Study design and participants
We conducted a randomized controlled trial in Taiyuan, which is the capital of the Shanxi Province and has six districts (Yingze, Xiaodian, Xinghualing, Wanbailin, Jinyuan, and Jiancaoping). A multi-stage stratified random cluster sampling method was used to select participants for the intervention from the six districts. First, the six districts were divided into three economic levels using their gross domestic product (GDP) based on government data taken from the website: high (Yingze, Xiaodian), medium (Xinghualing, Jiancaoping), and low (Wanbailin, Jinyuan). Then one district was selected randomly at each level by drawing lots. Second, each community in the three selected districts was numbered according to the order of communities in the government website, then two communities were selected randomly in each district using a random-number table. All the empty-nest elderly living in the selected communities were considered candidates for participation in the study. The empty-nest elderly, who were aged 60 years and above, had no cognitive disorder or other mental illnesses, provided informed consent, were willing and able to complete the investigation, and were residing in the community for at least a year before the study, were eligible for the study. The exclusion criterion was having cognitive disorders or serious diseases, such as deafness, psychiatric disorders, or Alzheimer’s disease.
The sample size was estimated according to the following formula A total of 396 empty nesters were enrolled as participants, of which 204 and 192 were in the intervention and control group, respectively. Figure 1 presents the details of participant involvement. The study was approved by the institutional review boards of Shanxi Medical University. A written informed consent was obtained from all participants.
Interventions
From October 2016 to May 2017, the intervention group participated in a seven-month SMG-based intervention, which consisted of three stages: self-management (1–2 months), mutual-management (3–4 months), and group-management (5–7 months). The first stage was aimed at the empty-nest elderly individuals to develop their self-health management awareness and ability, such as self-care awareness, active medical awareness, self-health assessment ability, and self-service medical equipment use ability. At the mutual-management stage, the empty nesters in the same community were paired according to their age, gender, relationship status, home distance, and other factors to form mutual-management; if necessary, volunteers or study staff members were introduced to participate in the pairing. The third stage involved the implementation of group health management based on the first two levels. Groups were categorized by disease type and residential area. For the former, because the empty-nest elderly often face common health problems, there is a common interest in implementing certain goals. For the latter, the principle of proximity was considered; participants were expected not to drop out of the intervention because of distance problems. Details of the intervention are shown in Figure 2 In the entire intervention process, researchers and community workers played the role of health management instructors to assist in the implementation and development of the three types of management. The control group received a routine follow-up without any intervention. Finally, 350 empty-nest elderly completed the study continuously, of which 167 and 183 completed the seven-month SMG-based intervention and seven-month routine follow-up, respectively. Attrition rate was 11.6%.
Instruments
The general information questionnaire was used to assess the participants’ demographic information, including age, gender, education, marital status, frequency of children visit, employment status, monthly income, social activity participation, self-care ability, and chronic disease.
The primary outcome variable, that is, quality of life of the empty-nest elderly, was measured using the Short Form 36-Item Health Survey (SF-36), which was developed in the United States and designed to allow self-evaluation of quality of life 18. It made up of 36 questions and two summary scores, namely, Physical and Mental Component Summary (PCS and MCS, respectively). The PCS encompassed the following dimensions: physical function (PF), role physical (RP), bodily pain (BP), and general health (GH). The MCS contained the following dimensions: vitality (VT), social function (SF), mental health (MH), and role emotional (RE). Individual item scores were summed up and transformed into a 0–100 scale, ranging from the worst to the best possible quality of life 19. At present, the SF-36 has been used widely in evaluating quality of life, including both the physical and mental health of the empty nest elderly in China. In this study, the scale’s Cronbach’s α was 0.751.
Questionnaires were administered at baseline and post-intervention (seven months). The questionnaire was completed following a face-to-face interview between an interviewer and a participant and collected on the spot. To ensure quality, completed questionnaires were checked carefully by quality supervisors after the interview. The response rate was 100%.
Statistical analysis
EpiData was used for entering and checking the validity of data, and SPSS 22.0 software for statistical analysis. Data were expressed as mean ± SD (standard deviation). Difference between groups in terms of baseline characteristics was tested using chi-square test. The effect of the intervention versus control conditions was examined using ANCOVA analysis on post-intervention measurement values, controlling for pre-intervention scores. Cohen’s d was provided to evaluate the effect size with a guideline: trivial (<0.20), small (0.20 to <0.50), moderate (0.50 to <0.80), and large (≥0.80) 20. P value was statistically significant at (P < 0.05).