Models For Binary Outcomes Research Articles

Effect Modification by Obesity on Nonsurgical Periodontal Treatment.

Obesity is associated with increased periodontal disease prevalence and incidence. This retrospective cohort study examined whether body mass index (BMI) is an effect modifier of periodontal treatment outcomes in patients attending an urban dental school clinic. Data were extracted from electronic health records of 344 patients at a large urban dental school clinic who had at least 1 tooth with a probing pocket depth (PD) ≥5 mm at baseline and who subsequently received nonsurgical periodontal treatment. BMI was computed from self-reported weight and height and categorized as obese (≥30 kg/m2), overweight (25-29.9 kg/m2), or healthy (18-24.9 kg/m2). The primary treatment outcome of interest was defined as having no teeth with PD ≥5 mm in a quadrant on follow-up after nonsurgical periodontal therapy. That outcome was considered to represent treatment success in this study. Secondary outcomes included changes in mean PD and clinical attachment loss (CAL). Analyses included 879 treated quadrants among 344 patients (185 males, 159 females; mean age 49 ± 12 y at baseline; mean posttreatment follow-up of 6 ± 2 mo). Clinical outcomes in patients who were overweight or obese were compared to healthy-weight patients using generalized linear models for binary or continuous outcomes, accounting for clustering within patients. Covariates were age, gender, tobacco use, history of diabetes, insurance type, and number of baseline sites ≥5 mm. Obesity was associated with a significantly lower likelihood of successful nonsurgical treatment (odds ratio = 0.47; 95% confidence interval, 0.25-0.88) than healthy weight. Being overweight was not associated with treatment success. Posttreatment reductions in the percentage of sites with pockets ≥5 mm and CAL ≥5 mm were greater in patients with healthy weight as compared to those either overweight or obese. However, posttreatment changes in mean PD and CAL did not differ among the BMI groups. Obesity adversely modifies the effectiveness of nonsurgical periodontal treatment among dental school clinic patients. The results of this study may be used by dental providers to better understand and manage periodontal therapy in patients with obesity. Furthermore, patients will be better informed about their therapeutic options and outcome success.

Development of a software service for stock price forecasting based on sentiment analysis and autoregressive models

This paper addresses the critical need for efficient market analysis tools in the era of big data and artificial intelligence. We present a novel software service that integrates real-time news sentiment analysis with stock market prediction, enhancing the accuracy and speed of trading decisions. The system employs APIs for data collection, FinBERT for sentiment analysis, and MongoDB for data storage, overcoming limitations of existing platforms like Investing.com and MarketWatch. Our methodology combines sentiment analysis with autoregressive models to forecast stock prices for 11 major companies. The experiment utilized 141 observations, applying multiple regression and binary outcome models. Results demonstrate that investor sentiment significantly affects stock prices for 2 out of 11 companies, with Meta showing a 70 % determination coefficient in price direction changes based on sentiment. The study reveals that incorporating both quantitative (previous stock prices) and qualitative (sentiment) data improves forecast accuracy for certain stocks. This research contributes to the field of financial analytics by providing a more comprehensive approach to stock price prediction, integrating ML models and data analytics to support informed decision-making in dynamic financial markets.

Nā Kānaka Maoli ma nā 'Āina 'Ē: Exploring Place of Residency as a Native Hawaiian Health Predictor During the COVID-19 Pandemic.

Little is known about the impacts of living in diaspora from the Hawaiian Islands on Native Hawaiian health. To address this, the authors conducted an exploratory analysis using cross-sectional data from the 2021 Native American COVID-19 Alliance Needs Assessment. A total of 1418 participants identified as Native Hawaiian (alone or in any combination), of which 1222 reported residency in the continental US and 196 in Hawai'i. Residency status in the continental US vs Hawai'i was evaluated as a predictor of survey outcomes using likelihood ratio tests on linear and logistic regression models for linear and binary outcomes, respectively. Results showed that NH residency in the continental US was significantly associated with increased odds of reporting fair or poor self-rated health; increased odds for screening positive for anxiety, depression, and suicidality; and increased odds of health insurance loss (P's < .05). Residency in the continent was also associated with lower odds of reporting a diagnosed chronic health condition (P < .05). Residency in the continental US had no observed effect on the odds that participants engaged cultural activities or cultural coping strategies. These results support the role of place of residency as an important Native Hawaiian health predictor during and beyond the COVID-19 pandemic.

