Survey of work-related knee-stressing activities in selected occupations

Project No. IFA 4188


completed 01/2012


The code of practice for formally recognized occupational disease (BK) No 2112, concerning knee osteoarthritis resulting from an activity performed in a kneeling posture or comparable load upon the knee with cumulative exposure over a vocational life of at least 13,000 hours and a minimum total exposure duration of one hour per shift, states that work-related knee loading caused by kneeling, squatting, and crawling are risk factors for the emergence of a relevant disease. There are however a lack of valid data on the quality and quantity of the loading concerned. In addition, a wealth of epidemiological studies on this subject are based upon self-reports. The present project therefore had two objectives: firstly, to describe the quality and quantity of occupational knee-loading based upon valid measurement data; and secondly, to test the validity of information provided by self-reports concerning occupational knee-loading. The results are to be presented in the form of an IFA Report and in two international publications in relevant research journals.


The data on occupational loading caused by kneeling, squatting, and crawling measured in the course of IFA Project 4138, "Measured value register of knee-straining work activities (GonKatast)", were to be interpreted beyond the pure duration of knee-straining postures per work shift. This was to yield information on the knee angle ranges assumed during the postures, the number and duration of individual phases of knee-loading and the symmetry of the loading, for each of the 16 vocations studied in the "GonKatast" project. In addition, the measured data on the duration of five different knee postures (unsupported kneeling, supported kneeling, sitting on heels, squatting and crawling) were to be compared with the self-reported information obtained by means of questionnaires from 190 test subjects. For this purpose, the results of two questionnaires for each test subject (the first recorded immediately following measurement, the second six months later) were to be interpreted and compared with the corresponding measured data from GonKatast. Correlation analyses, Wilcoxon tests and Bland-Altman plots were to be employed as the statistical methods.


As with the daily exposure duration of the occupational knee-loading, which had already been shown in the preceding studies (GonKatast) to vary widely, a wide variation was now also observed in the number and duration of knee stress phases per working shift. For the sample group, the number of knee stress phases per day ranged from 20 to 384 processes and the average duration of a single process from 0.2 to 13.5 minutes. This enabled both relatively static loading and highly dynamic processes to be distinguished. Study of the knee angle ranges assumed during the knee-straining postures showed the knee joints to be at least strongly bent (>=120°) for approximately 75% of the time and extremely strongly bent (>= 150°) for 25% of the time. The knee-joint flexion would therefore appear to be stronger than has previously been assumed. The analysis by measurement of knee-straining postures also showed that the loading do not generally affect both knee joints in equal measure, but rather that - for example as a result of kneeling on one knee only - the loading is asymmetrical, therefore affecting one knee joint more strongly than the other. In summary, the results of the measurement analysis constitute an important element in research into the damaging mechanisms of kneeling and squatting body postures.
With regard to the quality of the test subjects' own estimates, the study demonstrated that the test subjects were not able to provide adequately valid estimations of the duration of their knee-loading neither immediately following the task concerned, nor six months later. The deviations from the objectively measured duration of the knee loading were in some cases considerable, at over 100% for a large proportion of the test subjects and in some cases over 1,000%. In contrast to the quantification of the loading, the test subjects were however quite able to state correctly in both surveys whether certain postures were or were not assumed. These results are to be taken into consideration during the planning of studies and the development of questionnaires.

Last Update:

20 Aug 2012


Financed by:
  • Deutsche Gesetzliche Unfallversicherung e. V. (DGUV)
Research institution(s):
  • Institut für Arbeitsschutz der Deutschen Gesetzlichen Unfallversicherung (IFA)
  • BG BAU - Berufsgenossenschaft der Bauwirtschaft
  • Berufsgenossenschaft für Transport und Verkehrswirtschaft (BG Verkehr)
  • Berufsgenossenschaft Holz und Metall (BGHM)
  • Berufsgenossenschaft Rohstoffe und chemische Industrie (BG RCI)
  • Berufsgenossenschaft Energie Textil Elektro Medienerzeugnisse
  • Universität Witten-Herdecke
  • Universitätsklinikum Tübingen

-cross sectoral-

Type of hazard:

work-related diseases


musculoskeletal disorders (except cancer)

Description, key words:

IFA report, work-related knee loading, knee angle ranges, validity of self-reports, comparison between methods