Study protocol of a multi-center RCT testing a social-cognitive intervention to promote volunteering in older adults against an active control

Ethical approval for this RCT was obtained by the Human Research Ethics Committee (HREC) of the Education University of Hong Kong (HREC number 2015–2016-0324). The study had been prospectively registered in the Clinical Trials Registry of the Center for Clinical Research and Biostatistics (CCRB) recommended by the World Health Organization (Primary Registry and Trial Identifying Number ChiCTR-IIC-17010349, secondary CCRB trial number CUHK_CCRB00543, www2.​ccrb.​cuhk.​edu.​hk/​registry/​public/​416). The RCT is funded by the Public Policy Research (PPR) Funding Scheme, Policy Innovation and Co-ordination Office (PICO) of the Hong Kong Special Administrative Region Government (project number 2016.A5.023.16C).

This is the first version of the study protocol. No further study protocols will be published.

A sample of community-dwelling adults aged 50 and older is recruited via invitation of local community centers within all districts of Hong Kong. The eligibility criteria are: 1) 50 years of age and older; 2) not acutely physically impaired or disabled; 3) not seriously cognitively impaired or severely depressed; 4) not working on a full-time or part-time basis (fewer than 20 h per week); 5) able to read and write simple Chinese; and 6) engaged in formal volunteer activities fewer than four times in the past year. Except for meeting the eligibility criteria there were no further exclusion criteria.

Through local community centers, participants are scheduled for one of several group orientation meetings held at their nearest center, during which details of the study’s purpose, time commitment involved in study participation, and inclusion criteria are explained. Individuals who meet the criteria are asked to sign a written informed consent form. The research team ensures confidentiality of this data by keeping questionnaire data and consent forms in different locked filing cabinets, to which only research assistants, who signed a data protection form have access. All participants are reimbursed with HKD 200 (approx. 25.50 USD or 22 EUR) in the form of a voucher for a supermarket if they complete all intervention sessions and all four assessment points. Participants cannot be blinded as it becomes obvious whether they are randomized to the experimental group targeting volunteering or the active control group targeting physical activity during the social-cognitive intervention.

The study by Warner et al. [30] found an intervention effect with an effect size of 0.38 (Cohen’s d). For the power calculation of the intervention effect in the proposed study, an estimated 15% dropout rate over the study period, a power of 90% and an alpha level of 5% is set. To detect this effect size using T-tests, a minimum sample size of 240 is needed.

The study is conducted in 15 centers belonging to two NGOs’ elderly centers, located in 9 out of 18 districts in Hong Kong. In each center, participants are assigned individually and at random to either the experimental or control group by the facilitator. Experimental and control groups are expected to run with 10 to 15 participants. Due to randomization within clusters, the statistical power might be slightly diminished; therefore, the aim is to recruit 360 instead of 240 participants in total.

At baseline, a research assistant or a part-time interviewer interviews participants in a face-to-face format using the structured questionnaire. The intervention period for both groups comprises 4 weeks, with one session lasting approximately 1 hr conducted each week at the respective senior centers. After the intervention and active control intervention period, participants are interviewed three times in a face-to-face format, at 6 weeks, 3 months, and 6 months after the intervention period. Intensive training is provided to the part-time interviewers, to ensure reliability.

In accordance with the volunteering intervention developed by Warner et al. [30], which serves as a basis for this refined intervention, and in lack of a social-cognitive theory specifically for volunteering, the well-established social-cognitive framework of the health action process approach (HAPA; [29]) is chosen as theoretical backdrop for intervention development. This model postulates that behavior change is based upon developing an intention, which is built upon risk-perception (not assumed to be relevant for volunteering), outcome expectancies, and self-efficacy in the motivational phase, whereas intentions are translated into behavior via self-regulatory strategies such as planning and self-monitoring in the volitional phase [29]. As non-volunteers are the target group of the current study, motivational as well as volitional (self-regulatory) strategies are chosen to support participants in the initiation as well as maintenance phase of their new volunteer activity.

