Development and Validation of Risk Scores for All-Cause Mortality for a Smartphone-Based "General Health Score" App: Prospective Cohort Study Using the UK Biobank

Health literacy in relation to web-based measurement of cognitive function in the home: UK Women's Cohort Study

OBJECTIVE

Older adults may require additional support to comprehend written information due to inadequate health literacy, which involves components of cognitive function including reaction time. This study tested the acceptability of web-based reaction time testing in the UK Women's Cohort Study and possible sources of bias. Additionally, it assessed the association between health literacy and reaction time.

DESIGN

A cross-sectional analysis was conducted using data from the UK Women's Cohort Study, a prospective cohort study.

PARTICIPANTS

The study involved women aged 48-85 without cancer registration who participated in the 2010/2011 follow-up (n=768).

SETTING

Postal questionnaires and web-based cognitive function tests were administered in participants' homes.

METHODS AND ANALYSIS

Logistic regression identified predictors of volunteering for reaction time testing, used to calculate inverse probability weights for the primary analysis. Associations between health literacy and reaction time were estimated with linear regression models, adjusting for volunteer effects. Poisson regression models assessed associations between health literacy and choice reaction time errors.

PRIMARY AND SECONDARY OUTCOME MEASURES

The primary outcome was acceptability of web-based testing (response rate, task distress, task difficulty). Secondary outcomes were sources of volunteer bias and the association between health literacy and reaction time.

RESULTS

Web-based testing of cognitive function was attempted by 67% of women (maximum age 80), with little distress or difficulty reported. There was substantive volunteer bias. Women providing data on cognitive function were younger, had higher educational attainment and were higher in self-rated intelligence. Inadequate health literacy was associated with making fewer choice reaction time errors among those providing valid data but was also associated with not providing valid data. Health literacy was not associated with other aspects of reaction time (speed, variability). Additionally, selection bias may have restricted range on study variables, given that 2010/2011 volunteers were younger and more educated compared with those at recruitment in 1995/1998.

CONCLUSION

Brief web-based measures of cognitive function in the home are acceptable to women aged 48-80, but there are substantive selection effects and volunteer biases. Additionally, there are potentially vulnerable subgroups who provide poorer quality data.

Correlation between allostatic load index and cumulative mortality: a register-based study of Danish municipalities

OBJECTIVES

The aim of this study was to examine population-based allostatic load (AL) indices as an indicator of community health across 14 municipalities in Denmark.

DESIGN

Register-based study.

SETTING

Data derived from: the Lolland-Falster Health Study, the Copenhagen General Population Study and the Danish General Suburban Population Study. Nine biomarkers (systolic blood pressure, diastolic blood pressure, pulse rate, total serum cholesterol, high-density lipoprotein cholesterol, waist-to-hip ratio, triglycerides, C-reactive protein and serum albumin) were divided into high-risk and low-risk values based on clinically accepted criteria, and the AL index was defined as the average between the nine values. All-cause mortality data were obtained from Statistics Denmark.

PARTICIPANTS

We examined a total of 106 808 individuals aged 40-79 years.

PRIMARY OUTCOME MEASURE

Linear regression models were performed to investigate the association between mean AL index and cumulative mortality risk.

RESULTS

Mean AL index was higher in men (range 2.3-3.3) than in women (range 1.7-2.6). We found AL index to be strongly correlated with the cumulative mortality rate, correlation coefficient of 0.82. A unit increase in mean AL index corresponded to an increase in the cumulative mortality rate of 19% (95% CI 13% to 25%) for men, and 16% (95% CI 8% to 23%) for women but this difference was not statistically significant. The overall mean increase in cumulative mortality rate for both men and women was 17% (95% CI 14% to 20%).

CONCLUSIONS

Our findings indicate the population-based AL index to be a strong indicator of community health, and suggest identification of targets for reducing AL.

Development and validation of questionnaire-based machine learning models for predicting all-cause mortality in a representative population of China

Background

Considering that the previously developed mortality prediction models have limited applications to the Chinese population, a questionnaire-based prediction model is of great importance for its accuracy and convenience in clinical practice.

Methods

Two national cohort, namely, the China Health and Nutrition Survey (8,355 individual older than 18) and the China Health and Retirement Longitudinal Study (12,711 individuals older than 45) were used for model development and validation. One hundred and fifty-nine variables were compiled to generate predictions. The Cox regression model and six machine learning (ML) models were used to predict all-cause mortality. Finally, a simple questionnaire-based ML prediction model was developed using the best algorithm and validated.

