SEARCH · PubMed Health

Results for “predictive modelling”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

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

At least 1,621 records · Page 90Linked to original sources

PubMed2004

A food chain model to predict the levels of lipophilic organic contaminants in humans.

A fugacity-based, nonsteady state, mechanistic model called ACC-HUMAN was developed to describe bioaccumulation of lipophilic organic pollutants from air, water, and soil to humans. The physical environment was linked via a marine and an agricultural food chain model to a human bioaccumulation model. Contaminant uptake via the primary dietary sources of persistent lipophilic contaminants in industrialized countries was addressed, namely fish, dairy products, and beef. In addition, uptake from air and water was considered, allowing the model also to treat less lipophilic compounds. To evaluate the model, the food chain characteristics were parameterized for southern Sweden and historical scenarios of polychlorinated biphenyl (PCB) concentrations in air, water, and soil in this region were constructed from published data. The resulting model predictions of PCB concentrations in fish, milk, beef, and human tissue agreed well with measured concentrations from Swedish monitoring programs. This suggests that ACC-HUMAN is a useful tool for predicting human exposure to bioaccumulative organic compounds. It can be linked easily to existing multimedia fate and transport models.

Agriculture↗

PubMed1991

Development of a model to predict forage intake by grazing cattle.

A mathematical model to predict daily DMI and account for effects of energy supplementation on forage intake has been developed in several stages. A previously evaluated dynamic rumen model for sheep was adopted as the fermentation component of the intake model. Intake was adjusted to reach a given level of DM fill, which is the sum of the concentrations of each of the dietary fractions within the rumen. Differential equations described the rate of change of each nutrient fraction. Genetic size scaling rules based on mature body size relationships were used to adjust rate and fill parameters of the intake model from sheep to beef cattle. Nutrient fractions were partitioned into those that flow at the particulate passage rate vs the fluid passage rate. Forty-two data points representing perennial ryegrass, wheat pasture and range grasses were used to parameterize and evaluate the model. The model was relatively sensitive to the coefficient relating DMI to particulate rate of passage, the rate constant for the use of the potentially degraded fiber fraction of the forage, and to the composition constants for the amount of carbohydrate and nitrogen in the microbial mass. Relative insensitivity was observed for starch and protein nutrient use rate constants, for the coefficient relating DMI to fluid passage rate, and for constants relating to the growth of the microbial mass in the rumen. Feed intake of grazing cattle may be predicted by mechanistic models describing various nutrients' contribution to ruminal fill.

Animal Feed↗

PubMed2002

Assessment of mathematical models for predicting Staphylococcus aureus growth in cooked meat products.

The growth of Staphylococcus aureus in commercially available vacuum-packaged cooked ham, turkey breast meat, and chicken breast meat stored at 2.3, 6.5, 10, 13.5, and 17.7 degrees C was studied. Growth rates observed in these food products were compared with those predicted on the basis of various growth models found in the literature and with those generated by the Pathogen Modeling Program and the Food MicroModel software using graphical and mathematical analysis for performance evaluation. In general, the models studied overestimated the growth of S. aureus. The Dengremont and Membré model most closely matched the observed behavior of S. aureus in ham and chicken breast meat, with bias factors of 1.56 and 1.09, respectively. The Eifert et al. model accurately described the growth of S. aureus in turkey breast meat, with a bias factor of 1.51. The remaining models provided safe predictions of the growth rate of S. aureus, but with poor accuracy. Predictive microbiology models have an immediate practical application in improving microbial food safety and quality and are very useful decision support tools, but they should not be used as the sole determinant of product safety.

Animals↗

PubMed2006

Combining population-dynamic and ecophysiological models to predict climate-induced insect range shifts.

Hundreds of species are shifting their ranges in response to recent climate warming. To predict how continued climate warming will affect the potential, or “bioclimatic range,” of a skipper butterfly, we present a population‐dynamic model of range shift in which population growth is a function of temperature. We estimate the parameters of this model using previously published data for Atalopedes campestris. Summer and winter temperatures affect population growth rate independently in this species and therefore interact as potential range‐limiting factors. Our model predicts a two‐phase response to climate change; one range‐limiting factor gradually becomes dominant, even if warming occurs steadily along a thermally linear landscape. Whether the range shift accelerates or decelerates and whether the number of generations per year at the range edge increases or decreases depend on whether summer or winter warms faster. To estimate the uncertainty in our predictions of range shift, we use a parametric bootstrap of biological parameter values. Our results show that even modest amounts of data yield predictions with reasonably small confidence intervals, indicating that ecophysiological models can be useful in predicting range changes. Nevertheless, the confidence intervals are sensitive to regional differences in the underlying thermal landscape and the warming scenario.

