Backward Stepwise Approach Research Articles

Effect of basal metabolic rate on the bone mineral density in middle to old age women in Taiwan

BackgroundBasal metabolic rate (BMR) reflects a combination of cardiopulmonary function and lean body mass resulting from regular physical activity. Though many studies have examined the relationships between bone mineral density (BMD) and body composition, little is known regarding the relationship between BMD and BMR. ObjectiveThe purpose of this study was to investigate the relationship between BMR, anthropometric parameters, body composition and BMD in postmenopausal women in Taiwan. MethodsTwo hundred and eighty-nine women between the ages of 40 and 80 years were included in this cross-sectional study. The following parameters were assessed: height, body weight, total body fat (TBF), BMR, waist-to-hip ratio, grip strength, and back strength. Differences in all variables between osteoporotic and non-osteoporotic women (categorized according to decades in age) were calculated using a one-way analysis of variance (ANOVA) and a Bonferroni post-hoc test. Multiple linear regression with a backward stepwise approach was performed to evaluate the relationship between these measurements and BMD. ResultsAmong women over 50 years of age, those who were non-osteoporotic had higher BMR, BMI, and body fat by comparison to their osteoporotic counterparts (p<0.05 for all). Multiple linear regression revealed that BMR and body fat significantly predicted BMD of the femoral neck (adjusted beta coefficients of 0.304 and 0.190, respectively; p<0.05 for both). BMR and body fat also predicted an increased vertebral BMD (adjusted beta coefficients of 0.310 and 0.141, respectively; p<0.05 for both). ConclusionBMR is closely associated with BMD in elderly persons, and may be a novel target for interventions aimed at preventing the age-related decline in BMD.

Impact of physician specialty on classification of physician-perceived patient severity for patients with osteoarthritis

Purpose: To identify physician and patient characteristics that lead to a patient being perceived as having more severe OA. Methods: All data were analyzed from the Osteoarthritis IX Disease Specific Programme, a large cross-sectional non-hypothesis driven survey conducted in Germany, UK, and USA, collecting robust real-world data. Physicians recruited up to 10 consecutive consulting patients presenting with OA from September 2011 to January 2012. An ordinal logistic regression, controlling for physician clustering, was performed using a backward stepwise approach on preselected physician-reported patient attributes identified in the initial univariate selection process. This process produced an initial model identifying which attributes significantly effected physicians’ rating of OA severity. Refinement to the model included physician specialty and physician attributes (gender, qualification date) and use of diagnostic tools or techniques. McFadden's pseudo R-squared values were used to compare the fit of each model. Results: 363 physicians (220 primary care physicians (PCPs), 95 Rheumatologists (Rheums), 48 orthopedic surgeons (ORURGS)) recruited 3,561patients 24.9% of whom were assessed as mild, 52.0% moderate, and 23.1% severe; of these, 3332 (93.6%) had completed data for analysis. All physician-reported patient characteristics (demographics, pain rating, functionality rating, number of joints, analgesia level, symptoms ever suffered, concomitant condition), with the exception of patient gender, loss of movement, and number of autoimmune diseases, differed significantly between severity groups (p<.0001) at a univariate level. The multivariate model indicated that OSURGs (odds ratio 1.6, 95% Confidence interval 1.2 to 2.2) were more likely to perceive patients as more severe compared to PCPs and RHEUMs combined. The model also indicated that a greater age, body mass index (BMI), use of diagnostics [joint space narrowing based on X-ray, severity of pain symptom(s), impairment in the ability to function (e.g. walk, activities of daily living), severity of joint deterioration], and ever suffering from one or more of the symptoms (pain on movement, pain at rest, nocturnal wakening, loss of movement), are associated with greater severity. McFadden's R-squared increased from 0.35 to 0.37. Conclusions: Patient age, BMI, reported symptoms, disability and radiographic grade influenced physicians’ assessment of OA severity. Controlling for patient factors, OSURGs rated patient's severity as worse compared to RHEUMs and PCPS. Our results suggest that this effect could in part be due to a greater influence of radiographic findings on OA severity rating (potentially deemed more important by OSURGs in severity assessment). Further research is needed to understand other potential explanations for this difference.

Sickle-Cell Trait and Atopic and Inflammatory Disease in Children.

