| Title: | Outlier detection and influence diagnostics for meta-analysis |
|---|---|
| Description: | A R package for implementing outlier detection and influence diagnostics for meta-analysis. Bootstrap distributions of the influence statistics are calculated, and the thresholds to determine outliers are clearly provided. |
| Authors: | Hisashi Noma |
| Maintainer: | Hisashi Noma <[email protected]> |
| License: | GPL-3 |
| Version: | 1.1-1 |
| Built: | 2025-02-03 03:53:41 UTC |
| Source: | https://github.com/nomahi/boutliers |
A R package for implementing outlier detection and influence diagnostics for meta-analysis. Bootstrap distributions of the influence statistics are calculated, and the thresholds to determine outliers are clearly provided.
Hedges, L. V., and Olkins, I. (1985). Statistical Methods for Meta-Analysis. New York: Academic Press.
Noma, H., Gosho, M., Ishii, R., Oba, K., and Furukawa, T. A. (2020+). Outlier detection and influence diagnostics in network meta-analysis. Research Synthesis Methods. https://doi.org/10.1002/jrsm.1455
Viechtbauer, W., and Cheung, M. W. (2010). Outlier and influence diagnostics for meta-analysis. Research Synthesis Methods. 1(2): 112-125. https://doi.org/10.1002/jrsm.11
Converting binary outcome data to the effect size estimates and the within studies variances vector
convert_bin(m1, n1, m2, n2, type = c("logOR", "logRR", "RD"))convert_bin(m1, n1, m2, n2, type = c("logOR", "logRR", "RD"))
m1 |
A vector of the number of successes in experimental group |
n1 |
A vector of the number of patients in experimental group |
m2 |
A vector of the number of successes in contorol group |
n2 |
A vector of the number of patients in contorol group |
type |
the outcome measure for binary outcome data (default = "logOR").
|
This function implements methods for logarithmic odds ratio, logarithmic relative risk, and risk difference described in Hartung & Knapp (2001).
A data.frame of study data.
y: A numeric vector of the effect size estimates.
v: A numeric vector of the within studies variances.
Hartung, J., and Knapp, G. (2001). A refined method for the meta-analysis of controlled clinical trials with binary outcome. Stat Med. 20(24): 3875-3889. https://doi.org/10.1002/sim.1009
data(PPI) attach(PPI) dat1 <- convert_bin(d1, n1, d2, n2, type = "logOR") print(dat1) dat2 <- convert_bin(d1, n1, d2, n2, type = "logRR") print(dat2) dat3 <- convert_bin(d1, n1, d2, n2, type = "RD") print(dat3)data(PPI) attach(PPI) dat1 <- convert_bin(d1, n1, d2, n2, type = "logOR") print(dat1) dat2 <- convert_bin(d1, n1, d2, n2, type = "logRR") print(dat2) dat3 <- convert_bin(d1, n1, d2, n2, type = "RD") print(dat3)
Converting estimated means and standard deviations in experimental and contorol groups to the effect size estimates and the within studies variances vector
convert_mean(n1, m1, s1, n2, m2, s2, type=c("MD", "SMD"), pooled = FALSE)convert_mean(n1, m1, s1, n2, m2, s2, type=c("MD", "SMD"), pooled = FALSE)
n1 |
A vector of number of observations in experimental group |
m1 |
A vector of estimated mean in experimental group |
s1 |
A vector of standard deviation in experimental group |
n2 |
A vector of number of observations in experimental group |
m2 |
A vector of estimated mean in experimental group |
s2 |
A vector of standard deviation in experimental group |
type |
the outcome measure for continous outcome data (default = "MD").
|
pooled |
logical; if |
A data.frame of study data.
y: A numeric vector of the effect size estimates.
v: A numeric vector of the within studies variance estimates.
