Variation in Specificity of HIV Rapid Diagnostic Tests over Place and Time: An Analysis of Discordancy Data Using a Bayesian Approach

Background

Recent trends to earlier access to anti-retroviral treatment underline the importance of accurate HIV diagnosis. The WHO HIV testing strategy recommends the use of two or three rapid diagnostic tests (RDTs) combined in an algorithm and assume a population is serologically stable over time. Yet RDTs are prone to cross reactivity which can lead to false positive or discordant results. This paper uses discordancy data from Médecins Sans Frontières (MSF) programmes to test the hypothesis that the specificity of RDTs change over place and time.

Methods

Data was drawn from all MSF test centres in 2007-8 using a parallel testing algorithm. A Bayesian approach was used to derive estimates of disease prevalence, and of test sensitivity and specificity using the software WinBUGS. A comparison of models with different levels of complexity was performed to assess the evidence for changes in test characteristics by location and over time.

Results

106, 035 individuals were included from 51 centres in 10 countries using 7 different RDTs. Discordancy patterns were found to vary by location and time. Model fit statistics confirmed this, with improved fit to the data when test specificity and sensitivity were allowed to vary by centre and over time. Two examples show evidence of variation in specificity between different testing locations within a single country. Finally, within a single test centre, variation in specificity was seen over time with one test becoming more specific and the other less specific.

Conclusion

This analysis demonstrates the variable specificity of multiple HIV RDTs over geographic location and time. This variability suggests that cross reactivity is occurring and indicates a higher than previously appreciated risk of false positive HIV results using the current WHO testing guidelines. Given the significant consequences of false HIV diagnosis, we suggest that current testing and evaluation strategies be reviewed.     

The mutant form of lamin A that causes Hutchinson-Gilford progeria is a biomarker of cellular aging in human skin

Hutchinson-Gilford progeria syndrome (HGPS, OMIM 176670) is a rare disorder characterized by accelerated aging and early death, frequently from stroke or coronary artery disease. 90% of HGPS cases carry the LMNA G608G (GGC>GGT) mutation within exon 11 of LMNA, activating a splice donor site that results in production of a dominant negative form of lamin A protein, denoted progerin. Screening 150 skin biopsies from unaffected individuals (newborn to 97 years) showed that a similar splicing event occurs in vivo at a low level in the skin at all ages. While progerin mRNA remains low, the protein accumulates in the skin with age in a subset of dermal fibroblasts and in a few terminally differentiated keratinocytes. Progerin-positive fibroblasts localize near the basement membrane and in the papillary dermis of young adult skin; however, their numbers increase and their distribution reaches the deep reticular dermis in elderly skin. Our findings demonstrate that progerin expression is a biomarker of normal cellular aging and may potentially be linked to terminal differentiation and senescence in elderly individuals.     

The Evaluation of a Rapid In Situ HIV Confirmation Test in a Programme with a High Failure Rate of the WHO HIV Two-Test Diagnostic Algorithm

Background

Concerns about false-positive HIV results led to a review of testing procedures used in a Médecins Sans Frontières (MSF) HIV programme in Bukavu, eastern Democratic Republic of Congo. In addition to the WHO HIV rapid diagnostic test algorithm (RDT) (two positive RDTs alone for HIV diagnosis) used in voluntary counselling and testing (VCT) sites we evaluated in situ a practical field-based confirmation test against western blot WB. In addition, we aimed to determine the false-positive rate of the WHO two-test algorithm compared with our adapted protocol including confirmation testing, and whether weakly reactive compared with strongly reactive rapid test results were more likely to be false positives.

Methodology/Principal Findings

2864 clients presenting to MSF VCT centres in Bukavu during January to May 2006 were tested using Determine HIV-1/2® and UniGold HIV® rapid tests in parallel by nurse counsellors. Plasma samples on 229 clients confirmed as double RDT positive by laboratory retesting were further tested using both WB and the Orgenics Immunocomb Combfirm® HIV confirmation test (OIC-HIV). Of these, 24 samples were negative or indeterminate by WB representing a false-positive rate of the WHO two-test algorithm of 10.5% (95%CI 6.6-15.2). 17 of the 229 samples were weakly positive on rapid testing and all were negative or indeterminate by WB. The false-positive rate fell to 3.3% (95%CI 1.3–6.7) when only strong-positive rapid test results were considered. Agreement between OIC-HIV and WB was 99.1% (95%CI 96.9–99.9%) with no false OIC-HIV positives if stringent criteria for positive OIC-HIV diagnoses were used.

