Is It Time to Re-Shift the Research Agenda? A Scoping Review of Participation Rates in Workplace Health Promotion Programs

Workplace health promotion programmes (WHPPs) are among the most important measures to improve the health and motivation of the ageing workforce. However, they are accompanied with certain challenges, such as low participation rates and higher participation levels of the more health-conscious workers, often failing to engage those who need such interventions the most. Following the PRISMA guidelines, this scoping review examined participation rates reported in articles on WHPPs to identify potential knowledge gaps. The results are worrying: participation rates are not only infrequently reported, but also low. Of the 58 articles, 37 report participation rates, with the majority (20) reporting an average participation rate of less than 50%. Reported participation rates refer either to different target groups, the type of intervention, or to single points in time, which makes it difficult to establish consistent criteria for comparison. We argue that despite the importance of WHPP efficacy, research focus should shift to the determinants of participation, as well as the issue of standardising the reporting of participation rates, alongside the potential problem of reporting bias.


Introduction
According to the OECD data on retirement age, the duration of working life in most OECD countries has increased since 2000 [1] due to specific work patterns and demographic changes. The proportion of older workers has increased as well. Workplace health promotion programmes (WHPPs) are among the most important measures to improve the health and motivation of the ageing workforce, as well as the working conditions themselves. Thus, choosing an effective WHPP has a positive impact on maintaining and strengthening workers' physical [2] and mental health [3]. It contributes to actual vitality and work engagement and has the potential to prolong working life [2,4]. Workplace health promotion is necessary not only for employees but also for employers, as it brings many benefits to companies, such as improved work performance [5,6], higher work engagement [5], better work ability, higher job retention rate [6], and a lower financial burden due to a reduction in sick leave [7]. Thus, WHPPs can be profitable. An analysis of 51 studies with 261,901 participants and 122,242 controls from nine economic sectors in 12 countries, published between 1984 and 2012, found that overall weighted return on investment (ROI) yielded a total of USD 2.38 for every dollar invested [8].
However, WHPPs come with certain challenges and the reality is far from perfect. Studies suggest that participation rates in such programmes are often below 50 per cent [9] and typically range between 20 and 30 per cent [10]. Moreover, healthier and more healthconscious workers predominate among WHPP participants [11,12]. On the other hand, those in need of such interventions often fail to participate or to be activated by these programmes in the first place [13]. There is some limited evidence that preventive interventions or programmes, while effective, also have deadweight effects [14,15], the extent to which the (expected) effects would have occurred even without the intervention. This means that some participants in these programmes are already health conscious in the first place and are thus rewarded for performing something or being offered something for free that they would have implemented anyway. This significantly reduces the potential benefits and cost-effectiveness of these programmes [10,16].
Although 50 to 75 per cent of workers do not choose to participate in WHPPs, the field of employees' commitment to personal health is still fairly unexplored. Some of these challenges have been known for a long time, with the 2013 research agenda published in the American Journal of Health Promotion (AJHP) calling for more research to focus on organisational culture, incentives, and social marketing in WHPPs to address uptake issues [17]. The question, however, is just how much the research focus has actually shifted over the past 10 years. The authors fear not much. First, our literature search revealed that no review articles had examined WHPP participation rates. Second, a brief review of abstracts in the AJHP showed that such issues were addressed in only a few articles. Third, WHPP participation rates are (mostly, as we found in our scoping review) still below 50%.

Research Problem and Aim
WHPP researchers and practitioners seem to forget that the problem with the overall effectiveness of WHPPs is not the effects of WHPPs as such, but rather the low participation rates to begin with. Thus, the research field may have a problem that we are not even aware of. There is a large body of research in the field of WHPPs that focuses only on individual interventions and programmes and their effects. This calls into question its usefulness, given that the real problem appears to be low participation.
The only systematic review of participation rates we came across is Ryde et al., (2013) [18], which covers research up to 2010. However, it only considers workplace physical activity programmes and describes the characteristics of studies with high participation rates. The aim of our scoping review is more ambitious. First, our scoping review covers all articles up to April 2022, whereas the mentioned review only covers research from 1977 to 2010. Second, by including all WHPPs, we do not limit ourselves to physical activity programmes. Third, our analysis considers WHPP specifics that (on face value) relate to participation rates.
In this article, we present a scoping review of WHPP participation rates reported in research using the PRISMA model [19]. To shed light on this particular research topic, the analysis also considers some characteristics of WHPPs and organisations that seem to be related to WHPP participation rates. In addition to the level of employee participation, the analysis also takes into consideration the type of programme or intervention, the target group(s) of the intervention, the type of organisation where the intervention took place, and the methods used to recruit employees to participate in the intervention.
The article is structured as follows. The following section presents the application of the PRISMA method. The methods section is followed by the analysis of the results. Next, we discuss the results and point out the limitations of our scoping review. The article concludes with suggestions for future research and implications for the WHPP field.

