Table 3. Summary of the included studies (n = 24) grouped by main targeted health outcome.

Author, Country, Publication Year	Study Details	Intervention details	Measured Outcomes	Main findings
A. Lifestyle/ Chronic disease risk (n = 7)
Liu Z et al [22]., China, 2015	Type: Clustered RCT Participant: 589 staff of hospital health management centre Company size: Not provided	Length:12-month intervention Goal/Focus: Reduce overall CVD (cardiovascular disease) risk Mode of Delivery Intervention Group (IG): receive mobile phone-based lifestyle intervention including an individualised electronic prescription, follow-up 5 to 8 min phone calls and text messages targeting reducing CVD risk during the 12-month intervention. Control Group (CG): receive usual medical examination without follow-up calls and text messages. No theory applied.	Primary outcome: Change in 10-year CVD risk between baseline and follow-up point at 12-months, Change in components of risk score Secondary outcome: Diastolic blood pressure (BP), triglycerides (TG), high-density lipoprotein (HDL), and low-density lipoprotein (LDL), fasting plasma glucose (FPG), waist hip ratio (WHR)	Significant difference between baseline and 12-month in CG: 1.Increased mean CVD risk, Systolic BP (p<0.001), DBP mean value (p< 0.001) 2.Decreased HDL and LDL (p<0.05), overall Diastolic BP, FPG and WHR (p<0.001) Significant different between baseline and 12-months in IG: Decreased 10-year CVD risk (Systolic BP, TC, BMI) (p<0.05), DBP mean value (p<0.001), HDL and LDL (p<0.05) Mobile phone-based intervention may therefore be a potential solution for reducing CVD risk in China.
Jorvand R at al. [30], Iran, 2020	Type: Quasi-experiment Participants: 114 healthcare workers employed in two cities Company Size: 262 healthcare workers from two cities assessed for eligibility	Length: 2-week intervention with 6-month follow-up Goal/Focus: Effect of Health Belief Model (HBM)-based education on exercise Mode of delivery IG: Workers from one network received a Telegram-based intervention and supervised exercises, receiving education packages every 2 weeks and exercise reminder messaging, discussion and interlocution and their own exercise pictures sharing. CG: workers from the other network received uncontrolled individual exercise and self-reported. Underlying theory: HBM	Primary outcome: 1.HBM (perceived susceptibility, severity, benefits, barriers, and self-efficacy) 2.Exercise (daily and weekly in minutes) Secondary outcome: blood biochemical markers	Significant difference of HBM constructs mean score at pre and post intervention in IG: 1.Increased perceived severity (p = 0.000), perceived benefits (p = 0.010) and self-efficacy (p = 0.024) 2.Decreased perceived susceptibility (p = 0.018): directly related to increased preventive behaviours (doing exercise for CVDs) Significant improvements of exercise in IG: Daily exercise at post intervention (p = 0.001) Weekly exercise at pre and post intervention (p = 0.001) Educational interventions based-on Telegram messenger using HBM can improve exercise level.
Ramachandran A et al. [33], India, 2013	Type: Prospective RCT Participant: 537 male employees in public and private industrial units Company size: Not provided, 8741 for eligibility screening	Length: 24-month intervention with follow-up Goal/Focus: Effect of phone messages on lifestyle change (PA and diet) to reduce Type 2 diabetes (T2D) Mode of delivery IG: Frequent mobile phone messages about healthy lifestyle, (the benefits, cues to start, ways to avoid relapse and remain motivated in physical activity (PA) and healthy dietary habits) CG: standard lifestyle modification advice at baseline only Underlying theory: Transtheoretical Model of Behavioural Change	Primary: Incidence of T2D Secondary: body mass index (BMI), waist circumference, systolic and diastolic BP, lipid profile, total dietary energy intake, PA score, acceptability of mobile phone messaging assessed by questionnaire	50 (18%) men in the IG developed T2D over the 2 years compared with 73 (27%) CG (absolute risk reduction 9%) The intervention reduced the incidence of T2D during the study (β –0·447). The number needed to treat to prevent one case of T2D was 11 (95% CI 6–55). Mobile phone messaging is an effective and acceptable method to deliver advice and support towards lifestyle modification to prevent T2D in men at high risk.
