PDB11 COMPARISON OF MULTIVARIABLE-ADJUSTED LOGISTIC REGRESSION WITH PROPENSITY SCORE-MATCHED, PROPENSITY SCORE-STRATIFIED, AND PROPENSITY SCORE-ADJUSTED LOGISTIC REGRESSION MODELS

Introduction: Characterized by significant left ventricular (LV) dysfunction and clinical heart failure (HF), peripartum cardiomyopathy (PPCM) has an incidence of approximately 1/2200 live births (0.04%). Prior studies estimate that approximately 25% of those with recovered LV function will have recurrent clinical PPCM during subsequent pregnancies, compared to 50% of those without recovered LV function. Specific predictors of recurrent PPCM have not been studied in cohorts with large numbers. Methods: From 2005-2011, we identified 1,872,227 pregnancies by International Classification of Diseases, 9th Revision (ICD-9) codes in the California Healthcare Cost and Utilization Project (HCUP) database, which captures over 95% of the California hospitalized population. Excluding 15,765 women with prior cardiovascular disease (myocardial infarction, coronary artery disease, stroke, HF, valve disease, or congenital heart disease), yielded n=1,856,462 women. Among women without prior cardiovascular disease, we identified index and subsequent pregnancies with PPCM to determine episodes of recurrent PPCM. We considered the following potential predictors of PPCM recurrence in both univariate and age-adjusted logistic regression models: age, race, hypertension, diabetes, smoking, obesity, chronic kidney disease, family history, pre-eclampsia, ectopic pregnancy, income, and insurance status. Results: In HCUP, n=783 women had pregnancies complicated by PPCM (mean age=30.8 years). Among these women, n=133 had a subsequent pregnancy (17%; mean age=28.1 years), with a mean follow-up of 4.34 years (±1.71 years). In this group of 133 subsequent pregnancies, n=14 (10.5%) were complicated by recurrent PPCM, with a mean time-to-event of 2.2 years (±1.89 years). Among the risk factors studied, the only univariate predictor of recurrent PPCM was grand multiparity, defined as ≥ 5 previous deliveries (odds ratio: 22; 95% confidence interval 4.43-118.22). The other predictors we studied were not significantly associated with recurrent PPCM in either univariate or multivariable models. Conclusion: In a large population database in California with 783 cases of PPCM over a 6-year period, 17% of women had a subsequent pregnancy, of which 10.5% had recurrent PPCM. In age-adjusted logistic regression models, grand multiparity was the only statistically significant predictor of recurrent PPCM.

Download Full-text

Sleep and daytime problems during the COVID-19 pandemic and effects of coronavirus infection, confinement and financial suffering: a multinational survey using a harmonised questionnaire

BMJ Open ◽

10.1136/bmjopen-2021-050672 ◽

2021 ◽

Vol 11 (12) ◽

pp. e050672

Author(s):

Markku Partinen ◽

Brigitte Holzinger ◽

Charles M Morin ◽

Colin Espie ◽

Frances Chung ◽

...

Keyword(s):

