Estimating Population Distributions When Some Data Are Below a Limit of Detection by Using a Reverse Kaplan-Meier Estimator

Gillespie, Brenda W.; Chen, Qixuan; Reichert, Heidi; Franzblau, Alfred; Hedgeman, Elizabeth; Lepkowski, James M.; Adriaens, Peter; Demond, Avery H.; Luksemburg, William; Garabrant, David H.

doi:10.1097/ede.0b013e3181ce9f08

Cited by 89 publications

(74 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…2d/2e & Supplemental Fig. 3c, a log-rank test was utilized, thus accounting for the variation in the lower limit of detection between samples similar to what has been done for other assays (37–39). For all Mann-Whitney comparisons, a two-tailed test was used.…”

Section: Methodsmentioning

confidence: 99%

HIV persistence in tissue macrophages of humanized myeloid-only mice during antiretroviral therapy

Honeycutt

Thayer

Baker

et al. 2017

Nat Med

244

224

View full text Add to dashboard Cite

Despite years of fully suppressive antiretroviral therapy (ART), HIV persists in the host and is never eradicated. One major barrier to eradication is that multiple different cell types are infected that may individually contribute to HIV persistence. Tissue macrophages are critical contributors to HIV disease (1–3); however, their specific role in HIV persistence during long-term suppressive ART has not been established (4–6). Using humanized myeloid-only mice (MoM), we demonstrate that HIV infection of tissue macrophages is rapidly suppressed by ART, as determined by a rapid drop in plasma viral load and a dramatic drop in the levels of cell-associated viral RNA and DNA. No virus rebound was observed in the plasma of 67% of the ART treated animals at seven weeks post-ART interruption, and no replication competent virus was rescued from the tissue macrophages obtained from these animals. In contrast, in a subset of animals (~33%), a significantly delayed viral rebound was observed that is consistent with the establishment of persistent infection in tissue macrophages. These observations represent the first direct evidence of HIV persistence in tissue macrophages in vivo.

show abstract

Section: Methodsmentioning

confidence: 99%

HIV persistence in tissue macrophages of humanized myeloid-only mice during antiretroviral therapy

Honeycutt

Thayer

Baker

et al. 2017

Nat Med

244

224

View full text Add to dashboard Cite

show abstract

“…23,24 A more appropriate method to compare distributional differences is the use of the reverse Kaplan-Meier estimator. 24 This estimator is a left-censored analog to the common Kaplan-Meier estimator. This is a nonparametric method that makes no distributional assumptions regarding the trace elements.…”

Section: Methodsmentioning

confidence: 99%

“…It also accounts for the non-detects and has been advocated as a preferred method over other more popular ad hoc methods based on “filling in” a value for the non-detect (e.g., fill in the MDL or ½ × MDL). 24 …”

Section: Methodsmentioning

confidence: 99%

“…Several of the percentiles ≤25th percentile could not be estimated precisely for any of the trace elements. Where this was the case, we followed the procedure recommended by others, 24 and report an upper bound based on the minimum trace element value observed (e.g., 0.02 was the minimum observed arsenic value in both Appalachian and Jefferson, and thus “<0.02” is the value reported for the 10th percentiles in the results. As is the case with the Kaplan-Meier estimator, the reported P values based on reverse Kaplan-Meier estimation reflect overall group differences in trace element distributions between the relevant groups.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Concentrations of Arsenic, Chromium, and Nickel in Toenail Samples From Appalachian Kentucky Residents

