Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Predictive”

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

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

At least 703 records · Page 39Linked to original sources
PubMed2003

[Prediction of postoperative nausea and vomiting using an artificial neural network].

OBJECTIVE: Postoperative nausea and vomiting (PONV) are still frequent side-effects after general anaesthesia. These unpleasant symptoms for the patients can be sufficiently reduced using a multimodal antiemetic approach. However, these efforts should be restricted to risk patients for PONV. Thus, predictive models are required to identify these patients before surgery. So far all risk scores to predict PONV are based on results of logistic regression analysis. Artificial neural networks (ANN) can also be used for prediction since they can take into account complex and non-linear relationships between predictive variables and the dependent item. This study presents the development of an ANN to predict PONV and compares its performance with two established simplified risk scores (Apfel's and Koivuranta's scores). METHODS: The development of the ANN was based on data from 1,764 patients undergoing elective surgical procedures under balanced anaesthesia. The ANN was trained with 1,364 datasets and a further 400 were used for supervising the learning process. One of the 49 ANNs showing the best predictive performance was compared with the established risk scores with respect to practicability, discrimination (by means of the area under a receiver operating characteristics curve) and calibration properties (by means of a weighted linear regression between the predicted and the actual incidences of PONV). RESULTS: The ANN tested showed a statistically significant ( p<0.0001) and clinically relevant higher discriminating power (0.74; 95% confidence interval: 0.70-0.78) than the Apfel score (0.66; 95% CI: 0.61-0.71) or Koivuranta's score (0.69; 95% CI: 0.65-0.74). Furthermore, the agreement between the actual incidences of PONV and those predicted by the ANN was also better and near to an ideal fit, represented by the equation y=1.0x+0. The equations for the calibration curves were: KNN y=1.11x+0, Apfel y=0.71x+1, Koivuranta 0.86x-5. CONCLUSION: The improved predictive accuracy achieved by the ANN is clinically relevant. However, the disadvantages of this system prevail because a computer is required for risk calculation. Thus, we still recommend the use of one of the simplified risk scores for clinical practice.

Artificial Intelligence↗
PubMed1999

Severe brain injury in children: long-term outcome and its prediction using somatosensory evoked potentials (SEPs).

OBJECTIVE: To evaluate the outcome of children 1 and 5 years after severe brain injury (Glasgow Coma Score < 8) using a functional measure [Glasgow Outcome Scale (GOS)] and a health status measure (the Torrance Health State (HUI:1)) and to determine the ability of somatosensory evoked potentials (SEPs) to predict these long-term outcomes. DESIGN: Prospective study. SETTING: A 16-bed paediatric intensive care unit in a tertiary children's hospital. PATIENTS AND PARTICIPANTS: 105 children with severe brain injury. INTERVENTIONS: SEPs were recorded once in the first week after admission. Outcome was assessed 1 and 5 years after injury using the GOS and at 5 years after injury using HUI:1. MEASUREMENTS AND RESULTS: At 5 years, using the GOS, 46 (43.8%) children had a good outcome, 10 (9.5%) were moderately disabled, 2 (1.9%) severely disabled, 3 (2.9%) vegetative and 44 (41.9%) had died. At 5 years, 17 of 40 (42.5%) survivors from 1 year had changed outcomes: 12 had improved, 3 had worsened and 2 had died. For a normal SEP, positive predictive power was 85.4%, sensitivity 62.5%, specificity 87.8%, negative predictive power 67.2% and the positive likelihood ratio was 5.1. For bilaterally absent responses, positive predictive power was 90.9%, sensitivity 61.2%, specificity 94.6%; negative predictive power 73.6% and the positive likelihood ratio was 11.4. Outcomes using HUI:1 were: 30 (28.6%) had a good quality of life, 21 (20.0%) had a moderate quality of life, 7 (6.7%) a poor quality, 44 died (41.9%) and 3 (2.9%) survived in a state deemed worse than death. For a normal SEP, positive predictive power was 85.4%, sensitivity 68.6%, specificity 88.9%, negative predictive power 75.0% and the positive likelihood ratio was 6.2. For bilaterally absent responses, positive predictive power was 93.9%, sensitivity 57.4%, specificity 96.1%, negative predictive power 68.1% and the positive likelihood ratio was 14.6. CONCLUSION: The outcome for children with severe brain injury should be assessed 5 years after injury because important changes occur between 1 year and 5 years. Differences exist between outcomes assessed using the GOS and HUI:1 as they measure slightly different aspects of function. Consideration should therefore be given to using both measures. SEPs are excellent predictors of long-term outcome measured by either the GOS or the HUI:1.