Nurgul K et al. [36], Turkey, 2015	Type: Descriptive and quasi experimental study Participant: 30 Female staff from Sakarya University Company Size: Sakarya University (44 staff voluntary participation)	Length: 3-month intervention without follow-up Goal/Focus: Effect of web-based education on health knowledge and behaviour Mode of delivery IG: 3-month web-supported health training material in ppt or audio-visual (3 modules: nutrition and health, PA, damages of smoking and stress management), available 7 days a week and 24 hours a day. No CG or theory applied.	Primary outcome: 1.Knowledge on health promotion: assessed by multiple choice questionnaire (MCQ) Individual’s health behaviours: assessed via 52-item health promotion lifestyle profile (HPLSP) with 6 sub-dimensions (Health Responsibility, PA, Nutrition, Self-actualization, Interpersonal support, stress management)	Significant difference (p<0.05) between pre and post intervention 1.HPLSP total points (p = 0.001) 2.HPLSP sub-scale total points (p = 0.001) 3.MCQ (p<0.001) Web-based health education had a positive effect on healthy lifestyle behaviours of women staff working at Sakarya University and on their knowledge of health protection.
Rubinstein A et al. [38], Argentina, Guatemala, and Peru, 2016	Type: Parallel RCT Participant: 637 adults with prehypertension from workplaces, health care, and community centres Company size: Not provided, 2630 for eligibility screening	Length: 12-month intervention with 6-month follow-up Goal/Focus: Effect of mHealth on cardiometabolic profile in prehypertension people Mode of delivery IG: Received monthly motivational counselling calls and weekly personalised messages about diet quality and PA. CG: Received usual care Underlying theory: Transtheoretical Model of Behavioural Change and the Health Belief Model	Primary outcome: Change in Systolic BP (mmHg) and Diastolic BP (mmHg) Secondary outcome: Change in weight (kg), BMI (kg/m2), WC (cm), PA (metabolic equivalent of task (METs)/ min per week), daily intake of fruits and vegetables (F&V), high fat and sugar foods	Significant changes in highly engaged participants (received ≥75% of counselling calls in IG): Body weight (-4∙85kg), WC (-3∙31cm), Daily intake of F&V (+0.66), Daily intake of high-sodium foods (-0.42), Daily intake of high fats food (-1.52) Significant difference between IG and CG by country: 1.Body weight in Peru (IG): -1.24kg 2.Daily F&V intake in Peru (IG): +0.64 3.Daily F&V intake in Guatemala (IG): +0.04 The mHealth-based intervention was associated with a small reduction in bodyweight and some dietary habits. A dose-response effect signalling potential opportunities for larger effects from similar interventions in low-resource settings was seen.
Martinez C et al. [39], Bolivia, Guatemala and Paraguay, 2018	Type: Pre-post study Participant: 202 hospital workers from three organisations Company size: 1450 workers in three organization	Length: 6-month intervention with follow-up Goal/Focus: Effect of online training in smoking behaviour Mode of delivery IG: received an adapted version of a 5A’s (Ask, Advise, Assess, Assist and Arrange follow-up) training program developed by the online platform e-oncologia based on in-person courses. No CG or theory applied	Primary outcome: Cognitive and behavioural factors relating to smoking Secondary outcome: Self-reported performance level according to 5A’s, demographics characteristic and questions suggested by experts to explore behavioural factors	Significant Increase in performance of each of the 5A components (p<0.001) Significant improvement at post training 1. Performance score of 5A’s (p<0.001) 2. Cognitive, behavioural, and organisational factors affecting 5A’s   • Five identified barriers (p<0.001): self-reported preparedness, drug preparedness, competency in assisting smokers to quit, using additional resources, and having positive experience.   • Opportunities with score ≥7 (p<0.001): motivation to help patients to quit, importance of smoking cessation in job, seeking frequently for patients. Online education on smoking cessation is feasible and effective in improving smoking cessation interventions in these countries.
