % CI: 1.16–3.02) in males and 2.25 (95 % CI: 1.56–3.23) in females (P-value for interaction term = 0.361).
      Fig. 2.1
      Fig. 2.1Odds ratios for the association between childhood socioeconomic disadvantage and depression by sex.
      α: Interaction test (sex × childhood socioeconomic disadvantage score) P-value for interaction term = 0.094.
      β: Adjusted for age.
      Fig. 2.2
      Fig. 2.2Odds ratios for the association between childhood socioeconomic disadvantage and physical multimorbidity by sex.
      α: Interaction test (sex × childhood socioeconomic disadvantage score) P-value for interaction term = 0.361.
      β: Adjusted for age.
      We then assessed the mediating role of depression in the association between childhood SED and physical multimorbidity. Among participants free from any chronic conditions at baseline, the proportion of the association between childhood SED score and incident physical multimorbidity mediated by baseline status of depression was 20 % (Fig. 3). The mediating effect of depression was stronger among females (mediated percentage = 36 %) than in males (mediated percentage = 5 %).
      Fig. 3
      Fig. 3Mediation effect of baseline status of depression in the association between childhood socioeconomic disadvantage score and the incident physical multimorbidity.
      Model was adjusted for age.

      4. Discussion

      Based on a longitudinal, national study on Chinese adults, we found dose-response associations of childhood SED with the risks of depression and physical multimorbidity in later life. Furthermore, the status of depression mediated the association between childhood SED and physical multimorbidity, especially among females.
      In our study, childhood SED, indexed by food insecurity, highest education level of parents, and self-perceived household financial situation, was significantly related to the development of depression and physical multimorbidity, after controlling for age and sex. Our results were consistent with previous studies. Kim et al. found that suspension of school education due to financial strain in childhood was associated with increased OR of depression across stages of adulthood among participants in South Korea [
      • Kim S.S.
      • et al.
      Association between childhood adversities and adulthood depressive symptoms in South Korea: results from a nationally representative longitudinal study.
      ]. Chen and colleagues using CHARLS data found that experiencing severe starvation in childhood (defined as some family members had starved to death) was associated with higher odds of depression (OR = 1.42, 95 % CI: 1.15, 1.49) in old age (60 years old and above) [
      • Chen Y.
      • et al.
      Association between early life circumstances and depressive symptoms among chinese older adults: results from China health and retirement longitudinal study: early life circumstances and depression.
      ]. Another Japanese study showed adverse childhood experiences (≥2 experiences) were associated with depressive symptoms in later life, and the association was partly mediated by adult socioeconomic status [
      • Yazawa A.
      • et al.
      Early childhood adversity and late-life depressive symptoms: unpacking mediation and interaction by adult socioeconomic status.
      ]. Similarly, childhood financial hardship [
      • Tucker-Seeley R.D.
      • et al.
      Lifecourse socioeconomic circumstances and multimorbidity among older adults.
      ] and food shortage [
      • Henchoz Y.
      • et al.
      Childhood adversity: a gateway to multimorbidity in older age?.
      ] were associated with a higher number of chronic conditions or multimorbidity.
      Childhood SED might affect the development of depression and physical multimorbidity in different pathways. Children raised in a disadvantaged environment were more likely to develop insecure adult attachment [
      • Mickelson K.D.
      • Kessler R.C.
      • Shaver P.R.
      Adult attachment in a nationally representative sample.
      ] and acquire a deficit in escaping from aversive conditions after repeatedly facing negative and uncontrollable situations (i.e., learned helplessness), thus increasing their vulnerability to depression in later life [
      • Gilman S.E.
      • et al.
      Socioeconomic status in childhood and the lifetime risk of major depression.
      ]. On the other hand, children with lower socioeconomic status had more obstacles to adopting healthier behaviors and lacked access to health resources and adequate nutrition, leading to higher risks of chronic diseases [
      • Braveman P.
      • Barclay C.
      Health disparities beginning in childhood: a life-course perspective.
      ]. Furthermore, disadvantaged socioeconomic position in childhood was found to be associated with higher level of inflammation in later life [
      • Berger E.
      • et al.
      Multi-cohort study identifies social determinants of systemic inflammation over the life course.
      ], and inflammation is recognized as playing an important role in the pathophysiology of depression [
      • Miller A.H.
      • Raison C.L.
      The role of inflammation in depression: from evolutionary imperative to modern treatment target.
      ] and physical multimorbidity [
      • Friedman E.
      • Shorey C.
      Inflammation in multimorbidity and disability: an integrative review.
      ]. Considering childhood SED is a shared risk factor of depression and physical multimorbidity, improving childhood socioeconomic status is beneficial for both mental and physical health in older adults.
