Study protocol - assessing parkrun for walking rehabilitation for people living with, and beyond, cancer: acceptability, adherence, social support and physical function.

INTRODUCTION: Due to a variety of barriers, the majority of cancer survivors do not do enough physical activity to meet current recommendations. This study will assess the feasibility of participation in parkrun walk-run events as a novel mode of community rehabilitation exercise. METHODS: This protocol describes a single-arm intervention study with participants acting as their own controls. The study accepts adults diagnosed with any type of cancer, undergoing treatment or in remission. Participants must be able to walk and have medical clearance to exercise. A sample of 100 participants will be recruited across the Sunshine Coast over two years. Data will be collected over 9-months at 4 time points: Baseline (T1); after 4-weeks of usual daily activities and cancer management prior to parkrun participation(T2); after a 6-month parkrun intervention (T3); at 2-month follow-up (T4). The primary objectives are to assess the acceptability of, and adherence to, parkrun as rehabilitation exercise. Secondary outcomes include wellness, health-related quality of life, anxiety, depression, mood, physical function, parkrun metrics, dietary intake, and diet and exercise behaviour. CONCLUSION: This study will be the first to examine the long-term effects of parkrun as a cancer rehabilitation modality with regard to physical function, psychosocial outcomes and dietary intake. TRIAL REGISTRATION: Australian and New Zealand Clinical Trials Registry ACTRN12623000473662 registered 09/05/2023.Approved by UniSC Human Research Ethics Committee (A221828) and the UK parkrun Research Board. Original protocol. Authors SB, RB, HHW, MM, YK.

The effect of dietary interventions or patterns on the cardiometabolic health of individuals treated with androgen deprivation therapy for prostate cancer: A systematic review.

Prostate cancer survivors treated with androgen deprivation therapy may be at increased risk of cardiovascular disease. Dietary recommendations for the prevention and/or management of cardiovascular disease for these individuals are lacking. This review synthesizes the evidence on the effect of dietary interventions on cardiometabolic biomarkers and cardiovascular disease risk in prostate cancer survivors receiving androgen deprivation therapy. A systematic review was conducted across PubMed, CINAHL, Embase, and Cochrane CENTRAL. Intervention or observational cohort studies evaluating diets, nutrients, or nutraceuticals with or without concurrent exercise interventions on cardiovascular disease, cardiovascular events, or cardiovascular disease biomarkers in those treated with androgen deprivation therapy were included. Confidence in the body of evidence was appraised using Grading of Recommendations, Assessment, Development and Evaluations. Twelve studies reported across fifteen papers were included. Interventions were heterogenous, with most studies including an exercise co-intervention (n = 8). Few significant findings for the effects of diet on cardiometabolic markers were likely due to weak methodology and sample sizes. Strongest evidence was for the effect of a healthy Western dietary pattern with exercise on improved blood pressure (Confidence: moderate). The healthy Western dietary pattern with exercise may improve high-density lipoprotein cholesterol (Confidence: Low) and flow-mediated dilation. Soy may improve total cholesterol (Confidence: Very low). A low-carbohydrate diet with physical activity may improve high-density lipoprotein cholesterol, incidence of metabolic syndrome, and Framingham cardiovascular disease risk score. Evidence of the effect of dietary interventions on cardiometabolic biomarkers and cardiovascular disease risk of prostate cancer survivors receiving androgen deprivation therapy is insufficient to inform practice. Well-designed dietary interventions aimed at improving cardiometabolic outcomes of this population are warranted to inform future dietary recommendations.

Fasted C-Peptide Distribution and Associated Clinical Factors in Adults With Longstanding Type 1 Diabetes: Analysis of the Canadian Study of Longevity in Type 1 Diabetes.

