Common use of PATIENTS AND METHODS Clause in Contracts

PATIENTS AND METHODS. The PiKo-1 (Ferraris Cardiorespiratory, Louisville, CO, USA) is a lightweight, small, inexpensive electronic sensing device that can measure PEF and FEV1. The device can store 96 readings and report errors in the procedure. The reading itself is displayed along with a corresponding color zone, which can be adjusted according to the reference values. Data stored in the device can also be transferred to a computer or transmitted to other units1 (Figure 1). The Masterlab pneumotachograph (▇▇▇▇▇▇ ▇▇, Würzburg, Germany) is a device that is widely used in clinical practice for measuring spirometric variables. However, the size of the device is a hindrance, and personnel with specific training are required. For these reasons, other devices, such as the PiKo-1 are under development. The aim is to make these devices easier to operate so that, once their readings have been shown to agree with those of the pneumotachograph, physicians can use them in the clinical practice. In clinical research, the reliability of a device is usually assessed by comparing the results with those of another one widely used in clinical practice for agreement or discrepancies. If a more practical alternative to the reference device becomes available, the agreement between systems should be determined. When a variable in a comparative analysis is continuous and quantitative, the intraclass correlation coefficient (ICC) is more appropriate than the ▇▇▇▇▇▇▇ correlation coefficient (r), as it can indicate general agreement between 2 or more methods of measurement or different observations.8 Another simple and visual method is the so-called ▇▇▇▇▇-▇▇▇▇▇▇ analysis to assess agreement between 2 systems of measurement.9 We have undertaken a randomized, single-blind, cross- sectional study of agreement between 2 measurement devices in a specialist care setting—the lung function testing laboratory of the Hospital General Yagüe in Burgos, Spain, at 867 m above sea level. Patients were recruited from those attending our specialist laboratory between March 15 and April 24, 2004 for lung function testing. Patients aged between 20 years and 80 years were included, and those who did not understand the technique after a brief explanation were excluded in order to avoid procedural errors. A table of random numbers was used to select patients from among those who attended the lung function testing laboratory and who met the inclusion criteria. A sample size of 40 patients was calculated to be sufficient to detect mean differences between the 2 measurement devices.