Aims: Profound changes in family structure took place in many countries, during the second part of the previous century. The purpose of this paper is to analyse the association between the type of family residency and psychosomatic problems in younger and older adolescents, particularly focusing on alternate residency, and to examine the impact of child-parent relations. Methods: We used data collected in 2009 by Statistics Sweden among 172,298 Swedish students in Grade 6 and Grade 9 (approximate ages 12 and 15 years old); comprising 80% and 86%, respectively, of the entire population of students in those grades. We collected the data with a questionnaire, completed anonymously in school: We used the Psychosomatic Problems (PSP) scale as the outcome measure. Results: The type of family residency showed a weaker association with psychosomatic problems than the child-parent relationships did. Living in non-intact families increased the probability of adolescent psychosomatic problems by 0-0.05, compared to intact families. In Grade 9, there were no differences in psychosomatic problems between the students in alternate residency and those living with their two parents; and in Grade 6, these differences were relatively small. In comparison, a worse relationship with parents increased the probability of psychosomatic problems by 0.11-0.17, depending on the school grade and type of family residency. Conclusions: The structure of the family, as well as the child-parent relationships needs to be taken into account, to properly estimate the magnitude of the family situation as a determinant of adolescent psychosomatic problems. Our results justify universal intervention at the policy level.