Brief Report: Assessing and Interpreting the Association Between Continuous Covariates and Outcomes in Observational Studies of HIV Using Splines.
Ontology highlight
ABSTRACT:
SUBMITTER: Shepherd BE
PROVIDER: S-EPMC5303133 | biostudies-literature | 2017 Mar
REPOSITORIES: biostudies-literature
Publications
Brief Report: Assessing and Interpreting the Association Between Continuous Covariates and Outcomes in Observational Studies of HIV Using Splines.
Shepherd Bryan E BE Rebeiro Peter F PF
Journal of acquired immune deficiency syndromes (1999) 20170301 3
The HIV epidemiology literature is replete with studies that categorize continuous predictors, such as age and CD4 cell count. It is well known that such an approach is suboptimal, but it persists in part because results are easy to interpret after categorization. Splines may be used to incorporate continuous predictors with smoothed curves into regression models without categorization or linearity assumptions. Properly presenting and interpreting results from analyses with splines is critical f ...[more]
Similar Datasets
Project description:ObjectiveTo evaluate potential adverse associations of individual antiretroviral medications used in combination antiretroviral therapy regimens on cardiac structure and function in youth with perinatally-acquired HIV infection (PHIV).DesignPHIV youth (N = 325) enrolled in a prospective multisite cohort study had a single echocardiogram at age 7-16 years to evaluate cardiac function and structure.MethodsWe applied several statistical approaches to evaluate associations between use of 18 individual antiretroviral medications with Z-scores for 11 measures of left ventricular function and structure. These included simultaneously evaluating all antiretroviral medications in adjusted linear regression models controlling for the false discovery rate (FDR), applying hierarchical models to estimate individual antiretroviral medication effects as deviations from drug class means, and evaluating latent measures of cardiac function and structure underlying multiple echocardiographic parameters.ResultsYouth taking combination regimens with a protease inhibitor (69%) had significantly better cardiac function than those on other regimens. After FDR control and adjustment for other antiretroviral medications, no individual antiretroviral medication was significantly associated with any measure of left ventricular function, but zidovudine was associated with higher adjusted mean Z-scores for one measure of left ventricular structure (end-systolic wall stress). Factor analysis identified three latent factors: heart function, heart size, and heart wall stress. Lopinavir was associated with better heart function scores, whereas zidovudine was associated with higher wall stress scores. Zidovudine and nevirapine were associated with higher heart size factor scores.ConclusionsDespite cardioprotective effects of combination regimens in PHIV youth, individual antiretroviral medications were associated with altered cardiac structure, which could progress to symptomatic cardiomyopathy in adulthood.
| S-EPMC6519135 | biostudies-literature
Project description:Human papillomavirus (HPV) testing is recommended for primary screening for cervical cancer by several health authorities. Several countries that have implemented HPV testing programs have encountered resistance against extended screening intervals and older age of initiation. As Canada prepares to implement HPV testing programs, it is important to understand women's preferences toward cervical cancer screening to ensure a smooth transition. The objective of this study was to assess Canadian women's current preferences toward cervical cancer screening. Using a web-based survey, we recruited underscreened ( > 3 years since last Pap test) and adequately screened (< 3 years since last Pap test) Canadian women aged 21-70 who were biologically female and had a cervix. We used Best-Worst Scaling (BWS) methodology to collect data on women's preferences for different screening methods, screening intervals, and ages of initiation. We used conditional logistic regression to estimate preferences in both subgroups. In both subgroups, women preferred screening every three years compared to every five or ten years, and initiating screening at age 21 compared to age 25 or 30. Adequately screened women (n = 503) most preferred co-testing, while underscreened women (n = 524) preferred both co-testing and HPV self-sampling over Pap testing. Regardless of screening status, women preferred shorter screening intervals, an earlier age of initiation, and co-testing. Adequate communication from public health authorities is needed to explain the extended screening intervals and age of initiation to prevent resistance against these changes to cervical cancer screening.
| S-EPMC9366717 | biostudies-literature
Project description:BackgroundHematological malignancies have continued to be highly prevalent among people living with HIV (PLHIV). This study assessed the occurrence of, risk factors for, and outcomes of hematological and nonhematological malignancies in PLHIV in Asia.MethodsIncidence of malignancy after cohort enrollment was evaluated. Factors associated with development of hematological and nonhematological malignancy were analyzed using competing risk regression and survival time using Kaplan-Meier.ResultsOf 7455 patients, 107 patients (1%) developed a malignancy: 34 (0.5%) hematological [0.08 per 100 person-years (/100PY)] and 73 (1%) nonhematological (0.17/100PY). Of the hematological malignancies, non-Hodgkin lymphoma was predominant (n = 26, 76%): immunoblastic (n = 6, 18%), Burkitt (n = 5, 15%), diffuse large B-cell (n = 5, 15%), and unspecified (n = 10, 30%). Others include central nervous system lymphoma (n = 7, 21%) and myelodysplastic syndrome (n = 1, 3%). Nonhematological malignancies were mostly Kaposi sarcoma (n = 12, 16%) and cervical cancer (n = 10, 14%). Risk factors for hematological malignancy included age >50 vs. ≤30 years [subhazard ratio (SHR) = 6.48, 95% confidence interval (CI): 1.79 to 23.43] and being from a high-income vs. a lower-middle-income country (SHR = 3.97, 95% CI: 1.45 to 10.84). Risk was reduced with CD4 351-500 cells/µL (SHR = 0.20, 95% CI: 0.05 to 0.74) and CD4 >500 cells/µL (SHR = 0.14, 95% CI: 0.04 to 0.78), compared to CD4 ≤200 cells/µL. Similar risk factors were seen for nonhematological malignancy, with prior AIDS diagnosis showing a weak association. Patients diagnosed with a hematological malignancy had shorter survival time compared to patients diagnosed with a nonhematological malignancy.ConclusionsNonhematological malignancies were common but non-Hodgkin lymphoma was more predominant in our cohort. PLHIV from high-income countries were more likely to be diagnosed, indicating a potential underdiagnosis of cancer in low-income settings.
