Agreement Between Methods of Measurement with Multiple Observations Per Individual

Department of Health Sciences, University of York, York, UK.
Journal of Biopharmaceutical Statistics (Impact Factor: 0.59). 02/2007; 17(4):571-82. DOI: 10.1080/10543400701329422
Source: PubMed


Limits of agreement provide a straightforward and intuitive approach to agreement between different methods for measuring the same quantity. When pairs of observations using the two methods are independent, i.e., on different subjects, the calculations are very simple and straightforward. Some authors collect repeated data, either as repeated pairs of measurements on the same subject, whose true value of the measured quantity may be changing, or more than one measurement by one or both methods of an unchanging underlying quantity. In this paper we describe methods for analysing such clustered observations, both when the underlying quantity is assumed to be changing and when it is not.

1 Follower
51 Reads
  • Source
    • "Statistical significance was assumed at values of P < 0.05. Agreement between measured and predicted _ Vo 2 values with multiple observations per participant was assessed graphically with Bland–Altman plots (Bland and Altman 2007). Variables are presented as means AE SD, and 95% Confidence Intervals (CI), unless otherwise stated. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Estimation of human oxygen uptake (_ Vo 2) during exercise is often used as an alternative when its direct measurement is not feasible. The American College of Sports Medicine (ACSM) suggests estimating human _ Vo 2 during exercise on a cycle ergometer through an equation that considers individual's body mass and external work rate, but not pedaling rate (PR). We hypothesized that including PR in the ACSM equation would improve its _ Vo 2 prediction accuracy. Ten healthy male participants' (age 19–48 years) were recruited and their steady-state _ Vo 2 was recorded on a cycle ergometer for 16 combinations of external work rates (0, 50, 100, and 150 W) and PR (50, 70, 90, and 110 revolutions per minute). _ Vo 2 was calculated by means of a new equation, and by the ACSM equation for comparison. Kinematic data were collected by means of an infrared 3-D motion analysis system in order to explore the mechanical determinants of _ Vo 2. Including PR in the ACSM equation improved the accuracy for prediction of sub-maximal _ Vo 2 during exercise (mean bias 1.9 vs. 3.3 mL O 2 kg À1 min À1) but it did not affect the accuracy for prediction of maximal _ Vo 2 (P > 0.05). Confirming the validity of this new equation, the results were replicated for data reported in the literature in 51 participants. We conclude that PR is an important determinant of human _ Vo 2 during cycling exercise, and it should be considered when predicting oxygen consumption.
    09/2015; 3(9):e12500. DOI:10.14814/phy2.12500
  • Source
    • "For the longitudinal cohort we performed modified Bland-Altman analysis to adjust for repeated measures, utilizing the change in various retinal layers between serial scans [32]. Similar to the cross-sectional cohort, we calculated mean differences and LOA with 95% CI and Bland-Altman plots. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Background. The lack of segmentation algorithms operative across optical coherence tomography (OCT) platforms hinders utility of retinal layer measures in MS trials. Objective. To determine cross-sectional and longitudinal agreement of retinal layer thicknesses derived from an open-source, fully-automated, segmentation algorithm, applied to two spectral-domain OCT devices. Methods. Cirrus HD-OCT and Spectralis OCT macular scans from 68 MS patients and 22 healthy controls were segmented. A longitudinal cohort comprising 51 subjects (mean follow-up: 1.4 ± 0.9 years) was also examined. Bland-Altman analyses and interscanner agreement indices were utilized to assess agreement between scanners. Results. Low mean differences (−2.16 to 0.26 μm) and narrow limits of agreement (LOA) were noted for ganglion cell and inner and outer nuclear layer thicknesses cross-sectionally. Longitudinally we found low mean differences (−0.195 to 0.21 μm) for changes in all layers, with wider LOA. Comparisons of rate of change in layer thicknesses over time revealed consistent results between the platforms. Conclusions. Retinal thickness measures for the majority of the retinal layers agree well cross-sectionally and longitudinally between the two scanners at the cohort level, with greater variability at the individual level. This open-source segmentation algorithm enables combining data from different OCT platforms, broadening utilization of OCT as an outcome measure in MS trials.
    06/2015; 2015:1-10. DOI:10.1155/2015/136295
  • Source
    • "Simulated values of pHa were compared against measured pHa from the corresponding ABG taken at each PS level. All comparisons were performed using Bland-Altman analysis for repeated measures [24], taking account for numerous PS levels for each patient. Bias and limits of agreement are reported as bias ± 2 SDs for each Bland-Altman analysis . "
    [Show abstract] [Hide abstract]
    ABSTRACT: Purpose: This article evaluates how mathematical models of gas exchange, blood acid-base status, chemical respiratory drive, and muscle function can describe the respiratory response of spontaneously breathing patients to different levels of pressure support. Methods: The models were evaluated with data from 12 patients ventilated in pressure support ventilation. Models were tuned with clinical data (arterial blood gas measurement, ventilation, and respiratory gas fractions of O2 and CO2) to describe each patient at the clinical level of pressure support. Patients were ventilated up to 5 different pressure support levels, for 15 minutes at each level to achieve steady-state conditions. Model-simulated values of respiratory frequency (fR), arterial pH (pHa), and end-tidal CO2 (FeCO2) were compared to measured values at each pressure support level. Results: Model simulations compared well to measured data with Bland-Altman bias and limits of agreement of fR of 0.7 ± 2.2 per minute, pHa of − 0.0007 ± 0.019, and FeCO2 of − 0.001 ± 0.003. Conclusion: The models describe patients' fR, pHa, and FeCO2 response to changes in pressure support with low bias and narrow limits of agreement.
    Journal of critical care 05/2015; DOI:10.1016/j.jcrc.2015.05.010 · 2.00 Impact Factor
Show more


51 Reads
Available from