Conclusions

The WHO HIV two-test diagnostic algorithm produced an unacceptably high level of false-positive diagnoses in our setting, especially if results were weakly positive. The most probable causes of the false-positive results were serological cross-reactivity or non-specific immune reactivity. Our findings show that the OIC-HIV confirmation test is practical and effective in field contexts. We propose that all double-positive HIV RDT samples should undergo further testing to confirm HIV seropositivity until the accuracy of the RDT testing algorithm has been established at programme level.     

A metapopulation model of tuberculosis transmission with a case study from high to low burden areas

Tuberculosis (TB) is a growing problem worldwide, especially with the emergence and high prevalence of multidrug-resistant strains. We develop a metapopulation model for TB spread, which is particularly suited to investigating transmission between areas of high and low prevalence. A case study of cross-border transmission in the Torres Strait region of Australia and Papua New Guinea (PNG) is considered and a sensitivity analysis is conducted. We find that only 6 of the 50 parameters analysed are important to the cumulative number of clinically active TB patients in the entire region. Of these, only the detection rate in PNG is found to be an important intervention parameter. We therefore give insight into the extent the area with the high burden of TB (PNG in the case study) is dominating the TB dynamics of the entire region. Furthermore, the sensitivity analysis results give insight into the data that most important to collect and refine, which is found to be data relating to the PNG parameters.     

Effects of Light Crude Oil Contamination on the Physical and Mechanical Properties of Fine Sand

The use of oil-contaminated sand in building and construction is now being considered as an alternative and cost-effective way to minimize its adverse effect on the environment. To achieve this, the effect of oil contamination on the important mechanical properties of sand should be investigated first. This study investigated the effect of petroleum-derived contaminants on the water absorption, permeability, cohesion, friction angle, and shear strength of fine sand. Contaminated samples were prepared by mixing fine sand with different percentages of light crude oil (0 to 20%). The results indicated that the water absorption of fine sand decreases with an increase in crude oil. An increase in the cohesion was observed for sand with up to 1% of oil contamination, after which the cohesion began to decrease, which also results in the reduction in the permeability. A slight reduction in the friction angle was found for oil-contaminated fine sand. At a low normal stress of 50 kPa, as the percentage of light crude oil increased, the shear strength increased up to 1% of oil contamination and then it decreased. These results provided useful information on how oil-contaminated sand can be used safely and effectively in building and construction.     

Cervical Abnormalities Are More Common among Indigenous than Other Australian Women: A Retrospective Record-Linkage Study, 2000–2011

Indigenous Australian women have much higher incidence of cervical cancer compared to non-Indigenous women. Despite an organised cervical screening program introduced 25 years ago, a paucity of Indigenous-identified data in Pap Smear Registers remains. Prevalence of cervical abnormalities detected among the screened Indigenous population has not previously been reported. We conducted a retrospective cohort study of population-based linked health records for 1,334,795 female Queensland residents aged 20–69 years who had one or more Pap smears during 2000–2011; from linked hospital records 23,483 were identified as Indigenous. Prevalence was calculated separately for Indigenous and non-Indigenous women, for cytology-detected low-grade (cLGA) and high-grade abnormalities (cHGA), and histologically confirmed high-grade abnormalities (hHGA). Odds ratios (OR) were estimated from logistic regression analysis. In 2010–2011 the prevalence of hHGA among Indigenous women (16.6 per 1000 women screened, 95% confidence interval [CI] 14.6–18.9) was twice that of non-Indigenous women (7.5 per 1000 women screened, CI 7.3–7.7). Adjusted for age, area-level disadvantage and place of residence, Indigenous women had higher prevalence of cLGA (OR 1.4, CI 1.3–1.4), cHGA (OR 2.2, CI 2.1–2.3) and hHGA (OR 2.0, CI 1.9–2.1). Our findings show that Indigenous women recorded on the Pap Smear Register have much higher prevalence for cLGA, cHGA and hHGA compared to non-Indigenous women. The renewed cervical screening program, to be implemented in 2017, offers opportunities to reduce the burden of abnormalities and invasive cancer among Indigenous women and address long-standing data deficiencies.     

