Validity Of Predictive Equations Research Articles

PT10.3: Validity of Predictive Equations for Estimating Resting Energy Expenditure in Critically ILL Children: Results of a Systematic Review

PT10.3: Validity of Predictive Equations for Estimating Resting Energy Expenditure in Critically ILL Children: Results of a Systematic Review

Use of a Nonexercise Estimate for Prestroke Peak Vo 2 During the Acute Stroke Hospital Stay

For individuals with acute stroke, it is difficult to conduct an exercise test to assess peak oxygen consumption (peak Vo2). Therefore, the purpose of this study was to use a clinically feasible tool for assessing prestroke peak Vo2 using a nonexercise estimation equation to test whether estimated prestroke peak Vo2 was related to the functional outcome measures at discharge from the hospital in individuals after an acute stroke. We hypothesized that the estimated prestroke peak Vo2 would be significantly related to discharge Physical Performance Test (PPT), 6-minute walk test (6MWT), and lower extremity Fugl-Meyer (LEFM) assessment. Estimated prestroke peak Vo2 was calculated using a previously validated prediction equation using the following variables: body mass index, age, sex, resting heart rate, and a self-reported measure of physical activity. Outcome measures were assessed 4 days after enrollment or immediately before discharge (whichever occurred first). Thirty-four participants (mean age = 56.0, SD = 12.6 years; 20 men) with acute stroke were enrolled within 48 hours of admission. For all individuals, mean estimated prestroke peak Vo2 was 27.3 (SD = 7.4) mL·kg-1·min-1 and had a weak, nonsignificant relationship with the PPT (r = 0.19; P = .28), 6MWT (r = 0.10; P = .56), and LEFM (r = 0.32; P = .06). However, when considering sex, women, but not men, had a significant relationship with LEFM (r = 0.73; P = .005) and moderate but nonsignificant relationship with PPT (r = 0.53; P = .06) and 6MWT (r = 0.47; P = .10). Within 48 hours of stroke admission, we were able to administer a nonexercise equation to estimate prestroke peak Vo2. For the entire sample, functional measures conducted at discharge were not related to estimated prestroke peak Vo2. However, when considering sex, the relationship between prestroke Vo2 and the functional measures was strengthened.

Proposal of new body composition prediction equations from bioelectrical impedance for Indonesian men.

Bioelectrical impedance analysis (BIA) is one of the most affordable and feasible body composition assessment techniques for clinical and field settings. However, it is important to use an equation appropriate for the study population. This study aimed to propose and validate prediction equations to estimate body composition using BIA for Indonesian men. Total body water (TBW), fat-free mass (FFM) and fat mass (FM) were determined using the deuterium dilution technique in 292 Indonesian males. Participants were divided equally into development and validation groups to develop prediction equations and to cross-validate the proposed prediction equations, respectively. In addition, selected prediction equations using BIA were cross-validated. The proposed BIA equations were valid in our cross-validation samples. The best performance equations obtained from the absolute measure of body composition (TBW, FFM and FM) showed that r ranged between 0.89 and 0.91 and standard error of the estimate=1.8-2.6 kg. Cross-validation analysis indicated that the proposed equations had a bias of 0.1-0.3 kg, pure error of 1.3-1.8 kg and limits of agreement (mean difference±1.96 s.d.) of -0.26 to 0.13±4.09 to 5.59 kg. Among existing prediction equations examined, those by Deurenberg et al. (1989) and Lukaski et al. (1987) significantly overestimated FM by 4.0 and 3.2 kg, respectively, whereas the equation by Deurenberg et al. (1991) significantly (P<0.001) underestimated FFM by 5.0 kg compared with the reference FFM. The new BIA prediction equations may provide more precise and accurate estimation of body composition in Indonesian men than the existing equations.

