Nonlinear analysis of heart rate variability for evaluating the growing pig stress response to an acute heat episode.

Heart rate variability (HRV) is a proxy measure of autonomic function and can be used as an indicator of swine stress. While traditional linear measures are used to distinguish between stressed and unstressed treatments, inclusion of nonlinear HRV measures that evaluate data structure and organization shows promise for improving HRV interpretation. The objective of this study was to evaluate the inclusion of nonlinear HRV measures in response to an acute heat episode. Twenty 12- to 14-week-old growing pigs were individually housed for 7 days and acclimated to thermoneutral conditions (20.35°C ± 0.01°C; 67.6% ± 0.2% RH) before undergoing one of the two treatments: (1) thermoneutral control (TN; n = 10 pigs) or (2) acute heat stress (HS; n = 10 pigs; 32.6°C ± 0.1°C; 26.2% ± 0.1% RH). In Phase 1 of the experimental procedure (P1; 60 min), pigs underwent a baseline HRV measurement period in thermoneutral conditions before treatment [Phase 2; P2; 60 min once gastrointestinal temperature (Tg) reached 40.6°C], where HS pigs were exposed to heated conditions and TN pigs remained in thermoneutral conditions. After P2, all pigs were moved back to thermoneutral conditions (Phase 3; P3; 60 min). During each phase, Tg data were collected every 5 min and behavioural data were collected to evaluate the amount of time each pig spent in an active posture. Additionally, linear (time and frequency domain) and nonlinear [sample entropy (SampEn), de-trended fluctuation analysis, percentage recurrence, percentage determinism (%DET), mean diagonal line length in a recurrence plot] HRV measures were quantified. Heat stressed pigs exhibited greater Tg (P = 0.002) and spent less time in an active posture compared to TN pigs during P2 (P = 0.0003). Additionally, low frequency to high frequency ratio was greater in HS pigs during P3 compared to TN pigs (P = 0.02). SampEn was reduced in HS pigs during P2 (P = 0.01) and P3 (P = 0.03) compared to TN pigs. Heat stressed pigs exhibited greater %DET during P3 (P = 0.03) and tended to have greater %DET (P = 0.09) during P2 than TN pigs. No differences between treatments were detected for the remaining HRV measures. In conclusion, linear HRV measures were largely unchanged during P2. However, changes to SampEn and %DET suggest increased heat stress as a result of the acute heat episode. Future work should continue to evaluate the benefits of including nonlinear HRV measures in HRV analysis of swine heat stress.

Short communication: Assessment of disbudding pain in dairy calves using nonlinear measures of heart rate variability.

The purpose of this study was to evaluate whether pain-related stress caused by disbudding could be detected using nonlinear measures of heart rate variability (HRV). Twenty-five female Holstein calves (4-7 wk of age) were randomly assigned to 1 of 3 treatments: (1) sham disbud (SHAM; n = 9), (2) disbud with lidocaine-meloxicam pain mitigation (MED; n = 8), or (3) disbud without pain mitigation (NoMED; n = 8). Heart rate variability (sample entropy, percentage determinism, percentage recurrence, or mean length of diagonal lines in a recurrence plot) was recorded on d -1, 0, 1, 3, and 5 relative to the experimental procedure, with disbudding taking place on d 0. The short-term detrended fluctuation analysis scaling exponent was greater in MED calves than in SHAM calves, indicating a greater stress response to the disbudding procedure regardless of pain mitigation. These results indicate that calves in the MED group may have experienced pain-related stress as a result of the disbudding procedure. The remaining nonlinear HRV measures did not differ by treatment. Future research on this topic should address additional potential confounding factors, such as the effect of pain-mitigating drugs on autonomic function or the influence of the autonomic nervous system on wound healing, that may prohibit HRV measurement as an indicator of disbudding pain severity.

Analysis of lactation feed intakes for sows with extended lactation lengths.

