Assessing Digital-Age Lifestyle Health and Well-Being Among IT Professionals in Sri Lanka: A Healthy Lifestyle Index Approach

1. Introduction

The rapid expansion of digital technologies under Industry 4.0 has changed the nature of work in many service-based industries. The IT industry is one of the most digitally intensive sectors because employees depend on computers, online communication platforms, development environments, cloud services and continuous digital collaboration. These technologies increase productivity and flexibility; however, they also create lifestyle-related health risks when work routines are not managed in a sustainable manner.

Sri Lanka's IT industry has become an important contributor to employment generation, innovation and economic growth. The long-term success of this sector depends not only on technical skills and organizational performance, but also on the health and well-being of the workforce. IT professionals commonly experience long sitting time, prolonged screen exposure, irregular meals, insufficient physical activity, mental fatigue, disrupted sleep and reduced separation between work and personal life. These issues can reduce employee well-being, productivity and sustainable workforce participation.

Several studies have examined occupational stress, work–life balance or mental well-being in professional groups. However, a clear gap remains in the Sri Lankan context: there is limited evidence on integrated, IT-specific lifestyle health assessment models that combine nutrition, physical activity, stress, sleep, social connection and substance-related behaviours into a single data-driven framework. Existing approaches often focus on one health dimension and therefore provide limited support for holistic workplace decision-making. Addressing this gap is important because IT work has unique patterns, such as hybrid work, remote work, long coding sessions, project deadlines and continuous digital connectivity.

Lifestyle medicine provides a useful foundation for this study because it emphasizes health promotion through daily behavioural practices. The six-pillar framework of lifestyle medicine includes healthy nutrition, regular physical activity, stress management, restorative sleep, positive social relationships and avoidance of risky substance use. When these dimensions are measured together, they can provide a practical profile of lifestyle health and identify areas that require targeted intervention.

The main aim of this research was to develop and analyse a digitally driven lifestyle health assessment framework for Sri Lankan IT professionals. The specific objectives were to construct a composite Healthy Lifestyle Index (HLI), examine its relationship with subjective well-being measured through the WHO-5 items, evaluate the influence of demographic and workplace factors on lifestyle health and propose the relevance of digital lifestyle analytics for sustainable workforce development.

2. Literature Review

2.1. Lifestyle Medicine and Workforce Health

Lifestyle medicine focuses on preventing and managing lifestyle-related health problems through evidence-based behaviour change. Its core pillars provide a multidimensional approach rather than considering health only as the absence of disease. In a workforce setting, these pillars are important because daily work routines directly shape eating patterns, movement, sleep, stress responses and social connection.

Healthy nutrition and regular physical activity are strongly linked with energy level, concentration and long-term non-communicable disease prevention. Stress management and adequate sleep are also essential for cognitive performance, emotional regulation and recovery from work demands. Social connection supports psychological resilience, while avoiding harmful substance use reduces lifestyle-related health risk. Therefore, the six-pillar model is suitable for assessing the health sustainability of digitally intensive workers.

2.2. Occupational Health Challenges in the IT Sector

IT professionals face occupational risks that differ from many traditional workforces. A large portion of their work is performed through screens, keyboards, digital meetings and continuous online collaboration. This creates sedentary behaviour, static posture, eye strain, cognitive overload and limited opportunities for movement during working hours.

Project deadlines, rapid technology changes and the expectation of constant availability can increase work-related stress. Remote and hybrid work arrangements may improve flexibility, but they can also weaken work–life boundaries and extend working hours into personal time. These work characteristics show why a general health questionnaire may not be sufficient for the IT workforce and why a local, occupation-specific assessment is required.

2.3. Digitalization, Industry 4.0 and Sustainable Well-Being

Industry 4.0 environments depend on automation, data analytics, cloud computing, artificial intelligence and digitally connected work systems. In such settings, workforce sustainability should be understood as the ability of employees to perform effectively while maintaining physical, mental and social well-being. Digital health assessment can support this aim by converting self-reported lifestyle data into interpretable health indicators.

A digital lifestyle assessment model can help organizations identify common risk areas, compare lifestyle patterns across work arrangements and design well-being programmes based on evidence. It can also create a foundation for future AI-based predictive systems that identify early signs of lifestyle imbalance before they develop into serious health or performance problems.

2.4. Research Gap and Contribution

Although international literature supports holistic lifestyle assessment, there is a lack of empirical research that applies this approach to Sri Lankan IT professionals. Most local discussions are fragmented across stress, work–life balance or occupational satisfaction. Few studies integrate multiple lifestyle medicine domains with a standardized well-being outcome and workplace variables such as work arrangement, working hours, job role and company size.