Search Strategies
The scoping review was based on a literature search in bibliographic databases, namely PubMed, SAGE Journals, JSTOR, and Emerald in 2021-2022. In addition, a manual search of Google Scholar was conducted to identify articles that may have been missed in the database search. The search used a combination of terms and keywords, which were adjusted and refined during the search to yield more appropriate results. The first search used a combination of terms (workplace OR worksite OR employee*) AND (health promotion OR well-being OR well-being OR wellness) AND (program* OR intervention) AND (participation) AND (motives OR factors OR determinants). As the results were not entirely suitable, we adjusted the search terms and used a new combination of terms in the second search: (workplace OR worksite OR employee*) AND (health promotion OR well-being OR well-being OR wellness) AND (program* OR intervention) AND (participation rate OR recruitment OR engagement OR attendance OR "take-up") AND (rate). In the third search iteration, the term "take-up rate" was excluded from the search. An obstacle arose when searching the JSTOR article database, where the length and number of search terms and keywords are limited, which meant that the search terms had to be adapted again: (workplace OR worksite OR employee*) AND (health promotion OR wellbeing OR wellness) AND (program* OR intervention) AND (participation OR recruitment OR engagement) AND (rate).

Inclusion and Exclusion Criteria
The titles and abstracts of a total of 1374 papers were screened for relevance, and duplicates were excluded. In the next step, studies were included if the content of the abstract included WHPP intervention(s). There were no restrictions regarding the year in which the study was published, the type of research, or the type of health promotion programme. Studies involving clinical populations were excluded. Our search yielded 58 research articles, the full texts of which were examined to see if the study mentioned the participation rate or data from which we could calculate the WHPP participation rate. A total of 37 articles met this condition. The elimination process according to the PRISMA model is presented in Figure 1.

Data Extraction and Analysis
In our scoping review, the missing participation rates were calculated as the number of employees participating in the WHPP divided by the number of employees initially invited, multiplied by 100. One third of the articles were independently reviewed by both authors to ensure the accuracy of the extracted data. The articles were then assessed for the level of employee participation, the type of programme or intervention, the target group(s) of the intervention, the type of business where the intervention took place, the strategies used to recruit or activate employees to participate in the intervention, and the study characteristics. Table 1 provides a summary of this analysis.  [27] 79.8 From USD 400 to USD 600 Ott-Holland et al., (2019) [28] n.a. Up to USD 200

Frequency of Reporting on Participation Rates
Only a small number of articles on WHPPs were found to report the participation rate. However, the number of articles reporting on participation rates was found to have increased in recent years. Of the 58 relevant articles found in selected bibliographic databases, only 37 studies (64%) reported participation rates. In terms of the year of publication, seven articles were published before 2000, and seven articles were published between 2000 and 2010. Since 2010, the number of articles has increased significantly with 10 or more in each subsequent 5-year period. The oldest study dates from 1988, and the most recent study dates from 2020.
Participation rates were found to be reported in different ways. A total of 23 studies reported the average or overall participation rate in health promotion programmes. Overall participation rates were reported for the specific programme as a whole, whereby the average participation rates were calculated across different interventions or groups. For the remaining 13 studies, average participation rates were calculated by the authors from individual participation rates associated with participant groups or interventions. However, the studies by Ott-Holland, Shepherd, and Ryan (2019) [28] and Brill et al., (1991) [29] differed substantially as they spanned three years between 2010 and 2012, with participation rates for the second and third years recorded in three time periods per year.
For the programmes with the highest participation rates, we found no obvious correlation with company size, i.e., the number of employees involved. What is typical of these programmes is that employees were either enrolled or they volunteered. One of such programmes also offered a monetary prize/incentive [27]. A closer look at the programmes with the lowest participation rates showed that these were predominantly dealing with very large populations, however the strategies used to recruit or activate employees were not described. In one of the studies, a financial incentive was offered [20]. We also note that in the studies on WHPPs with low participation rates, the data relevant to understanding participation are more deficient than in the studies with higher participation rates, at least in terms of describing the process of recruiting employees. Lower participation rates appear more often in physical activity interventions than in those related to counselling and training.

