Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Linear Regression”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 721 records · Page 40Linked to original sources
PubMed2006

Variants in the glutamate-cysteine-ligase gene are associated with cystic fibrosis lung disease.

BACKGROUND: Chronic progressive lung disease is the most serious complication of cystic fibrosis (CF). Glutathione plays an important role in the protection of the CF lung against oxidant-induced lung injury. OBJECTIVES: We hypothesized that a polymorphism in a novel candidate gene that regulates glutathione synthesis might influence CF lung disease. METHODS: In a cross-sectional study, subjects were recruited from CF clinics in Seattle and multiple centers in Canada. We tested for an association between CF lung disease and a functional polymorphism in the glutamate-cysteine ligase catalytic subunit (GCLC) gene. Multiple linear regression was used to test for association between polymorphisms of GCLC and severity of CF lung disease while adjusting for age, Pseudomonas aeruginosa infection, and cystic fibrosis transmembrane conductance regulator (CFTR) genotype. Analysis was repeated for patients with CF stratified by CFTR genotype. MEASUREMENTS AND MAIN RESULTS: A total of 440 subjects with CF participated in the study (51% male; mean [+/- SD] age, 26 +/- 11 yr; mean FEV(1), 62 +/- 28% predicted). In the total population, there was a trend toward an association between GCLC genotypes and CF lung disease (linear regression coefficient [SEM], 1.68 [1.0]; p = 0.097). In the stratified analysis, there was a highly significant association between GCLC genotype and CF lung function in subjects with a milder CFTR genotype (linear regression coefficient [SEM], 5.5 (1.7); p = 0.001). CONCLUSIONS: In patients with CF with a milder CFTR genotype, there is a strong association between functional polymorphisms of the GCLC gene and CF lung disease severity.

Adolescent↗
PubMed2006

Correction for errors in measuring adherence to prenatal multivitamin/mineral supplement use among low-income women.

Adherence to prenatal multivitamin/mineral supplement use is often measured by self-reports or pill counts. Although both measures were shown to overestimate adherence, measurement error is rarely considered. In this study, we examined measurement error in adherence to prenatal supplement use among pregnant women and demonstrated a calibration method to adjust for error. In a validation subsample (n=51) from a larger clinical study of supplementation, adherence was assessed by self-reports, pill counts, and a Medication Event Monitoring System (MEMS) bottle cap that recorded the date and time of each opening of the pill bottle. Mean adherence in the validation sample as measured by the MEMS (the gold standard) was 68%; thus, adherence measured by self-report (77%) and pill count (84%) reflected overestimation. The Pearson correlation coefficients of self-reports and pill counts to MEMS were 0.35 and 0.62, respectively. When adherence was defined as taking >or=75% of the pills prescribed, sensitivity and specificity were greater for pill counts (93 and 52%, respectively) than for self-reports (88 and 44%). The regression coefficient for pill count adherence from a linear regression on MEMS adherence was applied to pill counts from a larger sample (n=244). The adjustment significantly lowered the estimate of adherence from 74 to 64% (P<0.001) in this larger sample. In conclusion, our data show that both self-reports and pill counts overestimate adherence and that linear regression in comparison to an external standard such as MEMS can be used to correct for measurement error in adherence.

Adolescent↗
PubMed1999

Determination of copolymer ratios of poly(lactide-co-glycolide) using near-infrared spectroscopy.