One experienced social worker and one research assistant are hired to deliver the intervention to the experimental group and the active control group conditions for all sessions, respectively. They learn all session protocols by heart and read the standardized intervention manual before every session. The research team provides them with intensive training with the manual and role-play exercises. Moreover, the sessions are structured using standardized PowerPoint presentations to ensure the comparability of session contents. To enhance fidelity to the protocol and quality of intervention delivery, at least five sessions in the intervention condition and five sessions in the active control condition are observed and rated by the fourth and fifth author of this manuscript.

The following behavior change techniques (BCTs) are used in four interactive group intervention sessions to prompt volunteering and categorized according to Michie’s BCT taxonomy [30]: information about the benefits of volunteering in old age (prompt for outcome expectancies), focus on past success (prompt for self-efficacy in biography worksheet), goal setting behavior and outcome (prompt for intention formation in worksheet), action planning and use of cues (prompt for use of planning strategy in worksheet), modeling behavior (prompt for self-efficacy in a 5-min video clip with older person as role model). Furthermore, information material on volunteer opportunities for older adults in their residential area is available for free.

Four one-hour sessions are conducted: 1) introduction to the study and the meaning of retirement; 2) quiz regarding volunteer engagement; 3) ideas for volunteer engagement; 4) self-regulation techniques to translate intentions into behavior. These topics are delivered to participants by means of weekly one-hour sessions with two breaks.

In the first session, the facilitator briefly introduces the research team, the goals of the study, the target population, schedule, follow-up questionnaires, and financial incentive. An ice-breaking game gives group members the chance to get to know each other. The facilitator discusses the personal meaning of retirement with participants by asking them about their ideas (e.g., positive or negative connotations, retirement expectations, elements of their ideal retirement, favorite activities during their retirement). The facilitator writes down participants’ responses on a whiteboard to acknowledge their ideas; she also provides more ideas, such as the development of hobbies, activities with friends, activities with family, doing work around the flat, travelling, and cultural events. Finally, the facilitator introduces the idea that one further important aspect of retirement could be doing something for others or taking an active part in volunteering by mentioning the social relevance of volunteer activities as well (retaining knowledge, caring for those in need, strengthening social integration and older people’s role in society). Some individual benefits of volunteering (effects on physical and psychological health) and societal advantages as well as economic aspects are presented to participants at the end of this first session.

In the second session, the facilitator first discusses the positive outcomes of engaging in volunteer activities with the group members. The facilitator asks if participants expect any benefits from being volunteers, whether they perceive those benefits as being relevant to themselves, and their underlying reasons for wanting to volunteer. Information on the positive consequences of volunteering in old age is delivered through a quiz on volunteer engagement. The quiz includes information on the prevalence of volunteering in the general population in Hong Kong and among Hong Kong’s older adults, future trends in volunteering in old age, financial rewards for volunteer activities, reasons for engaging in or barriers for not engaging in volunteer activities (too old, too busy, poor physical health, etc.), and ways to volunteer (e.g., through NGOs, religious organizations, educational organizations, community organizations). The facilitator discusses how to overcome barriers for volunteering in old age with the group members. Finally, group members are asked to reflect on whether they want to be a volunteer, the reasons for this decision, and to share their thoughts with other group members.

