Estimating Incidence Curves of Several Infections Using Symptom Surveillance Data
Author Information
Author(s): Edward Goldstein, Benjamin J. Cowling, Allison E. Aiello, Saki Takahashi, Gary King, Ying Lu, Marc Lipsitch
Primary Institution: Harvard School of Public Health
Hypothesis
Can symptom surveillance data be used to estimate the incidence of co-circulating infectious pathogens?
Conclusion
The study presents a method for estimating the incidence of various infections based on symptom profiles, demonstrating its application through simulations and real data.
Supporting Evidence
- The method utilizes symptom data from a defined population to estimate infection rates.
- Numerical simulations were used to illustrate the method's effectiveness.
- Data from a survey conducted on the University of Michigan campus was analyzed.
- The study highlights the importance of accurate symptom profile distributions for reliable estimates.
Takeaway
Researchers found a way to use symptoms reported by people to figure out how many infections are happening, even when many cases go unreported.
Methodology
The study used a deconvolution method to analyze weekly symptom surveillance data and estimate infection incidence.
Potential Biases
Potential misattribution of non-flu cases to flu could skew incidence estimates.
Limitations
The method's accuracy depends on the correct estimation of symptom profile distributions, which can vary by population.
Participant Demographics
Participants were from the University of Michigan campus, with a sample size ranging from 830 to 902 respondents weekly.
Statistical Information
P-Value
0.011
Confidence Interval
2.2%, 28.6%
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website