Ways of Promoting Employee Involvement in WHPPs
Financial incentives are the most common form of encouraging participation in WH-PPs, but do not seem to guarantee higher participation. Financial incentives were reported in nine studies. Eight articles reported voluntary participation in the programme and two reported assigning employees to a specific programme. Seventeen studies did not report how employees joined the programme or how participation was encouraged. Incentives and their relation to participation rates are difficult to compare directly, since various incentive sizes and incentive schemes were applied-some with fixed incentives, some with variable incentives, and some with stepped incentives. When examining the used financial incentives and the respective response rates (see Table 2), it appears that higher (financial) incentives tend to be associated with higher participation rates, although financial incentives do not guarantee higher participation rates in WHHPs.

Research Design and Duration
There is an apparent diversity in both the duration of the WHPP and the research design used. Only 13 studies reported the duration of the programme/intervention. The shortest study lasted three weeks [32] and the longest four years [38]. Some used qualitative or quantitative approaches, while others used only cross-sectional designs or questionnaires.

Types of WHPPs/Interventions
The types of interventions were found to vary widely, ranging from specific to combined interventions. Analysis revealed 17 WHPPs which specified only the general area of intervention, and 20 WHPPs which explicitly described the interventions, such as physical activity programmes that included stretching, exercise programmes, gym availability, and standard walking [28,29,42,55]; health education programmes included guidelines on fruit and vegetable consumption or nutrition in general, biometric screening tests and health risk assessments [27,36]. Health and life coaches were also available. One WHPP even offered free flu vaccinations [35]. Online programmes focusing on wellness and health or general self-care were also offered, including programmes on diabetes, depression, stress management, insomnia, back pain management, relaxation techniques, weight management, and binge eating [38,53]. Five WHPPs also offered smoking cessation interventions [35,38,42,52,54]. Some WHPPs were organised in the form of small group lectures, others offered cross-site activities, and some also included individual counselling [52]. Many WHPPs offered more than one intervention; whereby in some cases, employees were free to choose the intervention in which they wished to participate, and in others they were assigned a specific intervention.

Company Branches and Employee Positions
The analysis revealed that limited information was reported about the industry and even less about the position of participating employees, but a wide variety of work settings was reported. A total of 12 articles contained no information at all about the company, while 18 articles reported only on the industry or area in which the company operated. Moreover, as only 11 studies were found to report on the specific position of employees within the company, we did not include this factor in the additional analysis. A cluster randomised controlled trial, with departments within companies as the unit.
The intervention was compared with a standard programme consisting of a physical health check with face-to-face advice and personal feedback on a website.  A cross-sectional survey design-the tailored design method

2-years
The research examined the relationships between managerial support for health promotion, participation in wellness programmes, work stress, and positive health behaviours.
Participation in wellness activities, workplace stress, and health behaviours were measured. The public university and bank offered biometric screenings, on-site flu shots, on-site fitness facilities, and classes offered during the workday, health insurance take-up incentives, and wellness coaching services. The wholesaler's wellness programme included biometric screenings, free flu shots, a diabetes prevention programme, coaching for blood pressure, diabetes, weight management, exercise, proper nutrition, and smoking cessation. Biggest Loser and Maintain Not Gain competitions, and provision of healthy snacks and water. The private university's wellness programme included biometric screenings, free Zumba and yoga classes, a free lunch when an employee goes for a walk before or after lunch, and health-related "Lunch and Learn" sessions with information and discounts.   with an internal programme.