Animals↗

PubMed2004

A mechanistic model for predicting the nutrient requirements and feed biological values for sheep.

The Cornell Net Carbohydrate and Protein System (CNCPS), a mechanistic model that predicts nutrient requirements and biological values of feeds for cattle, was modified for use with sheep. Published equations were added for predicting the energy and protein requirements of sheep, with a special emphasis on dairy sheep, whose specific needs are not considered by most sheep-feeding systems. The CNCPS for cattle equations that are used to predict the supply of nutrients from each feed were modified to include new solid and liquid ruminal passage rates for sheep, and revised equations were inserted to predict metabolic fecal N. Equations were added to predict fluxes in body energy and protein reserves from BW and condition score. When evaluated with data from seven published studies (19 treatments), for which the CNCPS for sheep predicted positive ruminal N balance, the CNCPS for sheep predicted OM digestibility, which is used to predict feed ME values, with no mean bias (1.1 g/100 g of OM; P > 0.10) and a low root mean squared prediction error (RMSPE; 3.6 g/100 g of OM). Crude protein digestibility, which is used to predict N excretion, was evaluated with eight published studies (23 treatments). The model predicted CP digestibility with no mean bias (-1.9 g/100 g of CP; P > 0.10) but with a large RMSPE (7.2 g/100 g of CP). Evaluation with a data set of published studies in which the CNCPS for sheep predicted negative ruminal N balance indicated that the model tended to underpredict OM digestibility (mean bias of -3.3 g/100 g of OM, P > 0.10; RMSPE = 6.5 g/100 g of OM; n = 12) and to overpredict CP digestibility (mean bias of 2.7 g/100 g of CP, P > 0.10; RMSPE = 12.8 g/100 g of CP; n = 7). The ability of the CNCPS for sheep to predict gains and losses in shrunk BW was evaluated using data from six studies with adult sheep (13 treatments with lactating ewes and 16 with dry ewes). It accurately predicted variations in shrunk BW when diets had positive N balance (mean bias of 5.8 g/d; P > 0.10; RMSPE of 30.0 g/d; n = 15), whereas it markedly overpredicted the variations in shrunk BW when ruminal balance was negative (mean bias of 53.4 g/d, P < 0.05; RMSPE = 84.1 g/d; n = 14). These evaluations indicated that the Cornell Net Carbohydrate and Protein System for Sheep can be used to predict energy and protein requirements, feed biological values, and BW gains and losses in adult sheep.

Animal Feed↗

PubMed1997

A mathematical model that predicts skeletal muscle force.

This study demonstrates the validity of a mathematical model that predicts the force generated by rat skeletal muscles during brief subtetanic and tetanic isometric contractions. The model consists of three coupled differential equations (ODE's). The first two equations represent the calcium dynamics and the third equation represents force dynamics. The model parameters were identified from brief trains of regularly spaces pulses [constant-frequency trains (CFT's)] that produce subtetanic muscle responses. Using these parameters, the model was able to predict isometric forces from other stimulation patterns. For the gastrocnemius muscles predictions were made for responses to CFT's with interpulse intervals (IPI's) ranging from 10 to 50 ms and variable-frequency trains (VFT's), where the initial IPI = 10 ms and the remaining IPI's were identical to those used for the CFT's. For the soleus muscles predictions were made for 10-100-ms CFT's. The shape of the predicted responses closely match the experimental data. Comparisons between experimental and modeled force-time integrals, peak forces, and time-to-peak also suggest excellent agreement between the model and the experiment data. Many physiological parameters predicted by the model agree with values obtained independently by others. In conclusion, the model accurately predicts isometric forces generated by rat gastrocnemius and soleus muscles produced by brief stimulation trains.

Actins↗

PubMed1999

A physiological model for predicting carboxyhemoglobin formation from exposure to carbon monoxide in rats.

