Ethics of early detection of disease risk factors: A scoping review.

BACKGROUND: Scientific and technological advancements in mapping and understanding the interrelated pathways through which biological and environmental exposures affect disease development create new possibilities for detecting disease risk factors. Early detection of such risk factors may help prevent disease onset or moderate the disease course, thereby decreasing associated disease burden, morbidity, and mortality. However, the ethical implications of screening for disease risk factors are unclear and the current literature provides a fragmented and case-by-case picture. METHODS: To identify key ethical considerations arising from the early detection of disease risk factors, we performed a systematic scoping review. The Scopus, Embase, and Philosopher's Index databases were searched for peer-reviewed, academic records, which were included if they were written in English or Dutch and concerned the ethics of (1) early detection of (2) disease risk factors for (3) disease caused by environmental factors or gene-environment interactions. All records were reviewed independently by at least two researchers. RESULTS: After screening 2034 titles and abstracts, and 112 full papers, 55 articles were included in the thematic synthesis of the results. We identified eight common ethical themes: (1) Reliability and uncertainty in early detection, (2) autonomy, (3) privacy, (4) beneficence and non-maleficence, (5) downstream burdens on others, (6) responsibility, (7) justice, and (8) medicalization and conceptual disruption. We identified several gaps in the literature, including a relative scarcity of research on ethical considerations associated with environmental preventive health interventions, a dearth of practical suggestions on how to address expressed concerns about overestimating health capacities, and a lack of insights into preventing undue attribution of health responsibility to individuals. CONCLUSIONS: The ethical concerns arising with the early detection of risk factors are often interrelated and complex. Comprehensive ethical analyses are needed that are better embedded in normative frameworks and also assess and weigh the expected benefits of early risk factor detection. Such research is necessary for developing and implementing responsible and fair preventive health policies.

Exposure to images showing (non)adherence to physical distancing rules: Effect on adherence behavior and perceived social norms.

INTRODUCTION: Adherence to behavioral measures such as physical distancing are key to mitigating the effects of viral pandemics such as the COVID-19 pandemic. Adherence depends in part on people's perception of what others do (descriptive norms) or approve of (injunctive norms). This study examines the effects that exposure to images depicting people following or breaking physical distancing rules have on perceptions of descriptive and injunctive norms and subsequent adherence behavior. METHODS: An online between-subjects experiment (n = 315) was conducted, in which participants were exposed to a set of five photographs of different public spaces in which people either did or did not adhere to physical distancing rules (pre-registration: https://www.osf.io/uek2p). Participants' adherence behavior was measured using a triangulation of measures (incentivized online behavioral task, vignettes, intention measure). Perceptions of relevant social norms were also measured. RESULTS: Mann-Whitney tests showed no effects of condition on perceptions of descriptive and injunctive norms or on adherence behavior. Linear regressions showed that both component paths of the indirect effect (condition on norm perceptions, and norm perceptions on adherence behavior) were non-significant, hence mediation analyses were not conducted. CONCLUSIONS: Exposure to images of people following (compared to breaking) physical distancing rules did not affect adherence to such rules or perceived norms. We surmise that a single exposure to such images, especially in the context of COVID-19, is insufficient to affect behavior. We therefore recommend performing a comparable experiment in which participants are exposed repeatedly to images showing people (non)adhering to a specific behavior in a particular context for a longer period.

Accurate training of the Cox proportional hazards model on vertically-partitioned data while preserving privacy.

BACKGROUND: Analysing distributed medical data is challenging because of data sensitivity and various regulations to access and combine data. Some privacy-preserving methods are known for analyzing horizontally-partitioned data, where different organisations have similar data on disjoint sets of people. Technically more challenging is the case of vertically-partitioned data, dealing with data on overlapping sets of people. We use an emerging technology based on cryptographic techniques called secure multi-party computation (MPC), and apply it to perform privacy-preserving survival analysis on vertically-distributed data by means of the Cox proportional hazards (CPH) model. Both MPC and CPH are explained. METHODS: We use a Newton-Raphson solver to securely train the CPH model with MPC, jointly with all data holders, without revealing any sensitive data. In order to securely compute the log-partial likelihood in each iteration, we run into several technical challenges to preserve the efficiency and security of our solution. To tackle these technical challenges, we generalize a cryptographic protocol for securely computing the inverse of the Hessian matrix and develop a new method for securely computing exponentiations. A theoretical complexity estimate is given to get insight into the computational and communication effort that is needed. RESULTS: Our secure solution is implemented in a setting with three different machines, each presenting a different data holder, which can communicate through the internet. The MPyC platform is used for implementing this privacy-preserving solution to obtain the CPH model. We test the accuracy and computation time of our methods on three standard benchmark survival datasets. We identify future work to make our solution more efficient. CONCLUSIONS: Our secure solution is comparable with the standard, non-secure solver in terms of accuracy and convergence speed. The computation time is considerably larger, although the theoretical complexity is still cubic in the number of covariates and quadratic in the number of subjects. We conclude that this is a promising way of performing parametric survival analysis on vertically-distributed medical data, while realising high level of security and privacy.

