Health and care data: approaches to data linkage for evidence-informed policy

An indispensable prerequisite for answering research questions in health services research is the availability and accessibility of comprehensive, high quality data. It can be assumed that health services research in the comingyears will be increasingly based on data linkage, i.e., the linking, or connecting, of several data sources based on suitable common key variables. A range of approaches to data collection, storage, linkage and availability exists across countries, particularly for secondary research purposes (i.e., the use of data initially collected for other purposes), such as health systems research. The main goal of this review is to develop an overview of, and gain insights into, current approaches to linking data sources in the context of health services research, with the view to inform policy, based on existing practices in high-income countries in Europe and beyond. In doing so, another objective is to provide lessons for countries looking for possible or alternative approaches to data linkage. Thirteen country case studies of data linkage approaches were selected and analyzed. Rather than being comprehensive, this review aimed to identify varied and potentially useful case studies to showcase different approaches to data linkage worldwide. A conceptual framework was developed to guide the selection and description of case studies. Information was first identified and collected from publicly available sources and a profile was then created for each country and each case study; these profiles were forwarded to appropriate country experts for validation and completion.

Core outcome set for early intervention trials to prevent obesity in childhood (COS-EPOCH): Agreement on "what" to measure.

BACKGROUND: Heterogeneity in the outcomes collected and reported in trials of interventions to prevent obesity in the first five years of life highlights the need for a core outcome set to streamline intervention evaluation and synthesis of effects. This study aimed to develop a core outcome set for use in early childhood obesity prevention intervention studies in children from birth to five years of age (COS-EPOCH). METHODS: The development of the core outcome set followed published guidelines and consisted of three stages: (1) systematic scoping review of outcomes collected and reported in early childhood obesity prevention trials; (2) e-Delphi study with stakeholders to prioritise outcomes; (3) meeting with stakeholders to reach consensus on outcomes. Stakeholders included parents/caregivers of children aged ≤ five years, policy-makers/funders, researchers, health professionals, and community and organisational stakeholders interested in obesity prevention interventions. RESULTS: Twenty-two outcomes from nine outcome domains (anthropometry, dietary intake, sedentary behaviour, physical activity, sleep, outcomes in parents/caregivers, environmental, emotional/cognitive functioning, economics) were included in the core outcome set: infant tummy time; child diet quality, dietary intake, fruit and vegetable intake, non-core food intake, non-core beverage intake, meal patterns, weight-based anthropometry, screentime, time spent sedentary, physical activity, sleep duration, wellbeing; parent/caregiver physical activity, sleep and nutrition parenting practices; food environment, sedentary behaviour or physical activity home environment, family meal environment, early childhood education and care environment, household food security; economic evaluation. CONCLUSIONS: The systematic stakeholder-informed study identified the minimum outcomes recommended for collection and reporting in early childhood obesity prevention trials. Future work will investigate the recommended instruments to measure each of these outcomes. The core outcome set will standardise guidance on the measurement and reporting of outcomes from early childhood obesity prevention interventions, to better facilitate evidence comparison and synthesis, and maximise the value of data collected across studies.

Landscape of clinical trial activity focusing on Indigenous health in Australia: an overview using clinical trial registry data from 2008-2018.

BACKGROUND: Aboriginal and Torres Strait Islander peoples (hereafter respectfully referred to as Indigenous Australians) represent about 3% of the total Australian population. Major health disparities exist between Indigenous and Non-Indigenous Australians. To address this, it is vital to understand key health priorities and knowledge gaps in the current landscape of clinical trial activity focusing on Indigenous health in Australia. METHODS: Australian-based clinical trials registered on the Australian New Zealand Clinical Trials Registry or ClinicalTrials.gov from 2008 to 2018 were analysed. Australian clinical trials with and without a focus on Indigenous health were compared in terms of total numbers, participant size, conditions studied, design, intervention type and funding source. RESULTS: Of the 9206 clinical trials included, 139 (1.5%) focused on Indigenous health, with no proportional increase in Indigenous trials over the decade (p = 0.30). Top conditions studied in Indigenous-focused trials were mental health (n = 35, 28%), cardiovascular disease (n = 20, 20%) and infection (n = 16, 16%). Compared to General Australian trials, Indigenous-focused trials more frequently studied ear conditions (OR 20.26, 95% CI 10.32-37.02, p < 0.001), infection (OR 3.11, 95% CI 1.88-4.85, p < 0.001) and reproductive health (OR 2.59, 95% CI 1.50-4.15, p < 0.001), and less of musculoskeletal conditions (OR 0.09, 95% CI 0.00-0.37, p < 0.001), anaesthesiology (OR 0.16, 95% CI 0.01-0.69, p = 0.021) and surgery (OR 0.17, 95% CI 0.01-0.73, p = 0.027). For intervention types, Indigenous trials focused more on prevention (n = 48, 36%) and screening (n = 18, 13%). They were far less involved in treatment (n = 72, 52%) as an intervention than General Australian trials (n = 6785, 75%), and were less likely to be blinded (n = 48, 35% vs n = 4273, 47%) or have industry funding (n = 9, 7% vs 1587, 17%). CONCLUSIONS: Trials with an Indigenous focus differed from General Australian trials in the conditions studied, design and funding source. The presented findings may inform research prioritisation and alleviate the substantial burden of disease for Indigenous population.

Associations between industry involvement and study characteristics at the time of trial registration in biomedical research.

BACKGROUND: Commercial or industry funding is associated with outcomes that favour the study funder in published studies, across various areas of research. However, it is currently unclear whether there are differences between trials with and without industry involvement at the stage of trial registration. OBJECTIVE: To determine whether industry involvement (industry sponsorship, funding, or collaboration) is associated with trial characteristics at the time of trial registration. METHODS: We conducted a cross-sectional analysis of all interventional studies registered on the Australian New Zealand Clinical Trials Registry in 2017 and classified them by industry involvement. We analysed whether there were differences in study characteristics (including type of control, sample size, study phase, randomisation, registration timing, and purpose of study) by industry involvement. RESULTS: Industry involvement was reported by 21% of the 1,433 included trials. Only 40% of trials with industry involvement used an active control compared to 58% of non-industry trials (OR = 0.49, 95%CI = 0.38 to 0.63, p < .001), and industry trials reported smaller sample sizes (Median(IQR)industry = 45(24-100), Median(IQR)non-industry = 70(35-160), Mean Difference = -153, 95% CI = -233 to -75, p < .001). Industry trials were more likely to be earlier phase trials (Χ2(df) = 71.46(4), p < .001). There was no difference in use of randomisation between industry (70%) and non-industry trials (73%) (OR = 0.88, 95%CI = 0.67-1.20, p = .38). Eighty-three percent of industry trials compared to 70% of non-industry trials were prospectively registered (OR = 2.02, 95%CI = 1.47-2.82, p < .001). Industry trials were more likely to assess treatment (85%), rather than prevention, education or diagnosis compared to non-industry trials (64%) (OR = 3.02, 95%CI = 2.17-4.32, p < .001). CONCLUSION: The current study gives insight into differences in trial characteristics by industry involvement at registration stage. There was a reduced use of active controls in trials with industry involvement which has previously been proposed as a mechanism behind more favourable results. Non-industry funders and sponsors are crucial to ensure research addresses not only treatments, but also prevention, diagnosis and education questions.