Correlates of vitamin D in psychotic disorders: A comprehensive systematic review

James Adamson*, John Lally, Fiona Gaughran, Amir Krivoy, Lauren Allen, Brendon Stubbs

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

Abstract

People with psychosis have high prevalence of low vitamin D levels but the correlates and relevance of this deficiency are unclear. A systematic search of major databases from inception to 03/2016 was undertaken investigating correlates of vitamin D in people with psychosis. Data was summarised with a best evidence synthesis. Across 23 included studies (n=1770 psychosis, n=8171 controls) a mean difference in vitamin D levels between both groups of −11.14 ng/ml±0.59 was found. 53 unique correlations between vitamin D and outcomes in people with psychosis were identified. The evidence base was broadly equivocal although season of blood sampling (67% of studies found a positive correlation with warmer seasons) and parathyroid hormone (100% of studies found a negative correlation) were associated with vitamin D levels. The most commonly non-correlated variables were: BMI (83% found no correlation), age (73%), gender (86%), smoking (100%), duration of illness (100%) and general assessment of functioning score (100%). In conclusion, whilst many unique correlates have been investigated, there is weak and inconclusive evidence regarding the consistency and meaning of the correlates of vitamin D levels in people with psychosis. Future longitudinal studies should consider the correlates of vitamin D in people with psychosis.

Original languageEnglish
Pages (from-to)78-85
Number of pages8
JournalPsychiatry Research
Volume249
DOIs
StatePublished - 1 Mar 2017
Externally publishedYes

Keywords

  • 25-OHD
  • 25-hydroxyvitamin D
  • Hypovitaminosis D
  • Psychosis
  • Schizophrenia
  • Vitamin D

Fingerprint

Dive into the research topics of 'Correlates of vitamin D in psychotic disorders: A comprehensive systematic review'. Together they form a unique fingerprint.

Cite this