Introducing a short-term life-style programme with exercise intervention as the core approach within a worksite prevention strategy initiative was associated with long-term minor improvements of physical fitness, abdominal obesity, systemic inflammation, glucose metabolism as well as 10-year-cardiovascular risk by PROCAM (Table 2). Of note, this intervention only consisted of a two-day seminar focusing on behavioural changes, a six weeks supervised exercise intervention twice weekly, a follow-up seminar of a half day and encouragement to continue exercise intervention, which was supported by monthly supervised exercise sessions. Data shown are the result of a long-term assessment after one year, far beyond the active intervention phase of six weeks. Moreover, the programme was introduced in a large proportion of employees involving 60% of the whole work force at one company, which is, as a refinery, not per se a classical preventive strategies open employee cohort, characterized by a proportion of 50% working night or weekend shifts (Table 1). However, the achieved changes were not clinically relevant and no systematic changes of vascular function parameters, lipids or fatty acid profile could be established in this healthy sample (Table 2).

With regard to the relatively small effects on metabolic and vascular components after 12 months, it has to be taken into account that the study participants were a relatively healthy cohort (10-year cardiovascular risk at baseline: 3.7%; smoking rate at baseline: 8.3%). These findings are in line with previous studies and systematic reviews [7, 16]. Workplace interventions have shown to reduce body weight and waist circumference, but associations with biochemical markers as well as on blood pressure were inconclusive and small, respectively [17]. An ecological study, which examined trends of cardiovascular risk factors from 2008 – 2017 in oil refinery workers, found increasing rates only of hypertension and diabetes. Rates of low HDL, high LDL, high cholesterol, smoking and coronary artery risk decreased. The authors stated a medium cardiovascular risk in oil refinery workers and called for systematic health promotion at the workplace [18].

However, when starting with abnormal baseline levels, improvements might be far larger. In our study, with regard to endothelial function measured by RHI as primary endpoint, mean baseline values of RHI 2.00 ± 0.72 and RHI 1.87 ± 0.49 after 12 months were low, but still were above the cut-off value for endothelial dysfunction of RHI ≤ 1.67 [13, 19]. Previous studies revealed little effect of exercise training on endothelial function in healthy individuals, but endothelial function improved particularly in those with abnormal baseline endothelial function [20].

Nutritional advice or counselling beyond the introductory seminar was not given, which may explain that body weight was not considerably reduced over 12 months. Moreover, no changes of the Omega-3-Index, a parameter dependent on the consumption of unsaturated fatty acids, was observed (Table 2), although values were very low at baseline [14]. Therefore, individual nutritional counselling could have been a valid prevention strategy in addition to exercise intervention, as has shown by previous investigations [21,22,23]. We, however, did not include individual nutritional advice into our programme, as it was reported from previous experience by the corporate medical department and the employee representation that the employees were reluctant to adopt changes in nutrition to a large extend. It was felt that focussing on exercise intervention was more feasible and long-lasting. This approach was confirmed by our 12-months programme results that exercise training was only associated with improved exercise capacity also long-term. These findings are beyond most studies and even our experience that improvements observed during supervised intervention can mostly not be maintained during follow-up [24]. Obviously, the monthly training sessions offered onsite the factory area seem to be a practical and successful approach for maintaining adherence to exercise programmes. In a sub-analysis (data not shown) we classified the trainings adherence of all participants to low, moderate and high. In all three groups there was a slight reduction of 0.5% to 0.7% in the 10-year cardiovascular risk by PROCAM after 12 months.

In general, physical activity have multiple health benefits, particularly on metabolic control, vascular function and cardiovascular morbidity and mortality and is strongly recommended in recent guidelines [25, 26]. However, there seems to be an exercise paradox when comparing exercise performed during leisure time or during the occupational setting. Data from the Copenhagen General Population Study with 104,046 adults revealed that higher leisure time physical activity is associated with a 15% reduction of risk for major cardiovascular events (MACE) and a 40% reduction of all-cause mortality risk [27]. At the same time, higher occupational physical activity was associated with a 35% increase of MACE risk and a 27% increase of risk for death [27]. These paradoxical results may be explained by the different characters of leisure and occupational physical activities and by occupational various stressors, e. g. shift work [28].

Regarding monetary benefits, occupational health programmes offer a lot of potential. However, cost effectiveness of corporate wellness programmes has been challenged in a recent review [16]. The economic impact of the programmes is hard to monetise. Given the small changes and unknown cost implications, the economic evidence of workplace-interventions remains uncertain. We did not apply any systematic cost-analysis, but as we have calculated approximately €300 for each individual for one year including seminars and exercise sessions, we are convinced that this programme is cost-effective. This amount, however, does not include risk assessment examinations.

The strongest limitation is the lack of a control group. Consequently, no causal conclusions may be drawn from the results. However, the changes during intervention are certainly induced by the life-style changes, while a control group would perhaps even deteriorate. Nonetheless, a randomized trial would have been optimal. Regarding inclusion, maybe there is a selection bias, because health-conscious people are more likely to apply to health promotion programmes. Moreover, inclusion of more females would have been relevant to see changes in that group and compare these with male counterparts. However, females are not represented at an oil refinery because of the heavy labour of most jobs. Therefore, the number was overall low. Nonetheless, the strength of the current study is that 60% of employees participated in the 12 months programme and adherence was high.

In our experience this is mainly linked to the strong support of the company´s executive board in combination with the internal media and occupational health department. The health seminar added by medical examinations was clearly seen as an incentive for workers, which improved adherence to the programme. Previous studies have shown that unionization and management support were the strongest predictors of the adoption of health programmes [29].