Introduction

Cardiovascular diseases (CVD) are a significant cause of disability and premature death, accounting for one-third of total deaths globally each year [1, 2]. Approximately 80% of these deaths occur in low and middle-income countries (LMIC) [3, 4]. Global costs attributable to CVD were estimated at $863 billion in 2010, with 55% due to direct healthcare costs and 45% due to loss of productivity from disability and premature death or time lost from work due to illness. These costs are projected to increase to $1,044 billion by 2030 [5, 6]. In LMIC, it was estimated that nearly 2% of the Gross Domestic Product (GDP) was lost due to CVD between 2011 and 2015 [7].

A sedentary lifestyle, poor diet, and excessive body weight are well-established major risk factors for CVD [8, 9]. The evidence for the ability of lifestyle interventions to improve diet and increase exercise to reduce cardiovascular risk is substantial [10, 11]. Despite the evidence supporting the use of lifestyle interventions to prevent CVD, their translation into real-world settings has been challenging. Both individual and group-level interventions have effectively promoted behavior change; however, attrition rates in these programs are high, and behavior change is typically not sustained [12, 13]. Social ecological approaches that address social and environmental factors to prevent cardiometabolic risk reach broadly to target large segments of the population rather than focusing on a small number of selected individuals [14, 15]. Because many people spend most of their time at work, worksites provide unique opportunities for health promotion and disease prevention programs, (24) and worksites have a number of characteristics that could support multi-component, multi-level interventions. For example, worksites typically feature existing social networks, formal communication systems, readily available eating environments, and access to employer support programs.

Worksite-based health programs have positively impacted employee health [16, 17] and significantly improved cardiovascular risk factor profiles [18]. Worksite interventions encompassing environmental changes (i.e., low-cost healthy food options or spaces for physical activity, such as gyms, have been highlighted as components of successful worksite interventions [18, 19]. Previous studies that evaluated environmental level changes in worksite settings have shown effects for preventing obesity [20, 21], and lowering lipids [20–22], blood glucose levels [20–22], and blood pressure [14, 23, 24]. However, few studies have evaluated an independent effect of environmental intervention, while others examined the combined effect of environmental and education interventions [8, 10, 15]. Moreover, much of the evidence comes from high-income countries, and there is limited data from low-income settings. The Nepal Pioneer Worksite Intervention Study [25] aims to assess the effectiveness of a workplace environmental, dietary modification intervention and a nutrition education intervention both alone and in combination. In this paper, we report the changes in cardiovascular risk factors and dietary consumption (servings per week) among high-risk employees in a worksite setting after six months of an environmental dietary modification.

Methods

Results

Study enrollment and follow-up

We invited 830 employees for screening, of which 35 (4%) refused. The remaining 795 employees attended the screening. From these, 435 (55%) were excluded (337 were ineligible, 58 were unwilling, and 34 were for other reasons (Fig 1). Of the remaining 360 employees, 313 (86%) participated in the study. Thirty-six (13%) dropped out (25 were unwilling, 11 resigned, or were on long-term leave) before the start of the cafeteria intervention. Hence, 277 participants received cafeteria intervention. After 6 months of cafeteria intervention, 20 (7%) dropped out (4 unwilling, 10 resigned, 6 pregnant). Fig 1 shows the details of study enrollment and follow-up. For the analysis of food groups, complete data was available for 245 participants.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. Study flow diagram.

https://doi.org/10.1371/journal.pone.0301826.g001

Characteristics of participants

The characteristics of the 277 participants (155 women and 122 men) at baseline are shown in Table 1. The mean age of the participants was 32.3 (SD = 7.9) years. Half of the participants (52%) were from the Newars, and 35% were Brahmin/Chhetri. The majority were Hindu (89%). More women had a high-school education compared to men.

Download:

• PNG
  larger image
• TIFF
  original image

Table 1. Characteristics of the study participants at baseline (n = 277).

https://doi.org/10.1371/journal.pone.0301826.t001

The distribution of CVD risk factors in the study participants at baseline is shown in Table 2. More men were smokers, drank alcohol, had quite high alcohol intake on average, and had higher levels of physical activity compared to women. More men had high blood pressure. More women were classified as overweight and obese compared to men.

