The burden of conveyor belt work in the canteen kitchen: A question of working height?

Abstract

BACKGROUND:

Working in forced postures and standing continuously can be classified as straining the musculoskeletal system.

OBJECTIVE:

Since such postures are frequently used in hospital canteen kitchens, we used kinematic analysis to determine the working postures of canteen kitchen staff.

METHODS:

In this study, the daily work routine of 18 (11 w/7 m) workers of a hospital canteen kitchen (Frankfurt Main/Germany) aged 21–62 years (46±13 years) was examined by means of kinematic analysis (CULEA system; IFA; Sankt Augustin/Germany) and a detailed computerized analysis of the activities performed on-site. Angle values of the head and trunk were evaluated in accordance with ergonomic standards and presented using percentile values (P05-P95). The OWAS method was also employed to capture the proportions of standing, walking and sitting work.

RESULTS:

The kinematic posture analysis showed for all activities on the conveyor belt a tendency towards a dorsally inclined body position: trunk inclination (–7.5° to 0), thoracic spine inclination or a bending forward (–11.3° to 0°) and curvature of the back within the thoracic spine (–15.2° to 0°). In addition, >90% of the “activities on the belt” (46% of the daily working routine) were carried out standing.

CONCLUSION:

The activities on the conveyor belt were characterized by a tendency towards hyperextension of the trunk, possibly due to a too high working environment. Furthermore, an increased burden on body structures while standing can be concluded. From a primary prevention perspective, this increased standing load should be reduced by behavioral and relational prevention measures.

Keywords

Musculoskeletal diseases canteen kitchen kinematic analysis CUELA occupational medicine

1 Introduction

Musculoskeletal disorders (MSDs) are the main cause of lost work days due to the incapacity to work. MSDs affect both employees and employers and are the most expensive occupational diseases in developed countries, such as Germany, as well as in developing countries [1 –3]. Working in forced postures that are adopted for long periods of time with little or no possibility of posture change, repetitive activities, load handling or standing with no possibility of relief through brief sitting or walking, are common risk factors for the development of work-related MSDs [4, 5].

Risk factors such as these are evident when working in a restaurant or canteen kitchen. Shankar & Subramaniam et al. [6, 7] found an increased incidence of unspecific lower back pain in chefs (79.2%), assistant chefs (71.4%) and kitchen assistants (30.0%) using a standardized survey of 114 male employees working in commercial kitchens. However, previous studies have mostly analysed kitchens that prepared meals in the traditional manner. A study by Nanyan et al. [8] showed a significantly increased prevalence of MSDs among employees of canteen kitchens compared to those of traditional or fast food restaurant kitchens. The kitchens of restaurants that serve traditional cuisine exhibit differences in their working methods compared to fast food restaurants or canteen kitchens. The working processes in canteen kitchens focus on portioning patients’ meals on the conveyor belt, preparing dishes, and commissioning and logistics activities [9 –11]. Here, the main tasks for kitchens that work according to such cook-and-freeze principle [9 –11] are portioning patients’ meals on the conveyor belt, and commissioning and logistics activities. In this context, the daily preparation of numerous special menus is one of the central requirements for a German hospital canteen kitchen [9 , 12–16]. These differences also affect the demands on the musculoskeletal system.

The share of conveyor belt work is supposedly much higher in kitchens that work according to the cook-and-freeze principle. The resulting long standing times of workers in these kitchens are cited, in particular, as a physical strain on the musculoskeletal system [17]. In addition, the risk assessment of standing work indicates a clear positive correlation between total time spent per shift and the risk of health consequences [18]. The health consequences of standing for long periods range from local pain and functional disorders to the damage of anatomical structures. Garcia et al. [19], for example, demonstrated that significant fatigue of the muscles in the lower extremities occurs after five hours of standing work. In addition, the working height is crucial for the ergonomics of conveyor belt work. Worktops which are positioned too high cause workers to complain about the upper extremities, as well as the shoulder and neck areas [20], and the lower back [21]. Overall, Pekkarinen et al. [20] showed that the work surfaces were too high for one third of the workers. A kinematic study of supermarket employees when restocking the vegetable counters [22] also provides comparable results. Here, the shelf levels could be identified at which the load on the musculoskeletal system was maximum or minimum when filling shelves by using EMG and capturing the range of motion of trunk, shoulder, elbow, hip, knee and ankle joints. The relationship between working height and the load on the lower back has also been described by Iwakiri al. [23]. In this study, cooks were working in upper body postures that were bent forward during food preparation. This position was described as contributing to the complaints caused by a worktop that was not height-adjustable.

The aim of the current study was, therefore, to determine the daily working postures of employees in a hospital canteen kitchen. The present analysis focuses on the following activity categories: “portioning belt activities”, “logistical activities”, “rinsing belt activities”, “administrative activities” and “other activities”. In particular, working in forced postures, including long periods of standing with a forced posture and static holding work, are taken into account. The kinematic measuring method to be used in this study (CUELA system) has already been successfully used in other professional groups, such as nurses [24], educators [25], employees in emergency services [26], dentists [28] and orthodontists [27] to analyze working postures. Furthermore, the data analysis not only clarifies the joint angles of the individual body segments, but also provides basic information about the posture in which work is performed per se (e.g. walking, sitting, standing). This is provided using the OWAS method [28].

2 Materials and methods

2.1 Participants

Eighteen volunteers (11 w/7 m) aged between 21 and 62 years, with an average age of 46±13 years, participated in the study.

All participants were employees of the hospital canteen kitchen of the patient care department of the University Hospital of the Goethe University Frankfurt am Main (Germany) and worked in the areas of the scullery, the collecting of ward supplies, the office, or the portioning of meals at the conveyor belt. All kitchen employees are assigned by the head at which position they have to work for a certain time. Basically, a rotation of the workplace is the case after 3 h at the latest. However, this time depends on the current work situation and the number of workers. It should be noted that all employees were able to change the work area in which they worked during their working day. Rotations of the workstations must be carried out in particular due to the fact that the portioning belt is only running three times a day for 1–2.5 hours.

