Study design and participants
This study is a prospective cohort study conducted among older adults residing in communities in Shanghai, China. The baseline survey (wave 1) was conducted from June 2020 to December 2020, the first follow-up (wave 2) was completed between June 2021 and December 2021, and the second follow-up (wave 3) is currently being conducted. The participants, who were aged 60 years and above, were randomly selected using a two-stage sampling method. Sixteen communities in Shanghai were conveniently chosen, and 300 older adults were randomly selected from each community. To be eligible for participation, individuals had to be aged 60 years or older and without progressive tumors or severe mental disorders. Trained interviewers either visited the participants at their homes or invited them to a Community Healthcare Center. The surveys were conducted face-to-face using a self-administered electronic questionnaire. For this study, data from wave 2, including a sample of 4,522 individuals, was analyzed. The research protocol involving human participants was reviewed and approved by the Research Ethics Committee of the Medical Research at the School of public health, Fudan University. Before participating in the study, the participants or illiterate older people’s family members provided written informed consent.
Measures
Basic information
Basic information regarding the participants included age (year), sex (man/ woman), education (illiterate or not complete primary school, primary school, middle school, and high school or above), marital status (married/ the other (including unmarried, divorced, and widowed)), monthly income (< 2000 China Yuan (CNY), 2000–4999 CNY, and ≥ 5000 CNY), live alone (yes/ no), self-reported 20 chronic diseases diagnosis (yes/ no, including Hypertension, Diabetes, Coronary Heart Disease, Stroke, Arrhythmia, Myocardial Infarction, Congestive Heart Failure, Cervical Spondylosis, Herniated Disk, Arthritis, Osteoporosis, Asthma, Chronic Lung Disease, Chronic Kidney Disease, Digestive Diseases, Thyroid Disease, Parkinson’s Disease, Cancer, Dementia, and Epilepsy), and cognitive function (measured using the Mini-Mental State Examination (MMSE) [24]).
Frailty
Frailty was measured by the Chinese Frailty Screening scale (CFSS), which includes 10 items that are scored on 0 (no) and 1 (yes). The total score ranged from 0 to 10, with a higher score indicating greater severity of frailty [25]. The internal consistency of the CFSS-10 in the present study showed good (Cronbach’s α = 0.689).
Depressive symptoms
The measurement of DS was conducted using the Patient Health Questionnaire (PHQ) [26]. The scale consists of 9 items that are scored on a scale of 0 (not at all) to 1 (several days or more), with total score ranges from 0 to 9 in this study. A higher score of the PHQ-9 indicates a greater severity of DS. In the present study, the internal consistency of the PHQ-9 was found to be very good, with a Cronbach's α of 0.852.
Attitudes to aging
AA were assessed using the short form of the Attitudes to Aging Questionnaire (AAQ) [27], which consisted of 12 items scored on a 6-point Likert scale ranging from 1 (very disagree) to 6 (very agree). The total score on the AAQ-12 ranged from 12 to 72, with higher scores indicating more positive attitudes to aging and lower scores indicating more negative attitudes to aging. The internal consistency of the AAQ-12 in our study was good, with a Cronbach’s α of 0.798.
Ageism
Ageism was measured using a questionnaire consisting of 5 questions, which aimed to report experiences of age-based discrimination in everyday life over the past year [28, 29]. These items included situations where people act as if you are not smart, people act as if they are afraid of you, being treated prejudice or unfair, being treated with less respect than others, being treated badly such as insulting, abusing, or refusing services. Respondents were asked to rate the frequency of each event on a 5-point Likert scale, ranging from 0 (not at all) to 4 (always). The total score ranged from 0 to 20, with 0 indicating not ageism and 1 or above indicating ageism [28]. In the current study, the internal consistency of the 5 items demonstrated excellent reliability (Cronbach’s α = 0.934).
Statistical analysis
Descriptive analyses were conducted to present the characteristics and examine correlations of the four variables (frailty, DS, AA, and ageism). The moderated mediation models were employed, which included the independent variable (frailty/ DS), dependent variable (DS/ frailty), mediator (AA), and moderator (ageism). These models were analyzed using Model 15 in the PROCESS Marco (Fig. 1) [30]. To assess mediation and moderation effects, bootstrapping with 5000 bootstrap samples was performed to estimate direct and indirect effects, accompanied by 95% confidence intervals (CI). Additionally, potential confounding variables, such as age, sex, education, marital status, monthly income, live alone, chronic diseases, and cognitive function, were adjusted as covariates in the models. A 95% CI that includes zero indicates no significant mediating (indirect) or moderating effects at a significance level of 0.05. All analyses were conducted using SPSS 22.0 for Windows.