Bootstrap Confidence Intervals Research Articles

Bootstrap confidence intervals of process capability indices Spmk, Spmkc and Cs for Frechet distribution

AbstractIn this paper, we considered the Bayesian estimators under reference and Jeffery's priors and maximum likelihood estimators to estimate the unknown parameters of the process capability indices Spmk, Spmkc, and Cs forFrechetdistribution. Further, we developed bootstrap confidence intervals for aforementioned process capability indices based on above‐mentioned estimators. Monte Carlo simulations are performed to investigate the performance of process capability indices through skewness, kurtosis, mean square error and widths of bootstrap confidence intervals for small, moderate, and large sample sizes. Simulations results indicate that the Bayesian estimator under reference prior outperforms even in small sample sizes, and all performed equally well for larger sample sizes. Moreover, the average width for bootstrap confidence interval for Cs is least in all. Finally, real data is analyzed for illustration purposes.

Trending and analysis of payload vs. All low earth conjunction data messages below 1000 km, from 2016 through 2021

The Air Force Safety Center is developing a program to quantify the distribution of risks to the orbital environment across time and space by trending and analyzing orbital conjunction events. Previous efforts described automating the intake of Conjunction Data Messages (CDM) previously generated by the 18th and 19th Space Defense Squadrons (18 SDS, 19 SDS) and developing pre-processing scripts necessary to populate information germane to orbital collision risk assessment. The intent of the current effort is to account not only for the probability of collision (Pc), but also the consequence of the debris products from a collision to the low Earth orbital environment, should it occur. To accomplish this, the team analyzed all reported CDMs of conjunction events below 1000 km in altitude, between 1 Jan 2016 and 31 Dec 2021. 18 SDS reported Pcs were re-computed using the Foster-92 method. A circular 2-dimensional interaction area was applied instead of a square, to reduce conservatism and lend closer towards realism. Hard Body Radii estimates were pooled from externally sourced information, including The Aerospace Corporation and various open sources. When not available, radius estimates were imputed based on available data. Covariance information was sourced from CDMs. For consequence, the amount of damaging, lethal non-trackable and trackable debris products were estimated using The Aerospace Corporation's IMPACT 8.0 model. The model accounts for relative velocity, object masses, mass distribution, and structure. Object distribution and structure were inferred from the object type field in the CDMs. Relative velocity was also sourced from CDMs. The object masses were sourced from data furnished by The Aerospace Corporation and Union of Concerned Scientists satellite database. The expected number of damaging, lethal non-trackable, and trackable debris products were multiplied by the Pc to produce a quantitative risk metric for each conjunction event. Risks were adjusted to account for likely collision avoidance maneuvers based on satellite age, mass, and Pc. The results show a logarithmic increase in total risk over time, particularly after 2020. This change correlates to the launch of Starlink and OneWeb satellites into orbit, and analysis confirmed that the uptick in risk is attributable to the launch of those large constellations into orbit. The risk metric for Starlink vs. Starlink conjunctions was further adjusted to more accurately account for proprietary processes and high accuracy ephemerides which enables better information for performing collision avoidance maneuvers. This adjustment reduced the slope of the risk trends starting in 2020, but they are still positive and logarithmic. Spatial domain analysis revealed that the preponderance of debris production risk for reported conjunctions below 1,000 km resides between the altitudes of 450 km and 850 km. The spatial distribution of risk by altitude changed starting in 2020 due to the launch of large constellations, resulting in a higher and more evenly distributed risk of debris production between 350 and 1,000 km. Results also indicate that more trackable debris would have likely been generated by collisions between payloads and trackable objects, with perigees below 1,000 km than the 37 that have been confirmed and cataloged during that time period. However, analysis of the four confirmed payload collision events with cataloged debris produced a bootstrapped confidence interval of trackable debris generated per year suggesting the trackable debris production rate estimated here is within a plausible range. Planned future work looks to address known assumptions and limitations. This work is the first known effort to quantitatively analyze historical risks of conjunction events using full sets of CDM data and fuse them with the latest semi-analytical breakup algorithms. Analyses such as these are necessary to better inform ongoing and future Space Safety and Space Traffic Management policy development efforts within the US Government and internationally.

