Prevalence and influencing factors of chronic back pain among staff at special schools with multiple and severely handicapped children in Germany: results of a cross-sectional study

Study design

Our cross-sectional study was carried out between August 2010 and August 2012 at 13 special schools focusing on motoric and/or holistic development of severely and multiple handicapped children in Rhineland-Palatinate (Germany). The main objective of the study was to survey data on physical and psychological strains and several health outcomes of teachers and educational staff at this type of school. Between August and September 2010, an information letter with a brief description of the course of the study was sent to the head teachers of all 15 schools focusing on motoric and/or holistic development of severely and multiple handicapped children in Rhineland-Palatinate. Altogether, 13 schools comprising 245 teachers and 417 educational staff were interested in participating. The head teachers of the two schools which did not participate asked the principal investigators to participate at a later time when the study was already completed. Subsequently the project was presented to the employees of the interested schools at teachers´ councils between October and December 2010. Directly following the presentation, an information letter about the whole project along with a written informed consent form was deposited in the personal “letter boxes” of all teachers and educational staff at the interested schools. The informed consent form has been signed by all participants. The individual registration deadline for a participation in the study was four weeks after the teacher´s council.

All participating schools received a detailed plan of action, including dates of interviews. In order to minimize any disturbances of regular teaching activities, the participants were allowed to determine the date of participation. The study was carried out directly at the schools. Data were surveyed using written questionnaires.

The study was conducted by the Institute of Occupational, Social and Environmental Medicine and the Institute for Teachers´ Health at the University Medical Center of the Johannes Gutenberg University of Mainz.

The study was approved by the State Ministry of Rhineland-Palatinate for Education, Science, Further Education and Culture (MBWWK), the supervision and service administration body (ADD) in Trier, the main staff councils of the special schools for children with learning difficulties, the ethical committee of the medical association of the German State of Rhineland-Palatinate, and the responsible principals and the local staff councils.

Questionnaire

The data for the present article were surveyed using a medical history questionnaire, containing 84 questions covering the topics basic sociodemographic information, professional qualification/requirements, personal state of health, musculoskeletal strains and disorders, prevention and risk of infections/vaccination status, psychological health, work-related accidents, and health-related behavior. The majority of the questions included in the medical history questionnaire were developed by the Institute of Occupational, Social and Environmental Medicine. In addition, 21 questions were adopted from the questionnaire of the GEDA (Ge sundheit in D eutschland a ktuell) study [10–12]. The latter study was commissioned by the German Federal Ministry of Health and carried out by the Robert Koch-Institute.

Variable construction

The dependent variable “chronic back pain” was surveyed by the following two questions adopted from the GEDA-questionnaire:

If the respondents agreed to both questions, they were categorized as having had chronic back pain in the last 12 months. If one of the questions was answered with “No” or “I don´t know” they were coded as not having suffered from chronic back pain during the aforementioned period.

Regarding sociodemographic variables, information on age, sex, marital status, highest level of education, and profession was used. Age was defined as the present age (in years) when completing the questionnaire. Information on the educational background of the respondents was obtained by asking for the highest level of education, using the categories “Abitur”, “Fachhochschulreife/Fachoberschule”, and “Mittlere Reife”. Generally speaking, German school-leaving qualifications do not have exact counterparts in most other countries. While the “Abitur” takes 12 to 13 years of school and is the entry qualification for the universities, “Fachhochschulreife/Oberschulreife” takes 12 years of school and permits entering an advanced technical college, whereas “Mittlere Reife” is obtained after 10 years of school without being able to study at a university or college. The profession of the respondents was obtained by asking whether they were actually working as teachers or as educational staff. The term “educational staff” is not clearly defined and is rather a collective name for all non-teachers working at special schools (e.g. physiotherapists, ergotherapists, nursery school teachers, remedial teachers, social education workers, speech therapists, etc.). Apart from nursing activities, people working as educational staff are also involved in teaching, but unlike regular teachers do not have a higher education degree in special needs education.

With regard to health-related variables we collected information on smoking status, alcohol consumption, body mass index (kg/m² as reported by the respondents), days of physical activity per week, and whether participants ever having had a diagnosed depression. People were classified as having had a depression/depressive mood when they agreed to the statement that a physician or a psychotherapist ever diagnosed a depression or a depressive mood in the past. Concerning the body mass index (BMI), we applied the thresholds given by the World Health Organization to categorize respondents as having normal weight (BMI < 25), overweight (25 ≤ BMI < 30) or being obese (BMI ≥ 30) [13].

The work-related exposures included the duration of employment (in years), hours working per week (in hours), frequency of exposure to carrying and lifting heavy loads (> 20 kg), frequency of working in an uncomfortable position, frequency of exposure to environmental impacts (e.g. noise, heat, cold, moisture, fume, dust), and frequency of psychosocial stressors in the working atmosphere (e.g. conflicts with colleagues/superiors or bullying). The response format concerning exposure frequency comprised “frequently”, “sometimes”, “rarely”, and “never”. We subsumed “rarely” and “never” into one item category afterwards. Information on nursing activities (carrying, lifting, and transferring pupils, washing pupils, toilet assistance for pupils, changing pupils´ diapers, (un-)dressing pupils) was surveyed by asking if the respective activity is an integral part of everyday work (Yes/No).

Statistical analyses

All statistical analyses were carried out using STATA/IC 12.1 (StataCorp LP, College Station, TX). Relative frequencies were calculated for a general description of the sample. In order to identify influencing factors of chronic back pain we conducted several multivariable logistic regression analyses. Firstly, we carried out regression analyses adjusted for age and sex for all variables potentially associated with chronic back pain. Secondly, all variables being significantly associated with chronic back pain in the age- and sex-adjusted models were eligible to enter into a final multivariable logistic regression model. We applied backward elimination with p < 0.05 to determine the variables to retain in the final model. Age and sex were entered into the final multivariable model independent of any considerations of significance.