Exclusion criteria included current injuries to the musculoskeletal system within the last year, genetic muscle diseases and severely restrictive malpositions or functional limitations of the spine or extremities. As a result, 18 of the 22 employees of the canteen kitchen were able to participate in the study.

An approved ethics application from the Goethe University Frankfurt am Main (No. 46/17) was obtained. All participants gave consent to participate.

2.2 Measuring systems CUELA

The personal CUELA measuring system (IFA; Sankt Augustin/Germany), developed by the Institute for Occupational Safety and Health of the German Social Accident Insurance, provides position and angle information via sensors (acceleration sensors [ADXL 103/203] which measure the acceleration, gyroscopes [muRata ENC-03R] for the head, arms, legs and back which measure the angular velocity so that conclusions can be drawn about the position in space; inclinometers additionally map the tilt of the upper body and a potentiometer [Contelect] for measuring back torsion). In this way, a kinematic reconstruction of movement is created and allows a differentiation between standing, walking and sitting postures. The extremity sensors were fixed on both arms and forearms and both thighs and legs. The extremity sensors were attached directly to the skin and were fixed with elastic adhesive plasters (Fixomull, BSN medical GmbH, Hamburg, Germany). The potentiometer and gyroscopes for the thoracic and lumbar spine area were attached to a pelvic belt and a back vest both were connected with a telescopic rod to capture the rotation. In addition, the participants wore a headband with sewn-on sensors [29 –31] to measure head movements. All sensors had a resolution of 0.1° or better. To assess validity, comparison was made with biomechanical model calculations as well as with angular values obtained from VICON marker coordinates recorded in parallel. These showed good and reproducible results with a deviation between the two measurement systems of <10% of the comparable angle measurements of knee, hip and upper body [32]. A mobile data storage unit, which was carried together with batteries in a belt bag on the participant, stored the information recorded at a sampling frequency of 50 Hz. All sensors were connected to the storage unit.

2.3 Objective activity analysis

In order to enable a reliable allocation of the kinematic data collected by the CUELA system to the activities performed in a commercial kitchen (such as positioning or cleaning), a simultaneous activity analysis was performed. The objective activity analysis is needed so that all joint angles can be accurately assigned to the respective activity and so that the OWAS tool can be used appropriately. This meant that the respondent was accompanied throughout the entire period of the study and the activities were recorded in real time using a portable PC (UMPC, Samsung Q1, Samsung Electronics GmbH, Schwalbach, Germany) and a computer program specially developed for objective activity analysis [33 –36]. For this purpose, the compilation of selectable activities was adopted in advance to the work spectrum of canteen kitchen employees. The same participants were observed beforehand and it was noted which activities they performed. The software of the objective activity analysis was programmed in such a way that afterwards these activities could be assigned exactly to the body postures that were recorded by means of the CUELA system.

2.4 Experimental protocol

For data acquisition, all participants were equipped with a back vest, pelvic belt, extremity sensors and headband before the start of the shift. The back vest and pelvic belt were connected by a telescopic rod so the rotations were fed to the potentiometer. The extremity sensors were attached directly to the skin with elastic adhesive plasters (Fixomull, BSN medical GmbH, Hamburg). In addition, due to frequent temperature changes between the commission room, cold rooms, band room and scullery, it was sometimes necessary to wear a cold protection jacket. This was the main reason why the sensors of the extremities had to be worn underneath the clothing. This was also worn over the pelvic belt and back vest. Parallel to the recording, an observer accompanied the participants and documented each activity with the help of the activity analysis on the portable handheld computer throughout a complete shift, of an average of eight hours, on a randomized chosen working day. For randomization, a day was drawn from the possible shifts for the participant.

2.5 Evaluation

A prerequisite for the analysis of the working posture is the assignment of the postures adopted to the activities carried out. The WIDAAN software enables the postures recorded by the CUELA measuring system in the form of joint angles to be linked to the objective activity analysis recorded simultaneously. The representation of the distribution of the assumed joint angles is achieved by means of percentile values (P05-P95). The percentile 05 describes the joint angle value which is undershot by 5% of all measured angle values at this joint and exceeded by 95%.

For postural movement components within the sagittal plane, ventral was defined as the reference direction, unless otherwise specified. Angular values with a positive sign were, therefore, in the ventral direction, whilst angular values with a negative sign were in the dorsal direction. Henceforth, the addition “anterior” is largely omitted for the specifications of the thoracic spine inclination, back curvature and trunk inclination. For postural movement components within the frontal plane to the right was defined as the reference direction. Accordingly, angles with a positive sign were to the right of the median sagittal plane, while values with a negative sign were to the left of the median sagittal plane. The addition “to the right” is largely omitted for the specifications of the thoracic spine lateral inclination, lumbar spine lateral inclination, trunk lateral inclination and back torsion. Torsion is a movement within a horizontal plane that rotates around the longitudinal axis.

The assessment of joint angles was carried out according to ergonomic aspects. This was based on the deviation from the neutral-zero position and a division of the angles into a neutral, medium/moderate and final/unfavorable angle range using a traffic light system [36]. The general standards ISO 11226 and DIN EN 1005-4 were used as the bases for this assessment [37, 38].

In addition, the OWAS method was used to examine the time proportion of certain postures (e.g., sitting, walking, standing) during a performed activity [28 , 37–39]. This allows conclusions to be drawn about the resulting burden on the musculoskeletal system [40, 41].

Furthermore, the interindividual variance of movement with which a certain activity was performed must be taken into account. This is illustrated by means of the modified interquantile range (mIR = [(P50-P05)+(P95-P50)]/2).