Joseph-Shehu EM et al. [42], Nigeria, 2019	Type: Pre-post study Participant: 22 university staff in Nigeria Company Size: 1349 staff	Length: 12-week intervention with follow-up Goal/Focus: Effect of information and communication technology on health promotion Mode of delivery IG: Adopted a nurse-client interactive Android phone app, Tertiary Staff Health Promotion App to access health promotion information, to monitor health status, increase PA, minimise sitting hours, with reminder of activities to improve health and quality of life. CG: N/A Underlying theory: Health promotion model	Primary outcome: Health promoting lifestyle behaviour Health status (BP, BMI, WHR, and fasting blood sugar)	Significant difference between pre- and post-intervention: Increased nutrition score (p = 0.0001), PA score (p = 0.0001), health responsibility score (p = 0.0001), stress management score (p = 0.001), Interpersonal relation subscales score (p = 0.009) Decreased BMI (p = 0.038) and diastolic BP (p = 0.04) The health-promoting lifestyle behaviour and health status of workers and other population groups showed improvements through information and communication technology.
B. Weight management (n = 5)
He C et al. [23], China, 2017	Type: Cohort study Participant: 15,310 employees from 134 government agencies and enterprises Company Size: Not provided, 15818 for eligibility screening	Length: 6-month intervention with 6-month follow-up Goal/Focus: Effect of WeChat-based multicomponent program on weight loss IG: Social media-promoted intervention. Participants willing to use the research team’s official WeChat account were enrolled in a WeChat group that provided feedback on weight, diet, and exercise weekly, microvideos and popular science knowledge on weight loss, community area for communication. CG: Participants not willing to use official WeChat account given routine publicity such as slogan “take the stairs and lose weight” on weight loss No theory applied.	Primary outcome: Weight loss: height, weight, waist circumferences before and after intervention Demographic characteristics: gender, age, educational level, and telephone number online registered with WeChat account	Weight loss: IG (2.09±3.43kg) > CG (1.78±2.96kg) Effect of WeChat on weight loss (Assessed with propensity method, p<0.05): 1.Males in IG (active or inactive) had higher probability of maintaining weight with 1-2kg or > 2kg weight loss than CG (0-1kg) 2.Active participants in WeChat groups were more likely to lose weight. The weight loss intervention campaign based on an official WeChat account focused on an occupation-based population in Shunyi District was more effective for males than females.
Yu Y et al. [24], China, 2018	Type: RCT Participant: 802 employees from institutions or enterprises from 4 areas, 44.9% overweight & obese, 49.6% normal weight and 0.5% underweight Company Size: Not provided, 904 for eligibility screening	Length: 3-month intervention Goal/Focus: Effect of pedometer-based walking and diet guidance on weight management Mode of delivery IG: Receive self-monitored intervention trial (exercise prescription and dietary guidance), synchronise pedometer exercise data to the Internet-based Health System Centre daily (at least weekly) No CG or theory applied	Primary: Changes in body weight (kg) or BMI, waist circumference (cm), and BP (mmHg). Secondary: Changes in lifestyle behaviour (scores), body fat percentage (%), fasting blood glucose/ fasting serum glucose (mmol/L), and serum lipid (mmol/L)	Normal weight participants: 1. Weight decreased 0.7% (p < 0.01). 2. Body fat percentage decreased 2.5% (p < 0.01). 3. BP and FSG decreased significantly (p<0.05) Underweight participants: 1. Weight gain of 1.0% (p< 0.05), Overweight participants: 1. 68.2% (208/305) experienced weight loss, with an average reduction of 3.5%, with 20.2% (42/208) of them achieving weight loss 5%. 2. BP and FSG decreased significantly (p<0.05) The incidence of hypertension was significantly lower and lifestyle behaviour significantly improved (p < 0.05)
Abdi J et al. [28], Iran, 2015	Type: Three-arm RCT Participant: 435 governmental employees with overweight or obesity (BMI>25kgm/m2) Company Size: Not provided,1200 for eligibility screening	Length: 6-month intervention with 3-month follow-up and maintenance Goal/Focus: Effect of communication technologies and social-cognitive based education on weight loss Mode of delivery Two IG received lifestyle program and general brochures: IG1 (Web-assisted group): educational content provided on a website. IG2 (Telephone-assisted group): educational content provided through SMS every two weeks CG: only receive general brochures about lifestyle and overweight Underlying theory: Social cognitive theory (SCT)	Primary: anthropometric measures include weight (kg), waist circumference (cm) and blood pressure (mmHg) Secondary: SCT measures	The lifestyle intervention resulted in a weight loss of: -1.08 kg in IG1, -1.92kg in IG2 IG1: mean scores of the constructs of self-efficacy (P = 0.001) and outcome expectancies (P = 0.020) increased. IG2: mean scores of the constructs of self-efficacy (P = 0.001), environment (P = 0.001), outcome expectations (P = 0.040), and outcome expectancies (P = 0.001) increased. A significant difference (P = 0.03) was observed in weight loss among the groups over the course of time.
