-
The Angle Orthodontist May 2022To map the statistical methods applied to assess reliability in orthodontic publications and to identify possible trends over time.
OBJECTIVES
To map the statistical methods applied to assess reliability in orthodontic publications and to identify possible trends over time.
MATERIALS AND METHODS
Original research articles published in 2009 and 2019 in a subset of orthodontic journals were downloaded. Publication characteristics, including publication year, number of authors, single vs multicenter study, geographic origin of the study, statistician involvement, study category, subject category, types of reliability assessment, and statistical methods applied to assess reliability, were recorded. Descriptive statistics, Chi-square tests, and logistic regression analyses were performed to investigate associations between reliability analysis and study characteristics.
RESULTS
A total of 768 original research articles were analyzed. The most prevalent study category was observational (69%) with a statistician involved in 16% of studies. Overall, reliability was assessed in 47% of studies, and the most frequent methods applied to assess reliability were intraclass correlation coefficients or kappa statistics (60.4%). The odds of applying appropriate methods were greater in 2019 than in 2009 (odds ratio [OR]: 2.43; 95% confidence interval [CI]: 1.75, 3.37; P < .001). Involvement of a statistician resulted in greater odds of applying appropriate methods compared to no statistician involvement (OR: 1.88; 95% CI: 1.23, 2.87; P < .01).
CONCLUSIONS
Over the past decade (2009 vs 2019), reliability assessment became more common in the orthodontic literature, and studies applying correct statistical methods to assess reliability significantly increased. This trend was more apparent in studies that involved a statistician, which may highlight the role of the statistician.
Topics: Odds Ratio; Reproducibility of Results; Research Design
PubMed: 35099528
DOI: 10.2319/081021-625.1 -
Statistics in Medicine Jun 2022Studies of critically ill, hospitalized patients often follow participants and characterize daily health status using an ordinal outcome variable. Statistically,... (Randomized Controlled Trial)
Randomized Controlled Trial
Studies of critically ill, hospitalized patients often follow participants and characterize daily health status using an ordinal outcome variable. Statistically, longitudinal proportional odds models are a natural choice in these settings since such models can parsimoniously summarize differences across patient groups and over time. However, when one or more of the outcome states is absorbing, the proportional odds assumption for the follow-up time parameter will likely be violated, and more flexible longitudinal models are needed. Motivated by the VIOLET Study (Ginde et al), a parallel-arm, randomized clinical trial of Vitamin in critically ill patients, we discuss and contrast several treatment effect estimands based on time-dependent odds ratio parameters, and we detail contemporary modeling approaches. In VIOLET, the outcome is a four-level ordinal variable where the lowest "not alive" state is absorbing and the highest "at-home" state is nearly absorbing. We discuss flexible extensions of the proportional odds model for longitudinal data that can be used for either model-based inference, where the odds ratio estimator is taken directly from the model fit, or for model-assisted inferences, where heterogeneity across cumulative log odds dichotomizations is modeled and results are summarized to obtain an overall odds ratio estimator. We focus on direct estimation of cumulative probability model (CPM) parameters using likelihood-based analysis procedures that naturally handle absorbing states. We illustrate the modeling procedures, the relative precision of model-based and model-assisted estimators, and the possible differences in the values for which the estimators are consistent through simulations and analysis of the VIOLET Study data.
Topics: Biometry; Critical Illness; Humans; Likelihood Functions; Longitudinal Studies; Odds Ratio
PubMed: 35253265
DOI: 10.1002/sim.9366 -
American Journal of Obstetrics and... Oct 2022Hysterectomy is the most common nonobstetrical medical procedure performed in US women. Evaluating hysterectomy prevalence trends and determinants is important for...
BACKGROUND
Hysterectomy is the most common nonobstetrical medical procedure performed in US women. Evaluating hysterectomy prevalence trends and determinants is important for estimating gynecologic cancer rates and management of uterine conditions.
OBJECTIVE
This study aimed to assess hysterectomy prevalence trends and determinants using the Behavioral Risk Factor Surveillance System (2006-2016).
STUDY DESIGN
We estimated crude hysterectomy prevalences and multivariable-adjusted odds ratios and 95% confidence intervals for associations of race or ethnicity, age group (5-year), body mass index (categorical), smoking status, education, insurance, income, and US region with hysterectomy. Missing data were imputed. The number of women in each survey year ranged from 220,302 in 2006 to 275,631 in 2016.
