SEARCH · PubMed Health

Results for “cross-validation”

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 343 records · Page 19Linked to original sources

PubMed1991

Predictors of long-term sexual assault trauma among a national sample of victimized college women.

Several multivariate studies have attempted to delineate the major predictors of long-term, sexual assault-induced aftereffects. The present study was an attempt to extend previous work by supplementing known preassault, assault, and postassault predictors of trauma with cognitive measures. The data were cross-sectional and included responses from 1,213 victims of sexual assault who were recruited from a national sample of higher education students. Utilizing hierarchical multiple regression analyses and a cross-validation procedure, the results suggested that the prior mental status of the victims, the forcefulness of the assaults, as well as two cognitive variables related to a victim's beliefs about sex and relationships predicted victims' scores on standardized measures of anxiety and depression. The relationships were such that the most traumatized victims were those women who had evidenced a history of mental health problems, who had experienced the more aggressive assaults, who tended to believe that people in relationships are not trustworthy, and tended to place conservative restrictions on the sexual acts and circumstances under which sex should occur. Taken together these variables accounted for between 9%-14% of the variance in sexual assault aftereffects. The discussion emphasizes the contribution of cognitive processes to the understanding of sexual assault trauma.

Adult↗

PubMed1991

Morphometry of the human thigh muscles. A comparison between anatomical sections and computer tomographic and magnetic resonance images.

The present study examined the relative accuracy and precision of MR and CT procedures for determining the CSA of individual muscles from the human thigh. Serial AN, CT and MR cross-sections were obtained from three cadaveric lower limbs. The MR measurements provided accurate and precise values for the CSAs of most thigh muscles, generally within +/- 7.5% of the AN standard. In contrast, CT tended systematically to overestimate the AN CSA by 10-20%. Retest procedures indicated that highly reliable measurements could be obtained from both MR and CT images. However, subjective interpretations of boundaries between closely apposed muscle bellies, particularly for muscles with more than one head, were necessary for resolving entities in the imaging records and this decreased the relative accuracy of MR and CT measures. Interestingly, MR records demonstrated an incomplete septum between vastus lateralis and vastus intermedius more prominently than AN cross-sections. The novel cross-validation procedures used in this study also highlighted several system-based errors in the MR records that, if undetected and left uncorrected, would have seriously biased the morphometric data obtained with this technique. In general, MR provides high resolution images of the human thigh muscles which may be used to obtain valid measures of the CSA of these structures.

Cadaver↗

PubMed1991

Midwest wrestling study: prediction of minimal weight for high school wrestlers.

This study determined the validity of previously published or newly derived equations to predict fat-free body mass (FFB) in high school wrestlers from the midwestern United States. Five laboratories participated in the data-pooling study (total sample of 860 subjects). Measures included body composition by underwater weighing and anthropometric measurements of body mass, stature, and selected circumferences, diameters, and skinfolds. Cross-validation of selected equations to predict FFB revealed the lowest levels of error from the equations of Lohman, Thorland et al., Katch and McArdle, and Behnke and Wilmore. Modification of the constants in these equations or generation of new equations did not substantially reduce prediction error. Overall, total error for these top equations ranged from 2.44 to 2.59 kg. However, based on observed trends, this error was of lower magnitude with the younger and lighter subjects and of higher magnitude with the older and heavier subjects. We conclude that these equations could be used singularly or collectively to determine FFB, and a minimal weight could then be derived and assigned to a scholastic wrestler.

Adolescent↗

PubMedDate not supplied

[Validity of 5 bioelectric impedance equations for the estimation of lean body mass in women].

Fat free mass (FFM) predicted by five bioelectrical impedance (BIA) available formulae (Segal et al., Lukaski et al., Kushner et al., Deurenberg et al., manufacturer's predictive equation) was compared with densitometrically-determined FFM (as a reference method) in two groups of young women with different anthropometric characteristics. In the farmers (very short and light individuals), none of the studied formulae was able to accurately predict FFM: this latter was overestimated in subjects with low and underestimated in those with high FFM. In the other group (INCAP employees), only values predicted by Lukaski formula were in perfect agreement with FFM, showing no bias depending on FFM changes. In conclusion, the available BIA formulae seem to be specific for population from which they were derived and could not be used on samples from other populations if not cross-validated.