Classification Agreement for FITNESSGRAM® Aerobic Capacity Between 1-Mile Run and the New PACER Formula

To assess aerobic capacity (AC), FITNESSGRAM® offers 1-mile run/walk (1MRW) and PACER tests. AC is estimated from 1MRW time, gender, age, and BMI from a validated prediction equation. Since 2011, PACER laps were converted to 1MRW times using a test-equating method, and AC was estimated from equated 1MRW performance. A new formula for estimating FITNESSGRAM® AC directly from PACER laps and age has been recently adopted. PURPOSE: The purpose of this study was to examine the classification agreement between AC estimated from 1MRW times and AC estimated from the new PACER formula. METHODS: Subjects were 440 sixth-grade boys and girls, ages 11-13 years, who completed each of the FITNESSGRAM® test components of their yearly physical education assessment. The percent meeting the age and gender AC standards estimated from 1MRW times were 55% for the Healthy Fitness Zone (HFZ), 19% for Some Risk (SR), and 26% in the High Risk (HR) category. The percent meeting the AC standards estimated from the new PACER formula were 32% for HFZ, 32% for SR, and 35% for HR. RESULTS: The correlation between AC estimated from 1MRW times and the new PACER formula was .64, compared to .93 between AC estimated from 1MRW times and the old PACER test-equating method. The observed agreement comparing AC estimated from 1MRW times with the new PACER formula for distinguishing HFZ from Needs Improvement (NI) was .68, Kappa = .40, and for distinguishing SR from HR was .71, Kappa = .50. The observed agreement comparing AC estimated from 1MRW times with the old PACER test-equating method for distinguishing HFZ from NI was .94, Kappa = .88, and for distinguishing SR from HR was .96, Kappa = .92. CONCLUSIONS: Classification agreement between AC estimated from 1MRW times and the new PACER formula was much lower than for the old PACER test-equating method. This result was due at least in part by the exclusion of BMI in the new PACER formula. The correlation between BMI and AC estimated from the new PACER formula was .47, compared to a correlation of .82 between BMI and AC estimated from 1MRW times. For AC estimated from the new PACER formula there is little variation based on body size, whereas body size is a significant predictor of AC estimated from 1MRW performance. Accurate assessment and classification are important for appropriate interventions, especially for those in the HR category.

Prediction equations for diffusing capacity (transfer factor) of lung for North Indians.

Background:Prediction equations for diffusing capacity of lung for carbon monoxide (DLCO), alveolar volume (VA), and DLCO/VA using the current standardization guidelines are not available for Indian population. The present study was carried out to develop equations for these parameters for North Indian adults and examine the ethnic diversity in predictions.Materials and Methods:DLCO was measured by single-breath technique and VA by single-breath helium dilution using standardized methodology in 357 (258 males, 99 females) normal nonsmoker adult North Indians and DLCO/VA was computed. The subjects were randomized into training and test datasets for development of prediction equations by multiple linear regressions and for validation, respectively.Results:For males, the following equations were developed: DLCO, −7.813 + 0.318 × ht −0.624 × age + 0.00552 × age2; VA, −8.152 + 0.087 × ht −0.019 × wt; DLCO/VA, 7.315 − 0.037 × age. For females, the equations were: DLCO, −44.15 + 0.449 × ht −0.099 × age; VA, −6.893 + 0.068 × ht. A statistically acceptable prediction equation was not obtained for DLCO/VA in females. It was therefore computed from predicted DLCO and predicted VA. All equations were internally valid. Predictions of DLCO by Indian equations were lower than most Caucasian predictions in both males and females and greater than the Chinese predictions for males.Conclusion:This study has developed validated prediction equations for DLCO, VA, and DLCO/VA in North Indians. Substantial ethnic diversity exists in predictions for DLCO and VA with Caucasian equations generally yielding higher values than the Indian or Chinese equations. However, DLCO/VA predicted by the Indian equations is slightly higher than that by other equations.

Validating prediction equations of metabolizable energy of soybean meal for growing pigs

The aim of this study was to evaluate equations to predict the metabolizable energy (ME) of soybean meal (SBM) for swine. Seven SBM were used, which were analyzed for dry matter, crude protein, ether extract, neutral detergent fiber (NDF), acid detergent fiber (ADF), ash, calcium, phosphorus, solubility in potassium hydroxide (KOH) and urease index. To determine the ME of SBM, 32 barrows, with an average initial weight of 29.01 ± 3.64 kg, were used and distributed in a randomized blocks design, with seven treatments and four replicates. To validate the prediction equations, linear regression models were adjusted, using observed values of ME (metabolism trial) as a function of the estimated ME (obtained by applying the chemical composition of the SBM in selected equations found in the literature). The existence of regression was evaluated by the “t” test, partially applied to each parameter (?0 and ?1). The validation of the prediction models of first degree was obtained by accepting the joint null hypothesis ?0 = 0 and ?1 = 1. The equations ME = 5.42 - 17.2FDN - 19.4MM + 0.709GE and ME= 1099 + 0.740GE - 5.5MM - 3.7NDF are effective for estimating the ME of SBM for growing pigs.

