This RCT aimed to investigate the effectiveness of a tailored multicomponent intervention to improve lifting behavior, strength endurance of the lumbar extensors, LBP, and functional impairment caused by back pain for elderly care nurses. Additionally, participants’ adherence to the intervention was assessed.
The results showed positive effects on intervention group nurses’ lifting performance, illustrated by higher lifted weight with concomitant changes in termination criteria. Furthermore, the analysis revealed significant differences between both groups in back pain (after IGs’ participation in ergonomics training). However, the effect on LBP did not remain within the intention-to-treat analysis.
The further evaluation revealed no significant differences at any measurement time point concerning the strength endurance of the lumbar extensors and the functional impairment caused by LBP. As expected, participants’ adherence showed positive ratings.
Regarding the lifting performance and behavior as one common risk factor for increasing LBP, the analysis of the PILE-Test revealed significant improvements in the intervention group after participation in ergonomics training. This effect in the intervention group remained similarly positive after participating in strength training. Additionally, the control group showed comparable improvements after their ergonomics intervention, including the same training contents. In line with Ewert and colleagues (2009), our data support that ergonomics training successfully increases lifting performance for both groups.
Notably, the groups also differed in their quality of movement execution. The intervention group demonstrated a simultaneous increase in terminations due to muscular fatigue in addition to reduced terminations due to dysfunctional posture. We assume a reason might be the increased lifted weight during the test and individuals' low physical capacity (Rasmussen et al., 2013). However, the differences were only present at the second measurement point, indicating that both groups benefited equally from the training. In line with Wollesen and colleagues (2016), ergonomic training reduced dysfunctional lifting behavior successfully. Therefore, we suggest that the program is suitable to reduce a major risk factor of LBP. Moreover, we can summarize that the training content was successfully adapted to the nursing field and promises to be useful for further implementation in other care settings.
In addition, the intervention group increased their Biering-Sørensen-Test time by approximately 23 seconds after participating in the strength training, while test time of the control group decreased. However, these observations failed to be significant. We assume that the results are caused by interindividual differences between the nurses’ strength endurance, which are reflected in the standard deviations (cf. Table 2). Our results align with Stevens and colleagues (2018), even though they measured subjectively perceived strength. The lack of significant group differences can probably be attributed to methodological problems we observed, such as motivation, tolerance of the discomfort of fatiguing muscles or pain, or fear of pain (Mannion et al., 2011). This observation suggests that another test is needed to obtain valid data (e.g., electromyography).
Moreover, the duration of the intervention might have been insufficient to increase strength. The nurses in our study were analyzed when they reached a 70% participation rate, which corresponds to participation in at least nine units over nine weeks. A recent meta-analysis recommends training over 12 to 16 weeks to gain significant improvements in strength. Furthermore, according to the specific training stimulus (Prieske et al., 2019), criteria of load control, stimulus scope, and stimulus density must be considered to generate adaptation effects. In our study, despite dividing the strength training into three phases with progression, the degree of difficulty had to be increasingly adapted individually to the participants. Therefore, the overly high level of difficulty resulted in simplifying the exercises, leading to inadequate progression control. Overall, future strength interventions should be performed for extended periods with a higher stimulus density and include exercises with a lower intensity and slower progression. This highlights the need to address nurses’ time constraints and offer alternate dates for training sessions that could not attend.
The analysis of the LBP within the VAS revealed significant differences between both groups. Participation in the ergonomics training resulted in the same Level of LBP in the intervention group, whereas pain slightly increased in the control group. As a clinically significant change does not occur until a value of 20 mm, the results should be interpreted with caution (Ostelo and Vet, 2005). Nevertheless, the further aggravation of the staff shortage may have affected results in the control group (Michel and Ecarnot, 2020). Unfortunately, this effect did not remain in the intention-to-treat analysis. One might assume that the reason, again, was the heterogeneity of the recorded data, exposed by high standard deviation. However, the participants failed to follow-up were not significantly different from the considered participants concerning baseline characteristics and primary and secondary outcomes studied.
Future studies should verify the pain situation over at least 17-20-weeks, consistent with the literature (Prieske et al., 2019). Nevertheless, we suppose that our tailoring, the individual adaptation of the training content to the conditions of the nursing home facility, and the movement experiences, including body awareness, recognition of dysfunctional movements, and understanding positive and negative work behavior, might affect back pain positively. This highlights the need for interventions, considering the bottom-up approach, taking work-related burdens, wishes, barriers, and the facility’s condition into account (van Hoof et al., 2018).
Surprisingly, the intensity of LBP in the intervention group and control group was relatively low, compared to the high prevalence of LBP (around 50%) in this target group which is, additionally, repeatedly demonstrated in the literature (Rasmussen et al., 2013; Simon et al., 2008; Wollesen et al., 2019). The discrepancy might indicate a lack of ability of the nurses to discriminate between the presence of pain and assessing the pain intensity which is associated with a possible underestimation of the severity of LBP. As a result, nurses reported low functional impairment caused by back pain. The evaluation of the ODI and the classification of the percentages showed a minimal disability in both groups and no significant differences in the study (Mannion et al., 2011). However, the intervention period may not have been long enough to reduce functional impairment and should be considered more closely in future investigations. Overall, the low intensity in LBP and the low functional impairment of all included nurses may also indicate that only those who already live health-consciously and suffer less from pain and functional impairment were motivated to participate in the intervention. For this reason, we recommend conveying education about the importance, the mode of action, and the effects of health-promoting interventions in future research to motivate the non-participating nurses.
Regarding the participation, it is newsworthy to report that the intervention program achieved a positive adherence. The participants of the control group determined the high dropout rate. Within the participants that joined the program regularly, the dropout quote was only 24%. According to our results, the nurses equally accepted and tolerated the intervention. Nurses regularly participated in the intervention despite the barriers, such as time pressure. This highlights the need for the participatory approach in the process and the tailored structure of the training with the individually adapted intensity of the exercises (Wollesen et al., 2016).
Limitations
Next to the study’s strengths, some limitations need to be addressed. The participants took part in the study during working hours. Therefore, time pressure, lack of time, or motivation could have influenced the data assessment. Moreover, the LBP and functional impairment were assessed with self-administered questionnaires and scale. This may have led to a possible underestimation of the severity of LBP and disability due to a lack of ability for discrimination between the presence of pain and the assessment of the pain intensity.
The results were not controlled for the duration of employment, the type of employment (e.g., full time, part-time), or the activities during leisure time which might be a relevant factor for burdens and strains. Moreover, the high level of difficulty in back fitness and the resulting simplification of the exercises could have led to the progression not being sufficiently controlled. Finally, the dropout rate of 37% was high. However, considering the difficulties in implementing a randomized controlled study in elderly care, the number of participants achieved can be considered a success and fulfilled the prior calculated power.