Effectiveness of an ecological theory-driven health promotion program on improvement of retirement adjustment, psychological well-being, and coping strategies in Iranian retirees: a randomized controlled trial Authors: Elaheh Shoushtari-Moghaddam, Mohammad Hossein Kaveh, Masoud Karimi, Mohammad Javad Tarrahi, Esmail Soltani Published: 2026 ABSTRACT BackgroundRetirement is a major life transition that affects psychological well-being and coping. This study evaluated the effect of a health promotion program grounded in an ecological framework on retirement adjustment, psychological well-being, and coping strategies among Iranian retirees.MethodsA randomized controlled trial was conducted with 178 retirees from retirement centers affiliated with the Departments of Education and the Oil Industry in Fars province, southern Iran. Participants were systematically selected and randomly assigned to intervention (n = 90) and control (n = 88) groups using stratified block randomization. The intervention comprised a multicomponent program including blended learning (face-to-face sessions and mobile-based delivery via WhatsApp) and support from family and organizations. Educational strategies were guided by the PRECEDE-PROCEED model and Lazarus’ transactional model of stress and coping. Data were collected at baseline, three months, and six months post-intervention using the Retirement Adjustment Scale, Ryff Psychological Well-Being Scale, and Ways of Coping Questionnaire. Statistical analyses included repeated measures ANOVA, MANCOVA (adjusting for baseline scores), and independent samples t-tests.ResultsGroups did not differ significantly at baseline. Over time, the intervention group showed improvements in retirement adjustment, total psychological well-being, subjective well-being, psychological well-being dimension, ways of coping, and problem-focused coping (P < 0.001). Emotion-focused coping showed no significant change. Significant time-by-group interaction effects were observed for all these variables (P < 0.001).ConclusionAn ecological, theory-driven health promotion program improved well-being, coping skills, and adjustment in retirees. Multilevel interventions addressing individual, family, and organizational levels may support healthy aging and successful retirement.Trial registrationIranian Registry of Clinical Trials (IRCT) with a code of IRCT20180516039690N2, registered 16th September 2022.Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1186/s12889-026-27764-9. Background Retirement is a significant life transition that can profoundly impact an individual’s psychological well-being and overall quality of life [1]. While it marks the end of a professional career, it also presents opportunities for personal growth, leisure, and new social interactions [2]. However, for many retirees, this transition is accompanied by psychological challenges such as loss of identity, reduced social engagement, and uncertainty about the future [3]. The structure and purpose once provided by work can be difficult to replace, leading to emotional distress, social isolation, and a diminished sense of purpose [4]. The extent to which retirees successfully adjust to this phase of life largely depends on their ways of coping [5]. While some individuals adapt by engaging in meaningful activities, maintaining strong social connections, and cultivating a positive outlook, others experience heightened stress, loneliness, and emotional distress [6]. Ways of coping are crucial in shaping retirees’ psychological well-being and adaptation to post-work life [7]. These strategies are broadly categorized into problem-focused and emotion-focused approaches [8]. Problem-focused coping involves proactive efforts to manage challenges, such as setting new goals, participating in social or community activities, and seeking knowledge or support [9]. Studies have shown that retirees who engage in structured programs, such as volunteer work, educational workshops, and recreational activities, report higher life satisfaction and psychological resilience [10]. Emotion-focused coping, on the other hand, helps individuals regulate their emotional responses to stressors through techniques such as cognitive reframing, relaxation methods, and emotional support-seeking [11]. Research suggests that retirees who flexibly combine these ways of coping tend to experience better mental health outcomes compared to those who rely solely on one approach [12]. Most educational and empowerment interventions focus on content like skill development while lacking a strong theoretical foundation, limiting replicability and effectiveness. A comprehensive framework enhances shared understanding, facilitates comparative studies, and improves intervention outcomes. Using a guiding framework such as the PRECEDE-PROCEED model for planning interventions and a specific theory suited to the research problem, like Lazarus and Folkman’s Stress and Coping Model for promoting coping skills, enhances intervention effectiveness. The transactional model of stress and coping explains how individuals appraise and respond to stressors, highlighting the role of cognitive appraisal in shaping well-being [8]. Meanwhile, the PRECEDE-PROCEED model structures intervention design by addressing predisposing (e.g., retirees’ beliefs and psychological preparedness), enabling (e.g., social networks and access to mental health resources), and reinforcing factors (e.g., ongoing family and community support) [13]. Although extensive research has explored retirement adaptation and psychological well-being, several critical gaps remain. The effectiveness of coping interventions over the long term has not been sufficiently examined, as most studies focus on short-term improvements without assessing whether these benefits are sustained [14]. For example, Heaven et al. (2010) found that retirees who participated in organized programs initially reported improved well-being, but these effects diminished after a year due to a lack of continued