Abstract
BACKGROUND:
Repositioning patients is a frequent task for healthcare workers causing substantial stress to the low back. Patient handling methodologies that reduce low back load should be used. Some studies have observed the effect of bed height on back forces using a limited range of heights. This study details a wider range.
OBJECTIVE:
The aim of this study was to discover an optimal bed height for reducing low back force when boosting a patient.
METHODS:
11 university students and local residents participated by completing a series of boosts with a 91.6 kg research assistant acting as dependent. The bed was adjusted 3% of participant height and 3 boosts were completed at each height which resulted in 8–10 different bed heights depending on the height of the participant. Motion and force data were collected to estimate low back forces via 3DSSPP. Pearson’s R was performed to observe the correlation between caregiver height and low back forces.
RESULTS:
There were significant negative correlations between bed height and low back compression force at L4-L5 (r = –0.676, p = <0.001) and L5-S1 (r = –0.704, p = <0.001). There were no significant correlations with any shear forces.
CONCLUSION:
The highest bed height led to decreased low back compression forces regardless of participant height, but there was not a significant difference in shear forces. Thus, healthcare workers may experience less low back stress with the bed at a higher height. There may be a force tradeoff between the low back and other parts of the body that needs further exploration. Healthcare workers need to be made aware of the implications of adjusting the environment when performing patient handling tasks.
Background
Work-related musculoskeletal injuries are a common problem among healthcare workers and steps are being taken to understand and potentially alleviate this issue [1]. These workers experience musculoskeletal injury rates comparable to construction workers and those in other professions that require heavy lifting [2]. These injuries are due to the high rate of daily lifting and moving of patients. The maximum recommended lifting limit is 156 N at the hands for all healthcare workers, and that is only if the patient is cooperative and the task does not include trunk rotation, excessive bending, or other awkward postures [3]. Attempts have been made to alleviate the impact of musculoskeletal injury on healthcare workers, but the risk remains high for nurses, nursing aids, and other related professions such as occupational therapists, occupational therapy assistants, physical therapists, and physical therapy assistants [2, 4].
The low back is consistently among the most common sites for musculoskeletal injury in healthcare workers [5] specifically the L4-L5 and L5-S1 vertebral levels. These vertebral levels are commonly implicated in low back pain caused by heavy loads, repetitive actions, high forces, and awkward postures [5, 6]. Much ongoing effort has gone into developing methodologies to accurately measure forces at these vertebral levels with varying degrees of success depending on what specifically is being sought with different biomechanical models [7–9]. Force data at these levels have been used to establish safety guidelines for a variety of professions that require moving and lifting heavy objects, including healthcare workers. High forces are defined as low back compression forces of greater than 3,400 N during repetitive tasks and shear forces of 500–700 N for repetitive tasks, depending on the source, while awkward postures have been described as those where the individual is in a bent or twisted posture [3, 11]. In an effort to reduce low back forces and keep healthcare workers safe, these low back force recommendations coupled with the maximum hand force lifting limit of 156 N, is only valid when working with patients who are capable and cooperative and the healthcare worker is able to maintain a neutral trunk posture [12]. If the ideal conditions are violated, lower force limits are indicated. Lifting more than these limits places healthcare workers at risk of high low back forces and consequently musculoskeletal injury. These numbers can potentially be influenced by the height of the hospital bed upon which the patient is lying which can influence the posture of theworker.
Bed height, frequently too low during manual patient handling tasks, is a likely factor in healthcare workers sustaining musculoskeletal injuries [13, 14]. This could be due to poor trunk posture during these tasks which can place undue stress on the low back. Unfortunately, the recommendations for bed height to safely move patients are inconsistent and nonspecific [15–17]. There is consensus that there is increased risk of low back musculoskeletal injury to healthcare workers when they use bed heights that are lower than the hip of the individual performing the task, but other recommendations use the elbow or waist of the worker in addition to the hip [15–17]. These landmarks can be highly variable within and between individuals. What seems to be commonly accepted is that the bed needs to be raised prior to any patient handling task, but there is a lack of consensus and empirical evidence showing the impact of this technique on the body of those performing patient handling tasks.