OBJECTIVE: Although insulin production is reportedly retained in many people with longstanding type 1 diabetes (T1D), the magnitude and relevance of connecting peptide (C-peptide) production are uncertain. In this study, we aimed to define fasted C-peptide distributions and associated clinical factors. METHODS: In a cross-sectional analysis of the Canadian Study of Longevity, fasted serum and urinary C-peptide was measured in 74 patients with longstanding T1D (duration ≥50 years) and 75 age- and sex-matched controls. Extensive phenotyping for complications was performed and patient-reported variables were included. C-peptide distributions were analyzed, and multivariable logistic regression was used to assess the variable association in participants with T1D. RESULTS: The 74 participants with T1D had a mean age of 66±8 years, a disease duration of 54 (interquartile range 52 to 58) years, and a glycated hemoglobin (A1C) of 7.4%±0.8% (56.8±9.15 mmol/mol). The 75 controls had a mean age of 65±8 years and an A1C of 5.7%±0.4% (38.4±4.05 mmol/mol). Participants with T1D had lower fasted serum C-peptide than controls (0.013±0.022 vs 1.595±1.099 nmol/L, p<0.001). Of the participants with T1D, C-peptide was detectable in 30 of 73 (41%) serum samples, 32 of 74 (43%) urine samples, and 48 of 74 (65%) for either serum or urine. The variables independently associated with detectable serum or urinary C-peptide were lower total daily insulin requirement (odds ratio 2.351 [for 1 lower unit/kg], p=0.013) and lower hypoglycemia worry score (odds ratio 1.059 [for 1 point lower on the worry subscore of the Hypoglycemia Fear Survey], p=0.030). CONCLUSIONS: Although detectable C-peptide in longstanding diabetes was common, the magnitude of concentration was extremely low when compared with age- and sex-matched controls. Despite minimal detectability, its presence is validated by lower insulin requirements and strongly associated with lower hypoglycemia worry.

Commentary on the Revision of SAMHSA's National Outcome Measures (NOMs) for Discretionary Programs.

Data collection is an integral part of government agencies like the Substance Abuse and Mental Health Administration (SAMHSA), for reporting program outcomes and accountability. SAMHSA-funded community behavioral health programs have been evaluated by the National Outcome Measures (NOMs) since 2007. NOMs collects data on important aspects of client health including but not limited to mental health symptoms, functioning, and social connectedness through interviews with clients. Since its inception, NOMs has gone through a number of revisions. The most recent revision in 2021 has significant implications for program evaluation and research. This commentary provides an overview of the history of the NOMs followed by a review and critique of the recent changes with a particular attention to revisions in how responses are recorded. Implications of the NOMs are discussed with respect to its utility in evaluation, practice, and research.

A phase II trial of nivolumab followed by ipilimumab and nivolumab in advanced non-clear-cell renal cell carcinoma.

OBJECTIVE: To evaluate the efficacy of sequential treatment with ipilimumab and nivolumab following progression on nivolumab monotherapy in individuals with advanced, non-clear-cell renal cell carcinoma (nccRCC). MATERIALS AND METHODS: UNISoN (ANZUP1602; NCT03177239) was an open-label, single-arm, phase 2 clinical trial that recruited adults with immunotherapy-naïve, advanced nccRCC. Participants received nivolumab 240 mg i.v. two-weekly for up to 12 months (Part 1), followed by sequential addition of ipilimumab 1 mg/kg three-weekly for four doses to nivolumab if disease progression occurred during treatment (Part 2). The primary endpoint was objective tumour response rate (OTRR) and secondary endpoints included duration of response (DOR), progression-free (PFS) and overall survival (OS), and toxicity (treatment-related adverse events). RESULTS: A total of 83 participants were eligible for Part 1, including people with papillary (37/83, 45%), chromophobe (15/83, 18%) and other nccRCC subtypes (31/83, 37%); 41 participants enrolled in Part 2. The median (range) follow-up was 22 (16-30) months. In Part 1, the OTRR was 16.9% (95% confidence interval [CI] 9.5-26.7), the median DOR was 20.7 months (95% CI 3.7-not reached) and the median PFS was 4.0 months (95% CI 3.6-7.4). Treatment-related adverse events were reported in 71% of participants; 19% were grade 3 or 4. For participants who enrolled in Part 2, the OTRR was 10%; the median DOR was 13.5 months (95% CI 4.8-19.7) and the median PFS 2.6 months (95% CI 2.2-3.8). Treatment-related adverse events occurred in 80% of these participants; 49% had grade 3, 4 or 5. The median OS was 24 months (95% CI 16-28) from time of enrolment in Part 1. CONCLUSIONS: Nivolumab monotherapy had a modest effect overall, with a few participants experiencing a long DOR. Sequential combination immunotherapy by addition of ipilimumab in the context of disease progression to nivolumab in nccRCC is not supported by this study, with only a minority of participants benefiting from this strategy.