This study contributes by developing a composite HLI for Sri Lankan IT employees and testing its relationship with subjective well-being. It also responds to reviewer recommendations by explicitly examining sampling, self-reported data bias, demographic variables and workplace-related differences. The proposed framework can be used as a practical base for digital health dashboards, organizational well-being assessments and future AI-enabled lifestyle prediction models.

3. Materials and Methodology

3.1. Research Design

A quantitative cross-sectional research design was used. The design was selected because the study aimed to measure current lifestyle health patterns and examine statistical relationships between lifestyle dimensions, demographic variables, occupational characteristics and well-being outcomes among IT professionals in Sri Lanka.

The study used a structured online questionnaire to support digital data collection. This method was suitable for the IT workforce because participants were familiar with digital forms and could respond from different work settings, including remote, hybrid and on-site environments.

3.2. Participants and Sampling

The target population consisted of IT professionals working in Sri Lanka. Participants represented different job roles such as software engineering, quality assurance, business analysis, data science, system and network administration, cloud and DevOps engineering, cybersecurity, UI/UX engineering, project management, technical leadership, IT support and internship positions.

A non-probability online distribution approach with broad outreach was used to reach respondents across job roles, experience levels, company sizes and work arrangements. Although the original plan described random distribution, the final survey process is more accurately described as voluntary online participation through professional and academic networks. This clarification is important because online voluntary sampling may introduce selection bias; for example, professionals with stronger interest in health or higher availability may have been more likely to respond. After removing incomplete entries, 481 valid responses were included in the final analysis.

3.3. Questionnaire and Measures

The questionnaire contained demographic and occupational questions followed by lifestyle health items. Demographic variables included age, gender, marital status and education level. Occupational variables included current job role, years of experience in the IT field, average working hours per week, work arrangement and company size.

Lifestyle health was assessed using items aligned with the six pillars of lifestyle medicine. Nutrition items measured balanced diet, label checking, limiting high-fat and high-sugar foods, breakfast habits and regular meal timing. Physical activity items measured vigorous activity, moderate activity and daily movement beyond work. Stress management items measured coping with work stress and leisure time, while negatively worded items on emotional exhaustion and loss of interest were reverse scored. Substance avoidance was assessed by reverse scoring tobacco, alcohol and medication overuse items and including moderate caffeine use. Sleep was measured through restfulness, consistent sleep timing and 7–8 hours of sleep. Social connection was measured through personal support, affection, sharing worries and work–life balance.

Subjective well-being was assessed using five WHO-5-type items: feeling cheerful and in good spirits, calm and relaxed, active and vigorous, fresh and rested, and interested in daily life. All lifestyle and well-being items were measured using a five-point Likert scale. Higher scores were interpreted as healthier lifestyle behaviour or better well-being after reverse scoring was applied where necessary.

3.4. Data Processing and Index Construction

Data cleaning was performed before analysis. Incomplete entries and blank spreadsheet rows were removed. Negatively worded items were reverse scored using the transformation shown in Equation (1), where X is the original Likert score.

Reverse scored value = 6 − X

(1)

Each lifestyle domain was calculated as the mean of its relevant items. The Healthy Lifestyle Index (HLI) was constructed as the unweighted average of the six domain scores, as shown in Equation (2).

HLI = (Nutrition + Physical Activity + Stress Management + Substance Avoidance + Sleep + Social Connection) / 6

(2)

For easier interpretation, HLI and WHO-5 mean scores were transformed to a 0–100 scale using Equation (3).

Transformed score = ((Mean score − 1) / 4) × 100

(3)

3.5. Statistical Analysis

Statistical analysis included descriptive statistics, reliability inspection, multicollinearity checking, Pearson correlation analysis, ordinary least squares regression, sensitivity analysis and group comparisons. Variance inflation factors (VIFs) were used to check whether the six lifestyle domains were sufficiently distinct for formative index construction. One-way ANOVA was used to examine differences in HLI across demographic and occupational groups.

Self-reported data were treated carefully because responses may be affected by recall bias and social desirability bias. To reduce interpretation risk, the study focused on patterns and associations rather than causal claims. The cross-sectional design also means that the analysis cannot prove that lifestyle health causes well-being; it can only show whether the two are statistically related in the sample.

3.6. Ethical Considerations

Participation was voluntary and anonymous. The questionnaire included a consent statement before participation. No personally identifiable information was required for the analysis. Data were used only for academic research purposes, and results were presented in aggregated form.