Logistic Regression ◽

Sleep Quality ◽

Regression Models ◽

Sleep Problems ◽

Poor Sleep ◽

Poor Sleep Quality ◽

Logistic Regression Models ◽

Excessive Sleepiness ◽

Adjusted Logistic Regression ◽

Daytime Function

ObjectivesSleep is important for human health and well-being. No previous study has assessed whether the COVID-19 pandemic impacts sleep and daytime function across the globe.MethodsThis large-scale international survey used a harmonised questionnaire. Fourteen countries participated during the period of May–August 2020. Sleep and daytime problems (poor sleep quality, sleep onset and maintenance problems, nightmares, hypnotic use, fatigue and excessive sleepiness) occurring ‘before’ and ‘during’ the pandemic were investigated. In total, 25 484 people participated and 22 151 (86.9%) responded to the key parameters and were included. Effects of COVID-19, confinement and financial suffering were considered. In the fully adjusted logistic regression models, results (weighted and stratified by country) were adjusted for gender, age, marital status, educational level, ethnicity, presence of sleep problems before COVID-19 and severity of the COVID-19 pandemic in each country at the time of the survey.ResultsThe responders were mostly women (64%) with a mean age 41.8 (SD 15.9) years (median 39, range 18–95). Altogether, 3.0% reported having had COVID-19; 42.2% reported having been in confinement; and 55.9% had suffered financially. All sleep and daytime problems worsened during the pandemic by about 10% or more. Also, some participants reported improvements in sleep and daytime function. For example, sleep quality worsened in about 20% of subjects and improved in about 5%. COVID-19 was particularly associated with poor sleep quality, early morning awakening and daytime sleepiness. Confinement was associated with poor sleep quality, problems falling asleep and decreased use of hypnotics. Financial suffering was associated with all sleep and daytime problems, including nightmares and fatigue, even in the fully adjusted logistic regression models.ConclusionsSleep problems, fatigue and excessive sleepiness increased significantly worldwide during the first phase of the COVID-19 pandemic. Problems were associated with confinement and especially with financial suffering.

Download Full-text

Multiple Myeloma in the Afro-Caribbean Inner City Population: Conventional Cytogenetics and Survival

Blood ◽

10.1182/blood.v124.21.5710.5710 ◽

2014 ◽

Vol 124 (21) ◽

pp. 5710-5710

Author(s):

Dhvani Thakker ◽

Charles Yun ◽

Adam Goldrich ◽

Helzner Elizabeth ◽

Daniel Fein ◽

...

Keyword(s):