3 Results

Following the acquisition of the data, an examination of the temporal distribution of the individual activities, in addition to the movement profiles determined by the CUELA system, could be made.

3.1 Time distribution of the activities

A data set totaling 3995.56 minutes (66.59 hours) was obtained for the analysis, minus irrelevant measurements such as breaks or toilet visits and incorrect measurements, such as disconnected sensors. Apart from toilet trips (1-2 times/measurement), these irrelevant measurements occurred quite rarely.

The category “portioning belt activities” accounted for the largest share of time with 1548.31 min (39%). A similarly large field of tasks was formed by “logistical activities” with 34% (1367.82 min). Almost a third of the time was spent on “washing up” (263.07 min) and “administrative activities” (282.42 min), each accounting for 7% of the time, while “other activities” accounted for 13% of the time. If all the activities carried out on the conveyor belt (“portioning belt activities” and “rinsing belt activities”) were combined, this would correspond to 46% of the working time (1811.38 min).

After analysis using the OWAS method (Table 1), it was determined that more than 95% of all four activities in the category “portioning belt activities” were carried out standing (97.1–99.3%). The remaining proportion was allocated to walking (0.6–2.8%). Most of the work was performed in a neutral back position, i.e., with a straight back (79.2–93.7%). Only 14.1% and 20.4% of the activities “portioning” and “opening packaging”, respectively, were carried out with a twisted or sideways inclined back. “Loading/unloading dishwasher” was the only activity with a measurable amount of time spent on in the area of “rinsing belt activities” which, according to the OWAS analysis, showed the percentage of standing work as 92.1% (Table 1). The category “walking” comprises 6.2% and the category “sitting” 0.2% for this activity. The largest proportion of work was performed with a straight back (88.2%), followed by twisted or sideways inclined backs with 10.6%. Only 1.1% of the work was performed with an inclined back. The combination of bent and sideways inclined or twisted back was negligible (0.1%).

Table 1
OWAS evaluation of the “portioning belt activities” and the “rinsing belt activities”. The values represent the percentage of time spent in a certain posture while doing a certain activity

Activities Sitting (%) Standing (%) Walking (%) Straight back (%) Forward inclined back (%) Twisted or laterally inclined back (%) Inclined forward and twisted/ laterally inclined (%)

Arranging 0 97.2 2 91.6 1.1 7.1 0.2

Final inspection 0 99.3 0.6 93.7 0 6.3 0

Portioning 0 97.1 2.3 84.8 0.8 14.1 0.3

Opening packaging 0 97.1 2.8 79.2 0.3 20.4 0.1

Loading/unloading dishwasher 0.2 92.1 6.2 88.2 1.1 10.6 0.1

Activities	Sitting (%)	Standing (%)	Walking (%)	Straight back (%)	Forward inclined back (%)	Twisted or laterally inclined back (%)	Inclined forward and twisted/ laterally inclined (%)
Arranging	0	97.2	2	91.6	1.1	7.1	0.2
Final inspection	0	99.3	0.6	93.7	0	6.3	0
Portioning	0	97.1	2.3	84.8	0.8	14.1	0.3
Opening packaging	0	97.1	2.8	79.2	0.3	20.4	0.1
Loading/unloading dishwasher	0.2	92.1	6.2	88.2	1.1	10.6	0.1

For the “portioning belt activities” in the head and neck areas, slight tendencies towards the posterior for the movement components of the sagittal plane were apparent, as shown in Table 2. For example, the P05 of the head tilted to the front (HT_f) contained negative angle values for all activities; these were classified as “unfavorable” according to the coding (–4.8° to –13°, mIR: 11.95° to 19.8°). For the neck curvature to the front (NC_f), the “arranging” activity showed negative values not only for P05 (–15.7° to –8°, mIR: 14.1° to 18.6°) but also for P25 “arranging” with –4.8°. In contrast, the P95 of the “final inspection” activity revealed moderate, forward inclining values for both head tilted to the front and neck curvature to the front. The remaining percentile values were between 3.1° and 19.8° for HT_f and 0.9° and 23° for NC_f.

Table 2

“Portioning belt activities”: duration of the individual activities; percentile values (P05, P25, P50, P75, P95); its classification into neutral(green), moderate (yellow), and unfavorable (red) postures in accordance with the DIN EN 1005-4/ ISO 11226 standard and the modified interquantile range (mIR) in the head and neck areas

For the lateral movement components, the percentile values at the edges were mostly in the unfavorable range, thus, for the activity “arranging” the angle value for the P05 of the head tilted to the right (HT_r) was –9.6° (mIR: 10.2°) in the neutral, left area and –13.9° (mIR: 14.1°) in the unfavorable left area. The angle values of the P05 for the neck curvature to the right (NC_r) were also in the unfavorable left range (–12.5° to –11.4°). The P95 of HT_r and NC_r showed a mirror image result with values in the unfavorable right range (HT_r: 10.8° to 14.3°; NC_r: 12.3° to 13.7°). However, all percentile values of P25, P50 and P75 were found to lie in the neutral range between –10° and 10° (HT_r: –3° to 6.7°; NC_r: –3.4° to 6.9°).