Adolescent↗

PubMed1991

Modeling the cannabinoid receptor: a three-dimensional quantitative structure-activity analysis.

The structure-activity relationship studies that have been reported for cannabinoids suggest that 1) the conformation of the C-ring at the C9 position, 2) the A-ring phenolic hydroxyl, and 3) the hydrophobic side chain are important determinants for the production of analgesia, as well as other cannabinoid effects. However, either these previous structure-activity studies described for cannabinoid compounds have not been quantitative in nature or the prediction of the activity of known and unknown compounds based on molecular structure has not been tested in a comprehensive manner. In this study we describe a three-dimensional molecular modeling program using comparative molecular field analysis to derive quantitative structure-activity relationships fitting pharmacological potencies and binding affinities of cannabinoids. The analysis has proven to accurately fit the pharmacological activity of cannabinoid analogs, with cross-validated r2 values of greater than 0.3 and final analysis r2 values of greater than 0.88. Additionally, this study has further characterized the steric and electrostatic properties that account for the variations in their potency. The results from this study indicate that steric repulsion behind the C-ring is associated with decreased predicted binding affinity and pharmacological potency. On the other hand, the steric bulk of a side chain that is extended up to seven carbons contributes to predictions of increased binding affinity and potency. The electrostatic fields of cannabinoid analogs also contribute to the predicted in vitro and in vivo potencies. If the biological activities we have investigated are assumed to be the result of interaction with a single binding site, this method indicates the structural and physicochemical properties necessary for binding to the receptor and producing an effect. By defining cannabinoid binding affinity and behavioral activity pharmacophores, this method can be used for designing cannabinoid agonists and it is capable of predicting the activity of unknowns, thereby serving to facilitate rational drug design.

Animals↗

PubMed1991

The middle schooling process: influences on science and mathematics achievement from the longitudinal study of American youth.

The purpose of this longitudinal study was to develop and test a model of middle school science and mathematics achievement with a national probability sample of 3,116 seventh-grade public school students. Eighth-grade achievement was viewed as a function of readiness attributes on entry into seventh grade and intervening parent, peer, self, and classroom affective and behavioral measures. Data were collected from students and parents over three time periods in seventh and eighth grade. Results of a revised structural model showed that prior achievement in science and math was a strong mediator of effects in the process of schooling. Grades in sixth grade, parental expectations, parent educational attainment, and motivation had moderately strong indirect effects on eighth-grade achievements. Also notable was the positive direct influence of perceptions of classroom context on science and math achievement growth and the negative direct influence of sex (in favor of girls) on science achievement growth. Cross-validation on a split-half sample did not disconfirm the model. It was concluded that while prior achievement had a dominant influence in the schooling process, other variables including parental expectations, motivation, and classroom context do contribute to the schooling process and can be a focal point for improving school success. These and other factors, though helpful, may be most effective well before the middle school years.

Achievement↗

PubMed1990

Stress, health and performance in university students.

The hypothesis tested was that cognitive factors in the generation of stress, namely perceived coping incapacity (PCI), relate to the extent of psychosomatic ailments. We developed a new scale, CAST (Cognitive Appraisal Stress Test), with a 10 point self-rating of PCI using 16 questions on different aspects of perceived demands versus perceived capabilities. For the assessment of ailments relevant to stress we constructed a scale (PSYSOM) based on a translation of the Scheuch-Vogel symptom list (developed in the GDR from self-report of teachers who could not work as a result of work stress). A special scoring system combined duration, intensity and frequency of 30 symptoms and provided normalisation of distributions. The scales were administered on separate days to 75 female students in their second year (working to obtain admission to an honours degree course). The hypothesis was strongly supported, since the overall scale of PSYSOM correlated with CAST: r = 0.42, p less than .001. The physiologically defined subscale of 'skeletal muscle tension' correlated r = 0.65 with CAST. Subsequently CAST and PSYSOM have been shortened and factor analysed to give, in each case, a general principal component scale and several identifiable varimax subscales. Research using these scales on employees has now provided cross-validation of the first findings reported here.