Multiple Myeloma ◽

Logistic Regression ◽

Survival Data ◽

Poor Prognosis ◽

Regression Models ◽

Chromosome Banding ◽

Newly Diagnosed ◽

Cytogenetic Abnormalities ◽

Logistic Regression Models ◽

Adjusted Logistic Regression

Abstract Background: Multiple Myeloma (MM) is the second most common hematologic malignancy in the United States. African Americans have among the highest risks of MM and MGUS with several distinct features compared to existing literature. Furthermore, the prevalence of MM is even higher in the Afro-Caribbean population. Cytogenetic and molecular genetic abnormalities predict outcome in patients with MM. Hyperdiploid MM (H-MM) generally has a better prognosis than nonhyperdiploid MM (NH-MM). In addition, patients with additional chromosome 1 abnormalities, loss of chromosome 13, translocation t(14;16) and t(4;14) tend to have a worse survival while patients with translocations t(11;14) are associated with improved survival. In our patient population, the most common cytogenetic abnormalities and their effect on survival remain unknown. Objective: This study was performed to establish a profile of Afro-Caribbean patients with newly diagnosed Multiple Myeloma in order to gain further insight into unique cytogenetic abnormalities and their effects on survival. Methods: Patients with Multiple Myeloma at Kings County Hospital Center and University Hospital at Brooklyn from 2000-2013 were identified by our tumor registries (n=311). We included all the newly diagnosed patients from 2000-2013 who underwent a bone marrow biopsy and conventional cytogenetic by chromosome banding and FISH (n= 173). Patients who did not have a cytogenetic analysis were excluded. Data was collected at the time of initial presentation to include demographics and cytogenetic abnormalities. Survival data was obtained from Social Security Death Index. Differences in frequency of each cytogenetic abnormality by mortality status were examined using Chi-Square or Fisher’s Exact Tests. Two sets of age-adjusted logistic regression models were used to examine potential cytogenetic correlates of both poor (less than two years) and good (4 years or more) survival. Data analysis was performed using SPSS Advanced Statistics. Results: The median age at the time of diagnosis was 65 (Range 36-90). Chromosome banding and FISH showed abnormal cytogenetics in 46% of our patients (n=79). These patients were also found to have multiple abnormal clones. NH-MM was found in 24% (n=19) and H-MM was found in 39% (n=31) of the 79 patients. The most commonly affected abnormalities were trisomiesof odd-numbered chromosomes; +1 (47%), +3 (19%), +5 (21%), +7 (24%), +9 (47%), +11 (42%), +15 (44%), +17 (9%) and +19 (29%). Thirty five percent of 173 patients have expired (n=60). The median survival in the deceased patients was 6.2 years (Range 0.34-12.9). When we examined all patients who lived greater than four years post-diagnosis (n=152), we found significant abnormalities including +5 (p=0.052), NH-MM (p=0.009) and t(11;14) (p=0.03) (See Table 1). Indicators of poor prognosis including 1q gain (p=0.13), loss of chromosome 13 (p=0.21) and del17 (p=0.08) were not significant. In patients who are living, 19% (n=29) have not yet reached the four-year post-diagnosis survival. Less than ten percent underwent autologous stem cell transplantation. Excludes patients who lived less than 3 months post diagnosis August 5 2014 Table 1: Age-Adjusted Logistic Regression Models Predicting Good Survival (lived 4 years or more post-diagnosis) Chromosome abnormality ( + gain, - loss) Age-Adjusted Odds Ratio (95% CI) N=152 P-value 1+ 0.77 (0.26, 2.29) 0.63 1- 2.91 (0.58, 14.57) 0.19 3+ 1.05 (0.35, 3.17) 0.93 5+ 0.47 (0.22, 1.00) 0.052 7+ 0.39 (0.14, 1.10) 0.08 11+ 0.80 (0.36, 1.75) 0.57 14+ 2.07 (0.62, 6.91) 0.24 15+ 0.74 (0.34,1.60) 0.44 19+ 1.20 (0.46, 3.13) 0.71 X- 0.42 (0.11, 1.50) 0.18 Y- 0.40 (0.13, 1.26) 0.12 Hyperdiploidy 0.88 (0.39, 2.00) 0.88 Nonhyperdiplody 0.24 (0.08, 0.70) 0.009 t(4;14) 0.76 (0.27, 2.15) 0.60 t(11;14) 0.18 (0.04, 0.86) 0.03 Conclusion: In this group of Afro-Caribbean patients, median survival (6 years) was higher than Surveillance, Epidemiology, and End Results (SEER) data and more recent review of literature. Gain of chromosome 5 and t(11;14) are consistent with existing data for good prognosis. However, NH-MM which is usually an indicator of poor prognosis was also highly significant in the four-year post-diagnosis survival. This further supports the notion that prognostic value of cytogenetic analysis in this population requires further exploration. Disclosures No relevant conflicts of interest to declare.

Download Full-text

Racial differences in crossing major growth percentiles in infancy

Archives of Disease in Childhood ◽

10.1136/archdischild-2016-311238 ◽

2017 ◽

Vol 103 (8) ◽

pp. 795-797 ◽

Cited By ~ 4

Author(s):

Summer Sherburne Hawkins ◽

Sheryl L Rifas-Shiman ◽

Matthew W Gillman ◽

Elsie M Taveras

Keyword(s):

Body Mass Index ◽

Logistic Regression ◽

Weight Gain ◽

Racial Differences ◽

Regression Models ◽

Black Children ◽

Rapid Weight Gain ◽

White Children ◽

Logistic Regression Models ◽

Adjusted Logistic Regression

Background/aimsWe examined associations of ever crossing upwards ≥2 (vs <2) major weight-for-length (WFL) percentiles in the first 24 months with obesity at 5 years among white and black children.MethodsWe included 10 979 white and 1245 black children from the Linked CENTURY Study with percentile crossing data in all four 6-month periods in the first 24 months and obesity (age-specific and sex-specific body mass index ≥95th percentile) at 5 years. We used adjusted logistic regression models and stratified by race.Results64% of children crossed upwards ≥2 major WFL percentiles in the first 2 years. Among white children, 12% were obese vs 7% for <2 crossings, while among black children the frequencies were 23% vs 9%. Black children (adjusted OR 2.94, 2.04 to 4.23) who had ever crossed upwards ≥2 major WFL percentiles had a higher odds of obesity at age 5 than white children (adjusted OR 1.89, 1.64 to 2.18) (interaction p=0.02).ConclusionsOur results suggest that rapid weight gain in infancy is more deleterious among black than white children for later obesity.