The objectives of this research were to quantify and model daily feed intakes to 28 d of lactation in modern sows. A total of 4,512 daily feed intake (DFI) records were collected for 156 Hypor sows from February 2015 to March 2016. The mean lactation length was 27.9 ± 2.0 d. The data included 9 parity 1, 33 parity 2 and 114 parity 3+ sows. Data were collected using a computerized feeding system (Gestal Solo, JYGA Technologies, Quebec, Canada). The feeding system was used to set an upper limit to DFI for the first 7 d of lactation. Overall, the least-squares means of a model including the random effect of sow indicated that DFI's continued to slowly increase to 28 d of lactation. The DFI data were fitted to Generalized Michaelis-Menten (GMM) and polynomial functions of day of lactation (t). The GMM function [DFIi,t (kg/d) = DFI0 + (DFIA - DFI0)(t/K)C/[1 + (t/K)C]] was fitted with 2 random effects for DFI (dfiAi) and intercept (dfi0i) using the NLMIXED procedure in SAS®. The polynomial function DFIi,t (kg/d) = [B0 + B1 t + B2 t2 + B3 t3 + B4 t4] was fitted with three random effects for B0, B1, and B2 using the MIXED procedure in SAS®. Fixed effects models of the two functions had similar Akaike's Information Criteria (AIC) values and mean predicted DFI's. The polynomial function with 3 random effects provided a better fit to the data based on R2 30 (0.81 versus 0.79), AIC (14,709 versus 15,158) and RSD (1.204 versus 1.321) values than the GMM function with two random effects. The random effect for B2 in the polynomial function allowed for the fitting of the function to lactation records that had decreased DFI after 15 d of lactation. The random effects for the polynomial function were used to sort the lactation records into three groups based on the derivative of the function at 21 d of lactation. Lactation records of the three groups had similar DFI the first two weeks of lactation (P > 0.40). The three groups of sows had substantially different DFI's after 18 d of lactation (P < 0.028). The differences in both actual and predicted DFI's between the three groups increased with each day of lactation to day 28 (P < 0.001). Mixed model polynomial functions can be used to identify sows with different patterns of DFI after 15 d of lactation.

Evaluation of the impact of errors in the measurement of backfat depth on the prediction of fat-free lean mass.

The development of regression equations to predict carcass composition typically assumes that the independent variables, such as backfat depth, are measured without error. However, technological and operator-specific types of measurement errors do exist. To evaluate the impact of measurement error for backfat depth, Monte Carlo simulation was used to model carcass fat-free lean mass (FFLM) in pigs. In the simulation, FFLM was a linear function of carcass weight and actual backfat depth (ABFD). Carcass weight was assumed to be measured without error, but measurement errors were generated such that the correlation (r(BF)) of the measured backfat depth (BFD) and ABFD ranged from 0.70 to 0.95. Two types of measurement errors were simulated: 1) constant variation that was additive to the variance of ABFD, and 2) variation proportional to the ABFD that was additive to the variance in ABFD. A total of 1,000 replications of 1,000 pigs were simulated. Within each type of measurement error, the absolute values of the regression coefficients and R2 values of the equations decreased as r(BF) decreased. The probability of the backfat depth squared (BFD2) being significant (P < 0.05) in the regression equation was increased when the measurement errors were proportional to ABFD. The occurrence of a significant BFD2 variable was 792 times out of 1,000 replications when r(BF) = 0.95 and increased to 996 times out of 1,000 when r(BF) = 0.85 for BFD with type 2 measurement errors. The inclusion of a CW x BFD variable in the regression equations (P < 0.05) increased (270 to 423 times out of 1,000) as r(BF) decreased from 0.85 to 0.70 for BFD with type 2 errors. Equations developed from BFD with measurement errors resulted in biased predictions of FFLM and changes in FFLM per unit change in BFD. The level and type of measurement errors that exist in the independent variables should be evaluated.

Analysis of gene expression during the onset of muscle hypertrophy in callipyge lambs.

The callipyge mutation causes postnatal muscle hypertrophy in heterozygous lambs that inherit a paternal callipyge allele (+/CLPG). Our hypothesis was that the up-regulation of one or both of the affected paternally expressed genes (DLK1 or PEG11) initiates changes in biochemical and physiological pathways in skeletal muscle to induce hypertrophy. The goal of this study was to identify changes in gene expression during the onset of muscle hypertrophy to identify the pathways that are involved in the expression of the callipyge phenotype. Gene expression was analysed in longissimus dorsi total RNA from lambs at 10, 20, and 30 days of age using the Affymetrix Bovine Expression Array. An average of 40.6% of probe sets on the array was detected in sheep muscle. Data were normalized and analysed using a two-way anova for genotype and age effects with a false discovery rate of 0.10. From the anova, 13 genes were significant for the effect of genotype and 13 were significant for effect of age (P < 0.10). No significant age-by-genotype interactions were detected (P > 0.10). Of the 13 genes indicating an effect of genotype, quantitative PCR assays were developed for all of them and tested on a larger group of animals from 10 to 200 days of age. Nine genes had significantly elevated transcript levels in callipyge lambs. These genes included phosphofructokinase, a putative methyltransferase protein, a cAMP phosphodiesterase, and the transcription factor DNTTIP1.