Most of the studies that take bed height into consideration, including an article by Wiggermann et al. [18], allow the healthcare worker the choice to adjust the bed to his or her desired height, which is good for observing typical practice, but is not a standardized height and does not compare heights [19]. Another study used a single bed height without accounting for the different heights of the healthcare workers [20]. The studies that used bed height as a percentage of caregiver height have been brought into question for using bed heights that are too low for boosting tasks and thus don’t reduce low back forces as much as they could [13, 14]. One study attempted to discover an optimal bed height for passive manual tasks, such as when healthcare workers perform passive range of motion on a patient [21]. The researchers found that when the bed height was raised by the participants the lumbar flexion angle was reduced during patient handling tasks [21]. However, by observing only lumbar angle paired with perceived exertion rather than lumbar force, a complete picture of what is happening in the low back cannot be deduced based only on the results of their study [21]. Therefore, investigating low back force during the related task of boosting a dependent patient up in bed may lead to further insights into what is occurring at the low back using different bed heights and lead to better recommendations for adjusting the bed prior to manual patient handling tasks.
When considering the compression and shear force relationship as uncovered by Murtagh and Rice [14], as the compression force decreases the sagittal shear force increases at the L4-L5 joint. With this tradeoff between forces, it was theorized that the trend would continue as bed height increases. This would identify an optimal bed height for boosting tasks where there is lower shear and lower compressionforce.
Purpose
This project sought to ascertain if there is an optimal bed height as a percentage of participant height to reduce low back forces during the task of boosting a patient up in bed. The researchers hypothesized that when the bed was in the lowest position it would generate the highest amount of compressive force and the least amount of shear force, while the highest bed position would generate the least amount of compression force and the greatest amount of shear force. With this assumption, it was thought that there would be a point where the reduction in compression force would be surpassed by the increase in shear force somewhere between the lowest and the highest bed heights. It was further hypothesized that even at the ideal bed height, hand forces would be higher than the recommended limit.
Methods
This study was a quasi-experimental cross-sectional crossover design with a randomized starting bed position. The independent variable was bed height incrementation. Control variables include a consistent research assistant being boosted as well as a consistent research assistant helping with the boosting task, and the data collection equipment. The dependent variables in this study were L4-L5 and L5-S1 compression and shear forces and calculated hand forces.
This study complied with ethical standards in accord with the Declaration of Helsinki of 1964 and its later amendments. This research was approved by the IRB at Brigham Young University (IRB protocol # F19017). All participants reviewed and signed an informed consent form prior to participating. A convenience sample of 11 university students and local residents was used for this study consisting of five females and six males, but one participant’s data were excluded due to overstepping the bounds of the force plate, which resulted in lost data resulting in data from 10 participants being included in the analysis. This was determined to be a sufficient sample size based on a power analysis using data by Murtagh and Rice comparing compression force at L5-S1 between the low and high bed conditions with an alpha of.05 and a beta of.8. A scale with a built-in height rod was used to procure the height and weight of each participant prior to the boosting tasks. The mean height of the participants was 175.8 cm (Standard Deviation: 10.0 cm) with a range of 157.5–188.0 cm. The mean weight was 72.1 kg (Standard Deviation: 20.8 kg) with a range of 47.6–122.5 kg. The mean age was 28.0 years (Standard Deviation: 4.05 years) with a range of 21–33 years.
A ten-camera Oqus motion capture system (Qualisys, Göteborg, Sweden) was used to collect motion data using a collection rate of 100 Hz. Two in-ground force plates that were flush with the floor were used for the collection of ground reaction forces during the boosting tasks at a collection rate of 400 Hz. These rates were deemed sufficient for the relatively slow motion of boosting a patient up in bed as described elsewhere [22].
Nineteen reflective markers were placed on the body. The markers were placed as recommended by C-Motion with slight modification [23] (see Fig. 1) for the purpose of use in 3DSSPP.

Marker set (C-Motion, 2010).