Validating prediction equations of metabolizable energy of soybean meal for growing pigs

&lt;p&gt;The aim of this study was to evaluate equations to predict the metabolizable energy (ME) of soybean meal (SBM) for swine. Seven SBM were used, which were analyzed for dry matter, crude protein, ether extract, neutral detergent fiber (NDF), acid detergent fiber (ADF), ash, calcium, phosphorus, solubility in potassium hydroxide (KOH) and urease index. To determine the ME of SBM, 32 barrows, with an average initial weight of 29.01 ± 3.64 kg, were used and distributed in a randomized blocks design, with seven treatments and four replicates. To validate the prediction equations, linear regression models were adjusted, using observed values of ME (metabolism trial) as a function of the estimated ME (obtained by applying the chemical composition of the SBM in selected equations found in the literature). The existence of regression was evaluated by the “t” test, partially applied to each parameter (?0 and ?1). The validation of the prediction models of first degree was obtained by accepting the joint null hypothesis ?0 = 0 and ?1 = 1. The equations ME = 5.42 - 17.2FDN - 19.4MM + 0.709GE and ME= 1099 + 0.740GE - 5.5MM - 3.7NDF are effective for estimating the ME of SBM for growing pigs.&lt;/p&gt;

Development and ex post validation of prediction equations of corn energy values for growing pigs

The aim of this study was to determine and validate prediction equations for digestible (DE) and metabolizable energy (ME) of corn for growing pigs. The prediction equations were developed based on data on the chemical composition, digestible and metabolizable energy of corn grain (30 samples) evaluated in experiments in Embrapa Suinos e Aves, Brazil. The equations were evaluated using regression analysis, and adjusted R² was the criterion for selection of the best models. Two equations were tested for DE and ME, each. To validate the equations, 1 experiment with 2 assays was performed to determine the values of DE and ME of 5 corn cultivars. In each assay, we used 24 growing pigs with initial average weight of 54.21 ± 1.68 kg in complete randomized block design with 6 treatments and 4 replicates. Treatments consisted of a reference diet and 5 ration tests composed of 60% of the reference diet and 40% of corn (1 of the 5 cultivars). Based on the results of the metabolic experiment and predicted values obtained in the equations, the validation of the equations was conducted using the lowest prediction error (pe) as a criterion for selection. The equations that produced the most accurate estimates of DE and ME of corn were as follows: DE = 11812 – 1015.9CP – 837.9EE – 1641ADF + 2616.3Ash + 47.5(CP2) + 114.7(CF2) + 46(ADF2) – 1.6(NDF2) – 997.1(Ash2) + 151.9EECF + 23.2EENDF – 126.4CPCF + 136.4CPADF – 4.0CPNDF, with R2 = 0.81 and pe = 2.33; ME = 12574 – 1254.9CP – 1140.5EE – 1359.9ADF + 2816.3Ash + 77.6(CP2) + 92.3(CF2) + 54.1(ADF2) – 1.8(NDF2) – 1097.2(Ash2) + 240.6EECF + 26.3EENDF – 157.4CPCF + 96.5CPADF – 4.4CPNDF, with R2 = 0.89 and pe = 2.24. Thus, using the data on chemical composition, it is possible to derive prediction equations for DE and ME of corn for pigs; these equations seem to be valid because of the small prediction errors suggestive of high accuracy of these models.

A novel risk score to predict cardiovascular disease risk in national populations (Globorisk): a pooled analysis of prospective cohorts and health examination surveys

