Environmental ultraviolet-B exposures and measured health outcomes in English children

Abstract

Purpose
Vitamin D status has been shown to significantly impact extra-skeletal health outcomes. In the UK, the ability of the population to synthesize vitamin D is highly sensitive to environmental factors including ultraviolet-B wavelength light. This thesis aims to investigate the impact of environmental determinants of vitamin D status on child population health. Asthma, lower respiratory tract infection (LRTI) and gastroenteritis admission rates in English children will be analysed with meteorological variables at multiple geographical levels.

Methods
Five ecological studies were carried out. Three sets of multivariate linear regression analyses were performed to examine the associations between regional meteorological variables (hours of sunshine, rainfall and air temperature) and admission rates of asthma, LRTI and gastroenteritis in children in England between 2002 and 2011. Regional income deprivation affecting children index (IDACI) scores were used to control for population socioeconomic deprivation. Met Office and NHS Hospital Episode Statistics (HES) datasets were matched to create variables in these models. Extended generalised estimating equations were also used to explore these relationships. The remaining two studies focused on the relationships between acute paediatric asthma and LRTI admission rates on a GP-practice level, and ultraviolet-B (UVB) radiation doses in England between 2013 and 2018. HES admission data and Tropospheric Emission Monitoring Internet Service (TEMIS) satellite data on UVB levels of wavelengths specific to the vitamin D action spectrum (D-UVB) were used to carry out multivariate linear regression models on a GP and Clinical Commissioning Group aggregate levels. Maternal smoking rates, GP practice IDACI scores, latitude, longitude, age and gender were also included in these models.

Results
The average English child population from 2002 to 2011 in the first three sets of analyses was 9,731,440, with average admission rates of 265.3 per 100,000 per annum, 390.0 per 100,000 per annum and 1375 per 100,000 per annum for paediatric asthma, LRTI and gastroenteritis respectively. Asthma, LRTI and gastroenteritis admission rates were found to be significantly inversely associated with hours of sunshine (HoS) (p<0.001). In the linear regression analyses, total annual rainfall and average temperature in English regions were also significantly associated with LRTI and gastroenteritis admissions (p<0.001). Regional IDACI scores were found to be significantly associated with asthma and LRTI admissions. In the studies focusing on D-UVB levels and child health, between 2013 and 2018 the average asthma admission rate was 314 per 100,000 per annum, and the average LRTI rate between 2016 and 2019 was 678 per 100,000 per annum. The average D-UVB levels corrected for cloud cover between 2013 and 2018 was 794.98 kJ/m2. Both asthma and LRTI admission rates were found to be significantly associated with cloud corrected D-UVB levels (p<0.001) at both GP and CCG levels. There were varying associations between asthma and LRTI admission rates and the other covariates included in these models.

Conclusions
Meteorological factors were associated with hospital admission rates for common paediatric conditions; the most reliably significant factor associated with a decrease in admissions was hours of sunshine or D-UVB levels. These findings support the hypothesis that population vitamin D status has a significant influence on child health, and that targeting this with encouraging judicious sunlight exposure, outdoor activity, weight loss or a comprehensive supplementation programme could be effective preventative measures.