Command Suite Research Articles

Data management and techniques for best–worst discrete choice experiments

In this article, we present software that is suitable for use with Stata’s choice modeling suite of commands, which begin with cm. Within the context of choice models, we focus on best–worst data. In such data, respondents are presented a set of choices and are required to select a best and a worst choice from among the alternatives. Optionally, respondents may indicate an opt-out choice, in which no best or worst choice exists in the choice set. Such data are simplified versions of experiments in which respondents rank all the choices. Once best–worst data are collected, there are specific types of data expansions that analysts use to take advantage of both explicit and implicit information. The commands described in this article support data expansion and model estimation.

Adverse effects of immunotherapies for multiple sclerosis: a network meta-analysis.

Adverse effects of immunotherapies for multiple sclerosis: a network meta-analysis.

Hdps: A suite of commands for applying high-dimensional propensity-score approaches

Large healthcare databases are increasingly used for research investigating the effects of medications. However, a key challenge is capturing hard-to-measure concepts (often relating to frailty and disease severity) that can be crucial for successful confounder adjustment. The high-dimensional propensity score has been proposed as a data-driven method to improve confounder adjustment within healthcare databases and was developed in the context of administrative claims databases. We present hdps, a suite of commands implementing this approach in Stata that assesses the prevalence of codes, generates high-dimensional propensity-score covariates, performs variable selection, and provides investigators with graphical tools for inspecting the properties of selected covariates.

Finite mixture models for linked survey and administrative data: Estimation and postestimation

Researchers use finite mixture models to analyze linked survey and administrative data on labor earnings, while also accounting for various types of measurement error in each data source. Different combinations of error-ridden and error-free observations characterize latent classes. Latent class probabilities depend on the probabilities of the different types of error. We introduce a suite of commands to fit finite mixture models to linked survey-administrative data: there is a general model and seven simpler variants. We also provide postestimation commands for assessment of reliability, marginal effects, data simulation, and prediction of hybrid variables that combine information from both data sources about the outcome of interest. Our commands can also be used to study measurement errors in other variables besides labor earnings.

Reproductive health in humanitarian settings in Lebanon and Iraq: results from four cross-sectional studies, 2014\u20132015

BackgroundReproductive health is an important component of humanitarian response. Displaced women need access to family planning, antenatal care, and the presence of a skilled birth attendant at delivery. Since the beginning of the Syrian conflict in 2011, Lebanon and Iraq have been hosting large numbers of refugees, thereby straining local capacities to provide these services. In order to identify salient health needs, Médecins Sans Frontières conducted a survey in several sites hosting refugees and internally displaced persons across the region. Here we describe the reproductive health profile of Syrian refugees, Iraqi displaced persons, and vulnerable Lebanese and their use of services.MethodsWe conducted four cross-sectional surveys in 2014–2015 in two sites in Lebanon and two sites in Iraq. Depending on the site, two-stage cluster sampling or systematic sampling was intended, but non-probability methods were employed at the second stage due to implementation challenges. We collected information on overall health (including reproductive health) and demographic information from heads of households on the basis of a standardized questionnaire. Pearson chi-square tests were used to compare proportions, and generalized linear models were used to calculate odds ratios with regard to risk factors. All analyses were performed using the survey suite of commands in Stata version 14.1.ResultsA total of 23,604 individuals were surveyed, including 5925 women of childbearing age. Overall, it was reported that 7.5% of women were currently pregnant and 12.8% had given birth within the previous 12 months. It was reported that pregnancy was unplanned for 57% of currently pregnant women and 66.7% of women who had delivered in the previous year. A slight majority of women from both groups had accessed antenatal care at least once. Amongst women who had delivered in the previous year, 84.5% had done so with a skilled birth attendant and 22.1% had had a cesarean section. Location and head of household education were predictors of unplanned pregnancy in multivariable analysis. Head of household education was also significantly associated with higher uptake of antenatal care.ConclusionsConsidering the large number of pregnant women and women having recently delivered in these settings, addressing their sexual and reproductive health needs emerges as a crucial aspect of humanitarian response. This study identified unmet needs for family planning and high cesarean section rates at all sites, suggesting both lack of access to certain services (contraception, antenatal care), but also over-recourse to cesarean section. These specific challenges can impact directly on maternal and child health and need today to be kept high on the humanitarian agenda.