Supermarket Transaction Records In Dietary Evaluation: the STRIDE study: validation against self-reported dietary intake.

OBJECTIVE: Scalable methods are required for population dietary monitoring. The Supermarket Transaction Records In Dietary Evaluation (STRIDE) study compares dietary estimates from supermarket transactions with an online FFQ. DESIGN: Participants were recruited in four waves, accounting for seasonal dietary variation. Purchases were collected for 1 year during and 1 year prior to the study. Bland-Altman agreement and limits of agreement (LoA) were calculated for energy, sugar, fat, saturated fat, protein and sodium (absolute and relative). SETTING: This study was partnered with a large UK retailer. PARTICIPANTS: Totally, 1788 participants from four UK regions were recruited from the retailer's loyalty card customer database, according to breadth and frequency of purchases. Six hundred and eighty-six participants were included for analysis. RESULTS: The analysis sample were mostly female (72 %), with a mean age of 56 years (sd 13). The ratio of purchases to intakes varied depending on amounts purchased and consumed; purchases under-estimated intakes for smaller amounts on average, but over-estimated for larger amounts. For absolute measures, the LoA across households were wide, for example, for energy intake of 2000 kcal, purchases could under- or over-estimate intake by a factor of 5; values could be between 400 kcal and 10000 kcal. LoA for relative (energy-adjusted) estimates were smaller, for example, for 14 % of total energy from saturated fat, purchase estimates may be between 7 % and 27 %. CONCLUSIONS: Agreement between purchases and intake was highly variable, strongest for smaller loyal households and for relative values. For some customers, relative nutrient purchases are a reasonable proxy for dietary composition indicating utility in population-level dietary research.

Contrasting socio-economic influences on colorectal cancer incidence and survival in England and Wales.

Colorectal cancer (CRC) is the third most commonly diagnosed cancer in the world and second most common cause of cancer death. The relationship between socio-economic deprivation and CRC incidence is unclear and previous findings have been inconsistent. There is stronger evidence of an association between area-level deprivation and CRC survival; however, few studies have investigated the association between individual-level socio-economic status (SES) and CRC survival. Data from the Office for National Statistics Longitudinal Study (LS) in England and Wales was used. LS members aged 50+ were stratified by individual-level educational attainment, social class, housing tenure and area deprivation quintile, measured at the 2001 Census. Time-to-event analysis examined associations between indicators of SES and CRC incidence and survival (all-cause and CRC death), over a 15-year follow-up period. Among 178116 LS members, incidence of CRC was lower among those with a degree, compared to those with no degree and higher among those employed in manual occupations compared to non-manual occupations. No clear relationship was observed between CRC incidence and the area-based measure of deprivation. Disparities were greater for survival. Among 5016 patients diagnosed with CRC aged 50+, probability of death from all-causes was lower among those with a degree, compared to no degree and higher among those employed in manual occupations, compared to non-manual occupations and among those living in social-rented housing, compared to owner-occupiers. Individual indicators of SES were also associated with probability of death from CRC. Those living in the most deprived areas had a higher probability of death (from all-causes and CRC) compared to those in the least deprived areas. Both individual and area-based indicators of SES were associated with CRC survival, and the relationships were stronger than those observed for CRC incidence. These findings could help inform more effective targeting of public health interventions for CRC.

Supermarket top-up of Healthy Start vouchers increases fruit and vegetable purchases in low-income households.