How to develop, validate, and update clinical prediction models using multinomial logistic regression

ObjectivesMulticategory prediction models (MPMs) can be used in health care when the primary outcome of interest has more than two categories. The application of MPMs is scarce, possibly due to added methodological complexities compared to binary outcome models. We provide a guide of how to develop, validate, and update clinical prediction models based on multinomial logistic regression. Study Design and SettingWe present guidance and recommendations based on recent methodological literature, illustrated by a previously developed and validated MPM for treatment outcomes in rheumatoid arthritis. Prediction models using multinomial logistic regression can be developed for nominal outcomes, but also for ordinal outcomes. This article is intended to supplement existing general guidance on prediction model research. ResultsThis guide is split into three parts: 1) outcome definition and variable selection, 2) model development, and 3) model evaluation (including performance assessment, internal and external validation, and model recalibration). We outline how to evaluate and interpret the predictive performance of MPMs. R code is provided. ConclusionWe recommend the application of MPMs in clinical settings where the prediction of a multicategory outcome is of interest. Future methodological research could focus on MPM-specific considerations for variable selection and sample size criteria for external validation.

Average treatment effects on binary outcomes with stochastic covariates.

When evaluating the effect of psychological treatments on a dichotomous outcome variable in a randomized controlled trial (RCT), covariate adjustment using logistic regression models is often applied. In the presence of covariates, average marginal effects (AMEs) are often preferred over odds ratios, as AMEs yield a clearer substantive and causal interpretation. However, standard error computation of AMEs neglects sampling-based uncertainty (i.e., covariate values are assumed to be fixed over repeated sampling), which leads to underestimation of AME standard errors in other generalized linear models (e.g., Poisson regression). In this paper, we present and compare approaches allowing for stochastic (i.e., randomly sampled) covariates in models for binary outcomes. In a simulation study, we investigated the quality of the AME and stochastic-covariate approaches focusing on statistical inference in finite samples. Our results indicate that the fixed-covariate approach provides reliable results only if there is no heterogeneity in interindividual treatment effects (i.e., presence of treatment-covariate interactions), while the stochastic-covariate approaches are preferable in all other simulated conditions. We provide an illustrative example from clinical psychology investigating the effect of a cognitive bias modification training on post-traumatic stress disorder while accounting for patients' anxiety using an RCT.

Efficacy and Safety of Allopurinol and Febuxostat in Patients With Gout and CKD: Subgroup Analysis of the STOP Gout Trial