A Bootstrap-Based Tooth Surface Errors Statistics Methodology for Batch Hypoid Gears after Heat Treatment

In the manufacturing and production of hypoid gears, it is a necessary key problem to improve the tooth surface heat treatment precision and production efficiency of the hypoid gears. How to use advanced statistical theory and methods to evaluate the whole batch machining quality of the tooth surface after heat treatment is particularly urgent. In this connection, for the same batch of hypoid gears with the same gear material, numerical control gear milling method, and heat treatment specifications, a bootstrap-based statistics scheme of tooth surface errors after heat treatment is proposed in this paper. The bootstrap statistics model of the tooth surface errors for the batch hypoid gears is established. The bootstrap probability eigenvalues and confidence intervals of the measurement sequence points on the tooth surface errors are solved, and the optimizing selection of the single sampling numbers and the repeated sampling times is completed. On this basis, by applying the cubic NURBS surface fitting method, the mean value difference surface of the batch tooth surface errors data is constructed, the statistics laws of the whole batch tooth surface errors after heat treatment is determined, and the effective evaluation of the whole batch tooth surface accuracy is realized. Finally, the feasibility and correctness of the bootstrap-based statistics scheme are verified by the tooth surface errors bootstrap statistics application experiment of one kind of hypoid gear. The results show that with the help of the bootstrap statistics method proposed in this paper, it is not necessary to accurately measure the tooth surface errors of the whole batch of hypoid gears one by one. Only by randomly selecting 10 tooth surface samples and repeatedly sampling 2000 times, the original sample characteristic values of the whole batch tooth surface errors can be accurately estimated, and the heat treatment deformation statistics laws of the whole patch tooth surfaces can be also counted.

Uses of a new asymmetric loss-based process capability index in the electronic industries

This article suggests a novel process capability index (PCI) termed as , which is based on an asymmetric loss function (linear exponential) for a normal process and offers a specific method of incorporating the loss in capability analysis. Next, we estimate the suggested PCI using the moment estimation approach when the process follows a normal distribution, and we compare the effectiveness of the investigated estimation methods in terms of their mean squared errors through simulation analysis. Additionally, the confidence intervals for the index are constructed using the generalized confidence interval (GCI) and parametric bootstrap confidence interval (BCI) approach. Using Monte Carlo simulation, the performance of the GCI and BCI is compared in terms of average width, associated coverage probabilities, and relative coverage. Finally, three real data sets from the electronic industries are re-analyzed to show the usefulness of the suggested index, MOM estimation, GCI and BCI.

Analysis of Dependent Variables Following Marshal- Olkin Bivariate Distributions in the Presence of Progressive Type II Censoring

In this paper, the likelihood function under progressive Type II censoring is generalized for Marshal-Olkin bivariate class of distributions and applied it on the bivariate Dagum distribution. Maximum likelihood estimation is considered for the model unknown parameters. Asymptotic and bootstrap confidence intervals for the unknown parameters are evaluated under progressive Type II censoring. Bayesian estimation is also considered in both complete and progressive Type II censored samples; moreover, the Bayes estimators are obtained explicitly with respect to square error loss function in both cases.

When to Use Different Inferential Methods for Power Analysis and Data Analysis for Between-Subjects Mediation