RESULTS
Although overall hysterectomy prevalence changed little between 2006 and 2016 (21.4% and 21.1%, respectively), hysterectomy prevalence was lower in 2016 than in 2006 among women aged ≥40 years, particularly among non-Hispanic Black and Hispanic women. Current smoking (odds ratio, 1.38; 95% confidence interval, 1.35-1.41), increasing age (odds ratio, 1.40; 95% confidence interval, 1.39-1.40), living in the South compared with the Midwest (odds ratio, 1.36; 95% confidence interval, 1.34-1.39), higher body mass index (odds ratio, 1.26; 95% confidence interval, 1.25-1.27), Black race compared with White (odds ratio, 1.10; 95% confidence interval, 1.07-1.13), and having insurance compared with being uninsured (odds ratio, 1.26; 95% confidence interval, 1.22-1.30) were most strongly associated with increased prevalence. Hispanic ethnicity and living in the Northeast were most strongly associated with decreased prevalence (odds ratio, 0.73; 95% confidence interval, 0.70-0.76; odds ratio, 0.67; 95% confidence interval, 0.65-0.69).
CONCLUSION
Nationwide hysterectomy prevalence decreased among women aged ≥40 years from 2006 to 2016, particularly among non-Hispanic Black and Hispanic women. Age, non-Hispanic Black race, having insurance, current smoking, and living in the South were associated with increased odds of hysterectomy, even after accounting for possible explanatory factors. Further research is needed to better understand associations of race and ethnicity and region with hysterectomy prevalence.
Topics: Ethnicity; Female; Hispanic or Latino; Humans; Hysterectomy; Odds Ratio; Prevalence; United States
PubMed: 35764133
DOI: 10.1016/j.ajog.2022.06.028 -
Acta Neurologica Scandinavica Nov 2022Vestibular neuritis is a common neuro-otological entity. Therapeutically, corticosteroids are advised, although the evidence is limited. The objective of this review is... (Meta-Analysis)
Meta-Analysis Review
Vestibular neuritis is a common neuro-otological entity. Therapeutically, corticosteroids are advised, although the evidence is limited. The objective of this review is to update meta-analyses of clinical trials that address the question of whether patients with vestibular neuritis treated with corticosteroids show better recovery than control patients. The electronic databases Medline, Scopus and Cochrane were searched for clinical trials for the years 1970-2020 without language restriction. Data were extracted, and outcome parameters were subjected to conventional and cumulative meta-analysis using a commercially available software program (www.meta-analysis.com). Finally, 15 trials with 363 participants in the treatment and 489 in the control groups were identified and could be included. Eight studies were judged to be at high risk of bias. The odds ratio (OR) for good outcome in the acute phase was 3.1 (95% CI 1.2-7.8; p = .015) in favour of steroid treatment leading to the number needed to treat (NNT) = 6 (95% CI 4-23). The odds ratio (OR) for restoration of vestibular function in the follow-up was 2.4 (95% CI 1.3-4.4; p = .004) for the benefit of steroid treatment resulting in a NNT = 7 (95% CI 5-18). The results of the cumulative statistics did not differ. The risk of adverse effects was higher in patients treated with steroids with an OR of 10.9 (95% CI 1.3-93.8; p = .015) and an estimated number needed to harm (NNH) = 4 (95% CI 3-19). The advantage for corticosteroids remained when differentiating between patients who participated in randomized or non-randomized clinical trials. Steroid treatment in vestibular neuritis resulted in a statistically significant benefit compared to control therapies. However, broad heterogeneity of the studies, mostly low-grade quality of studies, high risk of bias and broad confidence intervals put the findings into perspective allowing only a careful judgement of some benefit of corticosteroids. The findings, however, support the call for an adequately powered and well-designed randomized controlled trial to re-evaluate the effectiveness of corticosteroids.
Topics: Adrenal Cortex Hormones; Humans; Odds Ratio; Steroids; Vestibular Neuronitis
PubMed: 36029039
DOI: 10.1111/ane.13676 -
Journal of Epidemiology Oct 2023In case-cohort studies with binary outcomes, ordinary logistic regression analyses have been widely used because of their computational simplicity. However, the...
BACKGROUND
In case-cohort studies with binary outcomes, ordinary logistic regression analyses have been widely used because of their computational simplicity. However, the resultant odds ratio estimates cannot be interpreted as relative risk measures unless the event rate is low. The risk ratio and risk difference are more favorable outcome measures that are directly interpreted as effect measures without the rare disease assumption.
METHODS
We provide pseudo-Poisson and pseudo-normal linear regression methods for estimating risk ratios and risk differences in analyses of case-cohort studies. These multivariate regression models are fitted by weighting the inverses of sampling probabilities. Also, the precisions of the risk ratio and risk difference estimators can be improved using auxiliary variable information, specifically by adapting the calibrated or estimated weights, which are readily measured on all samples from the whole cohort. Finally, we provide computational code in R (R Foundation for Statistical Computing, Vienna, Austria) that can easily perform these methods.