Stark, widening health and income inequalities in the United Kingdom underpin the need for increased support for low-income families to access affordable and nutritious foods. Using anonymised supermarket loyalty card transaction records, this study aimed to assess how an additional Healthy Start voucher (HSV) top-up of £2, redeemable only against fruit and vegetables (FVs), was associated with FV purchases among at-risk households. Transaction and redemption records from 150 loyalty card-holding households, living in northern England, who had engaged with the top-up scheme, were analysed to assess the potential overall population impact. Using a pre-post study design, 133 of these households' records from 2021 were compared with equivalent time periods in 2019 and 2020. Records were linked to product, customer and store data, permitting comparisons using Wilcoxon matched-pairs sign-ranked tests and relationships assessed with Spearman's Rho. These analyses demonstrated that 0.9 more portions of FV per day per household were purchased during the scheme compared to the 2019 baseline (p = 0.0017). The percentage of FV weight within total baskets also increased by 1.6 percentage points (p = 0.0242), although the proportional spend on FV did not change. During the scheme period, FV purchased was higher by 0.4 percentage points (p = 0.0012) and 1.6 percentage points (p = 0.0062) according to spend and weight, respectively, in top-up redeeming baskets compared to non-top-up redeeming baskets with at least one FV item and was associated with 5.5 more HSV 'Suggested' FV portions (p < 0.0001). The median weight of FV purchased increased from 41.83 kg in 2019 to 54.14 kg in 2021 (p = 0.0017). However, top-up vouchers were only redeemed on 9.1% of occasions where FV were purchased. In summary, this study provides novel data showing that safeguarding funds exclusively for FV can help to increase access to FV in low-income households. These results yield important insights to inform public policy aimed at levelling up health inequalities.

Development of a key performance indicator for breast cancer in Queensland, Australia.

PURPOSE: Using population-based data for women diagnosed with stage I-III breast cancer, our aim was to examine the impact of time to treatment completion on survival and to identify factors associated with treatment delay. METHODS: This retrospective study used clinical and treatment data from the Queensland Oncology Repository. Time from diagnosis to completing surgery, chemotherapy and radiation therapy identified a cut-off of 37 weeks as the optimal threshold for completing treatment. Logistic regression was used to identify factors associated with the likelihood of completing treatment > 37 weeks. Overall (OS) and breast cancer-specific survival (BCSS) were examined using Cox proportional hazards models. RESULTS: Of 8279 women with stage I-III breast cancer, 31.9% completed treatment > 37 weeks. Apart from several clinical factors, being Indigenous (p = 0.002), living in a disadvantaged area (p = 0.003) and receiving ≥ two treatment modalities within the public sector (p < 0.001) were associated with an increased likelihood of completing treatment > 37 weeks. The risk of death from any cause was about 40% higher for women whose treatment went beyond 37 weeks (HR 1.37, 95%CI 1.16-1.61), a similar result was observed for BCSS. Using the surgery + chemotherapy + radiation pathway, a delay of > 6.9 weeks from surgery to starting chemotherapy was significantly associated with poorer survival (p = 0.001). CONCLUSIONS: Several sociodemographic and system-related factors were associated with a greater likelihood of treatment completion > 37 weeks. We are proposing a key performance indicator for the management of early breast cancer where a facility should have > 90% of patients with a time from surgery to adjuvant chemotherapy < 6.9 weeks.

Objectively measuring the association between the built environment and physical activity: a systematic review and reporting framework.

BACKGROUND: Objective measures of built environment and physical activity provide the opportunity to directly compare their relationship across different populations and spatial contexts. This systematic review synthesises the current body of knowledge and knowledge gaps around the impact of objectively measured built environment metrics on physical activity levels in adults (≥ 18 years). Additionally, this review aims to address the need for improved quality of methodological reporting to evaluate studies and improve inter-study comparability though the creation of a reporting framework. METHODS: A systematic search of the literature was conducted following the PRISMA guidelines. After abstract and full-text screening, 94 studies were included in the final review. Results were synthesised using an association matrix to show overall association between built environment and physical activity variables. Finally, the new PERFORM ('Physical and Environmental Reporting Framework for Objectively Recorded Measures') checklist was created and applied to the included studies rating them on their reporting quality across four key areas: study design and characteristics, built environment exposures, physical activity metrics, and the association between built environment and physical activity. RESULTS: Studies came from 21 countries and ranged from two days to six years in duration. Accelerometers and using geographic information system (GIS) to define the spatial extent of exposure around a pre-defined geocoded location were the most popular tools to capture physical activity and built environment respectively. Ethnicity and socio-economic status of participants were generally poorly reported. Moderate-to-vigorous physical activity (MVPA) was the most common metric of physical activity used followed by walking. Commonly investigated elements of the built environment included walkability, access to parks and green space. Areas where there was a strong body of evidence for a positive or negative association between the built environment and physical activity were identified. The new PERFORM checklist was devised and poorly reported areas identified, included poor reporting of built environment data sources and poor justification of method choice. CONCLUSIONS: This systematic review highlights key gaps in studies objectively measuring the built environment and physical activity both in terms of the breadth and quality of reporting. Broadening the variety measures of the built environment and physical activity across different demographic groups and spatial areas will grow the body and quality of evidence around built environment effect on activity behaviour. Whilst following the PERFORM reporting guidance will ensure the high quality, reproducibility, and comparability of future research.

