9 research outputs found
Twelve cities: does lowering speed limits save pedestrian lives?
No full textWe investigate whether American cities can expect to achieve a meaningful reduction in pedestrian deaths by lowering the posted speed limit. We present our work in three sections. First we briefly motivate the problem and provide a description of the dataset. Second we fit a log-linear model and compare sources of variation with an analysis of variance. Finally we demonstrate a sample use case. We evaluate the decision to lower many of New York City's posted speed limits from 30 mph to 25 mph. In our evaluation we assume the assignment of speed limits to roads is ignorable given measured covariates, and we calculate the number of lives saved by estimating the causal effect of lowering the speed limit on New York City roads from 30 mph to 25 mph on 25 mph roads. We find some evidence that a lower speed limit does in fact reduce fatality rates, and our estimated causal effect is similar to the traditional before-after analysis espoused by policy analysts. Nevertheless, we conclude that adjusting the posted speed limit in urban environments does not correspond with a reliable reduction in pedestrian fatalities.Code and data available at the stancon_talks GitHub repository (https://github.com/stan-dev/stancon_talks
Measuring the impact of spatial perturbations on the relationship between data privacy and validity of descriptive statistics
No full textAbstract
Background
Like many scientific fields, epidemiology is addressing issues of research reproducibility. Spatial epidemiology, which often uses the inherently identifiable variable of participant address, must balance reproducibility with participant privacy. In this study, we assess the impact of several different data perturbation methods on key spatial statistics and patient privacy.
Methods
We analyzed the impact of perturbation on spatial patterns in the full set of address-level mortality data from Lawrence, MA during the period from 1911 to 1913. The original death locations were perturbed using seven different published approaches to stochastic and deterministic spatial data anonymization. Key spatial descriptive statistics were calculated for each perturbation, including changes in spatial pattern center, Global Moran’s I, Local Moran’s I, distance to the k-th nearest neighbors, and the L-function (a normalized form of Ripley’s K). A spatially adapted form of k-anonymity was used to measure the privacy protection conferred by each method, and its compliance with HIPAA and GDPR privacy standards.
Results
Random perturbation at 50 m, donut masking between 5 and 50 m, and Voronoi masking maintain the validity of descriptive spatial statistics better than other perturbations. Grid center masking with both 100 × 100 and 250 × 250 m cells led to large changes in descriptive spatial statistics. None of the perturbation methods adhered to the HIPAA standard that all points have a k-anonymity > 10. All other perturbation methods employed had at least 265 points, or over 6%, not adhering to the HIPAA standard.
Conclusions
Using the set of published perturbation methods applied in this analysis, HIPAA and GDPR compliant de-identification was not compatible with maintaining key spatial patterns as measured by our chosen summary statistics. Further research should investigate alternate methods to balancing tradeoffs between spatial data privacy and preservation of key patterns in public health data that are of scientific and medical importance.http://deepblue.lib.umich.edu/bitstream/2027.42/173714/1/12942_2020_Article_256.pd
Vaccine efficacy for binary post-infection outcomes under misclassification without monotonicity
No full textIn order to meet regulatory approval, pharmaceutical companies often must demonstrate that new vaccines reduce the total risk of a post-infection outcome like transmission, symptomatic disease, severe illness, or death in randomized, placebo-controlled trials. Given that infection is a necessary precondition for a post-infection outcome, one can use principal stratification to partition the total causal effect of vaccination into two causal effects: vaccine efficacy against infection, and the principal effect of vaccine efficacy against a post-infection outcome in the patients that would be infected under both placebo and vaccination. Despite the importance of such principal effects to policymakers, these estimands are generally unidentifiable, even under strong assumptions that are rarely satisfied in real-world trials. We develop a novel method to nonparametrically point identify these principal effects while eliminating the monotonicity assumption and allowing for measurement error. Furthermore, our results allow for multiple treatments, and are general enough to be applicable outside of vaccine efficacy. Our method relies on the fact that many vaccine trials are run at geographically disparate health centers, and measure biologically-relevant categorical pretreatment covariates. We show that our method can be applied to a variety of clinical trial settings where vaccine efficacy against infection and a post-infection outcome can be jointly inferred. This can yield new insights from existing vaccine efficacy trial data and will aid researchers in designing new multi-arm clinical trials
Bayesian Methods for Modeling Cumulative Exposure to Extensive Environmental Health Hazards