The participant approached the adjustable height hospital bed (with a range of 55.9 cm to 96.5 cm as measured from the floor to the top of the mattress with a tape measure, the mattress being a 15.2 cm foam mattress common to home use adjustable medical beds) with the bed either at its lowest or highest position as determined by a randomization sequence. The participant was instructed in the manual patient handling task of boosting a patient up in bed and allowed to practice prior to data collection. To complete the boosting task, the participant placed one foot on each force plate initially (see Fig. 2), but was allowed to step laterally as needed while staying within the borders of the force plates. The participant was instructed not to come in bodily contact with the bed as that would invalidate the force measurements. On the count of three the participant and the researcher slid the “patient” approximately 15 cm up in the hospital bed as delineated by two pieces of electrical tape placed 15 cm apart on the fitted sheet. This was done using a Cotton draw sheet. The patient was a 91.6 kg female who acted as a dependent patient for all boosting tasks. A single, trained occupational therapist was on the other side of the hospital bed to assist in making the patient’s movement symmetrical by pulling on the draw sheet at the same rate as the research participant. After three boosts at the same bed height, the bed was raised or lowered 3% of the participant’s height, and the boosts repeated. Three percent was chosen as the interval as previous studies used 5% differences and a higher degree of sensitivity to any changes that are related to bed height was desired [13, 14]. This process continued until the hospital bed was in the opposite extreme compared to its starting position with the boosting task having been performed three times at each bed height. This resulted in different numbers of bed heights for different participants as taller participants had larger incremental advances in bed height than did shorter participants. At times there was a final increment smaller than 3% of participant height in order to reach the extreme bed position and these numbers were also included in the analysis. The tallest participants completed a total of 24 boosts while the shortest participantsperformed 30.

Research lab setup. Note that during data collection there was another researcher on the other side of the bed assisting with the boosting task.
Motion and force plate data were filtered using a low pass Butterworth filter with a 10 Hz cutoff frequency. The moment of peak ground reaction force was isolated and this single time point was used for further analysis. The participant’s body weight was subtracted from the vertical force component of the ground reaction forces and the remaining x, y, and z components were resolved. The resolved force was divided in two and applied to the hands in a method based on work by Dutta et al. [24]. The formula used to account for body weight and resolve the forces was as follows: SQRT(Fx∧2+Fy∧2+(Fz-body weight)∧2) = resultant force. The resultant force vector angles were calculated and used to inform the force angles for 3DSSPP. Custom MATLAB code was used to compute body segment angles as specified by 3DSSPP [25]. This was done by finding the reference line for each specific part of the body and calculating how far the part of the body was from that line in two directions (horizontal and vertical) for each limb segment and three directions (flexion, lateral bend, and axial rotation) for the axial skeleton including the pelvis and inputting those values into the 3DSSPP computer program. 3DSSPP then gave specific compression and shear force information for the L4-L5 and L5-S1 joint forces. The three trials at each height were averaged to alleviate some of the error in the hand force calculationmethod.
The data were analyzed using Pearson’s r to determine correlations between the variables. Compression, frontal plane shear, sagittal plane shear data at L4-L5 and L5-S1 levels and calculated hand forces were compared to percentage of participant height with Pearson’s r and a Holm-Bonferroni correction for multiple comparisons.
Results
After observing plots of the data, it was clear that the expected outcome of a reduction in compression force accompanied by an increase of shear force was not present. There were negative trends or no trends. As displayed in Fig. 3 and Table 1, the results showed a strong negative correlation between bed height as a percentage of caregiver height and low back compression force at both vertebral levels (L4-L5 r = –0.676, p = <0.001, L5-S1 r = –0.704, p = <0.001). This showed that the lowest bed height yielded the highest amount of compression force and the highest bed height yielded the lowest amount of compression force with a steady decline between the two points. With the directional hypothesis that shear forces would increase with an increase in bed height there was no correlation found, as Pearson’s r was negative for all of the correlations suggesting that there may be a decrease in shear forces in addition to the decreases in compression forces and the assumptions of the researchers were incorrect. There was also no significant correlation between bed height and hand force after the Holm-Bonferronicorrection.

Low back and hand forces compared to bed height as a percentage of caregiver height.
Bed height correlations with forces
*Indicates a significant correlation. Pearson correlations between bed height as a percentage of participant height and low back/hand force.
