International Journal for Population Data Science最新文献

{"title":"Identifying drivers of food insecurity through linked data- the Priority Places for Food Index","authors":"Francesca Pontin, Peter Baudains, Emily Ennis, Michelle Morris","doi":"10.23889/ijpds.v8i3.2277","DOIUrl":"https://doi.org/10.23889/ijpds.v8i3.2277","url":null,"abstract":"Introduction & Background15.5% of all UK households are food insecure; either unable to afford to eat, skipping meals or reducing meal sizes despite being hungry. Drivers of food insecurity include both access to and affordability of food, with those most in need often left unable to access healthy and affordable food. Taking a place-based approach to understand the drivers of food insecurity allows for targeted support from government, third sector and private organisations to mitigate growing food insecurity in the UK.
 Objectives & ApproachThis research presents the methodology for the co-production and construction of the Priority Places for Food Index (PPFI) and supporting dashboard, co-designed by the Consumer Data Research Centre (CDRC) and consumer champions Which? in response to the 'cost-of-living crisis'.
 The PPFI equally weights measures of access to affordable food and indicators of barriers to affording food across seven domains; Proximity to supermarket retail facilities, Accessibility of supermarket retail facilities, Access to online deliveries, Proximity to non-supermarket food provision, Socio-economic barriers, Fuel Poverty and Family Food for support. The PPFI uses open data combing traditional census data metrics, with government data (e.g., Healthy start voucher and free-school meals uptake), digital footprints data (web-scraped delivery addresses and food bank item request data) and scaled-survey data (fuel poverty, propensity to shop online).
 Relevance to Digital FootprintsDigital footprint data can complement traditional data sources to provide a more nuanced view of health inequalities. These data are typically less timely to collect than traditional data collection methods (census, survey) allowing a more reactive response to emergent issues such as the cost-of-living crisis.
 ResultsThe PPFI interactive map and underlying data have been published via the CDRC https://priorityplaces.cdrc.ac.uk/.
 Conclusions & ImplicationsWe demonstrate the value of data linkage across individual and population level data to provide localised insight into food insecurity and identify where digital footprints data can improve gaps in the current evidence base. We also reflect on the value of co-production and stakeholder engagement in creating a policy ready interactive map which has facilitated the lobbying of targeted practical support and policy change to address food insecurity.","PeriodicalId":132937,"journal":{"name":"International Journal for Population Data Science","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135154246","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Predicting health related deprivation using loyalty card digital footprints","authors":"Gavin Long, Georgiana Nica-Avram, John Harvey, Roberto Mansilla, Simon Welham, Evgeniya Lukinova, James Goulding","doi":"10.23889/ijpds.v8i3.2282","DOIUrl":"https://doi.org/10.23889/ijpds.v8i3.2282","url":null,"abstract":"Introduction & BackgroundIn England, The Indices of Deprivation (IoD) are a widely used and referenced measure to assess local levels of deprivation across a range of domains, including health and disability. However, due to their complex nature and the number of inputs required to generate these measures, they are only updated infrequently. Typically every 4-5 years, with the most recent versions released in 2019 and 2015.
 This study expands on previous research looking at the feasibility of using digital footprint data, in the form of retail loyalty card transactions, to predict local deprivation. This work focuses specifically on the health and disability subdomain of IoD. Our hypothesis is that retail behaviour relating to food purchases and their associated nutritional content, can be used to predict health deprivation.
 Objectives & ApproachThe work utilises loyalty card data from a large UK grocery retailer. Anonymised geo-location data for loyalty card members was used to assign retail grocery transactions to individual Lower Layer Super Output Areas (LSOAs) for each of the ten quarters in the study period (July 2019 - December 2021). A nutritional lookup was developed to enable the nutritional content of food transactions to be assigned to each LSOA.
 A number of metrics based on categories of food sold and their nutritional content were developed and used in a Machine Learning model, based on a Random Forest classifier, to predict areas with high levels of health related deprivation.
 Relevance to Digital FootprintsThis study uses data derived from digital footprint data of grocery transactions. It demonstrates the potential for utilising digital footprint data as a proxy for traditional demographic data without the need for expensive, both in terms of time and cost, surveying to be performed.