Cointegration Testing and Dynamic Simulations of Autoregressive Distributed Lag Models

In this article, we introduce dynamac, a suite of commands designed to assist users in modeling and visualizing the effects of autoregressive distributed lag models and in testing for cointegration. We discuss the bounds cointegration test proposed by Pesaran, Shin, and Smith (2001, Journal of Applied Econometrics 16: 289–326), which we have adapted into a command. Because the resulting models can be dynamically complex, we follow the advice of Philips (2018, American Journal of Political Science 62: 230–244) by introducing a flexible command designed to dynamically simulate and plot a variety of types of autoregressive distributed lag models, including error-correction models.

How Modeless Editing Came To Be

Several word processing and desktop publishing innovations grew out of 1970s research by Larry Tesler and his collaborators at Stanford and Xerox PARC. One prototype was PUB, a markup language for print publishing. Another was Gypsy, a text editor with a cut/copy/paste command suite that became the de facto standard in desktop publishing and most other applications.

Attrition Diagrams for Clinical Trials and Meta-analyses in Stata

In this article, we present attrition, a suite of commands to simplify the maintenance and documentation of implemented exclusion criteria and attrition conditions using standard Stata facilities and to generate an attrition diagram. attrition can be used, both from the command line and in do-files, to keep the diagram up to date with the analysis it documents. Six subcommands (set, exclude, count, tab, list, graph) allow the diagram to be constructed in a versatile way.

Spatial Panel-data Models Using Stata

xsmle is a new user-written command for spatial analysis. We consider the quasi–maximum likelihood estimation of a wide set of both fixed- and random-effects spatial models for balanced panel data. xsmle allows users to handle unbalanced panels using its full compatibility with the mi suite of commands, use spatial weight matrices in the form of both Stata matrices and spmat objects, compute direct, indirect, and total marginal effects and related standard errors for linear (in variables) specifications, and exploit a wide range of postestimation features, including the panel-data case predictors of Kelejian and Prucha (2007, Regional Science and Urban Economics 37: 363–374). Moreover, xsmle allows the use of margins to compute total marginal effects in the presence of nonlinear specifications obtained using factor variables. In this article, we describe the command and all of its functionalities using simulated and real data.

Spatial Panel Data Models Using Stata

xsmle is a new command for spatial analysis using Stata. We consider the quasi-maximum likelihood estimation of a wide set of both fi xed- and random- eff ects spatial models for balanced panel data. Of special note is that xsmle allows to handle unbalanced panels thanks to its full compatibility with the mi suite of commands, to use spatial weight matrices in the form of both Stata matrices and spmat objects, to compute direct, indirect and total e ffects according to the procedure outlined in LeSage and Pace (2009), and to exploit a wide range of postestimation features, extending to the panel data case the predictors proposed by Kelejian and Prucha (2007). This paper describes the command and all its functionalities using both simulated and real data.

Visualizing Assumptions and Results in Network Meta-analysis: The Network Graphs Package

Network meta-analysis has been established in recent years as a particularly useful evidence synthesis tool. However, it is still challenging to develop understandable and concise ways to present data, assumptions, and results from network meta-analysis to inform decision making and evaluate the credibility of the results. In this article, we provide a suite of commands with graphical tools to facilitate the understanding of data, the evaluation of assumptions, and the interpretation of findings from network meta-analysis.