The near infrared (NIR) spectroscopic technique was used to determine copolymer ratios of polylactide-co-glycolide samples. Appropriate quantities of DL-polylactic acid and lactic-co-glycolic acid polymers with 86:14, 75:25, 64:36 and 52:48 lactide to glycolide ratios were dissolved in methylene chloride to obtain 5% (w/w) solutions. NIR spectra of the samples were obtained from the solutions using a Polyol Analyzer operated in the transmittance mode. Linear regression calibration models were generated at 2130 and 2288 nm from the second derivative spectral data obtained from the NIR technique. The lowest and highest standard errors of calibration (SEC) at 2130 nm were 1.29 and 1.63%, whereas those obtained from the calibration models generated at 2288 nm were 2.00 and 2.03%, respectively. Partial least squares (PLS) calibration models were also generated from the second derivative spectral data from 1100 to 2500 nm. The lowest and the highest SEC for the models were 1.46 and 1.53%, respectively. The calibration models were then used to predict the lactide contents of the unknown (test) samples. The highest percent error of prediction was 2.56% for samples with 86% lactide content when the linear regression calibration at 2130 nm was used, whereas the highest percent error of prediction was 1.56% for samples with 64% lactide content when the linear regression calibration at 2288 nm was used. The highest percent error of prediction was 1.73% for samples with 75% lactide content when the two-factor PLS calibration model was used.

Calibration↗
PubMedDate not supplied

How significant is the difference between drug doses influencing the threshold for electroconvulsions?

In this study, three computational methods (log-probit, one-way analysis of variance and linear regression analysis) were applied to statistically determine and compare a minimal dose of the antiepileptic drug--levetiracetam (LEV), which significantly increased the threshold for maximal electroconvulsions in mice. Results analyzed with log-probit method revealed that such a dose for LEV was 50 mg/kg. In contrast, one-way ANOVA indicated that the minimal dose of LEV, fulfilling the criterion of the significant increment in the threshold for electroconvulsions, was 100 mg/kg. Moreover, linear regression analysis allowed for the determination of threshold increasing doses by 20% (TID20) and 50% (TID50) for LEV, which amounted to 44 mg/kg and 150 mg/kg. This comparative study demonstrated distinct differences between the electroconvulsive threshold-influencing doses of LEV calculated with these three methods, and only statistical evaluation of data using one-way ANOVA provided the most adequate determination of statistical significance. The application of log-probit method and linear regression analysis in the search for significant drug-dose effects should be substantially restricted to specific experimental conditions.

Analysis of Variance↗
PubMed1996

Chronic viral hepatitis enhances the risk of infection but not acute rejection in renal transplant recipients.

To assess the impact of chronic viral hepatitis on host immune response, we analyzed the incidence of acute rejection and the frequency of infections in 86 patients infected with hepatitis B and C viruses and had developed clinical evidence of chronic liver disease and 1283 control patients who were transplanted at our center during the same period, but had no evidence of chronic viral hepatitis. To compare the mean number of rejections and the mean number of infections between the two groups, we used multivariate linear regression analysis, which allowed us to adjust simultaneously for the effects of 10 other risk variables with potential impact on graft rejection and posttransplant infection. During a mean follow up of 5.3+/-5.2 years, 62% of hepatitis patients and 54% of control patients had experienced an acute rejection (P=NS). The mean rejections/patient in the hepatitis group was 1.3+/-0.14 versus 1.03+/-0.03 in control (P=NS). In the linear regression analysis, the number of acute rejections in the hepatitis group was 0.16 higher than in control (P=NS). With reference to infection, 84% of hepatitis patients experienced an infectious complication in the posttransplant period, compared with 75% in the control (P=0.05). The mean number of infections/patient was 5.7+/-0.73 in the hepatitis group compared with 3.9+/-0.14 in the control group (P=0.002). The linear regression model had shown that the hepatitis group had a relative increase of 1.18 infections/pt, compared with control. Of the different sites of infection, the hepatitis group had a significant increase in bloodstream (0.48+/-0.08 vs. 0.25+/-0.02) P=0.003; pulmonary (0.60+/-0.09 vs. 0.38+/-0.03) P=0.03; and CNS infections (0.08+/-0.03 vs. 0.02+/-0.004) P=0.05 compared with control. Among the different microorganisms causing infection, the hepatitis patients had a significant increase in gram negative bacterial infections compared with the control group (74% vs. 61%) P=0.04. Our data suggest that chronic viral hepatitis is associated with a significant increase in overall infections, and that of potentially fatal infections involving CNS, lungs and bloodstream. Since there is no significant increase in the rate of graft rejection, one could consider a cautious reduction in the doses of maintenance immunosuppressive agents in renal transplant patients with chronic viral hepatitis. The reduced immunosuppression may in turn lower the death rate from sepsis and progressive hepatic failure.