Achievement↗

PubMed1990

Secondary structure prediction and protein design.

For non-homologous proteins, and after cross-validation, the methods reviewed in this article exhibit a probability index (percentage of correctly predicted residues per predicted residues) of 59-65.5% according to the methods employed with a standard deviation of 7% for three conformational states: alpha-helix, beta-strand and coil. These present limitations in the accuracy of prediction are related both to the limited number of known structures and to the effect of long-range interactions. The methods based on sequence similarity can improve the accuracy of prediction by expressing the homology of the protein to be predicted explicitly with proteins in the database. Under these circumstances, the probability index can reach 87% with a standard deviation of 6.6%. This property was used for modelling homologous proteins by assisting in amino acid sequence alignments. Examples will be given for the alignment of serine proteinases.

Amino Acid Sequence↗

PubMed1990

Marital status, social support, and health transitions in chronic disease patients.

Married persons tend to be healthier, both physically and mentally, than unmarried persons. We tested the hypothesis that being married results in better physical and mental health outcomes for chronic disease patients (N = 1,817) by increasing social support. We modeled health outcomes one year later, controlling for initial health status. Cross-validation studies of two random halves of the sample supported an indirect effect of marital status on mental health through social support, but did not support a relationship, direct or indirect, of either marital status or social support with physical health outcomes. In addition, specific types of functional support were not differentially predictive of mental health status.

Adaptation, Psychological↗

PubMed1990

Prediction of body water and fat in adult Danes from measurement of electrical impedance. A validation study.

The purpose of the present study was to test the validity of the electrical impedance method for estimation of total body water and lean body mass in adult Danes with large differences of obesity and fat distribution, and to develop algorithms for estimation of body water, lean body mass and fat from measurement of impedance. The results of the electrical impedance method were compared in 139 Danes aged 35-65 years, to those of a four-compartment-model based on measurements of both total body potassium (whole body counting) and total body water (dilutometry). The comparison confirmed the validity of the impedance method. Equations for predicting body water and lean body mass in Danes are given. A significant difference was found between two instruments of the same model. In spite of the fact that the test resistance supplied by the manufacturer gave identical measurements, measurements on humans diverged on average by 31 ohm. A cross-validation study showed that most of the algorithms found in the literature for predicting lean body mass from impedance yield reliable results.

Adipose Tissue↗

PubMed1990

Mortality determinants and prediction of outcome in high risk newborns.

The aim of this study was to determine independent patient-related predictors of mortality in high risk newborns admitted at our centre. The study population comprised 100 consecutive newborns each, from the premature unit (PU) and sick baby care unit (SBCU), respectively. Thirteen high risk factors (variables) for each of the two units, were entered into a multivariate regression analysis. Variables with independent predictive value for poor outcome (i.e., death) in PU were, weight less than 1 kg, hyaline membrane disease, neurologic problems, and intravenous therapy. High risk factors in SBCU included, blood gas abnormality, bleeding phenomena, recurrent convulsions, apnea, and congenital anomalies. Identification of these factors guided us in defining priority areas for improvement in our system of neonatal care. Also, based on these variables a simple predictive score for outcome was constructed. The prediction equation and the score were cross-validated by applying them to a 'test-set' of 100 newborns each for PU and SBCU. Results showed a comparable sensitivity, specificity and error rate.

Forecasting↗

PubMed1989

Validity of anthropometric equations for determination of changes in body composition in adult males during training.