In the back area (Table 3), tendencies towards the posterior were particularly noticeable for the movement components in the sagittal plane. This is particularly obvious in the “final inspection” activity. In contrast, only P05 had a negative value for the thoracic spine inclination to the front (TSI_f) during “arranging” (–3° mIR: 11.65°). In the case of P95, the value was even in the moderate range when bending anteriorly (20.3°). In the “final inspection” activity, however, the values of P05-P75 were in the unfavorable range (–9.9° to –1.1°; mIR: 7.4°). The P25, P50 and P75 of “arranging” (2.1° to 9.9°) and P95 “final inspection” (4.9°) were found to be in the neutral range. A similar picture can be seen with the back curvature to the front (BC_f) and the inclination of the trunk to the front (TI_f). In the case of BC_f, the angle values of P05, P25 and P50 for both activities were in the unfavorable range (–7.3° to –0.4°; mIR: 6.05° to 6.5°), while the P75 and P95 angle values were in the neutral range for both activities. However, the “arranging” and “final inspection” operations differ in the case of TI_f; while only P05 was in the unfavorable range for the “arranging” activity (–0.8°; mIR: 9.35°) (P25-P95:3.1° to 17.9°), P05-P75 of the “final inspection” had negative values (–7.5° to –1°; mIR: 5.4°) which were assigned to the unfavorable range. The P95 value for the “final inspection” was in the neutral range at 3.3°. The lateral movement components were rather inconspicuous. The thoracic spine inclination to the right (TSI_ r) showed values in the neutral range for almost all percentiles (–9.8° to 7.5°; mIR: 8.05° to 10.3°); the only exception to this was the P95 for “arranging” which was in the moderate range with 10.8°. For the back curvature to the right (BC_r) and the inclination of the trunk to the right (TI_r), all values without exception fell in the neutral range (BC_r: –7.2° to 14°; mIR: 7.95° to 8.45°; TI_r: –9.3° to 8.1°; mIR: 5.75° to 7.9°). In the case of back torsion to the right (BT_r), moderate values were observed at the ends (P05 “arranging” –12° and P95 “final inspection” 12.8°, mIR: 9.75° to 11.2°). The angular values of the remaining percentiles fell in the neutral range (–9.6° to 7.5°).

Table 3

“Portioning belt activities”: duration of the individual activities; percentile values (P05, P25, P50, P75, P95); its classification into neutral (green), moderate (yellow), and unfavorable (red) postures in accordance with the DIN EN 1005-4/ ISO 11226 standard and the modified interquantile range (mIR) in the area of the back

For the “rinsing belt activities” in the head and neck areas, unfavorable and moderate values were found only in the percentiles P05 and P95 (Table 4). The centrally located percentiles P25, P50, P75 were all in the respective neutral range. For the movement components in the sagittal plane, i.e., HT_f and NC_f, the values fell between 3.2° and 20.8° (mIR: 19.65° to 21.15°). For the movement components within the frontal plane, the values were between –4.7° for HT_r and 9.5° for NC_r (mIR: 12.8° to 16.55°). The P95 of the head tilt and neck curvature were in the moderate range (32.5° and 29.9°, respectively), while the P05 of both angular dimensions were in the unfavorable range at –9.8° and –9.4°, respectively. The lateral movement components, HT_r and NC_r, were in the left unfavorable range for the values of P05 (–15.8° and –12.3°, respectively) and in the right unfavorable range for the values of P95 (17.3° and 13.3°, respectively). They did not show any significant tendency towards one side.

In the back region (Table 4), the unfavorable values were significantly more frequent for the postural components oriented ventrally or dorsally when compared to the lateral postural components.

Table 4

“Rinsing belt activities”: duration of the individual activities; percentile values (P05, P25, P50, P75, P95); its classification into neutral(green), moderate (yellow), and unfavorable (red) postures in accordance with the DIN EN 1005-4/ ISO 11226 standard and the modified interquantile range (mIR)

Thus, the P05, P25 and P50 of the thoracic spine inclination to the front were all in the negative range (–2.7° to –11.3°, mIR: 15.1°), which corresponds to a posterior tilt. The thoracic spine lateral inclination to the right showed a balanced image with P25, P50 and P75 values being in the neutral range (–4.9° to 3.7°) while the P05 and P95 values were in the moderate range (–15.7° to 12.8°). The backward curvature to the front showed values in the unfavorable range (–15.2° to –4.7°, mIR: 9.85°), even in the percentiles P05- P75; only P95 was in the neutral range with 4.5°. Similar, but less pronounced observations were also evident in TI_f; here, P05 and P25 had values in the unfavorable range (–5.4° and –1.7°, mIR: 11.65°), while P50, P75 and P95 varied between 1.3° and 17.9° in the neutral range. Neither BC_r nor TI_r had values in the unfavorable range; here, nearly all percentiles were in the neutral range, between –9° and 9.5° (BC_r: mIR 10.45°; TI_r: mIR 9.55°), with the exceptions being P05 for BC_r (–11.4°) and P95 of TI_r (10.1°) with moderate values. The back torsion to the right showed no prevalence towards either the right or the left body side. Thus, the values of P25, P50 and P75 were in the neutral range (–6.8° to 3.2°). Five percent of all collected data were below the P05 of –15.7° and 5% above the P95 of 10.6°, both of which were in the moderate ranges.

3.2 Static percentages of postures performed

The evaluation of static postures of duration > 4 s revealed that for all activities on the conveyor belt, no static postures outside the neutral posture either in the neck or trunk region could be recorded.

4 Discussion

The time distribution of the individual activities in this study of a canteen kitchen that operates according to the “cook - and - freeze” system, showed that a large proportion of the total working time (46%) was accounted for by activities on the conveyor belt. The postures evaluated, taking into account the above-mentioned DIN standards, were considered and classified using known risk factors for the musculoskeletal system during the conveyor belt work [29 , 42–44]. Almost all percentiles of the upper body movements in the sagittal plane ranged from hyperextension to the neutral range (TI_f: –7.5° to 17.9°, mIR: 5.4° to 11.65°; BC_f: –15.2° to 10.5°, mIR: 3.7° to 9.85°; TSI_f: –11.3° to 21.2°, mIR:6.9° to 15.1°). Moderate flexion angles of TSI_f in P95 can be considered to be measurement error. Therefore, it can be concluded that there were no remarkable unfavorable flexion movements within the upper body, although a considerable proportion of the total time was spent in hyperextension. From this it can be concluded that the working height of the conveyor belt does indeed have an influence on the flexion and extension in the sagittal plane of the kitchen workers. Iwakiri et al. [23] showed in their study, that unfavorable postures in the sagittal plane of the upper body could be reduced by changing the working height. Furthermore, Pekkarinen et al. [20] investigated the working height of workers in traditional canteen kitchens and found that the work surfaces were too high for one third of the workers. The increased hyperextension in the entire upper body in the present study would suggest that the current conveyor belt height was too high for the workers. If necessary, adjusting the height of the conveyor belt by using pedestals or an inclined conveyor belt with individually selectable position at the belt may lead to an improvement in workplace ergonomics and, hence, back posture.