Treatment of cardiovascular risk factors based on disease risk depends on valid risk prediction equations. We aimed to develop, and apply in example countries, a risk prediction equation for cardiovascular disease (consisting here of coronary heart disease and stroke) that can be recalibrated and updated for application in different countries with routinely available information. We used data from eight prospective cohort studies to estimate coefficients of the risk equation with proportional hazard regressions. The risk prediction equation included smoking, blood pressure, diabetes, and total cholesterol, and allowed the effects of sex and age on cardiovascular disease to vary between cohorts or countries. We developed risk equations for fatal cardiovascular disease and for fatal plus non-fatal cardiovascular disease. We validated the risk equations internally and also using data from three cohorts that were not used to create the equations. We then used the risk prediction equation and data from recent (2006 or later) national health surveys to estimate the proportion of the population at different levels of cardiovascular disease risk in 11 countries from different world regions (China, Czech Republic, Denmark, England, Iran, Japan, Malawi, Mexico, South Korea, Spain, and USA). The risk score discriminated well in internal and external validations, with C statistics generally 70% or more. At any age and risk factor level, the estimated 10 year fatal cardiovascular disease risk varied substantially between countries. The prevalence of people at high risk of fatal cardiovascular disease was lowest in South Korea, Spain, and Denmark, where only 5-10% of men and women had more than a 10% risk, and 62-77% of men and 79-82% of women had less than a 3% risk. Conversely, the proportion of people at high risk of fatal cardiovascular disease was largest in China and Mexico. In China, 33% of men and 28% of women had a 10-year risk of fatal cardiovascular disease of 10% or more, whereas in Mexico, the prevalence of this high risk was 16% for men and 11% for women. The prevalence of less than a 3% risk was 37% for men and 42% for women in China, and 55% for men and 69% for women in Mexico. We developed a cardiovascular disease risk equation that can be recalibrated for application in different countries with routinely available information. The estimated percentage of people at high risk of fatal cardiovascular disease was higher in low-income and middle-income countries than in high-income countries. US National Institutes of Health, UK Medical Research Council, Wellcome Trust.

New anthropometry-based age- and sex-specific reference values for urinary 24-hour creatinine excretion based on the adult Swiss population.

BackgroundUrinary creatinine excretion is used as a marker of completeness of timed urine collections, which are a keystone of several metabolic evaluations in clinical investigations and epidemiological surveys.The current reference values for 24-hour urinary creatinine excretion rely on observations performed in the 1960s and 1970s in relatively small and mostly selected groups, and may thus poorly fit to the present-day general European population.The aim of this study was to establish and validate anthropometry-based age- and sex-specific reference values of the 24-hour urinary creatinine excretion on adult populations with preserved renal function.MethodsWe used data from two independent Swiss cross-sectional population-based studies with standardised 24-hour urinary collection and measured anthropometric variables. Only data from adults of European descent, with estimated glomerular filtration rate (eGFR) ≥60 ml/min/1.73 m2 and reported completeness of the urinary collection were retained. A linear regression model was developed to predict centiles of the 24-hour urinary creatinine excretion in 1,137 participants from the Swiss Survey on Salt and validated in 994 participants from the Swiss Kidney Project on Genes in Hypertension.ResultsThe mean urinary creatinine excretion was 193 ± 41 μmol/kg/24 hours in men and 151 ± 38 μmol/kg/24 hours in women in the Swiss Survey on Salt. The values were inversely correlated with age and body mass index (BMI).Based on current reference values (177 to 221 μmol/kg/24 hours in men and 133 to 177 μmol/kg/24 hours in women), 56% of the urinary collections in the whole population and 67% in people >60 years old would have been considered as inaccurate.A linear regression model with sex, BMI and age as predictor variables was found to provide the best prediction of the observed values and showed a good fit when applied to the validation population.ConclusionsWe propose a validated prediction equation for 24-hour urinary creatinine excretion in the general European population, based on readily available variables such as age, sex and BMI, and a few derived normograms to ease its clinical application. This should help healthcare providers to interpret the completeness of a 24-hour urine collection in daily clinical practice and in epidemiological population studies.

Body composition by dual-energy X-ray spectrometry and bioelectrical impedance spectroscopy in a healthy population at age 75 and 80