 ResultsThe random forest classifier was able to predict neighbourhoods (at the LSOA level) with the top 20% of health related deprivation. A high level of predictive power was identified (Overall accuracy 80%). SHAP (SHapley Additive exPlanations) and Model Class Reliance (MCR) were used to determine the importance of the input features. Areas with higher proportional spending on cigarettes and soft drinks and lower spending on fish, wine and fruit and vegtables were found to be associated with extreme levels of health deprivation. In terms of nutrition, two derived metrics, calories per pound spend and the obesogenicity of food purchased, were found to be important predictors of health deprivation.
 Conclusions & ImplicationsDigital footprint data on grocery purchases have been shown to be highly effective at predicting areas of extreme health related deprivation at the LSOA level. Features related to proportional spend on food categories and proportions of nutrients associated with these purchases were identified as optimal for predicting health related deprivation.
 The number of calories per pou","PeriodicalId":132937,"journal":{"name":"International Journal for Population Data Science","volume":"224 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135154384","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Education for out-of-school children: Unpacking driving factors of vulnerability via digital-learning footprint data in East Africa","authors":"Bethany Huntington, Nicola Pitchford, James Goulding","doi":"10.23889/ijpds.v8i3.2293","DOIUrl":"https://doi.org/10.23889/ijpds.v8i3.2293","url":null,"abstract":"Introduction & BackgroundApproximately 617 million children and adolescents worldwide do not possess the foundational skills to live healthy and productive lives. Sub-Saharan Africa is profoundly affected, resulting in social and financial dependency and raising vulnerability to forced marriage, female genital mutilation, and mental health issues. Contextual factors are considered critical in this crisis, yet have received little attention due to a currently insurmountable deficit in traditional census and survey data. Digital footprint data offers a potential route to filling this information gap - particularly in education- and how a learner’s environment can impact digital learning and future well-being.
 To address the crisis, the ‘Global Learning’ XPRIZE competition challenged teams to develop software empowering marginalised out-of-school children to learn literacy and numeracy skills. Five finalist teams tested their technology with 2041 children using handheld tablets in 172 remote villages in Tanzania.
 Objectives & ApproachOur study examined factors that can predict improvements in learning outcomes, building on the digital footprint data collected from the children participating in the device intervention in the form of app usage and locational and activity data. Additional geospatial features were engineered based on village coordinates, distance to local amenities, services and transport, variables serving as additional potential indicators of isolation and connectedness. These data were linked with child-level factors, including household composition and literacy levels.
 After comparative assessment of machine learning regression models, tree-based models (XGB, RF) were used to establish the optimal predictive performance for literacy and numeracy. Variable importance using SHAP was used to determine which specific contextual variables should be considered before deploying digital interventions to support education and well-being.
 Relevance to Digital FootprintsUtilising digital footprint data to quantify the influence of geospatial and contextual data in digital interventions can offer comprehensive insights to understand and address factors impacting learning outcomes in this context.
 ResultsPrior school attendance, home reading environments and high familial literacy were found to be predictors of higher learning outcomes after a technology-based learning intervention. Environments featuring an unemployed caregiver and few siblings were surprisingly consistent positive predictors, suggesting accompanying and focussed caregiver support as valuable for effective development via digital interventions. Proximity to police stations and health centres were revealed as key predictors, indicating the importance of social and physical connectedness in positive learning environments.
 Conclusions & ImplicationsTargeted improvement of EdTech provision with out-of-school learners promises to help ","PeriodicalId":132937,"journal":{"name":"International Journal for Population Data Science","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135153267","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Daily, Weekly and Monthly Variation in Lunch Time Calories","authors":"Anya Skatova, Neil Stewart, Edward Flavahan, James Goulding","doi":"10.23889/ijpds.v8i3.2288","DOIUrl":"https://doi.org/10.23889/ijpds.v8i3.2288","url":null,"abstract":"Introduction & BackgroundDespite the level of attention that healthy and unhealthy eating receive from academic research, policymakers and the wider public, objective data on food consumption is limited. This is because studies of individual eating patterns using food diaries are subject to underreporting, particularly by people who are overweight. For example, the UK population is estimated to consume between 30% to 50% more calories than they report in surveys. New data sources such as office canteen ordering systems and individual records of supermarket transactions recorded through supermarket loyalty or bonus cards offer larger and potentially more robust data on real world individual eating behaviours.
