A comparison of weather variables linked to infectious disease patterns using laboratory addresses and patient residence addresses

Abdelmajid Djennad*, Giovanni Lo Iacono, Christophe Sarran, Lora E. Fleming, Anthony Kessel, Andy Haines, Gordon Nichols

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

4 Citations (Scopus)

Abstract

Background: To understand the impact of weather on infectious diseases, information on weather parameters at patient locations is needed, but this is not always accessible due to confidentiality or data availability. Weather parameters at nearby locations are often used as a proxy, but the accuracy of this practice is not known. Methods: Daily Campylobacter and Cryptosporidium cases across England and Wales were linked to local temperature and rainfall at the residence postcodes of the patients and at the corresponding postcodes of the laboratory where the patient's specimen was tested. The paired values of daily rainfall and temperature for the laboratory versus residence postcodes were interpolated from weather station data, and the results were analysed for agreement using linear regression. We also assessed potential dependency of the findings on the relative geographic distance between the patient's residence and the laboratory. Results: There was significant and strong agreement between the daily values of rainfall and temperature at diagnostic laboratories with the values at the patient residence postcodes for samples containing the pathogens Campylobacter or Cryptosporidium. For rainfall, the R-squared was 0.96 for the former and 0.97 for the latter, and for maximum daily temperature, the R-squared was 0.99 for both. The overall mean distance between the patient residence and the laboratory was 11.9 km; however, the distribution of these distances exhibited a heavy tail, with some rare situations where the distance between the patient residence and the laboratory was larger than 500 km. These large distances impact the distributions of the weather variable discrepancies (i.e. the differences between weather parameters estimated at patient residence postcodes and those at laboratory postcodes), with discrepancies up to ±10 °C for the minimum and maximum temperature and 20 mm for rainfall. Nevertheless, the distributions of discrepancies (estimated separately for minimum and maximum temperature and rainfall), based on the cases where the distance between the patient residence and the laboratory was within 20 km, still exhibited tails somewhat longer than the corresponding exponential fits suggesting modest small scale variations in temperature and rainfall. Conclusion: The findings confirm that, for the purposes of studying the relationships between meteorological variables and infectious diseases using data based on laboratory postcodes, the weather results are sufficiently similar to justify the use of laboratory postcode as a surrogate for domestic postcode. Exclusion of the small percentage of cases where there is a large distance between the residence and the laboratory could increase the precision of estimates, but there are generally strong associations between daily weather parameters at residence and laboratory.

Original languageEnglish
Article number198
JournalBMC Infectious Diseases
Volume18
Issue number1
DOIs
Publication statusPublished - 27 Apr 2018

Bibliographical note

Funding Information:
Acknowledgements We thank Francis Senyah in Public Health England for his collaboration in the production of Fig. 3a. Funding The research was funded in part by the UK Medical Research Council (MRC) and UK Natural Environment Research Council (NERC) for the MEDMI Project; the National Institute for Health Research Health Protection Research Unit (NIHR HPRU) in Environmental Change and Health at the London School of Hygiene and Tropical Medicine in partnership with Public Health England (PHE), and in collaboration with the University of Exeter, University College London, and the Met Office. It also included collaboration with the NIHR HPRU in Gastrointestinal Infection Programme. The views expressed are those of the authors not necessarily those of the NERC, MRC, NHS, the NIHR, the Department of Health and Social Care or Public Health England.

Publisher Copyright:
© 2018 The Author(s).

Keywords

  • Campylobacter
  • Cryptosporidium
  • Data linkage
  • MEDMI
  • MIDAS
  • Rainfall
  • SGSS
  • Temperature

Fingerprint

Dive into the research topics of 'A comparison of weather variables linked to infectious disease patterns using laboratory addresses and patient residence addresses'. Together they form a unique fingerprint.

Cite this