The third session starts with a review of the first two sessions. Then, the facilitator shows two five-minute video clips in which older volunteers are presented as positive role models. In the video, a woman and a man introduce themselves as being over 70 years old. The man has problems walking, while the woman did not receive formal education. Nevertheless, the role models actively volunteer in social areas by visiting older people, who live alone and by teaching older adults how to use computers in senior centers. The models express how they love their volunteer work, because it makes them feel needed and that giving pleasure to others renders them pleasure as well. In the video, the models speak of their experiences with volunteering. The models explain how they started their work as volunteers, how they planned to volunteer, and how they enjoyed volunteering immensely. After showing the video chips, the facilitator asks participants to discuss what they feel towards the role models and the most important benefits of being a volunteer from their perspectives. Then, participants are asked to think about the areas in which they would like to be active volunteers. The facilitator provides a decision aid by showing different age groups as recipients of volunteer support, such as children, adolescents, adults, and older adults as well as some specific vulnerable groups, such as migrants, families with financial difficulties, people with disabilities, and people with chronic medical conditions. The facilitator introduces some concrete examples of volunteer work, such as escorting others, doing household chores, acting as receptionist, measuring blood pressure, fundraising, visiting other people, designing posters, taking photos, and reading newspapers or storybooks to others. Different categories of volunteering, including clerical work, elderly care, child care, and organizing activities, are also presented along with different venues for volunteer activities, such as community centers, social service organizations, hospitals, religious groups, cultural organizations, animal shelters, and political parties. Volunteering opportunities at the elderly center are briefly mentioned by the staff of the senior center hosting the intervention group. A list of volunteering opportunities in each district of Hong Kong and the websites to search for volunteer opportunities are presented to the group. Finally, the facilitator discusses the amount of time that participants would like to spend on volunteering and warns them that excessive volunteering may be detrimental to older adults’ health. Participants are informed that they should identify their optimal level of volunteering per week and should not exceed this level, to prevent burnout. Before the end of this session, participants set goals related to volunteer activities: 1) the organization; 2) the target group; 3) the nature of the volunteer activities; 4) personal goals; 5) activities or services; and 6) frequency of participation.

In the fourth session, participants are asked to focus on past experiences using a biography worksheet. On this worksheet, participants are invited to look back on different periods within their own biography. They are asked: 1) to state the hobbies and volunteer engagements they participated in during the course of their lifes; 2) whether they participated in these activities regularly; 3) whether they enjoyed performing these activities back then; and 4) for how long they engaged in these activities. The lifespan is divided into childhood and youth (under 20 years), young adulthood (20–30 years old), and adulthood (40–60 years old). After these recollections, participants are asked if they had engaged in volunteer activities based on their hobbies and about volunteer engagements in their late adulthood. The facilitator discusses different personal goals related to engaging in volunteer activities (e.g., helping others, doing something meaningful, spending time with others, self-actualization, use of skills or experiences, improving the community, religious reasons, and acquiring new skills or knowledge) and participants are asked to discuss their personal goals to engage in volunteer activities.

Once participants have established their personal goals and plans to volunteer, the facilitator introduces the Proper-Effective-Practicable-Plannable-Test (PEPP-Test, similar to setting SMART goals). In this test, participants are asked to think about a volunteer engagement that would suit them, whether they would enjoy this kind of engagement, and whether this engagement matches their areas of interest for the first P (Proper). In the E-test (Effective) participants are asked whether they think the chosen volunteer activity is effective to achieve their personal goals. In the third P-test (Practical) participants are asked, whether the plans they made are practicable and feasible. In the last P-test (Plannable) participants are asked whether their formulated volunteering plan fits within their weekly schedules and is not too demanding.

Afterwards, participants are asked to complete if-then implementation intentions on a worksheet, to initiate action planning and use of cues [31]. In this worksheet participants are asked to come up with different situations, objects, or people to which they could link a certain volunteering activity, using if-then sentences. Examples include “If I had breakfast, then I will go to the agency for volunteer engagement”; “If I see my calendar, then I will be reminded to look for a volunteer engagement”; “If I meet my friends, I will ask them about their experiences with volunteer engagements.” Participants are engaged in a discussion about tricks they could use on themselves to render becoming active volunteers easier. Besides the tricks participants mention themselves, the facilitator introduces a diary to help them self-monitor their progress towards more volunteering. In this diary, participants are asked to record whether they engage in volunteer activities and, if they do, the nature of the volunteer activities, the amount of time they dedicate to volunteer activities, and how they feel after carrying out or not carrying out planned volunteer activities (negative or positive) in the coming 6 months. In addition, a six-month calendar is provided, in which participants can schedule their volunteer appointments.

