Dataset Information

On a Shape-Invariant Hazard Regression Model with application to an HIV Prevention Study of Mother-to-Child Transmission.

ABSTRACT: In survival analysis, Cox model is widely used for most clinical trial data. Alternatives include the additive hazard model, the accelerated failure time (AFT) model and a more general transformation model. All these models assume that the effects for all covariates are on the same scale. However, it is possible that for different covariates, the effects are on different scales. In this paper, we propose a shape-invariant hazard regression model that allows us to estimate the multiplicative treatment effect with adjustment of covariates that have non-multiplicative effects. We propose moment-based inference procedures for the regression parameters. We also discuss the risk prediction and the goodness of fit test for our proposed model. Numerical studies show good finite sample performance of our proposed estimator. We applied our method to the HIVNET 012 study, a milestone trial of single-dose nevirapine in prevention of mother-to-child transmission of HIV. From the HIVNET 012 data analysis, single-dose nevirapine treatment is shown to improve 18-month infant survival significantly with appropriate adjustment of the maternal CD4 counts and the virus load.

SUBMITTER: Zheng C

PROVIDER: S-EPMC7730855 | biostudies-literature | 2020 Dec

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

On a Shape-Invariant Hazard Regression Model with application to an HIV Prevention Study of Mother-to-Child Transmission.

Zheng Cheng C Chen Ying Qing YQ

Statistics in biosciences 20191019 3

In survival analysis, Cox model is widely used for most clinical trial data. Alternatives include the additive hazard model, the accelerated failure time (AFT) model and a more general transformation model. All these models assume that the effects for all covariates are on the same scale. However, it is possible that for different covariates, the effects are on different scales. In this paper, we propose a shape-invariant hazard regression model that allows us to estimate the multiplicative trea ...[more]

PMID: 33312265

Similar Datasets

Project description:BackgroundWe performed a systematic review to assess the effect of integrated perinatal prevention of mother-to-child transmission of HIV interventions compared to non- or partially integrated services on the uptake in low- and middle-income countries.MethodsWe searched for experimental, quasi-experimental and controlled observational studies in any language from 21 databases and grey literature sources.ResultsOut of 28 654 citations retrieved, five studies met our inclusion criteria. A cluster randomized controlled trial reported higher probability of nevirapine uptake at the labor wards implementing HIV testing and structured nevirapine adherence assessment (RRR 1.37, bootstrapped 95% CI, 1.04-1.77). A stepped wedge design study showed marked improvement in antiretroviral therapy (ART) enrolment (44.4% versus 25.3%, p<0.001) and initiation (32.9% versus 14.4%, p<0.001) in integrated care, but the median gestational age of ART initiation (27.1 versus 27.7 weeks, p = 0.4), ART duration (10.8 versus 10.0 weeks, p = 0.3) or 90 days ART retention (87.8% versus 91.3%, p = 0.3) did not differ significantly. A cohort study reported no significant difference either in the ART coverage (55% versus 48% versus 47%, p = 0.29) or eight weeks of ART duration before the delivery (50% versus 42% versus 52%; p = 0.96) between integrated, proximal and distal partially integrated care. Two before and after studies assessed the impact of integration on HIV testing uptake in antenatal care. The first study reported that significantly more women received information on PMTCT (92% versus 77%, p<0.001), were tested (76% versus 62%, p<0.001) and learned their HIV status (66% versus 55%, p<0.001) after integration. The second study also reported significant increase in HIV testing uptake after integration (98.8% versus 52.6%, p<0.001).ConclusionLimited, non-generalizable evidence supports the effectiveness of integrated PMTCT programs. More research measuring coverage and other relevant outcomes is urgently needed to inform the design of services delivering PMTCT programs.

Project description:BackgroundPrevention of Mother-to-Child HIV Transmission (PMTCT) coverage has been low in Ethiopia and the service has been implemented in a fragmented manner. Solutions to this problem have mainly been sought on the supply-side in the form of improved management and allocation of limited resources. However, this approach largely ignores the demand-side factors associated with low PMTCT coverage in the country. The study assesses the factors associated with the utilization of PMTCT services taking into consideration counts of visits to antenatal care (ANC) services in urban high-HIV prevalence and rural low-HIV prevalence settings in Ethiopia.MethodsA multivariate regression model was employed to identify significant factors associated with PMTCT service utilization. Poisson and negative binomial regression models were applied, considering the number of ANC visits as a dependent variable. The explanatory variables were age; educational status; type of occupation; decision-making power in the household; living in proximity to educated people; a neighborhood with good welfare services; location (urban high-HIV prevalence and rural low-HIV prevalence); transportation accessibility; walking distance (in minutes); and household income status. The alpha dispersion test (a) was performed to measure the goodness-of-fit of the model. Significant results were reported at p-values of < 0.05 and < 0.001.ResultsHousehold income, socio-economic setting (urban high-HIV prevalence and rural low-HIV prevalence) and walking distance (in minutes) had a statistically significant relationship with the number of ANC visits by pregnant women (p < 0.05). A pregnant woman from an urban high-HIV prevalence setting would be expected to make 34% more ANC visits (counts) than her rural low-HIV prevalence counterparts (p < 0.05). Holding other variables constant, a unit increase in household income would increase the expected ANC visits by 0.004%. An increase in walking distance by a unit (a minute) would decrease the number of ANC visits by 0.001(p < 0.001).ConclusionLong walking distance, low household income and living in a rural setting are the significant factors associated with low PMTCT service utilization. The primary strategies for a holistic policy to improve ANC/PMTCT utilization should thus include improving the geographical accessibility of ANC/PMTCT services, expanding household welfare and paying more attention to remote rural areas.

Dataset Information

On a Shape-Invariant Hazard Regression Model with application to an HIV Prevention Study of Mother-to-Child Transmission.

Publications

On a Shape-Invariant Hazard Regression Model with application to an HIV Prevention Study of Mother-to-Child Transmission.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets