,
Marc Choisy [aut] | Title: | Model Infectious Disease Parameters from Serosurveys |
|---|---|
| Description: | An easy-to-use and efficient tool to estimate infectious diseases parameters using serological data. Implemented models include SIR models (basic_sir_model(), static_sir_model(), mseir_model(), sir_subpops_model()), parametric models (polynomial_model(), fp_model()), nonparametric models (lp_model()), semiparametric models (penalized_splines_model()), hierarchical models (hierarchical_bayesian_model()). The package is based on the book "Modeling Infectious Disease Parameters Based on Serological and Social Contact Data: A Modern Statistical Perspective" (Hens, Niel & Shkedy, Ziv & Aerts, Marc & Faes, Christel & Damme, Pierre & Beutels, Philippe., 2013) <doi:10.1007/978-1-4614-4072-7>. |
| Authors: | Anh Phan Truong Quynh [aut, cre],
Nguyen Pham Nguyen The [aut],
Long Bui Thanh [aut],
Tuyen Huynh [aut],
Thinh Ong [aut] ,
Marc Choisy [aut] |
| Maintainer: | Anh Phan Truong Quynh <[email protected]> |
| License: | MIT + file LICENSE |
| Version: | 1.0.0.9000 |
| Built: | 2024-11-07 05:32:19 UTC |
| Source: | https://github.com/oucru-modelling/serosv |
An easy-to-use and efficient tool to estimate infectious diseases parameters using serological data. Implemented models include SIR models (basic_sir_model(), static_sir_model(), mseir_model(), sir_subpops_model()), parametric models (polynomial_model(), fp_model()), nonparametric models (lp_model()), semiparametric models (penalized_splines_model()), hierarchical models (hierarchical_bayesian_model()). The package is based on the book "Modeling Infectious Disease Parameters Based on Serological and Social Contact Data: A Modern Statistical Perspective" (Hens, Niel & Shkedy, Ziv & Aerts, Marc & Faes, Christel & Damme, Pierre & Beutels, Philippe., 2013) doi:10.1007/978-1-4614-4072-7.
Maintainer: Anh Phan Truong Quynh [email protected]
Authors:
Nguyen Pham Nguyen The [email protected]
Long Bui Thanh
Tuyen Huynh [email protected]
Thinh Ong [email protected] (ORCID)
Marc Choisy [email protected] (ORCID)
Useful links:
Report bugs at https://github.com/OUCRU-Modelling/serosv/issues
Compute confidence interval
compute_ci(x, ci = 0.95, le = 100, ...)compute_ci(x, ci = 0.95, le = 100, ...)
x |
- serosv models |
ci |
- confidence interval |
le |
- number of data for computing confidence interval |
... |
- arbitrary argument |
confidence interval dataframe with 4 variables, x and y for the fitted values and ymin and ymax for the confidence interval
Compute confidence interval for fractional polynomial model
compute_ci.fp_model(x, ci = 0.95, le = 100, ...)compute_ci.fp_model(x, ci = 0.95, le = 100, ...)
x |
- serosv models |
ci |
- confidence interval |
le |
- number of data for computing confidence interval |
... |
- arbitrary argument |
confidence interval dataframe with 4 variables, x and y for the fitted values and ymin and ymax for the confidence interval
Compute confidence interval for local polynomial model
compute_ci.lp_model(x, ci = 0.95, ...)compute_ci.lp_model(x, ci = 0.95, ...)
x |
- serosv models |
ci |
- confidence interval |
... |
- arbitrary arguments |
confidence interval dataframe with 4 variables, x and y for the fitted values and ymin and ymax for the confidence interval
Compute confidence interval for mixture model
compute_ci.mixture_model(x, ci = 0.95, ...)compute_ci.mixture_model(x, ci = 0.95, ...)
x |
- serosv mixture_model object |
ci |
- confidence interval |
... |
- arbitrary arguments |
list of confidence interval for susceptible and infected. Each confidence interval is a list with 2 items for lower and upper bound of the interval.
Compute confidence interval for penalized_spline_model
compute_ci.penalized_spline_model(x, ci = 0.95, ...)compute_ci.penalized_spline_model(x, ci = 0.95, ...)
x |
- serosv models |
ci |
- confidence interval |
... |
- arbitrary arguments |
list of confidence interval for seroprevalence and foi Each confidence interval dataframe with 4 variables, x and y for the fitted values and ymin and ymax for the confidence interval
Compute confidence interval for Weibull model
compute_ci.weibull_model(x, ci = 0.95, ...)compute_ci.weibull_model(x, ci = 0.95, ...)
x |
- serosv models |
ci |
- confidence interval |
... |
- arbitrary argument |
confidence interval dataframe with 4 variables, x and y for the fitted values and ymin and ymax for the confidence interval
Estimate force of infection
est_foi(t, sp)est_foi(t, sp)
t |
- time (in this case age) vector |
sp |
- seroprevalence vector |
computed foi vector
Refers to section 11.2 - 11.4
estimate_from_mixture( age, antibody_level, threshold_status = NULL, mixture_model, s = "ps", sp = 83, monotonize = TRUE )estimate_from_mixture( age, antibody_level, threshold_status = NULL, mixture_model, s = "ps", sp = 83, monotonize = TRUE )
age |
- vector of age |
antibody_level |
- vector of the corresponding raw antibody level |
threshold_status |
- sero status using threshold approach in line listing (optional, for visualization and comparison only) |
mixture_model |
- mixture_model object generated by serosv::mixture_model() |
s |
- smoothing basis used to fit antibody level |
sp |
- smoothing parameter |
monotonize |
- whether to monotonize seroprevalence (default to TRUE) |
a list of class estimated_from_mixture with the following items
df |
the dataframe used for fitting the model |
info |
a fitted "gam" model for mu(a) |
sp |
seroprevalence |
foi |
force of infection |
threshold_status |
serostatus using threshold method only if provided |
[mgcv::gam()] for more information about the fitted gam object
Refers to section 6.1.2.
farrington_model( age, start, pos = NULL, tot = NULL, status = NULL, fixed = list() )farrington_model( age, start, pos = NULL, tot = NULL, status = NULL, fixed = list() )
age |
the age vector. |
start |
Named list of vectors or single vector. Initial values for optimizer. |
pos |
the positive count vector (optional if status is provided). |
tot |
the total count vector (optional if status is provided). |
status |
the serostatus vector (optional if pos & tot are provided). |
fixed |
Named list of vectors or single vector. Parameter values to keep fixed during optimization. |
a list of class farrington_model with 5 items
datatype |
type of datatype used for model fitting (aggregated or linelisting) |
df |
the dataframe used for fitting the model |
info |
fitted "glm" object |
sp |
seroprevalence |
foi |
force of infection |
[stats::glm()] for more information on the fitted glm object
df <- rubella_uk_1986_1987 model <- farrington_model( df$age, pos = df$pos, tot = df$tot, start=list(alpha=0.07,beta=0.1,gamma=0.03) ) plot(model)df <- rubella_uk_1986_1987 model <- farrington_model( df$age, pos = df$pos, tot = df$tot, start=list(alpha=0.07,beta=0.1,gamma=0.03) ) plot(model)
Refers to section 6.2.
