scholarly journals Effect of a preliminary test of homogeneity of stratum-specific odds ratios on their confidence intervals

2012 ◽  
Vol 6 (0) ◽  
pp. 672-685 ◽  
Author(s):  
Paul Kabaila ◽  
Dilshani Tissera
Circulation ◽  
2007 ◽  
Vol 116 (suppl_16) ◽  
Author(s):  
George A Diamond ◽  
Sanjay Kaul

Background A highly publicized meta-analysis of 42 clinical trials comprising 27,844 diabetics ignited a firestorm of controversy by charging that treatment with rosiglitazone was associated with a “…worrisome…” 43% greater risk of myocardial infarction ( p =0.03) and a 64% greater risk of cardiovascular death ( p =0.06). Objective The investigators excluded 4 trials from the infarction analysis and 19 trials from the mortality analysis in which no events were observed. We sought to determine if these exclusions biased the results. Methods We compared the index study to a Bayesian meta-analysis of the entire 42 trials (using odds ratio as the measure of effect size) and to fixed-effects and random-effects analyses with and without a continuity correction that adjusts for values of zero. Results The odds ratios and confidence intervals for the analyses are summarized in the Table . Odds ratios for infarction ranged from 1.43 to 1.22 and for death from 1.64 to 1.13. Corrected models resulted in substantially smaller odds ratios and narrower confidence intervals than did uncorrected models. Although corrected risks remain elevated, none are statistically significant (*p<0.05). Conclusions Given the fragility of the effect sizes and confidence intervals, the charge that roziglitazone increases the risk of adverse events is not supported by these additional analyses. The exaggerated values observed in the index study are likely the result of excluding the zero-event trials from analysis. Continuity adjustments mitigate this error and provide more consistent and reliable assessments of true effect size. Transparent sensitivity analyses should therefore be performed over a realistic range of the operative assumptions to verify the stability of such assessments especially when outcome events are rare. Given the relatively wide confidence intervals, additional data will be required to adjudicate these inconclusive results.


Author(s):  
Seo-Hee Park ◽  
Byung-Jin Park ◽  
Dong-Hyuk Jung ◽  
Yu-Jin Kwon

Household food insecurity has been associated with noncommunicable diseases. The aim of this study was to investigate the association between household food insecurity and asthma in Korean adults. Household food security statuses were classified into three groups: Food-secure household, food-insecure household without hunger, and food-insecure household with hunger. The odds ratios and 95% confidence intervals for the presence of asthma according to household food security status were calculated using multiple logistic regression analyses after adjusting for confounding factors. A total of 14,770 participants were included in the analysis. The prevalence of asthma was 2.6% in those with a secure food status, 3.2% in those with an insecure food status without hunger, and 7.6% in those with an insecure food status with hunger (p < 0.001). Compared with that in participants with a household food secure status, the odds ratios (95% confidence intervals) for asthma were 1.12 (0.73–1.73) in those with a food-insecure household without hunger status and 2.44 (1.33–4.46) in those with a food-insecure household with hunger status after additionally adjusting for confounding factors. We found that household food insecurity with hunger was significantly associated with asthma prevalence in Korean adults. Implementation of household food security screening and public health intervention could be helpful to prevent and reduce asthma in adults.


2020 ◽  
Vol 12 ◽  
pp. 1759720X2091286
Author(s):  
Yu-Jin Kwon ◽  
Il-Hyun Koh ◽  
Kwangho Chung ◽  
Yong-Jae Lee ◽  
Hyoung-Sik Kim

Background: Osteoarthritis (OA) is a multifactorial disease involving inflammatory processes. Platelets play important roles in both hemostasis and the inflammatory response; however, the relationship between platelet count and OA is unclear. Our aim was to evaluate the association between platelet count and knee and hip OA in Korean women. Methods: In this cross-sectional designed study, we included a total of 6011 women aged ⩾50 years from the 2010–2013 Korea National Health and Nutrition Examination Survey. Knee and hip OA were defined as Kellgren–Lawrence grade ⩾2 and presence of knee or hip pain, respectively. Platelet counts were divided into quartiles as follows: Q1, 150–212 (103/µl); Q2, 213–246 (103/µl); Q3, 247–283 (103/µl); and Q4, 284–450 (103/µl). Multiple logistic-regression analysis was conducted to calculate odds ratios and 95% confidence intervals. Receiver operating characteristic analysis was performed to determine the optimal platelet count cut-off with which to discriminate participants with knee and/hip OA versus those without OA. Results: Of the 6011 participants, 1141 (18.1%) had knee or hip OA. The mean age of participants without OA was 60.6 years, and that of participants with OA was 68.0 years. Compared with the lowest quartile, odds ratios (95% confidence intervals) for OA were 1.08 (0.84–1.39) for Q2, 0.94 (0.73–1.23) for Q3, and 1.35 (1.08–1.69) for Q4 after adjusting for confounders. The prevalence of OA was significantly higher with platelet counts ⩾288 × 103/µl, compared with platelet counts <288 × 103/µl. Conclusion: High platelet counts within the normal range are significantly associated with knee and hip OA.