If one compares the present kinematic posture analysis of employees of a canteen kitchen with the CUELA activity analyses of other occupational groups, such as nurses [45], educators [25] or rescue service employees [26], it is noticeable that in this case significantly fewer unfavorable postures were recorded and that the unfavorable postures, compared with other occupational groups, do not consist of bending forward but, rather, a tendency to bend backward. In a study of the working postures of nurses by Freitag et al. [45] using the CUELA system, frequent trunk bending was shown to be particularly associated with patient contact.

In contrast to other activities [27 , 44–47], no increased ergonomic risk due to excessive forward bending of the upper body was observed for activities in the canteen kitchens. This could be due to the fact that the activities (arranging, final inspection, portioning, opening packaging and loading/unloading dishwasher) were primarily performed at an unadjustable working height. As already elaborated, the working height used in this study was deemed too high, so that a measurable duration of loading was observed in back hyperextension.

The OWAS analysis of the areas of activity revealed the percentage distributions for the divisions of “standing”, “walking” and “sitting” activities. The “portioning belt activities” were carried out for between 97.1 to 99.3% of the time while standing. In comparison, 92.1% of the “rinsing belt activities” were carried out in a standing position. Whether an activity can be performed standing or sitting depends largely on the working environment; in this study, the work on the portioning belt or rinsing belt was considered. The working posture was thus determined for the conveyor belt and adjacent equipment (in the case of the hospital canteen kitchen, for example, this consisted of containers with stored menu components, cutlery or trays), however, the option of carrying out the activities while seated would scarcely be practicable in this environment. Unsurprisingly, using the OWAS analysis, it was found that the proportion of work carried out in a sitting position was between 0–0.2% for all activities on the conveyor belt. In a meta-analysis by Prince et al. [48], the physical activities of different occupational groups were compared; the average share of time spent in a sitting position in the total working time for workers with physical work was determined to be 46.9%, which was below the overall average of 58.8%. Therefore, it can be clearly seen that the 0–0.2% recorded for kitchen workers in this study is far below the average for physical workers. It should be borne in mind that working on the conveyor belt accounted for only 46% of the total working time of a commercial kitchen worker examined in this study; in comparison, office workers sit for much longer, on average, in their work, accounting for 66.1% of their total working time. In a study by the Federal Institute for Occupational Safety and Health [5] on the health effects of continuous standing work, both the effects on the venous vascular system and, in connection with this, the cardiovascular system, and the direct effect on the musculoskeletal system were considered. In accordance with the instructions for the ergonomic design of continuous standing work by the Federal Committee for Occupational Safety and Safety Engineering [18], the risk associated with continuous standing can be divided into four risk areas ranging from low to high standing stress. Varnai et al. [49] assigned the standing work of hospital canteen employees to risk area 3, with a significantly increased load on the muscular skeletal system while standing in relation to all areas of activity in an objective activity analysis. In addition to external factors, the degree of curvature in the lower back also influences the development of lower back pain [50].

In the present study, the analysis of the static postures with a duration of > 4 s outside the neutral body position revealed no relevant static posture components for all the activities on the conveyor belt outside the neutral position in the trunk and head and neck areas. Furthermore, the subjective observation supports the established assumption that during conveyor belt work, in particular, repetitive activities are performed without assuming a static posture in the non-neutral area. In further analytical steps, an analysis of these static posture components on the one hand in the neutral angle ranges and on the other hand in all angle ranges would, therefore, be recommended in order to gain a more detailed insight into static holding work. In this context, the use of another ergonomic evaluation method, such as RULA, might be interesting to include in the analysis [41, 51].

Taking into account, among other things, the risk assessment for conveyor belt work of the German statutory accident insurance [17], different risk factors for the occurrence of work-related MSD could be identified in relation to canteen kitchen employees. In addition to working in incorrect or forced postures (e.g., prolonged standing), other factors which may contribute to MSD include psychosocial factors, repetitive activities and load handling [4 , 52–59]. In addition to the pure posture analysis for work on the conveyor belt, it is, therefore, necessary to consider the other risk factors listed in order to estimate the total load on the musculoskeletal system even if these exceed a purely kinematic consideration [59]. Silverstein et al. [60] described four different risk groups for the occurrence of MSD due to cumulative trauma during repetitive activities. Here, as well as in the work of Kilbom [54 , 61] and Moore & Garg [62], frequency of the executed movements is a central factor in assessing the risk. A further investigation of repetitive activities in conveyor belt work in a canteen kitchen would be a useful addition to this study. Manual load handling also plays a major role. The “Instructions for the assessment of pulling and pushing” drafted by Jürgens et al. [56] for the “Exclusion of the Federal States for Occupational Safety and Safety Engineering” (LASI) includes the following guiding characteristics in addition to posture: duration/frequency, mass to be moved/industrial truck, positioning accuracy/movement speed and execution conditions.

The CUELA system enables an exact recording and evaluation of posture, whereas the guidance feature method is based on observation. All sensors had a resolution of 0.1° or better. To assess validity, comparison was made with biomechanical model calculations as well as with angular values obtained from VICON marker coordinates recorded in parallel. These showed good and reproducible results with a deviation between the two measurement systems of <10% of the comparable angle measurements of knee, hip and upper body [32]. However, in order to make a conclusive statement about the burden on the musculoskeletal system caused by the given working conditions, it is necessary to consider the other assessment criteria mentioned above.