Adolescent↗
PubMed2003

Presentation assessment of minor histocompatibility antigens by predictive proteasomal cleavage analysis.

Minor histocompatibility peptides (mHps) derived from polymorphic segments of endogenous proteins are thought to be targets for graft-versus-host and graft-versus-leukemia reactions after HLA-identical stem cell transplantation. A great majority of antigenic peptides is generated by fragmentation of proteins in the course of proteasomal processing. An algorithm was recently developed to predict cleavage sites during proteasomal processing. We tested the accuracy of the algorithm to predict mHps using 18 amino acid (AA) sequences of minor histocompatibility antigens (mHags) encoded by autosomal genes representing single nucleotide polymorphisms or by Y-chromosomal genes. The algorithm correctly predicted the C-termini of 11 of 13 experimentally confirmed mHps: 1) Correct prediction of C- and N-termini, e.g., for HA-1(H); 2) Correct prediction of C- and N-termini while anticipating intra-epitope cleavage sites, e.g., for SMCY-A*0201; 3) Correct prediction of C-termini and N-terminal extensions, e.g., for HA-8(R/V); and 4) Correct prediction of C-termini and N-terminal extensions while anticipating intra-epitope cleavage sites, e.g., for UTY-B8. Analysis of experimentally unconfirmed allelic counterparts of four autosomal mHags showed that AA substitutions either led to the insertion of an epitope-destroying cleavage site (e.g., in HA-1(R)) or abolished the correct C-terminus (e.g., in HA-2(M)). The proteasomal processing algorithm provides reliable data on the generation of mHps and forecasts their presence or absence. Combined with MHC class I ligand prediction, it can be a useful tool for the prediction of generation and presentation of new CTL epitopes derived from minor histocompatibility antigens.

Algorithms↗
PubMed2004

Multivariate prediction of clarified butter composition using Raman spectroscopy.

Raman spectroscopy has been used to predict the abundance of the FA in clarified butterfat that was obtained from dairy cows fed a range of levels of rapeseed oil in their diet. Partial least squares regression of the Raman spectra against FA compositions obtained by GC showed good prediction for the five major (abundance >5%) FA with R2 = 0.74-0.92 and a root mean SE of prediction (RMSEP) that was 5-7% of the mean. In general, the prediction accuracy fell with decreasing abundance in the sample, but the RMSEP was <10% for all but one of the 10 FA present at levels >1.25%. The Raman method has the best prediction ability for unsaturated FA (R2 = 0.85-0.92), and in particular trans unsaturated FA (best-predicted FA was 18:1 t delta9). This enhancement was attributed to the isolation of the unsaturated modes from the saturated modes and the significantly higher spectral response of unsaturated bonds compared with saturated bonds. Raman spectra of the melted butter samples could also be used to predict bulk parameters calculated from standard analyzes, such as iodine value (R2 = 0.80) and solid fat content at low temperature (R2 = 0.87). For solid fat contents determined at higher temperatures, the prediction ability was significantly reduced (R2 = 0.42), and this decrease in performance was attributed to the smaller range of values in solid fat content at the higher temperatures. Finally, although the prediction errors for the abundances of each of the FA in a given sample are much larger with Raman than with full GC analysis, the accuracy is acceptably high for quality control applications. This, combined with the fact that Raman spectra can be obtained with no sample preparation and with 60-s data collection times, means that high-throughput, on-line Raman analysis of butter samples should be possible.