At the individual/interpersonal
level, it provided one-to-one, small group, and worksite-wide activities: quitting smoking, healthy eating or nutrition, workplace health and safety, physical activity or exercise. Examples of worksite-wide activities included a physical activity contest, a fruit and vegetable challenge, and large displays that were tailored to each worksite.
A total of 131 companies met the study eligibility criteria; of these, 26 agreed to participate; of 26 worksites, follow-up was completed on 24 sites. The majority were men, approximately half of the workers had no more than a high school education, and the median age range was 35 to 50 years. Thirteen sites were randomised to the intervention condition and 13 to the minimal-intervention control condition.
Recruitedworksite representative agreed to be randomly assigned to the intervention condition.
The survey response rate at the final time was 77% Of physical activity and exercise programs, 19% participation rate for NHIS and HD-SB had 54.4% participation rate. A total of 28% of NHIS respondents and 62% of HD-SB participated in nutrition education.
50.4% of NHIS respondents and 54% of HD-SB workers reported participation in workplace health and safety. 6% of NHIS respondents and 25% of HD-SB workers reported participating in smoking cessation programs.  Employees were assigned to one of three arms. Assignment to a treatment arm versus the nontreatment arm was determined by management. Assignment to an arm among those eligible for treatment was randomised by office. The main dependent measure in the study was HRA completion rate.

weeks
One reminder e-mail was sent each week for 4 weeks.
The study was conducted in a health management company in all 14 branches spread across the country with 15 or more employees, resulting in 1299 eligible employees. The sample was predominantly female (85%), with an average age of 41 years and an average length of service of 2.86 years. The median household income, estimated based on the median income for the employee's postcode, was USD 43,084.
All employees were eligible to receive USD 25 for completing the HRA. Those in the lottery condition were assigned to teams of four to eight people and, conditional on HRA completion, were entered into a lottery with a prize of USD 100 (expected value, USD 25) and a bonus value of an additional USD 25 if 80% of team members participated. Those in the grocery gift certificate condition who completed an HRA received a USD 25 grocery gift certificate. Those in the comparison condition received no additional incentive.

Calculated by author:
49.33% The HRA completion rate was 64% in the lottery group (n = 489), 44% in the grocery voucher group (n = 184), and 40% in the comparison group (n = 626).    The RE-AIM (Reach, Efficacy-Adoption, Implementation, Maintenance) evaluation Using the RE-AIM framework, they summarise the characteristics and findings of selected studies to document the reporting of the intervention's reach, adoption, implementation, and maintenance.

Basic WHPP intervention
The authors reviewed a total of 24 publications from 11 leading journals on the topic of health behaviour.
The median adoption participation rate was 56.5%.
The range of participation rates among eligible workers varied across the studies-from 8% (34) to 97%. Participation rates at the individual worker level: average participation rate of 63.5% across studies.
The authors' own calculation of the participation rate in the WHPP.

Discussion
The conclusions that emerge from our analysis represent a valuable contribution to several aspects of the WHPP research (paragraphs one and two) and practice (paragraph three).
(1) The question of redirecting the research focus in WHPPs should be raised, as programme organisers/facilitators (sponsors, coaches) waste valuable resources on programmes of limited general impact, given that the benefits to the target population and those who would need the interventions most may be (very) small. Considering that the scoping review showed that the focus of WHPP research avoids the burning issue of low participation rates, it is not surprising that the incidence of modern work-related diseases (obesity, hypertension, burnout, etc.) continues to rise despite the abundance of programmes and funding. This raises the broader question of how the effects and impacts of WHPP interventions introduced can be considered positive when only a (relatively) small proportion of workers participate in such programmes. At the very least, the effectiveness and efficiency of such programmes are questionable, as many resources are spent on (very) limited effects, which more often than not benefit those who are already health conscious. (2) Reporting on participation rates in WHPPs should be made more consistent and comparable. In particular, it would be useful to report participation rates for each type of intervention (physical activity, workshops, active breaks, awareness programmes, etc.) by target group or type of incentive, and to summarise participation rates for composite interventions. In short, it would be useful to monitor participation rates by at least the most important factors so as to obtain comparable data on uptake. We also believe that an in-depth meta-analysis focusing on the determinants of participation in WHPPs would be imperative to understand the reasons for such low participation rates. (3) Not to overreach the aim of this study, we need to highlight an issue that is often left unaddressed. While we found that participation rates were reported in a few studies, in the vast majority of articles on WHPP they were not mentioned. Our scoping review thus draws attention to the potential problem of scientific misconduct [57], more specifically a type of reporting bias: "a systematic distortion that arises from the selective disclosure or withholding of information by parties involved in the design, conduct, analysis, or dissemination of a study or research findings" called publication bias [58]. Publication bias occurs when the research findings influence the decision to publish or otherwise disseminate them [59].
Chan and Altman (2005) [60] list many different reasons why research results go unreported. Researchers seem to make the decision to omit certain results for a combination of reasons, such as space limitations in journals, significance of the results, and specific statistical results. To this we could add the pressure of having to publish. We might hypothesise that in the world of academic research with the prevailing "publish or perish" culture, studies with higher participation rates are more likely to be published than those with "disturbingly low" participation rates despite the similar quality of conduct and design. On the other hand, authors might refrain from reporting participation rates if they appear worryingly low in the hope that this issue will not be raised during the review process. Whatever the reason, one possible consequence is that the balance of outcomes tips in favour of positive outcomes [59], which further distances the research focus from what needs to be addressed: low participation rates in WHPPs.