The purpose of this study was to explore low back forces at different bed heights during a common patient handling task and see if there is an optimal bed height as a percentage of caregiver height for reducing those forces. The main result of this study was unexpected. It was hypothesized that there would be a tradeoff of compression forces and shear forces and an optimal bed height could be inferred based on where the transition occurred [12–14]. Thus, the data were expected to show an ideal bed height for the boosting task somewhere between the highest and lowest bed height. However, this was not what the results uncovered.
When observing the separate force components, there was no increase in shear force with the increase in bed height. Rather, there was a decrease in each force category, although not significant for shear forces for reasons stated earlier. That is somewhat contrary to what Murtagh and Rice saw with an increase in L4-L5 sagittal shear force at three different bed heights [14]. Compressive forces showed an obvious negative correlation both at L4-L5 and at L5-S1. Other research using percentage of caregiver height either showed little difference between two heights that were 5% different [13], or that there was a decrease in low back compression force between 3 bed heights up to 51% of caregiver height without exploring higher bed positions [14].
The maximum compression and lateral shear forces at the low back exceeded the recommended limits for these types of tasks when the bed was at its lowest point with a maximum compression value of 7,277 N at the L4-L5 level and 6,855 N at L5-S1, which are almost twice the 3,400 N compression limit for repetitive actions and even exceed the 6,000 N one-time action limit [3]. Maximum shear forces were similar with the maximum Sagittal Plane Shear values being 231 N and 44.5 N and the Frontal Plane Shear values being 601 N and 818 N at L4-L5 and L5-S1 respectively, showing that the Frontal Plane Shear forces are regularly exceeding the 500–700 N repetitive task recommended limit while the Sagittal Plane Shear forces are not [11]. The highest hand force value was 628 N, which is also much higher than the 156 N limit, and even the lowest hand force was above this limit [12]. This highlights the risks with the high levels of low back and hand forces required for this patient handling task.
Biomechanically speaking, choosing the highest bed height makes sense for reducing compressive force at the low back as this puts the vertebral column in a more upright position and decreases the moment arm of the trunk to the low back, which aligns well with the findings of Alperovitch-Najenson [21] who found that less lumbar flexion when engaging in patient handling tasks resulted in a reduction in perceived exertion in healthcare workers. Decreased lumbar flexion paired with decreased L4-L5 and L5-S1 compression forces can have a positive impact on healthcare workers.
Wiggerman et al. [18] investigated low back forces during a boosting task at a self-chosen bed height with healthcare workers acting as participants and found L5/S1 compression forces consistent with the current study. Their study was looking primarily at sheet differences rather than bed height differences, but still serves to validate the current methodology. Vinstrup et al. [26] showed that consistent use of assistive devices during patient handling tasks reduced the intensity of low back pain, possibly due to reduced loads on the low back, but bed height was not a specific factor in this study. What is consistently unobserved in published works is what placing the bed in the highest position does to low back forces. In the current study, having the bed in the highest position resulted in low back forces that were below the recommended limit for repetitive action for some participants during this boosting task. While this was not consistent for all participants, all compression forces were lower at the highest bed height.
These findings also highlight potential practice limitations for many healthcare workers as higher bed heights reduce low back force. Hospital beds are generally adjustable to a height of 91.4–96.5 cm (including the mattress), which may be too low for some caregivers. There is a need for hospital beds to have a wider adjustable range in order to accommodate taller healthcare workers and caregivers.
This study did not observe other joints in the kinetic chain. There could be a negative effect on the shoulder with the increase in bed height which could be problematic as the shoulder is the second most injured body part of healthcare workers after the low back [27]. For this reason, the recommendation to raise the bed height is made cautiously. Interestingly, hand forces remained relatively unchanged across bed heights in this population, which means participants used the same amount of force for the task independent of bed height.
It is important to note that while many healthcare workers use a typical cotton sheet for this type of repositioning task, it is by no means the only option or the preferred option. There are many friction reducing devices available that serve to alleviate the amount of force on caregiver bodies to some degree, and for very heavy patients it may be required to use a mechanical lift to aid with repositioning patients in order to keep lifting forces below the recommended 156 N.