find_best_fp_powers(age, pos, tot, p, mc, degree, link = "logit")find_best_fp_powers(age, pos, tot, p, mc, degree, link = "logit")
age |
the age vector. |
pos |
the pos vector. |
tot |
the tot vector. |
p |
a powers sequence. |
mc |
indicates if the returned model should be monotonic. |
degree |
the degree of the model. Recommended to be <= 2. |
link |
the link function. Defaulted to "logit". |
list of 3 elements:
p |
The best power for fp model. |
deviance |
Deviance of the best fitted model. |
model |
The best model fitted |
df <- hav_be_1993_1994 best_p <- find_best_fp_powers( df$age, df$pos, df$tot, p=seq(-2,3,0.1), mc=FALSE, degree=2, link="cloglog" ) best_pdf <- hav_be_1993_1994 best_p <- find_best_fp_powers( df$age, df$pos, df$tot, p=seq(-2,3,0.1), mc=FALSE, degree=2, link="cloglog" ) best_p
Refers to section 6.2.
fp_model(age, p, pos = NULL, tot = NULL, status = NULL, link = "logit")fp_model(age, p, pos = NULL, tot = NULL, status = NULL, link = "logit")
age |
the age vector. |
p |
the powers of the predictor. |
pos |
the positive count vector (optional if status is provided). |
tot |
the total count vector (optional if status is provided). |
status |
the serostatus vector (optional if pos & tot are provided). |
link |
the link function for model. Defaulted to "logit". |
a list of class fp_model with 5 items
datatype |
type of data used for fitting model (aggregated or linelisting) |
df |
the dataframe used for fitting the model |
info |
a fitted glm model |
sp |
seroprevalence |
foi |
force of infection |
[stats::glm()] for more information on glm object
df <- hav_be_1993_1994 model <- fp_model( df$age, pos = df$pos, tot = df$tot, p=c(1.5, 1.6), link="cloglog") plot(model)df <- hav_be_1993_1994 model <- fp_model( df$age, pos = df$pos, tot = df$tot, p=c(1.5, 1.6), link="cloglog") plot(model)
A study of the prevalence of HAV antibodies conducted in the Flemish Community of Belgium in 1993 and early 1994
hav_be_1993_1994hav_be_1993_1994
A data frame with 3 variables:
Age group
Number of seropositive individuals
Total number of individuals surveyed
Beutels, M., Van Damme, P., Aelvoet, W. et al. Prevalence of hepatitis A, B and C in the Flemish population. Eur J Epidemiol 13, 275-280 (1997). doi:10.1023/A:1007393405966
# Reproduce Fig 4.1 (upper left panel), p. 63 age <- hav_be_1993_1994$age pos <- hav_be_1993_1994$pos tot <- hav_be_1993_1994$tot plot( age, pos / tot, pty = "s", cex = 0.06 * tot, pch = 16, xlab = "age", ylab = "seroprevalence", xlim = c(0, 86), ylim = c(0, 1) )# Reproduce Fig 4.1 (upper left panel), p. 63 age <- hav_be_1993_1994$age pos <- hav_be_1993_1994$pos tot <- hav_be_1993_1994$tot plot( age, pos / tot, pty = "s", cex = 0.06 * tot, pch = 16, xlab = "age", ylab = "seroprevalence", xlim = c(0, 86), ylim = c(0, 1) )
A subset of the serological dataset of Varicella-Zoster Virus (VZV) and Parvovirus B19 in Belgium where only individuals living in Flanders were selected
hav_be_2002hav_be_2002
A data frame with 2 variables:
Age of individual
If the individual is seropositive or not
Thiry, N., Beutels, P., Shkedy, Z. et al. The seroepidemiology of primary varicella-zoster virus infection in Flanders (Belgium). Eur J Pediatr 161, 588-593 (2002). doi:10.1007/s00431-002-1053-2
# Reproduce Fig 4.1 (upper right panel), p. 63 library(dplyr) df <- hav_be_2002 %>% group_by(age) %>% summarise(pos = sum(seropositive), tot = n()) plot( df$age, df$pos / df$tot, pty = "s", cex = 0.06 * df$tot, pch = 16, xlab = "age", ylab = "seroprevalence", xlim = c(0, 86), ylim = c(0, 1) )# Reproduce Fig 4.1 (upper right panel), p. 63 library(dplyr) df <- hav_be_2002 %>% group_by(age) %>% summarise(pos = sum(seropositive), tot = n()) plot( df$age, df$pos / df$tot, pty = "s", cex = 0.06 * df$tot, pch = 16, xlab = "age", ylab = "seroprevalence", xlim = c(0, 86), ylim = c(0, 1) )
A cross-sectional survey conducted in 1964 in Bulgaria. Samples were collected from schoolchildren and blood donors.
hav_bg_1964hav_bg_1964
A data frame with 3 variables:
Age group
Number of seropositive individuals
Total number of individuals surveyed
Keiding, Niels. "Age-Specific Incidence and Prevalence: A Statistical Perspective." Journal of the Royal Statistical Society. Series A (Statistics in Society) 154, no. 3 (1991): 371-412. doi:10.2307/2983150
# Reproduce Fig 4.1 (lower panel), p. 63 age <- hav_bg_1964$age pos <- hav_bg_1964$pos tot <- hav_bg_1964$tot plot( age, pos / tot, pty = "s", cex = 0.08 * tot, pch = 16, xlab = "age", ylab = "seroprevalence", xlim = c(0, 86), ylim = c(0, 1) )# Reproduce Fig 4.1 (lower panel), p. 63 age <- hav_bg_1964$age pos <- hav_bg_1964$pos tot <- hav_bg_1964$tot plot( age, pos / tot, pty = "s", cex = 0.08 * tot, pch = 16, xlab = "age", ylab = "seroprevalence", xlim = c(0, 86), ylim = c(0, 1) )
A seroprevalence study conducted in St. Petersburg (more information in the book)
hbv_ru_1999hbv_ru_1999
A data frame with 4 variables:
Age group
Number of seropositive individuals
Total number of individuals surveyed
Gender of cohort (unsure what 1 and 2 means)
Mukomolov, S., L. Shliakhtenko, I. Levakova, and E. Shargorodskaya. Viral hepatitis in Russian federation. An analytical overview. Technical Report 213 (3), 3rd edn. St Petersburg Pasteur Institute, St Petersburg, 2000.