Avian influenza H5N1 transmission in households, Indonesia

Background

Disease transmission patterns are needed to inform public health interventions, but remain largely unknown for avian influenza H5N1 virus infections. A recent study on the 139 outbreaks detected in Indonesia between 2005 and 2009 found that the type of exposure to sources of H5N1 virus for both the index case and their household members impacted the risk of additional cases in the household. This study describes the disease transmission patterns in those outbreak households.

Methodology/Principal Findings

We compared cases (n = 177) and contacts (n = 496) in the 113 sporadic and 26 cluster outbreaks detected between July 2005 and July 2009 to estimate attack rates and disease intervals. We used final size household models to fit transmission parameters to data on household size, cases and blood-related household contacts to assess the relative contribution of zoonotic and human-to-human transmission of the virus, as well as the reproduction number for human virus transmission. The overall household attack rate was 18.3% and secondary attack rate was 5.5%. Secondary attack rate remained stable as household size increased. The mean interval between onset of subsequent cases in outbreaks was 5.6 days. The transmission model found that human transmission was very rare, with a reproduction number between 0.1 and 0.25, and the upper confidence bounds below 0.4. Transmission model fit was best when the denominator population was restricted to blood-related household contacts of index cases.

Conclusions/Significance

The study only found strong support for human transmission of the virus when a single large cluster was included in the transmission model. The reproduction number was well below the threshold for sustained transmission. This study provides baseline information on the transmission dynamics for the current zoonotic virus and can be used to detect and define signatures of a virus with increasing capacity for human-to-human transmission.     

Successful Control of Ebola Virus Disease: Analysis of Service Based Data from Rural Sierra Leone

IntroductionThe scale and geographical distribution of the current outbreak in West Africa raised doubts as to the effectiveness of established methods of control. Ebola Virus Disease (EVD) was first detected in Sierra Leone in May 2014 in Kailahun district. Despite high case numbers elsewhere in the country, transmission was eliminated in the district by December 2014. We describe interventions underpinning successful EVD control in Kailahun and implications for EVD control in other areas.MethodsInternal service data and published reports from response agencies were analysed to describe the structure and type of response activities, EVD case numbers and epidemic characteristics. This included daily national situation reports and District-level data and reports of the Sierra Leone Ministry of Health and Sanitation, and Médecins Sans Frontières (MSF) patient data and internal epidemiological reports. We used EVD case definitions provided by the World Health Organisation over the course of the outbreak. Characteristics assessed included level of response activities and epidemiological features such as reported exposure (funeral-related or not), time interval between onset of illness and admission to the EVD Management Centre (EMC), work-related exposures (health worker or not) and mortality. We compared these characteristics between two time periods—June to July (the early period of response), and August to December (when coverage and quality of response had improved). A stochastic model was used to predict case numbers per generation with different numbers of beds and a varying percentage of community cases detected.ResultsThere were 652 probable/confirmed EVD cases from June-December 2014 in Kailahun. An EMC providing patient care opened in June. By August 2014 an integrated detection, treatment, and prevention strategy was in place across the district catchment zone. From June-July to August-December 2014 surveillance and contact tracing staff increased from 1.0 to 8.8 per confirmed EVD case, EMC capacity increased from 32 to 100 beds, the number of burial teams doubled, and health promotion activities increased in coverage. These improvements in response were associated with the following changes between the same periods: the proportion of confirmed/probable cases admitted to the EMC increased from 35% to 83% (χ² p-value<0·001), the proportion of confirmed patients admitted to the EMC <3 days of symptom onset increased from 19% to 37% (χ² p-value <0·001), and reported funeral contact in those admitted decreased from 33% to 16% (χ² p-value <0·001). Mathematical modelling confirmed the importance of both patient management capacity and surveillance and contact tracing for EVD control.DiscussionOur findings demonstrate that control of EVD can be achieved using established interventions based on identification and appropriate management of those who are at risk of and develop EVD, including in the context of ongoing transmission in surrounding regions. Key attributes in achieving control were sufficient patient care capacity (including admission to specialist facilities of suspect and probable cases for assessment), integrated with adequate staffing and resourcing of community-based case detection and prevention activities. The response structure and coverage targets we present are of value in informing effective control in current and future EVD outbreaks.