Limaye T at al. [34], India, 2017	Type: RCT Participant: 265 young Indians of IT industry with normoglycemia and at high risk of developing diabetes. Company Size: Not provided, 437 for eligibility screening	Length: 1-year intervention Goal/Focus: Effect of technology-based lifestyle program on reducing T2D risk IG: Receive lifestyle modification in Informative technology (LMIT) program 1. Mobile phone messages and emails with infographics which contain lifestyle modification information provided. No message was repeated. 2. Additional support through a website and a closed Facebook group. CG: no intervention received Underlying theory: goal setting	Primary outcome: prevalence of overweight/obesity (BMI ≥ 25 kg/m2) Secondary outcomes: Change in weight, waist circumference, blood pressure, glucose, lipid, lifestyle choices, diabetes awareness score, acceptability, and cost-effectiveness of the intervention	Significant weight loss in IG (p<0.05): Overweight/obese participants decreased from 104 (78.2%) to 96 (72.2%) Significant weight gain in CG (p<0.05): Overweight/obese participants increased from 101 (76.5%) to 110 (83.3%) The number needed to treat/prevent one case of overweight/obesity in 1 year was 9. A virtual assistance-based lifestyle intervention was effective, cost-effective and acceptable in reducing risk factors for diabetes in young employees in the IT industry and is potentially scalable.
Beleigoli A et al. [41], Brazil, 2020	Type: Three-arm parallel RCT Participant: 1,298 overweight and obese students and staff of a Brazilian university Company Size: Not provided, 3745 for eligibility screening	Length: 24-week intervention Goal/Focus: Effect of personalized web-based coaching on weight loss with overweight and obese people Mode of delivery Platform-only group (IG1): received reminder emails to report weight and habits at 12 and 24 weeks after baseline trial and get access to 24-week web-based weight loss program with personalised computer-delivered feedback. Platform and coaching group (IG2): received 24-week web-based weight loss program with 12-week personalised dietician-delivered feedback. CG (waitlist): received a non-personalized minimal intervention based on dietary and PA recommendations delivered through downloadable e-booklet and four 5-min videos. Underlying theory: Behaviour Change Wheel	Primary outcome: changes in weight and BMI Secondary outcome: 1.Changes in dietary intake: assessed by the daily F&V portions, weekly consumption of sweetened beverages and ultra-processed foods 2.Changes in PA: Moderate and vigorous PA was assessed by the Brief PA Assessment Questionnaire.	After 24-week intervention: 1.Primary outcome (Weight & BMI change)   • CG: Weight = -0.66kg, BMI = -0.24kg   • IG1: Weight = -1.08kg, BMI = -0.38kg   • IG2: Weight = -1.57kg, BMI = -0.56kg   • Significant overall weight loss (p = 0.001): IG1 (83/420, 19.8%) and IG2 (64/408, 15.7%) > CG (61/270, 13.0%) 2.Secondary outcome:   • Significant increase in F&V consumption (p = 0.001): IG1 and IG2 > CG   • Significant reduction in ultra-processed food consumption (p = 0.005): IG1 and IG2 > CG   • Significant increase in sweetened beverage consumption (p = 0.02): IG1> IG2
C. Physical activity (PA) (n = 4)
Blake H et al. [25], China, 2019	Type: Two-arm clustered RCT Participant: 282 employees of IT private sector organizations in Beijing and Gaungzhou Company size: 690 employees from two sites	Length:12-week intervention without follow-up Goal/Focus: Effect of digital video-based exercise on PA and work performance Mode of delivery IG (Guangzhou): “Move-It” digital video-based worksite exercise intervention 1. Move-It website: Six 10-min Qigong exercise video clips, twice/day on working day 2. Reminder messages: The Move-it desktop icon popped up at the same time, twice/day 3. Exercise adherence info: Daily exercise logs were collected through Move-it icon CG (Beijing, wait-list): received the same intervention after the intervention period Underlying theory: Behaviour Change