Background
Work ability reflects the interaction between the individual’s resources (e.g., health, competence and know-how and physical and mental capacity) and the specific demands of the work task(s) [
1]. Impairments in work ability have been associated with musculoskeletal pain, chronic disease, sickness absence, early retirement and all-cause mortality [
2‐
7]. Moreover, individuals exposed to forceful and awkward working postures and reduced lack of recovery have an elevated risk of musculoskeletal disorders and impaired work ability [
8‐
12].
Healthcare work involves frequent exposure to risk factors for back pain such as sudden and high loadings, including spinal twisting and bending during patient handling [
13,
14]. Although the implementation of manual handling equipment has increased the preventive efforts in the healthcare sector the incidences of musculoskeletal pain remains high [
15]. Accordingly, effective initiatives for preventing this imbalance between exposure and individual capacity from critically impairing work ability among healthcare workers are needed. Moreover, in addition to the emerging global shortage in the healthcare workforce [
16], associations between low work ability and intention to quit the healthcare workforce highlights the importance of increasing (or sustaining) the individual resources of the healthcare worker [
17].
Several intervention studies conducted in different occupational settings have investigated the effect of physical exercise on work ability, however predominantly with little or no effect [
18‐
22]. Conversely, Pohjonen & Ranta was able to demonstrate that 9 months of supervised exercise intervention, twice a week at the workplace, was sufficient in preventing a decline in work ability after 1 and 5 years in female care aides [
21]. Nevertheless, low adherence and high drop-out rates in most of the aforementioned studies highlights the difficulties of implementing effective interventions at the workplace. Consequently, high qualitative randomized controlled trials investigating the effects of physical exercise on work ability are warranted.
Although supervised and group-based intervention protocols seem to enhance exercise adherence compared to home-based exercise interventions [
23‐
25] effective interventions at the workplace can be very costly. Nevertheless, some employees might disfavor exercising with their colleagues at work whereas others find it more motivating to exercise with colleagues and exercise instructors at the workplace. Altogether, a need exists to examine if increasing the healthcare workers individual capacity by means of physical exercise can increase work ability. Secondly, we need to know if physical exercise performed as a low cost home-based intervention is equally effective as a workplace based intervention that invests working hours, equipment and on-site training instructors.
The aim of the present study, therefore, was to investigate the effect of workplace-based versus home-based physical exercise on work ability among healthcare workers. We tested the null-hypothesis that supervised physical exercise at the workplace and home-based exercise will not have different effects on work ability.
Discussion
The present cluster randomized controlled trial demonstrated that 10 weeks of physical exercise at the workplace was more effective of preventing deteriorations in work ability compared with 10 weeks of home-based exercise in female healthcare workers.
The WAI score can be classified into four categories; “poor” (7–27),”moderate” (28–36), “good (37–43) and “excellent” (44–49) work ability [
1]. The baseline average WAI score for our participants was 43.1, which thus could be categorized as “good” or almost “excellent”. This may seem unexpected since almost half of the participants had chronic pain in the lower back, neck and shoulders [
27]. In addition, the participants with chronic pain in the neck and shoulders and lower back did demonstrate a slightly lower but still “good” WAI score of 41.9. The relatively high scores of perceived work ability may be explained by the fact that participants were active at the labor market and worked full-time, reflecting a will to sustain work engagement despite the presence of musculoskeletal pain problems.
We rejected the null-hypothesis based on the significant group by time interaction following the 10-week intervention. The group difference was largely driven by a reduction in the WAI score in HOME. The within-group reduction in WAI may not necessarily have been caused by the home-based exercise intervention per se, but may also include the natural seasonal variation in pain symptoms, sickness absence and general health altogether affecting the WAI. Persson and co-workers observed that subjective health complaints peaked from December until February and were lowest during summer among hospital caretakers [
32]. Furthermore, Takala and co-workers demonstrated a reduction in neck and shoulder pain symptoms from autumn towards spring among female office workers [
33]. In addition, data from the National Health Service (i.e. healthcare personal) in the UK clearly demonstrates seasonal variation in sickness absence during the last 5 years, with peaks in the late autumn/winter months and subsequent reductions towards summer [
34]. The above notion may be linked to seasonal affective disorders, which are characterized by low mood, concentration problems, loss of energy and fatigue which are typically increased during autumn and early winter [
35]. Thus, the present results in WAI could have been influenced by seasonal variation as baseline testing and questionnaire was collected in August and September and follow-up in December.
We analyzed the seven items of WAI separately and observed a between-group difference in item 2 (work ability in relation to the demands of the job) and item 5 (sickness absence during the last year). The workplace-based exercise resulted in higher gains in back extensors muscle strength,[
27] which theoretically may result in lower relative exposure during high-force task as manual patient handling. Lidegaard and co-workers reported lower relative work exposure of the neck muscles in office workers after 10-weeks of strength training [
36]. Thus, improved balance between physical work demands and individual physical capacity in WORK compared with HOME may explain the observed improvement of work ability in relation to the demands of the job (item 2).