On the basis of further investigation which takes into account the individually adjusted working height (e.g. by means of platforms) the assumption could be confirmed or disproved that the number or duration of the postures recorded can be reduced in the unfavorable area (i.e., posture inclined backwards). In the future, it should also be analyzed whether factors such as gender, age or body height influence the working method (e.g. in terms of body posture or performance of specified tasks) on the conveyor belt. In order to be able to estimate the extent of the ergonomic, biomechanical strain on the musculoskeletal system caused by the postures inclined backwards and which may lead to mechanical deformation of joints, tissues, tendons and ligaments, the analysis should include besides kinematic data but kinetic data as well. Without an assessment of the mechanical loads on the body via kinetic data, an accurate assessment of injury risk is not possible.

5 Conclusion

In the present kinematic posture analysis, a tendency for hyperextension of the upper body can be observed for the thoracic spine inclination to the front, back curvature to the front and back curvature to the front for activities on the conveyor belt. It was assumed that this was caused by a working environment that was physically too high in this case. Furthermore, an increased burden on the musculoskeletal system while standing can be proven. The “activities on the conveyor belt” account for 46% of the daily work routine and are carried out for more than 90% of the time while standing. Intermediate relief, e.g., by means of standing aids, is not possible due to the activities performed; this results in a significantly increased burden on the musculoskeletal system while standing for the entire working day. Based on these results, it can be concluded that the kinematic movement analysis via CUELA, in combination with the objective activity analysis, is a measurement procedure that is suitable for evaluating the canteen kitchen working environment and subsequently deriving behavioural preventive measures.

Ethics statement

This study was approved by the Ethics Committee of the Department of Medicine of the University Hospital of the Goethe University Frankfurt am Main (Number: 219/14). All participants signed an informed consent form to participate prior to participation. Minors were excluded as participants in this study.

Availability of data and materials

The datasets supporting the conclusions of this article are included within the article.

Conflict of interest

The authors declare that they have no competing interests.

Funding

No funding was obtained for this study.

Author contributions

DO, KK, WO, FH, LM, HA and DAG made substantial contributions to the conception and design of the manuscript. DO, KK and DAG made substantial contributions to the construction of the measurement protocol, and HA and DO have been involved in the statistical data analysis.

Footnotes

Acknowledgments

Not applicable.

References

Zoike

, Bungard

, Ganske

, Grothaus

, Hertle

, Tewes

, et al., Zukunft der Arbeit. BKK Gesundheitsreport. 2011.

Schmidt

, Raspe

, Pfingsten

, Hasenbring

, Basler

, Eich

, et al., Back pain in the German adult population: prevalence, severity, and sociodemographic correlates in a multiregional survey, Spine 2007;32(18):2005–11PubMed PMID: 17700449.

Versicherung

DGU

. Belastungen für Rücken und Gelenke - was geht mich das an? DGUV Information 208-033 2016.

Trippler

Belastungen des Muskell-Sekeltt-Systems. BGHM Aktuell. 2013;April/Mai 2013.

(BAuA) BfAuA. Stehend K.O.? –Wenn Arbeit durchgestanden werden muss. . . . 2008.

Shankar

, Shanmugam

, Srinivasan

, Workplace factors and prevalence of low back pain among male commercial kitchen workers, Journal of Back and Musculoskeletal Rehabilitation 2015;28(3):481–8.

Subramaniam

, Murugesan

, Investigation of work-related musculoskeletal disorders among male kitchen workers in South India, International journal of occupational safety and ergonomics: JOSE 2015;21(4):524–31. PubMed PMID 26694004.

Nanyan

, Ben Charrada

, Effect of restaurant types on compensation claims for work-related musculoskeletal disorders in France, International journal of occupational safety and ergonomics: JOSE 2018;24(2):324–8. PubMed PMID 27935746.

Risikobewertung” BfHygienemanagement in Krankenhausküchen –Maßnahmen zur Verhütung und Aufklärung lebensmittelbedingter Krankheitsausbrüche 2007.

10.

Blum

, Lehmann

, Trends in der Krankenhausküche, das Krankenhaus 2014;106(2014):04.

11.

Orsag

, Kurz

, Gestaltungsmöglichkeiten und Outsourcingoptionen für Verpflegungssysteme im Krankenhaus, das Krankenhaus. 2015;08.

12.

Serzisko

Verpflegungskatalog des Universitätsklinikums Jena 2016;.

13.

Institut

, Anforderungen an die Hygiene bei der Lebensmittelversorgung und ihre Qualität, Epidemiologisches Bulletin 2006;. Epidemiologisches Bulletin Nr29: 228 f .

14.

Gatternig

, Widhalm

, Bericht & Report: Speisenversorgung im Krankenhaus, Journal für Ernährungsmedizin 2013;15(2):24–55.

15.

von Eiff

Speisenversorgung im Krankenhaus: Marketing- und Kosteneffekte durch Prozessund Qualitätsmanagement Ernährungs Umschau. 2012;aktualisierter Sonderdruck 2/2012.

16.

Verpflegungsmanagement GBuBASuLB. Verpflegungskatalog des Universitätsklinikums Jena. 2016.

17.

DGUV, Metal BHu. Fließbandarbeit Gefährdungsbeurteilung u. Maßnahmen bei physischen und psychischen Belastungen. DGUV-Information. 2015;11/ 2015.

18.

(LASI) LfAuS. Bewegungsergonomische Gestaltung von andauernder Steharbeit - Eine Handlungsanleitung zur Beurteilung der Arbeitsbedingungen. 2009;LV 50.

19.

