SEARCH · Search PubMed

Results for “linear mixed model”

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.

SourceResearch connectionFrom yearThrough year

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

At least 1,351 records · Page 75Linked to original sources

PubMed2006

The impact of migraine on daily activities: effect of topiramate compared with placebo.

OBJECTIVE: Assess the impact of migraine preventive therapy on patient-reported routine daily activities using the Migraine Specific Questionnaire (MSQ) and the Medical Outcomes Study Short Form-36 (SF-36) in patients with migraine who participated in a 26-week, randomized, double-blind, placebo-controlled trial of topiramate for migraine prevention. METHODS: Patients were required to have 3-12 migraines and < or = 15 headache days/month during the baseline phase. Patients who failed > 2 adequate regimens of migraine preventive therapy were excluded. MSQ and SF-36 data were collected at baseline, weeks 8, 16, and 26 from 469 patients receiving either topiramate 50, 100, or 200 mg/day or placebo. Patients entered a double-blind, 8-week titration period followed by an 18-week maintenance period. Two activity-related MSQ domains (Role Restrictive [RR] and Role Prevention [RP]) and two activity-related SF-36 domains (Role Physical [SF-36-RP] and Vitality [SF-36-VT]) were prospectively designated as the outcome measures. Changes in MSQ and SF-36 scores during the double-blind phase relative to prospective baseline scores were compared between topiramate- and placebo-treated groups. Specifically, a mixed-effect piecewise linear regression model was used to estimate average domain score over time, and areas under the domain-over-time curve (AUC) were compared using a 2-sided t-test, with multiplicity adjustment. RESULTS: In the intent-to-treat population (N = 469), topiramate (all doses) significantly improved mean MSQ-RR domain scores versus placebo (topiramate 50 mg/day, p = 0.035; topiramate 100 mg/day; p < 0.001; topiramate 200 mg/day, p = 0.001). Topiramate-associated improvements in mean MSQ-RP domain scores were significant versus placebo only for topiramate 100 mg/day (p = 0.045). SF-36-RP and SF-36-VT domain scores improved (not significant versus placebo) for topiramate 100 and 200 mg/day. Changes in these MSQ and SF-36 domain scores significantly correlated with changes in mean monthly migraine frequency. CONCLUSION: Improvements in patient-reported outcomes specific for migraine (measured by the MSQ) were significantly better for patients receiving topiramate than for those receiving placebo. Improvements in the prospectively selected MSQ and SF-36 domains were significantly correlated with the decrease in mean monthly migraine frequency observed with topiramate treatment.

Activities of Daily Living↗

PubMed2026

Blood mitochondrial heteroplasmic variants and cognitive performance in late midlife: REGARDS study.

BACKGROUND: Studies linking mitochondrial DNA (mtDNA) variants to cognition yielded inconsistent findings, and the underlying mechanisms remain unclear. We investigated whether mtDNA heteroplasmic variants were associated with cognitive outcomes, including the Montreal Cognitive Assessment (MoCA), in 197 late midlife adults from the Reasons for Geographic and Racial Differences in Stroke (REGARDS) cohort with complete data. METHODS: MtDNA was sequenced from blood using targeted deep sequencing. Adjusted linear and mixed-effects models examined the associations by functional regions, genes, total variant burden, nonsynonymous variants, and control regions. RESULTS: Heteroplasmic variants in the control region (β = -0.44, 95% CI: -0.83, -0.05, p = 0.027) and transfer RNA (tRNA) genes (β = -1.34, 95% CI: -2.58, -0.11, p = 0.034) were associated with MoCA baseline scores. Individual variants in cytochrome c oxidase subunit 1 (CO1) (β = -1.51, 95% CI: -2.54, -0.47, p = 0.005), NADH dehydrogenase subunit 1 (ND1) (β = -2.63, 95% CI: -4.56, -0.70, p = 0.008), and Displacement Loop (D-LOOP2) (β = -2.25, 95% CI: -4.20, -0.30, p = 0.025) was associated with reduced baseline MoCA scores. The ND6 (β = −1.23, 95% CI: −2.09, − 0.37, p = 0.006), ND4 (β = −1.11, 95% CI: −2.02, − 0.20, p = 0.018), ATP Synthase Membrane Subunit 8 (ATP8; β = −1.38, 95% CI: −2.63, − 0.13, p = 0.031), and D-LOOP1 (β = −0.61, 95% CI: −1.20, − 0.01, p = 0.045) genes suggested a potential association with executive function. Longitudinal Animal Fluency Test (AFT) scores were inversely associated with heteroplasmic variants in coding regions (β = -0.10, 95% CI: -0.19, -0.006, p = 0.049), the total number of variants (β = -0.06, 95% CI: -0.11, -0.003, p = 0.037) and total nonsynonymous variants (β = -0.11, 95% CI: -0.21, -0.01, p = 0.040). Variants in the control region were associated with the greatest decline in verbal fluency (β = −0.20, 95% CI: −0.39 to − 0.002, p = 0.049). No associations were observed between mitochondrial variants and verbal memory performance or the MoCA composite scores. CONCLUSIONS: Our study indicates that mitochondrial variants measured in blood may provide insight into cognitive function during midlife. However, additional studies are needed to validate these associations and to address potential power limitations in our study.