By the end of the last session, participants are asked to form a self-support group. Each group receives the opportunity to meet again at the respective senior center to share their volunteering experiences, the benefits of participating in volunteer activities, and any barriers or difficulties preventing them from participating in volunteer activities. The facilitator also asks them to name three significant others, who could support them in implementing their plans related to volunteering, to prompt social support.

Staff members of the hosting senior center briefly introduce volunteering opportunities at their center to participants and provide them with the same list of volunteering opportunities in each district and information about websites to search for volunteer activities as in the intervention group. Apart from this brief and purely informational intervention content focusing on volunteering, the active control group receives the same behavior change techniques as the volunteering intervention group in four parallel interactive group sessions with similar length, but all material is adapted to increase physical activity. Physical activity is chosen as the target behavior for the active control group as it resembles volunteering in several aspects: 1) It has similarly positive consequences for mental as well as physical health and longevity among older adults if performed on a regular basis [32, 33]. 2) The recommended amount of physical activity is only reached by a minority of older adults in Hong Kong (only 29% of the adult population aged 65 and older engage in 150 min of physical activity per week as recommended by the World Health Organization; [34]). 3) And social-cognitive motivational and self-regulatory intervention techniques are well applicable to the initiation and maintenance of regular physical activity, ensuring that parallel intervention content can be set up for both group conditions [35]. The only difference in group session content is that no PEPP-test and no self-monitoring is performed in the physical activity group and that this group learns some mild stretching and strengthening exercises during the group sessions in addition to the social-cognitive intervention techniques to ensure that participants’ expectations of an exercise group are met and to keep attrition rates low.

The facilitator in the active control group is the same as in the intervention group. This facilitator is supported by an additional trainer for physical activity in the elderly, who leads the actual physical exercises in the active control group (stretching and low-impact cardio exercises). For a detailed description of the active control group as compared to the intervention group, see Additional file 1.

See Table 1 containing the SPIRIT information [36] for an overview of time-points and measures in this RCT.

Self-reported minutes of formal volunteering minutes per month is chosen as the primary outcome measure as it has shown to have a reliable link to positive health outcomes in previous research (as compared to further possible outcomes such as the frequency of volunteering, or the number of organizations; [37]). The definition of volunteering – formal volunteering for organizations – is explained to participants before they answer the questionnaires. Monthly volunteering minutes are assessed at baseline and 6 weeks, 3 months, and 6 months after the completion of the intervention. Two items validated by Ayalon [38] and based on the study by Warner et al. [30] are used 1) “During the past four weeks, on how many days per week did you do volunteer work?” and 2) “If you did volunteer work, how many minutes did one session last on average?”. Sessions per week and minutes per session are multiplied to derive the average minutes of volunteering per month.

In addition to this primary outcome, further information on volunteering is assessed: 3) the type of volunteer activity “If you did volunteer work, what type of volunteer work did you perform?”, 10 categories (i.e., recreational, manual labor, keeping company, domestic, educational, caring in hospices, sociocultural, administrative, social, and managerial) based on the German Survey on Volunteering is provided as answering options [39]; 4) benefits of volunteering for younger generations are assessed using the item “Did the volunteer work you did in the past four weeks benefit younger generations?” with “yes” and “no” as answering options; and 5) “For how many organizations did you serve as a volunteer during the past four weeks?” is assessed with an open format.

The hypothesized mediators and active ingredients of the social-cognitive intervention to increase volunteering derived from the HAPA model are perceived benefits of volunteering, self-efficacy, intentions, planning and self-monitoring.

Perceived benefits of volunteering are assessed with the Chinese version of the perceived benefits subscale developed by Warburton et al. [26]. The subscale consists of five items (e.g., feeling useful, helping those in need, being busy and active). These positive expectations are rated on seven-point bipolar scales ranging from 1 (extremely unlikely) to 7 (extremely likely).

Self-efficacy for volunteering is measured by Wang’s volunteering self-efficacy scale, which comprises three items [40]. An example item is “How much confidence do you have in your ability to participate in volunteer activities?”. All items are rated on a five-point scale from 1 (not at all) to 5 (an extreme amount).