A time-dependent simulation model, based on the Coburn-Forster-Kane equation, was written in Advanced Continuous Simulation Language to predict carboxyhemoglobin (HbCO) formation and dissociation in F-344 rats during and after exposure to 500 parts/million CO for 1 h. Blood-gas analysis and CO-oximetry were performed on samples collected during exposure and off-gassing of CO. Volume displacement plethysmography was used to measure minute ventilation (VE) during exposure. CO diffusing capacity in the lung (DLCO) was also measured. Other model parameters measured in the animals included blood pH, total blood volume, and Hb concentration. Comparisons between model predictions using values for VE, DLCO, and the Haldane coefficient cited in the literature and predictions using measured VE, DLCO, and calculated Haldane coefficient for individual animals were made. General model predictions using values for model parameters derived from the literature agreed with published HbCO values by a factor of 0.987 but failed to simulate experimental data. On average, the general model overpredicted measured HbCO level by nearly 9%. A specific model using the means of measured variables predicted HbCO concentration within a factor of 0.993. When experimentally observed parameter fluctuations were included, the specific model predictions reflected experimental effects on HbCO formation.

Algorithms↗

PubMed2004

Artificial neural network modeling to predict the plasma concentration of aminoglycosides in burn patients.

The goal was to use an artificial neural network model to predict the plasma concentration of aminoglycosides in burn patients and identify patients whose plasma antibiotic concentration would be sub-therapeutic based on the patients' physiological data and taking into account burn severity. Physiological data and some indicators of burn severity were collected from 30 burn patients who received arbekacin. A three-layer artificial neural network with five neurons in the hidden layer was used to predict the plasma concentration of arbekacin. Linear modeling for prediction of plasma concentration and logistic regression modeling for the classification of patients were also used and the predictive performance was compared to results from the artificial neural network model. Dose, body mass index, serum creatinine concentration and amount of parenteral fluid were selected as covariates for the plasma concentration of arbekacin. Area of burn after skin graft was a good covariate for indicating burn severity. Predictive performance of the artificial neural network model including burn severity was much better than linear modeling and logistic regression analysis. An artificial neural network model should be helpful for the prediction of plasma concentration using patients' physiological data, and burn severity should be included for improved prediction in burn patients. Because the relationship between burn severity and plasma concentration of aminoglycosides is thought to be nonlinear, it is not surprising that the artificial neural network model showed better predictive performance compared to the linear or logistic regression models.

Adult↗

PubMed1999

Coal tits, Parus ater, lose weight in response to chases by predators.

Theoretical models predict that birds should decrease their body mass in response to increased predation risk because lighter birds take off faster and are more manoeuvrable. We studied the effect of predation risk by chasing coal tits in large outdoor aviaries thus simulating an attempt to capture them. With this increase in predation risk, both perceived and actual, coal tits lost significantly more weight than in a control situation when they were not pursued. This pattern was attributable to a smaller gain in weight only during the day; nocturnal weight did not change in relation to diurnal predation risk. The lower daily weight gain was not consistent with predictions from models of interrupted foraging, but was consistent with predictions from risk adjustment models. Moreover, there was no difference in weight gain over 2-h periods that included a 1-h fast and those in which feeding was ad libitum, suggesting that coal tits could easily regain their body mass after a predator had interrupted their feeding. Our results therefore suggest that pursuit by predators leads to a decrease in the body mass of small birds. Copyright 1999 The Association for the Study of Animal Behaviour.

Journal Article↗

PubMed2004

The utility of structure-activity relationship (SAR) models for prediction and covariate selection in developmental toxicity: comparative analysis of logistic regression and decision tree models.

Structure-activity relationship (SAR) models can be used to predict the biological activity of potential developmental toxicants whose adverse effects include death, structural abnormalities, altered growth and functional deficiencies in the developing organism. Physico-chemical descriptors of spatial, electronic and lipophilic properties were used to derive SAR models by two modeling approaches, logistic regression and Classification and Regression Tree (CART), using a new developmental database of 293 chemicals (FDA/TERIS). Both single models and ensembles of models (termed bagging) were derived to predict toxicity. Assessment of the empirical distributions of the prediction measures was performed by repeated random partitioning of the data set. Results showed that both the decision tree and logistic regression derived developmental SAR models exhibited modest prediction accuracy. Bagging tended to enhance the prediction accuracy and reduced the variability of prediction measures compared to the single model for CART-based models but not consistently for logistic-based models. Prediction accuracy of single logistic-based models was higher than single CART-based models but bagged CART-based models were more predictive. Descriptor selection in SAR for the understanding of the developmental mechanism was highly dependent on the modeling approach. Although prediction accuracy was similar in the two modeling approaches, there was inconsistency in the model descriptors.

