Recent Infection Testing Algorithms (RITA) are used in cross-sectional surveys to identify individuals whose HIV infections are likely to have been recency aquired. The methods of Fellows (2022) are used to calculate incidence adjusting for the RITA process.
| Authors: | Ian E. Fellows |
| Github: | https://github.com/fellstat/recent |
| Documentation: | https://fellstat.github.io/recent/index.html |
| Technical Paper: | https://arxiv.org/abs/2204.00048 |
The Triangulator is a Shiny user interface designed to help derive consensus estimates of a population quantity (e.g. a population size, a proportion, a mean, etc.) from multiple empirical estimates. Stakeholders may add additional information regarding the methodological quality of the studies and prior knowledge of the metric. Triangulated estimates are statistically defensible, reproducible and openly inspectable.
| Authors: | Ian E. Fellows and Carl Corcoran |
| Github: | https://github.com/fellstat/triangulator |
| Manual: | https://fellstat.github.io/triangulator/ |
Implements user interfaces for log-linear models, Bayesian model averaging and Bayesian Dirichlet process mixture models.
| Authors: | Ian E. Fellows |
| Video Tutorial: | https://www.youtube.com/watch?v=PgmyUnFlo5Y&feature=youtu.be |
| Manual: | https://fellstat.github.io/shinyrecap/ |
| Github: | https://github.com/fellstat/shinyrecap |
Implements a user interface for a Bayesian hierarchical model used to estimate the size of local and national populations. The model incorporates multiple commonly used data sources including mapping data, surveys, interventions, capture-recapture data, estimates or guesstimates from organizations, and expert opinion
| Authors: | Jacob Parsons using the algorithm developed by Le Bao, Adrian E. Raftery and Kyongwon Kim |
| CRAN Repository: | https://CRAN.R-project.org/package=SizeEstimation |
| Reference: | Bao, L., Raftery, A. E., & Reddy, A. (2015). Estimating the sizes of populations at risk of HIV infection from multiple data sources using a Bayesian hierarchical model. Statistics and its Interface, 8(2), 125-136. |
Utilizes crosssectional survey data containing information on participants' testing history and diagnosis to estimate incidence.
| Authors: | Ian E. Fellows |
| Video Tutorial: | https://www.youtube.com/watch?v=YVPcLLs9zxc&t=08s |
| Manual: | https://github.com/fellstat/TestingHistoryIncidence/wiki/ Shiny-App-Documentation |
| Github: | https://github.com/fellstat/TestingHistoryIncidence |
| Example Data: | https://raw.githubusercontent.com/fellstat/ TestingHistoryIncidence/master/inst/shiny_ui/tstdat.csv |
Time location sampling (TLS) is an important tool for surveying populations that attend venues. It is particularly useful in populations (such as injection drug users and men who have sex with men) that are difficult to reach via other methods.
| Authors: | Ian E. Fellows |
| Github: | https://github.com/fellstat/shinytls |
While numerous methodologies exist to estimate proportions and means from an respondent driven sample (RDS), estimating population size is more challenging. A privatized network sample (PNS) is similar to an RDS, but also collects identifiers for each individual’s network connections protecting privacy through the use of a hash function.
| Authors: | Ian E. Fellows |
| Github: | https://github.com/fellstat/pnspop |
| Manual: | https://fellstat.github.io/pnspop/ |