Humans↗

PubMed2004

Education and the course of cognitive decline in Alzheimer disease.

OBJECTIVE: To test the hypothesis that higher level of education is related to more rapid cognitive decline in Alzheimer disease (AD). METHODS: Participants are older persons with clinically diagnosed AD recruited from health care facilities in the Chicago area. At 6-month intervals for up to 4 years, they underwent uniform structured clinical evaluations that included administration of nine cognitive performance tests from which a composite measure of global cognition was derived. Analyses are based on 494 persons with follow-up data (89.3% of those eligible). In mixed models that allowed for linear and nonlinear decline, the authors first accounted for the effects of age on cognition and then tested the relation of education to rate of cognitive decline. RESULTS: Global cognitive decline had linear and nonlinear components, resulting in a gradually accelerating course of decline. Age was related to linear but not nonlinear decline, with more rapid decline observed in younger compared with older persons. Higher educational level was related to more rapid global cognitive decline, as hypothesized, with education related to the nonlinear but not the linear component of decline. CONCLUSION: Higher educational attainment is associated with a slightly accelerated rate of cognitive decline in Alzheimer disease.

Adult↗

PubMed2005

Birth outcomes and prenatal exposure to ozone, carbon monoxide, and particulate matter: results from the Children's Health Study.

Exposures to ambient air pollutants have been associated with adverse birth outcomes. We investigated the effects of air pollutants on birth weight mediated by reduced fetal growth among term infants who were born in California during 1975-1987 and who participated in the Children's Health Study. Birth certificates provided maternal reproductive history and residence location at birth. Sociodemographic factors and maternal smoking during pregnancy were collected by questionnaire. Monthly average air pollutant levels were interpolated from monitors to the ZIP code of maternal residence at childbirth. Results from linear mixed-effects regression models showed that a 12-ppb increase in 24-hr ozone averaged over the entire pregnancy was associated with 47.2 g lower birth weight [95% confidence interval (CI), 27.4-67.0 g], and this association was most robust for exposures during the second and third trimesters. A 1.4-ppm difference in first-trimester carbon monoxide exposure was associated with 21.7 g lower birth weight (95% CI, 1.1-42.3 g) and 20% increased risk of intrauterine growth retardation (95% CI, 1.0-1.4). First-trimester CO and third-trimester O3 exposures were associated with 20% increased risk of intrauterine growth retardation. A 20-microg/m3 difference in levels of particulate matter < or = 10 microm in aerodynamic diameter (PM10) during the third trimester was associated with a 21.7-g lower birth weight (95% CI, 1.1-42.2 g), but this association was reduced and not significant after adjusting for O3. In summary, O3 exposure during the second and third trimesters and CO exposure during the first trimester were associated with reduced birth weight.

Adult↗

PubMed2001

Lack of effect of a single dose of ketoconazole on the pharmacokinetics of citalopram.