There was a high degree of variability in individual forces. This could be due to the lack of training for the individuals as they were participating in an unfamiliar task. Thus, there were likely technique differences with initiating and following through with the boosting motion which registered even though there was a consistent person acting as the patient. This highlights the importance of training healthcare workers in proper techniques and good communication so they can decrease forces on theirbodies.
This information serves to confirm other research in this area and needs to be validated and disseminated to healthcare workers in the clinic and those in training in order to reduce the overall biomechanical stress on the low back. While this area has been under study for some time, there is new information coming out regularly to support the idea of being more mindful of the environment during patient handling tasks that can positively impact healthcare workers and their bodies (21-22). This information, combined with other recent research in the field, will potentially lead to a decrease in low back injury rates in this population and help these workers stay healthy and prolong their careers. Therefore, this study underscores the importance of using an appropriate bed height for healthcare workers who are involved in patient handling tasks, and training new healthcare workers to do the same[10, 28–30].
Limitations and future research
This study was completed in a lab setting rather than a clinical setting with nonhealthcare workers, which could have led to results atypical of a healthcare worker population. Healthcare workers receive specific training regarding biomechanics during patient handling tasks while the participants for this study received only a perfunctory overview. The training healthcare workers receive could affect how the task is performed and possibly reduce low back forces during the task at any height or change the hand forces involved. Healthcare workers perform these tasks on a daily basis as well, so they have more experience with patient handling and are likely skilled in their chosen methods which could also change the biomechanics of the task. Another limitation is that only one patient handling procedure was analyzed which, while very common, is not the only manual task caregivers are required to perform on the job.
There was only one healthy, young individual who acted as a patient. This was important as a control factor in order to make direct comparisons between forces, but this individual was atypical of a general patient population. Patients tend to be sickly, older, and less cooperative. This research assistant also only represents one height and weight, whereas genuine patients are much more varied and have individual differences that can affect patient handlingtechnique.
There was a small sample size in this study, which could have affected the outcomes of the research or give an incomplete picture of what is actually occurring with low back force during this type of patient handling task.
It would be enlightening to compare low back forces with similar patient handling tasks starting with the bed at 61% of caregiver height and then at even higher levels, such as elbow height and sternal height, to compare low back and shoulder forces to assess the relationship between the two areas of the body. This could inform healthcare workers if there are bed heights that are too high to be considered optimal. Incorporating multiple patient handling techniques and comparing low back forces between them is also warranted. These techniques could include bending the patient’s legs and putting their feet on the bed prior to the boosting task, placing the bed at a decline prior to the boost, or utilizing even more healthcare workers with the boosting task along with friction reducing draw sheets.
When determining the direction of the hypothesis for shear forces the assumption of the researchers was incorrect, which limits how the data can be viewed. In the future it will be possible to better understand the relationship of shear forces to bed height and lead to more accurate hypotheses. Finally, the methodology used for calculating hand forces based on ground reaction forces may lead to some inaccuracies. In the future, this can be alleviated by using hand held force gauges which would capture these forces with a higher degree of accuracy and allow for more precise low back force estimates.
Conclusions
The results from this study suggest that setting a hospital bed to the highest level prior to performing patient handling procedures reduces low back compression force regardless of the height of the caregiver. Heights above this are as yet untested. This simple and relatively quick adjustment can potentially reduce the risk of low back injury among healthcare workers. However, it may be important to find a balance between reducing low back forces and the impact at the shoulder for each healthcare worker. Simple choices that take a little extra time are worth the time in the long run if they reduce the risk of experiencing a musculoskeletal injury. This also highlights an important area of training for new healthcare workers as they begin their careers in their respective professions.
Footnotes
Acknowledgments
The authors would like to thank Lauren Adams and Spencer Peterson for tirelessly assisting with data collection and processing throughout the course of this project. It took wisdom and dedication to complete.
Ethical approval
This research was approved by the institutional review board (IRB) at Brigham Young University (IRB protocol # F19017).
Informed consent
All participants reviewed and signed an informed consent form prior to participation.
Conflict of interest
The authors declare that they have no conflict of interest.
Funding
The authors report no funding.