Impact on purchasing behaviour of implementing 'junk free checkouts': A pre-post study.

In 2015, Tesco Express convenience stores implemented a healthy checkouts initiative; products high in fat, salt or sugar were removed from in-queue areas. We compare purchasing of less healthy foods before and after its introduction. Tesco provided store-level sales data (n = 1151) for Express stores in England over two 8-week periods, May-July 2014 and 2015. Paired t-tests examined if spending on less healthy foods (biscuits, cakes, crisps and confectionery), as a proportion of total spend, changed between 2015 and 2014. Analyses were repeated for the quantity of less healthy products sold. Compliance was measured through unannounced store visits (n = 41). Complete sales data were available for 1101 stores (96%). Mean overall spend increased in 2015 compared with 2014 (£666 079.70 [SD 406 385.00] vs. £653 786.59 [SD 447 580.77]; p < 0.001). The proportion of total spend from less healthy foods decreased in 2015 versus 2014 (8.03% [SD 2.07] vs. 8.21% [SD 2.17]; p < 0.001). Confectionery accounted for the largest proportion of less healthy product spend, showing the biggest reduction (3.91% [SD 1.16] in 2015 vs. 4.12% [SD 1.24] in 2014; p < 0.001). Results were similar for quantity of less healthy products sold. Like-for-like sales data from major supermarkets revealed spend on less healthy products rose across the UK over this period. Thirty-nine per cent of stores were fully compliant. In conclusion, following implementation of Tesco's healthier checkouts initiative, there was a small reduction in sales of less healthy foods, largely accounted for by confectionery products. These findings suggest that removal of less healthy products from checkouts might lead to healthier purchasing behaviour. However, store compliance was poor, suggesting scope for improvement.

Trends in surgery and adjuvant treatment for early-stage breast cancer: a population-based study in Queensland, Australia.

PURPOSE: Our aim was to describe variations in the treatment of early-stage breast cancer and to examine factors associated with disease-specific survival (DSS). METHODS: The study used linked data for 24,190 women with a T1 (≤ 20 mm) breast cancer who underwent surgery from 2005 to 2019. Multivariate logistic regression was used to model predictors of receiving breast-conserving surgery (BCS) compared to mastectomy and a multinomial model was used to examine factors associated with type(s) of treatment received. RESULTS: Overall, 70.3% had BCS, with a reduced likelihood of BCS observed for younger women (p < 0.001), rural residence, (p < 0.001), socioeconomic disadvantage (p = 0.004), higher tumour grade (p < 0.001) and surgery in a public versus private hospital (p < 0.001). Compared to women who received BCS and radiation therapy (RT), those having mastectomy alone or mastectomy plus RT were more likely to be younger (p < 0.001), live in a rural area (p < 0.001), have higher-grade tumours (p < 0.001) and positive lymph nodes (p < 0.001). Overall 5-year survival was 95.3% and breast cancer-specific survival was 98.3%. Highest survival was observed for women having BCS and RT and lowest for those having mastectomy and RT (p < 0.001). CONCLUSION: Our results indicate some variation in the management of early-stage breast cancer. Lower rates of BCS were observed for rural and disadvantaged women and for those treated in a public or low-volume hospital. Whilst survival was high for this cohort, differences in tumour biology likely explain the differences in survival according to treatment type.

A foresight whole systems obesity classification for the English UK biobank cohort.

