BACKGROUND

SOLAR II is the 2nd follow-up of a population-based cohort study that follows the participants of ISAAC Phase Two recruited in Munich and Dresden in 1995/6. A first follow-up study was conducted 2002 and 2003 (SOLAR I). The aims of SOLAR II were to investigate the course of atopic diseases over puberty taking environmental and occupational risk factors into account. This paper describes the methods of the 2nd follow-up carried out from 2007 to 2009 and the challenges we faced while studying a population-based cohort of young adults.

METHODS

Wherever possible, the same questionnaire instruments were used throughout the studies. They included questions on respiratory and allergic diseases, domestic and occupational exposure and work related stress. Furthermore, clinical examinations including skin prick tests, spirometry and bronchial challenge with methacholine, exhaled nitric oxide (FeNO) and blood samples were employed at baseline and 2nd follow-up. As information from three studies was available, multiple imputation could be used to handle missing data.

RESULTS

Of the 3053 SOLAR I study participants who had agreed to be contacted again, about 50% had moved in the meantime and had to be traced using phone directories and the German population registries. Overall, 2904 of these participants could be contacted on average five years after the first follow-up. From this group, 2051 subjects (71%) completed the questionnaire they received via mail. Of these, 57% participated at least in some parts of the clinical examinations. Challenges faced included the high mobility of this age group. Time constraints and limited interest in the study were substantial. Analysing the results, selection bias had to be considered as questionnaire responders (54%) and those participating in the clinical part of the study (63%) were more likely to have a high parental level of education compared to non-participants (42%). Similarly, a higher prevalence of parental atopy (e.g. allergic rhinitis) at baseline was found for participants in the questionnaire part (22%) and those participating in the clinical part of the study (27%) compared to non-participants (11%).

CONCLUSIONS

In conclusion, a 12-year follow-up from childhood to adulthood is feasible resulting in a response of 32% of the baseline population. However, our experience shows that researchers need to allocate more time to the field work when studying young adults compared to other populations.