Intention to volunteer is assessed with four items based on previous volunteer intention measures [40, 41]. An item example is “Do you agree that you have decided to participate in future volunteering?” Answers range from 1 (strongly disagree) to 5 (strongly agree).

In lack of specific scales to assess self-regulation for volunteering, planning and self-monitoring for volunteering are assessed by adapting scales from the physical activity domain developed by Sniehotta et al. [42]. An item example for one of the four action planning items is “I have made a detailed plan regarding when to do volunteer work”. The two items for self-monitoring are “During the last four weeks, I have constantly monitored myself whether I volunteer frequently according to my plans”; and “…watched carefully that I volunteer as often as I intend to”. Planning and self-monitoring items are rated on a six-point scale from 1 (not at all true) to 6 (exactly true).

To test for further motives and values that might change through the intervention, additional exploratory mediators are assessed at all measurement points in time.

The Chinese version of the Loyola Generative Concern Scale [43, 44] is used to measure generative concerns with 20 items, in which participants are asked to what degree items describe themselves on a four-point scale ranging from 0 (never) to 3 (very often). Example items are “I try to pass along the knowledge I have gained through my experience” and “I feel as though I have made a difference to many people”.

Sense of community is assessed using the 12-item Sense of Community Index [45], which has been translated into Chinese and validated in the local community [46]. Participants are asked to rate items on a four-point scale from 1 (strongly agree) to 4 (strongly disagree). Example items are “I think the street/housing estate that I am living in is a good place for me to live” and “Very few of my neighbors know me” (recoded).

The Volunteer Functions Inventory (VFI; [47]) is used in its Chinese version [48] to measure participants’ motives to volunteer (i.e., the expected outcomes of volunteering). Participants are asked to indicate how important or accurate each reason for volunteering is for them on a seven-point scale from 1 (not important or not accurate at all) to 7 (very important or very accurate). This scale comprises 30 items, measuring six volunteering motives with five items each: 1) values (providing opportunities to express values regarding humanitarian concerns for other vulnerable people and helping others); 2) understanding (providing opportunities to acquire knowledge, skills, and capacities through novel learning experiences); 3) career (providing opportunities to gain career-related benefits); 4) protection (providing opportunities to reduce feelings of guilt and to solve personal problems); 5) enhancement (providing opportunities to enhance one’s ego); and 6) social (providing opportunities to maintain or improve friendships and to engage in activities admired by peers or significant others).

Further motives for volunteering are assessed with the Pleasure and Pressure based Prosocial Motivation Scale [49]. This scale contains a 4-item Pleasure based Prosocial Motivation subscale (e.g., “Supporting other people makes me very happy.”) and a 4-item Pressure based Prosocial Motivation subscale (e.g., “I feel indebted to stand up for other people.”). Participants respond using a 5-point scale ranging from 1 (strongly disagree) to 5 (strongly agree).

The perceived costs of volunteering are assessed with 5 items from the perceived costs and benefits scale validated by Warburton et al. [26]. Perceived negative consequences of volunteering (e.g., being overcommitted, being taken for granted, being tied down) are rated on a 7-point bipolar scale ranging from 1 (extremely unlikely) to 7 (extremely likely).

Depressive symptoms are assessed with the Geriatric Depression Scale-Short Form [50, 51]. Participants are asked to tick 1 (no) or 0 (yes) for whether they have experienced any of 15 symptoms within the past 2 weeks (e.g., “Have you dropped many of your activities or interests?”, “Do you feel that your life is empty?”).

Meaning in Life is assessed with the Chinese version of Steger’s scale [52, 53]. Items read for example “I have discovered a satisfying life purpose” and “I understand my life’s meaning.” and are rated on a scale from 0 (absolutely not true) to 6 (absolutely true).