STUDY OBJECTIVE: To determine whether the pharmacokinetics of the antidepressant citalopram are affected by ketoconazole, a potent inhibitor of cytochrome P450 (CYP) 3A4. DESIGN: Single-center, double-blind, randomized, three-way crossover trial. SETTING: Research facility. PARTICIPANTS: Eighteen healthy male and female volunteers. INTERVENTION: Subjects received three treatments with a 14-day washout period: single dose of ketoconazole 200 mg plus placebo, single dose of citalopram 40 mg plus placebo, and single dose of ketoconazole 200 mg plus single dose of citalopram 40 mg. MEASUREMENTS AND MAIN RESULTS: Pharmacokinetic parameters were determined after each treatment. The pharmacokinetic profile of citalopram administered alone was essentially identical to that when administered with ketoconazole. Similarly, the pharmacokinetics of the metabolite desmethylcitalopram were unaltered by ketoconazole. CONCLUSION: No changes in pharmacokinetics of citalopram were observed after coadministration of ketoconazole, suggesting that ketoconazole and other CYP3A4 inhibitors may be administered safely with citalopram. Furthermore, no adjustment of citalopram dosage should be necessary in most patients who receive the drug in combination with a CYP3A4 inhibitor.

Adult↗

PubMed2004

Does antidepressant adherence have an effect on glycemic control among diabetic antidepressant users?

OBJECTIVE: To examine the relationship between adherence to antidepressant medications and HbA1c levels among patients with diabetes in a managed care setting. METHOD: The analysis included measures of HbA1c levels before, during, and after initial antidepressant use among 568 patients with diabetes enrolled in the Harvard Pilgrim Health Care insurance plan from 1991-1995. Adherence was defined as four refills in a six-month period after the first antidepressant prescription. General linear models using SAS PROC MIXED were used to estimate the effects of covariates including antidepressant adherence on HbA1c levels over time, comparing patients who were adherent to antidepressant medications to those patients who were non-adherent to antidepressant medications. RESULTS: Adherence to antidepressant treatment was not significantly associated with HbA1c levels among diabetic patients who are antidepressant users. Younger age, use of insulin and oral medications, and female gender were all significantly associated with HbA1c levels over time. CONCLUSIONS: Although we did not observe any association between level of adherence to antidepressant therapy among diabetic patients and levels of glucose control, our results confirm previously established associations between patient characteristics and glycemic control. Further research is needed to disentangle the complex relationship among antidepressant treatment adherence and diabetes outcomes.

Adolescent↗

PubMed2006

Assessing the ability of topiramate to improve the daily activities of patients with migraine.

OBJECTIVE: To assess the impact of topiramate on the daily activities of patients with migraine. PATIENTS AND METHODS: We performed a randomized, double-blind, placebo-controlled multicenter trial Initiated on March 1, 2001, and completed on April 4, 2002. Patient-reported data from the Migraine Specific Questionnaire (MSQ) and the Medical Outcomes Study 36-item Short-Form Health Survey (SF-36) were collected at baseline and at weeks 8, 16, and 26 from an intent-to-treat population receiving either topiramate, 50, 100, or 200 mg/d, or placebo. Two activity-related MSQ domains (role restrictive [MSQ-RR] and role prevention [MSQ-RP]) and 2 activity-related SF-36 domains (role physical [SF36-RP] and vitality [SF36-VT]) were the prospectively designated secondary outcome measures. The changes in MSQ and SF-36 scores for each treatment group were calculated by measuring the area under the curve from week 8 (the beginning of the maintenance period) through week 26 of the double-blind phase, relative to the prospective baseline. A mixed-effect piecewise linear regression model was used to estimate average domain score over time. RESULTS: Patients receiving topiramate, 100 or 200 mg/d, had significantly reduced mean monthly (28-day) migraine frequency (P = .008 and P < .001, respectively) compared with placebo, but not patients receiving topiramate, 50 mg/d (P = .48). Topiramate significantly improved mean MSQ-RR domain scores (50 mg/d [P = .02], 100 mg/d [P< .001], and 200 mg/d [P < .001]) and mean MSQ-RP domain scores (50 mg/d [P = .007], 100 mg/d [P = .001], and 200 mg/d [P= .002]) vs placebo. Topiramate, 100 and 200 mg/d, significantly improved mean SF36-RP domain scores vs placebo (P = .02). Topiramate (all doses) improved SF36-VT domain scores, although not significantly vs placebo. Changes in prospectively designated domain scores were significantly correlated with changes in mean monthly migraine frequency (P < or = .001 [MSQ domains], P < or = .002 [SF-36 domains]). CONCLUSION: Patient-reported migraine-specific outcomes measured by the MSQ-RR and MSQ-RP domains improved significantly for those receiving topiramate (all doses) vs placebo. The SF36-RP domain scores improved significantly for patients receiving 100 or 200 mg/d of topiramate. Improvements in all 4 prospectively selected MSQ and SF-36 domains were significantly correlated with decreases in mean monthly migraine frequency.