engagement [15]. Despite its success in guiding behavioral change in other health-related fields, the PRECEDE-PROCEED model has limited application in retirement-focused interventions [16]. Additionally, most research on ways of coping has been conducted in Western contexts, leaving a gap in understanding how cultural values and social norms influence retirement adaptation in non-Western societies [17]. Studies indicate that in Western cultures, retirees often emphasize individualistic ways of coping such as self-improvement and independent decision-making [18], whereas in collectivist societies, family and community support play a greater role in managing retirement-related stress [19]. Given these variations, culturally tailored interventions are needed to address the specific coping needs of diverse populations. Furthermore, while problem-focused coping has been widely studied, emotion-focused ways of coping, such as mindfulness, cognitive reframing, and acceptance-based approaches, have received less attention in retirement adaptation programs [9]. Given that many retirement-related stressors, such as loss of professional identity or social roles, cannot be directly altered, effective emotional regulation is essential for maintaining psychological well-being [20]. Studies on mindfulness-based stress reduction and positive psychology interventions have demonstrated their effectiveness in reducing anxiety and depression among older adults [21, 22], yet these approaches are rarely incorporated into well-planned retirement programs. This study aims to investigate the role of ways of coping in shaping retirees’ psychological well-being, using Lazarus and Folkman’s stress and coping model and the PRECEDE-PROCEED model as guiding frameworks. By identifying the most effective ways of coping, developing structured intervention programs based on the PRECEDE-PROCEED model, exploring cultural variations in coping mechanisms, assessing the effectiveness of digital interventions, and examining the long-term impact of ways of coping, this research seeks to contribute to the development of evidence-based policies and interventions that enhance retirees' mental health and facilitate successful transitions into Retirement and old age. Hypotheses Null Hypothesis (H₀) There is no difference between the intervention and control groups in changes in retirement adjustment, psychological well-being (total, subjective, and dimension scores), or ways of coping (total, problem-focused, and emotion-focused) after adjusting for baseline values. Alternative Hypothesis (H₁) The intervention group will show greater improvements than the control group in retirement adjustment, psychological well-being (total, subjective, and dimension scores), and ways of coping (total, problem-focused, and emotion-focused) after adjusting for baseline values. Materials and methods Study design This was a two-arm parallel randomized controlled educational intervention study using a pre-test, post-test design. The intervention was developed using the PRECEDE-PROCEED model, which includes eight phases. Lazarus’ transactional model of stress and coping is embedded within this framework to address stress appraisal and coping processes. This integrated approach guided the design of the intervention across individual, family, and organizational levels, ensuring alignment with the theoretical constructs of both models. The detailed study protocol has been previously published [23]. Study population and sampling The study population consisted of retirees registered at retirement centers affiliated with the Departments of Education and the Oil Industry in Fars province, southern Iran. These two large public-sector retirement systems maintain updated registries, enabling the development of structured sampling frames. A total of 178 retirees were enrolled through systematic random sampling from approximately 4,000 retirees in the Oil Industry and 5,000 retirees in the Department of Education. The sampling interval was calculated by dividing the total number of retirees in each sector by the required sample size, yielding intervals of 31 and 25, respectively. To ensure ethical recruitment and avoid undue influence, initial contact was not made directly by the research team. Instead, a designated administrative staff member at each retirement center served as a gatekeeper solely to disseminate study information and identify individuals meeting the predefined inclusion criteria. Gatekeepers informed eligible retirees about the study via telephone or mobile-based communication platforms and referred interested individuals to the research team. Importantly, gatekeepers had no role in random sequence generation, allocation procedures, eligibility confirmation beyond administrative criteria, or outcome assessment, thereby minimizing potential selection bias. After referral, the research team contacted interested individuals by telephone to explain study objectives, procedures, potential risks and benefits, and to confirm eligibility. Written informed consent was obtained before baseline assessment. If a selected retiree declined participation or was ineligible, the next or previous individual on the systematic sampling list was approached to preserve sampling integrity. During recruitment and intervention phases, attrition occurred. To compensate and maintain statistical power, additional eligible retirees were approached using the same inclusion criteria through text messages, WhatsApp notifications, and announcements at retirement centers. These supplementary efforts followed the same structured eligibility verification and enrollment procedures as the initial phase. Randomization was conducted after completion of baseline assessments. The allocation sequence was generated by an independent researcher not involved in recruitment, enrollment, intervention delivery, or outcome assessment. A computer-based stratified block randomization procedure was used, with stratification by organizational affiliation