Bioelectrical impedance spectroscopy (BIS) is a versatile field tool to obtain information about body composition (BC), if prediction equations are available that are valid for the group under consideration. We aimed to validate prediction equations for total body skeletal muscle mass (TBSMM) developed in healthy 75-year olds from the Gothenburg area (Sweden) in a similar sample examined at age 80. We give new prediction equations for fat mass (FM), fat-free mass (FFM), and TBSMM that are based on the entire group of elderly subjects, using values from Dual-energy X-ray spectrometry (DXA) as reference standard. Finally, we describe the longitudinal change in BC in the subset of subjects, who participated at both age 75 and 80. Body composition was measured by DXA and multiple-frequency BIS at age 75 (n=111) and 80 (n=91) years. Ordinary and mixed linear regression was used to examine previous and new prediction equations. Longitudinal change was assessed by paired sample t-test. The prediction equations for TBSMM developed in 75-year old subjects showed good validity when applied to the sample of 80 year olds. The results were independent of the presence of a metal prosthesis in hip or knee. New, sex-specific prediction equations for FM, FFM, and TBSMM were derived for 75 - 80 year-old subjects that further improved the accuracy of prediction. Using DXA-derived BC at ages 75 and 80, we observed a decline in TBSMM in both sexes, which was mainly due to muscle loss in the legs. In men, the decrease in trunk FFM was accompanied by an increase in FM, leaving no overall weight change in the trunk. When the 5-year change in TBSMM was based on longitudinal BIS measurements, the results from DXA were reproduced only when the prediction was based on BIS-values at 50kHz. Good cross-sectional validity of BIS equations for TBSMM was found in a population sample of 80-year olds. The presence of a metal prosthesis did not influence the quality of prediction. New prediction equations for FM, FFM, and TBSMM need to be evaluated in an independent sample.

Validation of Prediction Equations for VO2max Using Bruce Protocol

Validation of Prediction Equations for VO2max Using Bruce Protocol

Serum osmolarity and haematocrit do not modify the association between the impedance index (Ht2/Z) and total body water in the very old: The Newcastle 85+ Study

Purpose of the researchBioelectrical impedance is a non-invasive technique for the assessment of body composition; however, information on its accuracy in the very old (80+years) is limited. We investigated whether the association between the impedance index and total body water (TBW) was modified by hydration status as assessed by haematocrit and serum osmolarity. Materials and methodsThis was a cross-sectional analysis of baseline data from the Newcastle 85+Cohort Study. Anthropometric measurements [weight, height (Ht)] were taken and body mass index (BMI) calculated. Leg-to-leg bioimpedance was used to measure the impedance value (Z) and to estimate fat mass, fat free mass and TBW. The impedance index (Ht2/Z) was calculated. Blood haematocrit, haemoglobin, glucose, sodium, potassium, urea and creatinine concentrations were measured. Serum osmolarity was calculated using a validated prediction equation. Principal results677 men and women aged 85 years were included. The average BMI of the population was 24.3±4.2kg/m2 and the prevalence of overweight and obesity was 32.6% and 9.5%, respectively. The impedance index was significantly associated with TBW in both men (n=274, r=0.76, p<0.001) and women (n=403, r=0.96, p<0.001); in regression models, the impedance index remained associated with TBW after adjustment for height, weight and gender, and further adjustment for serum osmolarity and haematocrit. The impedance index values increased with BMI and the relationship was not modified by hydration status in women (p=0.69) and only marginally in men (p=0.02). Major conclusionsThe association between the impedance index and TBW was not modified by hydration status, which may support the utilisation of leg-to-leg bioimpedance for the assessment of body composition in the very old.

Bioimpedance in 7-Year-Old Children: Validation by Dual X-Ray Absorptiometry - Part 1: Assessment of Whole Body Composition

Aim: To validate the bioimpedance analyzer (BIA) Tanita BC-418 for its clinical and epidemiological use in children compared to dual-energy X-ray absorptiometry (DXA). Methods: A cross-sectional validation study was performed in 7-year-old children using anthropometry, BIA and DXA. Whole body fat and lean masses were assessed through BIA (BIA_outputs) and DXA. Fat mass index (FMI) was calculated. Predictive equations were derived from raw impedance and anthropometric measures; results obtained from these predictive equations (BIA_regressions) were also compared to DXA. Results: 171 children (84 boys) were studied. BIA_outputs and DXA results revealed small differences for lean mass (1%) and moderate differences for fat mass (13%). BIA_regressions results showed small differences for both body lean and fat masses (0.21 and 4.62%, respectively). Sensitivity and specificity to correctly classify children >90.8th percentile of FMI was 84.6 (64.3-94.9) and 95.9% (90.8-98.3) for BIA_outputs and 100 (98.1-100.0) and 95.9% (92.3-99.4) for BIA_regressions, respectively. Conclusions: Tanita BC-418 may be valid for epidemiological studies assessing whole body composition. Its measurements may help in the diagnosis and monitoring of childhood overweight and obesity. The validation of predictive equations in specific populations may increase the precision of the technique.

Correlations of skin fold thickness and validation of prediction equations using DEXA as the gold standard for estimation of body fat composition in Pakistani children

ObjectiveTo determine the correlation between dual energy X-ray absorptiometry (DEXA) and skin fold thickness (SKF) equations for estimation of body fat (BF) composition in secondary school children and validation of...