Limitations
The following limitations may have appeared in our scoping review. First, our literature search was limited to the selected electronic article databases, where there is considerable overlap in content. It is also possible that some useful studies were overlooked, as we only focused on English language publications in scientific journals. However, we assume this does not to have a significant impact on the results. Another potential limitation is that many interventions or studies had been conducted in the field, and the results of such studies were either not evaluated or not accurately recorded in the literature.

Conclusions
The main findings of the systematic review can be divided into three sections pertaining to the reporting of participation in the WHPP, the level of reported participation rates with the indication of trends in higher or lower participation rates, and the issue of comparability of reported rates. The proportion of appropriate studies relative to the baseline articles identified is relatively small, with research still primarily concerned with reporting the effects of interventions rather than focusing (specifically) on the circumstances of the uptake. In more than half of the studies reviewed, reported participation rates were relatively low-below 50%. There seems to be no obvious interconnection pattern between the size of the company, i.e., the number of employees involved, and reported participation rates. Higher (financial) incentives tend to be associated with higher participation rates; however, monetary incentives do not guarantee higher participation rates in WHHPs. Lower participation rates are more striking for physical activity interventions than for those related to counselling and training. The types of WHPPs vary, from specific to composite, as do the types of research designs used to measure their impact. A great deal of variability was also found in the delivery of the programmes. Reporting on the industry and workers' position in the company was found to be very sparse. Most importantly, reported participation rates are tied either to different target groups or to the type of intervention or to single points in time in the research. As a result, it is difficult to establish uniform criteria for comparing reported participation rates, as they are difficult to compare directly. Nevertheless, the relationship between participation rates and intervention effectiveness should also be thoroughly explored.
Several implications emerge from these findings. Despite the importance of the WHPP design and its efficacy, research in the field of WHPPs should include analysing the mechanisms needed to improve WHPP participation rates, while reporting on WHPP participation rates should be made more consistent and comparable. An in-depth metaanalysis focusing on the determinants of participation in WHPPs would be imperative to shed light on the reasons for low participation rates. Although our research has only briefly touched on the (mezzo-level) determinants of participation, other possible micro-level (workers), mezzo-level (organisation, WHPPs), and macro-level (legislation, taxation, etc.) determinants should be thoroughly investigated. Second, the scientific community should closely examine the possibility of reporting bias in studies with low WHPP participation rates and, if identified, address it openly. Solutions to this problem are already known but do not appear to have been fully implemented, at least in the case of WHPPs. First, researchers and journal editors should ensure that complete data are made available for all studies, regardless of their values. In addition, as suggested by Richards and Onakpoya (2019) [58], post-study measures should be implemented. Reporting guidelines and other checklists and tools have been developed to assess the risk of reporting bias in studies, including the Cochrane risk of bias tool, GRADE, and ORBIT-II [61].

Conflicts of Interest:
The authors declare that we have no competing interests.