Chronic Disease↗
PubMed1998

The association between diabetic complications and exercise capacity in NIDDM patients.

OBJECTIVE: Exercise capacity has been used as a noninvasive parameter for predicting cardiovascular events. It has been demonstrated previously in NIDDM patients that several risk factors (i.e., obesity, smoking, hypertension, and African-American race) are associated with an impaired exercise capacity. We studied 265 male and 154 female NIDDM patients who underwent graded exercise testing with expired gas analyses to determine the possible influences of diabetic neuropathy, nephropathy, and retinopathy on exercise capacity. RESEARCH DESIGN AND METHODS: Univariate and multiple linear regression analyses were performed to determine the relationship between diabetic neuropathy, urinary albumin excretion (UAE), and retinopathy with respect to peak oxygen consumption (VO2). Neuropathy was assessed by neurological symptom and disability scores, autonomic function testing, and quantitative sensory exams involving thermal and vibratory sensation. Three categories of UAE were used: normal albuminuria (< 20 micrograms/min), microalbuminuria (20-200 micrograms/min), and overt albuminuria (> 200 micrograms/min). Retinopathy was assessed by stereoscopic fundus photographs. Multiple linear regression analyses were then performed controlling for age, sex, length of diagnosed diabetes, duration of hypertension, race and ethnicity, GHb, BMI, and smoking to determine whether there was an independent effect of these diabetic complications on exercise capacity. RESULTS: Univariate analyses revealed that the presence of diabetic retinopathy (P = 0.03), neuropathy (P = 0.002), microalbuminuria (P = 0.04), and overt albuminuria (P = 0.06) were associated with a lower peak VO2. Multiple linear regression analyses were performed to determine independent relationships with peak VO2. The results revealed that increasing retinopathy stage (Parameter estimate [PE] = -0.59 +/- 0.3 ml.kg-1.min-1; P = 0.026) and increasing UAE stage (PE = -0.62 +/- 0.3 ml.kg-1.min-1; P = 0.04) were associated with a decrease in peak VO2. CONCLUSIONS: In the present study of NIDDM subjects, a significant independent association was demonstrated between diabetic nephropathy and retinopathy with exercise capacity. These results were obtained controlling for age, sex, length of diagnosed diabetes, hypertension, race, and BMI. Thus the findings in this large NIDDM population without a history of coronary artery disease indicate a potential pathogenic relationship between microvascular disease and exercise capacity.

Adult↗
PubMed1987

Evaluation of three dosage-prediction methods for initial in-hospital stabilization of warfarin therapy.

Three dosage-prediction methods for initial in-hospital stabilization of warfarin therapy were evaluated. Adult inpatients who had received warfarin sodium 10 mg daily for less than three days were eligible for the study. After receiving their third warfarin dose, patients were randomly assigned to have their warfarin dosages adjusted using one of three dosage-prediction methods: by analog computer (n = 31), linear regression (n = 22), or empiric dosing by the physician (n = 34). A prothrombin time (PT) ratio (patient PT divided by control PT) between 1.3 and 2.5 was considered to be in the therapeutic range. For patients who achieved a stable PT ratio (defined as a PT ratio between 1.3 and 2.5 that varied by less than 0.05 on two consecutive days or by less than 0.1 on three consecutive days without a dosage change) before discharge, the number of days (time to stabilization) from administration of the first warfarin dose to achievement of the warfarin dosage that produced a stable PT ratio (stabilization dosage) was compared. A total of 54 patients met the study criteria for a stable PT ratio before hospital discharge (analog computer, n = 20; linear regression, n = 15; empiric dosing, n = 19). The mean times to stabilization were 6.8 days in the analog-computer group, 7.3 days in the linear-regression group, and 8.4 days in the empiric-dosing group; these times were not significantly different. All 20 stabilized patients in the analog-computer group achieved a stable PT ratio by the fourth dosage prediction.(ABSTRACT TRUNCATED AT 250 WORDS)

Aged↗
PubMed1980

Application of conditional probability analysis to the clinical diagnosis of coronary artery disease.