data(SMT) attach(SMT) dat1 <- convert_mean(n1, m1, s1, n2, m2, s2, type="MD") print(dat1) dat2 <- convert_mean(n1, m1, s1, n2, m2, s2, type="SMD") print(dat2)data(SMT) attach(SMT) dat1 <- convert_mean(n1, m1, s1, n2, m2, s2, type="MD") print(dat1) dat2 <- convert_mean(n1, m1, s1, n2, m2, s2, type="SMD") print(dat2)
center: Center ID
n1: Number of observations in finasteride group
m1: Mean of the change of Boyarsky score from baseline in finasteride group
s1: SD of the change of Boyarsky score from baseline in finasteride group
n0: Number of observations in placebo group
m0: Mean of the change of Boyarsky score from baseline in placebo group
s0: SD of the change of Boyarsky score from baseline in placebo group
data(PPI)data(PPI)
A data frame with 29 rows and 7 variables
Gormley, G. J., Stoner, E., Bruskewitz, R. C., et al. (1992). The effect of finasteride in men with benign prostatic hyperplasia. The Finasteride Study Group. New England Journal of Medicine. 327: 1185-1191. https://www.nejm.org/doi/full/10.1056/NEJM199210223271701
Gould, A. L. (1998). Multi-centre trial analysis revisited. Statistics in Medicine. 17: 1779-1797.
Implementing the likelihood ratio tests using the mean-shifted model for the DerSimonian-Laird-type random-effects model. The bootstrap p-values are provided.
LRT(y, v, B=2000, alpha=0.05)LRT(y, v, B=2000, alpha=0.05)
y |
A vector of the outcome measure estimates (e.g., MD, SMD, log OR, log RR, RD) |
v |
A vector of the variance estimate of |
B |
The number of bootstrap resampling (defalt: 2000) |
alpha |
The significance level (default: 0.05) |
Results of the likelihood ratio tests involving bootstrap p-values. The outputs are ordered by the p-values.
id: ID of the study.
LR: The likelihood ratio statistic for based on the mean-shifted model.
Q: 1-alphath percentile for the bootstrap distribution of the likelihood ratio statistic.
P: The bootstrap p-value for the likelihood ratio statistic.
data(SMT) attach(SMT) dat <- convert_mean(n1, m1, s1, n2, m2, s2, type="MD") LRT(dat$y, dat$v) detach(SMT)data(SMT) attach(SMT) dat <- convert_mean(n1, m1, s1, n2, m2, s2, type="MD") LRT(dat$y, dat$v) detach(SMT)
Implementing the likelihood ratio tests using the mean-shifted model for the fixed-effect model. The bootstrap p-values are provided.
LRT_FE(y, v, B=2000, alpha=0.05)LRT_FE(y, v, B=2000, alpha=0.05)
y |
A vector of the outcome measure estimates (e.g., MD, SMD, log OR, log RR, RD) |
v |
A vector of the variance estimate of |
B |
The number of bootstrap resampling (defalt: 2000) |
alpha |
The significance level (default: 0.05) |
Results of the likelihood ratio tests involving bootstrap p-values. The outputs are ordered by the p-values.
id: ID of the study.
LR: The likelihood ratio statistic for based on the mean-shifted model.
Q: 1-alphath percentile for the bootstrap distribution of the likelihood ratio statistic.
P: The bootstrap p-value for the likelihood ratio statistic.
data(SMT) attach(SMT) dat <- convert_mean(n1, m1, s1, n2, m2, s2, type="MD") LRT_FE(dat$y, dat$v) detach(SMT)data(SMT) attach(SMT) dat <- convert_mean(n1, m1, s1, n2, m2, s2, type="MD") LRT_FE(dat$y, dat$v) detach(SMT)
ID: Study ID
d1: Number of events in PPI intervention group
n1: Number of observations in PPI intervention group
d2: Number of events in non-PPI intervention group
n2: Number of observations in non-PPI intervention group
data(PPI)data(PPI)
A data frame with 21 rows and 5 variables
Crocker, J. C., Ricci-Cabello, I., Parker, A., Hirst, J. A., Chant, A., Petit-Zeman, S., Evans, D., Rees, S. (2018). Impact of patient and public involvement on enrolment and retention in clinical trials: systematic review and meta-analysis. BMJ. 363: k4738. https://doi.org/10.1136/bmj.k4738
ID: Study ID
Souce: First author name and year of publication
m1: Estimated mean in experimental group
s1: Standard deviation in experimental group
n1: Number of observations in experimental group
m2: Estimated mean in control group
s2: Standard deviation in control group
n2: Number of observations in control group
data(SMT)data(SMT)
A data frame with 23 rows and 8 variables
Rubinstein, S. M,, de Zoete, A., van Middelkoop, M., Assendelft, W. J. J., de Boer, M. R., van Tulder, M. W. (2019). Benefits and harms of spinal manipulative therapy for the treatment of chronic low back pain: systematic review and meta-analysis of randomised controlled trials. BMJ. 364: l689. https://doi.org/10.1136/bmj.l689
Calculating the studentized residuals by leave-one-out analysis (studentized deleted residuals) and the percentiles of their bootstrap distributions.
