Analysis
The analyses were performed separately for the two occupational groups. Information from the open-ended question on perceived causes or aggravating factors was categorised and described qualitatively. Descriptive statistics were presented as the percent, median and Inter Quartile Range (IQR). To gain insight in the characteristics and representativeness of the respondents, we tested i) for differences in age between responders and the initial sample using a t- test and ii) the relation between age (years) and having MSDs at baseline, and the relation between having or having not MSDs at baseline and response at follow-up using univariate logistic regression analysis. Statistical significance was set to an alpha level of 0.05. The prevalence and 95% confidence interval (CI) of the MSDs was calculated using the baseline data. Confidence intervals were calculated by using the Wald method when the sample consisted of more than 150 persons and the adjusted Wald method when the sample consisted of less than 150 persons [32].
Problems experienced during work was recoded into three categories (0–3 = no/little, 4-6= medium, 7–10 = many problems). Changes in the problems experienced during work were assessed by calculating the difference in the score between baseline and follow-up. We defined a difference over time in problems during work as a decrease or increase of 2 points or more. Furthermore, we assessed for every body region the frequencies and percentages of the reported tasks or activities that caused or aggravated the complaints of the body region under consideration. Next, for every task or activity a top-three of involved body regions was constructed and reported. The IBM SPSS Statistics 19 software was used to analyse the data.
Results
Characteristics of the study population
At baseline, the overall response rate was 39% for the bricklayers (292/750) and 34% for the construction supervisors (256/750). After excluding partially completed questionnaires, response rate was 36% (n=267) for the bricklayers and 31% (n=232) for the supervisors. No female workers responded. At follow-up, the overall response rate was 80%; 83% for the bricklayers (222/267) and 76% for the supervisors (177/232). The characteristics of the participants who responded at both baseline and follow-up, are shown in Table 1.
We found no difference in age between the responding bricklayers and the initial sample of bricklayers. However, we did find a statistically significant difference in age for the construction supervisors. At baseline, the responding construction supervisors were older (5.5 years, p