The effect of physical exercise on sickness absence has been investigated in numerous studies, however, with conflicting results. Svensson and co-workers demonstrated that a multifactorial 14-month prevention program combining physical training, patient transfer technique and stress management reduces self-reported sickness absence compared to a control group among assistant nursing students [
37]. However, other multifactorial randomized controlled trials including physical exercise have shown little effect on registered sickness absence among cleaners, construction workers and healthcare workers [
19,
38,
39]. Furthermore, a recent Cochrane review concluded that for workers with acute back pain, physical conditioning may not affect sickness absence duration [
40]. Moreover, Andersen and co-workers recently demonstrated that a threshold in pain intensity (above 3 on a 0–9 scale) significantly increases the risk for long-term sickness absence among female healthcare workers [
41]. Since almost half of our participants reported chronic musculoskeletal pain (≥3 on a 0–10 scale ≥ 3 months) in the neck and shoulder regions and lower back the base for reducing pain and sickness absence were present among this population [
27]. Accordingly, along with a reduction in average pain intensity of the neck and shoulder and lower back, workplace-based exercise reduced self-reported sickness absence during the last year (item 5) compared to home-based exercise. Notably, the potential difference in registered and self-reported sickness absence should be taken into account when interpreting the present results.
Numerous studies have investigated the effects of physical exercise on work ability, however, predominantly with little or no effect [
18‐
22]. Low adherence, ranging from 0–2 training sessions per week and high dropout rates may explain the lacking results shown in these studies. Accordingly, Nurminen and co-workers failed to see a change in WAI and sickness absence after performing one year of individually tailored physical exercise (strength training, aerobic training and stretching) once a week among female laundry workers [
20]. However, they concluded that perceived work ability and sickness absence cannot be affected very positively using a single-component exercise intervention and argued that work ability promotion may need a more multifactorial approach [
20]. The WORK intervention could be seen as a multifactorial approach since including daily supervised training- and motivational coaching sessions. The influence of the specific setting, i.e. exercising together with colleagues at work vs exercising alone at home should be considered in a bio-psychosocial aspect. The bio-psychosocial model focuses on the interaction between biological, psychological, and social factors in the neurological perception of pain [
42]. Hence, besides the physiological training effects, the fact that the workplace group performed group-based training- and coaching sessions at the department seemed to have great impact on the social relationship among colleagues and psychological wellbeing of the individual [
27]. Accordingly, the positive interplay between the bio-psychosocial factors may have had an additive effect on the perception of WAI in the workplace group.
The present study design aimed at comparing a low cost home-based intervention modality with an intervention protocol that involves investments in training instructors, coaches, working hours and additional exercise equipment. These interventional differences could potentially have influenced training adherence and training intensity. Especially, the higher adherence in WORK compared with HOME may explain our results. By training 2.2 times ten minutes per week the workplace group prevented deterioration of WAI, whereas a decrease in WAI was observed in the home group that trained on average only one time per week. The use of dedicated instructors throughout the intervention may not only ensure proper training intensity and safe exercise execution but also increase participant adherence, since supervised interventions is known to enhance exercise adherence [
23]. Furthermore, the provision of 5 coaching sessions for the workplace group may have increased their motivation for attending the daily training and thus increased training adherence.
Implementing longer training sessions may potentially have yielded more positive results. However, the reason for choosing only ten minutes was to offer a low duration, yet effective and realistic training program that could be implemented during the busy working day of a healthcare worker. We were informed by the hospitals health and safety board, prior to the study, that in order to increase training adherence the training sessions should be a short as possible. Thus, although shorter training sessions may compromise the physiological training effect the shorter duration may, however, increase training adherence in this specific hospital setting. Nonetheless, Andersen et al. have previously shown that brief daily sessions of 2 min of shoulder resistance training was equally effective as 12 min sessions in reducing neck and shoulder pain among office workers [
43]. However, in order to implement exercises targeting both the neck, shoulder and lower back region we chose to implement 10 min sessions which, when performed twice a week at the workplace, proved not only to be effective in relieving pain intensity but also preventing deterioration of work ability. In line with the adherence and findings observed in WORK, we showed, in a recent study in slaughterhouse workers, that brief sessions (~20 min) of high-intensity strength training performed twice a week at the workplace was effective of preventing a further decline in work ability among workers with chronic musculoskeletal pain [
44]. Moreover, Pohjonen & Ranta demonstrated that 9 months of supervised exercise (~1 h. sessions) twice a week improves physical capacity and prevents the decline in work ability after 1 and 5 years in female care aides [
21]. Accordingly, performing physical exercise at least twice a week may prevent the age-related deterioration of work ability, health and physical capacity among workers with physically demanding work. However, the small-to-moderate between-group effect size shown in this study indicates that the clinical implications of workplace-based exercise on work ability should be handled with caution.
Strength and limitations
As perceived work ability may be influenced by outcome expectations a limitation of this study, and behavioral interventions in general, is that blinding of participants and those administrating the intervention was not possible. However, to minimize this type of bias we informed the participants that neither intervention was known to be superior to the other. Moreover, we included two active intervention groups rather than comparing treatment with a waiting list group [
45,
46]. A strength of the study is the low loss of participants at follow-up and inclusion of drop-outs in the statistical analysis which allowed us to test the actual effect of the interventions.
Competing interests
The authors of the article declare that they have no conflict of interest what so ever. Further, the research has not received any funding or grant from any commercial source.
Authors’ contribution
MDJ and LLA designed and led the study and MDJ, ES, MB and KJ collected the data. MDJ and LLA analyzed the data and all authors were involved in the data interpretation. MDJ drafted the manuscript and all co-authors revised it critically for important intellectual content. All authors have read and approved the final manuscript.