Abstract
This cross-sectional study explores the association of adverse childhood experiences and frailty index among community-living adults aged at least 45 years in China.
Introduction
Adverse childhood experiences (ACEs) are recognized as an important public health issue.1 The most widely used ACE scale, the Kaiser Permanente ACE Study, includes 10 items across domains of child abuse, neglect, and household dysfunctions. It is generated based on a sample of primarily White and educated individuals.2 Recently, a set of expanded ACEs (eg, socioeconomic [SE] hardships, negative peer relationships, and community-level ACEs) has been measured to understand their health implications.1 However, the results differ by sample population. For example, a prospective cohort study in the UK showed no difference in the association between ACEs and health outcomes across SE groups, whereas in China, childhood SE status was significantly associated with later health.2,3 Thus, in terms of health, what constitutes ACEs remains unknown in a specific context, which was thought to be shaped by social contexts.4 Because China is a unique case owing to Confucian culture and massive social shifts, this study attempts to explore health-based ACEs based on the association between 18 ACEs and frailty index (FI) in China.5
Methods
This population-based cross-sectional study used the China Health and Retirement Longitudinal Study (CHARLS) data, providing a wide range of information from SE status to health conditions of community-living adults aged at least 45 years.6 After excluding missing values, those participated in the CHARLS surveys 2011, 2013, 2015, and 2018, and life history survey in 2014 were included. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline and is not subject to ethical approval according to the London School of Economics and Political Science research ethics policy.
Individual ACEs, covering intrafamilial aggression, family dynamic, SE deprivation, and neighborhood quality, were obtained from a life history survey. Based on 41 measurements collected from CHARLS 2011, 2013, 2015, and 2018, FI was calculated and categorized as robust, prefrail, and frail, with its trajectories classified as stable at robust and prefrail, and rapidly rising to frail.4
Weighted generalized ordered logistic and logistic models were applied adjusting for sociodemographic factors and health behaviors in adulthood. Two-tailed P < .05 indicated statistical significance, with OR and 95% CIs reported. Data analysis was performed from December 1 to 30, 2021. Stata version 15 (StataCorp LLC), was used. See eMethods in the Supplement for details.
Results
Among the 11 568 respondents (mean [SD] age, 57.95 [9.16] years; 6230 female respondents [54%]), 1869 (16%) reported frail status, with 2342 (20%) having rapidly rising to frail trajectory (Table 1). An increase in the number of ACEs was associated with a 20% (odds ratio [OR], 1.20; 95% CI, 1.16-1.23) and 19% (OR, 1.19; 95% CI, 1.16-1.23) increase in the likelihood of being in frail status, and rapidly rising trajectory. Loss or threat of loss within the family (eg, household mental or serious illness, serious childhood illness or injury), socioeconomic deprivation (excluding child hunger), low-quality neighbors, and peer bullying victimization were related to FI, whereas the prevalence of exposure to family dynamics (1154 of the 11 568 respondents) was less than 10%, and not significantly associated with FI. After adjustments of co-occurring ACEs, a limited association of intrafamilial aggression was found, which may be due to the prevailing views of stricter parenting practices in China (Table 2).
Table 1. Participant Descriptive Statistics Before and After Imputation and Weighting.