Several options exist for conducting inference on indirect effects in mediation analysis. Although methods that use bootstrapping are the preferred inferential approach for testing mediation, they are time-consuming when the test must be performed many times for a power analysis. Alternatives that are more computationally efficient are not as robust, meaning accuracy of the inferences from these methods is more affected by nonnormal and heteroskedastic data. Previous research has shown that different sample sizes are needed to achieve the same amount of statistical power for different inferential approaches with data that meet all the statistical assumptions of linear regression. By contrast, we explore how similar power estimates are at the same sample size, including when assumptions are violated. We compare the power estimates from six inferential methods for between-subjects mediation using a Monte Carlo simulation study. We varied the path coefficients, inferential methods for the indirect effect, and degree to which assumptions are met. We found that when the assumptions of linear regression are met, three inferential methods consistently perform similarly: the joint significance test, the Monte Carlo confidence interval, and the percentile bootstrap confidence interval. When the assumptions were violated, the nonbootstrapping methods tended to have vastly different power estimates compared with the bootstrapping methods. On the basis of these results, we recommend using the more computationally efficient joint significance test for power analysis only when no assumption violations are hypothesized a priori. We also recommend the joint significance test to pick an optimal starting sample size value for power analysis using the percentile bootstrap confidence interval when assumption violations are suspected.

Mediating Role of Depression Between Diabetes Management Self-Efficacy and Diabetes Self-Care Behavior Among Elderly Type 2 Diabetes Mellitus Patients in China.

To investigate the current status of diabetes self-care behavior and the association between depression, self-efficacy and self-care in a sample of Chinese elderly type 2 diabetes mellitus (T2DM) patients. A cross-sectional study with a convenient sample including 240 elderly T2DM patients collected the data of demographic characteristics, diabetes self-care behavior, self-efficacy and depression status. The difference of self-care behavior in different sample characteristics was compared by independent t-test. The Personal correlation analysis was employed to examine the correlation of study variables. The method of bootstrap was used to analyze mediating role of depression. Only 22.5% of patients reported better diabetes self-care behavior and depression partly mediated the association between self-efficacy and self-care behavior. The significant coefficient of path a (B = -0.052, p < 0.001) and path b (B = -0.423, p < 0.05) indicated negative associations of self-efficacy on depression, and depression on self-care behavior. The indirect effect (Path a × b) between self-efficacy and self-care behavior through depression was significant (B = 0.022, p < 0.05), the 95% bias-corrected bootstrap confidence interval was 0.004 to 0.006. Meanwhile, the mediating role of depression was not found significant among the participants reported 60-74 years old (B = 0.104, p < 0.001). But depression completely mediated this association among the participants reported 75-89 years old (B = 0.034, p > 0.05). The level of diabetes self-care behavior among the elderly T2DM patients in Dahu community of Anqing city was hardly optimistic. The self-efficacy focused intervention could be encouraged for community and clinicians to improve diabetes self-care behavior. Moreover, the prevalence of depression and T2DM is increasing in younger population. More work is needed to confirm these findings, especially conducting cohort studies on different populations.

Impact of social capital on entrepreneurial intention of polytechnic students: The mediating role of innovativeness

Entrepreneurship is seen as a crucial component of any nation's economic progress. The most significant predictor of entrepreneurship is thought to be the entrepreneurial intention. Therefore, it is crucial to understand the variables that shape entrepreneurial intent. The purpose of this study is to investigate how social capital and innovativeness, two significant predictors of entrepreneurial intention, interact to determine the entrepreneurial intention of polytechnic students from the Indian state of Arunachal Pradesh. A questionnaire based on a 5-point Likert scale was completed by 261 polytechnic students (214 male,47 female) from six polytechnic (5 Govt,1 Private) colleges across Arunachal Pradesh. The structure model was examined using structural equation modelling (SEM) analysis in SPSS AMOS (Analysis of Moment Structure), and bootstrap confidence intervals were computed to test the mediating role. The findings showed that both social capital and innovativeness have a significant positive impact on entrepreneurial intention, and innovativeness partially mediates the association between social capital and entrepreneurial intention. Our study's findings have significant ramifications for vocational educators because they show that emphasizing the importance of social capital and innovativeness might increase polytechnic students' likelihood to become entrepreneurs.

