Study design, participants and procedure
This study adopted a cross-sectional design using an online questionnaire survey. To be eligible for inclusion, participants needed to meet the eligibility criteria: 1) aged 55 years or older; 2) have not been infected with the COVID-19; 3) have access to a mobile phone or laptop with internet connection; and 4) are able to read Chinese (for Chinese samples) or German (for German sample).
In China, participants were recruited using a convenient sampling approach from Wuhan, Hubei Province of China, which was the most seriously infected region during the COVID-19 pandemic in China in 2020. Data collection started on June 15, 2020 and was completed on July 10, 2020 (the lockdown had been withdrawn for around two months). We invited 434 older adults to attend the survey and 356 participants (Mean age = 67.75, SD = 6.24, age range: 58-89) completed the online survey (82% response rate). The survey was constructed and administered using an online survey platform in China, namely SOJUMP (Changsha Ranxing Information Technology Co., Ltd., China). All recruitment posters and the hyperlink for the survey were disseminated via a mobile Short Message Service (SMS) and popular social media platforms in China (e.g., WeChat, Weibo, and QQ). Three approaches were used for recruiting participants: 1) Relying on the researchers’ personal social networks in Wuhan, the eligible family members, friends and relatives of the researchers were invited. The participants then encouraged their friends to attend the survey; 2) Researchers contacted the directors of community neighborhood committees in Wuhan, respectively and sought their collaboration and support. Upon receiving the agreement of directors, researchers were permitted to enter into their community neighborhood WeChat groups to recruit eligible participants; 3) Researchers contacted officials who were in charge of the retirement in two universities in Wuhan. With the support of officials, a recruitment poster and survey hyperlink were delivered to their internal WeChat group, especially for retirement colleagues.
In Germany, participants were recruited through snowball sampling (personal contacts), press releases, social media groups (such as Facebook groups) and newspaper articles. Data collection was not limited to a city or a state. Rather a national wide recruitment was performed. The survey was set up and administered via the online survey tool Unipark. All data was collected anonymously. Data collection started on June 16, 2020 and was completed on February 17, 2021. 264 participants attended the survey. After excluding participants aged below 55 years, 222 eligible participants (Mean age = 69.09, SD = 6.9, age range: 55-86) were involved in the study.
All participants in China and Germany were asked to sign an informed consent form on the first page of the survey platform before completing the questionnaires. Ethical approval for the study in China was obtained from the Research Ethics Committee of Hong Kong Baptist University (REC/19-20/0490). For the German study, ethical approval was obtained from the Ethics Committee of Jacobs University (Application Number: 2020_09).
The sample in China was different from the sample in Germany concerning the majority of demographic variables, including age (t 576 = -2.41, p < .05), Body Mass Index (BMI) (t 576 = -7.72, p < .001), gender (females China =39.6%, females Germany = 63.5%; χ21 = 31.28, p < .001), marital status (single china = 16.9%, single Germany = 86.9%; χ21 = 275.07, p < .001), education level (secondary school and above china = 93.7%, secondary school and above Germany = 84.3%; χ21 = 21.8, p < .001), occupation status (unemployed china = 98.6%, unemployed Germany = 76.9%; χ21 = 72.79, p < .001), household income (average and above china = 79.8%, average and above Germany = 89.7%; χ21 = 70.78, p < .001), living situation (living with children/spouse china = 91%, living with children/spouse Germany = 68.5%; χ21)= 47.81, p < .001), chronic disease (yes china = 53.1%, yes Germany = 42.8%; χ21 = 5.57, p < .05), infected acquaintances (yes china = 12.6%, yes Germany = 37.4%; χ21 = 51.96, p < .001), and perceived health status (satisfactory and above china = 91.6%, satisfactory and above Germany = 85.5%; χ2 1 = 29.92, p < .001). See Table 1.
A series of questionnaires were used to investigate older adults’ demographic information, preventive behaviors, motivational and volitional factors of preventive behaviors. All questionnaires were adapted from well-established ones in previous studies and back-translated to Chinese and German by 2 independent bilingual translators. Each participant took 15-20 minutes to complete all online questionnaires. The questionnaire items and reliability are presented as follows:
The demographic characteristics included age, gender, marital status, country, living situation, education level, occupational status, household income, children status. Participants were also invited to report their chronic disease situation, infected acquaintances, perceived health status, height (cm) and weight (kg).
Preventive behaviors during the COVID-19 pandemic
Hand washing behavior was measured using two items in accordance with the World Health Organization’s (WHO) recommendations. The frequency of hand washing behavior was evaluated with the stem “During the previous week, how frequently did you wash your hands with soap and water or alcohol-based hand rub (for at least 20 seconds, all surfaces of the hands)… ”, followed by two kinds of situations, i.e., “in the daily life situations (e.g., before preparing food; before eating; after defecation)” or “in disease-related situations (e.g., after blowing nose or sneezing; before and after caring for the sick)”. Older adults were asked to rate the two items on a 4-point Likert scale ranging from (1) never to (4) always. A higher total score indicated better hand washing behavior.
Facemask wearing behavior was measured with two items in accordance with the WHO recommendations. The questions were asked using the stem “During the previous week, I have usually worn a facemask properly…” followed by two different situations relevant to older adults, i.e., “when visiting public places (e.g., public transportation, supermarket)”, and “caring for a person with suspected COVID-19 infection”. Responses were scored on a 4-point Likert scale ranging from (1) strongly disagree to (4) strongly agree.