Animals↗

PubMed2025

Genomic and Developmental Models to Predict Cognitive and Adaptive Outcomes in Autistic Children.

IMPORTANCE: Although early signs of autism are often observed between 18 and 36 months of age, there is considerable uncertainty regarding future development. Clinicians lack predictive tools to identify those who will later be diagnosed with co-occurring intellectual disability (ID). OBJECTIVE: To predict ID in children diagnosed with autism. DESIGN, SETTING, AND PARTICIPANTS: This prognostic study involved the development and validation of models integrating genetic variants and developmental milestones to predict ID. Models were trained, cross-validated, and tested for generalizability across 3 autism cohorts: Simons Foundation Powering Autism Research (SPARK), Simons Simplex Collection, and MSSNG. Autistic participants were assessed older than 6 years of age for ID. Study data were analyzed from January 2023 to July 2024. EXPOSURES: Ages at attaining early developmental milestones, occurrence of language regression, polygenic scores for cognitive ability and autism, rare copy number variants, de novo loss-of-function and missense variants impacting constrained genes. MAIN OUTCOMES AND MEASURES: The out-of-sample performance of predictive models was assessed using the area under the receiver operating characteristic curve (AUROC), positive predictive values (PPVs), and negative predictive values (NPVs). RESULTS: A total of 5633 autistic participants (4574 male [81.2%]) were included in this analysis. On average, participants were diagnosed with autism at 4 (IQR, 3-7) years of age and assessed for ID at 11 (8-14) years of age, with 1159 participants (20.6%) being diagnosed with ID. The model integrating all predictors yielded an AUROC of 0.653 (95% CI, 0.625-0.681), and this predictive performance was cross-validated and generalized across cohorts. This modest performance reflected that only a subset of individuals carried large-effect variants, high polygenic scores, or presented delayed milestones. However, combinations of genetic variants that are typically not considered clinically relevant by diagnostic laboratories achieved PPVs of 55% and correctly identified 10% of individuals developing ID. The addition of polygenic scores to developmental milestones specifically improved NPVs rather than PPVs. Notably, the ability to stratify ID probabilities using genetic variants was up to 2-fold higher in individuals with delayed milestones compared with those with typical development. CONCLUSIONS AND RELEVANCE: Results of this prognostic study suggest that the growing number of neurodevelopmental condition-associated variants cannot, in most cases, be used alone for predicting ID. However, models combining different classes of variants with developmental milestones provide clinically relevant individual-level predictions that could be useful for targeting early interventions.

Humans↗

PubMed1994

Prediction of axillary lymph node status in breast cancer patients by use of prognostic indicators.

BACKGROUND: If axillary lymph node status of breast cancer patients could be accurately predicted from basic clinical information and from characteristics of their primary tumors, then many patients could be spared axillary lymph node dissection. Tumor size alone does not allow the identification of groups with very low or high risk of being axillary node positive. PURPOSE: Our goal was to investigate the possibility of using prognostic indicators to predict axillary node status of patients with primary breast cancer. METHODS: Data from 26,683 patients from the National Breast Cancer Tissue Resource were used in this study. Patients in this dataset were randomly assigned to a training set (patient information used to construct predictive models) or a validation set (patient information used to prospectively evaluate predictive models). The records of a total of 11,964 case patients that had complete prognostic factors and pathologic data were analyzed: 5963 patients in the training set and 6001 patients in the validation set. All of the patients studied had tumors 5 cm or less in size and at least 15 axillary lymph nodes that had been examined. Data used for construction of the predictive models were available for all patients and included tumor size, number of nodes positive, patient age, quantitative estrogen receptor levels, quantitative progesterone receptor (PgR) levels, DNA flow cytometry-derived ploidy, and S-phase fraction. Logistic regression models were used to predict nodal status. RESULTS: Multivariate predictive models were produced that used tumor size, patient age, S phase, and PgR as independent predictors. These models allowed identification of patient risks of being node positive ranging from 6%-79% and as having 10 or more positive nodes ranging from less than 1% to slightly more than 30%. CONCLUSION: Addition of prognostic indicator information to tumor size can refine estimates of whether a patient is likely to be node positive. However, no patient subsets could be identified as having greater than 95% chance of being node negative or node positive. IMPLICATIONS: These predictive models cannot alleviate the necessity of axillary node dissection for staging of breast cancer patients in situations in which nodal status would affect therapeutic decisions. Subsets of patients could be identified who had a less than 5% chance of having 10 or more positive nodes. Thus, some patients could be spared axillary dissection if it was being performed solely to identify patients with this high-risk feature.