 Objectives & ApproachWe used 2,831,403 machine-recorded ‘meal deal’ transactions from 205,781 individuals over the course of one year from one of the UK’s largest suppliers of lunch time foods to investigate whether there is a relationship between patterns of choice and higher calorie consumption. A meal deal comprises three items; a main (e.g., a sandwich or a salad), a snack (e.g., crisps, fruit or a chocolate bar) and a drink (e.g., a smoothie or a bottle of water). In contrast to diary studies or aggregate transactional data from supermarkets, our dataset included “meal deal’ purchase which is highly likely to be made by an individual for their own consumption or soon afterwards.
 Relevance to Digital FootprintsLunch time food consumption can reflect the overall diet the individual is exposed to, helping to understand population level patterns of people’s food choices through a type of digital footprints data - shopping history records.
 ResultsControlling for gender, general index of variety in the choice of lunch food items, income and education, we found that individuals who vary in their calorie consumption most across the time of day, day of the week, and month of the year are the individuals who consume the greatest number of calories overall. These time sensitivity effects are large, together explaining a substantial amount of variance in calorie consumption. Time sensitivity effects are strongly correlated across all three time scales suggesting they measure a stable underlying trait.
 Conclusions & ImplicationsIndividuals vary calorific composition of their lunch over time of the day, day of the week and month of the year by 100 calories per meal between highest and lowest in sensitivity which is about 9% of the recommended amount of lunchtime calories. Those whose consumption varies the most with time consume the most calories, independently of income and gender. The variation in calories at all three time scales demonstrates the properties of an individual disposition. These findings can be used to understand why and when people make unhealthy food choices.","PeriodicalId":132937,"journal":{"name":"International Journal for Population Data Science","volume":"121 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135153268","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The Kids’ Environment and Health Cohort: a national, linked data resource for environmental child health research","authors":"Alister Baird, Emilie Courtin, Steven Cummins, Samantha Hajna, Vahé Nafilyan, Alison Macfarlane, Jess Walkeden, Pia Hardelid","doi":"10.23889/ijpds.v8i3.2291","DOIUrl":"https://doi.org/10.23889/ijpds.v8i3.2291","url":null,"abstract":"Introduction & BackgroundEvidence is mounting that children’s physical environment (e.g. in and around the home, school, and neighbourhood) is critical for their long-term health and education. Early life exposure to factors such as indoor and outdoor air pollution, or a lack of access to greenspaces are associated with the development of long-term health conditions such as asthma or mental health problems. Local and central government in England are implementing numerous policies to improve air quality and housing, and mitigate climate change. Further, England has seen large scale changes to local service provision (including childcare and libraries) due to austerity policies and the COVID-19 pandemic. Currently, there is no national, linked data resource for England that allows research into how the local environment impacts children’s health and education.
 Objectives & ApproachThe Kids’ Environment and Health Cohort will be a new, linked national data resource for England currently being developing by researchers from UCL, London School of Hygiene and Tropical Medicine, London School of Economics and Political Science, Brock University, and City, University of London in collaboration with the Office for National Statistics (ONS), and funded by Administrative Data Research-UK (ADR-UK). The Kids’ Environment and Health Cohort will be a de-identified and annually updated national birth cohort of all children born in England from 2006 onwards – around 10.5 million children until 2023. The cohort will be constructed using linked administrative data from vital registration (live and stillbirth, and death registration), Census (housing and socio-economic indicators), health (hospital contacts, mental health referrals, and community dispensing data), and education (key stage results, special educational needs, absenteeism). Environmental exposure data can be securely linked to the Cohort via longitudinal residential unique property reference numbers (UPRNs) and postcodes from the Personal Demographic Service, and school location from education records.
 Relevance to Digital FootprintsThe Kids’ Environment and Health Cohort will, for the first time, link health, education, Census and environmental data at national level in England. It will allow researchers to integrate data on local environments, including physical characteristics (such as temperature, building energy efficiency, or greenspace access) or the social environment (including proximity to food outlets, or services like libraries) with individual level data on health and education outcomes in children. This will be done using the ONS’s 5 safes framework, ensuring highest standards of data security and confidentiality.
