Clinicians often need to know whether a new method of measurement is equivalent to an established one already in clinical use. This article reviews the methodology of a method-comparison study to assist the clinician with the conduct and evaluation of such studies. Temperature data from 1 subject are used to illustrate the procedures. Although one would not make decisions on the basis of the findings from 1 subject, the large number of paired measurements in the data set permits its use for illustrative purposes. Currently available software eliminates the need for tedious statistical computation but does not reduce the burden of understanding the concepts underlying a method-comparison study and accurate interpretation of the findings.
Investigators commonly assess intra- and inter-assay coefficients of variation (CVs) to estimate the precision of salivary cortisol enzyme immunoassay (EIA). However, little guidance is available as to which samples to use for CV assessment. The purposes of this methodological study were to compare differences in intra- and inter-assay CVs (a) among controls, standards, and/or unknown samples; and (b) between fresh and previously frozen saliva. A total of 174 duplicates (controls = 58, standards = 48, and unknowns = 68) were tested. The unknowns were from 34 students; all student saliva was assayed as both fresh and frozen samples. All samples were assayed in duplicate, using a commercial salivary cortisol EIA kit, by the same technician with the same equipment. A priori criteria for intra- and inter-assay CV, respectively, were ≤ 4% and ≤ 7%, and a was .05 for CV differences. Mean intra-assay CVs for controls, standards, unknowns, and combined samples were ≤ 2.5%, and mean inter-assay CVs were ≤ 2.8%. Mean intra-assay CVs were 2.2% for fresh saliva and 1.5% for frozen samples. Comparisons showed no significant differences in intra- or inter-assay CV among controls, standards, and/or unknown samples. Inter-assay CV was significantly different between fresh and previously frozen saliva (p = .043), with fresh saliva CV higher than frozen; the difference was not meaningful because all evaluations showed minimal measurement error. In conclusion, results indicate that estimation of precision can be achieved by testing of controls, standards, or unknowns and with either fresh or frozen saliva in this population.
The purpose of these experiments was to test the equivalence of pulmonary artery, urinary bladder, tympanic, rectal and femoral artery methods of temperature measurement in healthy and critically ill swine under clinical intensive care unit (ICU) conditions using a prospective, time series design. First, sensors were tested for error and sensitivity to change in temperature with a precision-controlled water bath and a laboratory-certified digital thermometer for temperatures 34-42°C. There was virtually no systematic (bias) or random (precision) error (<0.2°C). The bladder sensor had the slowest response time to change in temperature (105-120 s). Next, testing was done in an experimental porcine ICU in a non-profit research institution with four male, sedated, and mechanically ventilated domestic farm pigs. The in vivo experiments were conducted over periods of 41-168 h with temperatures measured every 1-5 s. The bladder, tympanic and rectal methods had unacceptable bias (≥0.5°C) and/or precision (≥0.2°C). Response time varied from 7 s with the femoral artery method to 280 s (4.7 min) with the tympanic method. We concluded that equivalence of the methods was insufficient for them to be used interchangeably in the porcine ICU. Intravascular monitoring of core body temperature produces optimal measurement of porcine temperature under varying conditions of physiological stability.
Experimental control and mathematical techniques increase confidence that results of circadian temperature rhythm studies reflect true changes in the circadian timing system versus coupling with exogenous synchronizers. Masking effects represent confounding influences in studies that are concerned with the endogenous temperature rhythm. Because it is technically difficult to measure directly the behavior of the endogenous timing system, marker rhythms are used as proxy measures. However, in addition to entraining, the external environment exerts a direct masking effect on the monitored rhythm. Methods for measuring circadian temperature rhythm are reviewed in this article. Constant routine, forced desynchrony, and purification methods represent attempts, at an experimental or mathematical level, to remove masking effects and more accurately capture the endogenous circadian temperature rhythm. Exogenous factors have not been subjected to the same scrutiny as the endogenous features of circadian temperature rhythm. But it is the environmental context, the extent to which the endogenous features are adaptively modified by the field environment, that will ultimately determine the biological value of circadian temperature rhythm to the organism. Thus, nurse investigators are encouraged to use rigorous methods to study both endogenous circadian temperature rhythm and exogenous rhythms.
This study examined the association between pressure injuries and complexity of abdominal temperature measured in residents of a nursing facility. The temperature served as a proxy measure for skin thermoregulation. Refined multiscale sample entropy and bubble entropy were used to measure the irregularity of the temperature time series measured over two days at 1-min intervals. Robust summary measures were derived for the multiscale entropies and used in predictive models for pressure injuries that were built with adaptive lasso regression and neural networks. Both types of entropies were lower in the group of participants with pressure injuries () relative to the group of non-injured participants (). This was generally true at the longer temporal scales, with the effect peaking at scale min for sample entropy and min for bubble entropy. Predictive models for pressure injury on the basis of refined multiscale sample entropy and bubble entropy yielded 96% accuracy, outperforming predictions based on any single measure of entropy. Combining entropy measures with a widely used risk assessment score led to the best prediction accuracy. Complexity of the abdominal temperature series could therefore serve as an indicator of risk of pressure injury.