Butter↗
PubMed1992

Differences between predictive characteristics of signal-averaged electrocardiographic variables for postinfarction sudden death and ventricular tachycardia.

Several studies indicate that the electrophysiologic substrate for sustained ventricular tachycardia differs from that of ventricular fibrillation. This prospective study examined whether there were clinically relevant differences between the predictive values of the standard time-domain signal-averaged (SA) electrocardiographic (ECG) variables for ventricular tachycardia and sudden death after myocardial infarction. Predischarge SA electrocardiograms were recorded in 332 patients after infarction. During a follow-up period of greater than or equal to 6 months, there were 12 sudden deaths (3.6%), 14 patients (4.2%) developed spontaneous sustained ventricular tachycardia and 20 patients (6%) died of circulatory failure. The sensitivity, specificity and positive predictive accuracy of the numerical values of the time-domain SA electrocardiographic variables for predicting sudden death and ventricular tachycardia were compared. The optimal criteria for predicting ventricular tachycardia required the positivity of greater than or equal to 2 of the standard time-domain SA variables, whereas the optimal criteria for predicting sudden death required the positivity of all 3 variables. A high specificity was sustained over a wider range of sensitivity for sudden death than it was for ventricular tachycardia and the values of the variables which provided the same sensitivity for sudden death and ventricular tachycardia were different. For a sensitivity of 70%, the positive predictive accuracy was 31% for predicting sudden death and 13% for predicting ventricular tachycardia. The study concludes that differences in the predictive characteristics of variables for ventricular tachycardia and sudden death may be used to refine postinfarction risk stratification.

Adult↗
PubMed1993

Are routine non-invasive tests useful in prediction of outcome after myocardial infarction in elderly people?

Many patients with acute myocardial infarction undergo tests to identify ischaemia, left-ventricular dysfunction, and arrhythmias. We examined the usefulness of these tests in clinical practice by comparing the ability of a cardiologist to predict outcome at 1 year after infarction with and without knowledge of the results of an exercise test, radionuclide angiogram, and 24 h Holter electrocardiographic (ECG) recording. The study was limited to patients older than 65 years, who have a greater risk of cardiovascular sequelae and undergo fewer interventional procedures. The patient's own cardiologist predicted outcome on a standard rating scale, based on clinical findings and routine hospital tests. He then made a second prediction after seeing the non-invasive test results. Two other cardiologists not involved in the care of the patient independently made similar predictions. Success in predicting outcome was assessed by comparison of differences between the first and second predictions in the area under receiver operating characteristic curves. During 1 year's follow-up there were 24 cardiovascular deaths and 3 recurrent myocardial infarctions among the 147 patients. There were no significant differences in mean curve areas between the first and second predictions for the patients' own cardiologist (0.62 [SE 0.06] vs 0.60 [0.06]) or the other cardiologists (0.63 [0.06] vs 0.64 [0.06] and 0.61 [0.06] vs 0.65 [0.06]). All predictions were significantly (p < 0.05) better than chance. Prediction of outcome in older patients after myocardial infarction is not improved by knowledge of the results of an exercise test, radionuclide angiogram, or 24 h Holter ECG recording.

Aged↗
PubMed2005

Development of radiology prediction models using feature analysis.

RATIONALE AND OBJECTIVES: This article provides an introduction to prediction models and their application in diagnostic imaging research. Prediction models capitalize on the different degrees of association among variables to make a prediction of a health state, formulate a rule, or quantify individual contributions of various predictor variables. The purpose of this article is to elucidate the rationale, implication, and interpretation of prediction models using imaging features. MATERIALS AND METHODS: The techniques and challenges of developing, testing, and implementing prediction models are described. Prediction model development methods are similar to data-mining techniques. RESULTS: Learning objectives are to review prediction rule (model) methods, learn how prediction models may be applied to feature analysis, and understand the challenges of developing, testing, and implementing prediction models.

Decision Support Techniques↗
PubMed2004

Venous Doppler in the prediction of acid-base status of growth-restricted fetuses with elevated placental blood flow resistance.