Garcia

, Läubli

, Martin

, Long-Term Muscle Fatigue After Standing Work, HUMAN FACTORS 2015;57(7):1162–73.

20.

Pekkarinen

, Anttonen

, The effect of working height on the loading of the muscular and skeletal systems in the kitchens of workplace canteens, Appl Ergon 1988;19(4):306–8.PubMed PMID: Epub /12/01. eng.

21.

Nagasu

, Sakai

, Ito

, Tomita

, Temmyo

, Ueno

, et al., Prevalence and risk factors for low back pain among professional cooks working in school lunch services, BMC Public Health. 2007;7:171–8.PubMed PMID: Pubmed Central PMCID: PMC070. Epub /07/26. eng.

22.

Silvetti

, Mari

, Ranavolo

, Forzano

, Iavicoli

, Conte

, et al., Kinematic and electromyographic assessment of manual handling on a supermarket green- grocery shelf, Work 2014;51(2):7:261–71.PubMed PMID: Epub /06/26. eng.

23.

Iwakiri

, Kunisue

, Sotoyama

, Udo

, Postural support by a standing aid alleviating subjective discomfort among cooks in a forward-bent posture during food preparation, J Occup Health.-. PubMed PMID 2008;50(1), :57–62. PubMed PMID 18285645.

24.

Freitag

, Fincke-Junod

, Seddouki

, Dulon

, Hermanns

, Kersten

, et al., Frequent bending–an underestimated burden in nursing professions, Ann Occup Hyg 2012;56(6):697–707. PubMed PMID 22356807.

25.

Sinn-Behrendt

, Sica

, Bopp

, Bruder

, Brehmen

, Groneberg

, et al., IFA Report 2/2015: Projekt ErgoKiTa –Prävention vonMuskel-Skelett-Belastungen bei Erzieherinnen und Erziehern in Kindertageseinrichtungen: Deutsche Gesetzliche Unfallversicherung e. V. (DGUV) 2015.

26.

Prairiea

, Corbeil

, Paramedics on the job: Dynamic trunk motion assessment at the workplace, Applied Ergonomics 2014;45(4).

27.

Ohlendorf

, Erbe

, Hauck

, Nowak

, Hermanns

, Ditchen

, et al., Restricted posture in dentistry - a kinematic analysis of orthodontists, BMC Musculoskelet Disord 2017;18(1):275. PubMed PMID: Pubmed Central PMCID: PMC963. Epub /06/25. eng

28.

Karhu

, Kansi

, Kuorinka

, Correcting working postures in industry: A practical method for analysis, Appl Ergon 1977;8(4):199–201. PubMed PMID: Epub /12/01. eng.

29.

Ellegast

Messung von Muskel-Skelett-Belastungen mit dem CUELA-Messsystem Aus der Arbeit des IFA. 2013;2/2013.

30.

Franzen

, GmbH

IfG

Darstellung von Methoden zur Untersuchung und Bewertung von Arbeitsplätzen unter ergonomischen Gesichtspunkten, http://www.indina.de/fileadmin/indina/Bilderpool_indina/Publikationen/INDINA_Leitfaden_Ergonomie.pdf.

31.

Kiermayer

, Hoehne-Huckstadt

, Brielmeier

, Brutting

, Ellegast

, Schmidt

, Musculoskeletal load in and highly repetitive actions of animal facility washroom employees, J Am Assoc Lab Anim Sci 2011;50(5):665–74 PubMed PMID: 22330713. Pubmed Central PMCID: 3189670.

32.

Ellegast

Personengebundenes Meßsystem zur automatisierten Erfassung von Wirbelsäulenbelastungen bei beruflichen Tätigkeiten. BIA-Report. 1998;5/98.

33.

Mache

, Scutaru

, Vitzthum

, Gerber

, Quarcoo

, Welte

, et al., Development and evaluation of a computer-based medical work assessment programme, Journal of occupational medicine and toxicology (London, England) 2008;3(35):35.PubMed PMID: 19094213. Pubmed Central PMCID: 2628342.

34.

Mache

, Groneberg

, Medical work Assessment in German hospitals: a Real-time Observation study (MAGRO) - the study protocol, Journal of occupational medicine and toxicology (London, England) 2009;4:12. PubMed PMID: Pubmed Central PMCID: PMC922. Epub /06/10. eng

35.

Mache

, Vitzthum

, Kusma

, Nienhaus

, Klapp

, Groneberg

, Pediatricians’ working conditions in German hospitals: a real-time task analysis, Eur J Pediatr 2010;169(5):551–5. PubMed PMID 19774393.

36.

Ohlendorf

, Schwarzer

, Rey

, Hermanns

, Nienhaus

, Ellegast

, et al., Medical work assessment in German hospitals: a study protocol of a movement sequence analysis (MAGRO-MSA), Journal of occupational medicine and toxicology (London, England) 2015;10(1):1. PubMed PMID: Pubmed Central PMCID: PMC873. Epub /01/22. eng

37.

Karhu

, Harkonen

, Sorvali

, Vepsalainen

, Observing working postures in industry: Examples of OWAS application, Appl Ergon 1981;12(1):13–7.PubMed PMID: Epub /03/01. eng

38.

de Bruijn

, Engels

, van der Gulden

, A simple method to evaluate the reliability of OWAS observations, Appl Ergon 1998;29(4):281–3.PubMed PMID: 543. Epub /08/13. eng.

39.

Gómez-Galán

, Pérez-Alonso

, Callejón-Ferre

Á J

, López-Martínez

, Musculoskeletal disorders: OWAS review, Ind Health 2017;55(4):314–37.PubMed PMID: Pubmed Central PMCID: PMC841. Epub /05/10. eng

40.

Wang

, Chen

, Chiou

, Using the OVAKO working posture analysis system in cleaning occupations, Work. 2019;64(3):613–21.PubMed PMID: Epub /10/29. eng.