| S-EPMC6375805 | biostudies-literature
Project description:In this paper we use Cox's regression model to fit failure time data with continuous informative auxiliary variables in the presence of a validation subsample. We first estimate the induced relative risk function by kernel smoothing based on the validation subsample, and then improve the estimation by utilizing the information on the incomplete observations from non-validation subsample and the auxiliary observations from the primary sample. Asymptotic normality of the proposed estimator is derived. The proposed method allows one to robustly model the failure time data with an informative multivariate auxiliary covariate. Comparison of the proposed approach with several existing methods is made via simulations. Two real datasets are analyzed to illustrate the proposed method.
| S-EPMC4651204 | biostudies-literature
Project description:When validating a risk model in an independent cohort, some predictors may be missing for some subjects. Missingness can be unplanned or by design, as in case-cohort or nested case-control studies, in which some covariates are measured only in subsampled subjects. Weighting methods and imputation are used to handle missing data. We propose methods to increase the efficiency of weighting to assess calibration of a risk model (i.e. bias in model predictions), which is quantified by the ratio of the number of observed events, $\mathcal{O}$, to expected events, $\mathcal{E}$, computed from the model. We adjust known inverse probability weights by incorporating auxiliary information available for all cohort members. We use survey calibration that requires the weighted sum of the auxiliary statistics in the complete data subset to equal their sum in the full cohort. We show that a pseudo-risk estimate that approximates the actual risk value but uses only variables available for the entire cohort is an excellent auxiliary statistic to estimate $\mathcal{E}$. We derive analytic variance formulas for $\mathcal{O}/\mathcal{E}$ with adjusted weights. In simulations, weight adjustment with pseudo-risk was much more efficient than inverse probability weighting and yielded consistent estimates even when the pseudo-risk was a poor approximation. Multiple imputation was often efficient but yielded biased estimates when the imputation model was misspecified. Using these methods, we assessed calibration of an absolute risk model for second primary thyroid cancer in an independent cohort.
| S-EPMC9608650 | biostudies-literature
Project description:BackgroundAmong men who have sex with men, there is now clear evidence that the risk of HIV transmission through condomless sex when the HIV-positive partner is virally suppressed is effectively zero. However, an understanding of the accuracy of reporting of viral load among serodiscordant same-sex male couples is missing from the literature.SettingThis analysis uses data from the baseline sample of Stronger Together, a randomized controlled efficacy trial of an innovative dyadic intervention to enhance antiretroviral therapy adherence for HIV serodiscordant male couples in 3 US cities (Atlanta, Boston, and Chicago).MethodsBiomarker-confirmed and self-reported measures of viral load were used to assess the accuracy of self-report of viral suppression. In this descriptive analysis, the percentage of men who inaccurately reported being virally suppressed is compared across demographic, relationship, and HIV care characteristics.ResultsResults confirm those of other recent studies that have shown relatively high levels of inaccuracy in reporting of viral suppression. Although 72.5% of men could accurately report their viral load status, 20% reported that they were virally suppressed when they did not have a biomarker-confirmed measure of viral suppression.ConclusionThese results highlight the need to provide interventions to men who have sex with men living with HIV to support access to care and ensure current knowledge of viral load and to continue to support primary prevention of HIV through condom use and pre-exposure prophylaxis. For couples, particularly serodiscordant male couples, interventions that can teach the couple how to collaborate to achieve and maintain viral suppression for the positive partner are an urgent and pragmatic programmatic priority that can equip couples with the knowledge required to correctly implement U = U strategies.