Download Full-text

Is Greater Than 0.5 MAC Inhalational Agent Use Post-Bypass Related to Need for Inotropic and/or Vasoconstrictor Support?

Hearts ◽

10.3390/hearts2020017 ◽

2021 ◽

Vol 2 (2) ◽

pp. 213-223

Author(s):

Tara A. Lenk ◽

Carlos E. Guerra-Londono ◽

Thomas E. Graul ◽

Marc A. Murinson ◽

Prabhdeep K. Hehar ◽

...

Keyword(s):

Logistic Regression ◽

Cardiac Surgery ◽

Propensity Score ◽

Regression Models ◽

Independent Predictor ◽

Care Hospital ◽

Elective Cardiac Surgery ◽

Logistic Regression Models ◽

Adjusted Analysis ◽

Inhalational Agent

Background and Aims: We hypothesized that maintaining a patient on moderate–high doses of potent inhalational agent for greater than 30 min during the post-bypass period would be an independent predictor of initiation and usage of either inotropic and/or vasopressor infusions. Setting and Design: This study is a retrospective design and approved by the institutional review board. The setting was a single-center, academic tertiary care hospital in Detroit, Michigan. Materials and Methods: Three-hundred, ninety-seven elective cardiac surgery patients were identified for chart review. Electronic medical records were reviewed to collect demographics and perioperative data. Statistics used include a propensity score regression adjusted analysis utilizing logistic regression models and a multivariable model. Results: A propensity score regression adjusted analysis was performed and then applied in both univariate and multivariate logistic regression models with a p value of <0.05 reaching statistical significance. Fifty-six percent of the participants had an exposure of greater than 30 min of a minimum alveolar concentration of isoflurane greater than 0.5 (ETISO ≥ 0.5MAC, 30 min) in the post-bypass period. After adjusting for propensity score, this was found to be a significant predictor of inotrope and/or vasoconstrictor use post-bypass (OR 2.49, 95% CI 1.15–5.38, p = 0.021). In the multivariate model, pulmonary hypertension (OR 5.9; 95% CI 1.33–26.28; p = 0.02), Euroscore II (2.73; 95% CI 1.35–5.5; p = 0.005), and cardiopulmonary bypass hours (OR 1.86; 95% CI 1.02–3.4; p = 0.042) emerged as significant. Conclusions: This study showed that an ETISO ≥ 0.5MAC, 30 min exposure during the immediate post-bypass period during elective cardiac surgery was an independent predictor of a patient being started on inotrope or vasoconstrictor infusions. Further research should consider a prospective design and examine depth of anesthesia during the post-bypass period.

Download Full-text

PMC10 COMPARISON BETWEEN PROPENSITY SCORE AND TRADITIONAL COVARIATE ADJUSTMENT WITH LOGISTIC REGRESSION MODELS IN ESTIMATING THE AVERAGE TREATMENT EFFECTS (ATES): RESULTS FROM MONTE CARLO SIMULATIONS

Value in Health ◽

10.1016/s1098-3015(10)73163-6 ◽

2009 ◽

Vol 12 (3) ◽

pp. A21

Author(s):

QA Le ◽

JW Hay

Keyword(s):

Monte Carlo ◽

Logistic Regression ◽

Propensity Score ◽

Monte Carlo Simulations ◽

Regression Models ◽

Treatment Effects ◽

Covariate Adjustment ◽

Logistic Regression Models ◽

Average Treatment ◽

Average Treatment Effects

Download Full-text

Biopsy in emergency gastroscopy does not increase the risk of rebleeding in patients with Forrest I acute nonvariceal upper gastrointestinal bleeding combined with suspected malignant gastric ulcer: a multicenter retrospective cohort study