STR(y, v, B=2000, alpha=0.95)STR(y, v, B=2000, alpha=0.95)
y |
A vector of the outcome measure estimates (e.g., MD, SMD, log OR, log RR, RD) |
v |
A vector of the variance estimate of |
B |
The number of bootstrap resampling (defalt: 2000) |
alpha |
The bootstrap percentiles to be outputted; 0.5(1-alpha)th and (1-0.5(1-alpha))th pecentiles. Default is 0.95; 2.5th and 97.5th percentiles are calculated. |
The studentized residuals by leave-one-out analysis. The outputs are ordered by the sizes of the studentized residuals.
id: ID of the study.
psi: The studentized residuals by leave-one-out analysis (studentized deleted residuals).
Q1: 0.5(1-alpha)th percentile for the bootstrap distribution of the studentized residual (default: 2.5th percentile).
Q2: 1-0.5(1-alpha)th percentile for the bootstrap distribution of the studentized residual (default: 97.5th percentile).
data(PPI) attach(PPI) dat <- convert_bin(d1, n1, d2, n2, type = "logOR") print(dat) STR(dat$y, dat$v) detach(PPI)data(PPI) attach(PPI) dat <- convert_bin(d1, n1, d2, n2, type = "logOR") print(dat) STR(dat$y, dat$v) detach(PPI)
Calculating the studentized residuals by leave-one-out analysis (studentized deleted residuals) for the fixed-effect model and the percentiles of their bootstrap distributions.
STR_FE(y, v, B=2000, alpha=0.95)STR_FE(y, v, B=2000, alpha=0.95)
y |
A vector of the outcome measure estimates (e.g., MD, SMD, log OR, log RR, RD) |
v |
A vector of the variance estimate of |
B |
The number of bootstrap resampling (defalt: 2000) |
alpha |
The bootstrap percentiles to be outputted; 0.5(1-alpha)th and (1-0.5(1-alpha))th pecentiles. Default is 0.95; 2.5th and 97.5th percentiles are calculated. |
The studentized residuals by leave-one-out analysis. The outputs are ordered by the sizes of the studentized residuals.
id: ID of the study.
psi: The studentized residuals by leave-one-out analysis (studentized deleted residuals).
Q1: 0.5(1-alpha)th percentile for the bootstrap distribution of the studentized residual (default: 2.5th percentile).
Q2: 1-0.5(1-alpha)th percentile for the bootstrap distribution of the studentized residual (default: 97.5th percentile).
data(PPI) attach(PPI) dat <- convert_bin(d1, n1, d2, n2, type = "logOR") print(dat) STR_FE(dat$y, dat$v) detach(PPI)data(PPI) attach(PPI) dat <- convert_bin(d1, n1, d2, n2, type = "logOR") print(dat) STR_FE(dat$y, dat$v) detach(PPI)
Calculating the variance ratio influential statistics by leave-one-out analysis and the percentiles of their bootstrap distributions.
VRATIO(y, v, B=2000, alpha=0.05)VRATIO(y, v, B=2000, alpha=0.05)
y |
A vector of the outcome measure estimates (e.g., MD, SMD, log OR, log RR, RD) |
v |
A vector of the variance estimate of |
B |
The number of bootstrap resampling (defalt: 2000) |
alpha |
The bootstrap percentile to be outputted (default: 0.05) |
The variance ratio influential statistics by leave-one-out analysis and their bootstrap percentiles. The outputs are ordered by the sizes of the variance ratio statistics.
id: ID of the study.
VR: The VRATIO statistic (relative change of the variance of the overall estimator) by leave-one-out analysis.
Q1: alphath percentile for the bootstrap distribution of the VRATIO statistic.
TR: The TAU2RATIO statistic (relative change of the heterogeneity variance) by leave-one-out analysis.
Q2: alphath percentile for the bootstrap distribution of the TAU2RATIO statistic.
data(finasteride) attach(finasteride) dat <- convert_mean(n1, m1, s1, n0, m0, s0, type="MD") print(dat) VRATIO(dat$y, dat$v) detach(finasteride)data(finasteride) attach(finasteride) dat <- convert_mean(n1, m1, s1, n0, m0, s0, type="MD") print(dat) VRATIO(dat$y, dat$v) detach(finasteride)