41.

Kong

, Lee

, Kim

, Comparisons of ergonomic evaluation tools (ALLA, RULA, REBA and OWAS) for farm work, International journal of occupational safety and ergonomics: JOSE. 2017;24(2):218–23. PubMed PMID: Epub /03/18. eng.

42.

Binoosh

, Mohan

, Ashok

, Dhana Sekaran

, Virtual postural assessment of an assembly work in a small scale submersible pump manufacturing industry, Work 2017;58:567–78.

43.

Daneshmandi

, Kee

, Kamalinia

, Oliaei

, Mohammadi

, An ergonomic intervention to relieve musculoskeletal symptoms of assembly line workers at an electronic parts manufacturer in Iran, Work 2018;61:515–21.

44.

Ohlendorf

, Erbe

, Nowak

, Hauck

, Hermanns

, Ditchen

, et al., Constrained posture in dentistry - a kinematic analysis of dentists, BMC Musculoskelet Disord. 2017;18(1):291.PubMed PMID: Pubmed Central PMCID: PMC999. Epub /07/07. eng

45.

Freitag

, Ellegast

, Dulon

, Nienhaus

, Quantitative measurement of stressful trunk postures in nursing professions, Ann Occup Hyg. PubMed PMID 2007;51(4):385–95 PubMed PMID 17715425.

46.

Cetin

, Bilgin

, Köse

, A comparison of occupational groups using different working postures in terms of their low back and neck health status, Journal of Back and Musculoskeletal Rehabilitation 2018;31:475–80.

47.

Dutta

, Breloff

, Dai

, Sinsel

, Warren

, Carey

, et al., Effects of working posture and roof slope on activation of lower limb muscles during shingle installation, Ergonomics. 2020;63(9):1182–93. PubMed PMID: Pubmed Central PMCID: PMC978. Epub /05/22. eng

48.

Prince

, Elliott

, Scott

, Visintini

, Reed

, Device-measured physical activity, sedentary behaviour and cardiometabolic health and fitness across occupational groups: a systematic review and meta-analysis, The international journal of behavioral nutrition and physical activity. 2019;16(1):30. PubMed PMID: Pubmed Central PMCID: PMC868. Epub /04/04. eng.

49.

Varnai

, Nienhaus

, Groneberg

, Ohlendorf

, Körperhaltungen von Arbeitnehmern am Beispiel einer Krankenhausgroßküche Eine objektiveTätigkeitsanalyse, Zentralblatt für Arbeitsmedizin, Arbeitsschutz und Ergonomie 2018;68(Issue 4,):197–205.

50.

Viggiani

, Gallagher

, Sehl

, Callaghan

, The distribution of lumbar intervertebral angles in upright standing and extension is related to low back pain developed during standing, Clinical biomechanics (Bristol, Avon) 2017;49:85–90. PubMed PMID: Epub /09/13. eng

51.

McAtamney

, Nigel Corlett

. RULA: a survey method for the investigation of work-related upper limb disorders, Appl Ergon 1993;24(2):91–9.PubMed PMID: Epub /04/01. eng

52.

Liebers

, Brendler

, Latza

. Berufsspezifisches Risiko für das Auftreten von Arbeitsunfähigkeit durch Muskel-Skelett-Erkrankungen und Krankheiten des Herz-Kreislauf-Systems 2016

53.

Berger

, Caffier

, Schultz

, Trippler

, Bewegungsergonomische Gestaltung von andauernder Steharbeit: Länderausschuss für Arbeitsschutz und Sicherheitstechnik; 2009.

54.

Kilbom

, Repetitive work of the upper extremity: Part I Guidelines for the practitioner, International Journal of Industrial Ergonomics 1994;14, 51–7.

55.

Kilbom

. Repetitive work of the upper extremity: Part II - The scientific basis (knowledge base) for the guide, International Journal of Industrial Ergonomics 1994);14(59-86).

56.

Jürgens

, Mohr

, Pangert

, Pernack

, Schultz

, Steinberg

LV29 Handlungsanleitung zur Beurteilung der Arbeitsbedingungen beim Ziehen und Schieben von Lasten: Länderausschuss für Arbeitsschutz und Sicherheitstechnik; 2002.

57.

Nimbarte

, Sun

, Jaridi

, Hsiao

, Biomechanical loading of the shoulder complex and lumbosacral joints during dynamic cart pushing task, Appl Ergon 2013;44(5):841–9.PubMed PMID 23566675.

58.

Hoehne-Hückstädt

, Herda

, Ellegast

, Hermanns

, Hamburger

, Ditchen

, Bewertung des Risikofaktors “Ungünstige Haltungen und Bewegungen” Sankt Augustin: Hauptverband der gewerblichen Berufsgenossenschaften (HVBG); 2007.

59.

Tabatabaeifar

, Svendsen

, Johnsen

, Hansson

, Fuglsang-Frederiksen

, Frost

, Reversible median nerve impairment after three weeks of repetitive work, Scandinavian Journal of Work, Environment & Health 2022;43(2):163–70. PubMed PMID: Pubmed Central PMCID: PMID 28060387.

60.

Silverstein

, Fine

, Armstrong

, Hand wrist cumulative trauma disorders in industry, British Journal of Industrial Medicine 1986;43(11):779–70, PubMed PMID: 3790459. Pubmed Central PMCID: 1007752.

61.

Kilbom

, Armstrong

, Buckle

, Fine

, Hagberg

, Haring-Sweeney

, et al. Musculoskeletal Disorders: Work-related Risk Factors and Prevention, Int J Occup Environ Health 1996;2(3):239–46 PubMed PMID: 9933878.

62.

Moore

, Garg

, The Strain Index: a proposed method to analyze jobs for risk of distal upper extremity disorders, American Industrial Hygiene Association Journal 1995;56(5):443–58.