Back

Precision recruitment for high-risk participants in a COVID-19 research study

Mezlini, A.; Caddigan, E.; Shapiro, A.; Ramirez, E.; Kondow-McConaghy, H.; Yang, J.; DeMarco, K.; Naraghi-Arani, P.; Foschini, L.

2022-03-05 infectious diseases
10.1101/2022.03.03.22271504 medRxiv
Show abstract

Studies for developing diagnostics and treatments for infectious diseases usually require observing the onset of infection during the study period. However, when the infection base rate incidence is low, the cohort size required to measure an effect becomes large, and recruitment becomes costly and prolonged. We describe an approach for reducing recruiting time and resources in a COVID-19 study by targeting recruitment to high-risk individuals. Our approach is based on direct and longitudinal connection with research participants and computes individual risk scores from individually permissioned data about socioeconomic and behavioural data, in combination with predicted local prevalence data. When we used these scores to recruit a balanced cohort of participants for a COVID-19 detection study, we obtained a 4-7-fold greater COVID-19 infection incidence compared with similar real-world study cohorts.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

1
Nature Communications
○ 5641 papers in training set
Top 9%
18.3%
Similar papers in this journal
2
Nature Medicine
○ 125 papers in training set
Top 0.1%
9.7%
Similar papers in this journal
3
The Lancet Infectious Diseases
○ 73 papers in training set
Top 0.2%
5.4%
Similar papers in this journal
4
Clinical Infectious Diseases
◐ 235 papers in training set
Top 0.5%
5.4%
Similar papers in this journal
5
JAMA Network Open
● 130 papers in training set
Top 0.8%
4.3%
6
American Journal of Epidemiology
◐ 67 papers in training set
Top 0.3%
4.3%
Similar papers in this journal
7
eLife
● 5828 papers in training set
Top 31%
4.0%
50% of probability mass above
8
Med
○ 39 papers in training set
Top 0.1%
3.2%
9
Nature
○ 645 papers in training set
Top 5%
3.1%
10
International Journal of Epidemiology
◐ 88 papers in training set
Top 0.4%
2.8%
11
npj Digital Medicine
○ 118 papers in training set
Top 2%
2.6%
12
BMJ
● 51 papers in training set
Top 0.4%
2.1%
13
PLOS Medicine
● 110 papers in training set
Top 1%
1.9%
14
Scientific Reports
○ 3612 papers in training set
Top 49%
1.9%
15
Science Translational Medicine
● 127 papers in training set
Top 2%
1.7%
16
Nature Biomedical Engineering
○ 47 papers in training set
Top 0.9%
1.7%
17
Science
● 477 papers in training set
Top 5%
1.7%
18
Communications Medicine
○ 113 papers in training set
Top 2%
1.3%
19
Epidemiology
○ 32 papers in training set
Top 0.4%
1.1%
20
Annals of Internal Medicine
● 28 papers in training set
Top 0.4%
1.1%
21
New England Journal of Medicine
● 52 papers in training set
Top 0.4%
1.1%
22
eClinicalMedicine
○ 77 papers in training set
Top 2%
1.0%
23
PLOS Biology
● 486 papers in training set
Top 10%
0.9%
24
BMC Medical Research Methodology
○ 47 papers in training set
Top 1%
0.8%
25
International Journal of Infectious Diseases
○ 129 papers in training set
Top 3%
0.8%
26
The Lancet Digital Health
○ 25 papers in training set
Top 0.8%
0.8%
27
European Journal of Epidemiology
○ 43 papers in training set
Top 0.8%
0.8%
28
Vaccine
○ 203 papers in training set
Top 2%
0.6%
29
BMJ Open
● 601 papers in training set
Top 13%
0.6%
30
Eurosurveillance
● 83 papers in training set
Top 1%
0.6%

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.