This pilot randomized controlled trial was conducted to evaluate the feasibility, acceptability, and preliminary effectiveness of an SMS text messaging intervention focused on lifestyle self-management for weight management among employed men younger than 40 years.

To strengthen the theoretical basis of the intervention, this study incorporated internationally standardized BCTs into an SMS intervention in Japan.

Participants were randomly assigned to the intervention or control group. Owing to the nature of the intervention, neither the participants nor the principal investigator could be blinded to group allocation.

The inclusion criteria were as follows: male workers aged 18‐39 years with a BMI of ≥25 kg/m², possession of a smartphone or mobile phone and ability to use SMS text messaging services and access the internet, ability to read and write Japanese and understand the study content, and participation in workplace health checkups. The exclusion criteria were as follows: work involving night shifts and current medical treatment for a pre-existing condition that prevents participation in healthy eating or physical activity for weight management. Workers engaged in night-shift work were excluded to reduce heterogeneity related to work schedules and associated differences in lifestyle patterns, such as eating and sleeping patterns, in this pilot study.

Occupational physicians and occupational health nurses recruited participants from their workplaces using written or email invitations and workplace postings, as appropriate to each workplace. The participating workplaces were identified through the professional networks of the principal investigator and a coinvestigator. The researchers did not directly recruit employees. The recruitment period extended from December 2024 to April 2025 because the planned number of participants was not reached within the initial recruitment period. Participants who met the inclusion criteria received a QR code on the study website and provided informed consent.

After obtaining consent, the occupational health nurses measured the participants’ weights and waist circumferences at the participating workplaces. Body weight was measured using a TANITA BC-315 body composition scale (Tanita Corporation). Waist circumference was measured using a seca 201 ergonomic circumference measuring tape (seca GmbH & Co KG). Height data for the BMI calculations were obtained from workplace health examination records. The participants completed a questionnaire, including demographic characteristics, using Google Forms. Follow-up assessments were completed in August 2025.

Participants were randomly assigned to the intervention or control group in a 1:1 ratio using permuted block randomization with a block size of 2. Participants from the same workplace could be assigned to either the intervention or control group. Randomization was implemented using a Microsoft Excel–based tool incorporating the RAND function, which was prepared and provided by a statistician who was not involved in participant recruitment. After informed consent had been obtained and participant enrollment was completed, the principal investigator entered each participant into the randomization tool, which generated the group assignment. Thus, group allocation remained unknown until after participant enrollment. Owing to the nature of the intervention, neither the participants nor the principal investigator could be blinded to group allocation. However, the occupational health nurses who performed the outcome measurements were not informed of participants’ group allocation. The principal investigator conducted the statistical analyses and was aware of group allocation.

No formal a priori power calculation for intervention effectiveness was performed because the study was planned as a pilot randomized controlled trial. As recorded in the University Hospital Medical Information Network Clinical Trials Registry (UMIN-CTR; trial ID UMIN000054203), the initial target sample size was 90 participants, with 45 participants allocated to each group. The target was calculated to obtain approximately 30 evaluable participants per group after allowing for an anticipated attrition rate of 30% (30/0.70=42.9), which was rounded up to 45 participants per group. The anticipated attrition rate was based on previous mHealth weight-management studies that anticipated attrition rates of 20% and approximately 30%, respectively [30,31].

Owing to recruitment constraints during the planned study period, 35 participants were ultimately randomized, including 17 in the intervention group and 18 in the control group. A sensitivity analysis based on the achieved sample size indicated that, with a 2-sided α of .05 and 80% power, the minimum detectable standardized effect size for the binary primary outcome was approximately Cohen h=0.95. Although no participants were lost to follow-up, the randomized sample remained below the planned target; therefore, findings regarding intervention effectiveness should be interpreted as exploratory.

Statistical analyses were performed using IBM SPSS Statistics (version 31.0.1.0). Analyses were conducted according to the intention-to-treat (ITT) principle and included all randomized participants. When missing values were present, a per-protocol set (PPS) analysis was also used.

For the intervention and control groups, baseline age (years), height (cm), BMI (kg/m²), and waist circumference (cm) were calculated as mean (SD), and intergroup comparisons were performed using a 2-tailed independent samples t test.

Differences in the primary outcome, defined as the proportion of participants achieving a weight loss of 3% or more, were compared between the intervention and control groups using a χ² test. The Fisher exact test was applied when the expected frequency of any cell was ≤5. Odds ratios and 95% CIs were calculated to estimate effect sizes. Because binary classification of the achievement rate may result in a loss of information, weight change (postintervention minus baseline) was calculated as a supplementary analysis, and group differences were evaluated using a 2-tailed independent samples t test (assuming equal variances).