O65. Is there evidence to inform antihypertensive prescribing in pregnancy complicated by chronic hypertension: A systematic review

Introduction Chronic hypertension (CHT) complicates 1–5% of all pregnancies. The importance of optimising blood pressure control has been highlighted by the results of the Control of hypertension in pregnancy study. The frequency of severe hypertension in pregnancy directly affects maternal morbidity and mortality. Effective anti-hypertensive treatments (AHTs) for CHT in pregnancy that reduce this risk need to be identified. Objectives To perform a systematic review of randomised controlled trials (RCTs) in women with CHT in pregnancy to determine the incidence of severe hypertension with AHT use (vs. none) and by drug class, together with incidence of adverse maternal and fetal/neonatal outcomes. Method Medline (via OVID), Embase (via OVID) and the Cochrane Trials Register were searched from their earliest entries until 30.4.15. All RCTs evaluating AHTs for CHT in pregnancy were included. Additional data were found from references of papers identified from the original searches. Two review authors (LW and FCR) extracted and tabulated relevant data. Data were analysed in Stata 13.1 (StataCorp, College Station, Texas), using the metan suite of commands (Higgins et al. 2003). Where there was significant evidence of heterogeneity, the DerSimonian & Laird (1986) method was used. Results 10 RCTs of AHT vs other AHT or placebo/no AHT were identified for a pre-defined CHT cohort and an additional 5 RCTs reported outcomes for a CHT subgroup; all data were from studies completed prior to 2000. A clinically important reduction in the incidence of severe hypertension was seen with AHT use vs no AHT (4 studies, 244 women; risk ratio (RR) 0.37; 95% confidence interval (CI) 0.22 to 0.64; I 2 0.0%). Additionally two small studies reported incidence of severe hypertension when comparing β -blockers with methyldopa, but results are ambiguous (86 women; RR 0.85, CI 0.5 to 1.37; I 2 66.1%). There was no clear difference in the incidence of superimposed pre-eclampsia for women on AHTs vs no AHTs (6 studies, 525 women; RR 0.77, CI 0.50 to 1.18; I 2 37.3%). No significant difference in birthweight (g) was demonstrated with AHT treatment of CHT in pregnancy ( 7 studies, 731 women; weighted mean difference − 78.23; CI − 164.59 to 8.13; I 2 59.9%). Conclusion A considerable paucity of data exists from RCTs to guide AHT prescribing for CHT in pregnancy. Given the increasing incidence of CHT in pregnancy and that AHTs reduce the incidence of severe hypertension, head-to-head RCTs of the commonly used AHTs are urgently required to inform prescribing.

Collecting and Organizing Stata Graphs

Stata includes a powerful set of tools for constructing a wide array of graphs. Stata's graphing capabilities are well suited for describing exploratory or preliminary data analyses as well as producing publication-quality graphics. Currently, however, Stata does not have a built-in suite of commands for constructing various types of files (for example, HTML, TEX, or RTF files) to display multiple graphs. Such files can be invaluable for organizing and facilitating the interpretation of the numerous graphs needed throughout an analysis or in the final stage of a project. In this article, we provide an overview of two commands, graphsto and graphout, designed to organize and process multiple graphs across various file types.

Simulation-Based Sample-Size Calculation for Designing New Clinical Trials and Diagnostic Test Accuracy Studies to Update an Existing Meta-Analysis

In this article, we describe a suite of commands that enable the user to estimate the probability that the conclusions of a meta-analysis will change with the inclusion of a new study, as described previously by Sutton et al. (2007, Statistics in Medicine 26: 2479–2500). Using the metasim command, we take a simulation approach to estimating the effects in future studies. The method assumes that the effect sizes of future studies are consistent with those observed previously, as represented by the current meta-analysis. Two-arm randomized controlled trials and studies of diagnostic test accuracy are considered for a variety of outcome measures. Calculations are possible under both fixed- and random-effects assumptions, and several approaches to inference, including statistical significance and limits of clinical significance, are possible. Calculations for specific sample sizes can be conducted (by using metapow). Plots, akin to traditional power curves, can be produced (by using metapowplot) to indicate the probability that a new study will change inferences for a range of sample sizes. Finally, plots of the simulation results are overlaid on extended funnel plots by using extfunnel, described in Crowther, Langan, and Sutton (2012, Stata Journal 12: 605–622), which can help to intuitively explain the results of such calculations of sample size. We hope the command will be useful to trialists who want to assess the potential impact new trials will have on the overall evidence base and to meta-analysts who want to assess the robustness of the current meta-analysis to the inclusion of future data.