      Our results showed that females were relatively more vulnerable to the adverse impact of childhood SED on the risk of depression than males. Females with highest childhood SED score had 3.46 higher odds of depression, more than twice the number among males (OR in males = 1.50). A previous meta-analysis synthesizing data from >1.7 million subjects found that females had a significantly higher depression risk than males (OR = 1.95, 95 % CI: 1.88–2.03). Furthermore, the sex difference emerged in early life and peaked at age 16 [
      • Salk R.H.
      • Hyde J.S.
      • Abramson L.Y.
      Gender differences in depression in representative national samples: meta-analyses of diagnoses and symptoms.
      ]. A possible explanation of our results is that childhood SED was associated with earlier age at menarche [
      • James-Todd T.
      • et al.
      The impact of socioeconomic status across early life on age at menarche among a racially diverse population of girls.
      ]. Girls with early puberty were underprepared to cope with the life stresses following sexual maturity [
      • Ge X.
      • Conger R.D.
      • Elder Jr., G.H.
      Pubertal transition, stressful life events, and the emergence of gender differences in adolescent depressive symptoms.
      ] and were thus more vulnerable to low self-esteem and other symptoms related to depression than boys [
      • Gallo E.A.G.
      • et al.
      Gender differences in the effects of childhood maltreatment on adult depression and anxiety: a systematic review and meta-analysis.
      ].
      In our study, we found depression mediated the association between childhood SED and physical multimorbidity, which is biologically plausible: depression might beget unhealthy lifestyles (e.g., smoking and physical inactivity), thus increasing the risk of physical multimorbidity [
      • Jha M.K.
      • et al.
      Screening and management of depression in patients with cardiovascular disease: JACC state-of-the-art review.
      ]. Furthermore, we found that the mediating role of depression was stronger in females (mediated percentage = 36 %) than males (mediated percentage = 5 %). Given that depressive symptoms are treatable and easier to screen, prevention strategies to reduce the risk of depression among people with poor childhood socioeconomic status would further benefit public health by decreasing their risk of future physical multimorbidity, especially among females.
      Our findings must be interpreted in consideration of the study limitations. First, information on childhood SED, depression, and chronic conditions was self-reported and thus might be subject to recall bias. In addition, reports on childhood SED may be subject to social desirability response bias, where subjects tend to present a favourable image of themselves [
      • Van de Mortel T.F.
      Faking it: social desirability response bias in self-report research.
      ]. Second, we defined physical multimorbidity only by counting the number of chronic diseases without considering the different clusters and severity of chronic diseases. Third, although we chose both objective (highest education level of parents) and subjective (food insecurity and self-perceived household financial situation) indicators to reflect childhood SED from different dimensions, this measurement has not been validated. Further studies are needed to explore the most appropriate index to represent childhood SED. Fourth, bias might be introduced due to loss of follow-up and the process of selecting study population, where subjects without information on depression, physical multimorbidity, or childhood SED were excluded.
      In conclusion, childhood SED was significantly associated with higher risks of depression and physical multimorbidity in adulthood, and the association of childhood SED with physical multimorbidity was partially mediated by depression. Therefore, it's meaningful to integrate reducing childhood SED into public policy goals to promote healthy aging, considering its multiple long-term adverse health consequences. Furthermore, prevention strategies targeting depression, especially among females, may be helpful to mitigate the adverse impact of childhood SED on the incidence of physical multimorbidity.

      Contributors

      Chuyao Jin carried out the data analyses and wrote the initial draft of the manuscript.
      Xiaochen Dai revised the manuscript critically.
      Gita D. Mishra revised the manuscript critically.
      Yu Wang revised the manuscript critically.
      Xiaolin Xu contributed to the study conceptualization and supervised the project.
      All authors have approved the final manuscript.

      Funding

      This work was supported by the Hundred Talents Program Research Initiation Fund from Zhejiang University, and the Fundamental Research Funds for the Central Universities. The funder of the study had no role in study design, data collection, data analysis, data interpretation, or the writing of the manuscript.

      Ethical approval

      This study was a secondary analysis based on the publicly open dataset of the CHARLS. The CHARLS was approved by the Ethical Review Committee at Peking University and all participating respondents provided written informed consent. Ethical approval was not required for analyzing anonymized data.

      Provenance and peer review

      This article was not commissioned and was externally peer reviewed.

      Research data (data sharing and collaboration)

      The original data for this study are available on the website http://charls.pku.edu.cn/index/en.html.

      Declaration of competing interest

      The authors declare that they have no competing interest.

      Acknowledgments

      We are grateful for the CHARLS team to make the data publicly available. We are also grateful to the participants who provided the survey data.