Levels of general self-efficacy are measured with the validated Chinese version of the General Self-Efficacy Scale [54]. Item examples are “I can always manage to solve difficult problems if I try hard enough” and “If someone opposes me, I can find the means and ways to get what I want.” Items are rated from 1 (not at all true) to 4 (exactly true).

Perceived Autonomy [55] is assessed with a 4-item scale (e.g., “I live by my own choices now that I am old”, “I make my own decisions and don’t need others to protect me.”) and answered on a 4-point scale from 1 (strongly disagree) to 4 (strongly agree).

Some physical activity related questions are integrated into the questionnaire to test whether taking up volunteering activities also effects physical activities (diminishes vigorous sports activities and increases light leisure activities for example) or physical activity related cognitions. The Baecke Physical Activity Questionnaire [56, 57] is used to assess self-reported physical activity including light leisure time activities (“Do you or did you practice sports or physical exercise within the past 12 months?” and “Have you changed your habits on leisure activities within the past 12 months? (e.g., playing mahjong, morning exercises, watching TV, playing chess, reading, etc.)” followed by open fields to indicate the type of activity and options to tick how many hours per week. Barriers, pros and cons for exercising are assessed with the The Chinese Barriers to Exercise Scale (CBE scale) and the Chinese Outcome Expectations for Exercise (COEE) scale [57, 58]. Example items for the CBE scale are “The venue for exercising is too far” and “I feel embarrassed doing exercise in public” and for the COEE scale examples “Doing exercise makes you feel tired easily” and “Doing exercise makes you sick easier”. With answering options from 1 (disagree) to 5 (agree). Exercise Self-efficacy is assessed with items adopted from Marcus et al. [57, 59], such as “On a scale from 0-10, how much would you conform, that you will develop a habit of exercising for one year (at least three times a week for a minimum of 20 minutes at a time)?” and “Are you confident that you can do the following actions?” (0–100% to indicate confidence in walking, running slowly, carrying heavy items, climbing up stairs, doing sit-ups).

Stages of change for physical activity are assessed with one item and 5 answering options based on the stages of change of the Transtheoretical Model by Prochaska and DiClemente [57, 60] “In a typical week, do you do regular physical exercise – meaning at least three times 20 minutes of physical exercise per week?”. Answers can be given on a visual analog scale from 10 to 0 with verbal anchors at 10 ("I do regular exercise and have maintained it for more than 6 months"), 8 ("I do regular exercise but only started for no more than 6 months"), 5 ("I exercise sometimes"), 2 ("I don’t exercise regularly, but I’m thinking about starting to exercise within 6 months") and 0 ("I do not exercise regularly, nor do I consider starting to exercise within the next 6 months").

To determine, whether the volunteering intervention works better for individuals with a specific mindset or socio-economic background, several moderator variables are assessed. Participants, who report to have volunteered within the past 4 weeks, are asked to rate further cognitions and emotions towards their volunteering activity: Satisfaction with volunteer work is assessed with one item [61] “Overall, how satisfied are you with volunteer work” rated from 1 (extremely dissatisfied) to 10 (extremely satisfied). The Chinese version of the Volunteer Satisfaction Index [62] consisting of 26 items (e.g., “I receive help when I need while volunteering.”, “The actual conditions of volunteer work are consistent with my expectations.”) and answered on a scale from 1 (very dissatisfying) to 7 (very satisfying). General volunteering enjoyment is assessed with one item by Okun et al. [63] “Overall, how much do you enjoy volunteer work?” rated from 1 (not at all) to 5 (very enjoyable). More differentiated emotional responses to volunteering are assessed with a rating scale from 1 (not at all) to 7 (extremely) for five positive (rewarding, exciting, interesting, enjoyable, fulfilling) and four negative emotions (emotionally draining, frustrating, disappointing, depressing [64]). To assess possible symptoms of volunteer burnout, the Chinese version of the Maslach Burnout Inventory Human Service Survey Scale is administered [41]. Volunteers are asked to rate 22 items on a 5-point scale from 1 (disagree completely) to 5 (agree completely), e.g. “I feel emotionally drained from my volunteer work.” and “I worry that my volunteer work makes me indifferent.”