Analysis of multiple noninvasive tests offers the promise of more accurate diagnosis of coronary artery disease, but discordant test responses can occur frequently and, when observed, result in diagnostic uncertainty. Accordingly, 43 patients undergoing diagnostic coronary angiography were evaluated by noninvasive testing and the results subjected to analysis using Bayes' theorem of conditional probability. The procedures used included electrocardiographic stress testing for detection of exercise-induced ST segment depression, cardiokymographic stress testing for detection of exercise-induced precordial dyskinesis, myocardial perfusion scintigraphy for detection of exercise-induced relative regional hypoperfusion, and cardiac fluoroscopy for detection of coronary artery calcification. The probability for coronary artery disease was estimated by Bayes' theorem from each patient's age, sex, and symptom classification, and from the observed test responses. This analysis revealed a significant linear correlation between the predicted probability for coronary artery disease and the observed prevalence of angiographic disease over the entire range of probability from 0 to 100% (P less than 0.001 by linear regression). The 12 patients without angiographic disease had a mean posttest likelihood of only 7.0 +/- 2.6% despite the fact that 13 of the 60 historical and test responses were falsely "positive." In contrast, the mean posttest likelihood was 94.1 +/- 2.8% in the 31 patients with angiographic coronary artery disease, although 45 of the 155 historical and test responses were falsely "negative." In 8 of the 12 normal patients, the final posttest likelihood was under 10% and in 26 of the 31 coronary artery disease patients, it was over 90%. These estimates also correlated well with the pooled clinical judgment of five experienced cardiologists (P less than 0.001 by linear regression). The observed change in probability for disease for each of the 15 different test combinations correlated with their information content predicted according to Shannon's theorem (P less than 0.001 by linear regression). These results support the use of probability analysis in the clinical diagnosis of coronary artery disease and provide a formal basis for comparing the relative diagnostic effectiveness and cost-effectiveness of different test combinations.

Adult↗
PubMed1998

Pork carcass composition derived from a neural network model of electromagnetic scans.

We used an advanced computer logic system (NETS 3.0) to decipher electromagnetic (EM) scans in lieu of traditional linear regression for estimation of pork carcass composition. Fifty EM scans of pork carcasses were obtained on-line (prerigor) at a swine slaughter facility. Right sides were cut into wholesale parts and dissected into fat, lean, and bone to obtain total dissected carcass and primal cut lean. In this study, the input layer consisted of 81 nodes (80-point EM scan curve and warm carcass weight), one hidden layer of 42 nodes, and an output layer consisting of one node, which were run separately for outputs of ham, loin, or shoulder lean. The hidden layer connected to the output of total lean contained 50 nodes. Thirty-five scans were used for training of the network. The new network was then tested with 15 previously unseen input/output pairs. Separate neural networks were developed for the estimation of dissected total carcass, ham, loin, and shoulder lean. The NETS configuration improved on linear regression equations for estimation of total carcass lean by .31 kg, ham lean by .284 kg, and shoulder lean by .148 kg. Our results show that advanced computer logic systems have the capacity to improve upon traditional linear regression equations for prediction of pork carcass composition.

Animals↗
PubMed2004

Obstructive lower urinary tract symptoms correlate with erectile dysfunction.