# Reproduce Fig 4.2, p. 65 library(dplyr) hbv_ru_1999$age <- trunc(hbv_ru_1999$age / 1) * 1 hbv_ru_1999$age[hbv_ru_1999$age > 40] <- trunc( hbv_ru_1999$age[hbv_ru_1999$age > 40] / 5 ) * 5 df <- hbv_ru_1999 %>% group_by(age) %>% summarise(pos = sum(pos), tot = sum(tot)) plot( df$age, df$pos / df$tot, cex = 0.05 * df$tot, pch = 16, xlab = "age", ylab = "seroprevalence", xlim = c(0, 72) )# Reproduce Fig 4.2, p. 65 library(dplyr) hbv_ru_1999$age <- trunc(hbv_ru_1999$age / 1) * 1 hbv_ru_1999$age[hbv_ru_1999$age > 40] <- trunc( hbv_ru_1999$age[hbv_ru_1999$age > 40] / 5 ) * 5 df <- hbv_ru_1999 %>% group_by(age) %>% summarise(pos = sum(pos), tot = sum(tot)) plot( df$age, df$pos / df$tot, cex = 0.05 * df$tot, pch = 16, xlab = "age", ylab = "seroprevalence", xlim = c(0, 72) )
A study of HCV infection among injecting drug users. All injecting drug users were interviewed by means of a standardized face-to-face interview and information on their socio-demographic status, drug use history, drug use, and related risk behavior was recorded
hcv_be_2006hcv_be_2006
A data frame with 3 variables:
Duration of injection/Exposure time (years)
If the individual is seropositive or not
Mathei, C., Shkedy, Z., Denis, B., Kabali, C., Aerts, M., Molenberghs, G., Van Damme, P. and Buntinx, F. (2006), Evidence for a substantial role of sharing of injecting paraphernalia other than syringes/needles to the spread of hepatitis C among injecting drug users. Journal of Viral Hepatitis, 13: 560-570. doi:10.1111/j.1365-2893.2006.00725.x
# Reproduce Fig 4.3, p. 66 library(dplyr) # snapping age to aggregated age group # (credit: https://stackoverflow.com/a/12861810) groups <- c(0.5:24.5) range <- 0.5 low <- findInterval(hcv_be_2006$dur, groups) high <- low + 1 low_diff <- hcv_be_2006$dur - groups[ifelse(low == 0, NA, low)] high_diff <- groups[ifelse(high == 0, NA, high)] - hcv_be_2006$dur mins <- pmin(low_diff, high_diff, na.rm = TRUE) pick <- ifelse(!is.na(low_diff) & mins == low_diff, low, high) hcv_be_2006$dur <- ifelse( mins <= range + .Machine$double.eps, groups[pick], hcv_be_2006$dur ) hcv_be_2006 <- hcv_be_2006 %>% group_by(dur) %>% summarise(tot = n(), pos = sum(seropositive)) plot( hcv_be_2006$dur, hcv_be_2006$pos / hcv_be_2006$tot, cex = 0.1 * hcv_be_2006$tot, pch = 16, xlab = "duration of injection (years)", ylab = "seroprevalence", xlim = c(0, 25), ylim = c(0, 1) )# Reproduce Fig 4.3, p. 66 library(dplyr) # snapping age to aggregated age group # (credit: https://stackoverflow.com/a/12861810) groups <- c(0.5:24.5) range <- 0.5 low <- findInterval(hcv_be_2006$dur, groups) high <- low + 1 low_diff <- hcv_be_2006$dur - groups[ifelse(low == 0, NA, low)] high_diff <- groups[ifelse(high == 0, NA, high)] - hcv_be_2006$dur mins <- pmin(low_diff, high_diff, na.rm = TRUE) pick <- ifelse(!is.na(low_diff) & mins == low_diff, low, high) hcv_be_2006$dur <- ifelse( mins <= range + .Machine$double.eps, groups[pick], hcv_be_2006$dur ) hcv_be_2006 <- hcv_be_2006 %>% group_by(dur) %>% summarise(tot = n(), pos = sum(seropositive)) plot( hcv_be_2006$dur, hcv_be_2006$pos / hcv_be_2006$tot, cex = 0.1 * hcv_be_2006$tot, pch = 16, xlab = "duration of injection (years)", ylab = "seroprevalence", xlim = c(0, 25), ylim = c(0, 1) )
Refers to section 10.3
hierarchical_bayesian_model( age, pos = NULL, tot = NULL, status = NULL, type = "far3", chains = 1, warmup = 1500, iter = 5000 )hierarchical_bayesian_model( age, pos = NULL, tot = NULL, status = NULL, type = "far3", chains = 1, warmup = 1500, iter = 5000 )
age |
the age vector |
pos |
the positive count vector (optional if status is provided). |
tot |
the total count vector (optional if status is provided). |
status |
the serostatus vector (optional if pos & tot are provided). |
type |
type of model ("far2", "far3" or "log_logistic") |
chains |
number of Markov chains |
warmup |
number of warmup runs |
iter |
number of iterations |
a list of class hierarchical_bayesian_model with 6 items
datatype |
type of datatype used for model fitting (aggregated or linelisting) |
df |
the dataframe used for fitting the model |
type |
type of bayesian model far2, far3 or log_logistic |
info |
parameters for the fitted model |
sp |
seroprevalence |
foi |
force of infection |
df <- mumps_uk_1986_1987 model <- hierarchical_bayesian_model(age = df$age, pos = df$pos, tot = df$tot, type="far3") model$info plot(model)df <- mumps_uk_1986_1987 model <- hierarchical_bayesian_model(age = df$age, pos = df$pos, tot = df$tot, type="far3") model$info plot(model)
Refers to section 7.1. and 7.2.
lp_model( age, pos = NULL, tot = NULL, status = NULL, kern = "tcub", nn = 0, h = 0, deg = 2 )lp_model( age, pos = NULL, tot = NULL, status = NULL, kern = "tcub", nn = 0, h = 0, deg = 2 )
age |
the age vector. |
pos |
the positive count vector (optional if status is provided). |
tot |
the total count vector (optional if status is provided). |
status |
the serostatus vector (optional if pos & tot are provided). |
kern |
Weight function, default = "tcub". Other choices are "rect", "trwt", "tria", "epan", "bisq" and "gauss". Choices may be restricted when derivatives are required; e.g. for confidence bands and some bandwidth selectors. |
nn |
Nearest neighbor component of the smoothing parameter. Default value is 0.7, unless either h is provided, in which case the default is 0. |
h |
The constant component of the smoothing parameter. Default: 0. |
deg |
Degree of polynomial to use. Default: 2. |
a list of class lp_model with 6 items
datatype |
type of datatype used for model fitting (aggregated or linelisting) |
df |
the dataframe used for fitting the model |
pi |
fitted locfit object for pi |
eta |
fitted locfit object for eta |
sp |
seroprevalence |
foi |
force of infection |
[locfit::locfit()] for more information on the fitted locfit object
df <- mumps_uk_1986_1987 model <- lp_model( df$age, pos = df$pos, tot = df$tot, nn=0.7, kern="tcub" ) plot(model)df <- mumps_uk_1986_1987 model <- lp_model( df$age, pos = df$pos, tot = df$tot, nn=0.7, kern="tcub" ) plot(model)
Refers to section 11.1 - 11.4
mixture_model( antibody_level, breaks = 40, pi = c(0.2, 0.8), mu = c(2, 6), sigma = c(0.5, 1) )mixture_model( antibody_level, breaks = 40, pi = c(0.2, 0.8), mu = c(2, 6), sigma = c(0.5, 1) )
antibody_level |
- vector of the corresponding raw antibody level |
breaks |
- number of intervals which the antibody_level are grouped into |
pi |
- proportion of susceptible, infected |
mu |
- a vector of means of component distributions (vector of 2 numbers in ascending order) |
sigma |
- a vector of standard deviations of component distributions (vector of 2 number) |
a list of class mixture_model with the following items
df |
the dataframe used for fitting the model |
info |
list of 3 items parameters, distribution and constraints for the fitted model |
susceptible |
fitted distribution for susceptible |
infected |
fitted distribution for infected |
df <- vzv_be_2001_2003[vzv_be_2001_2003$age < 40.5,] data <- df$VZVmIUml[order(df$age)] model <- mixture_model(antibody_level = data) model$info plot(model)df <- vzv_be_2001_2003[vzv_be_2001_2003$age < 40.5,] data <- df$VZVmIUml[order(df$age)] model <- mixture_model(antibody_level = data) model$info plot(model)
Refers to section 3.4.
mseir_model(a, gamma, lambda, sigma, nu)mseir_model(a, gamma, lambda, sigma, nu)
a |
age sequence |
gamma |
time in maternal class. |
lambda |
time in susceptible class. |
sigma |
time in latent class. |
nu |
time in infected class. |
list of class mseir_model with the following parameters
parameters |
list of parameters used for fitting the model |
output |
matrix of proportion for each compartment over time |
model <- mseir_model( a=seq(from=1,to=20,length=500), # age range from 0 -> 20 yo gamma=1/0.5, # 6 months in the maternal antibodies lambda=0.2, # 5 years in the susceptible class sigma=26.07, # 14 days in the latent class nu=36.5 # 10 days in the infected class ) modelmodel <- mseir_model( a=seq(from=1,to=20,length=500), # age range from 0 -> 20 yo gamma=1/0.5, # 6 months in the maternal antibodies lambda=0.2, # 5 years in the susceptible class sigma=26.07, # 14 days in the latent class nu=36.5 # 10 days in the infected class ) model
a large survey of prevalence of antibodies to mumps and rubella viruses in the UK. The survey, covering subjects from 1 to over 65 years of age, provides information on the prevalence of antibody by age
mumps_uk_1986_1987mumps_uk_1986_1987
A data frame with 3 variables:
Age group
Number of seropositive individuals
Total number of individuals surveyed
Morgan-Capner P, Wright J, Miller C L, Miller E. Surveillance of antibody to measles, mumps, and rubella by age. British Medical Journal 1988; 297 :770 doi:10.1136/bmj.297.6651.770
# Reproduce Fig 4.4 (left panel), p. 67 age <- mumps_uk_1986_1987$age pos <- mumps_uk_1986_1987$pos tot <- mumps_uk_1986_1987$tot plot(age, pos / tot, cex = 0.008 * tot, pch = 16, xlab = "age", ylab = "seroprevalence", xlim = c(0, 45), ylim = c(0, 1) )# Reproduce Fig 4.4 (left panel), p. 67 age <- mumps_uk_1986_1987$age pos <- mumps_uk_1986_1987$pos tot <- mumps_uk_1986_1987$tot plot(age, pos / tot, cex = 0.008 * tot, pch = 16, xlab = "age", ylab = "seroprevalence", xlim = c(0, 45), ylim = c(0, 1) )
A seroprevalence survey testing for parvovirus B19 IgG antibody, performed on large representative national serum banks in Belgium, England and Wales, Finland, Italy, and Poland. The sera were collected between 1995 and 2004 and were obtained from residual sera submitted for routine laboratory testing.
parvob19_be_2001_2003parvob19_be_2001_2003
A data frame with 5 variables:
Age of individual
If the individual is seropositive or not
Year surveyed
Gender of individual
Parvo B19 antibody units per ml
MOSSONG, J., N. HENS, V. FRIEDERICHS, I. DAVIDKIN, M. BROMAN, B. LITWINSKA, J. SIENNICKA, et al. "Parvovirus B19 Infection in Five European Countries: Seroepidemiology, Force of Infection and Maternal Risk of Infection." Epidemiology and Infection 136, no. 8 (2008): 1059-68. doi:10.1017/S0950268807009661
# Reproduce Fig 4.5 (left upper panel), p. 68 library(dplyr) df <- parvob19_be_2001_2003 %>% group_by(age) %>% summarise(pos = sum(seropositive), tot = n()) plot(df$age, df$pos / df$tot, cex = 0.02 * df$tot, pch = 16, xlab = "age", ylab = "seroprevalence", xlim = c(0, 82), ylim = c(0, 1) )# Reproduce Fig 4.5 (left upper panel), p. 68 library(dplyr) df <- parvob19_be_2001_2003 %>% group_by(age) %>% summarise(pos = sum(seropositive), tot = n()) plot(df$age, df$pos / df$tot, cex = 0.02 * df$tot, pch = 16, xlab = "age", ylab = "seroprevalence", xlim = c(0, 82), ylim = c(0, 1) )
A seroprevalence survey testing for parvovirus B19 IgG antibody, performed on large representative national serum banks in Belgium, England and Wales, Finland, Italy, and Poland. The sera were collected between 1995 and 2004 and were obtained from residual sera submitted for routine laboratory testing.