Techniques (BCTs)	PA level, weekday sitting hours and work performance	Significant differences at post intervention: 1.Increased PA:   • IG (5.80 hrs/week, p = 0.04)   • CG (7.41 hrs/week, p = 0.00) 2.Work performance:   • Increased in CG (+0.69 units, p = 0.01)   • Reduced in IG (−0.03 units, p = 0.78) 3.Increased sitting hours: I  • G (+10.34 hrs/week, p = 0.00)   • CG (+5.68 hrs/week, p = 0.00)   • Between groups (−4.66 h/w, p<0.01) The intervention did not result in greater changes in the IG than in CG. Many participants perceived the Qigong exercises positively and reported positive benefits on physical and mental health including muscle relaxation, stress reduction and improved working mood.
Gu M et al. [26], China, 2020	Type: Quasi-experiment with self-controlled design Participant: 262 workers from 17 worksites Company Size: Not provided, 398 employees participated in study baseline	Length: 100-day intervention with follow-up Goal/Focus: Effect of pedometer and WeChat-based group program on PA and health-related outcome Mode of delivery IG: a pedometer- and group-based intervention program with 10–20 participants each in 47 groups. 1.Pedometer: to monitor the PA during waking hours and upload the data to a specific website. 2.We-Chat group: created by each group captain to share daily steps number, communicate, and motivate the participants to achieve the corresponding goals. CG: required to complete the measures and no intervention applied. Underlying theory: SCT	Primary outcome: 1.PA 2.Health related outcome: height (cm), weight (kg), Body Fat % (BF%), systolic diastolic BP, waist and hip circumference, BMI Secondary outcome: Job demand and control which measured based on Karasek’s Job Content Questionnaire (JCQ)	Baseline Significant difference between IG and CG (p<0.05): Waist and hip circumferences, BMI Post intervention   • Walking increased about 22% compared with baseline vigorous PA   • Significant increase in vigorous PA (p = 0.048) and walking (p<0.01) in IG   • Significant reduction in moderate PA in both IG (p<0.01) and CG (p<0.01)   • Significant reduction in health-related outcomes at post intervention (p<0.05) in IG: systolic BP, waist and hip circumferences, body fat and BMI Significant association between: 1.Gender and WHR (p<0.001): females showed larger decrease WHR 2.Age and BF% (p = 0025): age increased; BF% decreased 3.Age (p<0.027) and difference in METs for vigorous PA (p<0.001) with BF%: older and higher difference showed larger decreased BF% 4. Vigorous PA and BMI (p = 0.013): higher difference, larger decreased BMI
Pillay JD et al. [40], South Africa, 2014	Type: Pilot study Participant: 22 Staff members of the Faculty of Health Sciences, University of Cape Town Company Size: Not provided, 25 employees agreed to join	Length: 10-week intervention with 2-week follow-up Goal/Focus: Effect of pedometer-based program on PA level Mode of delivery IG: Received biweekly individual email feedback, motivational messages, and strategies to increase PA based on electronic receipt of pedometer data. CG: Received general motivational message biweekly without pedometer feedback Underlying theory: Theoretical model of behavioural change	Primary outcome:1. Participants’ perceptions: false, appeal, support and benefits of the intervention assessed through questionnaire during follow-up period. Secondary outcome: 1.PA level = measured at baseline and follow-up   • Steps per day: recorded by pedometer Biometric and clinical measure, including waist circumferences (cm), body fat (%), BMI (kg/m2), systolic and diastolic blood pressure (mmHg)	IG at 2-week follow up period after intervention: 1.Average daily aerobic steps: decreased (-54±2746 steps) 2.Daily aerobic time (min): Increased (+0.9±23.0) 3.Daily steps: increased (+996±1748 steps) CG at 2-week follow up period after intervention: 1.Daily steps: increased (+97±750 steps) This pilot study provides useful information on the potential for PA improvements through pedometry in an employed, adult group.