OBJECTIVES: To examine whether the severity of lower urinary tract symptoms (LUTS), when controlled for other factors, is related to erectile dysfunction (ED) in the male veteran population. Early evidence suggests that LUTS may be associated with ED in men. METHODS: A total of 181 male veterans were prospectively entered into the study. They completed the Sexual Health Inventory for Men (SHIM) and International Prostate Symptom Score (IPSS) questionnaires. Of the 181 men, 144 also underwent uroflowmetry, including determination of the postvoid residual urine volume. Demographic and medical history data were recorded. Pearson correlation coefficients and multiple linear regression analysis were used to examine the relationship between LUTS and ED, as well as the effects of age and comorbidities. RESULTS: The correlation coefficient (r) for the SHIM score with the total IPSS was -0.17 (P = 0.023); with the obstructive IPSS, it was -0.20 (P = 0.006); and with the irritative IPSS, -0.05 (P = 0.492). Age was the only other factor with a statistically significant correlation with the SHIM score (r = -0.23, P = 0.002). Multiple linear regression modeling showed obstructive IPSS (P = 0.001) and depression (P = 0.017) to be the only statistically significant predictors of the SHIM score. A consistent negative correlation was found between obstructive IPSS and the SHIM score across age groups, with the strongest effect for men aged 60 to 70 years (r = -0.412, P = 0.003). CONCLUSIONS: Obstructive LUTS correlated with, and were predictive of, ED, even after controlling for age and comorbidities. Although age correlated with ED, it did not add to the power of the multiple linear regression model composed of obstructive IPSS and depression.

Adult↗
PubMed2003

Bayesian hierarchical approach to estimate insulin sensitivity by minimal model.

We adopted Bayesian analysis in combination with hierarchical (population) modelling to estimate simultaneously population and individual insulin sensitivity (SI) and glucose effectiveness (SG) with the minimal model of glucose kinetics using data collected during insulin-modified intravenous glucose tolerance test (IVGTT) and made comparison with the standard non-linear regression analysis. After fasting overnight, subjects with newly presenting Type II diabetes according to World Health Organization criteria (n =65; 53 males, 12 females; age, 54 +/- 9 years; body mass index, 30.4 +/- 5.2 kg/m2; means+/-S.D.) underwent IVGTT consisting of a 0.3 g of glucose bolus/kg of body weight given at time zero for 2 min, followed by 0.05 unit of insulin/kg of body weight at 20 min. Bayesian inference was carried out using vague prior distributions and log-normal distributions to guarantee non-negativity and, thus, physiological plausibility of model parameters and associated credible intervals. Bayesian analysis gave estimates of SI in all subjects. Non-linear regression analysis failed in four cases, where Bayesian analysis-derived SI was located in the lower quartile and was estimated with lower precision. The population means of SI and SG provided by Bayesian analysis and non-linear regression were identical, but the interquartile range given by Bayesian analysis was tighter by approx. 20% for SI and by approx. 15% for SG. Individual insulin sensitivities estimated by the two methods were highly correlated ( rS=0.98; P <0.001). However, the correlation in the lower 20% centile of the insulin-sensitivity range was significantly lower than the correlation in the upper 80% centile ( rS=0.71 compared with rS=0.99; P <0.001). We conclude that the Bayesian hierarchical analysis is an appealing method to estimate SI and SG, as it avoids parameter estimation failures, and should be considered when investigating insulin-resistant subjects.

Adult↗
PubMed2005

Lack of association between polymorphism of the human cyclic GMP-dependent protein kinase gene and obesity.

OBJECTIVE: To investigate whether genetic variation in the cyclic GMP-dependent protein kinase gene (PRKG1) is associated with obesity. METHODS: The study included 143 individuals from New York City area, NY, USA. The subjects were sampled on the basis of body mass index (BMI): obese (BMI ranging from 33.8 to 89.5 kg/m(2)), and nonobese (BMI ranging from 16.0 to 29.4 kg/m(2)). The association between C2276T polymorphism in PRKG1 gene and obesity was tested using linear regression analysis. RESULTS: BMI levels were predicted by linear regression models adjusted for demographic factors. An analysis was performed twice: in individuals of all ethnic backgrounds and in European-Americans only. In both cases, genotype did not have a significant effect. CONCLUSION: We found no evidence that the C2276T polymorphism in the PKRG1 gene is associated with obesity.

