Dataset Information

Regional differences in prediction models of lung function in Germany.

ABSTRACT:

Background

Little is known about the influencing potential of specific characteristics on lung function in different populations. The aim of this analysis was to determine whether lung function determinants differ between subpopulations within Germany and whether prediction equations developed for one subpopulation are also adequate for another subpopulation.

Methods

Within three studies (KORA C, SHIP-I, ECRHS-I) in different areas of Germany 4059 adults performed lung function tests. The available data consisted of forced expiratory volume in one second, forced vital capacity and peak expiratory flow rate. For each study multivariate regression models were developed to predict lung function and Bland-Altman plots were established to evaluate the agreement between predicted and measured values.

Results

The final regression equations for FEV1 and FVC showed adjusted r-square values between 0.65 and 0.75, and for PEF they were between 0.46 and 0.61. In all studies gender, age, height and pack-years were significant determinants, each with a similar effect size. Regarding other predictors there were some, although not statistically significant, differences between the studies. Bland-Altman plots indicated that the regression models for each individual study adequately predict medium (i.e. normal) but not extremely high or low lung function values in the whole study population.

Conclusions

Simple models with gender, age and height explain a substantial part of lung function variance whereas further determinants add less than 5% to the total explained r-squared, at least for FEV1 and FVC. Thus, for different adult subpopulations of Germany one simple model for each lung function measures is still sufficient.

SUBMITTER: Schnabel E

PROVIDER: S-EPMC2873930 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:ObjectivesThe aim of our study was to explore patient types in general practitioner (GP) practices and to quantify the regional differences of the frequencies of these patient types in northern Germany.Design and settingWe conducted a mixed-methods study based on focus groups and standardised interviews with GPs. All counties and independent cities within a radius of 120 km around Hamburg were assigned one of three regional categories (urban areas, environs, rural areas). The focus groups were analysed using qualitative content analysis. Relative frequencies of consultations by patient types and differences between the regions were calculated. Logistic regression analyses were used to identify differences among regions.ParticipantsNine focus groups with 65 GPs (67.7% male). From the 280 initially recruited GPs 211 (65.4% male) could be personally interviewed.ResultsFour themes with 27 patient types were derived from the focus groups: patients classified by morbidity, sociodemographic characteristics, special care needs and patient behaviour. Five patient characteristics were significantly more prevalent in urban areas than rural areas: patients with migration background and culturally different disease concepts (OR 1.23; 95% CI 1.06 to 1.42), privately insured patients (OR 1.17; 95% CI 1.05 to 1.31), educationally disadvantaged patients with low health literacy (OR 1.11; 95% CI 1.04 to 1.19), patients with psychiatric disorders (OR 1.07; 95% CI 1.02 to 1.12) and senior citizens living on their own without caregivers (OR 1.05; 95% CI 1.05 to 1.31). Three patient types were significantly less prevalent in urban areas: minors accompanied by their parents (OR 0.71; 95% CI 0.61 to 0.83), patients with poor therapy adherence (OR 0.87, 95% CI 0.80 to 0.95) and patients with dementia (OR 0.90; 95% CI 0.82 to 0.99).ConclusionsGPs could compensate the specific needs of their patients with medical training aligned with the requirements of their region. Urban GPs need skills treating patients with psychiatric, social and cultural problems, rural GPs regarding the care for children or noncompliant patients.Trial registration numberNCT02558322.

Project description:IntroductionIdentification of individuals at high risk for lung cancer should be of value to individuals, patients, clinicians, and researchers. Existing prediction models have only modest capabilities to classify persons at risk accurately.MethodsProspective data from 70 962 control subjects in the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial (PLCO) were used in models for the general population (model 1) and for a subcohort of ever-smokers (N = 38 254) (model 2). Both models included age, socioeconomic status (education), body mass index, family history of lung cancer, chronic obstructive pulmonary disease, recent chest x-ray, smoking status (never, former, or current), pack-years smoked, and smoking duration. Model 2 also included smoking quit-time (time in years since ever-smokers permanently quit smoking). External validation was performed with 44 223 PLCO intervention arm participants who completed a supplemental questionnaire and were subsequently followed. Known available risk factors were included in logistic regression models. Bootstrap optimism-corrected estimates of predictive performance were calculated (internal validation). Nonlinear relationships for age, pack-years smoked, smoking duration, and quit-time were modeled using restricted cubic splines. All reported P values are two-sided.ResultsDuring follow-up (median 9.2 years) of the control arm subjects, 1040 lung cancers occurred. During follow-up of the external validation sample (median 3.0 years), 213 lung cancers occurred. For models 1 and 2, bootstrap optimism-corrected receiver operator characteristic area under the curves were 0.857 and 0.805, and calibration slopes (model-predicted probabilities vs observed probabilities) were 0.987 and 0.979, respectively. In the external validation sample, models 1 and 2 had area under the curves of 0.841 and 0.784, respectively. These models had high discrimination in women, men, whites, and nonwhites.ConclusionThe PLCO lung cancer risk models demonstrate high discrimination and calibration.