Main findings

The aim of our cross-sectional study was to investigate the prevalence and potential influencing factors of chronic back pain among teachers and educational staff at special schools with handicapped children in Rhineland-Palatinate, Germany. More than one third of the respondents reported having suffered from chronic back pain in the last twelve months. The multivariable logistic regression analyses showed that the main predictors for chronic back pain among teachers and educational staff at special schools were increasing age, current smoking, a diagnosed depression/depressive mood, frequently carrying and lifting heavy loads, and frequent exposure to environmental impacts.

Implications of the findings

The one-year prevalence of chronic back pain of 38.7% in our study is quite high, compared to telephone survey data of German adults (20.7%; n = 21.262) surveyed by the Robert Koch-Institute in 2009 [14]. Comparing age- and sex-specific prevalences of chronic back pain in our study with this data shows that differences were especially notable regarding women aged 40–49 (45.2% vs. 31.5%).

The corresponding figures of studies from Japan were respectively 44.9% (n = 975) [2], 38.4% (n = 975) [9], and 33.8% (n = 234) [3]. However, the results are not directly comparable since the aforementioned studies focused mostly on lower back pain and used different definitions and time periods to calculate the prevalence.

Several factors with a potentially negative impact on the musculoskeletal health of teachers and pedagogical staff were present in our study. Regarding sociodemographic factors, people were more likely to suffer from chronic back pain with increasing age. A similar result has been found in the study by Yamamoto et al. [9] where age was a significant factor predicting lower back symptoms in multivariable logistic regression, with adjusted odds ratios being highest for subjects in their fifties. Considering the mean age of 45 years in our sample and the demographic change in Western societies, this finding is even more worrying. We found no significant association between gender and chronic back pain, which is in accordance to recent findings of Yamamoto [9]. Likewise, only two out of 17 studies considered in a review on musculoskeletal disorders of nurses by Long et al. [15] found women to be more frequently affected by musculoskeletal disorders. The prevalence of back pain was higher in educational staff (43.8%) than in regular teachers (27.6%). However, working as an educational staff was not a risk factor for back pain in the final multivariable logistic regression model. We subsequently analyzed the relationship between nursing care and profession and found that all activities were significantly more frequently carried out by educational staff. This finding could explain the differences in the prevalence of chronic back pain between these two groups.

Regarding health-related variables, several studies have shown that there are indeed at least modest relationships between smoking, depression and back pain [2, 16, 17], although the mechanisms are not well understood. However, due to the cross-sectional design of our study (see limitations) it remains unclear if an unhealthy lifestyle and/or depression causes or aggravates back pain or if back pain influences depression and/or causes people to drink and smoke more. Further studies on this topic using a prospective study design are necessary.

Concerning work-related exposures, more than 40% (mostly women) of the respondents in our study reported frequently carrying and lifting heavy loads weighing more than 20 kg. Taking into account that working at special schools involves all-day care of pupils between 6 and 20 years, this is hardly surprising. In order to prevent overexertion or incorrect weight bearing of the back muscles at special schools with handicapped children, various measures should be provided to the employees. In order to positively influence the behavior of teachers and educational staff, ergonomic workouts, training courses, and information brochures could be appropriate measures of primary prevention. Furthermore, measures of situational prevention like ergonomic adjustments of workplaces, classrooms, and nursing rooms which fit the needs of the handicapped, organizational changes in workflows, and sufficient personnel are necessary. Regarding secondary prevention, suitable preventive medical check-ups should be established in order to diagnose diseases and risk factors at an early stage and to prevent disease progress and chronification by providing appropriate measures.

In view of the vast number of employees involved in nursing care in our study along with higher prevalences of chronic back pain in these persons it should be ensured that schools are sufficiently equipped with necessary auxiliary resources (e.g. lifters, wheelchairs, nursing rooms, etc.) to release the burden of teaching staff.

The prevalence of exposure to environmental impacts in our study (32.9%) is quite high, and has not been considered as a risk factor for back pain in past studies. Although the specific type of environmental impact from which teachers and educational staff suffered in our study (noise, heat, cold, moisture, fume and dust have been surveyed within the same question) is unknown, it is highly probable that noise accounts for the most part. School inspections and risk assessments by safety engineers and, if possible, subsequent elimination of identified sources of exposure could pose an appropriate measure to reduce the burden caused by environmental impacts.

Limitations

The results of the present study are limited in several respects. We reached a response rate of approximately 60%, which is comparable to other studies on the topic [2, 3]. However, since a substantial part of teachers refused to participate in our study, selection bias due to non-response cannot be ruled out. It is, for instance, possible that teachers and pedagogical staff were more likely to participate in our study when they were personally interested and/or affected by the topic. In this case the prevalence of chronic back pain in our study would be overestimated.

A further limitation of the present study which has to be acknowledged is the possibility for recall bias. For example, it is not clear if the respondents correctly remembered the presence of chronic back pain in the last twelve months which could have led to over- or underestimation. It is further possible, that respondents with back pain may recall their exposure more thoroughly than those without back pain, which could have led to an overestimation of the effect of a particular exposure. Moreover, the presence of back pain depends solely upon the subjective self-report of the respondents and is not based upon an objective clinically verified diagnosis of a specialist. A further point of criticism relates to the cross-sectional study design which makes it impossible to draw any inferences of causality. It is for example unclear if depression worsens back pain or if back pain has an adverse effect on the course of a depression.