Adolescent↗
PubMed2006

Comparison between GDx VCC parameter and achromatic perimetry in glaucoma patients.

PURPOSE: To evaluate the correlation between polarimetry parameters and standard threshold perimetry indices. PATIENT AND METHODS: One hundred and eight consecutive glaucomatous patients were recruited in this study. Glaucomatous patients were classified when an abnormal visual field and/or an abnormal optic nerve head were present. One eye was chosen randomly from all the subjects. All optic nerve heads were examined with the GDx variable corneal compensator (VCC) (Laser Diagnostic Technologies, Inc, San Diego, CA) and visual fields were assessed by Humphrey Perimeter, program 24-2, SITA standard (Humphrey Instrument, Inc, San Leandro, CA). GDx VCC parameters and visual field indices were considered. Correlation coefficient was used to compare the 2 sets of data. A linear regression model was also used to calculate the independent contribution of each GDx VCC parameter and to determine which parameter was the most predictive of visual field damage. RESULTS: The mean age (+/-standard deviation) was 61+/-10.3, the average of the mean deviation (MD) was -2.31+/-3.2 dB and the mean pattern standard deviation (PSD) was 3.32+/-2.83 dB. Significant (P<0.05) correlation was found between MD and the nerve fiber indicator (NFI) (r=-0.35), the superior average (r=0.28), the Temporal-Superior Nasal-Inferior Temporal average (r=0.24) and the inferior average (r=0.21). The linear regression model found NFI to be the predictor variable of MD. Significant (P<0.05) correlation was found between PSD and the NFI (r=0.39), the superior and the inferior average (r=0.30), and the Temporal-Superior Nasal-Inferior Temporal average (r=0.29). The linear regression model found NFI to be the predictor variable of PSD. CONCLUSIONS: Some of the GDx VCC indices were significantly correlated to visual field indices in glaucomatous patients, but NFI was the best indicator of visual field damage.

Adult↗
PubMed2000

Blood flow rate effects in continuous venovenous hemodiafiltration on blood urea nitrogen and creatinine reduction.

The effect of blood flow rates (BFRs) in continuous venovenous hemodiafiltration (CVVHDF) on the clearances of blood urea nitrogen (BUN) and creatinine was studied using a retrospective chart review. Fifteen patients, 9 males and 6 females aged 50-85 years, were included in the study. Each patient had a primary cardiovascular diagnosis with acute renal failure (ARF) and was treated with CVVHDF on the Gambro PRISMA continuous renal replacement therapy (CRRT) machine using the COBE AN69 filter. Independent sample t-test, simple linear regression, ANOVA, and multiple comparisons were used. Simple linear regression revealed that BUN, creatinine, and difference in creatinine (DIFFCRTN) were best predicted by the number of hours of treatment. For every hour of treatment the BUN decreased by .08 mg/dl, the creatinine decreased by .01 mg/dl, and the DIFFCRTN increased by .01 mg/dl. The study concluded that patients on CVVHDF for greater than 72 hours had the greatest changes in BUN and creatinine when a BFR between 135-145 cc/min was used.

Acute Kidney Injury↗
PubMed2000

How to program rate responsive pacemakers.

