Identifying long-term and imminent suicide predictors in a general population and a clinical sample with machine learning.

BACKGROUND: Machine learning (ML) is increasingly used to predict suicide deaths but their value for suicide prevention has not been established. Our first objective was to identify risk and protective factors in a general population. Our second objective was to identify factors indicating imminent suicide risk. METHODS: We used survival and ML models to identify lifetime predictors using the Cohort of Norway (n=173,275) and hospital diagnoses in a Saskatoon clinical sample (n=12,614). The mean follow-up times were 17 years and 3 years for the Cohort of Norway and Saskatoon respectively. People in the clinical sample had a longitudinal record of hospital visits grouped in six-month intervals. We developed models in a training set and these models predicted survival probabilities in held-out test data. RESULTS: In the general population, we found that a higher proportion of low-income residents in a county, mood symptoms, and daily smoking increased the risk of dying from suicide in both genders. In the clinical sample, the only predictors identified were male gender and older age. CONCLUSION: Suicide prevention probably requires individual actions with governmental incentives. The prediction of imminent suicide remains highly challenging, but machine learning can identify early prevention targets.

Differences in birth weight by sex using adjusted quantile distance functions.

Herein, we report results from a study of birth weight distribution among boys and girls born in Norway in 2008. As our primary interest was to detect differences in the variability between the two sexes, we used the quantile distance function to describe the difference between two distribution functions. We used an adjusted version of the quantile function to look into the relation of sex differences in birth weight conditioned on maternal age, gestational age, preeclampsia, maternal diabetes type 1, maternal smoking status, and parity. At term (⩾37 weeks of gestation), boys showed a greater variability in birth weight than did girls, and these differences were maintained in the adjusted model. We also found that maternal age and maternal smoking habits influenced both sexes equally, whereas gestational age, preeclampsia, maternal diabetes type 1, and parity influenced one sex more than the other. The adjusted quantile distance function proved efficient in analyzing and demonstrating how covariates influence sex differences in birth weight.

Dietary patterns among Norwegian 2-year-olds in 1999 and in 2007 and associations with child and parent characteristics.

Infant and childhood nutrition influences short- and long-term health. The objective of the present paper has been to explore dietary patterns and their associations with child and parent characteristics at two time points. Parents of Norwegian 2-year-olds were, in 1999 (n 3000) and in 2007 (n 2984), invited to participate in a national dietary survey. At both time points, diet was assessed by a semi-quantitative FFQ that also provided information on several child and parent characteristics. A total of 1373 participants in the 1999 sample and 1472 participants in the 2007 sample were included in the analyses. Dietary patterns were identified by principal components analysis and related to child and parent characteristics using the general linear model. Four dietary patterns were identified at each time point. The 'unhealthy' and 'healthy' patterns in 1999 and 2007 showed similarities with regard to loadings of food groups. Both the 'bread and spread-based' pattern in 1999 and the 'traditional' pattern in 2007 had high positive loadings for bread and spreads; however, the 'traditional' pattern did also include positive associations with a warm meal. The last patterns identified in 1999 and in 2007 were not comparable with regard to loadings of food groups. All dietary patterns were significantly associated with one or several child and parent characteristics. In conclusion, the 'unhealthy' patterns in 1999 and in 2007 showed similarities with regard to loadings of food groups and were, at both time points, associated with sex, breastfeeding at 12 months of age, parity, maternal age and maternal work situation.