Axilla↗

PubMed2002

Role of hydrophilic and hydrophobic contacts in folding of the second beta-hairpin fragment of protein G: molecular dynamics simulation studies of an all-atom model.

Predicting the folding mechanism of the second beta-hairpin fragment of the Ig-binding domain B of streptococcal protein G is unexpectedly challenging for simplified reduced models because the models developed so far indicated a different folding mechanism from what was suggested from high-temperature unfolding and equilibrium free-energy surface analysis based on established all-atom empirical force fields in explicit or implicit solvent. This happened despite the use of empirical residue-based interactions, multibody hydrophobic interactions, and inclusions of hydrogen bonding effects in the simplified models. This article employs a recently developed all-atom (except nonpolar hydrogens) model interacting with simple square-well potentials to fold the peptide fragment by molecular dynamics simulation methods. In this study, 193 out of 200 trajectories are folded at two reduced temperatures (3.5 and 3.7) close to the transition temperature T* approximately 4.0. Each simulation takes <7 h of CPU time on a Pentium 800-MHz PC. Folding of the new all-atom model is found to be initiated by collapse before the formation of main-chain hydrogen bonds. This verifies the mechanism proposed from previous all-atom unfolding and equilibrium simulations. The new model further predicts that the collapse is initiated by two nucleation contacts (a hydrophilic contact between D46 and T49 and a hydrophobic contact between Y45 and F52), in agreement with recent NMR measurements. The results suggest that atomic packing and native contact interactions play a dominant role in folding mechanism.

Amino Acids↗

PubMed2007

Survival and apoptotic pathways initiated by TNF-alpha: modeling and predictions.

We present a mathematical model which includes TNF-alpha initiated survival and apoptotic cascades, as well as nuclear transcription of IkappaB. These pathways play a crucial role in deciding cell fate in response to inflammation and infection. Our model incorporates known specific protein-protein interactions as identified by experiments. Using these biochemical interactions, we develop a mathematical model of the NF-kappaB-mediated survival and caspase-mediated apoptosis pathways. Using mass action kinetics, we follow the formation of the survival and late complexes as well as the dynamics of DNA fragmentation. The effect of TNF-alpha concentration on DNA fragmentation is modeled and compares well with experiment. Nuclear transcription is also modeled phenomenologically by means of time lagged cytosolic concentrations. This results in transcription related concentrations undergoing under-damped oscillations, in qualitative and quantitative agreement with experiment. Using a tumor cell as a hypothetical model, we explore the interplay between the components of the survival and apoptotic pathways. Results are presented which make predictions on the limits of cellular oscillations in terms of time delay, initial concentration ratios and other features of the model. The model also makes clear predictions on cell viability in terms of DNA damage within the framework of TNF-alpha stimulus duration.

Apoptosis↗

PubMed2005

The implantation of every embryo facilitates the chances of the remaining embryos to implant in an IVF programme: a mathematical model to predict pregnancy and multiple pregnancy rates.

BACKGROUND: We aimed to assess the validity of a theoretical mathematical model to predict the pregnancy rate and the multiple pregnancy rate in IVF/oocyte donation programmes on the basis of the implantation rate and the number of transferred embryos. METHODS: A total of 1835 embryo transfers corresponding to three different programmes in two centres with different implantation rates were analysed. Pregnancy and multiple pregnancy rates observed in the aforementioned programmes were compared with those obtained following different mathematical models. Four models were tested: binomial model, ground model, maternal variability model and collaborative model. The goodness of fit was performed by means of the maximum likelihood fit method. RESULTS: The binomial model could not predict the pregnancy rate, and especially the multiple pregnancy rate. The multiple pregnancy rate predicted following the binomial model was much lower than observed, up to 40-fold reduced. Ground model and maternal variability model adjusted to the data with more precision, but were still not accurate. Finally, the collaborative model reproduced with very great accuracy both pregnancy rate and the multiple pregnancy rate. A collaborative parameter of 22% was found, implying that the implantation probability of each embryo is increased by 22% for every embryo previously implanted. CONCLUSIONS: Embryonic implantation does not follow a binomial law, showing that the implantation is not independent from the number of embryos implanted. The best fit to the data is obtained following a collaborative model by which the implantation of one embryo is facilitated by the implantation of other embryo(s). The mathematical formula of the collaborative model predicts very accurately the pregnancy rate and the multiple pregnancy rate in IVF/oocyte donation programmes, based on the implantation rate of this specific programme and the number of embryos transferred up to five embryos. We recommend using the aforementioned formula to quantify the pregnancy rate and the risk of multiple pregnancy in the counselling of the infertile couple at embryo transfer. Such a formula is freely available at www.ifca.unican.es/matorras/mathpreg/.

