Background. Genetic effects upon behaviour are pervasive. To what extent are the many correlates of major depression (MD) due to individual-specific environmental experiences versus genetic factors correlated with risk for MD?

Methods. From a population-based twin registry, we identified 72 female monozygotic pairs discordant for a lifetime history of MD and compared the affected and unaffected members on a wide range of putative correlates of MD.

Results. The affected twin differed from her unaffected co-twin on many variables, eight of which were maximally discriminating: (i) maternal protectiveness; (ii) conflictual parent–child relationship; (iii) low optimism; (iv) current stressful life events; (v) financial difficulties and a history of (vi) phobia, (vii) nicotine dependence; and (viii) divorce. A cluster analysis suggested three ‘environmental pathways' to MD characterized by: (i) childhood vulnerability and anxiety; (ii) acting-out and demoralization; and (iii) interpersonal difficulties.

Conclusion. Important precursors and sequelae of MD originate in environmental experiences unique to the individual and are not mediated through genetic factors or family-of-origin effects. Such environmental factors cause pervasive differences in monozygotic twins discordant for MD, especially in the areas of interpersonal difficulties, psychopathology, social problems and self-concept. These findings should be interpreted in the context of possible retrospective recall bias and the difficulty of distinguishing risk factors from sequelae in co-twin–control studies.