Rationale & ObjectiveWe conducted a prespecified examination of the efficacy and safety of allopurinol and febuxostat administered using a treat-to-target strategy in trial participants with chronic kidney disease (CKD). Study DesignPrespecified sub cohort analysis of a randomized controlled trial. Setting& Participants: A sub study of the STOP Gout trial in participants with CKD. CKD was defined as an eGFR 30-59 mL/min/1.73 m2 at baseline. ExposureTrial participants with CKD and gout and serum urate (sUA) concentration ≥6.8 mg/dL were randomized 1:1 to receive allopurinol or febuxostat. Urate lowering therapy (ULT) was titrated during weeks 0-24 to achieve a goal sUA of <6.0 mg/dl (<5.0 mg/dl with tophi) (Phase 1) and maintained during weeks 25-48 (Phase 2). Gout flare was assessed between weeks 49-72 (Phase 3). OutcomeGout flare between weeks 49-72 (Phase 3) was the primary outcome. Secondary outcomes included sUA goal achievement and ULT dosing at end of Phase 2, and serious adverse events (SAEs). Analytical ApproachOutcomes between treatment groups were compared using logistic regression models for binary outcomes, and Poisson regression for flare rates. Multivariable models were subsequently used, adjusting for factors identified to be imbalanced by treatment arm. Results351 of 940 participants (37.3%) had CKD; 277 were assessed for the primary outcome. Fewer patients randomized to allopurinol had a flare during phase 3 (32% vs 45%; p=0.02) despite similar attainment of sUA goal (79% vs. 81%; p=0.6) by the end of Phase 2. Acute kidney injury (AKI) was more common in participants with stage 3 CKD randomized to allopurinol compared to febuxostat. LimitationsLimited power to assess infrequent safety events, largely male, older population. ConclusionsAllopurinol and febuxostat are similarly efficacious and well-tolerated in the treatment of gout in people with CKD when used in a treat-to-target regimen.

Estimation of average treatment effects for massively unbalanced binary outcomes

. The MLE of the ATE in the logit model for binary outcomes may have a significant second-order bias if the event has a low probability, which is the case we focus on in this article. We derive the second-order bias of the logit ATE estimator, and we propose a bias-corrected estimator of the ATE. We also propose a variation on the logit model with parameters that are elasticities. Finally, we propose a computational trick that avoids numerical instability in the case of estimation for rare events.

Improving Sleep Using Mentored Behavioral and Environmental Restructuring (SLUMBER)

ObjectivesTo evaluate the impact of a mentoring program to encourage staff-delivered sleep-promoting strategies on sleep, function, depression, and anxiety among skilled nursing facility (SNF) residents. DesignModified stepped-wedge unit-level intervention. Setting and ParticipantsSeventy-two residents (mean age 75 ± 15 years; 61.5% female, 41% non-Hispanic white, 35% Black, 20% Hispanic, 3% Asian) of 2 New York City urban SNFs. MethodsExpert mentors provided SNF staff webinars, in-person workshops, and weekly sleep pearls via text messaging. Resident data were collected at baseline, post-intervention (V1), and 3-month follow-up (V2), including wrist actigraphy, resident behavioral observations, Pittsburgh Sleep Quality Index (PSQI), Patient Health Questionnaire-9 (PHQ-9) depression scale, Brief Anxiety and Depression Scale (BADS), Brief Cognitive Assessment Tool (BCAT), and select Minimum Data Set 3.0 (MDS 3.0) measures. Linear mixed models were fit for continuous outcomes and mixed-effects logistic models for binary outcomes. Outcomes were modeled as a function of time. Planned contrasts compared baseline to V1 and V2. ResultsThere was significant improvement in PSQI scores from baseline to V1 (P = .009), and from baseline to V2 (P = .008). Other significant changes between baseline and V1 included decreased depression (PHQ-9) (P = .028), increased daytime observed out of bed (P ≤ .001), and increased daytime observed being awake (P < .001). At V2 (vs baseline) being observed out of bed decreased (P < .001). Daytime sleeping by actigraphy increased from baseline to V1 (P = .004), but not V2. MDS 3.0 activities of daily living and pain showed improvements by the second quarter following implementation of SLUMBER (P's ≤ .034). There were no significant changes in BADS or BCAT between baseline and V1 or V2. Conclusions and ImplicationsSNF residents had improvements in sleep quality and depression with intervention, but improvements were not sustained at 3-month follow-up. The COVID-19 pandemic led to premature study termination, so full impacts remain unknown.