parvob19_ew_1996parvob19_ew_1996
A data frame with 5 variables:
Age of individual
If the individual is seropositive or not
Year surveyed
Gender of individual
Parvo B19 antibody units per ml
MOSSONG, J., N. HENS, V. FRIEDERICHS, I. DAVIDKIN, M. BROMAN, B. LITWINSKA, J. SIENNICKA, et al. "Parvovirus B19 Infection in Five European Countries: Seroepidemiology, Force of Infection and Maternal Risk of Infection." Epidemiology and Infection 136, no. 8 (2008): 1059-68. doi:10.1017/S0950268807009661
# Reproduce Fig 4.5 (right upper panel), p. 68 # NB: This figure will look different to that of in the book, since we # believe that the original authors has made some errors in specifying # the sample size of the dots. library(dplyr) df <- parvob19_ew_1996 %>% mutate(age = round(age)) %>% group_by(age) %>% summarise(pos = sum(seropositive), tot = n()) plot(df$age, df$pos / df$tot, cex = 0.02 * df$tot, pch = 16, xlab = "age", ylab = "seroprevalence", xlim = c(0, 82), ylim = c(0, 1) )# Reproduce Fig 4.5 (right upper panel), p. 68 # NB: This figure will look different to that of in the book, since we # believe that the original authors has made some errors in specifying # the sample size of the dots. library(dplyr) df <- parvob19_ew_1996 %>% mutate(age = round(age)) %>% group_by(age) %>% summarise(pos = sum(seropositive), tot = n()) plot(df$age, df$pos / df$tot, cex = 0.02 * df$tot, pch = 16, xlab = "age", ylab = "seroprevalence", xlim = c(0, 82), ylim = c(0, 1) )
A seroprevalence survey testing for parvovirus B19 IgG antibody, performed on large representative national serum banks in Belgium, England and Wales, Finland, Italy, and Poland. The sera were collected between 1995 and 2004 and were obtained from residual sera submitted for routine laboratory testing.
parvob19_fi_1997_1998parvob19_fi_1997_1998
A data frame with 5 variables:
Age of individual
If the individual is seropositive or not
Year surveyed
Gender of individual
Parvo B19 antibody units per ml
MOSSONG, J., N. HENS, V. FRIEDERICHS, I. DAVIDKIN, M. BROMAN, B. LITWINSKA, J. SIENNICKA, et al. "Parvovirus B19 Infection in Five European Countries: Seroepidemiology, Force of Infection and Maternal Risk of Infection." Epidemiology and Infection 136, no. 8 (2008): 1059-68. doi:10.1017/S0950268807009661
# Reproduce Fig 4.5 (left bottom panel), p. 68 # NB: This figure will look different to that of in the book, since we # believe that the original authors has made some errors in specifying # the sample size of the dots. library(dplyr) df <- parvob19_fi_1997_1998 %>% mutate(age = round(age)) %>% group_by(age) %>% summarise(pos = sum(seropositive), tot = n()) plot(df$age, df$pos / df$tot, cex = 0.07 * df$tot, pch = 16, xlab = "age", ylab = "seroprevalence", xlim = c(0, 82), ylim = c(0, 1) )# Reproduce Fig 4.5 (left bottom panel), p. 68 # NB: This figure will look different to that of in the book, since we # believe that the original authors has made some errors in specifying # the sample size of the dots. library(dplyr) df <- parvob19_fi_1997_1998 %>% mutate(age = round(age)) %>% group_by(age) %>% summarise(pos = sum(seropositive), tot = n()) plot(df$age, df$pos / df$tot, cex = 0.07 * df$tot, pch = 16, xlab = "age", ylab = "seroprevalence", xlim = c(0, 82), ylim = c(0, 1) )
A seroprevalence survey testing for parvovirus B19 IgG antibody, performed on large representative national serum banks in Belgium, England and Wales, Finland, Italy, and Poland. The sera were collected between 1995 and 2004 and were obtained from residual sera submitted for routine laboratory testing.
parvob19_it_2003_2004parvob19_it_2003_2004
A data frame with 5 variables:
Age of individual
If the individual is seropositive or not
Year surveyed
Gender of individual
Parvo B19 antibody units per ml
MOSSONG, J., N. HENS, V. FRIEDERICHS, I. DAVIDKIN, M. BROMAN, B. LITWINSKA, J. SIENNICKA, et al. "Parvovirus B19 Infection in Five European Countries: Seroepidemiology, Force of Infection and Maternal Risk of Infection." Epidemiology and Infection 136, no. 8 (2008): 1059-68. doi:10.1017/S0950268807009661
# Reproduce Fig 4.5 (middle bottom panel), p. 68 # NB: This figure will look different to that of in the book, since we # believe that the original authors has made some errors in specifying # the sample size of the dots. library(dplyr) df <- parvob19_it_2003_2004 %>% group_by(age) %>% summarise(pos = sum(seropositive), tot = n()) plot(df$age, df$pos / df$tot, cex = 0.07 * df$tot, pch = 16, xlab = "age", ylab = "seroprevalence", xlim = c(0, 82), ylim = c(0, 1) )# Reproduce Fig 4.5 (middle bottom panel), p. 68 # NB: This figure will look different to that of in the book, since we # believe that the original authors has made some errors in specifying # the sample size of the dots. library(dplyr) df <- parvob19_it_2003_2004 %>% group_by(age) %>% summarise(pos = sum(seropositive), tot = n()) plot(df$age, df$pos / df$tot, cex = 0.07 * df$tot, pch = 16, xlab = "age", ylab = "seroprevalence", xlim = c(0, 82), ylim = c(0, 1) )
A seroprevalence survey testing for parvovirus B19 IgG antibody, performed on large representative national serum banks in Belgium, England and Wales, Finland, Italy, and Poland. The sera were collected between 1995 and 2004 and were obtained from residual sera submitted for routine laboratory testing.
parvob19_pl_1995_2004parvob19_pl_1995_2004
A data frame with 5 variables:
Age of individual
If the individual is seropositive or not
Year surveyed
Gender of individual
Parvo B19 antibody units per ml
MOSSONG, J., N. HENS, V. FRIEDERICHS, I. DAVIDKIN, M. BROMAN, B. LITWINSKA, J. SIENNICKA, et al. "Parvovirus B19 Infection in Five European Countries: Seroepidemiology, Force of Infection and Maternal Risk of Infection." Epidemiology and Infection 136, no. 8 (2008): 1059-68. doi:10.1017/S0950268807009661
# Reproduce Fig 4.5 (right bottom panel), p. 68 # NB: This figure will look different to that of in the book, since we # believe that the original authors has made some errors in specifying # the sample size of the dots. library(dplyr) df <- parvob19_pl_1995_2004 %>% mutate(age = round(age)) %>% group_by(age) %>% summarise(pos = sum(seropositive), tot = n()) plot(df$age, df$pos / df$tot, cex = 0.07 * df$tot, pch = 16, xlab = "age", ylab = "seroprevalence", xlim = c(0, 82), ylim = c(0, 1) )# Reproduce Fig 4.5 (right bottom panel), p. 68 # NB: This figure will look different to that of in the book, since we # believe that the original authors has made some errors in specifying # the sample size of the dots. library(dplyr) df <- parvob19_pl_1995_2004 %>% mutate(age = round(age)) %>% group_by(age) %>% summarise(pos = sum(seropositive), tot = n()) plot(df$age, df$pos / df$tot, cex = 0.07 * df$tot, pch = 16, xlab = "age", ylab = "seroprevalence", xlim = c(0, 82), ylim = c(0, 1) )
Monotonize seroprevalence
pava(pos = pos, tot = rep(1, length(pos)))pava(pos = pos, tot = rep(1, length(pos)))
pos |
the positive count vector. |
tot |
the total count vector. |
computed list of 2 items pai1 for original values and pai2 for monotonized value
Penalized Spline model
penalized_spline_model( age, pos = NULL, tot = NULL, status = NULL, s = "bs", link = "logit", framework = "pl", sp = NULL )penalized_spline_model( age, pos = NULL, tot = NULL, status = NULL, s = "bs", link = "logit", framework = "pl", sp = NULL )
age |
the age vector |
pos |
the positive count vector (optional if status is provided). |
tot |
the total count vector (optional if status is provided). |
status |
the serostatus vector (optional if pos & tot are provided). |
s |
smoothing basis to use |
link |
link function to use |
framework |
which approach to fit the model ("pl" for penalized likelihood framework, "glmm" for generalized linear mixed model framework) |
sp |
smoothing parameter |
a list of class penalized_spline_model with 6 attributes
datatype |
type of datatype used for model fitting (aggregated or linelisting) |
df |
the dataframe used for fitting the model |
framework |
either pl or glmm |
info |
fitted "gam" model when framework is pl or "gamm" model when framework is glmm |
sp |
seroprevalence |
foi |
force of infection |
[mgcv::gam()], [mgcv::gamm()] for more information the fitted gam and gamm model
data <- parvob19_be_2001_2003 model <- penalized_spline_model(data$age, status = data$seropositive, framework="glmm") model$gam$info plot(model)data <- parvob19_be_2001_2003 model <- penalized_spline_model(data$age, status = data$seropositive, framework="glmm") model$gam$info plot(model)
Refers to section 7.2.