Embryo Implantation↗

PubMed2001

Calibration of a nutrient flow model of energy utilization by growing pigs.

A computational framework to represent energy utilization for body protein and lipid accretion by growing pigs is presented. Nutrient and metabolite flows, and the biochemical and biological processes which transform these, are explicitly represented in this nutritional process model. A calibration procedure to adjust the marginal input-output response is described, and applied, using reported experimental results, to determine a complete set of parameters for representing energy utilization by growing pigs. A reasonable value for minimum basal energy requirements is also determined. Although model inputs and outputs need not at any time be converted to equivalent energy flows, to facilitate comparison of model response with that of conventional energy-based models, a simple means to estimate energy flows from model-predicted nutrient flows is described. The well-known hierarchy of marginal (biological) energetic efficiencies with which pigs use different classes of nutrients is predicted by the model, based only on simple biological and biochemical principles. The significance of independent diet and metabolic effects on both energetic efficiency and maintenance requirements is examined using model predictions from simulated experiments.

Animal Nutritional Physiological Phenomena↗

PubMed1995

Models of cooperativity in protein folding.

What is the basis for the two-state cooperativity of protein folding? Since the 1950s, three main models have been put forward. 1. In 'helix-coil' theory, cooperativity is due to local interactions among near neighbours in the sequence. Helix-coil cooperativity is probably not the principal basis for the folding of globular proteins because it is not two-state, the forces are weak, it does not account for sheet proteins, and there is no evidence that helix formation precedes the formation of a hydrophobic core in the following pathways. 2. In the 'sidechain packing' model, cooperativity is attributed to the jigsaw-puzzle-like complementary fits of sidechains. This too is probably not the basis of folding cooperativity because exact models and experiments on homopolymers with sidechains give no evidence that sidechain freezing is two-state, sidechain complementarities in proteins are only weak trends, and the molten globule model predicted by this model is far more native-like than experiments indicate. 3. In the 'hydrophobic core collapse' model, cooperativity is due to the assembly of non-polar residues into a good core. Exact model studies show that this model gives two-state behaviour for some sequences of hydrophobic and polar monomers. It is based on strong forces. There is considerable experimental evidence for the kinetics this model predicts: the development of hydrophobic clusters and cores is concurrent with secondary structure formation. It predicts compact denatured states with sizes and degrees of disorder that are in reasonable agreement with experiments.

Kinetics↗

PubMed2005

Pulmonary effects of indoor- and outdoor-generated particles in children with asthma.

Most particulate matter (PM) health effects studies use outdoor (ambient) PM as a surrogate for personal exposure. However, people spend most of their time indoors exposed to a combination of indoor-generated particles and ambient particles that have infiltrated. Thus, it is important to investigate the differential health effects of indoor- and ambient-generated particles. We combined our recently adapted recursive model and a predictive model for estimating infiltration efficiency to separate personal exposure (E) to PM2.5 (PM with aerodynamic diameter < or = 2.5 microm) into its indoor-generated (Eig) and ambient-generated (Eag) components for 19 children with asthma. We then compared Eig and Eag to changes in exhaled nitric oxide (eNO), a marker of airway inflammation. Based on the recursive model with a sample size of eight children, Eag was marginally associated with increases in eNO [5.6 ppb per 10-microg/m3 increase in PM2.5; 95% confidence interval (CI), -0.6 to 11.9; p = 0.08]. Eig was not associated with eNO (-0.19 ppb change per 10 microg/m3). Our predictive model allowed us to estimate Eag and Eig for all 19 children. For those combined estimates, only Eag was significantly associated with an increase in eNO (Eag: 5.0 ppb per 10-microg/m3 increase in PM2.5; 95% CI, 0.3 to 9.7; p = 0.04; Eig: 3.3 ppb per 10-microg/m3 increase in PM2.5; 95% CI, -1.1 to 7.7; p = 0.15). Effects were seen only in children who were not using corticosteroid therapy. We conclude that the ambient-generated component of PM2.5 exposure is consistently associated with increases in eNO and the indoor-generated component is less strongly associated with eNO.

Adolescent↗