A Bayesian Bernoulli-Exponential joint model for binary longitudinal outcomes and informative time with applications to bladder cancer recurrence data

BackgroundA variety of methods exist for the analysis of longitudinal data, many of which are characterized with the assumption of fixed visit time points for study individuals. This, however is not always a tenable assumption. Phenomenon that alter subject visit patterns such as adverse events due to investigative treatment administered, travel or any other emergencies may result in unbalanced data and varying individual visit time points. Visit times can be considered informative, because subsequent or current subject outcomes can change or be adapted due to previous subject outcomes.MethodsIn this paper, a Bayesian Bernoulli-Exponential model for analyzing joint binary outcomes and exponentially distributed informative visit times is developed. Via statistical simulations, the influence of controlled variations in visit patterns, prior and sample size schemes on model performance is assessed. As an application example, the proposed model is applied to a Bladder Cancer Recurrence data.Results and conclusionsResults from the simulation analysis indicated that the Bayesian Bernoulli-Exponential joint model converged in stationarity, and performed relatively better for small to medium sample size scenarios with less varying time sequences regardless of the choice of prior. In larger samples, the model performed better for less varying time sequences. This model’s application to the bladder cancer data showed a statistically significant effect of prior tumor recurrence on the probability of subsequent recurrences.

Unveiling challenges in Mendelian randomization for gene-environment interaction.

Gene-environment (GxE) interactions play a crucial role in understanding the complex etiology of various traits, but assessing them using observational data can be challenging due to unmeasured confounders for lifestyle and environmental risk factors. Mendelian randomization (MR) has emerged as a valuable method for assessing causal relationships based on observational data. This approach utilizes genetic variants as instrumental variables (IVs) with the aim of providing a valid statistical test and estimation of causal effects in the presence of unmeasured confounders. MR has gained substantial popularity in recent years largely due to the success of genome-wide association studies. Many methods have been developed for MR; however, limited work has been done on evaluating GxE interaction. In this paper, we focus on two primary IV approaches: the two-stage predictor substitution and the two-stage residual inclusion, and extend them to accommodate GxE interaction under both the linear and logistic regression models for continuous and binary outcomes, respectively. Comprehensive simulation study and analytical derivations reveal that resolving the linear regression model is relatively straightforward. In contrast, the logistic regression model presents a considerably more intricate challenge, which demands additional effort.

Individual-level probabilities and cluster-level proportions: Toward interpretable level 2 estimates in unconflated multilevel models for binary outcomes.

Individual-level probabilities and cluster-level proportions: Toward interpretable level 2 estimates in unconflated multilevel models for binary outcomes.

Supplemental Material for Individual-Level Probabilities and Cluster-Level Proportions: Toward Interpretable Level 2 Estimates in Unconflated Multilevel Models for Binary Outcomes

Supplemental Material for Individual-Level Probabilities and Cluster-Level Proportions: Toward Interpretable Level 2 Estimates in Unconflated Multilevel Models for Binary Outcomes

Multinomial Thompson sampling for rating scales and prior considerations for calibrating uncertainty

Bandit algorithms such as Thompson sampling (TS) have been put forth for decades as useful tools for conducting adaptively-randomised experiments. By skewing the allocation toward superior arms, they can substantially improve particular outcomes of interest for both participants and investigators. For example, they may use participants’ ratings for continuously optimising their experience with a program. However, most of the bandit and TS variants are based on either binary or continuous outcome models, leading to suboptimal performances in rating scale data. Guided by behavioural experiments we conducted online, we address this problem by introducing Multinomial-TS for rating scales. After assessing its improved empirical performance in unique optimal arm scenarios, we explore potential considerations (including prior’s role) for calibrating uncertainty and balancing arm allocation in scenarios with no unique optimal arms.