plot_gcv(age, pos, tot, nn_seq, h_seq, kern = "tcub", deg = 2)plot_gcv(age, pos, tot, nn_seq, h_seq, kern = "tcub", deg = 2)
age |
the age vector. |
pos |
the pos vector. |
tot |
the tot vector.#' |
nn_seq |
Nearest neighbor sequence. |
h_seq |
Smoothing parameter sequence. |
kern |
Weight function, default = "tcub". Other choices are "rect", "trwt", "tria", "epan", "bisq" and "gauss". Choices may be restricted when derivatives are required; e.g. for confidence bands and some bandwidth selectors. |
deg |
Degree of polynomial to use. Default: 2. |
plot of gcv value
df <- mumps_uk_1986_1987 plot_gcv( df$age, df$pos, df$tot, nn_seq = seq(0.2, 0.8, by=0.1), h_seq = seq(5, 25, by=1) )df <- mumps_uk_1986_1987 plot_gcv( df$age, df$pos, df$tot, nn_seq = seq(0.2, 0.8, by=0.1), h_seq = seq(5, 25, by=1) )
plot() overloading for result of estimate_from_mixture
## S3 method for class 'estimate_from_mixture' plot(x, ...)## S3 method for class 'estimate_from_mixture' plot(x, ...)
x |
the mixture_model |
... |
arbitrary params. |
ggplot object
plot() overloading for Farrington model
## S3 method for class 'farrington_model' plot(x, ...)## S3 method for class 'farrington_model' plot(x, ...)
x |
the Farrington model object. |
... |
arbitrary params. |
ggplot object
plot() overloading for fractional polynomial model
## S3 method for class 'fp_model' plot(x, ...)## S3 method for class 'fp_model' plot(x, ...)
x |
the fractional polynomial model object. |
... |
arbitrary params. |
ggplot object
plot() overloading for hierarchical_bayesian_model
## S3 method for class 'hierarchical_bayesian_model' plot(x, ...)## S3 method for class 'hierarchical_bayesian_model' plot(x, ...)
x |
hierarchical_bayesian_model object created by serosv. |
... |
arbitrary params. |
ggplot object
plot() overloading for local polynomial model
## S3 method for class 'lp_model' plot(x, ...)## S3 method for class 'lp_model' plot(x, ...)
x |
the local polynomial model object. |
... |
arbitrary params. |
ggplot object
plot() overloading for mixture model
## S3 method for class 'mixture_model' plot(x, ...)## S3 method for class 'mixture_model' plot(x, ...)
x |
the mixture_model |
... |
arbitrary params. |
ggplot object
plot() overloading for MSEIR model
## S3 method for class 'mseir_model' plot(x, ...)## S3 method for class 'mseir_model' plot(x, ...)
x |
the mseir_model object. |
... |
arbitrary params. |
ggplot object
plot() overloading for penalized spline
## S3 method for class 'penalized_spline_model' plot(x, ...)## S3 method for class 'penalized_spline_model' plot(x, ...)
x |
the penalized_spline_model object |
... |
arbitrary params. |
ggplot object
plot() overloading for polynomial model
## S3 method for class 'polynomial_model' plot(x, ...)## S3 method for class 'polynomial_model' plot(x, ...)
x |
the polynomial model object |
... |
arbitrary params. |
ggplot object
plot() overloading for SIR model
## S3 method for class 'sir_basic_model' plot(x, ...)## S3 method for class 'sir_basic_model' plot(x, ...)
x |
the sir_basic_model object. |
... |
arbitrary params. |
ggplot object
plot() overloading for SIR static model
## S3 method for class 'sir_static_model' plot(x, ...)## S3 method for class 'sir_static_model' plot(x, ...)
x |
the sir_static_model object. |
... |
arbitrary params. |
ggplot object
plot() overloading for SIR sub populations model
## S3 method for class 'sir_subpops_model' plot(x, ...)## S3 method for class 'sir_subpops_model' plot(x, ...)
x |
the sir_subpops_models object. |
... |
arbitrary params. |
list of ggplot objects, each object is the plot for the corresponding subpopulation
plot() overloading for Weibull model
## S3 method for class 'weibull_model' plot(x, ...)## S3 method for class 'weibull_model' plot(x, ...)
x |
the Weibull model object. |
... |
arbitrary params. |
ggplot object
Either 'status' (treated as line-listing dataset) or 'pos' & 'tot' (treated as aggregated dataset) must be provided
polynomial_model( age, k, type, pos = NULL, tot = NULL, status = NULL, link = "log" )polynomial_model( age, k, type, pos = NULL, tot = NULL, status = NULL, link = "log" )
age |
the age vector. |
k |
degree of the model. |
type |
name of method (Muench, Giffith, Grenfell). |
pos |
the positive count vector (optional if status is provided). |
tot |
the total count vector (optional if status is provided). |
status |
the serostatus vector (optional if pos & tot are provided). |
link |
link function. |
Refers to section 6.1.1
a list of class polynomial_model with 5 items
datatype |
type of datatype used for model fitting (aggregated or linelisting) |
df |
the dataframe used for fitting the model |
info |
fitted "glm" object |
sp |
seroprevalence |
foi |
force of infection |
data <- parvob19_fi_1997_1998[order(parvob19_fi_1997_1998$age), ] aggregated <- transform_data(data$age, data$seropositive) # fit with aggregated data model <- polynomial_model(aggregated$t, pos = aggregated$pos, tot = aggregated$tot, type = "Muench") # fit with linelisting data model <- polynomial_model(data$age, status = data$seropositive, type = "Muench") plot(model)data <- parvob19_fi_1997_1998[order(parvob19_fi_1997_1998$age), ] aggregated <- transform_data(data$age, data$seropositive) # fit with aggregated data model <- polynomial_model(aggregated$t, pos = aggregated$pos, tot = aggregated$tot, type = "Muench") # fit with linelisting data model <- polynomial_model(data$age, status = data$seropositive, type = "Muench") plot(model)
Rubella - Mumps data from the UK (aggregated)
rubella_mumps_ukrubella_mumps_uk
A data frame with 5 variables:
Age group
Number of individuals negative to rubella and mumps
Number of individuals negative to rubella and positive to mumps
Number of individuals positive to rubella and negative to mumps
Number of individuals positive to rubella and mumps
Morgan-Capner P, Wright J, Miller C L, Miller E. Surveillance of antibody to measles, mumps, and rubella by age. British Medical Journal 1988; 297 :770 doi:10.1136/bmj.297.6651.770
Prevalence of rubella in the UK, obtained from a large survey of prevalence of antibodies to both mumps and rubella viruses.
rubella_uk_1986_1987rubella_uk_1986_1987
A data frame with 3 variables:
Age group
Number of seropositive individuals
Total number of individuals surveyed
Morgan-Capner P, Wright J, Miller C L, Miller E. Surveillance of antibody to measles, mumps, and rubella by age. British Medical Journal 1988; 297 :770 doi:10.1136/bmj.297.6651.770
# Reproduce Fig 4.4 (middle panel), p. 67 age <- rubella_uk_1986_1987$age pos <- rubella_uk_1986_1987$pos tot <- rubella_uk_1986_1987$tot plot(age, pos / tot, cex = 0.008 * tot, pch = 16, xlab = "age", ylab = "seroprevalence", xlim = c(0, 45), ylim = c(0, 1) )# Reproduce Fig 4.4 (middle panel), p. 67 age <- rubella_uk_1986_1987$age pos <- rubella_uk_1986_1987$pos tot <- rubella_uk_1986_1987$tot plot(age, pos / tot, cex = 0.008 * tot, pch = 16, xlab = "age", ylab = "seroprevalence", xlim = c(0, 45), ylim = c(0, 1) )
Helper to adjust styling of a plot
set_plot_style( sero = "blueviolet", ci = "royalblue1", foi = "#fc0328", sero_line = "solid", foi_line = "dashed", xlabel = "Age" )set_plot_style( sero = "blueviolet", ci = "royalblue1", foi = "#fc0328", sero_line = "solid", foi_line = "dashed", xlabel = "Age" )
sero |
- color for seroprevalence line |
ci |
- color for confidence interval |
foi |
- color for force of infection line |
sero_line |
- linetype for seroprevalence line |
foi_line |
- linetype for force of infection line |
xlabel |
- x label |
list of updated aesthetic values
Refers to section 3.1.3.
sir_basic_model(times, state, parameters)sir_basic_model(times, state, parameters)
times |
time sequence. |
state |
the initial state of the model. |
parameters |
the parameters of the model. |
In state:
- S: number of susceptible
- I: number of infected
- R: number of recovered
In parameters:
- alpha: disease-related death rate
- mu: natural death rate (= 1/life expectancy)
- beta: transmission rate
- nu: recovery rate
- p: percent of population vaccinated at birth
list of class sir_basic_model with the following items
parameters |
list of parameters used for fitting the model |
output |
matrix of population for each compartment over time |
state <- c(S=4999, I=1, R=0) parameters <- c( mu=1/75, # 1 divided by life expectancy (75 years old) alpha=0, # no disease-related death beta=0.0005, # transmission rate nu=1, # 1 year for infected to recover p=0 # no vaccination at birth ) times <- seq(0, 250, by=0.1) model <- sir_basic_model(times, state, parameters) modelstate <- c(S=4999, I=1, R=0) parameters <- c( mu=1/75, # 1 divided by life expectancy (75 years old) alpha=0, # no disease-related death beta=0.0005, # transmission rate nu=1, # 1 year for infected to recover p=0 # no vaccination at birth ) times <- seq(0, 250, by=0.1) model <- sir_basic_model(times, state, parameters) model
Refers to section 3.2.2.
sir_static_model(a, state, parameters)sir_static_model(a, state, parameters)
a |
age sequence. |
state |
the initial state of the system. |
parameters |
the model's parameter. |
In state:
- s: proportion susceptible
- i: proportion infected
- r: proportion recovered
In parameters:
- lambda: natural death rate
- nu: recovery rate
list of class sir_static_model with the following items
parameters |
list of parameters used for fitting the model |
output |
matrix of proportion for each compartment over time |
state <- c(s=0.99,i=0.01,r=0) parameters <- c( lambda = 0.05, nu=1/(14/365) # 2 weeks to recover ) ages<-seq(0, 90, by=0.01) model = sir_static_model(ages, state, parameters) modelstate <- c(s=0.99,i=0.01,r=0) parameters <- c( lambda = 0.05, nu=1/(14/365) # 2 weeks to recover ) ages<-seq(0, 90, by=0.01) model = sir_static_model(ages, state, parameters) model
Refers to section 3.5.1.
sir_subpops_model(times, state, parameters)sir_subpops_model(times, state, parameters)
times |
time sequence. |
state |
the initial state of the model. |
parameters |
the parameters of the model. |
In state:
- s: Percent susceptible
- i: Percent infected
- r: Percent recovered
In parameters:
- mu: natural death rate (1/L).
- beta: transmission rate w.r.t population (beta tilde)
- nu: recovery rate
- k: number of subpopulations
list of class sir_subpops_model with the following items
parameters |
list of parameters used for fitting the model |
output |
matrix of proportion for each compartment over time |
k <- 2 state <- c( s = c(0.8, 0.8), i = c(0.2, 0.2), r = c( 0, 0) ) beta_matrix <- c( c(0.05, 0.00), c(0.00, 0.05) ) parameters <- list( beta = matrix(beta_matrix, nrow=k, ncol=k, byrow=TRUE), nu = c(1/30, 1/30), mu = 0.001, k = k ) times<-seq(0,10000,by=0.5) model <- sir_subpops_model(times, state, parameters) modelk <- 2 state <- c( s = c(0.8, 0.8), i = c(0.2, 0.2), r = c( 0, 0) ) beta_matrix <- c( c(0.05, 0.00), c(0.00, 0.05) ) parameters <- list( beta = matrix(beta_matrix, nrow=k, ncol=k, byrow=TRUE), nu = c(1/30, 1/30), mu = 0.001, k = k ) times<-seq(0,10000,by=0.5) model <- sir_subpops_model(times, state, parameters) model
A study of tuberculosis conducted in the Netherlands. Schoolchildren, aged between 6 and 18 years, were tested using the tuberculin skin test.