| S-EPMC8754005 | biostudies-literature
Project description:ContextPersons with HIV are at increased risk for adipose dysfunction, which could mediate metabolic complications such as cardiovascular disease, fatty liver disease, and diabetes. We have previously reported reduced browning and beiging capacity of the subcutaneous adipose depot in HIV.ObjectiveWe sought to evaluate how HIV-related parameters are related to the expression of brown and beige fat genes in the abdominal subcutaneous adipose tissue.DesignEighteen persons with HIV underwent punch biopsy of abdominal subcutaneous fat to determine mRNA expression of adipose-related genes using quantitative reverse transcriptase-polymerase chain reaction.ResultsDuration of antiretroviral therapy use, particularly related to protease inhibitor use, was significantly related to reduced expression of multiple brown and beige fat genes (including UCP1, PGC1α, PRDM16 and others, all P ≤ 0.04) in the abdominal subcutaneous fat. In addition, duration of HIV and CD4 T-cell count were significantly correlated with reduced expression of multiple brown and beige fat genes in the abdominal subcutaneous fat (PGC1α, P2XR5, TMEM26, CD137, all P ≤ 0.05 for duration of HIV; and PGC1α, ZIC1, PRDM16, PAT2, P2RX5, TMEM26, CD137, all P ≤ 0.04). In contrast, HIV viral load did not correlate with any brown or beige fat genes.ConclusionsKey HIV-related parameters reflective of nonacute infection (increased duration of HIV and duration of antiretroviral therapy use) or relatively reduced immunologic function (lower CD4 count) were linked to reduced expression of brown and beige fat gene in the abdominal subcutaneous adipose depot.Clinical trial registrationNCT01098045.
| S-EPMC6857717 | biostudies-literature
Project description:The relationships between HIV infection, monocyte activation, and endothelial colony-forming cells (ECFCs) are unknown. We compared ECFC, intermediate monocytes (CD14 CD16), and nonclassical monocytes (CD14 CD16) levels in HIV-infected participants virologically suppressed on antiretroviral therapy, HIV-infected treatment-naive participants, and HIV-uninfected healthy controls. ECFC levels were significantly higher in the HIV-infected virologically suppressed group compared with the uninfected controls. CD14 CD16 percentages (but not CD14 CD16 cells) were significantly higher in both HIV-infected groups vs. uninfected controls. In the HIV-infected groups, ECFCs and CD14 CD16 intermediate monocytes were significantly and inversely correlated. Lower availability of ECFCs may partly explain the relationship between greater intermediate monocytes and atherosclerosis in HIV.
| S-EPMC4357537 | biostudies-literature
Project description:ObjectiveWomen with HIV (WHIV) on antiretroviral therapy (ART) face an increased risk of cardiovascular disease (CVD) in the context of heightened systemic immune activation. Aortic stiffness, a measure of vascular dysfunction and a robust predictor of CVD outcomes, is highly influenced by immune activation. We compared aortic stiffness among women with and without HIV and examined interrelationships between aortic stiffness and key indices of systemic immune activation.MethodsTwenty WHIV on ART and 14 women without HIV group-matched on age and body mass index (BMI) were prospectively recruited and underwent cardiovascular magnetic resonance imaging, as well as metabolic and immune phenotyping.ResultsAge and BMI did not differ significantly across groups (age: 52 ± 4 vs. 53 ± 6 years; BMI: 32 ± 7 vs. 32 ± 7 kg/m). Aortic pulse wave velocity (aPWV) was higher among WHIV (8.6 ± 1.3 vs. 6.5 ± 1.3 m/s, P < 0.0001), reflecting increased aortic stiffness. Among the whole group and among WHIV, aPWV related to sCD163 levels (whole group: R = 0.65, P < 0.0001; WHIV: R = 0.73, P = 0.0003) and to myocardial fibrosis (extracellular volume; whole group: R = 0.54, P = 0.001; WHIV: R = 0.47, P = 0.04). Both HIV status and sCD163 levels independently predicted aPWV, controlling for age, BMI, cigarette smoking status, and systolic blood pressure (HIV status: β-estimate = 0.69, 95% CI [0.1 to 1.3], P = 0.02; sCD163: β-estimate = 0.002, 95% CI [0.0006 to 0.004], P = 0.01). Among WHIV, sCD163 levels independently predicted aPWV, controlling for duration of HIV, CD4 count, and HIV viral load (sCD163: β-estimate = 0.004, 95% CI [0.002 to 0.005], P = 0.0005).ConclusionsAsymptomatic WHIV on ART have increased aortic stiffness as compared to matched control subjects. Among WHIV, aPWV related to heightened monocyte activation (sCD163) and to downstream CVD pathology (myocardial fibrosis). CLINICALTRIALS.Gov registrationNCT02874703.
| S-EPMC8320532 | biostudies-literature
Project description:Viral load (VL) rebound timing and set point were analyzed in 235 participants undergoing analytic treatment interruption (ATI) in 6 AIDS Clinical Trials Group studies. There was no significant association between rebound timing and ATI VL set point for those who rebounded ?12 weeks. VL set points were lower in participants with rebound >12 weeks (P < 0.001) and participants treated during early infection (P < 0.001). Pre-antiretroviral therapy VL correlated with set point, though 68% of participants had a set point lower than pre-antiretroviral therapy VL. These results illustrate complex relationships between post-ATI virologic outcomes and the potential presence of biological factors mediating rebound timing and set point.
| S-EPMC4911279 | biostudies-literature