Comparative Study with Applications for Gompertz Models under Competing Risks and Generalized Hybrid Censoring Schemes

In reliability and survival analysis, the time-to-failure data play an important role in the development of the reliability and life characteristics of the products. In some cases, these kinds of data are modeled using a competing risks model. The problem of conducting comparative life testing under a competing risks model when the units come from different lines of production has recently been addressed. In this paper, we address this problem when the life of the unit is distributed using the Gompertz distribution, noting that the units come from two lines of production and two independent causes of failure are activated. The data are collected under a joint generalized type-II hybrid censoring scheme. Maximum likelihood estimators of the unknown parameters are derived, along with the corresponding asymptotic confidence intervals. We also adopt two bootstrap confidence intervals. Using independent gamma priors, the Bayes estimators relative to squared error loss function are obtained with credible intervals. The properties and quality of estimators are measured by performing a Monte Carlo simulation study. Finally, a real-life data set is analyzed to discuss the applicability of the proposed methods to real phenomena. The optimal plan with respect to comments on the numerical results is discussed in the conclusion.

On Predictive Modeling Using a New Three-Parameters Modification of Weibull Distribution and Application

In this article, a new modification of the Weibull model with three parameters, the new exponential Weibull distribution (E-WD), is defined. The new model has many statistical advantages, the heavy-tailed behavior and the regular variation property were offered. Many of the important statistical functions of the modified model are presented in closed forms. The flexibility of E-WD has been improved. The proposed model can be used to fit data with different shapes, it can be right-skewed, left-skewed, decreasing, curved and symmetric. Some distribution properties of the proposed model, including moment generating function, characteristic function, moment, quantile and identifiability property, have been derived. In addition to the information generating function, the Shannon entropy and information energy are also discussed. The maximum likelihood approach and Bayesian estimation are used to estimate the distribution parameters. In the Bayesian method, three different loss functions are used. The calculations show the biases and estimated risks to obtain the best estimator. The bootstrap confidence intervals, the asymptotic confidence intervals and the observed variance-covariance matrix are obtained. Metropolis Hastings’ MCMC procedure is used for the calculations. We apply the composite distribution to stock data for four variables. The goodness-of-fit results show that the model performs well compared to its competitors. The proposed model can be used for forecasting and decision making.

High-order Coverage of Smoothed Bayesian Bootstrap Intervals for Population Quantiles

We characterize the high-order coverage accuracy of smoothed and unsmoothed Bayesian bootstrap confidence intervals for population quantiles. Although the original (Rubin 1981) unsmoothed intervals have the same O(n−1/2) coverage error as the standard empirical bootstrap, the smoothed Bayesian bootstrap of Banks (1988) has much smaller O(n−3/2[log(n)]3) coverage error and is exact in special cases, without requiring any smoothing parameter. It automatically removes an error term of order 1/n that other approaches need to explicitly correct for. This motivates further study of the smoothed Bayesian bootstrap in more complex settings and models.

Mediating effect of BMI on the relation of dietary patterns and glycemic control inT2DM patients: results from China community-based cross-sectional study

ObjectiveTo analyze the effects of different dietary types on in type 2 diabetes mellitus (T2DM) and determine the mediating effects of Body Mass Index (BMI) on dietary type with Fasting Plasma Glucose (FPG), Glycosylated Hemoglobin (HbA1c) on the associations in T2DM.MethodsData of community-based cross-sectional study with 9602 participants including 3623 men and 5979 women were collected from the project ‘Comprehensive Research in prevention and Control of Diabetes mellitus (CRPCD)’ conducted by Jiangsu Center for Disease Control and Prevention in 2018. The dietary data were collected from a food frequency qualitative questionnaire (FFQ) and dietary patterns were derived through Latent Class Analysis (LCA). Then, Logistics regression analyses were used to evaluate the associations of FPG, HbA1c with different dietary patterns. The BMI (BMI = height/weight2) was used as a moderator to estimate the mediating effect. Mediation analysis was performed using hypothetical variables, the mediation variables, to identify and explain the observed mechanism of association between the independent and dependent variables while the moderation effect was tested with multiple regression analysis with interaction terms.ResultsAfter completing Latent Class Analysis (LCA), the dietary patterns were divided into three categories: TypeI, TypeII, TypeIII. After adjusting for confounding factors such as gender, age, education level, marital status, family income, smoking, drinking, disease course, HDL-C, LDL-C, TC, TG, oral hypoglycemic drugs, insulin therapy, Hypertension, Coronary heart disease, Stroke, Type III were all significantly associated with HbA1c compared to those with Type I (P < 0.05), and the research showed the patients with Type III had High glycemic control rate. Taking type I as the reference level, the 95% Bootstrap confidence intervals of the relative mediating effect of TypeIII on FPG were (-0.039, -0.005), except 0, indicating that the relative mediating effect was significant (αIII = 0.346*, βIIIFPG = -0.060*). The mediating effect analysis was performed to demonstrate that BMI was used as a moderator to estimate the moderation effect.ConclusionsOur findings demonstrate that consuming Type III dietary patterns associates with good glycemic control in T2DM and the BMI associations would be playing a two-way effect between diet and FPG in Chinese population with T2DM, indicated that Type III could not only directly affect FPG, but also affect FPG through the mediating effect of BMI.