Twenty-six males (26.5 +/- 6.0 yr; X +/- SD) were studied before and after a fourteen week endurance training program to determine the validity of anthropometric equations for estimating changes in body composition (BC). Anthropometric measures included skinfolds (SF), circumferences, and diameters. Body density (BD) was determined by underwater weighing corrected for residual lung volume. Training resulted in an increase in BD (1.061 +/- 0.002 to 1.067 +/- 0.002 g/ml; X +/- SEM) and decreases in body weight (73.0 +/- 2.1 to 71.4 +/- 2.0 kg), relative fat (16.6 +/- 0.9 to 14.1 +/- 0.8%), fat weight (12.4 +/- 1.0 to 10.2 +/- 0.8 kg), and seventeen of the anthropometric measures (p less than 0.05). Cross-validation of twenty-four equations revealed validity coefficients (r2) and total error in relative fat (RFE) of r2 = 0.40-0.77 and RFE = 2.60-10.15% before training and r2 = 0.14-0.61 and RFE = 2.62-9.45% after training. Linear and base 10 logarithmic (log10) equations using primarily SF measures tended to have higher r2 and lower RFE than equations based on quadratic and natural logarithmic (loge) models and other anthropometric measures. Paired t-tests revealed that of these equations with higher r2 and lower RFE, only the linear equation by Forsyth & Sinning (BD = 1.10647--0.00162(scapSF)--0.00144(abdSF)--0.00077(triSF++ +) + 0.00071(midaxSF] was a stable predictor of BD during training. These results suggest that many existing equations may not be accurate predictors of changes in BC during training.

Adult↗

PubMed1989

Spirometric standards for healthy male lifetime nonsmokers.

The FVC, FEV1.0, FEF25-75%, and FEV1.0/FVC (%) were measured in 162 males aged 18.9-78.6 yr using a Stead-Wells spirometer. Multiple regression equations were generated to predict these lung parameters from the best weighted combination (p less than or equal to 0.05) of age, standing height, sitting height, biacromial breadth, and chest expansion (FVC: R = 0.843, SEE = 513 ml BTPS; FEV1.0: R = 0.850, SEE = 436 ml BTPS; FEF25-75%: R = 0.665, SEE = 997 ml/s BTPS; FEV1.0/FVC: R = 0.537, SEE = 4.95%). The lower limit of normality was defined as the predicted value minus the 95% confidence interval (one-tailed test). Cross-validation of other FVC, FEV1.0, and FEF25-75% equations in the literature indicate that all the FVC and FEV1.0 ones are unsuitable for the sample and instrumentation used in this investigation.

Adult↗

PubMed1989

Lung volume and spirometric standards for healthy female lifetime nonsmokers.

The FRC, RV, VC, TLC, RV/TLC (%), FVC, FEV1.0, FEF25-75%, and FEV1.0/FVC (%) were measured in 161 South Australian females aged 18.4-81.2 yr using a Stead-Wells spirometer and helium analyzer. Multiple regression equations were generated to predict these lung volume and spirometric parameters from the best weighted combination of age, mass, standing height, and various other anthropometric variables (FRC: R = 0.715, SEE = 387 ml; RV: R = 0.684, SEE = 256 ml; VC: R = 0.815, SEE = 383 ml; TLC: R = 0.754, SEE = 468 ml; RV/TLC: R = 0.780, SEE = 4.2%; FVC: R = 0.839, SEE = 375 ml; FEV1.0: R = 0.869, SEE = 326 ml; FEV1.0/FVC: R = 0.644, SEE = 5.7%; FEF25-75%: R = 0.753, SEE = 802 ml/s). The range of normality for the lung volumes was defined as the predicted value plus or minus the 95% confidence interval (two-tailed test), and the lower limit of normality for the spirometric variables was designated as the predicted value minus the 95% confidence interval (one-tailed test). Cross-validation of other equations in the literature indicates that they are of limited use for the sample and instrumentation used in this study.

Adult↗

PubMed1989

Validity and intertester error of anthropometric estimations of body density.

The purposes of this study were to evaluate the intertester error in anthropometry and its effect on estimations of actual body density (ABD), as well as to assess the validity of predicted body density (PBD) from selected anthropometric equation. Sixteen adult males X +/- SD = 21.98 +/- 1.65 years) volunteered to be measured by three experienced and reliable testers at eight circumference, two diameter, and eight skinfold sites as well as by underwater weighing. Eighteen commonly used anthropometric equations which predict body density were selected for evaluation. Repeated measures ANOVA with Tukey post-hoc comparisons were used to determine differences between testers for each anthropometric measure and the PBD values, as well as between the PBD values and ABD. There were significant (p less than 0.01) differences between testers for all anthropometric measures except bi-iliac diameter and hips circumference. However, the magnitude of the differences between testers were considered to be quite small and of no practical concern. There were significant differences among the PBD values and ABD for all equations except the base 10 logarithmic transformations of Durnin and Womersley. Seven of the equations, however, resulted in a constant error of less than or equal to 0.0085 g.ml-1. Overall, there was no single anthropometric equation which satisfied all of the cross-validation criteria suggested by previous investigators, however the linear equation of Forsyth and Sinning which included the scapular and abdominal skinfolds, as well as the bitrochanteric diameter resulted in an extremely small constant error (0.0024 g.ml-1) and an appropriate standard deviation of the PBD values (0.0127 g.ml-1).