The role of socio-territorial conditions and distance to infrastructure for promoting recycling in Latin American cities: The case of Santiago de Chile

Convenient distances and higher density of drop-off points are expected to increase the probability of people participating in waste separation. However, although several analyses search for optimal distances between users and drop-off points, they tend to concentrate on European case studies. Few studies have analysed the actuallyexisting experiences regarding these factors in cities from the Global South. In this context, the authors provide an analysis on Santiago, the capital of Chile. A survey (N = 387) is generated to represent five neighbourhood clusters, and a logit binary outcome model is used to analyse the data. The research also considers territorial conditions based on the specific characteristics of the Latin American cities, particularly socioeconomic variables and location (i.e. centre versus periphery). The findings point towards a prominent role of educational levels, and an inverted U relationship between proximity to drop off points and the probability of recycling. Additionally, in the case of metals, results indicate that probability of recycling is higher both in the areas that concentrate the poor and in those that concentrate the rich. Such result is most likely associated to a contrasting relationship between recycling and socioeconomic variables: one linked to the need of generating income, and the other to privilege (i.e. conspicuous consumption and domestic workers).

Robit regression in Stata

Logistic and probit models are the most popular regression models for binary outcomes. A simple robust alternative is the robit model, which replaces the underlying normal distribution in the probit model with a Student’s t distribution. The heavier tails of the t distribution (compared with the normal distribution) mean that model outliers are less influential. Robit regression models can be fit as generalized linear models with the link function defined as the inverse cumulative t distribution function with a specified number of degrees of freedom; they have been advocated as being particularly suitable for estimating inverse-probability weights and propensity scoring more generally. Here we describe a new command, robit, that implements robit regression in Stata.

Sample size requirements are not being considered in studies developing prediction models for binary outcomes: a systematic review

BackgroundHaving an appropriate sample size is important when developing a clinical prediction model. We aimed to review how sample size is considered in studies developing a prediction model for a binary outcome.MethodsWe searched PubMed for studies published between 01/07/2020 and 30/07/2020 and reviewed the sample size calculations used to develop the prediction models. Using the available information, we calculated the minimum sample size that would be needed to estimate overall risk and minimise overfitting in each study and summarised the difference between the calculated and used sample size.ResultsA total of 119 studies were included, of which nine studies provided sample size justification (8%). The recommended minimum sample size could be calculated for 94 studies: 73% (95% CI: 63–82%) used sample sizes lower than required to estimate overall risk and minimise overfitting including 26% studies that used sample sizes lower than required to estimate overall risk only. A similar number of studies did not meet the ≥ 10EPV criteria (75%, 95% CI: 66–84%). The median deficit of the number of events used to develop a model was 75 [IQR: 234 lower to 7 higher]) which reduced to 63 if the total available data (before any data splitting) was used [IQR:225 lower to 7 higher]. Studies that met the minimum required sample size had a median c-statistic of 0.84 (IQR:0.80 to 0.9) and studies where the minimum sample size was not met had a median c-statistic of 0.83 (IQR: 0.75 to 0.9). Studies that met the ≥ 10 EPP criteria had a median c-statistic of 0.80 (IQR: 0.73 to 0.84).ConclusionsPrediction models are often developed with no sample size calculation, as a consequence many are too small to precisely estimate the overall risk. We encourage researchers to justify, perform and report sample size calculations when developing a prediction model.

Simultaneity in binary outcome models with an application to employment for couples

Two of Peter Schmidt’s many contributions to econometrics have been to introduce a simultaneous logit model for bivariate binary outcomes and to study estimation of dynamic linear fixed effects panel data models using short panels. In this paper, we study a dynamic panel data version of the bivariate model introduced in Schmidt and Strauss (Econometrica 43:745–755, 1975) that allows for lagged dependent variables and fixed effects as in Ahn and Schmidt (J Econom 68:5–27, 1995). We combine a conditional likelihood approach with a method of moments approach to obtain an estimation strategy for the resulting model. We apply this estimation strategy to a simple model for the intra-household relationship in employment. Our main conclusion is that the within-household dependence in employment differs significantly by the ethnicity composition of the couple even after one allows for unobserved household specific heterogeneity.

