{vaccineff 1.0.1} patches the package’s tests to fixfailing CRAN checks.
{vaccineff 1.0.0} refactors the package’s internalstructure for better maintainability.
estimate_vaccineff() replaceseffectiveness().
vaccineff.at parameter must always be provided for accurateresults.plot.vaccineff_data() replacesplot_coverage().
cohortdata has been simplified andreduced to improve examples and reduce computation time.
{vaccineff 0.0.4} simplifies data handling by usinglinelistobjects. Tags are assigned to the outcome, censoring, and vaccine datesusing the functionmake_vaccineff_data(), reducingredundancy in function input parameters.
The new pipeline includes the following three functions andcomplementary methods:summary andplot.
make_vaccineff_data(): Thisfunction returns an S3 object of the classvaccineff_data()with the study’s relevant information. It also allows the creation of amatched cohort to control for confounding variables by settingmatch = TRUE and passing the appropriateexactandnearest arguments. The methodsummary()can be used to check cohort characteristics, matching balance, and thesizes of matched, excluded, and removed populations.
plot_coverage(): This functionreturns a plot of the vaccine coverage or cumulative coverage. If thepopulation is matched, the plot includes the resulting count of dosesafter matching.
effectiveness(): This functionprovides methods for estimating VE using thesummary(), and a graphical representation ofthe methodology is generated byplot().
The following functions are no longer accessible to users, but theyare called withinmake_vaccineff_data():
make_immunization()match_cohort()Theplot() method returnslog-log andsurvival type plots when receiving an object of typeeffectiveness. This deprecates the functionsplot_survival() andplot_loglog().
This version introduces an iterative matching routine withinmatch_cohort(). After adjusting the exposure times of thepairs, new pairs are created between the removed ones and the unmatchedpopulation. The new matches with inconsistent exposure times are removedagain, and the procedure is repeated until no new pairs can be made. Theusage of all the functions remains unchanged by this update.
The number of functions and steps for computing vaccine effectivenesshas been drastically reduced in{vaccineff 0.0.2}. The newpipeline for estimation now consists of three main functions:
make_immunization(): Preparesinformation on immunization dates and vaccine status. It can handlemultiple columns for vaccine dates and custom vaccine statuses. In suchcases, it returns the name of the column selected as immunizing and thecustom name, if provided.
match_cohort(): This function hasbeen improved and generalized to reduce observation bias in cohorts. Thedefault matching strategy is static, based on nearest and exactcharacteristics using Mahalanobis distance. The exposure times of thepairs are adjusted after matching. In future releases, rolling calendarmatching will be introduced as a more accurate method to account forexposure times. The function returns an S3 object of classmatch, from which a summary and balance of the cohorts canbe printed using thesummary() method. The matched cohortcan be extracted using theget_dataset() method. Thematched cohort contains all the necessary information to estimatevaccine effectiveness.
effectiveness(): Receives a(matched) cohort and estimates vaccine effectiveness using the HazardRatio (HR). An S3 object of classeffectiveness isreturned, compatible with theplot() andsummary() methods. Future releases will provide relativerisk (RR) as an alternative for cases where the proportional hazardsassumption is not satisfied.
The following functions are no longer accessible to users. However,they are called withinmake_immunization():
get_immunization_date()get_immunization_dose()get_immunization_vaccine()set_status()Similarly, theeffectiveness() function deprecates theuse ofcoh_eff_noconf(), and theplot() methodnow returns a log-log plot, replacing theplot_loglog()function.