Roughly 9% of African-Americans carry 1 copy of the sickle β-globin allele [sickle-cell trait (SCT)]. The sickle allele may affect the host responses in a way that protects against sequelae of malaria. Besides the association with sporadic episodes of erythrocyte sickling, acute chest syndrome, and sudden death, the SCT is generally regarded as asymptomatic. Separately, and for unclear reasons, the African-Americans have an elevated risk compared to whites for several immune-mediated diseases, including asthma, eczema, and other respiratory conditions. We propose to determine whether the SCT contributes to the elevated risk of these inflammatory conditions. We performed a retrospective cohort study of children with and without the sickle trait to assess the risk for common immune-mediated conditions. The cohorts were individually matched, and multiple logistic regression was used with variables selected using a backward stepwise approach. A second approach (case-control design) assessed the odds of the SCT among African-Americans with and without asthma. We found 2,481 children with and 4,962 matched patients without the SCT in the cohort design. The sickle trait was associated with greater odds for several immune-mediated conditions in a multivariable analysis, but not associated with asthma (odds ratio, OR=1.10, P=0.212). In an adjusted case-control analysis (n=20,000), the sickle trait was weakly associated with asthma [adjusted odds ratio (aOR)=1.46, 1.01-2.13]. In both the designs and in all statistical models used, we found that the SCT was associated with bronchitis (aOR=1.71, 1.09-2.67) and eczema (aOR=1.74, 1.23-2.46). The SCT may contribute to an increased risk for eczema and asthma-like symptoms in the African-American children.

Forecasting peak asthma admissions in London: an application of quantile regression models

Asthma is a chronic condition of great public health concern globally. The associated morbidity, mortality and healthcare utilisation place an enormous burden on healthcare infrastructure and services. This study demonstrates a multistage quantile regression approach to predicting excess demand for health care services in the form of asthma daily admissions in London, using retrospective data from the Hospital Episode Statistics, weather and air quality. Trivariate quantile regression models (QRM) of asthma daily admissions were fitted to a 14-day range of lags of environmental factors, accounting for seasonality in a hold-in sample of the data. Representative lags were pooled to form multivariate predictive models, selected through a systematic backward stepwise reduction approach. Models were cross-validated using a hold-out sample of the data, and their respective root mean square error measures, sensitivity, specificity and predictive values compared. Two of the predictive models were able to detect extreme number of daily asthma admissions at sensitivity levels of 76 % and 62 %, as well as specificities of 66 % and 76 %. Their positive predictive values were slightly higher for the hold-out sample (29 % and 28 %) than for the hold-in model development sample (16 % and 18 %). QRMs can be used in multistage to select suitable variables to forecast extreme asthma events. The associations between asthma and environmental factors, including temperature, ozone and carbon monoxide can be exploited in predicting future events using QRMs.

234: Out-of-Hospital Factors and Their Association With Mortality and Functional Outcomes In Patients With Non-Traumatic Intra-Cranial Hemorrhage

234: Out-of-Hospital Factors and Their Association With Mortality and Functional Outcomes In Patients With Non-Traumatic Intra-Cranial Hemorrhage

The Correlation between Height of Macular Detachment and Visual Outcome in Macula-Off Retinal Detachments of ≤7 Days’ Duration

To determine the relationship between the height of macular detachment and visual recovery after treatment of macula-off retinal detachments (RDs) of < or = 7 days' duration. Prospective comparative case series. Fifty-two eyes of 52 consecutive patients who presented to one institution with macula-off RDs of < or = 7 days' duration were prospectively enrolled in the study. Gender, age, lens status, duration of macular detachment, and presenting visual acuity (VA) were recorded for all patients. Each patient underwent a 3-dimensional B-scan ultrasound examination at the time of presentation. The height of macular detachment was calculated using the mean measured distance between the retinal pigment epithelium and the outer neurosensory retina 3, 4, and 5 mm temporal to the center of the optic nerve head. All patients underwent primarily successful reattachment surgery within 7 days of macular detachment. Final vision was recorded with a minimum of 6 months' follow-up. Postoperative VA evaluated as a continuous variable. For the 52 subjects, a direct correlation between lower height of macular detachment and better final postoperative vision (dependent variable) was observed with linear regression (slope, 0.114; 95% confidence interval, 0.022-0.206; P = 0.016). Multivariate linear regression analysis with a backward stepwise approach did not demonstrate any significant association between postoperative vision and predictor variables (gender, age, lens status, duration of macular detachment, macular height, and preoperative vision), except for macular height. Lower height of macular detachment correlates with better visual recovery after treatment of macula-off RDs of < or = 7 days' duration.