Confidence intervals of the process capability index Cpc$C_{pc}$ revisited via modified bootstrap technique and ROC curves

AbstractWe consider the process capability index (PCI), a widely used quality‐related statistic used to assess the quality of products and performance of monitored processes in various industries. It is widely known that the conventional PCIs perform well when the quality process being monitored has a normal distribution. Unfortunately, using the indices to evaluate a non‐normally distributed process often leads to inaccurate results. In this article, we consider a new PCI, , that can be used in both normal and non‐normal scenarios. The objective of this article is threefold: (i) We provide a corrected form of the confidence interval for . (ii) We compare the performance of three nonparametric bootstrap confidence intervals (BCIs) for . Specifically, the standard bootstrap, percentile bootstrap, and bias‐corrected percentile bootstrap. Under various distributional assumptions such as the normal, chi‐square, Student t, Laplace, and two‐parameter exponential distributions, the estimated coverage probabilities and average width of the confidence intervals and BCIs for are compared. (iii) The power of the respective bootstrap approaches is evaluated by using the equivalence relation between confidence interval construction and two‐sided hypothesis testing. We also provide the receiver operating characteristic curves to evaluate their performance. Finally, as an illustrative example, an actual data set related to groove dimensions (in inches) measured from a manufacturing process of ignition keys is re‐analyzed to illustrate the utility of the proposed methods.

Bayesian estimation for the type-I hybrid xgamma distribution using asymmetric loss function

This article proposes the Bayes estimation of the parameter and reliability function for xgamma distribution in the presence of type-I hybrid censored observations. The Bayes estimate of the parameter has been obtained by assuming informative and non-informative priors using general entropy loss function. Obviously, censoring adds difficulties in estimation procedure; hence the Bayes estimators computed with type-I hybrid censored observation under the mentioned prior often do not assume any standard form. Therefore, Bayes estimates are computed using Tierney-Kadane approximation and Markov Chain Monte Carlo numerical technique. Further, different interval estimates namely asymptotic confidence interval, bootstrap confidence interval and highest posterior density interval along with the width of the interval and coverage probability are also discussed. The maximum likelihood estimate for the same has also been computed using non- linear maximization iterative procedure and compared with corresponding Bayes estimates using Monte Carlo simulations. The comparison of the estimators are made in terms of average loss over whole sample space and corresponding length of the interval. lastly, one medical data set has been considered for the real application of the proposed study.