All participants answer two questions on their level of religiousness. The two items are “I consider myself to be a spiritual person” [65], “I live according to religious principles” [66] and are answered from 1 (strongly disagree) to 4 (strongly agree).

All participants rate their attitudes towards older adults with the Chinese version of the Aging Semantic Differential [67, 68], that presents 10 bipolar semantic adjectives together with the question “Please indicate how you perceive older adults” (e.g., independent versus dependent, busy versus idle, secure versus insecure) to be rated on a 7-point scale between the two opposing poles.

All analyses are statistically controlled for the following covariates: age, sex, partner status (without versus with partner), education (high school degree versus no high school degree), subjective health (“In general, how would you rate your health today?” from 1 (very good) to 5 (very bad)), and number of chronic conditions (“Has a doctor ever diagnosed you with any of the following diseases?” followed by a list of 10 diseases, e.g. arthritis, heart disease, hypertension, and an open field for further chronic diseases) as these socio-demographic and health measures have repeatedly been shown to be associated with the amount of volunteering [9, 10]. Subjective health is assessed at each time point; the other covariates are assessed at baseline only.

Descriptive statistics, correlations of all variables, attrition analyses (T-tests comparing socio-demographic information as well as the level of expected outcome measures and mediators in participants, who drop out, to participants, who stay in the study until the last point of measurement) and randomization tests (T-tests comparing socio-demographic information as well as the level of expected outcomes and mediators in participants randomized to the intervention group to participants randomized to the active control group) are performed in SPSS 24.

The intervention effect on the primary outcome measure is first tested with repeated measures analyses of variance (ANOVA) to test for a time * group interaction over the six weeks, three months and six months follow-up. Based on the repeated measures ANOVAs, effect size calculations are performed using Eta².

For more refined analyses, latent change scores, modeling change in volunteering minutes from baseline to six weeks, three months, and six months is performed in Mplus 8 [69]. This latent change approach has several advantages over traditional variance analyses: It allows testing differences from baseline to all post-tests in a single model instead of multiple tests, has more power to detect treatment effects, is robust to non-normality, and provides information on individual variability and fit statistics [70, 71]. All models are statistically controlled for age, sex, marital status, education, subjective health, and number of chronic conditions. A conditional model with baseline as reference point is set up in accordance with Mun et al. [71]. The change scores are then regressed on group condition and on the above-mentioned control variables. If any of the control variables show significant interactions with the group condition on volunteering, subgroup analyses are performed to find possible moderators that increase or decrease the intervention effect. In the change models, effect sizes are calculated via R². Missing values are imputed via full information maximum likelihood estimation (FIML) in Mplus, as FIML makes use of all available data in model estimation [72]. To account for the cluster-randomization referring to participants clustered in groups and centers, analyses are conducted in a multilevel framework.

To test for possible mediators of the intervention effect, manifest path analyses with indirect effects are specified and tested with 95% bias-corrected confidence intervals resulting from 1000 bootstraps in Mplus 8. Respective mediators at six weeks and three months as well as total volunteering minutes per month at six months follow-up are regressed on their respective baseline scores, the above-mentioned control variables as well as on group condition to test for an indirect effect of group condition on volunteering via these mediators in separate models for each mediator.

The research team will publish one primary outcome article, describing the effect of the social-cognitive intervention on volunteering minutes per month. This article will include process analyses to elaborate on the active ingredients of the intervention material. Mediators named in the measurement section of this protocol are tested in the resulting publication. Another publication will focus on the secondary gains achieved by the intervention, testing effects on all secondary outcomes named in the measurement section of this protocol. Exploratory analyses are conducted to test for possible moderators of the intervention effect. If any of the covariates or moderators named in the measurement section of this protocol make a meaningful contribution to explaining the effects of the intervention on the primary or secondary outcomes, these analyses may result in further publications to inform future interventions on subgroups that are more or less receptive to the intervention.