Because oxygen uptake (VO2) increases linearly with heart rate during exercise, the oxygen pulse reserve (OPR) method (VO2 reserve divided by heart rate reserve) may provide a valid guide for rate responsive parameter tailoring. Using custom-made software (Pacing Rate Profile Software [PRPS]) it is possible to predict the exercise pacing rate profile with significant accuracy, according to the patient's functional class when ergospirometry apparatus is not available for a cardiopulmonary stress test (CPX). PRPS for Windows is based on the OPR method and some known workload/metabolic cost of exercise relationships during effort. The present study had two aims; first, to evaluate the reliability of PRPS in accurately predicting pacing rate profiles; and second, the suitability of activity and metabolic rate responsive sensors in supplying pacing rates sufficiently near to those predicted using CPX or PRPS. To test the reliability of PRPS we studied 244 patients, NYHA Class I-II, under two different stress test protocols. In one, the bicycle protocol (25 W, 2-minute steps), we tested 137 normal patients (94 men and 43 women, mean age 67 +/- 15 years). Sixty-eight of these were simultaneously CPX tested. PRPS predicted pacing rates were matched against the patients' sinus rhythms or their theoretical CPX measured VO2 heart rates (OPR method). Linear regression analysis was highly significant (r = 0.93 and r = 0.97, respectively). The other, the treadmill protocol, consisted of three different protocols. (1) Speed Incremental Treadmill Stress Test (SITST): 57 patients underwent CPX (33 men and 24 women, mean age 67 +/- 15 years, NYHA Class I-II). All had been pacemaker implanted for SSS and/or advanced atrioventricular block (AVB). PRPS pacing rates were matched against CPX VO2 OPR calculated heart rates (r = 0.93), (linear regression analysis). (2) CAEP: 30 patients underwent CPX (26 men and 4 women, mean age 61 +/- 11 years, NYHA Class I-II). Thirteen of them had been pacemaker implanted for SSS and/or advanced AVB. In all 30 patients the PRPS rates were matched against CPX VO2 calculated rates (r = 0.90). In the 17 normal nonimplanted patients, the PRPS rates were also matched against sinus rhythms, (r = 0.80). (3) Weber: 20 patients underwent CPX (16 men and 4 women, mean age 68 +/- 8 years, NYHA Class I-II). As above, in six normal nonimplanted patients, statistical analysis between PRPS rates and sinus rhythms was performed (r = 0.89). The comparison between PRPS theoretical pacing rates and VO2 predicted rates in all 20 patients was also statistically significant (r = 0.93). Finally, to test the reliability of PRPS also in NYHA Class III-IV patients, we tested 22 implanted patients (15 men and 7 women, mean age 70 +/- 9 years) and compared PRPS predicted rates against VO2 CPX measured rates (r = 0.92). To determine if the wide variety of RR pacers were able to supply pacing rates near to those predicted, whether by means of CPX or PRPS, we studied a total of 89 patients: 49 of these (26 men and 23 women, mean age 66 +/- 12 years) had been implanted with activity sensors; 12 patients (11 men and one woman, mean age 70 +/- 7 years) had been implanted with metabolic sensors, and finally 28 patients (19 men and 9 women, mean age 70 +/- 12 years) had been implanted with dual sensors (activity + QT or minute ventilation). Linear regression analysis showed r = 0.93 for activity sensors, r = 0.94 for metabolic sensors, and r = 0.92 for dual sensor. In conclusion, when rate responsive pacing causes symptoms or functional impairment, physicians must provide a personalized rate response tailoring derived from precise, simple physiological testing. OPR is an easy physiological method for tailoring rate response settings, suitable for activity and metabolic sensors. When ergospirometry apparatus is not available, PRPS can successfully replace CPX testing for tailoring.

Aged↗
PubMed1998

Mixed dentition analysis for Hong Kong Chinese.

Using simple linear regression analyses, prediction equations for the combined mesiodistal crown diameters of canines and premolars based on lower incisor size were generated from 97 Hong Kong Chinese (51 males and 46 females, average age 12.31 years) out of a sample of 112. The mesiodistal crown diameters of the permanent teeth were measured using calipers and recorded to the nearest 0.01 mm. Significant sex differences were found for the combined diameters of the canine-premolar segments. The coefficients of correlation between combined diameters of canines and premolars and lower incisors ranged from 0.65 to 0.79. Significant sex differences of the regression equations were found and thus four simple linear regression equations were generated. Coefficients for the slope ranged from 0.58 to 0.66, and coefficients for the intercept ranged from 6.66 to 8.82. The R2 values, standard errors of estimate, and absolute mean errors revealed that prediction models for females were less precise than those for males. Probability tables were constructed from the results of the present study. The prediction equations were found to differ from those of Tanaka and Johnston. Accuracy in the mixed dentition analysis for southern Chinese would be improved by applying the prediction equations or probability tables generated from the present study.