Abstract P405: Low Social Support Mediates the Effect of Marital Stress on 12-month Cardiac-Specific Quality of Life in Young Adults With Acute Myocardial Infarction

Background: Marital stress is associated with worse cardiac outcomes in young adults (≤55 years) with acute myocardial infarction (AMI), but whether psychosocial factors mediate this association remains largely unknown. We conducted a mediation analysis to investigate whether marital stress worsened quality of life (QoL) after AMI by increasing the likelihood of depression or low social support. Methods: There were 1,037 married/partnered AMI survivors aged 18-55 years enrolled in the VIRGO (Variation in Recovery: Role of Gender on Outcomes of Young AMI Patients) study (2008-2012) (67% women, mean age 47 years). Marital stress was measured by the Stockholm Marital Stress Scale at 1 month post-AMI; categorized as absent/moderate or severe. Depression (Patient Health Questionnaire-9 score ≥10), low social support (ENRICHD Social Support Instrument score ≤3 on ≥2 items and total score ≤18), and cardiac-specific QoL (Seattle Angina Questionnaire) were assessed at 1 year post-AMI. Natural direct and indirect effects of marital stress, depression, and low social support on QoL were estimated by causal mediation analysis with bias-corrected bootstrapped confidence intervals. Baseline QoL, sex, age, race, and socioeconomic factors (education, income, employment, and insurance status) were entered as covariates in all models. Results: There was a statistically significant direct effect from severe marital stress to lower 1-year cardiac-specific QoL after adjusting for covariates ( Figure ). Low social support and depression mediated 14.7% and 11.1% of the total relationship between marital stress and QoL, respectively; however, only the mediating effect through low social support was statistically significant. Conclusion: Marital stress was significantly associated with worse 1-year cardiac-specific QoL, and this effect was partially mediated by low social support. Interventions to decrease marital stress that also screen and provide resources for social support may help to improve AMI outcomes.

Confidence Intervals by Bootstrapping Approach: A Significance Review

A confidence interval is an interval estimate of a parameter of a population calculated from a sample drawn from the population. Bootstrapping method, which involves producing several new data sets that are resampled from the original data in order to estimate parameter for each newly created data set, allowing an empirical distribution for the parameter to be estimated. Since certain statistics are harder to estimate, confidence intervals are rarely employed. Several statistics might necessitate multi-step formulas assuming that are impractical for calculating confidence intervals. This paper reviews research on the concept of bootstrapping and bootstrap confidence interval. The current narrative analysis was developed to answer the main research question: (1) What is the concept of the bootstrap method and bootstrap confidence interval? (2) What are the methods of bootstrapping to obtain confidence interval? This study has found general bootstrap method idea, various techniques of bootstrap methods, its advantages and disadvantages, and its limitations. There are normal interval method, percentile bootstrap method, basic method, first-order normal approximation method, bias-corrected bootstrap, accelerated bias-corrected bootstrap and bootstrap-t method. This study concludes that the advantages of using bootstrap CI is that it does not require any assumptions about the shape of distribution and universality of the approach. Bootstrapping is a computer-intensive statistical technique that relies significantly on modern high-speed digital computers to do massive computations.

Estimation of stress‐strength reliability for beta log Weibull distribution using progressive first failure censored samples

AbstractIn this paper, we investigate classical and Bayes estimates for stress‐strength reliability (SSR) based on two independent progressive first failure censored samples from beta log Weibull distributions with different shape and scale parameters. The maximum likelihood estimation of SSR and its asymptotic confidence interval are obtained. Bayes estimates of SSR are derived under two different loss functions using the Tierney–Kadane's approximation algorithm. In addition, two different bootstrap confidence intervals are provided and the highest posterior density credible interval of SSR is also constructed by applying Monte Carlo Markov chain method. The performances of different point and interval estimates are assessed using the Monte Carlo simulation. Finally, three real data sets are analyzed for illustration purposes.

Impact of Bias Correction of the Least Squares Estimation on Bootstrap Confidence Intervals for Bifurcating Autoregressive Models

The least squares (LS) estimator of the autoregressive coefficient in the bifurcating autoregressive (BAR) model was recently shown to suffer from substantial bias, especially for small to moderate samples. This study investigates the impact of the bias in the LS estimator on the behavior of various types of bootstrap confidence intervals for the autoregressive coefficient and introduces methods for constructing bias-corrected bootstrap confidence intervals. We first describe several bootstrap confidence interval procedures for the autoregressive coefficient of the BAR model and present their bias-corrected versions. The behavior of uncorrected and corrected confidence interval procedures is studied empirically through extensive Monte Carlo simulations and two real cell lineage data applications. The empirical results show that the bias in the LS estimator can have a significant negative impact on the behavior of bootstrap confidence intervals and that bias correction can significantly improve the performance of bootstrap confidence intervals in terms of coverage, width, and symmetry.