Physical distancing behavior was measured with two items according to the WHO recommendations. Participants were asked to assess their physical distancing behavior during the past week, with items such as “a) usually stayed out of crowded places or mass gatherings, and b) usually kept space (at least 1.5 meters) between myself and other people who are coughing or sneezing.” Answers were given on a 4-point Likert scale from (1) strongly disagree to (4) strongly agree.
Past preventive behaviors before the COVID-19 pandemic
Participants were asked to recall their three preventive behaviors before the pandemic of COVID-19 respectively. Items of each past preventive behavior were identical to those during the COVID-19 pandemic aforementioned.
Motivational Factors of Preventive Behaviors
Risk perception was measured using one item for three preventive behaviors respectively, which was adapted from previous studies [26,27]. The participants were asked” Compared to an average person of your age and gender, what is your risk of COVID-19 infection from lack of frequent hand washing/facemask wearing/physical distancing?” with responses rated on a 6-point Likert scale from 1 = very low to 6 = very high.
Health knowledge was measured using one item for three preventive behaviors respectively, which was adapted from previous studies [28,29]. The participants were asked “Have you known how and in what situations to wash hands/ wear a facemask/ keep a safe physical distancing in accordance with the WHO recommendations?” with responses rated on a 4-point scale with 1 (do not know), 2 (a little), 3 (most) and 4 (all). The higher score represented more sufficient health knowledge.
Attitude was assessed using a common stem on three preventive behaviors. Such as “For me to wash hands frequently/wear a facemask/keep a safe physical distance during the outbreak of COVID-19 would be…” followed by two semantic differential items. Items were rated on a 6-point Likert scale: troubling-reassuring (1-6) and optional-necessary (1-6) [30,31]. A high total score means a positive attitude. The Cronbach alpha coefficient was .69 (China) and .75 (Germany) for hand washing behavior, .74 (China) and .77 (Germany) for mask wearing behavior and .80 (China) and .65 (Germany) for physical distancing behavior.
Subjective norm was assessed using one item measuring participants’ perceptions of important others’ approval on the three preventive behaviors [14,32]. The participants were asked “Most people who are important to me (e.g., my family members, friends, doctors) think that I should wear a facemask during the outbreak of COVID-19.” with responses rated on a 6-point Likert scale, from 1 = strongly disagree to 6 = strongly agree.
Intention was assessed with one item for three preventive behaviors respectively, which was adapted from previous studies [33,34]. The participants were asked “Today and in the near future, I intend to frequently wash my hands in various situations (e.g., before eating, after going to the washroom, after blowing my nose or sneezing)” for hand washing, “Today and in the near future, I intend to properly wear a facemask in various situations (e.g., visiting public places)” for mask wearing behavior, and “Today and in the near future, I intend to keep a safe physical distance in various situations (e.g., staying out of crowded places or mass gatherings when I go outside of my home)” for physical distancing. Items were rated on a 6-point Likert scale, from 1 = strongly disagree to 6 = strongly agree.
Motivational self-efficacy was assessed using one item measuring older adults’ level of confidence in starting to act on preventive behaviors. The participants were asked “I feel certain that I can begin to wash my hands frequently/ wear a facemask/ keep a safe physical distance, even if it would be difficult to change my routines.” with responses rated on a 6-point Likert scale, from 1 = totally disagree to 6 = totally agree [19,33].
Volitional Factors of Preventive Behaviors
Volitional self-efficacy was assessed using one item measuring participants’ confidence of recovery of the behaviors, respectively. The participants were asked “I feel certain that I can restart to wash my hands frequently/wear face mask/keep a secure physical distance even if I forgot to do it a few times” with responses rated on a 6-point Likert scale, from 1 = totally disagree to 6 = totally agree [19.33].
Planning included action planning and coping planning. Action planning was assessed with one item for three preventive behaviors respectively. The items were “I have already made a concrete action plan for hand washing regarding when, where and how to…” followed by “wash my hands/wear face mask/keep a safe physical distance”. Coping planning was assessed by the item “I have made a coping plan to maintain frequent hand washing/mask wearing/physical distancing if I am confronted with some barriers”. Answers were given on a 6-point Likert scale from 1 = totally disagree to 6 = totally agree [6,13, 33,35]. The Cronbach’s alpha coefficient was .75 (China) and .80 (Germany) for hand washing behavior, .84 (China) and .82 (Germany) for mask wearing behavior and .74 (China) and .83 (Germany) for physical distancing behavior.
Self-monitoring was assessed using one item measuring participants’ perceptions of their self-regulation over the preventive behaviors. The participants were asked “I have consistently monitored myself about how and in what situations to wash my hands/wear a face mask/keep a safe physical distance”, with responses rated on a 6-point Likert scale, from 1 = strongly disagree to 6 = strongly agree [13,35].
Data analysis was conducted using IBM SPSS 26.0 (Armonk, NY, USA). Descriptive analyses including percentages used to present demographic differences between Chinese and German older adults and were examined with an independent t-test and Chi-squared test. Moreover, the association of demographics and past preventive behaviors with the current three preventive behaviors were examined by t-tests, F-tests and Pearson/Spearman correlations. In addition, a series of univariate linear regressions were used to analyze the associations of social-cognitive factors with three preventive behaviors after control demographics and past behaviors. Furthermore, the moderating effect of the country on the association between social-cognitive factors and preventive behaviors was examined using multiple hierarchical linear regressions, where all independent variables were standardized using Z scores to avoid the collinearity problem. To further elaborate the magnitude of the association between preventive behaviors and their associated factors in regression analyses, effect size (f 2) was estimated with the conversion formula: f 2 = R2/ (1-R2), with .02, .15, and .35 indicating a small, medium and large effect, respectively.