      Appendix A. Supplementary data

      References

        • Li D.
        • et al.
        A meta-analysis of the prevalence of depressive symptoms in chinese older adults.
        Arch. Gerontol. Geriatr. 2014; 58: 1-9
        • Noel P.H.
        • et al.
        Depression and comorbid illness in elderly primary care patients: impact on multiple domains of health status and well-being.
        Ann. Fam. Med. 2004; 2: 555-562
        • Vancampfort D.
        • Stubbs B.
        • Koyanagi A.
        Physical chronic conditions, multimorbidity and sedentary behavior amongst middle-aged and older adults in six low- and middle-income countries.
        Int. J. Behav. Nutr. Phys. Act. 2017; 14: 147
        • Zhao Y.
        • et al.
        Physical multimorbidity, health service use, and catastrophic health expenditure by socioeconomic groups in China: an analysis of population-based panel data.
        Lancet Glob. Health. 2020; 8: e840-e849
        • Hopman P.
        • et al.
        Effectiveness of comprehensive care programs for patients with multiple chronic conditions or frailty: a systematic literature review.
        Health Policy. 2016; 120: 818-832
        • Ho I.S.
        • et al.
        Examining variation in the measurement of multimorbidity in research: a systematic review of 566 studies.
        Lancet Public Health. 2021; 6: e587-e597
        • Koyanagi A.
        • et al.
        Mortality in unipolar depression preceding and following chronic somatic diseases.
        Acta Psychiatr. Scand. 2018; 138: 500-508
        • Quinones A.R.
        • et al.
        Prospective disability in different combinations of somatic and mental multimorbidity.
        J. Gerontol. A Biol. Sci. Med. Sci. 2018; 73: 204-210
        • Braveman P.
        • Barclay C.
        Health disparities beginning in childhood: a life-course perspective.
        Pediatrics. 2009; 124: S163-S175
        • Cohen S.
        • et al.
        Childhood socioeconomic status and adult health.
        Ann. N. Y. Acad. Sci. 2010; 1186: 37-55
        • Lynch J.
        • Smith G.D.
        A life course approach to chronic disease epidemiology.
        Annu. Rev. Public Health. 2005; 26: 1-35
        • Chen Y.
        • et al.
        Association between early life circumstances and depressive symptoms among chinese older adults: results from China health and retirement longitudinal study: early life circumstances and depression.
        J. Affect. Disord. 2021; 292: 345-351
        • Henchoz Y.
        • et al.
        Childhood adversity: a gateway to multimorbidity in older age?.
        Arch. Gerontol. Geriatr. 2019; 80: 31-37
        • Kim S.S.
        • et al.
        Association between childhood adversities and adulthood depressive symptoms in South Korea: results from a nationally representative longitudinal study.
        BMJ Open. 2013; 3
        • Pavela G.
        • Latham K.
        Childhood conditions and multimorbidity among older adults.
        J. Gerontol. B Psychol. Sci. Soc. Sci. 2016; 71: 889-901
        • Tucker-Seeley R.D.
        • et al.
        Lifecourse socioeconomic circumstances and multimorbidity among older adults.
        BMC Public Health. 2011; 11: 313
        • Yang L.
        Childhood adversity and trajectories of multimorbidity in mid-late life: China health and longitudinal retirement study.
        J. Epidemiol. Community Health. 2020; 75: 593-600
        • Yazawa A.
        • et al.
        Early childhood adversity and late-life depressive symptoms: unpacking mediation and interaction by adult socioeconomic status.
        Soc. Psychiatry Psychiatr. Epidemiol. 2022; 57: 1147-1156
        • Triolo F.
        • et al.
        The complex interplay between depression and multimorbidity in late life: risks and pathways.
        Mech. Ageing Dev. 2020; 192111383
        • Qiao Y.
        • et al.
        Bidirectional association between depression and multimorbidity in middle-aged and elderly Chinese adults: a longitudinal cohort study.
        Aging Ment. Health. 2021; : 1-7
        • Ye B.
        • et al.
        Bidirectional association between physical multimorbidity and subclinical depression in chinese older adults: findings from a prospective cohort study.
        J. Affect. Disord. 2022; 296: 169-174
        • Birk J.L.
        • et al.
        Depression and multimorbidity: considering temporal characteristics of the associations between depression and multiple chronic diseases.
        Health Psychol. 2019; 38: 802-811
        • Salk R.H.
        • Hyde J.S.
        • Abramson L.Y.
        Gender differences in depression in representative national samples: meta-analyses of diagnoses and symptoms.
        Psychol. Bull. 2017; 143: 783-822
        • Alimohammadian M.