to ensure balanced group representation. The sequence was generated using specialized randomization software prior to implementation. Allocation concealment was maintained by separating sequence generation, enrollment, and assignment roles. The independent researcher prepared the allocation list, and group assignments were communicated to the intervention team only after baseline measurements were completed and participants formally enrolled. This process was designed to minimize selection bias. Due to the educational and behavioral nature of the intervention, blinding of participants and facilitators was not feasible, as participants were aware of the content and format of sessions received. However, outcome assessors were blinded to group allocation to reduce measurement bias. As shown in the CONSORT flow diagram (Fig. 1), each group included 66 participants from the Department of Education. Additionally, the intervention group included 24 participants from the Fars Oil Industry, and the control group included 22 participants from the same sector, resulting in final group sizes of 90 and 88 in the intervention and control arms, respectively.Fig. 1Study consort diagram Study consort diagram No restrictions were imposed based on gender, ethnicity, or socioeconomic status. All retirees meeting the predefined inclusion criteria were eligible to participate, reducing systematic selection bias within the sampled organizational contexts. Sample size calculation The required sample size was calculated using the formula: \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$n=\frac{{\left({Z}_{1-\raisebox{1ex}{$\propto $}\!\left/ \!\raisebox{-1ex}{$2$}\right.} +{Z}_{1-\beta } \right)}^{2}({S}_{1}^{2} +{S}_{2}^{2})}{{d}^{2}}$$\end{document}. Assuming α = 0.05, β = 0.1, Z₁ = 1.96, Z₂ = 1.28, S₁ = 15, S₂ = 16, and d = 8.85, a minimum of 66 participants per group was needed, accounting for a potential dropout rate of 20%. These values were based on a previous study by Kazazi et al. [24] on quality-of-life score changes pre- and post-intervention. Inclusion and exclusion criteria Participants were eligible if they agreed to participate and signed an informed consent form, had been officially retired for at least one year based on organizational records or Pension Center approval, had completed at least primary school education, did not hold Full-time Paid work, owned a smartphone, and actively used social networks such as WhatsApp or Telegram. Individuals were excluded if they were unwilling to participate after initial contact, did not meet the educational or technological access requirements, were employed full-time, or had medical conditions such as cancer, stroke, Parkinson’s disease, or severe psychological disorders that could interfere with their ability to participate in the intervention. Intervention: ecological health promotion program The intervention was developed based on the PRECEDE–PROCEED model and the integrated transactional model of stress and coping. Consistent with the ecological framework guiding the study, the program addressed determinants of retirement adaptation at three interconnected levels: individual, family, and organizational. The ecological approach emphasizes addressing health behavior determinants across multiple levels of influence, including intrapersonal, interpersonal, and institutional contexts [25, 26]. To ensure intervention fidelity and consistency across sites and facilitators, a comprehensive intervention protocol was developed before implementation. This protocol specified session objectives, theoretical constructs, educational content, instructional strategies, sequencing, duration, and required materials for each session. All facilitators participated in preparatory coordination meetings in which the protocol was reviewed in detail, and consensus was achieved regarding standardized delivery procedures. Throughout the intervention period, adherence to the predefined curriculum was supported through structured session plans and regular coordination meetings among members of the research and facilitation teams. Continuous communication was maintained to address procedural issues and ensure alignment with the theoretical framework and educational design. These procedures were implemented to minimize variability in content delivery and to maintain fidelity to the intervention model across participants and over time. At the individual level, retirees assigned to the intervention group participated in a blended educational program consisting of nine weekly 90-min face-to-face sessions supplemented by a mobile-based educational and communication component delivered via WhatsApp. The selection of educational topics was informed by scientific literature on health during retirement [27–29], and was guided by constructs derived from Phase 3 of the PRECEDE–PROCEED model and the transactional model of stress and coping [8, 30]. The curriculum addressed biological, psychological, and social transitions associated with retirement and their effects on health, the importance of self-care for maintaining quality of life during retirement, and core domains of self-care, including nutrition, physical activity, stress management, coping skills, spiritual well-being, planning, and social support. Face-to-face sessions were delivered using interactive and participatory teaching methods, including structured discussions, experience sharing, and cooperative learning techniques such as buzz groups [31] and the jigsaw method [32]. Peer-led instruction was also incorporated [33]. In the peer-led sessions, two retirees with relevant expertise facilitated educational components related to promoting physical activity and overcoming cognitive distortions. These components were designed to enhance social modeling, reinforce self-efficacy, and strengthen in-group social support mechanisms. The mobile-based component reinforced in-person content through the structured dissemination of texts, short audio and video materials, and