Ganesan A et al. [44], Asia (India), Europe, Africa, North America, South America, Australia, and New Zealand, 2016	Type: Prospective cohort study Participant: 26,562 Indian employees from private and public sector organizations Company Size: Not provided, 69,219 employees completed pre-event questionnaire	Length:100-day intervention annual program without follow-up Goal/Focus: Effect of mHealth on PA, sitting and weight Mode of delivery IG: Stepathlon workplace-based pedometer program 1.Pedometer: for step count challenge. 2.Stepathlon website (desktop and mobile version): provides personalized tools, educational content, and online community platform. 3.Daily encouraging emails: PA and nutrition messages, entertaining quizzes, and competitions to encourage online interface interaction No CG or theory applied.	PA measures: Change in step count, sitting duration and weight (kg)	Post program showed significant improvements (p<0.0001) in: 1.Recorded step count (+3,519 steps/day) 2.Exercise days (+0.89 days) 3.Sitting duration (-0.74 h) 4.Weight loss (-1.45 kg) Improvements occurred in women and men, in all geographic regions, and in both high and lower-middle income countries.
D. Job Performance (n = 2)
Guo YF et al. [27], China, 2020,	Type: RCT Participant: 73 clinical nurses from Chinese tertiary general hospital Company Size: 197 nurses	Length: 6-month intervention without follow-up Goal/Focus: Effect of WeChat-based psychotherapy on job performance and self-efficacy of people suffering burnout. Mode of delivery IG: WeChat-based 3GT-positive psychotherapy. 1.WeChat: to record three good things that were impressive each day and answer 2 questions: “Why did these good things happen?” and “What was your role in bringing them about?” in their WeChat circle 2.Reminder messages for recording 3 good things: sent to all the nurses at 8 pm by the researcher to remind them to increase the adherence of 3GT CG: No intervention received Underlying theory: Self Efficacy	Primary outcome: job performance measured by a 16-item scale Second outcome: Self-efficacy was measured by the 10-item GSS Tertiary outcome: burnout was measured by MBI-GS	Significant main intervention effect (p<0.05) and interactions (p<0.05) in both IG and CG on job performance and self-efficacy. Significant difference (p<0.05) between post-intervention scores for job performance and self-efficacy between IG and CG. Significant difference (p<0.05) between scores for job performance and self-efficacy of the IG before and after. Three Good Things are recommended to be included into the management systems to improve nurses’ physical and mental health and work outcomes over the long term.
Sasaki N et al. [43], Vietnam, 2021	Type: Three-arm RCT Participant: 951 nurses from national public tertiary hospital Company Size: 1256 nurses	Length: 7-month intervention with follow-up Goal/Focus: Effect of phone-based stress management program on work engagement Mode of delivery IG: Two 6-module CBT programs using ABC Stress Management app. Program A with free-choice multi-module, Program B with fixed-sequential order multi-module, both receive weekly messages and access to informal group chat (via social media apps such as Vober, Zalo, FB messenger) to receive intensive technical support CG (waitlist): 1.Free to use any other mental health services as usual treatment during the intervention period. 2.Received intervention after the 7-month intervention period No theory applied.	Work engagement scores	Work engagement scores in both IG: Increased from baseline to 3-month follow-up but decreased at the 7-month follow-up At 3-month follow-up: 1.Program A showed a non-significant trend (P = 0.07) toward improved engagement. 2.Program B showed a significant intervention effect on improving work engagement (P = 0.049) with a small effect size. 7-month follow up: neither program achieved effectiveness. A fixed order (program B) delivery of a smartphone-based stress management program improving work engagement in nurses in Vietnam effectively but with temporary and small effect.