Abstract MP16: The Validity of Predictive Equations for 24-Hour Urine Sodium Excretion Among Older Adults and Those with Hypertension: The MESA and CARDIA Urinary Sodium Study

Background: Excess dietary sodium (Na) intake is linked to adverse cardiovascular health; the population distribution of urinary sodium (UNa) values in the US is unknown. We examined the population distribution of UNa and the validity of existing equations predicting 24-h Na excretion from a single spot urine sample among older adults with and without hypertension. Methods: Demographic, anthropometric, lab, and diet data along with 24-h urine collections were obtained from 555 MESA and CARDIA participants aged 45-79y. One third provided a second 24-h urine collection. Four timed voids (morning, afternoon, evening, and overnight) and the 24-h collection were analyzed for Na, creatinine, potassium and chloride concentrations. We examined the distribution of 24-h excretion of each analyte overall and by gender-race subgroups and hypertensive (HTN) status. We then examined the mean differences (bias) and confidence intervals between measured 24-h UNa excretion and the predicted from spot urine using 4 published equations by specimen timing, race-gender subgroups, and HTN status. Results: Using preliminary data from 265 participants with completed laboratory analysis, 55% female, 61% Black, and 60% had HTN. Mean 24-h Na excretion was 4234 ± 1920 mg for white men, 2706 ± 1136 mg for white women, 3463 ± 1691 mg for Black men and 3415 ± 1635 mg for Black women and did not significantly differ by hypertensive status. Mean bias in predicting 24-h Na excretion overall ranged from -268 (95% CI: -443.5, -91.8) to 1045 (849.3, 1240.4) mg/d. (Table) Conclusion: Among this group of older adults and those with hypertension, the mean 24-h UNa excretion levels for all race-gender groups exceeded current recommendations of 2,300 mg/d, with significant variation by race and gender. All of the four published equations under or overestimated mean 24-h Na excretion when using a single, timed spot urine sample but using evening samples appeared to produce the least bias. These preliminary findings are tentative until confirmed with the full dataset.

Lean body mass: the development and validation of prediction equations in healthy adults

BackgroundThere is a loss of lean body mass (LBM) with increasing age. A low LBM has been associated with increased adverse effects from prescribed medications such as chemotherapy. Accurate assessment of LBM may allow for more accurate drug prescribing. The aims of this study were to develop new prediction equations (PEs) for LBM with anthropometric and biochemical variables from a development cohort and then validate the best performing PEs in validation cohorts.MethodsPEs were developed in a cohort of 188 healthy subjects and then validated in a convenience cohort of 52 healthy subjects. The best performing anthropometric PE was then compared to published anthropometric PEs in an older (age ≥ 50 years) cohort of 2287 people. Best subset regression analysis was used to derive PEs. Correlation, Bland-Altman and Sheiner & Beal methods were used to validate and compare the PEs against dual X-ray absorptiometry (DXA)-derived LBM.ResultsThe PE which included biochemistry variables performed only marginally better than the anthropometric PE. The anthropometric PE on average over-estimated LBM by 0.74 kg in the combined cohort. Across gender (male vs. female), body mass index (< 22, 22- < 27, 27- < 30 and ≥30 kg/m2) and age groups (50–64, 65–79 and ≥80 years), the maximum mean over-estimation of the anthropometric PE was 1.36 kg.ConclusionsA new anthropometric PE has been developed that offers an alternative for clinicians when access to DXA is limited. Further research is required to determine the clinical utility and if it will improve the safety of medication use.

Screening for impaired fasting glucose and diabetes using available health plan data

AimsTo develop and validate prediction equations to identify individuals at high risk for type 2 diabetes using existing health plan data. MethodsHealth plan data from 2005 to 2009 from 18,527 members of a Midwestern HMO without diabetes, 6% of whom had fasting plasma glucose (FPG) ≥110mg/dL, and health plan data from 2005 to 2006 from 368,025 members of a West Coast-integrated delivery system without diabetes, 13% of whom had FPG ≥110mg/dL were analyzed. Within each health plan, we used multiple logistic regression to develop equations to predict FPG ≥110mg/dL for half of the population and validated the equations using the other half. We then externally validated the equations in the other health plan. ResultsAreas under the curve for the most parsimonious equations were 0.665 to 0.729 when validated internally. Positive predictive values were 14% to 32% when validated internally and 14% to 29% when validated externally. ConclusionMultivariate logistic regression equations can be applied to existing health plan data to efficiently identify persons at higher risk for dysglycemia who might benefit from definitive diagnostic testing and interventions to prevent or treat diabetes.