OBJECTIVE: This study was undertaken to test which venous Doppler parameter offers the best prediction of acid-base status at birth in pregnancies complicated by intrauterine growth restriction (IUGR) caused by placental dysfunction. STUDY DESIGN: A prospective cross-sectional Doppler study of IUGR fetuses with abnormal umbilical artery Doppler and birth weight less than the 10th percentile. Absence of atrial systolic forward velocities in the ductus venosus (DV) (DV-RAV) and umbilical vein (UV) pulsations were noted and multiple venous indices were calculated for the inferior vena cava (IVC) and DV (IVC and DV preload index, peak velocity index [PVIV] and pulsatility index [PIV] and the DV S/a ratio). Doppler indices, UV pulsations, and DV- RAV were related to an umbilical artery cord pH <7.20, and a pH <7.00 and/or base deficit greater than -13 (severe metabolic compromise) in neonates delivered by cesarean section without labor. RESULTS: In 122 fetuses all venous Doppler indices were equally predictive of a pH <7.20, with the exception of the IVC PVIV. No Doppler index predicted severe metabolic compromise. Bayesian analysis of individual Doppler parameters showed comparable outcome prediction with the highest sensitivity for the IVC PIV (76%) and the highest specificity for DV-RAV (96%). Combined assessment of the IVC, DV, and UV provided the most accurate outcome prediction. Doppler abnormality in either vessel identified 89% of neonates with pH <7.20 (negative predictive value 92%) and 10 of 11 neonates with severe metabolic compromise. Prediction was most specific (84%) when Doppler parameters were abnormal in all 3 vessels. CONCLUSION: IVC, DV, and UV Doppler parameters correctly predict acid-base status in a significant proportion of IUGR neonates. Combination, rather than single vessel assessment provides the best predictive accuracy. While the choice of Doppler index can be guided by operator preference, familiarity with the examination technique of all 3 vessels is encouraged to offer the highest flexibility in clinical practice.

Acid-Base Imbalance↗
PubMed2006

Fetal transcerebellar diameter measurement for prediction of gestational age in twins.

OBJECTIVE: This study was undertaken to determine the accuracy of our previously published and prospectively validated institution-specific singleton transcerebellar diameter (TCD) nomogram in the prediction of gestational age (GA) in twin pregnancies. We further evaluated whether the prediction of GA in twin gestations using the singleton TCD nomogram differs between monochorionic and dichorionic twins. STUDY DESIGN: In our previously published studies, we retrospectively constructed a cross-sectional nomogram using TCD measurements in 24,026 well-dated, singleton fetuses, and prospectively validated the nomogram using 2,597 singleton fetuses. The current study comprised of 1,278 well-dated twins (19.6% monochorionic) seen in our ultrasound unit between August 1994 and May 2003, and the singleton TCD nomogram was validated in these twin gestations. The actual GA was subtracted from the GA predicted by the TCD nomogram and the concordance between actual and predicted GAs was assessed on the basis of the Pearson's correlation coefficient (r). This was performed separately for monochorionic and dichorionic twins. RESULTS: Concordance between the actual and predicted twin TCD measurements based on our previously published singleton TCD nomogram was high (Pearson's correlation, r = 0.95, P < .0001). Between 16 and 23 weeks' gestation, the predicted mean GA was within 6 days of actual GA. Between 24 and 30 weeks, the predicted mean GA was within 3 days, and at 32 weeks or more, the predicted mean GA was within 5 days of the actual GA. Prediction of GA based on the singleton TCD nomogram was equally accurate in both monochorionic and dichorionic twin gestations (P = .686). CONCLUSION: This study demonstrates that our previously validated singleton TCD nomogram is reliable and accurate in twins irrespective of placental chorionicity.

Adult↗
PubMed2005

Use of preoperative magnetic resonance imaging to predict rotator cuff tear pattern and method of repair.