E. Other health outcome such as stress (n = 3), sleep (n = 2), ergonomic condition (n = 1)
Aliakbari R et al. [29], Iran, 2020	Type: Quasi-experimental study Participant: 63 general dentists and dental specialists in Bojnourd city Company Size: 90 general dentists	Length: 3-month intervention without follow-up Goal/Focus: Effect of digital-based education to improve occupational health and ergonomic condition Mode of delivery IG: educational intervention developed based on predictive constructs using modern media. Messages: Receive 1–2 per day for a month about changing behaviour, share knowledge in a Telegram group to improve musculoskeletal conditions, and receive access to online educational material CG: received software package consisting of several applications, articles and corrective trainings as IG but not involved in Telegram group. Underlying theory: Theory of Planned Behaviour	Primary outcome: evaluated by a questionnaire about 1.Health condition: personal info, daily activities, and exercise 2.Knowledge and constructs of behavioural intention model Ergonomics condition: measured by Nordic questionnaire and Rapid Upper Limb Assessment (RULA)	Significant difference of mean scores of constructs between IG and CG at pre-intervention: Perceived control (p = 0.04) Significant difference of mean scores of constructs in IG at post-intervention: Attitude score (p = 0.03) Subjective norms, perceived control, attitude, and behavioural intention had the highest predictive power in improving the health and ergonomic position of dentists, respectively.
Nourian M et al. [31], Iran, 2021	Type: RCT Participant: 41 nurses working in 2 COVID-19 care wards in hospital. Company Size: 44 nurses in 2 COVID-19 care wards	Length: 7-week intervention Goal/Focus: Effect of WhatsApp-based training program on sleep quality Mode of delivery IG: Mindfulness-based stress reduction training program via WhatsApp group. Training content included educational media files of mediation, yoga exercise, speeches delivered by professionals CG: completed 2 questionnaires on Porsline website, received music and training file without WhatsApp application. No theory applied.	Sleep quality measured by Pittsburgh Sleep Quality Index tool (Score of 5 or higher indicates poor sleep quality)	Significant improvement on sleep quality score in IG: Subjective sleep quality (p<0.05), Sleep latency (p<0.05), Habitual sleep efficiency (p<0.05) Significant improvement on sleep quality score in CG: Subjective sleep quality (p<0.05), Daytime drowsiness (p<0.05), Total sleep quality score (p = 0.001) Significant difference on sleep quality score between IG and CG: Sleep latency (p<0.05), Subjective sleep quality (p<0.001) The total sleep quality did not change among the participants in the IG at pre- and post-intervention but increased significantly in the CG. The MBSR program may be effective in improving the sleep quality of nurses.
