Dementia risk prediction in individuals with mild cognitive impairment: a comparison of Cox regression and machine learning models.

BACKGROUND: Cox proportional hazards regression models and machine learning models are widely used for predicting the risk of dementia. Existing comparisons of these models have mostly been based on empirical datasets and have yielded mixed results. This study examines the accuracy of various machine learning and of the Cox regression models for predicting time-to-event outcomes using Monte Carlo simulation in people with mild cognitive impairment (MCI). METHODS: The predictive accuracy of nine time-to-event regression and machine learning models were investigated. These models include Cox regression, penalized Cox regression (with Ridge, LASSO, and elastic net penalties), survival trees, random survival forests, survival support vector machines, artificial neural networks, and extreme gradient boosting. Simulation data were generated using study design and data characteristics of a clinical registry and a large community-based registry of patients with MCI. The predictive performance of these models was evaluated based on three-fold cross-validation via Harrell's concordance index (c-index), integrated calibration index (ICI), and integrated brier score (IBS). RESULTS: Cox regression and machine learning model had comparable predictive accuracy across three different performance metrics and data-analytic conditions. The estimated c-index values for Cox regression, random survival forests, and extreme gradient boosting were 0.70, 0.69 and 0.70, respectively, when the data were generated from a Cox regression model in a large sample-size conditions. In contrast, the estimated c-index values for these models were 0.64, 0.64, and 0.65 when the data were generated from a random survival forest in a large sample size conditions. Both Cox regression and random survival forest had the lowest ICI values (0.12 for a large sample size and 0.18 for a small sample size) among all the investigated models regardless of sample size and data generating model. CONCLUSION: Cox regression models have comparable, and sometimes better predictive performance, than more complex machine learning models. We recommend that the choice among these models should be guided by important considerations for research hypotheses, model interpretability, and type of data.

Factors That Determine the Likelihood of Giving Birth to the First Child within 10 Months after Marriage.

BACKGROUND: One of the major aims of marriage is to procreate or give birth to a child. Childbirth is so crucial in marriage that it often determines the happiness of the couple. Too much delay in childbirth after marriage or the likelihood that one cannot give birth after marriage can lead to divorce. However, causes of delay in childbirth are often difficult to detect by both the Gynaecologist and the couple involved. This makes proposing solutions to issues related to childbirth usually unsuccessful. METHODS: It is against this background that we conducted this study to identify factors that determine childbirth within 10 months or after 10 months of marriage (birth length) among women in Ghana. This was achieved by using a logistic regression model for the dichotomous birth length variable, adjusting for risk factors/predictors of birth length. The data used for the study were obtained from the 2014 Ghana Demographic and Health Survey, consisting 6,525 complete cases with 18 predictor variables. Statistical analyses were carried out using STATA version 14.1. RESULTS: The results show that respondents who have ever terminated pregnancy are more likely (OR = 0.178, 95%CI = 0.044, 0.312) to deliver after 10 months, wives whose husbands have higher education are less likely (OR = -0.162, 95%CI = -0.236, -0.088) to give birth after 10 months of marriage, wives who reported that beating is justified if she goes out without her husband's notice are more likely (OR = 0.466, 95%CI = 0.305, 0.628) to give birth after 10 months, wives who reported that beating is justified if she neglects the child are more likely (OR = -0.305, 95%CI = -0.461, -0.149) to give birth within 10 months, and wives who reported that beating is justified when she argues with her husband are less likely (OR = -0.301, 95%CI = -0.451, -0.152) to give birth after 10 months of marriage. Every unit increase in the age of the respondent at marriage increases the likelihood of giving birth after 10 months of marriage, and a unit increase in the age of the respondent at first sex decreases the likelihood of giving birth after 10 months in marriage. CONCLUSIONS: For conception within 1 month of marriage, wives and husbands should/are encouraged to have frequent sex, any negative social behaviour or policies must be discouraged, experts' advice on contraceptive use must be sought, and women are encouraged to desist from termination of pregnancy at any time of their life. Husbands should openly express their desire and love for their children since this increases the likelihood of wives' desire to give birth. This leads to frequent sex, which then reduces conception time, and hence childbirth within the shortest possible time.