No full textMeasuring the impact of an environmental point source exposure on the risk of disease, like cancer or childhood asthma, is well-developed. Modeling how an environmental health hazard that is extensive in space, like a wastewater canal, impacts disease risk is not. We propose a novel Bayesian generative semiparametric model for characterizing the cumulative spatial exposure to an environmental health hazard that is not well-represented by a single point in space. The model couples a dose-response model with a log-Gaussian Cox process integrated against a distance kernel with an unknown length-scale. We show that this model is a well-defined Bayesian inverse model, namely that the posterior exists under a Gaussian process prior for the log-intensity of exposure, and that a simple integral approximation adequately controls the computational error. We quantify the finite-sample properties and the computational tractability of the discretization scheme in a simulation study. Finally, we apply the model to survey data on household risk of childhood diarrheal illness from exposure to a system of wastewater canals in Mezquital Valley, Mexico
Racial Disparities in Coronavirus Disease 2019 (COVID-19) Mortality Are Driven by Unequal Infection Risks
No full textBACKGROUND: As of 1 November 2020, there have been >230 000 deaths and 9 million confirmed and probable cases attributable to severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in the United States. However, this overwhelming toll has not been distributed equally, with geographic, race/ethnic, age, and socioeconomic disparities in exposure and mortality defining features of the US coronavirus disease 2019 (COVID-19) epidemic. METHODS: We used individual-level COVID-19 incidence and mortality data from the state of Michigan to estimate age-specific incidence and mortality rates by race/ethnic group. Data were analyzed using hierarchical Bayesian regression models, and model results were validated using posterior predictive checks. RESULTS: In crude and age-standardized analyses we found rates of incidence and mortality more than twice as high than for Whites for all groups except Native Americans. Blacks experienced the greatest burden of confirmed and probable COVID-19 (age-standardized incidence, 1626/100 000 population) and mortality (age-standardized mortality rate, 244/100 000). These rates reflect large disparities, as Blacks experienced age-standardized incidence and mortality rates 5.5 (95% posterior credible interval [CrI], 5.4-5.6) and 6.7 (95% CrI, 6.4-7.1) times higher than Whites, respectively. We found that the bulk of the disparity in mortality between Blacks and Whites is driven by dramatically higher rates of COVID-19 infection across all age groups, particularly among older adults, rather than age-specific variation in case-fatality rates. CONCLUSIONS: This work suggests that well-documented racial disparities in COVID-19 mortality in hard-hit settings, such as Michigan, are driven primarily by variation in household, community, and workplace exposure rather than case-fatality rates
Modeling Spatial Risk of Diarrheal Disease Associated with Household Proximity to Untreated Wastewater Used for Irrigation in the Mezquital Valley, Mexico
No full textstan-dev/math v2.15.0
No full textv.2.15.0 (13 April 2017)
New Features
<ul>
<li>Efficient blocking algorithm for gradient of the Cholesky(#384)</li>
<li>New distribution functions _lpdf / _lpmf / _lcdf / _lccdf to replace _log function (#320)</li>
<li>Univariate normal distribution on sufficient statistics(#38) </li>
<li>New to_matrix function for real arrays(real[]) (#467)</li>
<li>New specialization of stan::math::array_builder for matrix types(#496)</li>
</ul>
Bug Fixes
<ul>
<li>Fixes to hypergeometric functions(#487)</li>
</ul>
Other
<ul>
<li>Speedup for categorical_rng(#503)</li>
<li>Speedup for non-stiff ODE integration(#512)</li>
<li>Refactor VectorView into scalar_seq_view(#464)</li>
</ul>
stan-dev/math v2.16.0
No full textv.2.16.0 (15 June 2017)
New Features
<ul>
<li>New append_array function</li>
<li>Add categorical_logit_rng function</li>
</ul>
Bug Fixes
<ul>
<li>Align gamma_* function parameter names with documentation</li>
</ul>
Other
<ul>
<li>Update to Eigen 3.3.3</li>
<li>Support g++ 4.9</li>
<li>Fix overload logic in mdivide_left_tri_low so that it calls the var version of mdivide_left_tri where appropriate.</li>
<li>Check consistent size of state and dy_dt in ode_system</li>
<li>OperandsAndPartials refactor with new multivariate / nested container support</li>
<li>Update LLT to inplace decomposition per eigen 3.3 doc</li>
<li>Disable printf functions from CVODES</li>
</ul>
stan-dev/stan v2.13.0
No full textv2.13.0 (25 November 2016)
New Team Members
<ul>
<li>Thel Seraphim (Columbia University) -- Stan and Math libraries</li>
<li>Vincent Picaud (CEA, France) -- MathematicaStan</li>
</ul>
Bug Fixes
<ul>
<li>generated code for lower truncation fixed for discrete variables (#2054)</li>
<li>typo in error messages for RNGs only allowed in transformed data
block (#2124)</li>
<li>variables ending in _lpdf are now allowed (#2123)</li>
<li>cov_exp_quad() not compiling in C++ (#2113)</li>
<li>conditional operator in functions not compiling in C++ (#2101)</li>
<li>off-by-one error in error message for integrate_ode_bdf() fixed (#2073)</li>
</ul>
New User-Facing Features
<ul>
<li>vectorization of unary functions (#2119, #2037)</li>
<li>bernoulli_logit_rng() added to language (#2084)</li>
<li>Jacobian warning now suggests <code>target +=</code> (#2066)</li>
<li>matrix_exp() function now available (#2043)</li>
<li>compound declaration / definition statements (#1951)</li>
</ul>
New Internal Features
<ul>
<li>user-defined functions can be declared and not defined (#2068)</li>
</ul>
Documentation
<ul>
<li>manaul reorganization (#1599)</li>
<li>line too long (#2121)</li>
<li>added integer % operator documentation (#2065)</li>
<li>lots more (#2051)</li>
</ul>