Pendse et al. [32], India, 2012	Type: Pilot study with part 1 (P1) survey and part 2 (P2) intervention Participant: Service sector employees: 81 (P1) and 10 (P2) Company Size: Not provided	Length: 2-week intervention without follow-up Goal/Focus: Effect of web-based intervention on work life quality and mental well-being Mode of delivery IG: receive two 20–40 seconds stimuli in the forms of pictures, videos, and text through official emails every day for 10 working days. CG: No intervention received Underlying theory: Influential theory	Questionnaire in P1: measure career and job satisfaction, perceived absence of work stress Questionnaire in P2: measure affect balance and emotion/worry	Happiness positively and significantly related to 1.Quality of Work Life (R = 0.378, P<0.01) 2.Resilience (R = 0.365, P<0.05) CG on affect balance significantly lower than the gain scores of IG (U = 14, P<0.05) Web-based interventions show promise for enhancing employee’s’ happiness but this study was limited by there’s limitation on small sample size
Divya K et al. [35], India, 2021	Type: Pilot study with single arm pre-post design Participant: 92 healthcare providers (HCPs) Company Size: 7597 HCPs from different states	Length: 40-day intervention with follow-up Goal/Focus: Effect of online video-based education on mental wellbeing Mode of delivery IG: received a 4-day online breath and meditation workshop, Sudarshan Kriya Yoga (SKY) delivered by trained instructors with a 2-hour session/ day through video conference. Participants also learnt the 35-min home practice, including Pranayama, Bhastrika and SKY breathing to be practised at home daily. No CG or theory applied.	Primary outcome: 1.Depression and Anxiety: measured by self-reported Depression, Anxiety and Stress Scale (DASS-21). 2.Sleep Quality: measured by Pittsburgh Sleep Quality Index (PSQI). 3.Resilience: measured by self-rated Connor-Davidson Resilience Scale. Life satisfaction: measured by a 5-item Satisfaction with Life Scale.	Significant differences in the scale scores: 1.Depression, anxiety, and stress: reduced scores for all at post-intervention (p<0.001) 2.Resilience: Increased resilience at post-intervention (p<0.001) and greater increase at follow-up phase (p = 0.015). 3.Life satisfaction: increased life satisfaction at post-intervention (p<0.001) and greater increase at follow-up phase (p< 0.001). 4.Quality of sleep: reduced scores immediately after the program (p<0.001) SKY breathing technique had a positive impact on the well-being of healthcare professionals during the pandemic. Participants experienced improved quality of sleep, enhanced satisfaction with life, and increased resilience after SKY.
Dincer B and Inangil D [37], Turkey, 2021	Type: RCT Participant: 72 nurses caring for COVID-19 patients in a university hospital Company Size: Not provided, 80 nurses met criteria	Length: Not provided, without follow-up Goal/Focus: Effect of online group emotional freedom techniques (EFT) treatment on reducing stress, anxiety, and burnout level Mode of delivery IG: received a 20-mins guided online group EFT treatment by showing the participants a picture of the acupressure points and ways to apply pressure tap. CG: no EFT treatment received between the completion of 2 subjective units of distress (SUD) and burnout tests. No theory applied.	Primary outcome: 1.Stress levels: measured by SUD scale 2.Anxiety levels: measure the State Anxiety Scale 3.Burnout levels: measured by a 21-item Burnout Scale	Significant differences in IG at post intervention (p < .001): 1. Stress levels: Reduced mean SUD score 2.Anxiety levels Reduced anxiety score 3.Burnout levels: Reduced burnout score CG showed no statistically significant changes on these measures (p > .05)
Montagni I et al. [45], China, France, Spain, UK, 2019	Type: Pilot study Participant: 291 employees (T1) of eight company sites in four countries Company Size: Not provided, 834 employees in (T0)	Length: Two phase (T0 and T1) intervention of 5 days each. Follow-up (T1) after 6 months Goal/Focus: Effect of blended tablet-based survey to raise sleep awareness Mode of delivery IG: asked to use WarmUapp tablet application which consist of 27 screens (23 screens for questions, 3 screens for partial survey answers, 1 screen for survey results and personalised recommendations to improve sleep quality) No CG or theory applied	Primary outcome: 1.Change in sleep status which measured total sleep duration, sleep efficiency, sleep debt, insomnia, sleepiness 2.WarnUappTM effectiveness: measured user satisfaction, feedbacks and ideas through structured interview composed of 3 open-ended questions	Significant difference in sleep status at post intervention (follow-up phase): 1.Increased total sleep duration (p = 0.046) 2.Decreased sleep debt (p = 0.019), sleep difficulties between two phases (p<0.001), sleepiness (p = 0.026) 3.Sleep problem: females 2 times more likely to suffer in both phases (p = 0.006) Satisfaction with WarnUappTM: All interviewees were satisfied of the intervention Interventions blending face-to-face and web-based approaches show promise for effective promotion of sleep awareness at the workplace.