motivational messages. Participants were encouraged to ask questions and share ideas through the platform, thereby extending engagement beyond face-to-face sessions. A detailed description of the educational program is provided in Appendix A. At the family level, one primary family member of each retiree, typically the spouse, participated in a structured 100-min in-person training session. This session introduced the objectives and theoretical basis of the program, emphasized the importance of health for quality of life during retirement and aging, and highlighted the supportive role of family members in promoting adaptive behaviors and effective communication. Family participants were encouraged to review the educational materials provided to retirees during both in-person and digital sessions to reinforce program content within the home environment. At the organizational level, participatory strategies were implemented to facilitate institutional support for retiree well-being. Before intervention delivery, meetings were conducted with organizational representatives to secure administrative cooperation and clarify institutional support mechanisms. During the implementation phase, collaborative discussions involving managers and retirees were facilitated to identify feasible health-supportive initiatives within each setting. These efforts resulted in several supportive actions, including attendance of some managers at selected sessions, facilitation of peer-led educational activities utilizing retirees’ internal capacities, inclusion of retirement-related educational materials on the institutional website, installation of a suggestion mechanism to collect retirees’ feedback, and organization of recreational and physical activity events. These environmental strategies were designed to strengthen contextual support for behavioral and psychosocial change. Participants in the control group attended four educational sessions unrelated to the theoretical constructs or primary outcomes of the intervention. These sessions addressed home safety, medication safety, food safety, and Alzheimer’s disease awareness. To ensure ethical considerations were maintained, following completion of follow-up assessments, the control group received a summary of the intervention materials along with printed educational resources. Control participants also attended a recreational camp similar in format to that provided to the intervention group, without exposure to the coping skills or retirement-adjustment curriculum. Data collection Research data were collected in person at retirement centers over three stages: before, three months, and six months after the intervention, utilizing four questionnaires as follows. The Retirement Adjustment Scale (RAS), originally reported by Wells in the Healthy Retirement Project retiree population, consists of 13 items scored on a five-point Likert scale ranging from 1 (strongly agree) to 5 (strongly disagree), with total scores ranging from 13 to 65. Higher scores indicate more positive adjustment to retirement. Sample items include “I will adjusted to the changes” and “Retirement has been better than I expected.” The scale has demonstrated high internal consistency in previous studies, with Cronbach’s alpha coefficients ranging from 0.81 to 0.89 [34–36]. In the present study, the Persian version of the questionnaire was piloted among approximately 30 retirees to assess reliability. Cronbach’s alpha was 0.83, and the intraclass correlation coefficient (ICC) over a two-week interval was 0.91, indicating satisfactory internal consistency and test–retest reliability. Lazarus and folkman Ways of Coping Questionnaire (WCQ). This questionnaire contains 66 items across eight subscales (confrontive coping, distancing, self-control, seeking social support, accepting responsibility, escape–avoidance, planful problem solving, and positive reappraisal). The instrument measures two overarching coping styles: problem-focused coping and emotion-focused coping. Items are rated on a four-point Likert scale ranging from 0 (not used) to 3 (used a great deal). Higher scores in each domain indicate greater reliance on that coping strategy. Example items include “I made a plan of action and followed it” (problem-focused coping) and “I wished that the situation would go away or somehow be over with” (emotion-focused coping). The Persian version has been previously validated in Iran, with reported Cronbach’s alpha coefficients for subscales ranging from 0.60 to 0.84 [37]. In the present study, overall internal consistency was high, with a Cronbach’s alpha of 0.90. The 18-item version of the Ryff Psychological Well-Being Scale, developed by Ryff and Keyes. This scale assesses six dimensions: self-acceptance, positive relations with others, autonomy, environmental mastery, purpose in life, and personal growth. Items are rated on a six-point Likert scale ranging from strongly disagree to strongly agree, yielding total scores between 18 and 108. For analytical purposes in the present study, subscales were grouped into two broader dimensions: psychological well-being (autonomy, purpose in life, personal growth, and positive relations) and subjective well-being (environmental mastery and self-acceptance). Example items include “I judge myself by what I think is important, not by what others think” (autonomy) and “I am quite good at managing the responsibilities of my daily life” (environmental mastery). Previous research has reported Cronbach’s alpha coefficients of 0.75 and 0.74 for the respective well-being dimensions [38]. The Persian version has demonstrated acceptable reliability in Iran, with a reported Cronbach’s alpha of 0.71 [39]. In the current study, Cronbach’s alpha was 0.76. The demographic questionnaire collected detailed information on age (in years), gender (male or female), marital status (married or single), organizational affiliation (Education Department or Oil Industry), level of education (primary, secondary, diploma, associate, bachelor's, master's