BACKGROUND: The number of people living with obesity or who are overweight presents a global challenge, and the development of effective interventions is hampered by a lack of research which takes a joined up, whole system, approach that considers multiple elements of the complex obesity system together. We need to better understand the collective characteristics and behaviours of those who are overweight or have obesity and how these differ from those who maintain a healthy weight. METHODS: Using the UK Biobank cohort we develop an obesity classification system using k-means clustering. Variable selection from the UK Biobank cohort is informed by the Foresight obesity system map across key domains (Societal Influences, Individual Psychology, Individual Physiology, Individual Physical Activity, Physical Activity Environment). RESULTS: Our classification identifies eight groups of people, similar in respect to their exposure to known drivers of obesity: 'Younger, urban hard-pressed', 'Comfortable, fit families', 'Healthy, active and retirees', 'Content, rural and retirees', 'Comfortable professionals', 'Stressed and not in work', 'Deprived with less healthy lifestyles' and 'Active manual workers'. Pen portraits are developed to describe the characteristics of these different groups. Multinomial logistic regression is used to demonstrate that the classification can effectively detect groups of individuals more likely to be living with overweight or obesity. The group identified as 'Comfortable, fit families' are observed to have a higher proportion of healthy weight, while three groups have increased relative risk of being overweight or having obesity: 'Active manual workers', 'Stressed and not in work' and 'Deprived with less healthy lifestyles'. CONCLUSIONS: This paper presents the first study of UK Biobank participants to adopt this obesity system approach to characterising participants. It provides an innovative new approach to better understand the complex drivers of obesity which has the potential to produce meaningful tools for policy makers to better target interventions across the whole system to reduce overweight and obesity.

Opportunities to improve behavioral health crisis response: Results of a large urban county's community status assessment.

A comprehensive community status assessment of an Ohio urban county's crisis response (CR) system explored the experiences of its behavioral health services' clients and providers to surface themes characterizing the system's responsiveness and identifying opportunities for improvements. Forty-eight focus groups and two online surveys were conducted. Data were analyzed using qualitative content analysis and descriptive statistics. The greatest areas of needed improvement ascertained by this effort are in increased CR system resources, more efficient use of resources, and capacity enhancements in nine areas: the mobile crisis team, CR protocols, psychiatric inpatient and crisis stabilization beds, stabilization admission for eligible persons, stabilization services for in-crisis but admission-ineligible persons, continuity of care, research into child versus adult CR systems, Provider Emergency Support Program, and first responder crisis intervention training. The assessment provides a foundation for the county to identify further opportunities for system scale-up.

A systematic review of supermarket automated electronic sales data for population dietary surveillance.

CONTEXT: Most dietary assessment methods are limited by self-report biases, how long they take for participants to complete, and cost of time for dietitians to extract content. Electronically recorded, supermarket-obtained transactions are an objective measure of food purchases, with reduced bias and improved timeliness and scale. OBJECTIVE: The use, breadth, context, and utility of electronic purchase records for dietary research is assessed and discussed in this systematic review. DATA SOURCES: Four electronic databases (MEDLINE, EMBASE, PsycINFO, Global Health) were searched. Included studies used electronically recorded supermarket transactions to investigate the diet of healthy, free-living adults. DATA EXTRACTION: Searches identified 3422 articles, of which 145 full texts were retrieved and 72 met inclusion criteria. Study quality was assessed using the National Institutes of Health Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies. DATA ANALYSIS: Purchase records were used in observational studies, policy evaluations, and experimental designs. Nutrition outcomes included dietary patterns, nutrients, and food category sales. Transactions were linked to nutrient data from retailers, commercial data sources, and national food composition databases. CONCLUSION: Electronic sales data have the potential to transform dietary assessment and worldwide understanding of dietary behavior. Validation studies are warranted to understand limits to agreement and extrapolation to individual-level diets. SYSTEMATIC REVIEW REGISTRATION: PROSPERO registration no. CRD42018103470.

Clustering Accelerometer Activity Patterns from the UK Biobank Cohort.

Many researchers are beginning to adopt the use of wrist-worn accelerometers to objectively measure personal activity levels. Data from these devices are often used to summarise such activity in terms of averages, variances, exceedances, and patterns within a profile. In this study, we report the development of a clustering utilising the whole activity profile. This was achieved using the robust clustering technique of k-medoids applied to an extensive data set of over 90,000 activity profiles, collected as part of the UK Biobank study. We identified nine distinct activity profiles in these data, which captured both the pattern of activity throughout a week and the intensity of the activity: "Active 9 to 5", "Active", "Morning Movers", "Get up and Active", "Live for the Weekend", "Moderates", "Leisurely 9 to 5", "Sedate" and "Inactive". These patterns are differentiated by sociodemographic, socioeconomic, and health and circadian rhythm data collected by UK Biobank. The utility of these findings are that they sit alongside existing summary measures of physical activity to provide a way to typify distinct activity patterns that may help to explain other health and morbidity outcomes, e.g., BMI or COVID-19. This research will be returned to the UK Biobank for other researchers to use.