4. Results and Discussion

4.1. Demographic and Occupational Profile of Respondents

A total of 481 valid responses were included. The sample included 299 male respondents, 145 female respondents and 37 respondents who preferred not to state gender. Most respondents were married (n = 290), while 191 were single. In terms of education, 251 respondents held a bachelor's degree, 149 held a master's degree and 81 held a diploma. The age range was 21 to 54 years, showing representation from early-career to senior professionals.

Table 1. Demographic and occupational profile of respondents.

Variable	Distribution
Gender	Male: 299; Female: 145; Prefer not to say: 37
Marital status	Married: 290; Single: 191
Education level	Bachelor's: 251; Master's: 149; Diploma: 81
Work arrangement	Remote: 188; Hybrid: 152; On-site: 141
Working hours per week	<40: 38; 40–50: 214; 51–60: 139; 60+: 90
Experience	<1 year: 116; 1–3 years: 96; 4–6 years: 116; 7–10 years: 84; 10+ years: 69
Company size	Small: 139; Medium: 184; Large: 158

Table 1. Demographic and occupational profile of respondents.

Variable	Distribution
Gender	Male: 299; Female: 145; Prefer not to say: 37
Marital status	Married: 290; Single: 191
Education level	Bachelor's: 251; Master's: 149; Diploma: 81
Work arrangement	Remote: 188; Hybrid: 152; On-site: 141
Working hours per week	<40: 38; 40–50: 214; 51–60: 139; 60+: 90
Experience	<1 year: 116; 1–3 years: 96; 4–6 years: 116; 7–10 years: 84; 10+ years: 69
Company size	Small: 139; Medium: 184; Large: 158

4.2. Lifestyle Domain Scores

The overall HLI mean was 33.76 out of 100, indicating that the sample had considerable room for improvement across lifestyle health behaviours. Among the six lifestyle domains, substance avoidance had the highest mean score (52.38), while stress management (28.66), nutrition (29.54), physical activity (29.92) and sleep (30.02) were comparatively low. Social connection recorded a mean of 32.00. These results suggest that the main lifestyle risks among the respondents are not limited to a single behaviour, but are distributed across diet, movement, stress, sleep and social/work–life balance.

Table 2. Descriptive statistics for lifestyle domains, HLI and well-being (0–100 scale).

Measure	Mean	SD	Minimum	Maximum
Nutrition	29.54	14.97	0.00	90.00
Physical activity	29.92	18.81	0.00	83.33
Stress management	28.66	16.14	0.00	81.25
Substance avoidance	52.38	20.97	0.00	100.00
Sleep	30.02	19.05	0.00	83.33
Social connection	32.00	16.77	0.00	87.50
Healthy Lifestyle Index	33.76	10.83	7.78	77.01
WHO-5 well-being	31.47	17.21	0.00	80.00

Table 2. Descriptive statistics for lifestyle domains, HLI and well-being (0–100 scale).

Measure	Mean	SD	Minimum	Maximum
Nutrition	29.54	14.97	0.00	90.00
Physical activity	29.92	18.81	0.00	83.33
Stress management	28.66	16.14	0.00	81.25
Substance avoidance	52.38	20.97	0.00	100.00
Sleep	30.02	19.05	0.00	83.33
Social connection	32.00	16.77	0.00	87.50
Healthy Lifestyle Index	33.76	10.83	7.78	77.01
WHO-5 well-being	31.47	17.21	0.00	80.00

The relatively low nutrition score may reflect irregular meal timing, limited attention to food labels and difficulty maintaining balanced meals during demanding workdays. The low physical activity score is consistent with sedentary work patterns in the IT sector. The sleep score indicates that many respondents may not experience regular, restorative sleep, while the low stress management score suggests emotional fatigue and limited recovery time after work. The higher substance avoidance score is positive, but it should not be interpreted as a complete absence of risk because caffeine use, alcohol use and medication practices can still vary across individuals.

4.3. Index Validation and Reliability Inspection

The HLI was treated as a formative composite index because the six lifestyle domains represent different aspects of health rather than interchangeable items measuring one narrow construct. Therefore, very high internal consistency was not expected. The Cronbach's alpha values for individual domains ranged from 0.240 to 0.480, and the WHO-5-type items recorded an alpha of 0.463. These modest values reflect the breadth of the constructs and the short number of items per domain. For formative validation, multicollinearity was examined using VIF values. All six lifestyle domains recorded VIF values below 1.30 when a constant was included, indicating no problematic multicollinearity among the domains.

Table 3. Reliability inspection and VIF results.

Scale/Domain	Cronbach's Alpha	VIF
Nutrition	0.293	1.242
Physical activity	0.267	1.170
Stress management	0.313	1.269
Substance avoidance	0.480	1.201
Sleep	0.276	1.188
Social connection	0.240	1.196
WHO-5 well-being	0.463	–

Table 3. Reliability inspection and VIF results.