Wildfire-induced shifts in groundwater discharge to streams identified with paired air and stream water temperature analyses

Within the western United States, increasingly severe and frequent wildfires may alter the magnitude, timing, and quality of water exported from burned areas by streams. Post-fire hydrologic studies often focus on peak stream flow responses to shifts in runoff generation or on annual streamflow yield response to changes in evapotranspiration following fire. However, the magnitude and duration of wildfire effects on groundwater recharge, changes in subsurface routing, and consequences for stream low flows sourced predominately by baseflow are poorly understood. Here, we demonstrate an approach using the amplitude and phase of paired annual air and stream water temperature signals to broadly identify changes in watershed subsurface flow contributions after fire. Watersheds were classified using pre-fire temperature data, as having air-coupled (i.e., reduced apparent groundwater signature), deep groundwater, or shallow groundwater stream temperature signals. Changes in pre- and post-fire paired air and stream water temperature metrics were compared for locations (n = 17) spanning a large range of physiographic and climatic conditions across the western United States. Pre- and post-fire comparisons were computed by quantile using bootstrapped confidence intervals (ci = 95), as well as in aggregate using Kruskal-Wallis and post-hoc Dunn tests. Statistical comparisons of pre- and post-fire temperature metrics suggest that overall, watersheds classified as having minimal groundwater influence are the most likely to experience fire-induced subsurface hydrologic change. More specifically, watersheds classified as having air-coupled or shallow groundwater signals experienced increases in the magnitude of groundwater discharge, with more stable annual thermal regimes post-fire that are less-coupled to ambient air temperature. These findings form the basis of a conceptual framework for watershed resistance to subsurface hydrologic change following fire that can be broadly applied as a first approximation for water management, impacts on aquatic habitat, and post-wildfire response planning.

Impact of a Community Empowerment Intervention on Sexually Transmitted Infections Among Female Sex Workers in Baltimore, Maryland.

Female sex workers ( FSWs) are vulnerable to a number of health issues, but often delay seeking health care due to structural barriers. Multiservice drop-in centers have been shown to increase FSW access to health services globally, but their impact on FSW in the United States is lacking. This study seeks to evaluate the effect of a community-level empowerment intervention (the multiservice drop-in SPARC center) on cumulative sexually transmitted infections (STIs) among FSW in a city in the United States. Between September 2017 and January 2019, 385 FSWs were recruited in Baltimore. Participants from areas served by SPARC were recruited to the intervention; other areas of Baltimore were the control. Follow-up occurred at 6, 12, and 18 months. The primary outcome is cumulative STI (ie, positive gonorrhea or chlamydia test at any follow-up). We tested effect modification by condomless sex with paying clients reported at baseline. Logistic regressions with propensity score weighting were used to estimate intervention effect, accounting for loss to follow-up, with bootstrap confidence intervals. Participants completed 713 follow-up study visits (73%, 70%, 64% retention at 6, 12, and 18 months, respectively). Baseline STI prevalence was 28% and cumulative STI prevalence across follow-ups was 26%; these both did not differ between control and intervention communities in bivariate analyses. After adjusting for covariates, FSW in the intervention had a borderline-significant decrease in odds of cumulative STI compared with control (odds ratio, 0.61, P = 0.09 ) . There was evidence of effect modification by baseline condomless sex, such that FSW in the intervention who reported condomless sex had lower odds of cumulative STI compared with FSW in the control community who also reported baseline condomless sex (odds ratio, 0.29; P = 0.04). Results demonstrate the value of a low-barrier, multiservice model on reducing STIs among the highest-risk FSW.Clinical Trial Number: NCT04413591.