        • et al.
        Multimorbidity as an important issue among women: results of a gender difference investigation in a large population-based cross-sectional study in West Asia.
        BMJ Open. 2017; 7e013548
        • Janicki-Deverts D.
        • et al.
        Sex differences in the association of childhood socioeconomic status with adult blood pressure change: the CARDIA study.
        Psychosom. Med. 2012; 74: 728-735
        • Csajbók Z.
        • et al.
        Sex differences in the association of childhood socioeconomic position and later-life depressive symptoms in Europe: the mediating effect of education.
        Soc. Psychiatry Psychiatr. Epidemiol. 2021; 56: 1091-1101
        • Chapman B.P.
        • et al.
        Can the influence of childhood socioeconomic status on men’s and women’s adult body mass be explained by adult socioeconomic status or personality? Findings from a national sample.
        Health Psychol. 2009; 28: 419
        • Zhao Y.
        • et al.
        Cohort profile: the China health and retirement longitudinal study (CHARLS).
        Int. J. Epidemiol. 2014; 43: 61-68
        • Celeste R.K.
        • et al.
        Socioeconomic life course models and Oral health: a longitudinal analysis.
        J. Dent. Res. 2020; 99: 257-263
        • Laitinen T.T.
        • et al.
        Childhood socioeconomic disadvantage and risk of fatty liver in adulthood: the cardiovascular risk in young finns study.
        Hepatology. 2020; 71: 67-75
        • Parks C.G.
        • et al.
        Childhood socioeconomic factors and perinatal characteristics influence development of rheumatoid arthritis in adulthood.
        Ann. Rheum. Dis. 2013; 72: 350-356
        • Andresen E.M.
        • et al.
        Screening for depression in well older adults: evaluation of a short form of the CES-D (Center for Epidemiologic Studies Depression Scale).
        Am. J. Prev. Med. 1994; 10: 77-84
        • Fu H.
        • Si L.
        • Guo R.
        What is the optimal cut-off point of the 10-item Center for Epidemiologic Studies Depression Scale for screening depression among Chinese individuals aged 45 and over? An exploration using latent profile analysis.
        FrontPsychiatry. 2022; 13
        • Barnett K.
        • et al.
        Epidemiology of multimorbidity and implications for health care, research, and medical education: a cross-sectional study.
        Lancet. 2012; 380: 37-43
        • VanderWeele T.J.
        • Robins J.M.
        Directed acyclic graphs, sufficient causes, and the properties of conditioning on a common effect.
        Am. J. Epidemiol. 2007; 166: 1096-1104
        • Yung Y.-F.
        • Lamm M.
        • Zhang W.
        Causal mediation analysis with the CAUSALMED procedure.
        in: Proceedings of the SAS Global Forum 2018 Conference. SAS Institute Inc, Cary, NC2018
        • Mickelson K.D.
        • Kessler R.C.
        • Shaver P.R.
        Adult attachment in a nationally representative sample.
        J. Pers. Soc. Psychol. 1997; 73: 1092-1106
        • Gilman S.E.
        • et al.
        Socioeconomic status in childhood and the lifetime risk of major depression.
        Int. J. Epidemiol. 2002; 31: 359-367
        • Berger E.
        • et al.
        Multi-cohort study identifies social determinants of systemic inflammation over the life course.
        Nat. Commun. 2019; 10: 1-10
        • Miller A.H.
        • Raison C.L.
        The role of inflammation in depression: from evolutionary imperative to modern treatment target.
        Nat Rev Immunol. 2016; 16: 22-34
        • Friedman E.
        • Shorey C.
        Inflammation in multimorbidity and disability: an integrative review.
        Health Psychol. 2019; 38: 791
        • James-Todd T.
        • et al.
        The impact of socioeconomic status across early life on age at menarche among a racially diverse population of girls.
        Ann. Epidemiol. 2010; 20: 836-842
        • Ge X.
        • Conger R.D.
        • Elder Jr., G.H.
        Pubertal transition, stressful life events, and the emergence of gender differences in adolescent depressive symptoms.
        Dev. Psychol. 2001; 37: 404-417
        • Gallo E.A.G.
        • et al.
        Gender differences in the effects of childhood maltreatment on adult depression and anxiety: a systematic review and meta-analysis.
        Child Abuse Negl. 2018; 79: 107-114
        • Jha M.K.
        • et al.
        Screening and management of depression in patients with cardiovascular disease: JACC state-of-the-art review.
        J. Am. Coll. Cardiol. 2019; 73: 1827-1845
        • Van de Mortel T.F.
        Faking it: social desirability response bias in self-report research.
        Aust. J. Adv. Nurs. 2008; 25: 40-48