Scale/Domain	Cronbach's Alpha	VIF
Nutrition	0.293	1.242
Physical activity	0.267	1.170
Stress management	0.313	1.269
Substance avoidance	0.480	1.201
Sleep	0.276	1.188
Social connection	0.240	1.196
WHO-5 well-being	0.463	–

4.4. Relationship Between Lifestyle Health and Well-Being

Pearson correlation analysis showed a moderate positive and statistically significant relationship between HLI and WHO-5 well-being (r = 0.471, p < 0.001). This indicates that respondents with healthier lifestyle profiles tended to report better subjective well-being. Simple linear regression further showed that HLI explained 22.2% of the variance in well-being (R² = 0.222, F = 136.90, p < 0.001). The regression coefficient for HLI was 0.749, meaning that a one-point increase in HLI on the 0–100 scale was associated with an estimated 0.749-point increase in well-being score.

A second regression model using the six lifestyle domains as predictors explained 24.5% of the variance in well-being. Physical activity showed the strongest positive coefficient (β = 6.053, p < 0.001), followed by sleep (β = 3.301, p = 0.001), nutrition (β = 3.146, p = 0.014), social connection (β = 2.967, p = 0.008) and stress management (β = 2.634, p = 0.028). Substance avoidance was positive but not statistically significant in the multiple regression model (β = 0.913, p = 0.310). This suggests that daily movement, sleep quality, nutrition, social support and stress recovery may be especially important for well-being among IT professionals.

Table 4. Regression results for well-being.

Predictor	Coefficient	p-value	Model Fit (R²)
HLI predicting WHO-5	0.749	<0.001	0.222
Nutrition	3.146	0.014
Physical activity	6.053	<0.001	0.245 (model R²)
Stress management	2.634	0.028
Substance avoidance	0.913	0.310
Sleep	3.301	0.001
Social connection	2.967	0.008

Table 4. Regression results for well-being.

Predictor	Coefficient	p-value	Model Fit (R²)
HLI predicting WHO-5	0.749	<0.001	0.222
Nutrition	3.146	0.014
Physical activity	6.053	<0.001	0.245 (model R²)
Stress management	2.634	0.028
Substance avoidance	0.913	0.310
Sleep	3.301	0.001
Social connection	2.967	0.008

4.5. Demographic and Workplace Differences in HLI

The analysis of demographic and workplace variables addressed the reviewer comment that the discussion should explore demographic variables and workplace factors more deeply. HLI differed significantly by gender group, marital status, education level, experience, working hours and work arrangement. Differences across job roles and company sizes were weaker and not statistically significant at the 0.05 level.

Table 5. ANOVA summary for HLI by demographic and workplace variables.

Variable	F Statistic	p-value	Main Observed Pattern
Gender	29.872	<0.001	Female: 34.81; Male: 34.79; Prefer not to say: 21.26
Marital status	8.141	0.005	Married: 34.89; Single: 32.03
Education level	7.546	0.001	Bachelor's: 35.27; Master's: 33.20; Diploma: 30.07
Experience	8.092	<0.001	10+ years: 37.73; 7–10 years: 37.57; 1–3 years: 30.77
Working hours	11.587	<0.001	40–50 h: 36.34; 51–60 h: 33.62; 60+ h: 30.28; <40 h: 27.91
Work arrangement	21.415	<0.001	Hybrid: 38.12; Remote: 32.69; On-site: 30.47
Job role	1.301	0.220	No statistically significant difference
Company size	0.548	0.578	No statistically significant difference

Table 5. ANOVA summary for HLI by demographic and workplace variables.

Variable	F Statistic	p-value	Main Observed Pattern
Gender	29.872	<0.001	Female: 34.81; Male: 34.79; Prefer not to say: 21.26
Marital status	8.141	0.005	Married: 34.89; Single: 32.03
Education level	7.546	0.001	Bachelor's: 35.27; Master's: 33.20; Diploma: 30.07
Experience	8.092	<0.001	10+ years: 37.73; 7–10 years: 37.57; 1–3 years: 30.77
Working hours	11.587	<0.001	40–50 h: 36.34; 51–60 h: 33.62; 60+ h: 30.28; <40 h: 27.91
Work arrangement	21.415	<0.001	Hybrid: 38.12; Remote: 32.69; On-site: 30.47
Job role	1.301	0.220	No statistically significant difference
Company size	0.548	0.578	No statistically significant difference

Work arrangement showed a clear pattern. Hybrid workers recorded the highest HLI mean (38.12), followed by remote workers (32.69) and on-site workers (30.47). This may indicate that hybrid work provides a better balance between flexibility and social/work structure. Remote work can reduce commuting time, but it may also increase isolation and blur work–life boundaries. On-site work may provide social interaction, but commuting and fixed schedules can reduce time for exercise, sleep and healthy meals.

