Journal of Epidemiology and Community Health最新文献

Background: Children and young adults with attention-deficit/hyperactivity disorder (ADHD) may be at increased risk of criminal offending. This study examines the risk of first-time and repeat offending among individuals with ADHD and potential effect of stimulant medication.

Methods: A population-based, data linkage cohort study of individuals born in New South Wales, Australia between 1990 and 2005 and followed until May 2016. All individuals authorised for stimulant treatment for ADHD were frequency matched (1:10) to controls without ADHD. Proven criminal offences were identified using court records. First and repeat offences were examined separately using modified Cox regression and Prentice-Williams-Peterson models.

Results: The cohort included 75 650 individuals with ADHD (147 855 treated and 525 595 untreated person years) and 745 634 non-ADHD controls. The risk of first offence was higher among cases, but reduced by stimulants (males 10-17 years, untreated HR 2.02, 95% CI 1.95 to 2.10; treated HR 1.52, 95% CI 1.41 to 1.62). The effect estimates for males were reduced for repeat offences (10-17 years: untreated HR 1.09, 95% CI 1.05 to 1.13; treated HR 0.97, 95% CI 0.90 to 1.04). No association with reoffending was found in females, except treated individuals aged 10-17 years had a higher risk (HR 1.26, 95% CI 1.02 to 1.56).

Conclusions: ADHD was associated with higher risk of a first criminal offence, but stimulant prescription was associated with lower risk among individuals with ADHD. This association was reduced among repeat offenders, with only modest apparent benefit of treatment among males. Adequate treatment resources may help keep individuals with ADHD out of the criminal justice system.

Background: Food insecurity is associated with worse diabetes outcomes, but how that association may vary across individuals is unclear. We sought to better understand possible heterogeneity in the association between food insecurity and diabetes outcomes.

Methods: Longitudinal cohort study of adults with type 2 diabetes assessed for food insecurity in community-based health centres. Diabetes outcomes were haemoglobin A1c (HbA1c), systolic and diastolic blood pressure, and low-density lipoprotein cholesterol (LDL), 12 months after food insecurity assessment. We used three machine learning methods (generalised random forest (GRF), X-Learner and Doubly Robust Learner) to assess outcome heterogeneity across the following variables: age, sex, race and ethnicity, health insurance, income, comorbidity, baseline outcome and social vulnerability index.

Results: Among 41 581 individuals (mean age 56.2 (SD: 12.8) years, 56.9% female), mean baseline HbA1c was 7.8% (SD: 1.9%) and 10 399 (25.0%) reported food insecurity. Food insecurity was associated with 0.1% greater HbA1c at 12 months, with little heterogeneity. For example, using GRF, the mean difference in HbA1c in the quartile with the lowest estimated HbA1c difference associated with experiencing, versus not experiencing, food insecurity was 0.0%, while the mean difference in the quartile with the greatest estimated difference was 0.2%. Results were similar for other outcomes, except that there was no association between food insecurity and LDL.

Conclusions: In this study, we did not observe meaningful heterogeneity in the associations between food insecurity and diabetes outcomes.

Background: Global warming is expected to increase the risk of cardiovascular morbidity and mortality. The effects of climate variability on sudden cardiac arrest (SCA) have not been studied in tropical climates. We aimed to study this in Singapore, a city-state with a tropical climate.

Methods: We included all nationally reported SCA cases from 1 April 2010 to 31 December 2021. Using negative binomial regression adjusted for long-term trend, periodicity and public holidays, we estimated the temperature and absolute humidity effects on SCA risk within the Distributed Lag Non-Linear Model framework. We compared the seasonal trend components of SCA cases, temperature and absolute humidity.

