Longitudinal Analysis of Electronic Health Information to Identify Possible COVID-19 Sequelae.

Ongoing symptoms might follow acute COVID-19. Using electronic health information, we compared pre‒ and post‒COVID-19 diagnostic codes to identify symptoms that had higher encounter incidence in the post‒COVID-19 period as sequelae. This method can be used for hypothesis generation and ongoing monitoring of sequelae of COVID-19 and future emerging diseases.

Using SAS/STAT to Understand the NCI Joinpoint Regression Software: Testing for a Zero Slope Using Rates of Drug Overdose Deaths Involving Fentanyl, 2011-2016.

Background-The National Cancer Institute (NCI) Joinpoint regression software is a widely used software program for evaluating trends. In addition to producing model estimates for trend models, this software can search for changes in slope along the trend line. One component of the software, which tests whether line segment slopes are zero, is different from the usual t-test of zero slope that is used in linear models. This report will demonstrate this Joinpoint software procedure through replication using the SAS Institute's statistical software (that is, SAS) and discuss the implications of the different assumptions used by Joinpoint and a typical SAS model for the test of zero slope. Methods-First, Joinpoint's procedure for testing a zero slope is compared with a typical test of zero slope using SAS, and the assumptions behind both approaches are evaluated. Second, the test from the Joinpoint software is replicated in SAS using its PROC REG procedure and additional SAS programming. Trend analyses of rates of drug overdose deaths involving fentanyl from the general population and among females are used as examples. Results-In the evaluation of the trend of drug overdose deaths for the total population, Joinpoint produces a similar result to the linear model test in SAS. For the female subgroup, however, Joinpoint and SAS produce differing results for the test of zero slope. The replication of the Joinpoint test of zero slope using SAS demonstrates that Joinpoint's procedure is based on fewer degrees of freedom, which results in a larger standard error estimate. Conclusion-The Joinpoint approach accounts for the fact that the joinpoints are estimated and thus leads to a more conservative hypothesis test, particularly when the number of points in a trend analysis is small.

Late Conditions Diagnosed 1-4 Months Following an Initial Coronavirus Disease 2019 (COVID-19) Encounter: A Matched-Cohort Study Using Inpatient and Outpatient Administrative Data-United States, 1 March-30 June 2020.

BACKGROUND: Late sequelae of COVID-19 have been reported; however, few studies have investigated the time course or incidence of late new COVID-19-related health conditions (post-COVID conditions) after COVID-19 diagnosis. Studies distinguishing post-COVID conditions from late conditions caused by other etiologies are lacking. Using data from a large administrative all-payer database, we assessed type, association, and timing of post-COVID conditions following COVID-19 diagnosis. METHODS: Using the Premier Healthcare Database Special COVID-19 Release (release date, 20 October 2020) data, during March-June 2020, 27 589 inpatients and 46 857 outpatients diagnosed with COVID-19 (case-patients) were 1:1 matched with patients without COVID-19 through the 4-month follow-up period (control-patients) by using propensity score matching. In this matched-cohort study, adjusted ORs were calculated to assess for late conditions that were more common in case-patients than control-patients. Incidence proportion was calculated for conditions that were more common in case-patients than control-patients during 31-120 days following a COVID-19 encounter. RESULTS: During 31-120 days after an initial COVID-19 inpatient hospitalization, 7.0% of adults experienced ≥1 of 5 post-COVID conditions. Among adult outpatients with COVID-19, 7.7% experienced ≥1 of 10 post-COVID conditions. During 31-60 days after an initial outpatient encounter, adults with COVID-19 were 2.8 times as likely to experience acute pulmonary embolism as outpatient control-patients and also more likely to experience a range of conditions affecting multiple body systems (eg, nonspecific chest pain, fatigue, headache, and respiratory, nervous, circulatory, and gastrointestinal symptoms) than outpatient control-patients. CONCLUSIONS: These findings add to the evidence of late health conditions possibly related to COVID-19 in adults following COVID-19 diagnosis and can inform healthcare practice and resource planning for follow-up COVID-19 care.

National Center for Health Statistics Guidelines for Analysis of Trends.

Many reports present analyses of trends over time based on multiple years of data from National Center for Health Statistics (NCHS) surveys and the National Vital Statistics System (NVSS). Trend analyses of NCHS data involve analytic choices that can lead to different conclusions about the trends. This report discusses issues that should be considered when conducting a time trend analysis using NCHS data and presents guidelines for making trend analysis choices. Trend analysis issues discussed include: choosing the observed time points to include in the analysis, considerations for survey data and vital records data (record level and aggregated), a general approach for conducting trend analyses, assorted other analytic issues, and joinpoint regression. This report provides 12 guidelines for trend analyses, examples of analyses using NCHS survey and vital records data, statistical details for some analysis issues, and SAS and SUDAAN code for specification of joinpoint regression models. Several an lytic choices must be made during the course of a trend analysis, and the choices made can affect the results. This report highlights the strengths and limitations of different choices and presents guidelines for making some of these choices. While this report focuses on time trend analyses, the issues discussed and guidelines presented are applicable to trend analyses involving other ordinal and interval variables.

National Center for Health Statistics Data Presentation Standards for Proportions.

The National Center for Health Statistics (NCHS) disseminates information on a broad range of health topics through diverse publications. These publications must rely on clear and transparent presentation standards that can be broadly and efficiently applied. Standards are particularly important for large, cross-cutting reports where estimates cannot be individually evaluated and indicators of precision cannot be included alongside the estimates. This report describes the NCHS Data Presentation Standards for Proportions. The multistep NCHS Data Presentation Standards for Proportions are based on a minimum denominator sample size and on the absolute and relative widths of a confidence interval calculated using the Clopper-Pearson method. Proportions (usually multiplied by 100 and expressed as percentages) are the most commonly reported estimates in NCHS reports.

Small area random effects models for capture/recapture methods with applications to estimating coverage error in the U.S. Decennial Census.

The synthetic estimation approach currently in use for estimating net coverage error in the U.S. Census is evaluated using random effects models. The synthetic estimates from the 2000 Accuracy and Coverage Evaluation (ACE) Revision II are evaluated in two parts. First, a model is used, which produces the synthetic estimate components and, second, the model is enlarged to include random effects at the small area level. Retaining all the fixed effects that characterize the synthetic model produces an extremely large, saturated random effects model. Hence, we selectively reduce the random effects model with an aim towards keeping all fixed effects in order to fairly evaluate the synthetic model. A super-population model is used for the bivariate outcome of erroneous enumeration rate and census omission rate. Both these outcomes were previously estimated using the current synthetic estimation approach. A major hurdle in this project was the development of defensible input data for the small areas due to the large number of effects in the synthetic model, which render simple design-based estimates for small areas crossed with post-strata, mostly, unusable. For this initial approach, the small areas were the 540 local census offices. Bayesian methods are employed to evaluate these models. The advantage of this model is that it can evaluate a key assumption about the homogeneity of rates within a post-stratum and if the assumption holds, then this model reduces to the current synthetic model.