or higher), and perceived income adequacy (less than expenses, equal to expenses, or more than expenses). Statistical analysis Data were analyzed using SPSS version 26 (IBM Corp., Armonk, NY, USA). Descriptive statistics (means, standard deviations, frequencies, and percentages) summarized baseline characteristics. Group equivalence at baseline was assessed using independent samples t-tests for continuous variables and chi-square tests for categorical variables. Longitudinal changes across baseline, 3 months, and 6 months were examined using repeated measures ANOVA to evaluate the main effects of time, group, and the time × group interaction, which represented the primary test of intervention effectiveness. To compare groups at specific follow-up time points (3 and 6 months), multivariate analysis of covariance (MANCOVA) was conducted, adjusting for baseline values of the respective outcomes. This approach enabled simultaneous analysis of correlated dependent variables while controlling for baseline differences. Assumptions for each method were examined before analysis. For repeated measures ANOVA, independence, normality, and sphericity were assessed, with Greenhouse–Geisser corrections applied where necessary. For MANCOVA, assumptions of independence, multivariate normality, homogeneity of variance–covariance matrices, absence of multicollinearity, and homogeneity of regression slopes were evaluated and met. Effect sizes were reported using partial eta squared (η2p), with 0.01, 0.06, and 0.14 indicating small, medium, and large effects, respectively [40]. Statistical significance was set at p < 0.05 (two-tailed). Results A total of 178 retirees participated in the study, with 90 assigned to the intervention group and 88 to the control group. Among the participants, 74.2% were affiliated with the Department of Education and 25.8% with the Oil Industry. Demographic profiles of the retirees in the intervention and control groups are summarized in Table 1. No statistically significant differences were observed between the two groups in terms of organizational affiliation, gender, education level, marital status, or income status, as determined by the Chi-square test (P > 0.05). The mean age of participants in the intervention and control groups was 61.4 and 61.3 years, respectively, with no significant difference (P = 0.96, t = 0.05, df = 176).Table 1Demographic profiles of retirees in the two intervention and control groupsCategoriesIntervention Group(n = 90)Control Group(n = 88)Total(n = 178)P-value*Organizational Affiliation Oil Industry24 (26.7%)22 (25.0%)46 (25.8%)0.86 Education Department66 (73.3%)66 (75.0%)132 (74.2%)Gender Male30 (33.3%)28 (31.8%)58 (32.6%)0.87 Female60 (66.7%)60 (68.2%)120 (67.4%)Education Level High School Diploma18 (20.0%)20 (22.7%)38 (21.3%)0.68 Associate Degree22 (24.4%)27 (30.7%)49 (27.5%) Bachelor's Degree41 (45.6%)33 (37.5%)74 (41.6%) Master's and above9 (10.0%)8 (9.1%)17 (9.6%)Marital Status Single8 (8.9%)8 (9.1%)16 (9.0%)0.62 Married67 (74.4%)66 (75.0%)133 (74.7%) Divorced3 (3.3%)6 (6.8%)9 (5.1%) Widowed12 (13.3%)8 (9.1%)20 (11.2%)Income vs Expenses Income less than expenses54 (60.0%)46 (52.3%)100 (56.2%)0.49 Income equals expenses27 (30.0%)29 (33.0%)56 (31.5%) Income more than expenses9(10.0%)13(14.8%)22(12.4%)*Chi-square test Table 1Demographic profiles of retirees in the two intervention and control groupsCategoriesIntervention Group(n = 90)Control Group(n = 88)Total(n = 178)P-value*Organizational Affiliation Oil Industry24 (26.7%)22 (25.0%)46 (25.8%)0.86 Education Department66 (73.3%)66 (75.0%)132 (74.2%)Gender Male30 (33.3%)28 (31.8%)58 (32.6%)0.87 Female60 (66.7%)60 (68.2%)120 (67.4%)Education Level High School Diploma18 (20.0%)20 (22.7%)38 (21.3%)0.68 Associate Degree22 (24.4%)27 (30.7%)49 (27.5%) Bachelor's Degree41 (45.6%)33 (37.5%)74 (41.6%) Master's and above9 (10.0%)8 (9.1%)17 (9.6%)Marital Status Single8 (8.9%)8 (9.1%)16 (9.0%)0.62 Married67 (74.4%)66 (75.0%)133 (74.7%) Divorced3 (3.3%)6 (6.8%)9 (5.1%) Widowed12 (13.3%)8 (9.1%)20 (11.2%)Income vs Expenses Income less than expenses54 (60.0%)46 (52.3%)100 (56.2%)0.49 Income equals expenses27 (30.0%)29 (33.0%)56 (31.5%) Income more than expenses9(10.0%)13(14.8%)22(12.4%)*Chi-square test Demographic profiles of retirees in the two intervention and control groups *Chi-square test Ways of coping Repeated measures ANOVA showed a significant time effect in the intervention group (P < 0.001; η2 = 0.14). Post hoc comparisons indicated significant differences between baseline and post-test (P < 0.001) and between baseline and six-month follow-up (P = 0.001), while the difference between post-test and follow-up was not significant (P = 0.098). No significant time effect was observed in the control group (P = 0.253). The time × group interaction was significant (P = 0.001; η2 = 0.06). Baseline differences between groups were not significant (P = 0.368). MANCOVA analyses demonstrated significant adjusted between-group differences at three months (P = 0.005; η2 = 0.05) and six months (P = 0.001; η2 = 0.06) (Table 2).Table 2Comparison of mean scores of variables by time, group, and time-group interactionvariablesgroupPre-InterventionMean ± SDCohen’s d(P Value)a3 months post-interventionMean ± SDη2(P Value)b6 months post-interventionMean ± SDη2(P Value)bTimeη2(P Value)cInterventionη2(P Value)cInteractionη2(P Value)cWays of CopingIntervention97.01 ± 22.30105.17 ± 20.15103.91 ± 21.240.14(< 0.001)0.007(0.262)0.06(< 0.001) control99.71 ± 17.2599.68 ± 18.2197.70 ± 17.450.253(0.02)0.14 (0.368)0.05 (0.005)0.06 (0.001)problem-focused copingIntervention52.64 ± 12.1059.82 ± 11.7458.54 ± 13.390.24 (< 0.001)0.02 (0.049)0.10 (< 0.001) control54.51 ± 10.8454.70 ± 11.6752.38 ± 11.050.037 (0.04)0.16 (0.280)0.09 (< 0.001)0.14 (< 0.001)Emotion-focused copingIntervention44.63 ± 11.8145.35 ± 10.9444.84 ± 10.490.005 (0.531)0.000 (0.531)0.003 (0.967) control45.21 ± 8.1744.97 ± 8.3544.80 ± 8.180.717 (0.003)0.06 (0.703)0.03 (0.505)0.001 (0.658)Retirement AdjustmentIntervention42.37 ± 8.9945.20 ± 7.4145.87 ± 8.160.20 (< 0.001)0.03 (0.018)0.06 (< 0.001) control42.04 ± 5.2042.37 ± 6.1542.11 ± 5.730.746 (0.003)0.04 (0.764)0.04 (0.006)0.08 (< 0.001)Psychological Well-Being (Total)Intervention83.65 ± 8.4390.20 ± 7.1293.30 ± 8.470.44 (< 0.001)0.18 (< 0.001)0.31 (< 0.001) control83.37 ± 10.4580.00 ± 9.4581.17 ± 9.80 < 0.001 (0.18)0.03 (0.844)0.46 (< 0.001)0.39 (< 0.001)Subjective Well-Being (Dimension)Intervention 28.86 ± 4.0430.35 ± 3.2830.88 ± 3.550.18 (< 0.001)0.08 (< 0.001)0.09 (< 0.001) control27.97 ± 5.2527.31 ± 4.7927.73 ± 4.900.074 (0.03)0.19 (0.207)0.19 (< 0.001)0.16 (< 0.001)Psychological well-Being (Dimension)Intervention54.75 ± 5.6759.84 ± 4.9162.41 ± 5.830.47 (< 0.001)0.17 (< 0.001)0.37 (< 0.001) control55.39 ± 7.4852.68 ± 6.7553.43 ± 6.98 < 0.001 (0.26)0.09 (0.519)0.49 (< 0.001)0.47 (< 0.001)aIndependent Samples T-TestbMANCOVAcRepeated measure ANOVA Table 2Comparison of mean scores of variables by time, group, and time-group interactionvariablesgroupPre-InterventionMean ± SDCohen’s d(P Value)a3 months post-interventionMean ± SDη2(P Value)b6 months post-interventionMean ± SDη2(P Value)bTimeη2(P Value)cInterventionη2(P Value)cInteractionη2(P Value)cWays of CopingIntervention97.01 ± 22.30105.17 ± 20.15103.91 ± 21.240.14(< 0.001)0.007(0.262)0.06(< 0.001) control99.71 ± 17.2599.68 ± 18.2197.70 ± 17.450.253(0.02)0.14 (0.368)0.05 (0.005)0.06 (0.001)problem-focused copingIntervention52.64 ± 12.1059.82 ± 11.7458.54 ± 13.390.24 (< 0.001)0.02 (0.049)0.10 (< 0.001) control54.51 ± 10.8454.70 ± 11.6752.38 ± 11.050.037 (0.04)0.16 (0.280)0.09 (< 0.001)0.14 (< 0.001)Emotion-focused copingIntervention44.63 ± 11.8145.35 ± 10.9444.84 ± 10.490.005 (0.531)0.000 (0.531)0.003 (0.967) control45.21 ± 8.1744.97 ± 8.3544.80 ± 8.180.717 (0.003)0.06 (0.703)0.03 (0.505)0.001 (0.658)Retirement AdjustmentIntervention42.37 ± 8.9945.20 ± 7.4145.87 ± 8.160.20 (< 0.001)0.03 (0.018)0.06 (< 0.001) control42.04 ± 5.2042.37 ± 6.1542.11 ± 5.730.746 (0.003)0.04 (0.764)0.04 (0.006)0.08 (< 0.001)Psychological Well-Being (Total)Intervention83.65 ± 8.4390.20 ± 7.1293.30 ± 8.470.44 (< 0.001)0.18 (< 0.001)0.31 (< 0.001) control83.37 ± 10.4580.00 ± 9.4581.17 ± 9.80 < 0.001 (0.18)0.03 (0.844)0.46 (< 0.001)0.39 (< 0.001)Subjective Well-Being (Dimension)Intervention 28.86 ± 4.0430.35 ± 3.2830.88 ± 3.550.18 (< 0.001)0.08 (< 0.001)0.09 (< 0.001) control27.97 ± 5.2527.31 ± 4.7927.73 ± 4.900.074 (0.03)0.19 (0.207)0.19 (< 0.001)0.16 (< 0.001)Psychological well-Being (Dimension)Intervention54.75 ± 5.6759.84 ± 4.9162.41 ± 5.830.47 (< 0.001)0.17 (< 0.001)0.37 (< 0.001) control55.39 ± 7.4852.68 ± 6.7553.43 ± 6.98 < 0.001 (0.26)0.09 (0.519)0.49 (< 0.001)0.47 (< 0.001)aIndependent Samples T-TestbMANCOVAcRepeated measure ANOVA Comparison of mean scores of variables by time, group, and time-group interaction 0.14 (< 0.001) 0.007 (0.262) 0.06 (< 0.001) 0.253 (0.02) aIndependent Samples T-Test bMANCOVA cRepeated measure ANOVA Problem-focused coping A significant time effect was observed in the intervention group (P < 0.001; η2 = 0.24). Post hoc comparisons showed significant differences between baseline and both post-test and follow-up (P < 0.001). The difference between post-test and follow-up was marginal (P = 0.052). The control group also showed a significant time effect (P = 0.04). The time × group interaction was significant (P < 0.001; η2 = 0.10). Baseline differences were not significant (P = 0.280). MANCOVA indicated significant adjusted between-group differences at three months (P < 0.001; η2 = 0.09) and six months (P < 0.001; η2 = 0.14) (Table 2). Emotion-focused coping No significant time effect was observed in the intervention group (P = 0.531; η2 = 0.000). No significant time effect was observed in the control group (P = 0.717). The time × group interaction was not significant (P = 0.967; η2 = 0.003). No significant between-group differences were found at baseline (P = 0.703), three months (P = 0.505; η2 = 0.03), or six months (P = 0.658; η2 = 0.001) (Table 2). Retirement adjustment A significant time effect was observed in the intervention group (P < 0.001; η2 = 0.20). Post hoc analyses indicated significant differences between baseline and both post-test and follow-up (P < 0.001), with no significant difference between post-test and follow-up (P = 0.133). No significant time effect was observed in the control group (P = 0.746). The time × group interaction was significant (P < 0.001; η2 = 0.06). Baseline differences were not significant (P = 0.764). MANCOVA demonstrated significant adjusted between-group differences at three months (P = 0.006; η2 = 0.04) and six months (P < 0.001; η2 = 0.08) (Table 2). Total psychological well-being A significant time effect was observed in the intervention group (P < 0.001; η2 = 0.44). Post hoc comparisons showed significant differences across time points (P < 0.001).). No significant time effect was observed in the control group (P = 0.18). The time × group interaction was significant (P < 0.001; η2 = 0.31). Baseline differences were not significant (P = 0.844). MANCOVA indicated significant adjusted differences between groups at three months (P < 0.001; η2 = 0.46) and six months (P < 0.001; η2 = 0.39) (Table 2). Subjective well-being A significant time effect was found in the intervention group (P < 0.001; η2 = 0.18). Post hoc testing showed differences between baseline and both post-test and follow-up (P < 0.001), with no significant difference between post-test and follow-up (P = 0.070). No significant time effect was observed in the control group (P = 0.074). The time × group interaction was significant (P < 0.001; η2 = 0.09). Baseline comparisons were not significant (P = 0.207). MANCOVA confirmed significant adjusted group differences at three months (P < 0.001; η2 = 0.19) and six months (P < 0.001; η2 = 0.16) (Table 2). Psychological well-being dimension Significant time effects were observed in the intervention group (P < 0.001; η2 = 0.47). Post hoc analyses showed differences across all time points (P < 0.001). No