Assessing the adaptation of swine to fifty-seven hours of feed deprivation in terms of behavioral and physiological responses.

Dramatic changes in the livestock industry have brought about widespread concern regarding the welfare of animals in terms of the hunger that they may experience. Despite this concern, animal science has not been able to provide a methodology that can objectively determine whether the welfare of animals is compromised by hunger. The current work sought to provide data that characterized the physiological and behavioral responses of animals experiencing 57 h of food deprivation. For this purpose, 2 separate experiments were conducted, in which physiological or behavioral measures were collected from swine deprived of feed for 21 to 57 h or fed normally (physiology, n = 20/treatment; behavior, n = 8/treatment). Treatment x time interactions were found (P < 0.02) for insulin, beta-hydroxybutyrate, NEFA, drinking, standing, inactivity, lying sternal, lying lateral, and total lying. Animals appeared to adjust appropriately to the metabolic challenge imposed, as suggested by increases in alternative energy substrates (NEFA, beta-hydroxybutyrate). Additionally, feed-deprived animals appeared less active than control animals until after 45 h of feed deprivation, when the former appeared to be more active. Our results suggest that feed deprivation of finishing pigs for durations greater than 45 h produced behavioral changes that may be related to increased sensations of hunger and which possibly are dependent on the use of alternative energetic substrates. Our data link behavioral and physiological changes after the imposition of feed deprivation.

Incremental cost-effectiveness ratio and incremental net-health benefit: two sides of the same coin.

In recent years, an alternative framework for cost-effectiveness analyses has been growing in popularity. Instead of the incremental cost-effectiveness ratio for which statistical inference is often difficult, the incremental net-health benefit (INHB), a linear transformation of incremental costs and effectiveness, has been utilized. The linear structure of this statistic allows easy computation and interpretation of confidence intervals, hypothesis tests and acceptability curves. It is often difficult, however, to switch decision-making procedures without first verifying the appropriateness of the new methods. In this paper, we demonstrate the decision-making similarities between the INHB and the incremental cost-effectiveness ratio and describe how the INHB can be used to clarify inference of the incremental cost-effectiveness ratio. We also describe the two statistics in terms of the DeltaE-DeltaC plane, thus allowing both a mathematical and graphical comparison of these similarities. We conclude with a general discussion of cost-effectiveness analyses and advocate Bayesian, rather than frequentist inference as the more intuitive and powerful decision-making framework.

Model selection for quantitative trait locus analysis in polyploids.

Over the years, substantial gains have been made in locating regions of agricultural genomes associated with characteristics, diseases, and agroeconomic traits. These gains have relied heavily on the ability to statistically estimate the association between DNA markers and regions of a genome (quantitative trait loci or QTL) related to a particular trait. The majority of these advances have focused on diploid species, even though many important agricultural crops are, in fact, polyploid. The purpose of our work is to initiate an algorithmic approach for model selection and QTL detection in polyploid species. This approach involves the construction of all possible chromosomal configurations (models) that may result in a gamete, model reduction based on estimation of marker dosage from progeny data, and lastly model selection. While simplified for initial explanation, our approach has demonstrated itself to be extendible to many breeding schemes and less restricted settings.

Modeling approach to diameter breakpoint determination.

In determining zone diameter breakpoints, the error-rate bounded method focuses directly on the observed discrepancy percentages (very major, major, and minor). These percentages, however, are quite variable due to the number of isolates investigated, the drug-specific relationship between MIC and zone diameter, the location of the isolates relative to the MIC intermediate zone, and the inherent variability of each test. To overcome potential sampling problems, a hierarchical model is proposed which explicitly accounts for each of these factors and probabilities from this model are used to determine diameter breakpoints. A simulation study is performed to demonstrate the improved consistency of this model-based procedure. Application to three published scatterplots demonstrate its interpretability advantages.

Cost-effectiveness of azithromycin for preventing Mycobacterium avium complex infection in HIV-positive patients in the era of highly active antiretroviral therapy. The Swiss HIV Cohort Study.