Activities of Daily Living↗

PubMedDate not supplied

Changes in cognitive measures in a randomized controlled trial of a health promotion program for couples targeting diet and physical activity.

PURPOSE: To examine cognitive responses to a 4-month health promotion program targeting diet and physical activity in recently cohabiting couples. DESIGN: A three-group randomized controlled trial: no intervention (Controls), interactive group sessions and mail-outs (Interactive), and one group session followed by mail-outs (Mail). SETTING: Australian research studies unit. SUBJECTS: Seventy-eight of 137 couples cohabiting for < 2 years, recruited by advertising, completed the study. MEASURES: Stages of change; health beliefs; self-efficacy; and perceived barriers to change evaluated by questionnaires at baseline, postintervention, and 1-year follow-up. Data were analyzed using mixed models, factor analysis, and linear regression. RESULTS: In the Mail group, the perceived importance of barriers to dietary change decreased by 19% at postintervention and 16% at 1-year follow-up; dietary self-efficacy increased by 9% and 3%, respectively. In the Interactive group, the perceived importance of barriers to dietary change decreased by 26% and 20%, and dietary self-efficacy increased by 14% and 12%, respectively. Similar trends for physical activity were not statistically significant. However, changes in self-efficacy and perceived barriers to change significantly predicted dietary and physical activity behaviors. A preponderance of higher socio-economic groups and the proportion of drop-out may have biased results. CONCLUSIONS: Health promotion using partners' support and targeting diet and physical activity improves associated cognitive variables in couples. Most improvements were greater and more sustained with interactive group sessions.

Adult↗

PubMed2002

Prediction of additive genetic effects for the QTL-cluster on the basis of data on surrounding markers in outbred populations.

This paper described a method for predicting additive effects of a cluster of tightly linked QTLs for outbred populations of animals in the situation where the QTLs are located on a chromosome segment surrounded by multiple linked DNA markers. We present a mixed model method for best linear unbiased prediction (conditional to the marker data) of the additive effects of the QTL-cluster and of the remaining QTLs unlinked to the marker linkage group. This method takes into consideration the identity-by-descent proportion (IBDP) for the particular chromosomal segment, in contrast to some other methods which use IBD probabilities at one specific location. In this method, fully informative data on different flanking markers is used to calculate the values of the expectations of the IBDPs (EIBDPs) between gametes for animals to be evaluated. Then the expected values are used as the elements of the gametic relationship matrix required in the best linear unbiased prediction. Giving a small numerical example, we illustrate how the present method can be used for the prediction of the QTL-cluster effects and for genetic evaluation of animals in outbred populations. A computational strategy is discussed on the basis of the calculation of the EIBDPs and the inverted gametic relationship matrix in complex pedigrees.

Animals↗

PubMed1982

Extended Hildebrand solubility approach and the log linear solubility equation.