Download:

• PNG
  larger image
• TIFF
  original image

Table 2. Cardiovascular risk factors of the study participants at baseline (n = 277).

https://doi.org/10.1371/journal.pone.0301826.t002

The proportion of food intake in servings per week before and after the cafeteria intervention is presented in Table 3. Six months post cafeteria intervention, the median consumption of whole grains, mono/polyunsaturated fat, fruits increased, and nuts increased, whereas the median consumption of refined grains decreased. There were no significant changes in the consumption of potatoes, vegetables, saturated fats, lentils, red meat, SSB, and white meat.

Download:

• PNG
  larger image
• TIFF
  original image

Table 3. Change in food groups (servings per week) six months’ post cafeteria intervention (n = 245).

https://doi.org/10.1371/journal.pone.0301826.t003

The CVD risk factors and biomarkers before and after cafeteria intervention are presented in Table 4.

Download:

• PNG
  larger image
• TIFF
  original image

Table 4. Change in cardio-metabolic risk factors after cafeteria intervention (n = 257).

https://doi.org/10.1371/journal.pone.0301826.t004

The mean systolic blood pressure and diastolic blood pressure decreased significantly by 1.69 mm of Hg (95% CI: -2.81, -0.57; p = 0.003) and 1.08 mm of Hg (95% CI: -1.89, -0.33; p = 0.01) respectively after cafeteria intervention after adjusting for potential confounders such as age, gender, education, ethnicity, and religion. As shown in Table 4, the Mean LDL decreased significantly by 6.71 mg/dL (95% CI:-9.53, -3.89; p<0.001) after the cafeteria intervention after adjusting for confounders. The mean fasting blood sugar increased by 4.42 mg/dL (95% CI:3.11, 5.87; p<0.001) during the cafeteria intervention period in the linear mixed effect model adjusted for potential confounders such as age, gender, education, ethnicity, and religion. The mean HbA1c remained unchanged during cafeteria intervention. There were no significant changes in BMI, triglycerides, and total cholesterol.

Discussion

In a worksite in central Nepal, six months of cafeteria intervention to facilitate the intake of healthy foods and discourage the increase of unhealthy food resulted in positive changes in dietary intake and cardiometabolic risk factors. Specifically, after six months of cafeteria intervention, there was a significant increase in the intake of whole grains, fruits, nuts, mono/polyunsaturated oil, and fish, whereas the intake of refined grains decreased. There was a slight improvement in systolic and diastolic blood pressure and low-density lipoprotein. However, there was an increase in mean fasting blood sugar level.

In our study, we found an increase in the consumption of healthy foods and a decrease in the consumption of unhealthy foods. Systematic reviews have shown that strategies promoting healthy foods in the workplace, particularly environmental changes, have improved the intake of healthy foods [15, 38–41]. Our intervention used a multi-component environmental approach to promote healthy choices. A number of important factors led to the positive changes–increased availability of healthy food options, advertisement of healthy foods, and engagement of employees and cafeteria managers in planning and monitoring [42].

We observed a decrease in blood pressure six months after the cafeteria intervention in the workplace by adding whole grains, fruits, and vegetables and reducing refined grains and sugar-sweetened beverages. A large reduction in blood pressure (13.2 mmHg SBP and 14.9 mmHg DBP) was observed after introducing a Mediterranean diet in a workplace cafeteria over a year [24]. Another intervention study that implemented a low-fat vegan diet for 22 weeks resulted in no change in blood pressure, but it rose in the control group by approximately 5 mmHg at the same time [43]. The DASH (Dietary Approaches to Stop Hypertension) trial demonstrated a diet that emphasizes fruits, vegetables, and low-fat dairy products, including whole grains, poultry, fish, and nuts, while limiting red meat, sweets, and sugar-containing beverages. Total and saturated fat lowered blood pressure substantially in people with and without hypertension [44–46]. Our study has shown an increase in the intake of whole grains, fruits, nuts, mono/polyunsaturated fat, and fish and a reduction in the intake of refined grains. This could mediate blood pressure reduction and decrease the LDL level in the blood.

