Meta-analysis of adult height and birth length in schizophrenia

Abstract

ObjectiveAs a group, people with schizophrenia have a number of subtle anatomical abnormalities as well as physiological abnormalities that precede antipsychotic treatment. Some studies have also found shorter birth length or shorter adult height in people with schizophrenia compared to control subjects. We performed a systematic review and meta-analysis of studies of birth length and adult height in schizophrenia, following PRISMA guidelines (Prospero Registration # CRD42016043718). Data sourcesWe searched the PsycInfo, Web of Science, and PubMed databases for articles published 1947–2016. Study selectionArticles were included if they had data for patients diagnosed with schizophrenia and a matched control group of subjects without a psychotic disorder; both groups were measured for birth length and/or adult height (18years or older); and the paper was published in English. Data extractionOne author extracted the data, which was verified by the other. ResultsFor adult height, six studies with 1,122 patients and 250,200 control subjects were included in analyses. There were six birth length studies, which included 984 patients and 976,296 controls. The patients did not differ from comparison subjects in birth length (effect size estimate=−0.03; CI: −0.09, 0.03), but adults were slightly shorter than comparison subjects (−0.15; −0.24, −0.06). In meta-regression of adult studies, the variables of first episode versus clinical sample, and population registry versus non-registry were not significant. Matching for several important variables was usually lacking in these studies. ConclusionsWhile there appears to be no difference in birth length between people with schizophrenia and comparison subjects, the former may be slighter shorter in adult life. The cause of such a discrepancy, if confirmed, is not clear, and lack of matching on potentially confounding variables undermines confidence in any conclusion.