BMC Gastroenterology ◽

10.1186/s12876-021-01836-z ◽

2021 ◽

Vol 21 (1) ◽

Author(s):

Quchuan Zhao ◽

Tianyu Chi

Keyword(s):

Logistic Regression ◽

Gastric Ulcer ◽

Cohort Study ◽

Propensity Score ◽

Propensity Score Matching ◽

Retrospective Cohort ◽

Regression Models ◽

Upper Gastrointestinal Bleeding ◽

Logistic Regression Models ◽

Upper Gastrointestinal

Abstract Background Few studies have reported whether a biopsy in emergency gastroscopy (EG) increased the risk of rebleeding in patients with Forrest I acute nonvariceal upper gastrointestinal bleeding (ANVUGIB) combined with suspected malignant gastric ulcer (SMGU). This study aims to conduct a multicenter retrospective cohort study using propensity score matching to verify whether a biopsy in EG increases the risk of rebleeding in patients diagnosed with Forrest I ANVUGIB combined with SMGU. Methods Using the data for propensity-matched patients, logistic regression models were fitted using rebleeding as the dependent variable. Survival time was defined as the length of time the patient experienced from visiting the emergency department to rebleeding. We used the Kaplan–Meier (KM) method to analyze the 30-day survival of the patients with and without a biopsy after matching, and the log-rank test was performed to examine the differences in survival. Results With the use of propensity score matching, 308 patients who underwent a biopsy in EG were matched with 308 patients who did not. In the five logistic regression models, there were no significant group differences in the risk of rebleeding in patients with Forrest I ANVUGIB combined with SMGU between the biopsy and no-biopsy groups. The probability of survival was not significantly different between the no-biopsy and biopsy groups. Conclusions In this multicenter, retrospective propensity score matching cohort study, compared with patients without a biopsy, patients with a biopsy during EG had no increased risk of rebleeding, and there was no significant difference in the rate of rebleeding.

Download Full-text

Abstract WP39: Intraprocedural Systolic Blood Pressure Mean and Variability During Endovascular Thrombectomy Not Associated With Neurologic Outcome

Stroke ◽

10.1161/str.51.suppl_1.wp39 ◽

2020 ◽

Vol 51 (Suppl_1) ◽

Author(s):

Mohammad Anadani ◽

Adam De Havenon ◽

Linda M Baki ◽

Alejandro M Spiotta

Keyword(s):

Blood Pressure ◽

Logistic Regression ◽

Systolic Blood Pressure ◽

Regression Models ◽

Primary Outcome ◽

Neurologic Outcome ◽

Anterior Circulation ◽

Endovascular Thrombectomy ◽

Logistic Regression Models ◽

Adjusted Logistic Regression

Background: Prior research has shown that increased systolic blood pressure variability (BPV) is associated with worse neurologic outcome after endovascular thrombectomy (EVT). Those studies have typically included BPV from 24-72 hours after stroke onset, but none have focused specifically on the BPV during EVT. Methods: We included acute ischemic stroke patients who underwent EVT for anterior circulation large vessel occlusion from 2 stroke centers. All patients had a minimum of 10 blood pressure readings during EVT. The primary outcome is mRS 0-2 (good outcome) and the secondary outcome is death, both as close to 90 days as possible. We fit adjusted logistic regression models to our outcomes with the predictors of intraprocedural systolic mean, standard (SD), and coefficient of variation (CV). Results: We included 303 patients with a mean (SD) age of 65.7 years and 53.5% were female. The primary outcome of mRS 0-2 was met by 39.9% and 27.4% died. Systolic mean, SD, and CV did not differ in patients with mRS 0-2 versus 3-6 (Table 1) nor for patients who died versus were alive (all p>0.5). In the adjusted logistic regression models, systolic mean, SD, and CV were not associated with either mRS 0-2 or death at follow-up (Tables 2 & 3). Conclusions: Blood pressure variation during endovascular thrombectomy was not associated with the functional outcome or death in patients with anterior circulation strokes.