 ResultsThe Kids’ Environment and Health Cohort will be constructed using administrative datasets, including national linked vital statistics, health, education and Census data from multiple data providers (ONS, NHS England and Department for Education)","PeriodicalId":132937,"journal":{"name":"International Journal for Population Data Science","volume":"458 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135153436","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Studying insulin needs in Type 1 Diabetes by analysing the OpenAPS Data Commons","authors":"Isabella Degen, Kate Robson Brown, Zahraa S Abdallah","doi":"10.23889/ijpds.v8i3.2271","DOIUrl":"https://doi.org/10.23889/ijpds.v8i3.2271","url":null,"abstract":"Introduction & BackgroundType 1 Diabetes (T1D) is a chronic condition where the body produces too little insulin, a hormone required to regulate blood glucose (BG). Finding the correct insulin dose and time remains a complex and as yet unsolved control task. Many factors such as food, exercise, stress, menstrual cycle, etc. change how much insulin is required. Most of these factors remain unobserved and/or underexplored. The NHS national diabetes audit shows that in 2020-21 only 9.8% of people with T1D had a NICE recommended glycated haemoglobin (HbA1c) test result of 48 mmol/mol or less to avoid complications due to diabetes.
 Objectives & ApproachOur research aims to improve the understanding of underexplored factors that drive changes in insulin needs of people with T1D. We use unsupervised time series pattern detection methods such as time series k-means, heatmaps and matrix profile to identify underexplored patterns. These are times when insulin on board (IOB) does not rise with more carbohydrates on board (COB) and times when BG does not fall with more IOB and/or rise with more COB as expected. In the future, we aim to use causal feature selection methods to exclude COB as a causal driver for these patterns and identify other factors as causal drivers.
 Relevance to Digital FootprintsWe use a digital footprints dataset of n=187 people who use an open-source automated insulin delivery system and have donated their data via the OpenAPS Data Commons and the OpenHumans.org platform. The data has been collected in real-life conditions and contains system logs, BG sensor data and various user-entered annotations. The richness of the data provides a unique opportunity to study what happens in real-life conditions but also poses challenges for many methods. These include inconsistencies between users, irregular sampling between devices and missing data.
 ResultsOur pattern detection methods can successfully identify underexplored patterns in insulin needs that are not directly driven by COB. These include months that require more/less insulin without eating more/less carbohydrates and times of day when blood glucose does not rise with more COB and/or fall with more IOB.
 Conclusions & ImplicationsWhile our current methods can identify underexplored patterns in insulin needs, the principal pattern identified remains the well-known pattern of the main meal COB spikes. This is not surprising given the frequency and impact of this pattern. We are working on methods that can identify both frequent and less frequent patterns as well as patterns arising from the irregularity of the sampling. Digital footprints datasets like the OpenAPS Data Commons are promising to help increase understanding of complex conditions such as T1D. More of this type of multi-year data is needed. Especially data that includes multiple different sensor readings to help identify causal drivers behind underexplored patterns of insulin needs. And we ","PeriodicalId":132937,"journal":{"name":"International Journal for Population Data Science","volume":"170 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135154084","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Generating a Single Session Outcome Measure from Digital Mental Health Platform Footprints Using Natural Language Processing","authors":"Gregor Milligan, Aynsley Bernard, Liz Dowthwaite, Elvira Perez Vallejos, James Goulding","doi":"10.23889/ijpds.v8i3.2269","DOIUrl":"https://doi.org/10.23889/ijpds.v8i3.2269","url":null,"abstract":"Introduction & BackgroundThis work demonstrates the development of an Adult Session Wants and Needs Outcome Measure (Adult SWAN-OM) aimed at supporting service delivery within the digital mental health platform (DMHP), Qwell. Qwell is a DMHP commissioned by the United Kingdom’s National Health Service which provides access to an online community of peers, a team of experienced counsellors, and a cadre of emotional well-being practitioners. The service is anonymous at point of entry and free for users, and provides an extensive, person-centred approach which results in a wide range of user needs. Deriving outcome measures from the platform’s varied counselling sessions, aims to provide both insights into the want and needs of users and underpin improved mental health support. Objectives & ApproachThe objective of this research is to show how contemporary machine learning methods (Transformer Models and Contextualised Topic Modelling) may be combined with digital footprint data (in the form of seldom explored text data generated on DMHPs) to identify service user wants and needs. Specifically, with automated inference of patient outcomes currently scarce, we focus on the development of outcome measures in the context of ‘single sessions’, applying machine learning methods to extract topics related to the wants and needs of service users on DMHPs. Relevance to Digital FootprintsThe data used in this study consisted of transcripts between Qwell practitioners and service users (SU’s) (N=874) at conversation level (N=2323), a filter was applied to the dataset