Incorporation of hierarchical structure into estimation and projection package fitting with examples of estimating subnational HIV/AIDS dynamics

Xiaoyue Niu, Amy Zhang, Tim Brown, Robert Puckett, Mary Mahy, Le Bao

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

Objectives: The article aims to give Spectrum/estimation and projection package (EPP) users and the scientific community a basic understanding of the underlying statistical model used to incorporate hierarchical structure in HIV subnational estimation, and to show how it has been implemented in the Spectrum/EPP interface for improving subepidemic estimation. The article also provides recommended default settings for this new model. Methods: We apply a generalized linear mixed-effects model on antenatal clinics prevalence data to get area-specific prevalence and uncertainty estimates, and transform those estimates to auxiliary data. We then fit the EPP model to both the observed data and auxiliary data. Results: We apply the proposed methods to four countries with different levels of data availability. We compare the out-of-sample prediction accuracy of the proposed method with varying auxiliary sample sizes and EPP without auxiliary data. Conclusion: We find that borrowing information from data-rich areas to data-sparse areas using our proposed method improves EPP fit in data-sparse areas. We recommend using the sample size estimated from generalized linear mixed-effects model as the default auxiliary sample size.

Original languageEnglish (US)
Pages (from-to)S51-S59
JournalAIDS
Volume31
DOIs
StatePublished - Apr 1 2017

Fingerprint

Acquired Immunodeficiency Syndrome
HIV
Sample Size
Statistical Models
Uncertainty

All Science Journal Classification (ASJC) codes

  • Immunology and Allergy
  • Immunology
  • Infectious Diseases

Cite this

@article{5288153fcca24b239833d6a931f14587,
title = "Incorporation of hierarchical structure into estimation and projection package fitting with examples of estimating subnational HIV/AIDS dynamics",
abstract = "Objectives: The article aims to give Spectrum/estimation and projection package (EPP) users and the scientific community a basic understanding of the underlying statistical model used to incorporate hierarchical structure in HIV subnational estimation, and to show how it has been implemented in the Spectrum/EPP interface for improving subepidemic estimation. The article also provides recommended default settings for this new model. Methods: We apply a generalized linear mixed-effects model on antenatal clinics prevalence data to get area-specific prevalence and uncertainty estimates, and transform those estimates to auxiliary data. We then fit the EPP model to both the observed data and auxiliary data. Results: We apply the proposed methods to four countries with different levels of data availability. We compare the out-of-sample prediction accuracy of the proposed method with varying auxiliary sample sizes and EPP without auxiliary data. Conclusion: We find that borrowing information from data-rich areas to data-sparse areas using our proposed method improves EPP fit in data-sparse areas. We recommend using the sample size estimated from generalized linear mixed-effects model as the default auxiliary sample size.",
author = "Xiaoyue Niu and Amy Zhang and Tim Brown and Robert Puckett and Mary Mahy and Le Bao",
year = "2017",
month = "4",
day = "1",
doi = "10.1097/QAD.0000000000001426",
language = "English (US)",
volume = "31",
pages = "S51--S59",
journal = "AIDS",
issn = "0269-9370",
publisher = "Lippincott Williams and Wilkins",

}

Incorporation of hierarchical structure into estimation and projection package fitting with examples of estimating subnational HIV/AIDS dynamics. / Niu, Xiaoyue; Zhang, Amy; Brown, Tim; Puckett, Robert; Mahy, Mary; Bao, Le.

In: AIDS, Vol. 31, 01.04.2017, p. S51-S59.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Incorporation of hierarchical structure into estimation and projection package fitting with examples of estimating subnational HIV/AIDS dynamics

AU - Niu, Xiaoyue

AU - Zhang, Amy

AU - Brown, Tim

AU - Puckett, Robert

AU - Mahy, Mary

AU - Bao, Le

PY - 2017/4/1

Y1 - 2017/4/1

N2 - Objectives: The article aims to give Spectrum/estimation and projection package (EPP) users and the scientific community a basic understanding of the underlying statistical model used to incorporate hierarchical structure in HIV subnational estimation, and to show how it has been implemented in the Spectrum/EPP interface for improving subepidemic estimation. The article also provides recommended default settings for this new model. Methods: We apply a generalized linear mixed-effects model on antenatal clinics prevalence data to get area-specific prevalence and uncertainty estimates, and transform those estimates to auxiliary data. We then fit the EPP model to both the observed data and auxiliary data. Results: We apply the proposed methods to four countries with different levels of data availability. We compare the out-of-sample prediction accuracy of the proposed method with varying auxiliary sample sizes and EPP without auxiliary data. Conclusion: We find that borrowing information from data-rich areas to data-sparse areas using our proposed method improves EPP fit in data-sparse areas. We recommend using the sample size estimated from generalized linear mixed-effects model as the default auxiliary sample size.

AB - Objectives: The article aims to give Spectrum/estimation and projection package (EPP) users and the scientific community a basic understanding of the underlying statistical model used to incorporate hierarchical structure in HIV subnational estimation, and to show how it has been implemented in the Spectrum/EPP interface for improving subepidemic estimation. The article also provides recommended default settings for this new model. Methods: We apply a generalized linear mixed-effects model on antenatal clinics prevalence data to get area-specific prevalence and uncertainty estimates, and transform those estimates to auxiliary data. We then fit the EPP model to both the observed data and auxiliary data. Results: We apply the proposed methods to four countries with different levels of data availability. We compare the out-of-sample prediction accuracy of the proposed method with varying auxiliary sample sizes and EPP without auxiliary data. Conclusion: We find that borrowing information from data-rich areas to data-sparse areas using our proposed method improves EPP fit in data-sparse areas. We recommend using the sample size estimated from generalized linear mixed-effects model as the default auxiliary sample size.

UR - http://www.scopus.com/inward/record.url?scp=85016032457&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85016032457&partnerID=8YFLogxK

U2 - 10.1097/QAD.0000000000001426

DO - 10.1097/QAD.0000000000001426

M3 - Article

C2 - 28296800

AN - SCOPUS:85016032457

VL - 31

SP - S51-S59

JO - AIDS

JF - AIDS

SN - 0269-9370

ER -