We conducted a cost-effectiveness analysis to determine the clinical and economic consequences of Mycobacterium avium complex (MAC) prophylaxis in HIV-infected patients in the era of highly active antiretroviral therapy (HAART) in a health care system with access unrestricted by financial barriers. The analysis was performed from a health care perspective and compared azithromycin (1200 mg/week) with no prophylaxis over a period of 10 years based on data from the Swiss HIV Cohort Study (SHCS) and randomized controlled trials. The main outcome measures were: expected survival; average health care costs; and cost-effectiveness in 1997 Swiss francs ( pound1 corresponds to about 2.3 CHF) per life-year saved. In patients with an initial CD4 count <50 cells/mm(3) and no AIDS, azithromycin increased expected survival by 4 months. In patients with AIDS, HAART durability had a major impact on expected survival and costs. Incremental survival increased from 2 to 4 months if we assumed a 10 year, instead of a 3 year, HAART effect. The cost-effectiveness of azithromycin relative to no prophylaxis in patients without AIDS was between 47,000 CHF (3-year HAART effect) and 60,000 CHF (10-year HAART effect) per life-year saved. The cost-effectiveness ratio increased to 118,000 CHF per life-year saved in patients with symptomatic AIDS. In conclusion, in the era of HAART, MAC prophylaxis with azithromycin increases expected survival and reduces health care costs substantially. Starting MAC prophylaxis in patients without AIDS is more effective and cost-effective than in patients with AIDS.

Systematic validation of disease models for pharmacoeconomic evaluations. Swiss HIV Cohort Study.

Pharmacoeconomic evaluations are often based on computer models which simulate the course of disease with and without medical interventions. The purpose of this study is to propose and illustrate a rigorous approach for validating such disease models. For illustrative purposes, we applied this approach to a computer-based model we developed to mimic the history of HIV-infected subjects at the greatest risk for Mycobacterium avium complex (MAC) infection in Switzerland. The drugs included as a prophylactic intervention against MAC infection were azithromycin and clarithromycin. We used a homogenous Markov chain to describe the progression of an HIV-infected patient through six MAC-free states, one MAC state, and death. Probability estimates were extracted from the Swiss HIV Cohort Study database (1993-95) and randomized controlled trials. The model was validated testing for (1) technical validity (2) predictive validity (3) face validity and (4) modelling process validity. Sensitivity analysis and independent model implementation in DATA (PPS) and self-written Fortran 90 code (BAC) assured technical validity. Agreement between modelled and observed MAC incidence confirmed predictive validity. Modelled MAC prophylaxis at different starting conditions affirmed face validity. Published articles by other authors supported modelling process validity. The proposed validation procedure is a useful approach to improve the validity of the model.

A Bayesian approach to modelling the natural history of a chronic condition from observations with intervention.

To assess the costs and benefits of screening and treatment strategies, it is important to know what would have happened had there been no intervention. In today's ethical climate, however, it is almost impossible to observe this directly and therefore must be inferred from observations with intervention. In this paper, we illustrate a Bayesian approach to this situation when the observations are at separated and unequally spaced time points and the time of intervention is interval censored. We develop a discrete-time Markov model which combines a non-homogeneous Markov chain, used to model the natural progression, with mechanisms that describe the possibility of both treatment intervention and death. We apply this approach to a subpopulation of the Wisconsin Epidemiologic Study of Diabetic Retinopathy, a population-based cohort study to investigate prevalence, incidence, and progression of diabetic retinopathy. In addition, posterior predictive distributions are discussed as a prognostic tool to assist researchers in evaluating costs and benefits of treatment protocols. While we focus this approach on diabetic retinopathy cohort data, we believe this methodology can have wide application.

Cost effectiveness of highly active antiretroviral therapy in HIV-infected patients. Swiss HIV Cohort Study.

BACKGROUND: Highly active antiretroviral therapy (HAART) has become the most important strategy for treating HIV infection in developed countries; however, access to HAART might vary under different funding policies. The Swiss health care system provides unrestricted access to HAART for all patients who need these newer combination therapies. This study investigated the impact of this funding policy on the society and health care system. METHODS: A cost-effectiveness analysis with natural history data and productivity estimates was based on the Swiss HIV Cohort Study. A random sample of patient charts was used to estimate health care costs. In addition to a base-case scenario, a pessimistic and an optimistic scenario of natural disease history was developed. Costs were expressed in 1997 Swiss francs (100 CHF correspond to about US$67) and effects as projected years of life gained. RESULTS: In the analysis limited to health care costs, on the basis of projected survival in each scenario, the cost-effectiveness ratio was 33,000 CHF (base case), 14,000 CHF (optimistic), and 45,000 CHF (pessimistic) per year of life gained. When changes in productivity were included, cost savings occurred in the base-case and optimistic scenarios. The cost-effectiveness ratio was 11,000 CHF per year of life gained in the pessimistic scenario. CONCLUSIONS: HAART increases expected survival and health care costs. However, when productivity gains are included, society will probably save costs or pay a low price for substantial health benefits. The study provides strong arguments, from a societal perspective, to continue the current policy of providing unrestricted access to HAART in Switzerland. The presented results also suggest that this policy could be of interest for other developed countries. Decision makers in developed countries where access to HAART is limited should re-evaluate their policy for the benefit of the society at large.