A Lightweight Workbench for Database Benchmarking, Experimentation, and Implementation

We have developed a platform, called Cyclone Database Implementation Workbench (CyDIW), that can be used to implement new database prototypes, use existing command-based systems, and conduct experiments. The workbench allows seamless integration of multiple systems and provides useful services. To support database implementation page-based storage and buffer managers are built-in. A scripting language for batches of commands is included. Experiments are encapsulated as batches of commands on multiple systems. A simple and easy to use GUI is available that acts as an editor and a launchpad for execution of batches of commands. Emphasis in CyDIW is on simplifying the logistics surrounding setting up experiments that are comprehensive and self-contained. The benchmarking services in CyDIW can be used for lightweight benchmarking, where a benchmark consisting of a data set and a suite of commands is given. A benchmarking experiment collects performance statistics from multiple systems based on varying parameters and plots benchmarking results without leaving the GUI. Setup for the system is easy. All configuration settings are recorded in XML documents that are highly portable and readily visible. Once installed, batches representing experiments can be exchanged as text files and executed on CyDIW on any computer.

Prevalência de idosos restritos ao domicílio em região metropolitana de Belo Horizonte (Minas Gerais, Brasil)

The aim of this article is to estimate the prevalence and the socio-demographic and health profile of housebound elderly people registered at a Family Health Unit in the urban region of Belo Horizonte (Minas Gerais, Brazil). A household survey was conducted between May and July 2006 with 275 elderly people selected via cluster sampling. The svy suite of commands in Stata 9.0 was used to deal adequately with the cluster sample structure and to allow the incorporation of fractions of expansion in the analyses. Among the 275 elderly, 22.4% (IC95%: 14.7; 32.4) were restricted to their homes. The prevalence of this condition was greater among women, people over 80 and suspected of suffering from cognitive impairment (p-valor < 0.05). The majority of housebound people had incomes below the minimum wage, reported history of falls, depression and indicated physical disorders as the cause of the restriction. The large contingent of low-income housebound elderly with several health problems, reinforces the need for incorporation of proposals for promotion and vigilance of the health of the elderly, which extend beyond the boundaries of the healthcare units.

A Suite of Commands for Fitting the Skew-normal and Skew-t models

A Suite of Commands for Fitting the Skew-normal and Skew-t models

A Suite of Commands for Fitting the Skew-normal and Skew-t models

Nonnormal data arise often in practice, prompting the development of flexible distributions for modeling such situations. In this article, we describe two multivariate distributions, the skew-normal and the skew- t, which can be used to model skewed and heavy-tailed continuous data. We then discuss some inferential issues that can arise when fitting these distributions to real data. We also consider the use of these distributions in a regression setting for more flexible parametric modeling of the conditional distribution given other predictors. We present commands for fitting univariate and multivariate skew-normal and skew- t regressions in Stata (skewnreg, skewtreg, mskewnreg, and mskewtreg) as well as some postestimation features (predict and skewrplot). We also demonstrate the use of the commands for the analysis of the famous Australian Institute of Sport data and U.S. precipitation data.

Risk Management and Quality Assurance in the Emergency Room

A need existed to rapidly document the quality of care in an emergency room (ER). To fill this need, a computer-assisted, timely, clinically focused review of ER care was established in an Army hospital. Specialists were used as primary reviewers. Providers received rapid feedback. The command suite was provided with timely and specific quality assurance and risk management data, and patient complaints decreased.