For secondary outcomes, 2-tailed paired t tests were performed to evaluate the changes in BMI and waist circumference from baseline to postintervention within each group. For the WEL, NGSE-J, and behavior change stages regarding lifestyle improvement efforts, the Mann-Whitney U test was used to compare the intervention and control groups regarding the magnitude of change from baseline to postintervention, as well as the postintervention measurements. As the WEL, NGSE-J, and behavior change stages regarding lifestyle improvement efforts did not follow a normal distribution, descriptive statistics were presented as medians and interquartile ranges.

Evaluations of satisfaction, utilization, and willingness to recommend SMS text messages (sent only to the intervention group), as well as the frequency, timing, and duration of delivery, were presented using descriptive statistics. SMS delivery was summarized using frequency counts and delivery rates, and URL click-through rates were summarized as percentages. Descriptive statistics were used to calculate the costs associated with SMS text messaging during the delivery period.

Because this pilot study was designed to assess feasibility, acceptability, and preliminary effectiveness rather than to provide definitive evidence of effectiveness, we calculated the effect size (odds ratio) and 95% CI for the primary outcome, in addition to hypothesis testing, to evaluate the magnitude and precision of the intervention effect. For all statistical analyses, the significance level was set to 5% (2-tailed).

This study was approved by the Medical Research Ethics Review Committee of Mejiro University in March 2023 (approval number 23-I-054). The trial was prospectively registered in the UMIN-CTR (trial ID UMIN000054203). We adhered to the CONSORT-EHEALTH (Consolidated Standards of Reporting Trials of Electronic and Mobile Health Applications and online Telehealth) guidelines (see Checklist 1) [32] when reporting the study findings.

Participants were informed of the purpose and methods of the study, the voluntary nature of participation, and the protection of their personal information and were assured they would suffer no disadvantages if they declined to participate or withdrew from the study. Digital informed consent was obtained using the landing page service provided by CM.com, a provider of SMS messaging services. All collected data were anonymized before analysis to ensure participants’ confidentiality and privacy. All participants who completed the study were provided with a gift card worth JP¥ 2000 (approximately US $12.50) as a token of appreciation.

A total of 43 participants were screened in the study. Seven participants who did not meet the inclusion criteria and one who withdrew from the study were excluded. In all, 35 participants were included in the study and randomly assigned to the intervention group (n=17) or the control group (n=18). During the follow-up period, all 17 participants in the intervention group completed the 12-week intervention, and all 18 participants in the control group completed the trial. No participants dropped out in either group. The demographic and work-related characteristics of the participants are presented in Table 1.

Baseline data were analyzed for all randomized participants. For psychological and behavioral measures, because 4 participants in the control group had missing responses, the baseline values were imputed for missing data in the ITT analysis. In the PPS analysis, only participants with complete responses were included (Figure 1). Physical, psychological, and behavioral indicators at baseline are presented in Table 2.

A greater proportion of participants in the intervention group than in the control group achieved at least 3% weight loss, but the between-group difference was not statistically significant. The supplementary analysis of weight change likewise showed no significant between-group difference (Table 3).

At the 3-month follow-up, process evaluation data were obtained from all 17 participants in the intervention group, with no missing data. The mean (SD) ratings for satisfaction, utilization, and willingness to recommend were 3.5 (SD 1.1), 2.8 (SD 1.2), and 3.0 (SD 1.2), respectively (Table S5 in Multimedia Appendix 3). Delivery timing received less favorable ratings than those of delivery frequency and duration (Table S6 in Multimedia Appendix 3).

This pilot randomized controlled trial assessed the feasibility, acceptability, and preliminary effectiveness of an SMS text messaging intervention targeting lifestyle self-management among employed men younger than 40 years with obesity. The intervention was feasible in terms of SMS delivery and participant retention and was generally acceptable to participants. Regarding preliminary effectiveness, no statistically significant between-group differences were observed in the primary outcome or in the psychological and behavioral secondary outcomes. Waist circumference decreased significantly from baseline within the intervention group, whereas no significant within-group change was observed in BMI.

No statistically significant between-group difference was observed in the primary outcome. Although the point estimate of the odds ratio favored the intervention group, the 95% CI was wide and included 1, indicating considerable uncertainty in the effect estimate.

Given the pilot design and limited sample size, the study was not designed to provide a definitive estimate of intervention effectiveness. Therefore, these findings regarding effectiveness should be interpreted as preliminary and require confirmation in an adequately powered trial.

In a similar worksite-based RCT involving employed men with obesity and a mean age of approximately 41 years, no significant between-group difference in weight loss was observed with the addition of SMS text messaging [33]. In that study, the intervention group received tailored SMS text messages in addition to face-to-face education, monthly weight measurements, and brief counseling, whereas in the present study, SMS text messaging was used as the primary mode of intervention. These findings suggest that further refinement of SMS content and delivery strategies may be needed to support weight loss among employed men with obesity.