significant time effect was observed in the control group (P = 0.26). The time × group interaction was significant (P < 0.001; η2 = 0.37). Baseline differences were not significant (P = 0.519). MANCOVA showed significant adjusted between-group differences at three months (P < 0.001; η2 = 0.49) and six months (P < 0.001; η2 = 0.47) (Table 2). After applying Bonferroni correction for multiple pairwise comparisons (adjusted α = 0.0167), all statistically significant between-group differences and time × group interaction effects for ways of coping, problem-focused coping, retirement adjustment, total psychological well-being, and both well-being dimensions remained significant. In contrast, emotion-focused coping remained non-significant across all time points (Table 2). Discussion The present study evaluated the effect of a health promotion program, grounded in an ecological framework, on ways of coping, retirement adjustment, and psychological well-being among Iranian retirees. The findings demonstrated significant improvements in overall coping, problem-focused coping, retirement adjustment, psychological well-being, and subjective well-being among participants in the intervention group compared to the control group over six months. Within the PRECEDE–PROCEED framework, these outcomes can be interpreted as reflecting changes in Phase 3 determinants, including predisposing factors (knowledge, attitudes, self-efficacy), enabling factors (skills acquisition and access to structured resources), and reinforcing factors (family and organizational support). The significant increase in ways of coping, particularly problem-focused coping, is noteworthy. From a theoretical perspective, coping skills training directly targets predisposing factors such as cognitive appraisal and perceived self-efficacy, while simultaneously strengthening enabling factors through structured skill development. The improvement in problem-focused coping strategies after intervention and at follow-up in the present study, similar to the findings of Koolhaas et al., points to the long-term benefits of interventions that encourage problem-solving and active planning [41]. Loureiro et al. also showed that developing active problem-solving strategies helped to increase self-efficacy, which is a key factor for success in retirement [42]. Jurek reported similar findings, highlighting the effectiveness of coping skills training in enhancing resilience among retirees [43]. Our program's structured sessions on coping skills, mindfulness, and stress management appear to have strengthened participants' abilities to actively manage retirement-related challenges by operationalizing enabling resources within the ecological model. Gawande et al.'s study also showed that interventions designed to increase emotional regulation, through mindfulness and cognitive techniques, led to greater self-control and emotional stability in participants, findings that mirror the results of the present study [44]. Other study findings show that programs focused on self-management and stress control lead to more positive emotional outcomes in individuals [45]. However, unlike problem-focused strategies, emotion-focused coping did not change significantly in the present study. This partially contrasts with Lazarus, who posited that emotion-focused coping can be modified through psychological training [8]. From a theoretical standpoint, emotion-focused coping strategies may represent more stable, trait-like patterns embedded within long-standing cognitive–emotional schemas, particularly in later adulthood. While the intervention targeted enabling and reinforcing factors effectively, deeper modification of emotion-regulation processes may require more intensive, individualized, or longer-duration psychological interventions than those feasible within a structured health promotion program. The stability of emotion-focused coping across both groups suggests that the intervention exerted specificity toward skill-based coping mechanisms rather than broader dispositional emotional regulation processes. The improvement in retirement adjustment is also consistent with prior research and aligns closely with PRECEDE–PROCEED constructs. Wang and Shultz emphasized that proactive preparation, psychological resource building, and environmental support are crucial for successful retirement adjustment [46]. Within Phase 3 of the model, these components correspond to strengthening predisposing factors such as attitudes toward retirement, enhancing enabling factors through practical planning skills, and reinforcing adaptive behaviors via social and organizational support. Our intervention’s ecological framework—engaging family members, retirement organizations, and peer networks—operationalized reinforcing factors at multiple levels, thereby creating a supportive environment conducive to smoother adjustment. Similarly, the systematic review by Cassanet et al. confirmed that psychosocial interventions addressing both individual and contextual factors significantly enhance retirement outcomes [47]. In their review, Heaven et al. found that interventions involving meaningful social roles or retirement planning improved well-being and retirement adjustment by enhancing social support, life satisfaction, purpose in life, and planning [15]. This is relevant to the findings of the present study in that the trend in changes in retirement adjustment was significantly dependent on group membership, suggesting that targeted interventions, as well as the combined use of shared experiences, peer motivation, and structured goal setting, facilitate group-based adaptation processes. In older age, the experience of negative changes, including decreased physical function [48] and loss of loved ones [49], becomes more common. Therefore, promoting psychological adjustment is of great importance, and for this reason, psychological approaches are considered effective in promoting adjustment and psychological well-being and preventing psychological