Abstract MP78: Validity of Predictive Equations for 24-Hour Urine Sodium Excretion in Young Black and Other Adults

Introduction: Reducing mean population sodium intake by ~1200 mg is projected to reduce thousands of deaths from heart disease and stroke and save billions of health care dollars annually. Twenty-four hour urine collection is recommended for assessing changes in mean population sodium intake, but can be difficult to implement. Predicting 24-hour urine sodium excretion using spot urines is not recommended due to diurnal variations in excretion. Further, sodium excretion patterns differ between black and white persons. We assessed the validity of previously published prediction equations for 24-hour sodium excretion in black and other young adults by timing of spot urine collection. Design: Of 481 adult volunteers aged 18-39 years (50% Blacks) asked to participate in a 2011 study in the Metropolitan DC area, 407 collected each urine void in a separate container for 24 hours. Four timed voids from the 24-h urine collection were selected (morning, afternoon, evening, and overnight) to use with previously published predictive equations. Predictive equations were based on one of two approaches; 1) an indirect approach using spot urine sodium-to-creatinine concentrations and predicted 24-hour creatinine excretion ( Tanaka, Kawasaki, Mage ), and 2) a direct approach using spot urine sodium, potassium, and creatinine concentrations, and age, and body mass index with separate equations by sex ( Brown ). We assessed mean differences between predicted and measured 24-hour sodium excretion (bias) and individual differences across levels of sodium excretion using Bland-Altman plots. Results: Among participants, mean measured 24-hour sodium excretion was ~3300 mg (SD ~1400 mg). Of the equations evaluated, mean bias in predicted 24-hour sodium excretion was least from Brown equations when using morning (-165 mg, 95% confidence interval [CI], -295, -36 mg), afternoon (-90 mg, 95% CI, -208, 28 mg) or evening ( -120 mg, 95% CI -230, -11 mg) spot urines. When using overnight spot urines, mean bias from Brown equations was greatest and statistically significant (-247 mg, 95% CI, -348, -151 mg). When using overnight spot urines, mean bias from Tanaka (-23 mg) or Mage (-145 mg) equations was not significant, however, when stratified by sex, mean biases were significant and in opposite directions. Among Blacks, mean biases from Brown were not significant (-167 to 122 mg) except using overnight specimens among Black females (-267 mg, 95% CI, -525, -47 mg). Across equations and time periods, Bland-Altman plots indicated significant bias at the individual level. Conclusions: Of the evaluated equations, predicted 24-hour urine sodium excretion using the Brown equations with morning, afternoon, or evening specimens may provide the least biased estimates of group mean sodium intake among young US adults. None of the equations adequately predicted individual 24-hour sodium excretion measured on the same day.

Heat stress risk alert system for dairy cattle in Slovenia

For the estimation of heat stress risk and modelling of prediction equations, temperature-humidity conditions were recorded on three dairy farms with different ventilation systems: natural ventilation (NV), fan ventilation during daytime (FVD) and fan ventilation during day and nighttime (FVDN). First data set was collected from April 1, 2010 to September 30, 2010 and the second from April 1, 2011 to September 30, 2011. On each farm hourly data of outdoor and barn air temperature and humidity were recorded. Results showed that in all barns the time of exposure to heat stress risk was longer compared to outdoor conditions. However, the maximum exposure to the heat stress risk conditions was reached in July 2010 and August 2011 when it was 19.1 and 18.5 hours for NV facility, 17.9 and 16.9 hours for FDV facility and 13.8 and 13.3 hours for FVDN system. Barn temperature-humidity index (THI) was estimated on the first data set. The model included the linear effect of ambient THI, month and hour of measurement. Validation of prediction equations was performed on the second data set, showing a correlation between true and estimated THI of 83.7 for NV, 82.7 for FVD and 89.4 for FVDN facility. Finally, system for prediction of dairy cow heat stress risk based on the most recent data from the nearest automatic meteorological station was incorporated into web page Govedo (www.govedo.si) to alert farmers on possible risk of heat stress.