tb_nl_1966_1973tb_nl_1966_1973
A data frame with 5 variables:
Age group
Number of seropositive individuals
Total number of individuals surveyed
Gender of cohort (unsure what 0 and 1 means)
Birth year of cohort
Nagelkerke, N., Heisterkamp, S., Borgdorff, M., Broekmans, J. and Van Houwelingen, H. (1999), Semi-parametric estimation of age-time specific infection incidence from serial prevalence data. Statist. Med., 18: 307-320. doi:10.1002/(SICI)1097-0258(19990215)18:3<307::AID-SIM15>3.0.CO;2-Z
# Reproduce Fig 4.6, p.70 age <- tb_nl_1966_1973$age birthyr <- tb_nl_1966_1973$birthyr pos <- tb_nl_1966_1973$pos tot <- tb_nl_1966_1973$tot # left panel plot(age, pos / tot, pch = 16, cex = 0.00005 * tot, xlab = "age", ylab = "prevalence", xlim = c(6, 18) ) # right panel plot(birthyr, pos / tot, pch = 16, cex = 0.00005 * tot, xlab = "year", ylab = "prevalence" )# Reproduce Fig 4.6, p.70 age <- tb_nl_1966_1973$age birthyr <- tb_nl_1966_1973$birthyr pos <- tb_nl_1966_1973$pos tot <- tb_nl_1966_1973$tot # left panel plot(age, pos / tot, pch = 16, cex = 0.00005 * tot, xlab = "age", ylab = "prevalence", xlim = c(6, 18) ) # right panel plot(birthyr, pos / tot, pch = 16, cex = 0.00005 * tot, xlab = "year", ylab = "prevalence" )
Generate a dataframe with 't', 'pos' and 'tot' columns from 't' and 'seropositive' vectors.
transform_data(t, spos)transform_data(t, spos)
t |
the time vector. |
spos |
the seropositive vector. |
dataframe in aggregated format
df <- hcv_be_2006 hcv_df <- transform_data(df$dur, df$seropositive) hcv_dfdf <- hcv_be_2006 hcv_df <- transform_data(df$dur, df$seropositive) hcv_df
Age-specific seroprevalence of VZV antibodies, assessed in Flanders (Belgium) between October 1999 and April 2000. This population was stratified by age in order to obtain approximately 100 observations per age group.
vzv_be_1999_2000vzv_be_1999_2000
A data frame with 3 variables:
Age group
Number of seropositive individuals
Total number of individuals surveyed
Thiry, N., Beutels, P., Shkedy, Z. et al. The seroepidemiology of primary varicella-zoster virus infection in Flanders (Belgium). Eur J Pediatr 161, 588-593 (2002). doi:10.1007/s00431-002-1053-2
# Reproduce Fig 4.7 (left panel), p.71 age <- vzv_be_1999_2000$age pos <- vzv_be_1999_2000$pos tot <- vzv_be_1999_2000$tot plot(age, pos / tot, cex = 0.036 * tot, pch = 19, xlab = "age", ylab = "seroprevalence", xlim = c(0, 45), ylim = c(0, 1) )# Reproduce Fig 4.7 (left panel), p.71 age <- vzv_be_1999_2000$age pos <- vzv_be_1999_2000$pos tot <- vzv_be_1999_2000$tot plot(age, pos / tot, cex = 0.036 * tot, pch = 19, xlab = "age", ylab = "seroprevalence", xlim = c(0, 45), ylim = c(0, 1) )
The survey is the same as the one used to study the seroprevalence of parvovirus B19 in Belgium, as described above.
vzv_be_2001_2003vzv_be_2001_2003
A data frame with 4 variables:
Age of individual
If the individual is seropositive or not
Gender of individual
VZV milli international units per ml
MOSSONG, J., N. HENS, V. FRIEDERICHS, I. DAVIDKIN, M. BROMAN, B. LITWINSKA, J. SIENNICKA, et al. "Parvovirus B19 Infection in Five European Countries: Seroepidemiology, Force of Infection and Maternal Risk of Infection." Epidemiology and Infection 136, no. 8 (2008): 1059-68. doi:10.1017/S0950268807009661
# Reproduce Fig 4.7 (right panel), p.71 library(dplyr) df <- vzv_be_2001_2003 %>% mutate(age = round(age)) %>% group_by(age) %>% summarise(pos = sum(seropositive), tot = n()) plot(df$age, df$pos / df$tot, cex = 0.036 * df$tot, pch = 19, xlab = "age", ylab = "seroprevalence", xlim = c(0, 45), ylim = c(0, 1) )# Reproduce Fig 4.7 (right panel), p.71 library(dplyr) df <- vzv_be_2001_2003 %>% mutate(age = round(age)) %>% group_by(age) %>% summarise(pos = sum(seropositive), tot = n()) plot(df$age, df$pos / df$tot, cex = 0.036 * df$tot, pch = 19, xlab = "age", ylab = "seroprevalence", xlim = c(0, 45), ylim = c(0, 1) )
VZV and Parvovirus B19 serological data in Belgium (line listing)
vzv_parvo_bevzv_parvo_be
A data frame with 7 variables:
ID of individual
Age of individual
Gender of individual
Parvo B19 antibody units per ml
If an individual is positive for parvovirus B19
VZV milli international units per ml
If an individual is positive for VZV
MOSSONG, J., N. HENS, V. FRIEDERICHS, I. DAVIDKIN, M. BROMAN, B. LITWINSKA, J. SIENNICKA, et al. "Parvovirus B19 Infection in Five European Countries: Seroepidemiology, Force of Infection and Maternal Risk of Infection." Epidemiology and Infection 136, no. 8 (2008): 1059-68. doi:10.1017/S0950268807009661
Refers to section 6.1.2.
weibull_model(t, status = NULL, pos = NULL, tot = NULL)weibull_model(t, status = NULL, pos = NULL, tot = NULL)
t |
the time vector. |
status |
the serostatus vector (optional if pos & tot are provided). |
pos |
the positive count vector (optional if status is provided). |
tot |
the total count vector (optional if status is provided). |
list of class weibull_model with the following items
datatype |
type of datatype used for model fitting (aggregated or linelisting) |
df |
the dataframe used for fitting the model |
info |
fitted "glm" object |
sp |
seroprevalence |
foi |
force of infection |
[stats::glm()] for more information on the fitted "glm" object
df <- hcv_be_2006[order(hcv_be_2006$dur), ] model <- weibull_model( t=df$dur, status=df$seropositive ) plot(model)df <- hcv_be_2006[order(hcv_be_2006$dur), ] model <- weibull_model( t=df$dur, status=df$seropositive ) plot(model)