Results

In the internal validation set, all the ML models performed better than the traditional Cox model in predicting 6-year mortality and the random survival forest (RSF) model performed best. The questionnaire-based ML model, which only included 20 variables, achieved a C-index of 0.86 (95%CI: 0.80-0.92). On external validation, the simple questionnaire-based model achieved a C-index of 0.82 (95%CI: 0.77-0.87), 0.77 (95%CI: 0.75-0.79), and 0.79 (95%CI: 0.77-0.81), respectively, in predicting 2-, 9-, and 11-year mortality.

Conclusions

In this prospective population-based study, a model based on the RSF analysis performed best among all models. Furthermore, there was no significant difference between the prediction performance of the questionnaire-based ML model, which only included 20 variables, and that of the model with all variables (including laboratory variables). The simple questionnaire-based ML prediction model, which needs to be further explored, is of great importance for its accuracy and suitability to the Chinese general population.

A novel score for mobile health applications to predict and prevent mortality: Further validation and adaptation to US population using the US NHANES dataset

Background

The C-Score, which is an individual health score, is based on a predictive model validated in the UK and US populations. It was designed to serve as an individualized point-in-time health assessment tool that could be integrated into clinical counseling or consumer-facing digital health tools to encourage lifestyle modifications that reduce the risk of premature death.

Objective

Our study aimed to conduct an external validation of the C-Score in the US population and expand the original score to improve its predictive capabilities in the US population. The C-Score is intended for mobile health apps on wearable devices.

Methods

We conducted a literature review to identify relevant variables that were missing in the original C-Score. Subsequently, we used data from the 2005 to 2014 US National Health and Nutrition Examination Survey (NHANES; N=21,015) to test the capacity of the model to predict all-cause mortality. We used NHANES III data from 1988 to 1994 (N=1440) to conduct an external validation of the test. Only participants with complete data were included in this study. Discrimination and calibration tests were conducted to assess the operational characteristics of the adapted C-Score from receiver operating curves and a design-based goodness-of-fit test.

Results

Higher C-Scores were associated with reduced odds of all-cause mortality (odds ratio 0.96, P<.001). We found a good fit of the C-Score for all-cause mortality with an area under the curve (AUC) of 0.72. Among participants aged between 40 and 69 years, C-Score models had a good fit for all-cause mortality and an AUC >0.72. A sensitivity analysis using NHANES III data (1988-1994) was performed, yielding similar results. The inclusion of sociodemographic and clinical variables in the basic C-Score increased the AUCs from 0.72 (95% CI 0.71-0.73) to 0.87 (95% CI 0.85-0.88).

Conclusions

Our study shows that this digital biomarker, the C-Score, has good capabilities to predict all-cause mortality in the general US population. An expanded health score can predict 87% of the mortality in the US population. This model can be used as an instrument to assess individual mortality risk and as a counseling tool to motivate behavior changes and lifestyle modifications.

Comparison of Machine Learning Techniques for Mortality Prediction in a Prospective Cohort of Older Adults

As global demographics change, ageing is a global phenomenon which is increasingly of interest in our modern and rapidly changing society. Thus, the application of proper prognostic indices in clinical decisions regarding mortality prediction has assumed a significant importance for personalized risk management (i.e., identifying patients who are at high or low risk of death) and to help ensure effective healthcare services to patients. Consequently, prognostic modelling expressed as all-cause mortality prediction is an important step for effective patient management. Machine learning has the potential to transform prognostic modelling. In this paper, results on the development of machine learning models for all-cause mortality prediction in a cohort of healthy older adults are reported. The models are based on features covering anthropometric variables, physical and lab examinations, questionnaires, and lifestyles, as well as wearable data collected in free-living settings, obtained for the "Healthy Ageing Initiative" study conducted on 2291 recruited participants. Several machine learning techniques including feature engineering, feature selection, data augmentation and resampling were investigated for this purpose. A detailed empirical comparison of the impact of the different techniques is presented and discussed. The achieved performances were also compared with a standard epidemiological model. This investigation showed that, for the dataset under consideration, the best results were achieved with Random UnderSampling in conjunction with Random Forest (either with or without probability calibration). However, while including probability calibration slightly reduced the average performance, it increased the model robustness, as indicated by the lower 95% confidence intervals. The analysis showed that machine learning models could provide comparable results to standard epidemiological models while being completely data-driven and disease-agnostic, thus demonstrating the opportunity for building machine learning models on health records data for research and clinical practice. However, further testing is required to significantly improve the model performance and its robustness.