PURPOSE: To determine the magnetic resonance imaging (MRI) criteria for predicting rotator cuff tear pattern and method of repair. TYPE OF STUDY: Retrospective MRI/arthroscopy correlation. METHODS: Sixty-six preoperative MRI scans were evaluated. The maximum medial to lateral length (L) of the tear was measured on T2-weighted coronal cuts. The maximum anterior to posterior width (W) was measured on T2-weighted sagittal cuts. The cases were divided into 3 groups: group 1, short-wide tears, L < or = W, L < 2 cm; group 2, long-narrow tears, L > W, W < 2 cm; and group 3, long-wide tears, L > or = 2 cm, W > or = 2 cm. RESULTS: Of the 66 MRI scans, 55 were adequate for standardized measurement. Group 1, 16 cases: 15 were found at arthroscopy to be crescent-shaped tears repaired end-to-bone; 1 was repaired with interval slides. Group 2, 22 cases: all 22 were repaired side-to-side/margin convergence. Group 3, 17 cases: 12 required interval slides, 1 partial repair was performed, and 4 were repaired side-to-side/margin convergence. CONCLUSIONS: Tear pattern and method of repair can be predicted on high-quality MRI scan. Group 1, L < or = W and L < 2 cm, predicts a crescent-shaped tear and end-to-bone repair (positive predictive value, 93.8%). Group 2, L > W and W < 2 cm, predicts a longitudinal tear and side-to-side/margin convergence repair (positive predictive value 100%). Group 3, L > or = 2 cm and W > or = 2 cm, predicts a massive contracted tear and that primary end-to-bone or side-to-side repairs are usually not possible and that interval slides or partial repair may be necessary (positive predictive value, 76.5%). The overall diagnostic model based on usable MRI scans significantly predicted arthroscopic findings (P < .001 for chi-square test). LEVEL OF EVIDENCE: Level III, development of diagnostic criteria with universally applied reference (nonconsecutive patients).

Adult↗
PubMed2006

Spatio-temporal patient-individual assessment of synchronization changes for epileptic seizure prediction.

OBJECTIVE: Abnormal synchronization of neurons plays a central role for the generation of epileptic seizures. Therefore, multivariate time series analysis techniques investigating relationships between the dynamics of different neural populations may offer advantages in predicting epileptic seizures. METHODS: We applied a phase and a lag synchronization measure to a selected subset of multicontact intracranial EEG recordings and assessed changes in synchronization with respect to seizure prediction. RESULTS: Patient individual results, group results, spatial aspects using focal and extra-focal electrode contacts as well as two evaluation schemes analyzing decreases and increases in synchronization were examined. Averaged sensitivity values of 60% are observed for a false prediction rate of 0.15 false predictions per hour, a seizure occurrence period of half an hour, and a prediction horizon of 10 min. For approximately half of all 21 patients, a statistically significant prediction performance is observed for at least one synchronization measure and evaluation scheme. CONCLUSIONS: The results indicate that synchronization changes in the EEG dynamics preceding seizures can be used for seizure prediction. Nevertheless, the underlying pathogenic mechanisms differ and both decreases and increases in synchronization may precede epileptic seizures depending on the structures investigated. SIGNIFICANCE: The prediction method, optimized values of intervention times, as well as preferred brain structures for the EEG recordings have to be determined for each patient individually offering the chance of a better patient-individual prediction performance.

Adolescent↗
PubMed2006

Prediction of atrial fibrillation after coronary artery bypass grafting: the role of chemoreflex-sensitivity and P wave signal averaged ECG.