The log linear solubility equation, log S = log Sw + sigma f, was studied in relationship to the extended Hildebrand solubility approach. It is shown that the log linear form may be derived beginning with the extended Hildebrand approach. The log linear expression gives a good linear fit for semipolar drugs in a number of water-cosolvent mixtures. It is particularly successful when the solubility parameter, delta 1, of the cosolvent is 3 or more solubility parameter units larger than the solubility parameter, delta 2, of the drug. When the cosolvent tends to solvate the drug strongly, the log linear function may even hold where the solubility parameters of the drug and cosolvent are similar. It appears, however, not to be applicable to nonpolar cosolvent systems. An interfacial model for the solubility of drugs in polar mixed solvents is based on sigma, a parameter that also figures prominently in the log linear solubility equation. When used to describe mixed solvent systems, the interfacial model applies in the region of the solubility profile (solubility versus solvent composition) where the log linear relationships hold. The extended Hildebrand solubility approach is applicable over a wide range of cosolvent composition in mixed systems from nonpolar organic solvents to water.

Aminobenzoates↗

PubMed2005

On the surviving fraction in irradiated multicellular tumour spheroids: calculation of overall radiosensitivity parameters, influence of hypoxia and volume effects.

We model the heterogeneous response to radiation of multicellular tumour spheroids assuming position- and volume-dependent radiosensitivity. We propose a method to calculate the overall radiosensitivity parameters to obtain the surviving fraction of tumours. A mathematical model of a spherical tumour with a hypoxic core and a viable rim which is a caricature of a real tumour is constructed. The model is embedded in a two-compartment linear-quadratic (LQ) model, assuming a mixed bivariated Gaussian distribution to attain the radiosensitivity parameters. Ergodicity, i.e., the equivalence between ensemble and volumetric averages is used to obtain the overall radiosensitivities for the two compartments. We obtain expressions for the overall radiosensitivity parameters resulting from the use of both a linear and a nonlinear dependence of the local radiosensitivity with position. The model's results are compared with experimental data of surviving fraction (SF) for multicellular spheroids of different sizes. We make one fit using only the smallest spheroid data and we are able to predict the SF for the larger spheroids. These predictions are acceptable particularly using bounded sensitivities. We conclude with the importance of taking into account the contribution of clonogenic hypoxic cells to radiosensitivity and with the convenience of using bounded local sensitivities to predict overall radiosensitivity parameters.

Animals↗

PubMed2000

On extensions of approaches for heat recovery calculations in integrated chemical process systems

This paper deals with heat recovery calculations in integrated chemical process systems. A level of heat recovery in a heat exchanger network (HEN) determines the cost of heat energy consumed in a total system and influences environmental pollution. Even though well established approaches to the problem of calculating minimum cost of utilities in a HEN are available at present, they do not fully account for complex industrial scenarios. The aim of methods developed in this work is to extend applications for the following cases: 1. multiple utilities of great temperature spans (so called non-point utilities), and 2. disturbances of flow-rates of heat exchanging streams. To solve the problems, methods have been developed based on optimisation models. The models are linear though one of them involves mixed-integer variables. Linearity ensures the possibility of solving the models to global optimum, even for large-scale industrial problems. The description of solution methods and examples of applications are presented in this paper.

Journal Article↗

PubMed2006

Neural networks for longitudinal studies in Alzheimer's disease.

OBJECTIVE: Alzheimer's disease affects a growing population of elderly people today. The predictions about the course of the disease is a key component of health care decision making for patients with Alzheimer's. The physician's prognosis and predicted trajectory of cognitive decline often form the basis of treatment and health care decisions taken by patients and their families. These predictions are difficult to make because of the high variability and non-linearity exhibited by individual patterns of cognitive decline. This paper presents a new method of predicting the course of a disease using longitudinal data collected through multiple clinic visits. Longitudinal databases are similar to temporal databases, with some important differences--data is collected at irregular time intervals that are patient specific and also a varying number of observations are made for each patient, depending upon the number of times the patient visited the clinic. We propose a new type of neural network called the mixed effects neural network (MENN) model that can incorporate this type of longitudinal information. MATERIAL AND METHODS: We have used longitudinal data on 704 subjects enrolled at the Layton aging and research center (LAARC) at Oregon Health and Science University. A back-propagation algorithm, modified for longitudinal data is used to obtain the weight parameters of the MENN. The modified back-propagation algorithm is further embedded in an iterative procedure that estimates the noise variance and the parameters that capture the longitudinal (temporal) correlation structure. RESULTS: We have compared the performance of the MENN with linear mixed effects models and standard neural networks (NN). MENN show better performance (misclassification rate = 0.13 and relative MSE = 0.35) as compared to standard NN (misclassification rate = 0.34 and relative MSE = 2.74) and linear mixed effects models (misclassification rate = 0.14 and relative MSE = 0.4). CONCLUSION: The results show that this method can be a useful tool for predicting non-linear disease trajectories and uncovering significant prognostic factors in longitudinal databases.