Predicting Food Safety Compliance for Informed Food Outlet Inspections: A Machine Learning Approach.

Consumer food environments have transformed dramatically in the last decade. Food outlet prevalence has increased, and people are eating food outside the home more than ever before. Despite these developments, national spending on food control has reduced. The National Audit Office report that only 14% of local authorities are up to date with food business inspections, exposing consumers to unknown levels of risk. Given the scarcity of local authority resources, this paper presents a data-driven approach to predict compliance for newly opened businesses and those awaiting repeat inspections. This work capitalizes on the theory that food outlet compliance is a function of its geographic context, namely the characteristics of the neighborhood within which it sits. We explore the utility of three machine learning approaches to predict non-compliant food outlets in England and Wales using openly accessible socio-demographic, business type, and urbanness features at the output area level. We find that the synthetic minority oversampling technique alongside a random forest algorithm with a 1:1 sampling strategy provides the best predictive power. Our final model retrieves and identifies 84% of total non-compliant outlets in a test set of 92,595 (sensitivity = 0.843, specificity = 0.745, precision = 0.274). The originality of this work lies in its unique and methodological approach which combines the use of machine learning with fine-grained neighborhood data to make robust predictions of compliance.

Characterisation of Temporal Patterns in Step Count Behaviour from Smartphone App Data: An Unsupervised Machine Learning Approach.

The increasing ubiquity of smartphone data, with greater spatial and temporal coverage than achieved by traditional study designs, have the potential to provide insight into habitual physical activity patterns. This study implements and evaluates the utility of both K-means clustering and agglomerative hierarchical clustering methods in identifying weekly and yearlong physical activity behaviour trends. Characterising the demographics and choice of activity type within the identified clusters of behaviour. Across all seven clusters of seasonal activity behaviour identified, daylight saving was shown to play a key role in influencing behaviour, with increased activity in summer months. Investigation into weekly behaviours identified six clusters with varied roles, of weekday versus weekend, on the likelihood of meeting physical activity guidelines. Preferred type of physical activity likewise varied between clusters, with gender and age strongly associated with cluster membership. Key relationships are identified between weekly clusters and seasonal activity behaviour clusters, demonstrating how short-term behaviours contribute to longer-term activity patterns. Utilising unsupervised machine learning, this study demonstrates how the volume and richness of secondary app data can allow us to move away from aggregate measures of physical activity to better understand temporal variations in habitual physical activity behaviour.

A comparison of time to event analysis methods, using weight status and breast cancer as a case study.

Survival analysis with cohort study data has been traditionally performed using Cox proportional hazards models. Random survival forests (RSFs), a machine learning method, now present an alternative method. Using the UK Women's Cohort Study (n = 34,493) we evaluate two methods: a Cox model and an RSF, to investigate the association between Body Mass Index and time to breast cancer incidence. Robustness of the models were assessed by cross validation and bootstraping. Histograms of bootstrap coefficients are reported. C-Indices and Integrated Brier Scores are reported for all models. In post-menopausal women, the Cox model Hazard Ratios (HR) for Overweight (OW) and Obese (O) were 1.25 (1.04, 1.51) and 1.28 (0.98, 1.68) respectively and the RSF Odds Ratios (OR) with partial dependence on menopause for OW and O were 1.34 (1.31, 1.70) and 1.45 (1.42, 1.48). HR are non-significant results. Only the RSF appears confident about the effect of weight status on time to event. Bootstrapping demonstrated Cox model coefficients can vary significantly, weakening interpretation potential. An RSF was used to produce partial dependence plots (PDPs) showing OW and O weight status increase the probability of breast cancer incidence in post-menopausal women. All models have relatively low C-Index and high Integrated Brier Score. The RSF overfits the data. In our study, RSF can identify complex non-proportional hazard type patterns in the data, and allow more complicated relationships to be investigated using PDPs, but it overfits limiting extrapolation of results to new instances. Moreover, it is less easily interpreted than Cox models. The value of survival analysis remains paramount and therefore machine learning techniques like RSF should be considered as another method for analysis.