Leave-one-out cross-validation, penalization, and differential bias of some prediction model performance measures—a simulation study

BackgroundThe performance of models for binary outcomes can be described by measures such as the concordance statistic (c-statistic, area under the curve), the discrimination slope, or the Brier score. At internal validation, data resampling techniques, e.g., cross-validation, are frequently employed to correct for optimism in these model performance criteria. Especially with small samples or rare events, leave-one-out cross-validation is a popular choice.MethodsUsing simulations and a real data example, we compared the effect of different resampling techniques on the estimation of c-statistics, discrimination slopes, and Brier scores for three estimators of logistic regression models, including the maximum likelihood and two maximum penalized likelihood estimators.ResultsOur simulation study confirms earlier studies reporting that leave-one-out cross-validated c-statistics can be strongly biased towards zero. In addition, our study reveals that this bias is even more pronounced for model estimators shrinking estimated probabilities towards the observed event fraction, such as ridge regression. Leave-one-out cross-validation also provided pessimistic estimates of the discrimination slope but nearly unbiased estimates of the Brier score.ConclusionsWe recommend to use leave-pair-out cross-validation, fivefold cross-validation with repetitions, the enhanced or the .632+ bootstrap to estimate c-statistics, and leave-pair-out or fivefold cross-validation to estimate discrimination slopes.

A new pathway for considering trigger factors based on parallel-serial connection models and displaying the relationships of causal factors in low-probability events

BackgroundTo determine the effect size of observed factors considering trigger factors based on parallel-serial models and to explore how multiple factors can be related to the result of complex events for low-probability events with binary outcomes.MethodsA low-probability event with a true binary outcome can be explained by a trigger factor. The models were based on the parallel-serial connection of switches; causal factors, including trigger factors, were simplified as switches. Effect size values of an observed factor for an outcome were calculated as SAR = (Pe-Pn)/(Pe + Pn), where Pe and Pn represent percentages in the exposed and nonexposed groups, respectively, and SAR represents standardized absolute risk. The influence of trigger factors is eliminated by SAR. Actual data were collected to obtain a deeper understanding of the system.ResultsSAR values of < 0.25, 0.25–0.50, and > 0.50 indicate low, medium, and high effect sizes, respectively. The system of data visualization based on the parallel-serial connection model revealed that at least 7 predictors with SAR > 0.50, including a trigger factor, were needed to predict schizophrenia. The SAR of the HLADQB1*03 gene was 0.22 for schizophrenia.ConclusionsIt is likely that the trigger factors and observed factors had a cumulative effect, as indicated by the parallel-serial connection model for binary outcomes. SAR may allow better evaluation of the effect size of a factor in complex events by eliminating the influence of trigger factors. The efficiency and efficacy of observational research could be increased if we are able to clarify how multiple factors can be related to a result in a pragmatic manner.

A Bayesian (meta-)regression model for treatment effects on the risk difference scale.

In clinical settings, the absolute risk reduction due to treatment that can be expected in a particular patient is of key interest. However, logistic regression, the default regression model for trials with a binary outcome, produces estimates of the effect of treatment measured as a difference in log odds. We explored options to estimate treatment effects directly as a difference in risk, specifically in the network meta-analysis setting. We propose a novel Bayesian (meta-)regression model for binary outcomes on the additive risk scale. The model allows treatment effects, covariate effects, interactions and variance parameters to be estimated directly on the linear scale of clinical interest. We compared effect estimates from this model to (1) a previously proposed additive risk model by Warn, Thompson and Spiegelhalter ("WTS model") and (2) backtransforming the predictions from a logistic model to the natural scale after regression. The models were compared in a network meta-analysis of 20 hepatitis C trials, as well as in the analysis of simulated single trial settings. The resulting estimates diverged, in particular for small sample sizes or true risks close to 0% or 100%. Researchers should be aware that modelling untransformed risk can yield very different results from default logistic models. The treatment effect in participants with such extreme predicted risks weighed more heavily on the overall treatment effect estimate from our proposed model compared to the WTS model. In our network meta-analysis, this sensitivity of our proposed model was needed to detect all information in the data.