Johnson

Shelton

Hopenhayn

et al. 2011

J Environ Pathol Toxicol Oncol

View full text Add to dashboard Cite

Lung cancer rates in Appalachian Kentucky are almost twice national rates; colorectal cancer rates are also elevated. Although smoking prevalence is high, it does not explain all excess risk. The area is characterized by poverty, low educational attainment, and unemployment. Coal production is a major industry. Pyrite contaminants of coal contain established human carcinogens, arsenic (As), chromium (Cr), and nickel (Ni). We compared biological exposure to As, Cr, and Ni for adults living in Appalachian Kentucky with residents of Jefferson, a non-Appalachian, urban county. We further compared lung and colon cancer rates, demographics, and smoking prevalence across the study areas. Toenail clipping analysis measured As, Cr, and Ni for residents of 23 rural Appalachian Kentucky counties and for Jefferson County. Reverse Kaplan-Meier statistical methodology addressed left-censored data. Appalachian residents were exposed to higher concentrations of As, Cr, and Ni than Jefferson County residents. Lung cancer incidence and mortality rates in Appalachia are higher than Jefferson County and elsewhere in the state, as are colorectal mortality rates. Environmental factors may contribute to the increased concentration of trace elements measured in residents of the Appalachian region. Routes of human exposure need to be determined.

show abstract

“…The estimates provided by this method were comparable to the restricted ML method and its main advantage is its ease of implementation using Microsoft Excel Solver tool (Flynn, 2010). The Kaplan-Meier (K-M) method, which does not assume any distributional shape, estimates summary statistics by constructing a curve akin to an empirical CDF while accounting for censoring (Kaplan and Meier, 1958;Gillespie et al, 2010).…”

Section: Introductionmentioning

confidence: 99%

A Comparison of the β-Substitution Method and a Bayesian Method for Analyzing Left-Censored Data

Huynh

Quick

Ramachandran

et al. 2015

ANNHYG

View full text Add to dashboard Cite

Classical statistical methods for analyzing exposure data with values below the detection limits are well described in the occupational hygiene literature, but an evaluation of a Bayesian approach for handling such data is currently lacking. Here, we first describe a Bayesian framework for analyzing censored data. We then present the results of a simulation study conducted to compare the β-substitution method with a Bayesian method for exposure datasets drawn from lognormal distributions and mixed lognormal distributions with varying sample sizes, geometric standard deviations (GSDs), and censoring for single and multiple limits of detection. For each set of factors, estimates for the arithmetic mean (AM), geometric mean, GSD, and the 95th percentile (X 0.95 ) of the exposure distribution were obtained. We evaluated the performance of each method using relative bias, the root mean squared error (rMSE), and coverage (the proportion of the computed 95% uncertainty intervals containing the true value). The Bayesian method using non-informative priors and the β-substitution method were generally comparable in bias and rMSE when estimating the AM and GM. For the GSD and the 95th percentile, the Bayesian method with non-informative priors was more biased and had a higher rMSE than the β-substitution method, but use of more informative priors generally improved the Bayesian method's performance, making both the bias and the rMSE more comparable to the β-substitution method. An advantage of the Bayesian method is that it provided estimates of uncertainty for these parameters of interest and good coverage, whereas the β-substitution method only provided estimates of uncertainty for the AM, and coverage was not as consistent. Selection of one or the other method depends on the needs of the practitioner, the availability of prior information, and the distribution characteristics of the • 56Ann. Occup. Hyg., 2016, Vol. 60, No. 1, 56-73 doi:10.1093/annhyg/mev049 Advance Access publication 24 July 2015 measurement data. We suggest the use of Bayesian methods if the practitioner has the computational resources and prior information, as the method would generally provide accurate estimates and also provides the distributions of all of the parameters, which could be useful for making decisions in some applications.

show abstract

Estimating Population Distributions When Some Data Are Below a Limit of Detection by Using a Reverse Kaplan-Meier Estimator

Cited by 89 publications

References 18 publications

HIV persistence in tissue macrophages of humanized myeloid-only mice during antiretroviral therapy

HIV persistence in tissue macrophages of humanized myeloid-only mice during antiretroviral therapy

Concentrations of Arsenic, Chromium, and Nickel in Toenail Samples From Appalachian Kentucky Residents

A Comparison of the β-Substitution Method and a Bayesian Method for Analyzing Left-Censored Data

Contact Info

Product

Resources

About