Publications

Individual patient data (IPD) meta-analyses are increasingly common in the literature. In the context of estimating the diagnostic accuracy of ordinal or semi-continuous scale tests, sensitivity and specificity are often reported for a given threshold or a small set of thresholds, and a meta-analysis is conducted via a bivariate approach to account for their correlation. When IPD are available, sensitivity and specificity can be pooled for every possible threshold. Our objective was to compare the bivariate approach, which can be applied separately at every threshold, to two multivariate methods: the ordinal multivariate random-effects model and the Poisson correlated gamma-frailty model. Our comparison was empirical, using IPD from 13 studies that evaluated the diagnostic accuracy of the 9-item Patient Health Questionnaire depression screening tool, and included simulations. The empirical comparison showed that the implementation of the two multivariate methods is more laborious in terms of computational time and sensitivity to user-supplied values compared to the bivariate approach. Simulations showed that ignoring the within-study correlation of sensitivity and specificity across thresholds did not worsen inferences with the bivariate approach compared to the Poisson model. The ordinal approach was not suitable for simulations because the model was highly sensitive to user-supplied starting values. We tentatively recommend the bivariate approach rather than more complex multivariate methods for IPD diagnostic accuracy meta-analyses of ordinal scale tests, although the limited type of diagnostic data considered in the simulation study restricts the generalization of our findings.

BACKGROUND

Humoral immunity plays an important role against Pneumocystis jirovecii infection, yet clinical and environmental factors that impact bronchoalveolar antibody responses to P. jirovecii remain uncertain.

METHODS

From October 2008-December 2011 we enrolled consecutive HIV-infected adults admitted to San Francisco General Hospital (SFGH) who underwent bronchoscopy for suspected Pneumocystis pneumonia (PCP). We used local air quality monitoring data to assign ozone, nitrogen dioxide, and fine particulate matter exposures within 14 days prior to hospital admission. We quantified serum and bronchoalveolar lavage fluid (BALF) antibody responses to P. jirovecii major surface glycoprotein (Msg) recombinant constructs using ELISA. We then fit linear regression models to determine whether PCP and ambient air pollutants were associated with bronchoalveolar antibody responses to Msg.

RESULTS

Of 81 HIV-infected patients enrolled, 47 (58%) were diagnosed with current PCP and 9 (11%) had a prior history of PCP. The median CD4+ count was 51 cells/μl (IQR 15-129) and 44% were current smokers. Serum antibody responses to Msg were statistically significantly predictive of BALF antibody responses, with the exception of IgG responses to MsgC8 and MsgC9. Prior PCP was associated with increased BALF IgA responses to Msg and current PCP was associated with decreased IgA responses. For instance, among patients without current PCP, those with prior PCP had a median 73.2 U (IQR 19.2-169) IgA response to MsgC1 compared to a 5.00 U (3.52-12.6) response among those without prior PCP. Additionally, current PCP predicted a 22.5 U (95%CI -39.2, -5.82) lower IgA response to MsgC1. Ambient ozone within the two weeks prior to hospital admission was associated with decreased BALF IgA responses to Msg while nitrogen dioxide was associated with increased IgA responses.

CONCLUSIONS

PCP and ambient air pollutants were associated with BALF IgA responses to P. jirovecii in HIV-infected patients evaluated for suspected PCP.

Two papers published recently in the New England Journal of Medicine describe the utility of abiraterone acetate, an androgen biosynthesis inhibitor, in the early treatment of metastatic prostate cancer. In addition to establishing a new standard of care, these two articles pose a number of important questions for future investigation.