Results: There were 27 209 cases. The majority were 65 years of age and above (62.3%), male (63.4%) and of Chinese ethnicity (68.4%). SCA risk increased by 13% (RR: 1.13, 95% CI: 1.07 to 1.20) at 29°C and by 33% (RR: 1.33, 95% CI: 1.04 to 1.70) at 24°C (relative to 27°C) when these temperatures sustained over 8 days. SCA risk increased by 78% (RR: 1.78, 95% CI: 1.34 to 2.37) when absolute humidity was sustained over 7 days at 18 g/m³ (relative to 25 g/m³). The cooler temperature effects on SCA risk were higher at lower levels of absolute humidity, suggesting effect modification. SCA had its highest peak in the first half of the calendar year, occurring concurrent to a seasonal trough in absolute humidity while its second peak was preceded by the seasonal peak in temperature.

Conclusion: Temperature and absolute humidity were independently associated with SCA risk.

Background: Negative associations between neighbourhood socioeconomic disadvantage and health are well documented. However, limited longitudinal evidence exists on the dynamic nature of the neighbourhood context and implications for health disparities. Using rich register data, we examined how neighbourhood disadvantage in Denmark changed over three decades and explored associations with mortality rates.

Methods: This ecological study used nationwide register data from 1987 to 2018. A neighbourhood socioeconomic disadvantage index was created using aggregated data on neighbourhood-level educational attainment, unemployment and family income. Trends in the neighbourhood index were assessed using descriptive and geospatial analyses. Associations with age-standardised mortality rates were plotted and examined in regression-based models.

Results: Neighbourhood socioeconomic conditions generally improved over the three decades, and the mean difference between disadvantaged quartiles narrowed. The location of the most and least disadvantaged quartiles was largely stable across the period. Mortality rates generally decreased over time; however, rates of decline differed, leading to increased mortality disparities by quartile of neighbourhood disadvantage. Regression analysis demonstrated an increase in the association between neighbourhood disadvantage and mortality rates over time. First-difference regression models showed no association between change in neighbourhood disadvantage and change in mortality rates.

Conclusions: Despite general improvements in neighbourhood socioeconomic conditions and neighbourhood mortality rates across the last three decades in Denmark, we found growing health inequality between more and less disadvantaged neighbourhoods. A strong concentration of disadvantage over time and space and growing segregation of income levels may have contributed to this development, with important implications for public policy and health policy investments.

Introduction: Prior cross-sectional research has identified incarceration as a risk factor for food insecurity across the life course. However, there is a lack of longitudinal studies on the relationship between prior incarceration and food insecurity over time.

Methods: This study uses biennial data across 10 time points from the Health and Retirement Study (years 2012-2022) to examine the association between prior incarceration and longitudinal trajectories of food insecurity among adults aged 55 and older in the USA (N=8229). Group-based trajectory modelling was used to assess patterns of food insecurity status over time. Multinomial logistic regression assessed the relationship between prior incarceration and food insecurity trajectory group membership.

Results: Three food insecurity trajectory groups were identified: no food insecurity (86.2%), declining food insecurity (11.0%) and chronic food insecurity (2.8%). Results from the multinomial logistic regression demonstrated that a history of incarceration was significantly associated with a higher likelihood of membership in the Declining Food Insecurity (relative risk ratio (RRR)=1.80, 95% CI 1.24 to 2.60) and Chronic Food Insecurity groups (RRR=2.14, 95% CI 1.35 to 3.39), relative to No Food Insecurity group after adjusting for covariates. However, after controlling for household income and wealth, this association was attenuated and remained statistically significant only for the Declining Food Insecurity group (RRR=1.59, 95% CI 1.06 to 2.37).

Conclusions: A history of incarceration is associated with a greater risk of food insecurity across older adulthood, though this relationship appears to be largely due to disparities in socioeconomic status.

Background: Electrification of the light vehicle fleet is a core component of decarbonisation policy globally. As transport is an important determinant of health, this process will likely impact a range of pathways to population health, outside of averted climate change.

Methods: A transport-health multistate lifetable model for Aotearoa/New Zealand was used to model changes in the vehicle fleet and transport behaviour patterns that could be expected as part of electrification in Aotearoa/New Zealand. These changes are related to the new energy source and the costs of driving. The potential health, equity, health system cost and environmental impacts of this transition were modelled through the pathways of physical activity, injury and air pollution (tailpipe and non-tailpipe).

Results: Compared with a scenario of 2018 vehicles and travel patterns projected forward, under the electrification scenario modelled by 2050, there would be a 21% increase in per capita kilometres travelled by car, a loss of 270 health adjusted life years (95% uncertainty interval (UI) -1000 to 500) and health system costs incurred of NZ$44 million 2018 dollars (95% UI NZ$-12 to NZ$110). While health impacts from air pollution decreased, in the electrification scenario road injury increased and there was a net health loss. Increasing the cost of driving, through distance-based charging, attenuated the negative health impacts.

Conclusions: These results suggest that the electrification of the light vehicle fleet has mixed health impacts. Researchers and policy-makers should take a cautious approach to claims that light fleet electrification will have co-benefits for population health and health equity.

Background: Sub-Saharan Africa faces a double burden of disease due to the continued high prevalence of infectious diseases such as HIV and tuberculosis, and a concurrent increase in non-communicable diseases. The co-occurrence and clustering of multiple diseases can cause a syndemic, synergistically interacting epidemics, driven by context.

Methods: This cross-sectional study investigated potential syndemics in Eswatini, by determining geospatial disease concentration and estimating interaction between four diseases and the impact of context on these interactions. Using data from the WHOPEN@Scale household survey, we estimated generalised linear models with a quasi-Poisson link, incorporating three-way interaction terms. Joint effect estimates and the relative excess risk due to interaction (RERI) between diseases and contextual factors were estimated.

Results: Conditions with comorbid depression were concentrated in the middle of the country near the capital, and the combinations without depression mostly in the north. Additive interaction was found between HIV and diabetes across all three contextual factors, in particular for people who are the sole household member aged 30 or above RERI: 0.46 (95% CI 0.12 to 0.80) and those who had lost their partner RERI: 0.44 (95% CI 0.02 to 0.86).

Conclusions: We show that HIV and diabetes exhibit syndemic properties, indicated by geospatial variation, additive interaction as well as interaction with context. The lifelong chronicity and stigmatising nature of both diseases emphasise the importance of the social and financial context surrounding affected individuals. While the other disease pairs did not exhibit syndemic properties, the worst health outcomes among the double-exposed stipulate that comorbidity remains important within treatment guidelines.

Background: Evidence on socioeconomic inequalities in the prevalence of common long-term conditions and their variation across the life course is necessary for equitable service design and resource allocation. We used routinely collected electronic primary care records and a unified data extraction and analysis framework to estimate socioeconomic variations in the prevalence of 17 common long-term conditions by age and sex.

Methods: Electronic records for 2.2 m patients registered with 300 randomly selected primary care practices contributing to the Clinical Practice Research Datalink Aurum database were used to estimate observed, age-sex standardised and age-specific rates of disease prevalence on 31 March 2020 by Index of Multiple Deprivation quintile groups. Inequality in disease burden was expressed as the prevalence rate ratio (RR) between the most and least deprived fifths of the population.

Results: Age-sex standardised prevalence rates were higher in the most deprived compared with the least deprived fifth of the population for 16 of 17 conditions. The largest relative differences in disease prevalence were observed for chronic obstructive pulmonary disease (RR: 3.29; 95% CI: 3.19 to 3.38), severe mental illness (RR: 2.72; 95% CI: 2.60 to 2.85) and peripheral arterial disease (RR: 2.58; 95% CI: 2.46 to 2.72). For most conditions, the equity gap was largest in middle age and reduced with age thereafter.

Conclusions: Substantial socioeconomic inequalities in disease prevalence are evident in the English population. A catalogue of disease prevalence by socioeconomic quintile group, age and sex is provided to facilitate further analysis and modelling.