Aged↗

PubMed2006

Assessment of bone mineral density of the femoral head in dogs with early osteoarthritis.

OBJECTIVE: To compare the bone mineral density (BMD) of the proximal portion of the femur in dogs with and without early osteoarthritis secondary to hip dysplasia. ANIMALS: 24 dogs (3 Greyhounds, 6 Labrador-Greyhound crossbreeds, and 15 Labrador Retrievers). PROCEDURE: Computed tomography (CT) of the pelvis, including a bone-density phantom, was performed for each dog. Centrally located transverse CT slices and a computer workstation were used to identify 16 regions of interest (ROIs) in the proximal portion of the femur. For each ROI, the mean Hounsfield unit value was recorded; by use of the bone-density phantom and linear regression analysis, those values were converted to equivalent BMD (eBMD). Mean eBMD values for the subchondral and nonsubchondral ROIs in dogs with and without osteoarthritis (determined at necropsy) were compared. A mixed-model ANOVA and post hoc linear contrasts were used to evaluate the effects of osteoarthritis, breed, and sex on the BMD value. RESULTS: At necropsy, osteoarthritis was detected in 14 hip joints in 9 dogs; all lesions included early cartilage fibrillation. After adjusting for breed and sex, eBMD in subchondral ROIs 8 and 12 (adjacent to the fovea) were 8% and 6% higher, respectively, in osteoarthritis-affected dogs, compared with unaffected dogs; in the nonsubchondral ROIs, eBMD was 10% higher in osteoarthritis-affected dogs. CONCLUSIONS AND CLINICAL RELEVANCE: Compared with findings in unaffected dogs, increased eBMD in hip joints of dogs with early osteoarthritis supports a strong relationship between the subchondral and epiphyseal regions and articular cartilage in the pathogenesis and progression of osteoarthritis.

Animals↗

PubMed2001

Determinants of chlorpyrifos exposures and urinary 3,5,6-trichloro-2-pyridinol levels among termiticide applicators.

The exposures and work activities of 41 applicators in North Carolina using chlorpyrifos-containing termiticides were characterized. Personal air and urine samples were collected on multiple days within one week. Detailed information about chemical use, tasks, personal protective equipment and hygiene was recorded. During the 202 applicator-days monitored, 415 treatment jobs were performed. Full-shift chlorpyrifos exposures ranged from <0.048 to 110 microg/m(3) (N=184), with a geometric mean (GM) of 10 microg/m(3). Urinary 3,5,6-trichloro-2-pyridinol (TCP) levels ranged from 9.42 to 1960 microg/g creatinine (N=271) and varied significantly by day of the week (GM range: 169-262 microg/g creatinine). Predictive models for chlorpyrifos air exposure and urinary TCP levels were developed using mixed-effects stepwise linear regression. Determinants of airborne chlorpyrifos exposure included minutes chlorpyrifos applied and enclosed crawl space treated (yes/no). Determinants of TCP levels (depending on the model) included day-of-the-week, the chlorpyrifos air concentration one and two days before urine collection, minutes of chlorpyrifos applied one and two days before urine collection, enclosed crawl space treated (yes/no), and commercial structure treated (time-weighted). Within- and between-worker variability was similar for airborne chlorpyrifos; however, for TCP, between-worker variability exceeded within-worker variability by six times. The elimination half-life of TCP (26.9 h) and possibly the short sampling interval (one week) may explain the low TCP within-worker variability. Applicators' weekly mean ln(TCP levels) and weekly mean ln(chlorpyrifos air concentrations) were highly and positively linearly correlated (r(2)=0.73, P<0.0001). In summary, mixed-effects models were successfully constructed to predict airborne chlorpyrifos exposure and urinary TCP levels.