Atrial fibrillation (AF) after coronary artery bypass grafting (CABG) results in a prolonged hospital stay associated with higher costs. In our study P wave triggered P wave signal averaged ECG and chemoreflex-sensitivity (CHRS) was performed on 101 consecutive patients with sinus rhythm before CABG in order to evaluate the utility of these methods to predict AF. A CHRS below 3.0 ms/mm Hg was predefined as a pathological CHRS. Postoperative AF was observed in 37 (37%) of 101 patients. Patients with AF were older (68.4+/-6.9 vs. 63.8+/-9.4 years, p<0.01), had a longer filtered P-wave duration (FPD) (133.6+/-10.2 vs. 123.6+/-14.9 ms, p<0.0001), a lower root mean square voltage of the last 20 ms of the P wave (RMS 20) (2.86+/-0.88 vs. 5.10+/-2.73 microV, p<0.0001) and a significantly lower CHRS (3.32+/-1.83 vs. 4.17+/-2.19 ms/mm Hg, p<0.05). A cut-off point (COP) of FPD> or =124 ms and RMS 20< or =3.7 microV achieved a specificity of 75%, a sensitivity of 78%, a negative predictive value of 86%, a positive predictive value of 64% and an accuracy of 76% for prediction of AF. The predictive power was lower for a pathological CHRS which achieved a specificity of 63%, a sensitivity of 60%, a negative predictive value of 73%, a positive predictive value of 48% and an accuracy of 61%. A stepwise logistic regression analysis of all preoperative variables identified COP (odds ratio 8.21; 95% CI, 2.02-33.37, p<0.003) as independent predictor. Patients with postoperative AF stayed longer in the intensive care unit (2.9+/-1.7 vs. 1.3+/-0.5 days, p<0.0001) and in hospital (13.5+/-4.3 vs. 11.4+/-1.1 days, p<0.0004). The results of our study show that the risk for AF after CABG could preoperatively be predicted with P wave signal averaged ECG and an analysis of CHRS. The predictive power of the COP could be used for a preoperative risk stratification and a corresponding prophylactic therapy in order to reduce costs.

Aged↗
PubMed2006

Electrocardiographic prediction of left ventricular geometric patterns in patients with essential hypertension.

BACKGROUND: The present study sought to determine the diagnostic value of electrocardiographic voltage criteria in predicting geometry patterns in patients with essential hypertension. METHODS: Patients with essential hypertension (n=125) according to left ventricular mass index and relative wall thickness as determined by echocardiography were assigned in the following groups: normal geometry (N, n=50), concentric remodeling (CR, n=12), concentric hypertrophy (CH, n=28) and eccentric hypertrophy (EH, n=35). Each patient underwent 12-lead ECG followed by determination of conventional voltage criteria as well as peak to peak QRS lengths in each lead. RESULTS: Voltage criteria such as Sokolow-Lyon, Cornell, Cornell product >2440, D1R+D3S >25 mm, and AVL R >11 mm could not significantly predict and discriminate geometric patterns of LVH. However, they all were very specific (range 97-100%) and showed very high positive predictive values (range 94-100%) for detecting abnormal geometry. DI peak >12 mm had a sensitivity 61%, specificity 67%, accuracy 63%, positive predictive value 81%, and negative predictive value 42% in predicting to differentiate CH from CR. Sum of the calculated values from the peak of the R to the nadir of the S wave in all limb leads >60 mm had sensitivity 68%, specificity 75%, accuracy 70%, positive predictive value 86% and negative predictive value 50% in predicting to differentiate CH from CR. CONCLUSIONS: Conventional ECG voltage criteria could not significantly discriminate specific geometry patterns observed in patients with essential hypertension.

Echocardiography↗
PubMed2005

Prediction of pathogen growth on iceberg lettuce under real temperature history during distribution from farm to table.

The growth of pathogenic bacteria Escherichia coli O157:H7, Salmonella spp., and Listeria monocytogenes on iceberg lettuce under constant and fluctuating temperatures was modelled in order to estimate the microbial safety of this vegetable during distribution from the farm to the table. Firstly, we examined pathogen growth on lettuce at constant temperatures, ranging from 5 to 25 degrees C, and then we obtained the growth kinetic parameters (lag time, maximum growth rate (micro(max)), and maximum population density (MPD)) using the Baranyi primary growth model. The parameters were similar to those predicted by the pathogen modelling program (PMP), with the exception of MPD. The MPD of each pathogen on lettuce was 2-4 log(10) CFU/g lower than that predicted by PMP. Furthermore, the MPD of pathogens decreased with decreasing temperature. The relationship between mu(max) and temperature was linear in accordance with Ratkowsky secondary model as was the relationship between the MPD and temperature. Predictions of pathogen growth under fluctuating temperature used the Baranyi primary microbial growth model along with the Ratkowsky secondary model and MPD equation. The fluctuating temperature profile used in this study was the real temperature history measured during distribution from the field at harvesting to the retail store. Overall predictions for each pathogen agreed well with observed viable counts in most cases. The bias and root mean square error (RMSE) of the prediction were small. The prediction in which mu(max) was based on PMP showed a trend of overestimation relative to prediction based on lettuce. However, the prediction concerning E. coli O157:H7 and Salmonella spp. on lettuce greatly overestimated growth in the case of a temperature history starting relatively high, such as 25 degrees C for 5 h. In contrast, the overall prediction of L. monocytogenes under the same circumstances agreed with the observed data.