Adult↗

PubMedDate not supplied

Measurement of quality-of-life changes in patients with Alzheimer's disease.

A scale was developed that appears to be a reliable and sensitive indicator of differences and changes in the quality of life of patients with Alzheimer's disease (AD) as the disease progresses. The instrument was generated from interviews with spouses and/or caregivers of AD patients, and was designed to be completed easily by the caregiver. The bipolar analog scale was based on the daily behavior of AD patients and was developed in three steps involving interviews with caregivers (step I), testing of questionnaires, and validation of the instrument. At step II, four iterative trials, each with a different version of the questionnaire, were performed for validation. The final form of the Progressive Deterioration Scale (PDS) contained 27 quality-of-life factors and was cross-validated in a second, separate study. The Global Deterioration Scale (GDS) for cognitive function was used as the external criterion for the analysis of the quality-of-life scales. Statistical analysis indicates that the PDS provides a reliable, valid, and sensitive measure of changes in the quality of life as AD progresses. The PDS achieved 80% overall accuracy in discriminating non-AD elderly patients from patients in the early, middle, and late stages of AD. This scale may provide a useful means for clinical evaluation (phase III) of potential medications to alleviate the symptoms of AD.

Aged↗

PubMed1989

Sex determination with the head of the radius.

Analysis of a sample of 1108 radii corresponding to 567 black and white North Americans in the Terry Collection at the Smithsonian Institution demonstrates that the diameter of the radial head is an accurate sex discriminator for human remains. A simple "radial-head method" of sex determination consists of measuring the maximum and minimum diameters of the head and comparing such measurements with the test cutoff points. The subject is female when the maximum radial head diameter (either left or right) is less than or equal to 21 mm, and male when the maximum diameter (either left or right) is greater than or equal to 24 mm. The same decisions apply to the minimum diameters of 20 mm or less and 23 mm or more, respectively. When the maximum diameter is 23 mm or the minimum is 22 mm, the subject is more likely male; when the maximum diameter is 22 mm or the minimum is 21 mm, the subject is more likely female. The sample frequency of any one of these latter diameters is never more than 16%. Cross-validation of the method with a sample of 50 pairs of radii of the Terry Collection, different from the original specimens, resulted in 92% sexing accuracy when using the left radius singly, 94% accuracy when using the right radius singly, and 96% accuracy when using both radii jointly.

Anthropology, Physical↗

PubMed1987

Prediction of neuroleptic-induced dystonia.

For patients receiving neuroleptics, age, sex, neuroleptic potency, and dose all influence the likelihood of a dystonic reaction. Little is known, however, of the relative importance of these factors or of the feasibility of predicting dystonia in individual patients. We reviewed 135 charts of psychotic inpatients to examine these factors and their usefulness in predicting dystonia. Age, sex, neuroleptic type, dose, and occurrence of dystonia were recorded for the first 4 days of drug treatment and were used to construct a linear discriminant function that classified the cases as to whether dystonia was expected. Internal cross-validation was performed, and the error rate of this classification procedure was calculated. Forty-nine (36%) of the patients had dystonia. A younger age was the most powerful predictor of dystonia. Male gender was second in predictive power with minor effects from neuroleptic dose and potency. The overall error rate (false-positive and false-negative errors combined) of the discriminant function was 30%. These results suggest the possibility of predicting dystonia in individual patients but should be regarded with caution since the predictive procedure has not been tested prospectively. If confirmed, these data may allow treatment strategies that protect patients from dystonia while sparing patients not at risk unnecessary treatment with antiparkinson agents.

Adolescent↗