Adolescent↗

PubMed2004

Statistical correction for non-parallelism in a urinary enzyme immunoassay.

Our aim was to develop a statistical method to correct for non-parallelism in an estrone-3-glucuronide (E1G) enzyme immunoassay (EIA). Non-parallelism of serially diluted urine specimens with a calibration curve was demonstrated in an EIA for E1G. A linear mixed-effects analysis of 40 urine specimens was used to model the relationship of E1G concentration with urine volume and derive a statistical correction. The model was validated on an independent sample and applied to 30 menstrual cycles from American women. Specificity, detection limit, parallelism, recovery, correlation with serum estradiol, and imprecision of the assay were determined. Intra-and inter-assay CVs were less than 14% for high- and low-urine controls. Urinary E1G across the menstrual cycle was highly correlated with serum estradiol (r= 0.94). Non-parallelism produced decreasing E1G concentration with increase in urine volume (slope = -0.210, p < 0.0001). At 50% inhibition, the assay had 100% cross-reactivity with E1G and 83% with 17beta-estradiol 3-glucuronide. The dose-response curve of the latter did not parallel that of E1G and is a possible cause of the non-parallelism. The statistical correction adjusting E1G concentration to a standardized urine volume produced parallelism in 24 independent specimens (slope = -0.043+/-0.010), and improved the average CV of E1G concentration across dilutions from 19.5%+/-5.6% before correction to 10.3%+/-5.3% after correction. A statistical method based on linear mixed effects modeling is an expedient approach for correction of non-parallelism, particularly for hormone data that will be analyzed in aggregate.

Adult↗

PubMedDate not supplied

Regression models for calculating gas fluxes measured with a closed chamber.

Portable closed chambers provide a valuable tool for measuring crop photosynthesis and evapotranspiration. Typically, the rates of change of CO2 and water vapor concentration are assumed to be constant in the short time required to make the closed-chamber measurement, and a linear regression model is used to estimate the CO2 and H2O fluxes. However, due to the physical and physiological effects the measurement system has on the measured process, assuming a constant rate and using a linear model may underestimate the flux. Our objective was to provide a model that estimates the CO2 and H20 exchange rates at the time of chamber closure. We compared the linear regression model with a quadratic regression model using field measurements from two studies. Generally, 60 to 100% of all chamber measurement data sets were significantly nonlinear, causing the quadratic model to yield fluxes 10 to 40% greater than those calculated with the linear regression model. The frequency and degree of nonlinearity were related to the measured rate and chamber volume. Closed-chamber data should be tested for nonlinearity and an appropriate model used to calculate flux. The quadratic model provides users of well-mixed closed chambers an alternative to a simple linear model for data sets with significant nonlinearity.

Carbon Dioxide↗

PubMed2004

Generalized additive models for cancer mapping with incomplete covariates.

Maps depicting cancer incidence rates have become useful tools in public health research, giving valuable information about the spatial variation in rates of disease. Typically, these maps are generated using count data aggregated over areas such as counties or census blocks. However, with the proliferation of geographic information systems and related databases, it is becoming easier to obtain exact spatial locations for the cancer cases and suitable control subjects. The use of such point data allows us to adjust for individual-level covariates, such as age and smoking status, when estimating the spatial variation in disease risk. Unfortunately, such covariate information is often subject to missingness. We propose a method for mapping cancer risk when covariates are not completely observed. We model these data using a logistic generalized additive model. Estimates of the linear and non-linear effects are obtained using a mixed effects model representation. We develop an EM algorithm to account for missing data and the random effects. Since the expectation step involves an intractable integral, we estimate the E-step with a Laplace approximation. This framework provides a general method for handling missing covariate values when fitting generalized additive models. We illustrate our method through an analysis of cancer incidence data from Cape Cod, Massachusetts. These analyses demonstrate that standard complete-case methods can yield biased estimates of the spatial variation of cancer risk.

Algorithms↗