Download Full-text

Serious Psychological Distress and Emergency Room Use among Adults with Multimorbidity in the United States

Psychiatry Journal ◽

10.1155/2017/8565186 ◽

2017 ◽

Vol 2017 ◽

pp. 1-8 ◽

Cited By ~ 3

Author(s):

Khalid Alhussain ◽

Abdulkarim M. Meraya ◽

Usha Sambamoorthi

Keyword(s):

United States ◽

Logistic Regression ◽

Psychological Distress ◽

Emergency Room ◽

Regression Models ◽

The United States ◽

Serious Psychological Distress ◽

Cross Sectional ◽

Logistic Regression Models ◽

Adjusted Logistic Regression

Objectives. (1) To examine the association between serious psychological distress (SPD) and emergency room (ER) use in the past 12 months among adults with multimorbidity in the United States (US) and (2) to investigate the association between SPD and the reasons for ER use. Methods. The current study used a cross-sectional design with retrospective data from the 2015 National Health Interview Survey. Logistic regression models were used to assess the association between SPD and ER use among adults with multimorbidity. Among ER users, adjusted logistic regression models were conducted to examine the association between SPD and the reasons for the ER use. Results. After controlling for other variables, adults with multimorbidity and SPD were more likely to use ER than those with multimorbidity and no SPD (AOR = 1.61, 95% CI = 1.26, 2.04). Among ER users, there were no significant associations between SPD and the reasons for ER use after controlling for other variables. Conclusion. Adults with multimorbidity and SPD were more likely to use ER as compared to those with multimorbidity and no SPD. Among adults with multimorbidity, routine screening for SPD may be needed to reduce the ER use.

Download Full-text

Immigration, perceived discrimination and mental health: evidence from Venezuelan population living in Peru

Globalization and Health ◽

10.1186/s12992-020-00655-3 ◽

2021 ◽

Vol 17 (1) ◽

Author(s):

Benoît Mougenot ◽

Elard Amaya ◽

Edward Mezones-Holguin ◽

Alfonso J. Rodriguez-Morales ◽

Báltica Cabieses

Keyword(s):

Mental Health ◽

Logistic Regression ◽

Propensity Score ◽

Propensity Score Matching ◽

Regression Models ◽

Mental Health Problems ◽

Perceived Discrimination ◽

Health Problems ◽

Logistic Regression Models ◽

Venezuelan Population

Abstract Background The association between international migration and mental health is conditioned to several factors, and discrimination may play a significant role. Currently, Peru is one of the principal Venezuelan migrant-receiving countries in Latin America. There are around one million Venezuelan refugees and migrants in the country. This study evaluates the association between self-perceived discrimination and mental health problems in Venezuelan population living in Peru. Method We analyzed data from the Venezuelan Population Residing in Peru Survey 2018, a nationally representative urban sample aimed at collecting information on several dimensions of Venezuelan population wellbeing. We applied logistic regression models to assess the association between self-perceived discrimination and mental health problems. Moreover, we applied the propensity score matching method as a robustness check of our results. Results Of 9487 Venezuelans surveyed, 6806 included complete information. From this sample, 6.3% reported mental health problems related to fear, anger, anxiety, or stress. Logistic regression models showed that Venezuelans who perceived being discriminated against had 2.4 higher odds of presenting mental health problems than their non-discriminated counterparts. Moreover, propensity score matching models showed that Venezuelans who perceived being discriminated against increased by 3.5 percentage points their probability of presenting mental health problems compared to their non-discriminated counterparts. Conclusions There is evidence that self-perceived discrimination is associated with mental health deterioration in Venezuelan migrants living in Peru. Our findings are relevant in the current geopolitical context and could be useful in the decision making processes in international health.

Download Full-text