Working hours were also important. Respondents working 40–50 hours per week recorded the highest HLI mean (36.34), while those working more than 60 hours recorded a lower mean (30.28). Long working hours can reduce time for movement, meal planning, social contact and sleep. Interestingly, the <40 hour group also showed a low mean (27.91), which may reflect early-career, internship or unstable workload conditions in part of the sample rather than a simple linear relationship.

Experience was positively associated with lifestyle health. Respondents with 7–10 years and 10+ years of experience recorded higher HLI scores than early-career groups. This may be because experienced professionals have better control over work routines, stronger coping strategies or more stable employment conditions. Age also showed a positive correlation with HLI (r = 0.280, p < 0.001), supporting the idea that lifestyle management may improve with maturity and career stability.

4.6. Sensitivity Analysis

A sensitivity analysis was conducted by removing the sleep dimension from the HLI. The correlation between the revised HLI and WHO-5 remained positive and statistically significant (r = 0.446, p < 0.001). This result indicates that the relationship between lifestyle health and well-being was not driven only by the sleep domain. The composite measure was therefore robust across alternative index construction.

4.7. Discussion and Implications for Sustainable Workforce Well-Being

The findings show that lifestyle health among Sri Lankan IT professionals is multidimensional. The relatively low scores for nutrition, physical activity, stress management, sleep and social connection suggest that IT employees may face several simultaneous lifestyle pressures. These pressures are consistent with digitally intensive work environments where time spent on screens and projects can reduce healthy routines.

The significant relationship between HLI and well-being supports the practical value of the proposed assessment framework. A digital lifestyle dashboard based on the HLI could help organizations identify common risk areas and design targeted interventions. For example, low physical activity scores could support movement breaks, walking meetings or ergonomic wellness programmes. Low sleep and stress scores could support workload planning, digital disconnection policies and mental health support. Low nutrition scores could support healthy cafeteria options, meal timing awareness and nutrition education.

The workplace factor results are especially important for sustainable development. Hybrid workers showed better lifestyle scores, suggesting that flexible work models may support well-being when balanced with adequate structure. However, remote work still requires attention to social connection, boundaries and daily movement. Long working hours were associated with lower lifestyle health, highlighting the need for organizations to consider workload distribution, deadline planning and recovery time as part of workforce sustainability.

From an Industry 4.0 perspective, this research supports the use of digital health analytics in human-centred technology environments. As organizations adopt AI, automation and cloud-based work practices, they should also adopt data-driven well-being systems that protect the workforce behind digital transformation. The proposed HLI can be integrated into future AI-enabled predictive models to identify lifestyle risk profiles and recommend personalized well-being actions while maintaining privacy and ethical data handling.

5. Limitations and Future Work

This study has several limitations. First, the cross-sectional design prevents causal interpretation. The results show associations between lifestyle health and well-being, but they do not prove that one directly causes the other. Second, the data were self-reported; therefore, responses may be affected by recall error, under-reporting of unhealthy behaviours or social desirability bias. Third, the online voluntary sampling approach may not perfectly represent all IT professionals in Sri Lanka.

Future studies should use longitudinal designs to track lifestyle changes over time. Larger samples from multiple organizations could improve representativeness. Future work can also include objective digital indicators such as wearable activity data, sleep tracking, screen-time logs or workplace analytics with ethical consent. Advanced modelling methods such as structural equation modelling, clustering and machine learning can be used to identify risk groups and predict well-being outcomes. The final long-term direction is to develop an AI-assisted digital lifestyle health platform for Sri Lankan IT professionals.

6. Conclusion

This research developed and analysed a digitally driven lifestyle health assessment framework for Sri Lankan IT professionals. Using 481 valid survey responses, the study constructed a Healthy Lifestyle Index based on six lifestyle medicine domains and examined its relationship with WHO-5 well-being. The findings showed that lifestyle health was generally low to moderate in the sample and that healthier lifestyle profiles were significantly associated with better well-being.

The results also showed that workplace arrangement, working hours, experience, education and marital status influenced lifestyle health patterns. These findings respond directly to the need for deeper analysis of demographic and workplace factors. Overall, the study provides evidence that digital lifestyle health assessment can support sustainable workforce well-being in the Sri Lankan IT industry. The proposed framework can guide future occupational health initiatives, organizational well-being programmes and AI-enabled lifestyle analytics platforms.