problems in the elderly [50]. The observed improvement in retirement adjustment over time in this study may therefore reflect the synergistic effect of addressing predisposing, enabling, and reinforcing determinants within an ecological structure. The observed improvements in psychological well-being are consistent with previous research emphasizing the positive impact of retirement preparation programs on mental health. Heaven et al. demonstrated that retirees who participated in structured educational programs reported higher levels of psychological well-being compared to those who did not receive such interventions [15]. Similarly, Kim and Moen found that structured psychosocial interventions, especially those based on an ecological model, were associated with improved psychological outcomes during retirement transitions [18]. A study by Seckman highlighted that older adults who engaged in programs fostering social connections and purpose experienced better psychological outcomes during the COVID-19 pandemic [51]. These consistent findings suggest that strengthening predisposing factors such as knowledge and perceived control, while simultaneously enhancing enabling and reinforcing supports, can buffer the psychological challenges associated with retirement. As study findings show, people in retirement can strive for personal development through new learning opportunities and reflective practices [52]. In terms of subjective well-being, our results showed a statistically significant improvement in the intervention group. This finding aligns with Kubicek et al., who reported that active engagement in health-promoting and social activities post-retirement plays a critical role in maintaining high levels of subjective well-being [53]. Henning et al. also emphasized that retirees who continue to engage in meaningful activities experience sustained subjective well-being over time [54]. McKnight and Kashdan's study showed that interventions focusing on purpose in life increased both subjective well-being and life satisfaction in older adults [55]. Within the PRECEDE–PROCEED framework, these improvements may be interpreted as outcomes of strengthened reinforcing mechanisms through peer interaction and family involvement, as well as enhanced enabling resources through structured skill-building. The success of the present intervention in enhancing subjective well-being may stem from its blended learning format, social interaction elements, and active engagement strategies, such as peer-led sessions and problem-solving exercises, which fostered a sense of purpose and connectedness. The ecological design of the present program appears to be a major strength, consistent with recommendations from public health and aging research that emphasize multilevel interventions [56, 57]. By simultaneously targeting predisposing, enabling, and reinforcing determinants across individual, family, and organizational levels, the program addressed the complex and multifaceted nature of retirement adaptation while maintaining theoretical coherence with the PRECEDE–PROCEED model. Strengths and limitations This study had several strengths. The use of an ecological approach that integrated individual, family, and organizational support systems enhanced the intervention’s effectiveness. A blended learning format combining face-to-face sessions and digital tools increased accessibility and engagement. The randomized controlled design, along with the use of validated instruments adapted for the Iranian context, strengthened the study’s internal validity and reliability. Additionally, the three- and six-month follow-ups provided valuable insights into the sustainability of intervention effects over time. However, several limitations should be acknowledged. The sample was limited to retirees from two occupational sectors in Fars province, which may restrict generalizability to other professional, socioeconomic, or cultural contexts, despite evidence that retirement adjustment challenges are broadly shared across occupations [58]. The six-month follow-up period does not permit conclusions about long-term maintenance of outcomes. Outcomes were assessed exclusively through self-report instruments, which may introduce recall or social desirability bias, although such effects were likely non-differential across randomized groups. A per-protocol analytic approach was used due to practical attrition-related constraints in this older population; although baseline comparisons suggested no systematic differences between completers and non-completers, residual attrition bias cannot be entirely excluded. Finally, individual characteristics such as personality traits, prior coping styles, and baseline mental health status were not controlled and may have influenced responsiveness to the intervention. Future research should replicate these findings in more diverse occupational and sociocultural contexts, incorporate longer follow-up periods, and consider broader methodological approaches to further strengthen external validity. Conclusion The results highlight the potential of a multi-level, theory-based intervention to improve retirees' psychological well-being and their ability to adapt. As retirement becomes more complex, health promotion models that combine ecological views with cognitive-behavioral techniques are crucial. This method not only supports mental health but also helps retirees to face aging with confidence and purpose. Policymakers and health educators should consider incorporating such interventions into national aging strategies. Emphasis should be placed on building digital capacity for blended learning models and promoting intergenerational support mechanisms within families and communities. The findings will provide practical insights for mental health professionals, policymakers, and retirement program developers to design effective coping interventions tailored to retirees' diverse needs. Supplementary Information Supplementary Material 1. Supplementary Material 1.