The reduction in waist circumference from baseline in the intervention group, despite the absence of a significant change in BMI, may reflect an early change that was not captured by body weight–based measures. Ross et al [34] reported that abdominal and visceral fat can decrease even in the absence of weight loss.

No between-group differences were observed in the psychological and behavioral outcomes, including WEL, NGSE-J, and stages of behavior change. Previous studies have suggested that SMS text messages can function as prompts and reminders to encourage health behaviors [12,35]; however, in the present study, these messages may have been insufficient to produce measurable changes in self-efficacy or stages of behavior change. The one-way nature of the intervention may also have limited opportunities for active participant engagement. Future studies should explore strategies to increase engagement, such as personalizing the text message content or introducing two-way communication.

The SMS text messages in this study incorporated 10 BCTs to support lifestyle behavior change. However, these BCTs were delivered through standardized, one-way SMS text messages, with limited opportunities to tailor the techniques to participants’ individual goals, behaviors, or progress. Similarly, in the Game of Stones trial, theory-based BCTs were incorporated into text messages for men with obesity, but text messaging alone did not result in significantly greater weight loss than the control condition [22]. Further research should examine how BCT delivery through SMS can be optimized to support behavior change.

SMS text messaging showed a high delivery rate throughout the intervention period. A 12-month intervention study using SMS text messaging to support weight loss in individuals with obesity reported improvements in weight and waist circumference, even with relatively simple, regular SMS delivery alone [36]. Furthermore, while being a relatively simple method that does not involve interaction or face-to-face contact, it allows for the continuous provision of encouragement to improve lifestyle habits in the context of daily life [37]. As SMS text messaging does not require the installation of a dedicated app and can be received on mobile phones used in daily life, it is unlikely to impose a new burden on participants and has the potential to be easily used as a means of continuous support. These characteristics could enhance the feasibility of implementing workplace health guidance for employed men.

In contrast, the click-through rate for URLs included in text messages was relatively limited. Because the intervention involved one-way delivery via SMS text messaging and focused primarily on viewing health information, the participants may not have necessarily clicked on URLs. While the click-through rate is an indicator of the participants’ interest and engagement with the intervention content, it is not a direct indicator of the intervention’s effectiveness or behavior change. Given that the content of the text messages sent via SMS itself functions as support for behavior change, viewing the messages may have contributed to behavior changes. Follow-up was complete in the present study. The low-burden format, which does not require participants to reply or enter data, may have supported participant retention. This simplicity may have increased the acceptability of the intervention by busy working adults. Qualitative studies targeting young working adults have reported that time constraints associated with work act as barriers to practicing healthy eating and physical activity [38]. Therefore, SMS-based support, which is unlikely to impose a time burden, may be an intervention method that is easily accepted by working adults. Previous studies targeting men with obesity have reported the acceptability of interventions using story-based SMS [39,40]. Although messages in the present study were delivered between 12:00 PM and 1:00 PM so that participants could view them during their lunch break, this timing may not have suited all participants. Future interventions could therefore consider more flexible delivery times.

This study has some limitations. First, because this was a pilot study with a limited sample size, it was underpowered to detect significant differences between groups, limiting the certainty of the findings, particularly for the psychological and behavioral indicators. In addition, no adjustment for multiple testing was applied to the exploratory secondary outcomes; therefore, individual P values should be interpreted cautiously because of the type I error. Second, the Japanese version of the WEL used in this study has not yet been published, so these measurement results should be interpreted cautiously as exploratory findings. Third, blinding was not feasible, as participants in the intervention group were inherently aware they were receiving text messages. Furthermore, because participants from the same workplace could be assigned to either group and no specific measures were implemented to prevent the sharing of intervention messages, potential contamination between the intervention and control groups cannot be excluded. Complete follow-up should also be interpreted with caution, as the gift card provided after completion of the 3-month follow-up assessment may have contributed to participant retention.

Regarding the intervention itself, the reliance on one-way SMS text messages without individualized feedback or interactive support limited the overall intensity of the behavior-change support. Furthermore, while we measured SMS delivery and link click-throughs, the exact read rates of the messages and post-click on-page behavior could not be fully confirmed. Although the messages were grounded in BCTs, the study design did not allow us to isolate or verify the mechanisms of action of individual BCTs.

Finally, the short 12-week intervention period leaves the long-term maintenance of behavioral and weight changes unclear, and focusing exclusively on employed men younger than 40 years limits the generalizability of these findings. Future research should build on these exploratory results through adequately powered RCTs, longer follow-up periods, and study designs capable of clarifying the specific pathways and effects of individual BCTs.

This pilot RCT supported the feasibility and acceptability of an SMS text messaging intervention among employed men younger than 40 years with obesity, while evidence of its effectiveness remains preliminary. This approach may therefore be useful for providing health guidance to this population. Larger studies are needed to evaluate effectiveness and optimize intervention content and delivery.