Asian People↗
PubMed2001

Relationships between use of television during meals and children's food consumption patterns.

OBJECTIVE: We examined relationships between the presence of television during meals and children's food consumption patterns to test whether children's overall food consumption patterns, including foods not normally advertised, vary systematically with the extent to which television is part of normal mealtime routines. METHODS: Ninety-one parent-child pairs from suburbs adjacent to Washington, DC, recruited via advertisements and word of mouth, participated. Children were in the fourth, fifth, or sixth grades. Socioeconomic data and information on television use were collected during survey interviews. Three nonconsecutive 24-hour dietary recalls, conducted with each child, were used to construct nutrient and food intake outcome variables. Independent sample t tests were used to compare mean food and nutrient intakes of children from families in which the television was usually on during 2 or more meals (n = 41) to those of children from families in which the television was either never on or only on during one meal (n = 50). Multiple linear regression models, controlling for socioeconomic factors and other covariates, were used to test strength of associations between television and children's consumption of food groups and nutrients. RESULTS: Children from families with high television use derived, on average, 6% more of their total daily energy intake from meats; 5% more from pizza, salty snacks, and soda; and nearly 5% less of their energy intake from fruits, vegetables, and juices than did children from families with low television use. Associations between television and children's consumption of food groups remained statistically significant in multiple linear regression models that controlled for socioeconomic factors and other covariates. Children from high television families derived less of their total energy from carbohydrate and consumed twice as much caffeine as children from low television families. There continued to be a significant association between television and children's consumption of caffeine when these relationships were tested in multiple linear regression models. CONCLUSION: The dietary patterns of children from families in which television viewing is a normal part of meal routines may include fewer fruits and vegetables and more pizzas, snack foods, and sodas than the dietary patterns of children from families in which television viewing and eating are separate activities.

Adult↗
PubMed1997

Importance of using rigorous statistical methods to analyze low energy laser experimental data: Part two.

BACKGROUND AND OBJECTIVE: Numerous authors have reported successful alteration of peripheral nerve action potential characteristics through application of low energy laser irradiation (LELI). The statistical analysis that accompanies many of these reports frequently does not account for the special nature of the data generated in typical LELI experiments. The objective of this study was to evaluate the application of repeated measures linear regression techniques to the analysis of this type of data. Issues of analyzing raw versus normalized data, proper accounting for correlation between measurements, and discrete time point hypothesis testing were addressed. STUDY DESIGN/MATERIALS AND METHODS: The data analyzed in this work were generated from an experiment in which in vitro frog sciatic nerves were irradiated with a helium-neon laser using a variety of treatment protocols. Compound action potential (CAP) amplitude, latency, depolarization rate, and repolarization rate were recorded at 1-minute intervals for 135 minutes for each nerve. Laser-induced changes in CAP parameters were analyzed using various repeated measures linear regression models. RESULTS: The findings of statistical significance were highly dependent on the rigor of the regression model applied. Application of the same regression model to raw and normalized data produced different findings of significance. Determination of significant contrasts was highly dependent on how well the regression model accounted for the correlation between repeated measurements made on the same nerve. In general, models that failed to account adequately for this correlation produced more findings of significant contrasts than increasingly rigorous models. Finally, discrete time point hypothesis testing on normalized data can suggest improper statistical conclusions if the proper correlation structure is not applied to the data set. CONCLUSION: Linear regression analysis offers advantages over discrete time point hypothesis testing in the analysis of highly correlated serial data of this type. Trends in the behavior of the measured parameters are evident, rigorous accounting for correlation between measurements is facilitated, and hypothesis testing is highly flexible.

Action Potentials↗