P D Ghys1, E Kufa, M V George. 1. Joint United Nations Programme on AIDS (UNAIDS), Avenue Appia 20, Geneva CH-1211, Switzerland. ghysp@unaids.org
Abstract
OBJECTIVE: Review of recent data and practice to derive guidance on questions relating to the measurement and analysis of trends in HIV prevalence and incidence. RESULTS: HIV prevalence among pregnant women attending antenatal clinics (ANCs) remains the principal data source to inform trends in the epidemic. Other data sources are: less available, representative of a small section of the population (sex workers, occupational groups), subject to additional bias (for example, voluntary counselling and testing service statistics), or are not yet available for multiple years (national surveys). Validity of HIV prevalence results may change over time due to improvements in HIV tests per se and implementation of laboratory quality assurance systems. The newer laboratory tests for recent infections require further validation and development of methodology to derive estimates of HIV incidence. CONCLUSIONS: Issues to consider during statistical analyses of trends among ANC attendees are: inclusion of consistent sites only, use of confidence intervals, stratification by site when performing a statistical test for trend, the need for at least three observations in a surveillance system with data collection every one to two years, and sound judgement. Trends in HIV prevalence among pregnant 15-24 year olds attending ANCs can be used to approximate trends in incidence. Indepth small area research studies are useful to inform the interpretation of surveillance data and provide directly measured trends in prevalence and incidence. Modelling can assess changes over time in prevalence, incidence, and mortality at the same time. Modelling tools need to be further developed to allow incorporation of estimates of HIV incidence and mortality, as these data are likely to become available in the future. To increase their explanatory power, models should also be extended to incorporate programmatic inputs.
OBJECTIVE: Review of recent data and practice to derive guidance on questions relating to the measurement and analysis of trends in HIV prevalence and incidence. RESULTS: HIV prevalence among pregnant women attending antenatal clinics (ANCs) remains the principal data source to inform trends in the epidemic. Other data sources are: less available, representative of a small section of the population (sex workers, occupational groups), subject to additional bias (for example, voluntary counselling and testing service statistics), or are not yet available for multiple years (national surveys). Validity of HIV prevalence results may change over time due to improvements in HIV tests per se and implementation of laboratory quality assurance systems. The newer laboratory tests for recent infections require further validation and development of methodology to derive estimates of HIV incidence. CONCLUSIONS: Issues to consider during statistical analyses of trends among ANC attendees are: inclusion of consistent sites only, use of confidence intervals, stratification by site when performing a statistical test for trend, the need for at least three observations in a surveillance system with data collection every one to two years, and sound judgement. Trends in HIV prevalence among pregnant 15-24 year olds attending ANCs can be used to approximate trends in incidence. Indepth small area research studies are useful to inform the interpretation of surveillance data and provide directly measured trends in prevalence and incidence. Modelling can assess changes over time in prevalence, incidence, and mortality at the same time. Modelling tools need to be further developed to allow incorporation of estimates of HIV incidence and mortality, as these data are likely to become available in the future. To increase their explanatory power, models should also be extended to incorporate programmatic inputs.
Authors: Dilys Morgan; Cedric Mahe; Billy Mayanja; J Martin Okongo; Rosemary Lubega; James A G Whitworth Journal: AIDS Date: 2002-03-08 Impact factor: 4.177
Authors: J Steven McDougal; Christopher D Pilcher; Bharat S Parekh; Guy Gershy-Damet; Bernard M Branson; Kimberly Marsh; Stefan Z Wiktor Journal: AIDS Date: 2005-05 Impact factor: 4.177
Authors: Wolfgang Hladik; Kereng Masupu; Thierry Roels; Tanarak Plipat; Frank Kaharuza; Rebecca Bunnell; Nicole Seguy; Lawrence H Marum Journal: AIDS Date: 2005-05 Impact factor: 4.177
Authors: Fulgentius Baryarama; Rebecca E Bunnell; Raymond L Ransom; John Paul Ekwaru; Josephine Kalule; Esther B Tumuhairwe; Jonathan H Mermin Journal: J Acquir Immune Defic Syndr Date: 2004-09-01 Impact factor: 3.731
Authors: Simon Gregson; Constance A Nyamukapa; Geoffrey P Garnett; Peter R Mason; Tom Zhuwau; Michel Caraël; Stephen K Chandiwana; Roy M Anderson Journal: Lancet Date: 2002-06-01 Impact factor: 79.321
Authors: R S Janssen; G A Satten; S L Stramer; B D Rawal; T R O'Brien; B J Weiblen; F M Hecht; N Jack; F R Cleghorn; J O Kahn; M A Chesney; M P Busch Journal: JAMA Date: 1998-07-01 Impact factor: 56.272
Authors: Quarraisha Abdool Karim; Ayesha B M Kharsany; Janet A Frohlich; Lise Werner; May Mashego; Mukelisiwe Mlotshwa; Bernadette T Madlala; Fanelesibonge Ntombela; Salim S Abdool Karim Journal: Int J Epidemiol Date: 2010-11-03 Impact factor: 7.196
Authors: Yusufu Kumogola; Emma Slaymaker; Basia Zaba; Julius Mngara; Raphael Isingo; John Changalucha; Patrick Mwidunda; Daniel Kimaro; Mark Urassa Journal: BMC Public Health Date: 2010-09-13 Impact factor: 3.295
Authors: Na Wang; Guohui Wu; Rongrong Lu; Liangui Feng; Yan Xiao; Willi McFarland; Yuhua Ruan; Yiming Shao; H F Raymond Journal: AIDS Behav Date: 2016-12