2019 ◽  
Vol 47 (4) ◽  
pp. 1409-1416
Author(s):  
Meiming Yang ◽  
Xiaoli Du ◽  
Feng Zhang ◽  
Shifang Yuan

Background Several studies have reported correlations between BRCA1 polymorphisms rs799917 and rs1799966 with the risk of breast cancer (BC). However, this relationship remains controversial. Methods We conducted a meta-analysis of seven studies to assess the associations between BRCA1 rs799917 and rs1799966 and BC risk, with the aim of more accurately determining the potential correlation. Odds ratios (ORs) and 95% confidence intervals (CIs) were estimated to evaluate the correlation of rs799917 and rs1799966 with BC risk. Results There was no overall correlation between BRCA1 rs799917 and BC risk (TT vs CC: OR = 0.87, 95% CI = 0.66–1.16; CT vs CC: OR = 1.02, 95% CI = 0.89–1.15; dominant model: OR = 0.99, 95% CI = 0.88–1.11; recessive model: OR = 0.87, 95% CI = 0.65–1.16). Subgroup analysis by ethnicity also revealed no significant correlation between rs799917 and BC risk in either Asians or Caucasians. There was also no significant association between BRCA1 rs1799966 and BC risk (GG vs AA: OR = 0.70, 95% CI = 0.33–1.47; AG vs AA: OR = 0.68, 95% CI = 0.35–1.30; dominant model: OR = 0.76, 95% CI = 0.49–1.06; recessive model: OR = 0.82, 95% CI = 0.49–1.36). Conclusion BRCA1polymorphisms rs799917 and rs1799966 were not significantly associated with BC risk in this meta-analysis.


2021 ◽  
pp. 026988112110297
Author(s):  
Alberto Forte ◽  
Maurizio Pompili ◽  
Benedetta Imbastaro ◽  
Gabriele Pasquale De Luca ◽  
Martina Mastrangelo ◽  
...  

Background: Clozapine is the only treatment with regulatory-recognition of lowering suicidal risk, at least in schizophrenia patients. It remains uncertain whether such effects extend to other drugs for psychosis. Methods: We searched for reports on rates of suicidal behavior during treatment with clozapine and other modern drugs for psychosis (aripiprazole, olanzapine, risperidone, quetiapine, and ziprasidone) versus comparison or control treatments and analyzed the contrasts by random-effect meta-analysis to obtain pooled odds ratios (ORs) with 95% confidence intervals (CIs). Results: We identified 35 paired comparisons of modern drugs for psychosis versus comparison or control treatments in 18 reports. There was moderate overall superiority of all agents tested over alternatives (OR = 0.522, p = 0.004). With clozapine, this effect was large (OR = 0.229, p < 0.0001) and consistent (7/7 trials), but significant antisuicidal effects were not found with other drugs for psychosis in 28 other trials (OR = 0.941, p = 0.497). Apparent efficacy of specific agents ranked: risperidone ⩾ olanzapine ⩾ aripiprazole ⩾ ziprasidone ⩾ mixed drugs for psychosis ⩾ quetiapine, but none of these differences was significant. Conclusions: An ability of clozapine to reduce risk of suicides and attempts in schizophrenia patients appears to be a unique effect not shared with other modern medicines indicated for schizophrenia or bipolar disorder.