Privacy-preserving dataset combination and Lasso regression for healthcare predictions.

BACKGROUND: Recent developments in machine learning have shown its potential impact for clinical use such as risk prediction, prognosis, and treatment selection. However, relevant data are often scattered across different stakeholders and their use is regulated, e.g. by GDPR or HIPAA. As a concrete use-case, hospital Erasmus MC and health insurance company Achmea have data on individuals in the city of Rotterdam, which would in theory enable them to train a regression model in order to identify high-impact lifestyle factors for heart failure. However, privacy and confidentiality concerns make it unfeasible to exchange these data. METHODS: This article describes a solution where vertically-partitioned synthetic data of Achmea and of Erasmus MC are combined using Secure Multi-Party Computation. First, a secure inner join protocol takes place to securely determine the identifiers of the patients that are represented in both datasets. Then, a secure Lasso Regression model is trained on the securely combined data. The involved parties thus obtain the prediction model but no further information on the input data of the other parties. RESULTS: We implement our secure solution and describe its performance and scalability: we can train a prediction model on two datasets with 5000 records each and a total of 30 features in less than one hour, with a minimal difference from the results of standard (non-secure) methods. CONCLUSIONS: This article shows that it is possible to combine datasets and train a Lasso regression model on this combination in a secure way. Such a solution thus further expands the potential of privacy-preserving data analysis in the medical domain.

The Explanations People Give for Going to Bed Late: A Qualitative Study of the Varieties of Bedtime Procrastination.

Background/Objective: Bedtime procrastination is a prevalent cause of sleep deprivation, but little is known about why people delay their bedtimes. In the present research, we conducted a qualitative study with bedtime procrastinators to classify their self-reported reasons for later-than-intended bedtime. Participants: Participants (N = 17) were selected who frequently engaged in bedtime procrastination, but whose sleep was not otherwise affected by diagnosed sleep disorders or shift work. Method: We conducted in-depth, semistructured interviews and used thematic analysis to identify commonly recurring themes in the interviews. Results and conclusions: Three emerging themes were identified: deliberate procrastination, mindless procrastination, and strategic delay. For the form of procrastination we classified as deliberate procrastination, participants typically reported wilfully delaying their bedtime because they felt they deserved some time for themselves. For the category of mindless procrastination, a paradigmatic aspect was that participants lost track of the time due to being immersed in their evening activities. Finally, participants who engaged in strategic delay reported going to bed late because they felt they needed to in order to fall asleep (more quickly), which suggests that despite describing themselves as "procrastinating," their bedtime delay may actually be linked to undiagnosed insomnia. The conceptual distinctions drawn in this paper deepen our understanding of bedtime delay and may be helpful for designing effective interventions.

Too Depleted to Turn In: The Relevance of End-of-the-Day Resource Depletion for Reducing Bedtime Procrastination.

Bedtime procrastination is an important predictor of sleep insufficiency in the general population (Kroese et al., 2014b), but little is known about the determinants of this self-undermining behavior. As the phenomenon has been conceptualized in the literature as a form of self-regulation failure (Kroese et al., 2014a), we hypothesized that people's self-regulatory resources in the evening would be predictive of going to bed later than they intended. Specifically, we examined whether the cumulative effect of resisting desires, a measure of self-regulatory resource depletion (Hofmann et al., 2012b), relates to bedtime procrastination. Participants (N = 218) reported how many desires they had tried to resist during the previous day and the extent of their bedtime procrastination. Results show that people who attempted to resist more desires were more likely to engage in bedtime procrastination, suggesting that people may be less likely to stick to their intended bedtime after a particularly taxing day. Implications for intervention strategies are discussed.