Authors' reply

We would like to clarify a few points regarding that our results ‘strengthen the theory of a cardiovascular contribution to the aetiology of depression’. ¹

First, we talk about a ‘contribution’ which does not necessarily imply a direct causal pathway, stating that ‘our findings are not explanatory with respect to causal chains leading to the onset of depression’. In line with this, we did include a careful discussion about other possible confounding mechanisms, i.e. factors that may increase the risk for both poor fitness and depression – for example, childhood factors, personality, self-esteem and subsyndromal affective problems. By including parental educational level as a confounder and by performing subanalyses within full brother pairs, many of the early childhood risk factors could be accounted for.

Second, as the conscription routines included extensive questions regarding every possible previous and present mental health problem in combination with separate examinations by professional psychologists and physicians, we believe that subsyndromal affective problems were not often overlooked. Also, to further reduce baseline misclassification, we did perform separate analyses excluding incident cases in the first year.

Third, we would like to stress that not all study participants were fit for recruitment into national service, but that the conscription test was used to select suitable recruits. Participation in the conscription tests was compulsory according to Swedish law and exemptions were granted only for incarcerated males and severe chronic medical disabilities (approximately 2-3% of the yearly male population). We can therefore consider our study a population study. After conscription, about 40 000 individuals were considered ‘unfit’ due to a cardiovascular fitness stanine score 1-3. All these ‘unfit’ young men were included in our study.

Fourth, the question of whether cardiovascular fitness may be related to increased risk for other types of psychiatric disorders in adulthood is one that we will continue to pursue in future analyses of the national conscription data.

Taken together, we still argue that the data ‘strengthen the theory of a cardiovascular contribution to the aetiology of depression’, which in our paper stands in direct connection with the sentence: ‘although the results in the present population-based prospective study are compelling, a number of confounders could not be measured and intervention studies are needed to determine whether physical exercise in young adulthood can prevent future onset of depression.’ We fully agree with de Jonge & Roest in their editorial ² that a greater understanding of the mechanisms underlying these associations, including complex bidirectional models, may provide opportunities and strategies for prevention.