LSTM Home > LSTM Research > LSTM Online Archive

The spatiotemporal association of non-prescription retail sales with cases during the 2009 influenza pandemic in Great Britain

Todd, Stacy, Diggle, P. J., White, P. J., Fearne, A. and Read, J. M. (2014) 'The spatiotemporal association of non-prescription retail sales with cases during the 2009 influenza pandemic in Great Britain'. BMJ Open, Vol 4, Issue 4, e004869.

[img]
Preview
Text
BMJ_Open_4_4_e004869.pdf - Published Version
Available under License Creative Commons Attribution.

Download (2MB)

Abstract

Objective

To assess whether retail sales of non-prescription products can be used for syndromic surveillance and whether it can detect influenza activity at different spatial scales. A secondary objective was to assess whether changes in purchasing behaviour were related to public health advice or levels of media or public interest.

Setting

The UK.

Participants

National and regional influenza case estimates and retail sales from a major British supermarket.

Outcome measures

Weekly, seasonally adjusted sales of over-the-counter symptom remedies and non-pharmaceutical products; recommended as part of the advice offered by public health agencies; were compared with weekly influenza case estimates. Comparisons were made at national and regional spatial resolutions. We also compared sales to national measures of contemporaneous media output and public interest (Internet search volume) related to the pandemic.

Results

At a national scale there was no significant correlation between retail sales of symptom remedies and cases for the whole pandemic period in 2009. At the regional scale, a minority of regions showed statistically significant positive correlations between cases and sales of adult ‘cold and flu’ remedies and cough remedies (3.2%, 5/156, 3.8%, 6/156), but a greater number of regions showed a significant positive correlation between cases and symptomatic remedies for children (35.6%, 55/156). Significant positive correlations between cases and sales of thermometers and antiviral hand gels/wash were seen at both spatial scales (Cor 0.477 (95% CI 0.171 to 0.699); 0.711 (95% CI 0.495 to 0.844)). We found no significant association between retail sales and media reporting or Internet search volume.

Conclusions

This study provides evidence that the British public responded appropriately to health messaging about hygiene. Non-prescription retail sales at a national level are not useful for the detection of cases. However, at finer spatial scales, in particular age-groups, retail sales may help augment existing surveillance and merit further study.

Item Type: Article
Subjects: QW Microbiology and Immunology > Viruses > QW 160 Viruses (General). Virology
WA Public Health > WA 105 Epidemiology
WA Public Health > WA 20.5 Research (General)
WC Communicable Diseases > Virus Diseases > Viral Respiratory Tract Infections. Respirovirus Infections > WC 515 Human influenza
Faculty: Department: Clinical Sciences & International Health > Clinical Sciences Department
Digital Object Identifer (DOI): https://doi.org/10.1136/bmjopen-2014-004869
Depositing User: Lynn Roberts-Maloney
Date Deposited: 27 May 2015 09:52
Last Modified: 06 Feb 2018 13:10
URI: https://archive.lstmed.ac.uk/id/eprint/5171

Statistics

View details

Actions (login required)

Edit Item Edit Item