RESULTS
Through numerical analyses of artificially simulated data and the National Wilms Tumor Study data, accurate risk ratio and risk difference estimates were obtained using the pseudo-Poisson and pseudo-normal linear regression methods. Also, using the auxiliary variable information from the whole cohort, precisions of these estimators were markedly improved.
CONCLUSION
The ordinary logistic regression analyses may provide uninterpretable effect measure estimates, and the risk ratio and risk difference estimation methods are effective alternative approaches for case-cohort studies. These methods are especially recommended under situations in which the event rate is not low.
Topics: Humans; Odds Ratio; Japan; Cohort Studies; Probability; Risk; Models, Statistical
PubMed: 35753802
DOI: 10.2188/jea.JE20210509 -
American Journal of Epidemiology Feb 2021Case-control studies are an important part of the epidemiologic literature, yet confusion remains about how to interpret estimates from different case-control study...
Case-control studies are an important part of the epidemiologic literature, yet confusion remains about how to interpret estimates from different case-control study designs. We demonstrate that not all case-control study designs estimate odds ratios. On the contrary, case-control studies in the literature often report odds ratios as their main parameter even when using designs that do not estimate odds ratios. Only studies using specific case-control designs should report odds ratios, whereas the case-cohort and incidence-density sampled case-control studies must report risk ratio and incidence rate ratios, respectively. This also applies to case-control studies conducted in open cohorts, which often estimate incidence rate ratios. We also demonstrate the misinterpretation of case-control study estimates in a small sample of highly cited case-control studies in general epidemiologic and medical journals. We therefore suggest that greater care be taken when considering which parameter is to be reported from a case-control study.
Topics: Case-Control Studies; Data Interpretation, Statistical; Humans; Odds Ratio; Research Design
PubMed: 32889542
DOI: 10.1093/aje/kwaa167 -
JAMA Health Forum Dec 2021Cost-sharing requirements can discourage patients from seeking care and impose financial hardship. The Medicare program serves many older and disabled individuals with...
IMPORTANCE
Cost-sharing requirements can discourage patients from seeking care and impose financial hardship. The Medicare program serves many older and disabled individuals with multimorbidity and limited resources, but little has been known about the affordability of care in this population.
OBJECTIVE
To examine the affordability of medical care among Medicare enrollees, in terms of the prevalence of delaying medical care because of costs and having problems paying medical bills, and risk factors for these outcomes.
DESIGN SETTING AND PARTICIPANTS
Cross-sectional analyses conducted from November 1, 2019, to October 15, 2021, used logistic regression to compare the probability of outcomes by demographic and health characteristics. Data were obtained from the 2017 nationally representative Medicare Current Beneficiary Survey (response rate, 61.7%), with respondents representing 53 million community-dwelling Medicare enrollees.
MAIN OUTCOMES AND MEASURES
New questions about medical care affordability were included in the 2017 Medicare Current Beneficiary Survey: difficulty paying medical bills, ongoing medical debt, and contact by collection agencies. A companion survey question asked whether individuals had delayed seeking medical care because of worries about costs.
RESULTS
Respondents included 10 974 adults aged 65 years or older and 2197 aged 18 to 64 years; 54.2% of all respondents were women. The weighted proportions of Medicare enrollees with annual incomes below $25 000K were 30.7% in the older population and 67.4% in the younger group. Self-reported prevalence of delaying care because of cost was 8.3% (95% CI, 7.4%-9.1%) among enrollees aged 65 years or older, 25.2% (95% CI, 21.8%-28.6%) among enrollees younger than 65 years, and 10.9% (95% CI, 9.9%-11.9%) overall. Similarly, 7.4% (95% CI, 6.6%-8.2%) of older enrollees had problems paying medical bills, compared with 29.8% (95% CI, 25.6%-34.1%) among those younger than 65 years and 10.8% (95% CI, 9.8%-11.9%) overall. Regarding specific payment problems, 7.9% (95% CI, 7.0%-8.9%) of enrollees overall experienced ongoing medical debt, contact by a collection agency, or both. In adjusted analyses, older adults with incomes $15 000 to $25 000 per year had odds of delaying care more than twice as high as those with incomes greater than $50 000 (odds ratio, 2.47; 95% CI, 1.82-3.39), and their odds of problems paying medical bills were more than 3 times as high (odds ratio, 3.37; 95% CI, 2.81-5.21). Older adults with 4 to 10 chronic conditions were more than twice as likely to have problems paying medical bills as those with 0 or 1 condition.
CONCLUSIONS AND RELEVANCE
The findings of this study suggest that unaffordability of medical care is common among Medicare enrollees, especially those with lower incomes, or worse health, or who qualify for Medicare based on disability. Policy reforms, such as caps on patient spending, are needed to reduce Medical cost burdens on the most vulnerable enrollees.