It is unclear why fasting blood glucose increased during the cafeteria intervention period. Fasting blood glucose levels might have been affected by a recent diet. There was no change in HbA1c during the cafeteria intervention period. BMI did not decrease over the intervention year. However, mean BMI over the past decade has been increasing worldwide [47]; the observed lack of increase in BMI may be due to a beneficial effect of the intervention. A study of health appraisal found no change in BMI over the intervention but increased BMI in the control group that did not receive any appraisal over 2 years [23]. It is argued that the impact achieved in a wellness program should be compared to the likely natural decline in the health status of the population that results from aging and the natural progression of risk in the absence of intervention [4].

We did not find a significant change in total cholesterol, triglycerides, or HDL. However, mean LDL decreased significantly by 7 mg/dL six months after the start of the cafeteria intervention. Worksite interventions have found a similar decrease in LDL through a low-fat plant-based diet [21]. A factory-based intervention of a low-calorie, increased fiber, and unsaturated fat diet found significantly lower mean serum cholesterol of nearly 10%, increased HDL, and decreased LDL after two years [48]. Finally, an intervention with a low-fat vegan diet for 22 weeks led to a decrease in total and low-density lipoprotein that was not statistically significant, while HDL significantly decreased in the intervention group [43].

The environmental-level changes to the cafeteria were modifications to offer healthy food; there were no other changes in the worksite’s overall health policies. Therefore, employees had the choice of whether or not to take the healthier options introduced by the cafeteria intervention. Cafeteria interventions that mandate the employees to take only healthy foods may improve greater behavior change [49]. However, environmental-level changes may not be sufficient to bring the behavior changes that impact health. According to the Social Cognitive Theory, a person’s behavior is caused by an interaction of cognitive, affective, and environmental influences [50]. Healthy environments may keep low-risk employees at low risk and improve health outcomes for higher-risk employees [51].

The study has several strengths. This is the first study to report a positive effect of worksite-based environmental level changes on cardiometabolic disorders in Nepal. The employee participation rate was high. About 95% of those invited participants in the screening and 85% of eligible participants were enrolled, and 70% were retained at the end of the first year. The major reason for discontinuation was resignation or work leave for further study. This shows that it is feasible to implement a worksite CVD intervention in this setting. As mentioned in the published study protocol, the intervention was designed using a participatory approach, where catering and workplace stakeholders were involved in the study design and implementation of the interventions in the individual workplaces [25]. Participatory and theory-based approaches to workplace health promotion have been recommended to ensure the effectiveness of nutrition in workplace health promotion [52]. Finally, we used objective measurements to assess cardiometabolic risk factors, minimizing the information bias. The before-after design controls for confounding by all time-invariant risk factors.

Several limitations should be considered when interpreting these results. First, there is a possibility of regression to the mean bias [53], especially in the results six months post health risk appraisal, due to random within-person variation. Second, the study may have limited generalizability because the results are employees of one hospital in central Nepal. However, the results highlight the need for health assessments and more intensive health promotion efforts in similar worksites throughout Nepal, suitably contextualized and evaluated. Further, the socio-demographic characteristics of the participants who completed the study were mostly similar to those of those who dropped out (S2 Table. Comparison of characteristics of participants who completed the study and who dropped out of the study). However, those who dropped out were, in general, older than those who completed the study and had higher chances of being overweight and pre-diabetic. This might have limited the study’s generalizability. However, the age, overweight prevalence, and prediabetes prevalence of drop-outs were similar between the intervention and control groups. Third, there was no comparison group in this study because it was not feasible to intervene in some of the four worksite cafeterias and not others since employees were allowed to go to any cafeteria of their choice. The participants’ food intake and health outcomes were compared before and after the intervention, so we could not control for confounding by time and other variables that change with time. Fourth, social desirability reporting bias cannot be ruled out either, as employees with higher nutrition knowledge may have overestimated their intake of healthy foods in the dietary recalls.

Conclusion

Overall, consumption of healthy foods increased, and blood pressure and LDL had clinically irrelevant reductions six months after an environmental-level health intervention incorporating access to healthy foods in the cafeteria. There was no change in HbA1c and BMI. This is slightly beneficial compared to the expected increase in these indicators as a natural risk progression without intervention. Worksite can offer a unique intervention environment to improve the health of its employees by reducing risk factors and promoting healthy behaviors. However, more studies with modified interventions are needed to study its effect on cardiometabolic risk factors.

Supporting information

Acknowledgments

We acknowledge Dhulikhel Hospital- Kathmandu University Hospital for their permission and cooperation in conducting this study. We appreciate the contribution of all study participants.