Colony Count, Microbial↗
PubMed2005

Comparison of predictive equations for resting metabolic rate in healthy nonobese and obese adults: a systematic review.

BACKGROUND: An assessment of energy needs is a necessary component in the development and evaluation of a nutrition care plan. The metabolic rate can be measured or estimated by equations, but estimation is by far the more common method. However, predictive equations might generate errors large enough to impact outcome. Therefore, a systematic review of the literature was undertaken to document the accuracy of predictive equations preliminary to deciding on the imperative to measure metabolic rate. METHODS: As part of a larger project to determine the role of indirect calorimetry in clinical practice, an evidence team identified published articles that examined the validity of various predictive equations for resting metabolic rate (RMR) in nonobese and obese people and also in individuals of various ethnic and age groups. Articles were accepted based on defined criteria and abstracted using evidence analysis tools developed by the American Dietetic Association. Because these equations are applied by dietetics practitioners to individuals, a key inclusion criterion was research reports of individual data. The evidence was systematically evaluated, and a conclusion statement and grade were developed. RESULTS: Four prediction equations were identified as the most commonly used in clinical practice (Harris-Benedict, Mifflin-St Jeor, Owen, and World Health Organization/Food and Agriculture Organization/United Nations University [WHO/FAO/UNU]). Of these equations, the Mifflin-St Jeor equation was the most reliable, predicting RMR within 10% of measured in more nonobese and obese individuals than any other equation, and it also had the narrowest error range. No validation work concentrating on individual errors was found for the WHO/FAO/UNU equation. Older adults and US-residing ethnic minorities were underrepresented both in the development of predictive equations and in validation studies. CONCLUSIONS: The Mifflin-St Jeor equation is more likely than the other equations tested to estimate RMR to within 10% of that measured, but noteworthy errors and limitations exist when it is applied to individuals and possibly when it is generalized to certain age and ethnic groups. RMR estimation errors would be eliminated by valid measurement of RMR with indirect calorimetry, using an evidence-based protocol to minimize measurement error. The Expert Panel advises clinical judgment regarding when to accept estimated RMR using predictive equations in any given individual. Indirect calorimetry may be an important tool when, in the judgment of the clinician, the predictive methods fail an individual in a clinically relevant way. For members of groups that are greatly underrepresented by existing validation studies of predictive equations, a high level of suspicion regarding the accuracy of the equations is warranted.

Adolescent↗
PubMed2005

Computer-aided rodent carcinogenicity prediction.

The potential of the computer program PASS (Prediction Activity Spectra for Substances) to predict rodent carcinogenicity for chemical compounds was studied. PASS predicts carcinogenicity of chemical compounds on the basis of their structural formula and of structure-activity relationship analysis of known carcinogens and non-carcinogens. The data on structures and experimental results of 2-year carcinogenicity assays for 412 chemicals from the NTP (National Toxicological Program) and 1190 chemicals from the CPDB (Carcinogenic Potency Database) were used in our study. The predictions take into consideration information about species and sex of animals. For evaluation of the predictive accuracy we used two procedures: leave-one-out cross-validation (LOO CV) and leave-20%-out cross-validation. In the last case we randomly divided the studied data set 20 times into two subsets. The data from the first subset, containing 80% of the compounds, were added to the PASS training set (which includes about 46,000 compounds with about 1500 biological activity types collected during the last 20 years to predict biological activity spectra), the second subset with 20% of the compounds was used as an evaluation set. The mean accuracy of prediction calculated by LOO CV is about 73% for NTP compounds in the 'equivocal' category of carcinogenic activity and 80% for NTP compounds in the 'evidence' category of carcinogenicity. The mean accuracy of prediction for the CPDB database is 89.9% calculated by LOO CV and 63.4% calculated by leave-20%-out cross-validation. Influence of incorporation of species and sex data on the accuracy of carcinogenicity prediction was also investigated. It was shown that the accuracy was increased only for data on male animals.