to ensure that focus the SUs elicitation of wants and needs fit into the criteria of a single session. Individuals in the final selected cohort (n=192) are not significantly different from the wider Qwell SU population in the study period in terms of age, gender or ethnicity; suggesting that the cohort is representative of the wider target population. This study shows the potential of mental health digital footprints data when providing insight into the wants and needs of DMHP SUs. Conclusions & ImplicationsVia this analysis of mental health digital footprint data, this work establishes a process for creating a new outcome measure through the computational analysis of transcript data, incorporating insights from clinical experts and individuals with lived experience of engaging with DMHPs with textual data analysis. This methodological approach of Transformer Models and Contextualized Topic Modelling enables the analysis of a considerable volume of data faster than manually reviewing transcripts. We offer suggestions for the refinement of automated methods, in collaboration with direct support and feedback from both clinicians and individuals with lived experience of DMHPs to enable the understanding of wants and needs of service users within DMHPs.","PeriodicalId":132937,"journal":{"name":"International Journal for Population Data Science","volume":"170 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135154237","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Decoy Effects in a Massive Real-World Shopping Dataset","authors":"Sean Devine, James Goulding, Anya Skatova, Ross Otto","doi":"10.23889/ijpds.v8i3.2286","DOIUrl":"https://doi.org/10.23889/ijpds.v8i3.2286","url":null,"abstract":"Introduction & BackgroundA key premise of rational choice prescribes that decision-makers ought to ignore irrelevant, inferior alternative options. Consider for example the choice between two wines, where the value of an option is computed across two dimensions: quality and price. When deliberating about which wine to choose, one’s propensity to choose between two otherwise equally preferred wines should be not influenced by the introduction of a third clearly inferior option (being both of lower quality and more expensive than one of the original alternatives). Yet, a large body of work suggests that both people and animals routinely violate this premise in their decisions—in laboratory experiments, the introduction of irrelevant “decoys” into a choice set biases decision-making. However, these decoy effects are less understood in large-scale contexts of real-world decision-making, where choice sets can be large, and preference informed by consumers’ histories of experience.
 Objectives & ApproachWe explored whether the presence of irrelevant, “decoy” alternative options influenced wine purchases in a large real-world dataset of UK wine purchases. From shopping transaction records, we extracted all red and white wine purchases over a one month period. Our analyses examined 3.6M wine purchases made by 755,158 unique customers.
 Relevance to Digital FootprintsWe deployed shopping history data which is a popular example for digital footprints allowing us to track people’s choices and decisions over long periods of time.
 ResultsWe find that among pairs of wines that appear across many different contexts (i.e., stores with different product assortments) and trade off on quality and price, the presence of decoy options— similar options that were dominated by the focal option—made consumers more likely to purchase the focal option (a hallmark of the “attraction effect”). Furthermore, we find that sensitivity to this effect depended on consumers’ history of experience with the product, such that frequent shoppers were less likely to be sensitive to decoy effects in their purchase behaviour.
 Conclusions & ImplicationsWe examined whether real-world consumer decisions, evidenced in a large dataset of wine purchases in the United Kingdom, were subject to a canonical bias in multiattribute choice: the attraction effect. We found that wine purchases were systematically biased in favour of wines that dominated choice sets—a bias which was not observed when choice sets were not dominated. Together, these results extend laboratory-based accounts of decoy effects to real-world choices, and demonstrate how digital footprints data analysis can be linked to health, especially in terms of decision making which is associated with negative health outcomes.","PeriodicalId":132937,"journal":{"name":"International Journal for Population Data Science","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135154385","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Investigating the Relationship between Timing of Tweets and Mental Health, Well-being and Sleep Quality in a UK birth cohort","authors":"Oliver Davis, Nina Di Cara, Nello Cristianini, Claire Haworth","doi":"10.23889/ijpds.v8i3.2284","DOIUrl":"https://doi.org/10.23889/ijpds.v8i3.2284","url":null,"abstract":"Introduction & BackgroundSocial media use has been proposed as a cause of worsening mental health and wellbeing over the last decade, but its role in mitigating some of the effects of social distancing during the pandemic showed that it also has the potential to improve these outcomes. Whilst existing research disagrees on the degree to which social media use harms or helps, there is growing consensus around the need to move from global measures of social media use to specific measures of types of social media use. These new measures can enable an exploration of proposed mechanisms and causal pathways linking social media use and mental health and wellbeing. A commonly proposed mechanism is nighttime social media use reducing sleep quality, and consequently harming mental health and wellbeing.