| Characteristic | Before imputation and weighting, No. (%) | After imputation and weighting, % |
|---|---|---|
| Total sample (n = 11568) | ||
| FI categories | ||
| Robust (frailty index ≤0.10) | 1725 (14.91) | 17.44 |
| Prefrail (>0.10 to <0.25) | 7974 (68.93) | 67.60 |
| Frail (frailty index ≥0.25) | 1869 (16.16) | 14.96 |
| FI trajectory | ||
| Stable at robust or prefrail | 9206 (79.58) | 80.99 |
| Rapidly rising to frail | 2362 (20.42) | 19.01 |
| FI for stable at robust or prefrail group, mean (SD) | ||
| FI in 2011 | 0.13 (0.05) | 0.13 (0.001) |
| FI in 2013 | 0.14 (0.05) | 0.14 (0.001) |
| FI in 2015 | 0.14 (0.06) | 0.14 (0.001) |
| FI in 2018 | 0.15 (0.05) | 0.15 (0.001) |
| FI for rapidly rising to frail group, mean (SD) | ||
| FI in 2011 | 0.25 (0.09) | 0.25 (0.002) |
| FI in 2013 | 0.30 (0.10) | 0.29 (0.002) |
| FI in 2015 | 0.33 (0.11) | 0.33 (0.003) |
| FI in 2018 | 0.34 (0.12) | 0.35 (0.003) |
| No. of ACEs, mean (SD), No. | 3.26 (1.77) | 3.17 (0.02) |
| Age in 2011, mean (SD), y | 57.95 (9.16) | 58.47 (0.11) |
| Female | 6230 (53.86) | 52.24 |
| Male | 5338 (46.14) | 47.76 |
| Ever being unmarried from 2011 to 2018 | 2634 (22.77) | 23.68 |
| Education | ||
| Illiterate | 3262 (28.20) | 25.32 |
| Elementary school | 2129 (18.40) | 17.02 |
| Junior high school | 2544 (21.99) | 21.70 |
| High school or above | 3633 (31.41) | 35.96 |
| Ever being heavy drinker | 2921 (25.25) | 24.64 |
Abbreviations: ACE, adverse childhood experience; FI, frailty index.
Table 2. Association of ACE With FI Categories and Trajectories Using Imputed Data: Weighted Generalized Ordered Logistic Model and Logistic Model.
| Characteristic | Total samplea, No. (%) (N = 11 568) | Effect size for association, OR (95% CI)b | ||
|---|---|---|---|---|
| FI categoriesc | FI trajectoriesd | |||
| Robust vs prefrail | Prefail vs frail | |||
| Association between cumulative score of ACEs and FI | ||||
| ACEs, mean (SD), No. | 3.26 (1.77) | 1.19 (1.14-1.25) | 1.19 (1.15-1.23) | 1.20 (1.16-1.24) |
| Association between types of individual ACEs and FI | ||||
| Loss or threat of loss within the family | ||||
| Household mental illness | 420 (3.63) | 1.37 (0.84-2.02) | 1.57 (1.18-2.07) | 1.57 (1.20-2.06) |
| Severe illness in family | 2449 (21.17) | 1.37 (1.11-1.70) | 1.52 (1.31-1.77) | 1.50 (1.30-1.73) |
| Parental emotion issue | 3949 (34.14) | 1.20 (0.99-1.47) | 1.05 (0.92-1.21) | 1.09 (0.95-1.25) |
| Serious childhood illness or injury | 1001 (8.65) | 1.37 (1.03-1.81) | 1.67 (1.34-2.08) | 1.66 (1.36-2.03) |
| Childhood socioeconomic status | ||||
| Low parental education | 6079 (52.55) | 1.33 (1.14-1.55) | 1.22 (1.07-1.39) | 1.28 (1.13-1.45) |
| Parental unemployment | 1154 (9.98) | 1.33 (1.03-1.71) | 1.14 (0.94-1.41) | 1.08 (0.88-1.31) |
| Family financial problems | 4652 (40.21) | 1.23 (1.03-1.47) | 1.28 (1.12-1.46) | 1.36 (1.20-1.54) |
| Childhood hunger | 3319 (28.69) | 0.74 (0.63-0.86) | 0.71 (0.70-0.95) | 0.81 (0.70-0.94) |
| Childhood intrafamilial aggression | ||||
| Parental physical maltreatment | 3186 (27.54) | 1.12 (0.92-1.37) | 1.10 (0.95-1.27) | 1.11 (0.97-1.28) |
| Emotional neglect | 3949 (34.14) | 1.14 (0.98-1.33) | 1.09 (0.94-1.25) | 1.08 (0.95-1.23) |
| Sibling aggression victimization | 692 (5.98) | 1.24 (0.88-1.75) | 1.09 (0.86-1.38) | 1.09 (0.87-1.36) |
| Witnesses of inter-parental violence | 877 (7.58) | 1.19 (0.89-1.40) | 1.12 (0.90-1.40) | 1.15 (0.93-1.41) |
| Family dynamics | ||||
| Poor parent-child relationship | 204 (1.76) | 1.41 (0.57-3.50) | 1.49 (0.94-2.36) | 1.50 (0.97-2.33) |
| Parental separation or divorce | 100 (0.86) | 1.28 (0.47-3.50) | 1.37 (0.77-2.45) | 1.50 (0.88-2.55) |
| Household substance abuse | 802 (6.93) | 1.17 (0.86-1.60) | 1.13 (0.89-1.42) | 1.09 (0.87-1.36) |
| Household criminality | 128 (1.11) | 0.73 (0.26-2.02) | 0.96 (0.59-1.56) | 0.80 (0.49-1.30) |
| Neighborhood quality | ||||
| Low-quality neighbors | 5470 (47.29) | 1.22 (1.07-1.38) | 1.22 (1.07-1.38) | 1.21 (1.08-1.37) |
| Peer bullying victimization | 1723 (14.89) | 1.07 (0.86-1.37) | 1.19 (1.00-1.41) | 1.12 (0.98-1.29) |
Abbreviations: ACE, adverse childhood experience; FI, frailty index.