Topics: Aged; Chronic Disease; Cost Sharing; Cross-Sectional Studies; Female; Humans; Male; Medicare; Odds Ratio; United States
PubMed: 35977305
DOI: 10.1001/jamahealthforum.2021.4104 -
Jornal Brasileiro de Pneumologia :... Apr 2020
Topics: Clinical Trials as Topic; Humans; Logistic Models; Odds Ratio; Risk Factors
PubMed: 32321148
DOI: 10.36416/1806-3756/e20200137 -
BMC Medical Research Methodology Jun 2023Cochran's Q statistic is routinely used for testing heterogeneity in meta-analysis. Its expected value (under an incorrect null distribution) is part of several popular... (Meta-Analysis)
Meta-Analysis
BACKGROUND
Cochran's Q statistic is routinely used for testing heterogeneity in meta-analysis. Its expected value (under an incorrect null distribution) is part of several popular estimators of the between-study variance, [Formula: see text]. Those applications generally do not account for use of the studies' estimated variances in the inverse-variance weights that define Q (more explicitly, [Formula: see text]). Importantly, those weights make approximating the distribution of [Formula: see text] rather complicated.
METHODS
As an alternative, we are investigating a Q statistic, [Formula: see text], whose constant weights use only the studies' arm-level sample sizes. For log-odds-ratio (LOR), log-relative-risk (LRR), and risk difference (RD) as the measures of effect, we study, by simulation, approximations to distributions of [Formula: see text] and [Formula: see text], as the basis for tests of heterogeneity.
RESULTS
The results show that: for LOR and LRR, a two-moment gamma approximation to the distribution of [Formula: see text] works well for small sample sizes, and an approximation based on an algorithm of Farebrother is recommended for larger sample sizes. For RD, the Farebrother approximation works very well, even for small sample sizes. For [Formula: see text], the standard chi-square approximation provides levels that are much too low for LOR and LRR and too high for RD. The Kulinskaya et al. (Res Synth Methods 2:254-70, 2011) approximation for RD and the Kulinskaya and Dollinger (BMC Med Res Methodol 15:49, 2015) approximation for LOR work well for [Formula: see text] but have some convergence issues for very small sample sizes combined with small probabilities.
CONCLUSIONS
The performance of the standard [Formula: see text] approximation is inadequate for all three binary effect measures. Instead, we recommend a test of heterogeneity based on [Formula: see text] and provide practical guidelines for choosing an appropriate test at the .05 level for all three effect measures.
Topics: Humans; Computer Simulation; Probability; Odds Ratio; Algorithms; Sample Size
PubMed: 37344771
DOI: 10.1186/s12874-023-01939-z -
Prognostic value of capillary refill time in adult patients: a systematic review with meta-analysis.Critical Care (London, England) Dec 2023Acute circulatory failure leads to tissue hypoperfusion. Capillary refill time (CRT) has been widely studied, but its predictive value remains debated. We conducted a... (Meta-Analysis)
Meta-Analysis Review
PURPOSE
Acute circulatory failure leads to tissue hypoperfusion. Capillary refill time (CRT) has been widely studied, but its predictive value remains debated. We conducted a meta-analysis to assess the ability of CRT to predict death or adverse events in a context at risk or confirmed acute circulatory failure in adults.
METHOD
MEDLINE, EMBASE, and Google scholar databases were screened for relevant studies. The pooled area under the ROC curve (AUC ROC), sensitivity, specificity, threshold, and diagnostic odds ratio using a random-effects model were determined. The primary analysis was the ability of abnormal CRT to predict death in patients with acute circulatory failure. Secondary analysis included the ability of CRT to predict death or adverse events in patients at risk or with confirmed acute circulatory failure, the comparison with lactate, and the identification of explanatory factors associated with better accuracy.
RESULTS
A total of 60,656 patients in 23 studies were included. Concerning the primary analysis, the pooled AUC ROC of 13 studies was 0.66 (95%CI [0.59; 0.76]), and pooled sensitivity was 54% (95%CI [43; 64]). The pooled specificity was 72% (95%CI [55; 84]). The pooled diagnostic odds ratio was 3.4 (95%CI [1.4; 8.3]). Concerning the secondary analysis, the pooled AUC ROC of 23 studies was 0.69 (95%CI [0.65; 0.74]). The prognostic value of CRT compared to lactate was not significantly different. High-quality CRT was associated with a greater accuracy.
CONCLUSION
CRT poorly predicted death and adverse events in patients at risk or established acute circulatory failure. Its accuracy is greater when high-quality CRT measurement is performed.
Topics: Humans; Adult; Prognosis; Hemodynamics; Shock; Odds Ratio
PubMed: 38042855
DOI: 10.1186/s13054-023-04751-9