Estimating AIDS-free survival in a severely immunosuppressed asymptomatic HIV-infected population in the era of antiretroviral triple combination therapy. Swiss HIV Cohort Study.

BACKGROUND: Antiretroviral triple combination therapies have been evaluated in randomized controlled trials and cohort studies. Little is known about their impact on asymptomatic, severely immunosuppressed, HIV-infected individuals in a real world population. OBJECTIVES: To describe disease progression in a broad asymptomatic population of HIV-infected individuals with a CD4 count <100 cells/mm3 before and after the introduction of combination triple therapy. DESIGN: Six-month homogenous Markov chain consisting of four recurrent AIDS-free states and one absorbing AIDS state: (1) CD4 count > or =100 cells/mm3, (2) CD4 count 75 to 99 cells/mm3, (3) CD4 count 50 to 74 cells/mm3, (4) CD4 count 0 to 49 cells/mm3, and AIDS. SETTING: Swiss HIV Cohort Study database. PATIENTS: A total of 1027 patients contributed to 2634 pairs of 6-month observations from 1993 to 1995, and 681 patients contributed to 2077 pairs of 6-month observations from 1996 to 1997. MEASUREMENT: AIDS-free survival probabilities and the expected AIDS-free survival time. RESULTS: The expected number of AIDS-free months in a 3-year period was 17 (95% confidence interval [CI], 16-19) for patients starting in state 4 prior to 1996 versus 26 months (95% CI, 24-28) for patients starting in state 4 after 1996. For these two time periods, the corresponding expected numbers of AIDS-free months were 21 (95% CI, 20-22) versus 30 (95% CI, 28-32) for state 3 and 23 (95% CI, 21-24) versus 33 (95% CI, 32-34) for state 2. CONCLUSION: Expected 3-year AIDS-free survival in severely immunosuppressed individuals with CD4 counts <100 cells/mm3 improved significantly between 1993 to 1995 and 1996 to 1997.

Analysis of aerial survey data on Florida manatee using Markov chain Monte Carlo.

We assess population trends of the Atlantic coast population of Florida manatee, Trichechus manatus latirostris, by reanalyzing aerial survey data collected between 1982 and 1992. To do so, we develop an explicit biological model that accounts for the method by which the manatees are counted, the mammals' movement between surveys, and the behavior of the population total over time. Bayesian inference, enabled by Markov chain Monte Carlo, is used to combine the survey data with the biological model. We compute marginal posterior distributions for all model parameters and predictive distributions for future counts. Several conclusions, such as a decreasing population growth rate and low sighting probabilities, are consistent across different prior specifications.

Comparison of alternative diphasic lactation curve models under bovine somatotropin administration.

The multiphasic logistic function has recently been proposed as a model for lactation curves in dairy cattle. The applicability of several forms of this function was examined under three levels of recombinant bST administration beginning 100 d postpartum. Lactation curve models were fit to mean daily yields for each of the treatment groups and a control group. The traditional diphasic function was unable to model adequately the steep ascent and early peak in daily milk yield occurring immediately after calving in both treatment and control groups. The early portion of the lactation was more appropriately modeled using a diphasic function with a power transformation of time in the first phase, such that this phase became asymmetric. This modified diphasic function also resulted in smaller residuals and less autocorrelation than the triphasic function, which was used as a reference model, when both models were applied to the control group data. The increase in daily milk yield because of bST administration was modeled by the addition of a term consisting of a rising exponential function. Such a model may be useful for examining dosage effects in long-term bST studies in which injections begin in midlactation or later. Time series procedures reduced autocorrelation among residuals in both the control and treatment groups, and such models could be useful for removing autocorrelation in lactation curve applications that involve daily or weekly yield observations.