Includes 11 568 observations for each effect size.
Eighteen ACEs were conceptualized as a cumulative score based on the total number of ACEs experienced and as individual ACE types. The 18 individual ACEs were controlled simultaneously when examining the associations between types of ACEs and FI. Weighted generalized ordered logistic model and weighted logistic model were applied using imputed data with demographic characteristics (sex, marital status, and age), educational attainment, and ever being a heavy drinker controlled. The weight of the longitudinal data was calculated using the sample attrition adjustment method to correct the attribution bias in the cohort.
Three categories of FI were defined: robust (frailty index ≤0.10), prefrail (frailty index >0.10 to <0.25), and frail (frailty index ≥0.25).
FI trajectories were classified into 2 groups (eg, stable at robust and prefrail, rapidly rising and frail) using group-based trajectory modeling.
Discussion
Cumulative ACEs were associated with increased frailty events and a faster decline in FI in their middle and older age. However, the associations of different types of adversities were heterogeneous. Older Chinese people were sensitive to expanded ACEs including socioeconomic deprivation (excluding child hunger), low-quality neighbors, and peer bullying in FI.
Limitations of the study include retrospective self-evaluations, and the study did not include some common ACEs (eg, sex abuse) and identify all the health-based ACEs. Our findings should be applied to other countries with caution.
eMethods. Details on Methodology and Sample
References
- 1.Ports KA, Ford DC, Merrick MT, Guinn AS. ACEs: Definitions, measurement, and prevalence, IN Asmundson GJ, Afifi TG. Adverse Childhood Experiences. Academic Press; 2020:17-34. [Google Scholar]
- 2.Houtepen LC, Heron J, Suderman MJ, Fraser A, Chittleborough CR, Howe LD. Associations of adverse childhood experiences with educational attainment and adolescent health and the role of family and socioeconomic factors: A prospective cohort study in the UK. PLoS Med. 2020;17(3):e1003031. doi: 10.1371/journal.pmed.1003031 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Wang Q, Kang WW. Childhood socioeconomic circumstances, social status, and health in older age: are they related in China? Life Course Res. 2019;42:100289. doi: 10.1016/j.alcr.2019.100289 [DOI] [PubMed] [Google Scholar]
- 4.Ortiz R. Contextualizing adverse childhood experiences-intersections with structural adversity and imperatives for future research. JAMA Netw Open. 2021;4(10):e2130950. doi: 10.1001/jamanetworkopen.2021.30950 [DOI] [PubMed] [Google Scholar]
- 5.Cao M, Lu X, Qin Z, Liu X, Li F. Social-ecological system changes in China from 1990 to 2018. Ecol Indic. 2022;139:108926. doi: 10.1016/j.ecolind.2022.108926 [DOI] [Google Scholar]
- 6.Zhao Y, Hu Y, Smith JP, Strauss J, Yang G. Cohort profile: the China Health and Retirement Longitudinal Study (CHARLS). Int J Epidemiol. 2014;43(1):61-68. doi: 10.1093/ije/dys203 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eMethods. Details on Methodology and Sample