Blood ◽  
2004 ◽  
Vol 104 (11) ◽  
pp. 3722-3722
Author(s):  
Caroline M. Behler ◽  
Starley B. Shade ◽  
Kellan Gregory ◽  
Donald I. Abrams ◽  
Paul A. Volberding

Abstract BACKGROUND: Anemia remains the most common hematologic disorder in human immunodeficiency virus (HIV) infection despite the use of effective antiretroviral therapy, and is associated with decreased quality of life and survival. Hypogonadism is prevalent in advanced HIV disease, however low testosterone levels have not been customarily implicated in HIV-associated anemia. This study was undertaken to determine whether there is a relationship between testosterone levels and androgen use with anemia in HIV, and to characterize other clinical correlates of HIV-associated anemia. METHODS: This cross-sectional study examined the clinical characteristics of 200 HIV positive patients at a public hospital HIV clinic and clinical features associated with anemia. A written questionnaire detailed previous and current medication use, opportunistic infections and malignancies. Hematologic and virologic parameters, testosterone and erythropoietin levels were measured; CD4 count and viral load nadir and peak levels were obtained from the computerized medical record. Anemia was defined as hemoglobin <13.5 g/dL in men and <11.6 g/dL in women. RESULTS: Anemia was present in 24% of women and 28% of men. Anemia was negatively associated with female sex (adjusted OR 0.30, 95% CI 0.11–0.85), current antiretroviral therapy (adjusted OR 0.43, 95% CI 0.20–0.95), current androgen use (adjusted OR 0.20, 95% CI 0.05–0.84) and macrocytosis (adjusted OR 0.23, 95% CI 0.09–0.61). Anemia was positively associated with lymphopenia (adjusted OR 4.0, 95% CI 1.36–11.80), high erythropoieitin levels (adjusted OR 7.73, 95% CI 2.92–20.48) and low testosterone levels (adjusted OR 3.27, 95% CI 1.01–10.60). CONCLUSIONS: Low testosterone levels may have a positive association, and supplemental androgens a negative association with anemia in HIV disease. Predictors of Anemia Unadjusted odds ratios and 95% confidence intervals obtained by logistic regression. Variables that achieved a p-value of <0.1 in tests of interaction were included in a multivariable logistic regression model, which was used to obtain adjusted odds ratios and 95% confidence intervals. N % Anemic Unadjusted OR (95% CI) Adjusted OR (95% CI) Female 38 23.7 0.78 (0.34, 1.78) 0.30 (0.11, 0.85) Male/MTF Transgender 162 28.4 Lymphopenia (<1.0x109/L) 19 57.9 3.05 (1.24, 7.51) 4.00 (1.36, 11.80) Normal Lymphocyte Count 178 24.7 Macrocytosis (MCV>100fL) 71 14.1 0.34 (0.16, 0.74) 0.23 (0.09, 0.61) Normal MCV 117 32.5 Microcytosis (MCV<80fL) 11 63.6 2.91 (0.87, 9.77) 2.02 (0.50, 8.13) Current Antiretroviral Therapy 139 20.1 0.32 (0.17, 0.61) 0.43 (0.20, 0.95) No current Antiretroviral Therapy 61 44.3 Current Androgen Use 23 13.0 0.36 (0.10, 1.27) 0.20 (0.05, 0.84) No Current Androgen Use 171 28.7 EPO x Testosterone Interaction High EPO-High/nl Testosterone 46 41.3 2.50 (1.23, 5.11) 7.73 (2.92, 20.48) Low/nl EPO-Low Testosterone 24 37.5 2.39 (0.89, 6.39) 3.27 (1.01, 10.60) High EPO-Low Testosterone 11 27.3 0.33 (0.04, 2.50) 0.17 (0.012, 2.36) Low/nl EPO-High/nl Testosterone 115 20.9


2011 ◽  
Vol 28 (6) ◽  
pp. 363-367 ◽  
Author(s):  
Catherine M. Laing ◽  
James A. Rankin

Professional registered nurses (RNs) are active participants in seeking and interpreting research evidence. To facilitate knowledge transfer for RNs at the bedside, it behooves researchers to present their findings in a format that facilitates understanding. There is also an expectation that clinicians are capable of interpreting results in a meaningful way. It is important to be able to understand and interpret research reports where statistical methods are used as part of providing the safest and best care for patients. The purpose of this article is to describe the basic concepts of odds ratios and confidence intervals used in research. These statistical measures are used frequently in quantitative research and are often the principle measure of association that is reported. The more comfortable pediatric oncology clinicians are with the interpretation of odds ratios and confidence intervals, the better equipped they will be to bring relevant research results from the “bench” to the bedside.


Sign in / Sign up

Export Citation Format

Share Document