 Objectives & ApproachWe aimed to investigate the relationships between the time Twitter users post content and their mental health, wellbeing and sleep quality using direct measurements of Twitter use linked to standardised mental health measures in a well-characterized cohort.
 This study uses approximately 1.5 million Tweets harvested between January 2008 and March 2023 from 622 participants in the Avon Longitudinal Study of Parents and Children (ALSPAC). These Tweets have been linked to questionnaire data collected on six occasions spanning April 2019 to May 2021. These questionnaires included standard measures of depressive symptoms, anxiety symptoms, mental wellbeing and difficulty sleeping.We have taken two approaches to explore these relationships, using circular statistical methods novel to social media data analysis to account for day/night cycles. The first approach used mixed effect models to investigate the association between the time a Tweet was posted and the mental health, mental wellbeing and sleep quality of the poster. The second approach explored the relationships between the mean hour participants post Tweets in a given time period, and their mental health, mental wellbeing and sleep quality.
 Relevance to Digital FootprintsThis research is highly relevant to Digital Footprints, due to its use of data directly extracted from a social media site. The methodologies employed in analysing this alongside more traditional epidemiological survey data provides an example of how digital footprint data can complemented by high quality ground truths.
 ResultsThere was evidence that the timing of Twitter activity was predictive of the mental wellbeing and sleep quality of participants, even after adjustment for demographic, educational and socio-economic covariates. However, the hour a Tweet was posted at explained very little of the variation in the mental wellbeing or sleep quality of the participant who posted it (0.1% and less than 0.1% respectively). There was weak to no evidence that the timing of Twitter activity was predictive of the depressive and anxiety symptoms of participants.
 Conclusions & Implic","PeriodicalId":132937,"journal":{"name":"International Journal for Population Data Science","volume":"188 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135153270","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Healthy foods, healthy sales? Cross-category spillover effects of a reward program promoting sales of fruit and vegetables","authors":"Luca Panzone, Barbara Tocco, Ruzica Brecic, Matthew Gorton","doi":"10.23889/ijpds.v8i3.2270","DOIUrl":"https://doi.org/10.23889/ijpds.v8i3.2270","url":null,"abstract":"Introduction & BackgroundGlobally, consumption of Fruit and Vegetables (F&V) remains below nutritional guidelines. Marketing healthy products can be complex for retailers, and a key challenge is the design of strategies that benefit retailers, e.g., through improved loyalty, and deliver progress on societal goals.
 Objectives & ApproachThis study evaluates a point-plus-cash frequency reward program where participants received points by purchasing selected F&V, redeemable against a reward (plush toys in the shape of F&V). We estimate the impact of the program using a difference-in-difference-in-difference model, which compares expenditures in several categories before, during, and after the promotional period, across two different years, and separately for consumers who redeemed a reward and those who did not. Identification includes weighting for the propensity scores, and using an instrumental variable approach.
 Relevance to Digital FootprintsThe data refers to grocery expenditure in five categories in the focal retailer for over 268,000 consumers, over 27 weeks for 2 years, as recorded through their loyalty card.
 ResultsThe reward program significantly increased expenditures in F&V in the focal retailer during the promotional period. However, results differed depending on reward redemption. For reward-redeemers, the program increased expenditures in F&V as well as in other food categories, an effect that persisted – at a declining rate – after the program stopped. The program had a short-lived effect on non-rewards redeemers, who only increased F&V expenditures during the promotional period.
 Conclusions & ImplicationsResults indicate that a loyalty program promoting sales of F&V can create win-win benefits to both society and the retailer: it increases expenditures on healthy foods (F&V), while improving overall loyalty (i.e., expenditures) to the retailer.","PeriodicalId":132937,"journal":{"name":"International Journal for Population Data Science","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135154240","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}