Animals↗
PubMed1996

The unassisted respiratory rate-tidal volume ratio accurately predicts weaning outcome.

PURPOSE: To assess the accuracies of four commonly used parameters in predicting weaning outcome and whether breathing pattern changes during weaning. PATIENTS AND METHODS: We prospectively examined the predictive accuracies of four weaning parameters in mechanically ventilated patients in the medical and cardiac intensive care units of a 270-bed community teaching hospital. The spontaneous respiratory rate:tidal volume ratio (RVRi), negative inspiratory force (NIF), and spontaneous minute volume (VE) at the onset of weaning, and the RVR at 30 to 60 minutes of weaning (RVR30) were measured. Weaning decisions were made by patients' primary physicians independent of this study. Threshold values for computations of predictive values were as follows: RVR 100 < or = breaths per minute/L, NIF < or = -20 cm H2O, VE < or = 10 Lpm. Receiver operator curves were generated for each parameter. RESULTS: One hundred medical/cardiac intensive care unit patients were studied. Their mean age was 64.6 +/- 15.8 years, mean APACHE II score of 15.8 +/- 6.7 and mean duration of mechanical ventilation before the study of 4.9 +/- 8.1 days. RVRi sensitivity was 89%, specificity was 41%, positive predictive value was 72%, negative predictive value was 68%, and accuracy was 71%. The RVR30 sensitivity was 98%, specificity was 59%, positive predictive value was 83%, negative predictive value was 94%, and accuracy was 85%. Accuracies for the NIF and VE were 66% and 62%, respectively. The area under the receiver operator curve of the RVR30 (0.92 +/- 0.03) was higher than the RVRi (0.74 +/- 0.05), NIF (0.68 +/- 0.06) and VE (0.54 +/- 0.06) (p < 0.05). CONCLUSIONS: The RVR is more accurate than other commonly utilized clinical tools in predicting the outcome of weaning from mechanical ventilation. The RVR measured at 30 minutes is superior to the RVR in the first minute of weaning. The predictive accuracy and unique simplicity of the RVR justify its use in the care of mechanically ventilated patients.

Aged↗
PubMed2006

Clinical prediction rules: what are they and what do they tell us?

QUESTION: Clinical prediction rules are research-based tools that quantify the contributions of relevant patient characteristics to provide numeric indices that assist clinicians in making predictions. Clinical prediction rules have been used to describe the likelihood of the presence or absence of a condition, assist in determining patient prognosis, and help the classification of patients for treatment. The recent rapid rise in the use of clinical prediction rules raises questions about the conditions under which they may be used most appropriately. What is the potential role of clinical prediction rules in physiotherapy practice and what are the strategies by which clinicians can determine their appropriate use for a given clinical setting? CONCLUSION: Clinical prediction rules use quantitative methods to build upon the body of literature and expert opinion and can provide quick and inexpensive estimates of probability. Clinical prediction rules can be of great value to assist clinical decision making but should not be used indiscriminately. They are not a replacement for clinical judgment and should complement rather than supplant clinical opinion and intuition. The development of valid clinical prediction rules should be a goal of physiotherapy research. Specific areas in need of attention include deriving and validating clinical prediction rules to screen patients for potentially serious conditions for which current tests lack adequate diagnostic accuracy or have unacceptable cost and risk, and to assist in classification of patients for treatments that are likely to result in substantially different outcomes in heterogeneous groups of patients.

Decision Support Techniques↗