The importance of quality control in both general and respiratory medicine has increased in parallel with the complexity of healthcare provision. Only a few decades ago, the respiratory physician and/or scientist had a very limited number of diagnostic and therapeutic tools available and, moreover, medical practice was based almost exclusively on the personal interaction between doctor and patient. Consequently, at that time the quality of the respiratory healthcare depended entirely on the professional competence of the doctor. Although nowadays the relationship between physician and patient undoubtedly still lies at the heart of respiratory medical practice, the quality of the medical service received by the patient also depends on many other participants in a complex healthcare network: various medical specialists, lung function technicians, nurses, respiratory therapists, social workers and administrative staff. Accordingly, several quality control programmes are applied in order to avoid, or at least to reduce, errors in diagnosis, improper performance of procedures, errors in medication, and failure to supervise or monitor care or recognise complications associated with treatment 1.
An adequate quality control scheme seeks to cover the various levels involved in a healthcare system, all of them with a potential influence on clinical and socioeconomic outcomes. The attending physician is at the centre of the system and, therefore, is perhaps the main element that needs to be assessed in quality control procedures 2. The results achieved by a given respiratory physician and/or scientist depend, however, on the performance of several professional collaborators in the hospital, each with their own quality assessment programme: specimen laboratories 3, 4, screening 5 and surgical 6, 7 procedures, and intensive care units 8, 9. It is interesting to note that, in the light of the increasing complexity of the procedures in use, some quality assessment techniques previously developed for industrial environments have recently been applied to the monitoring of healthcare processes 10, 11. Home mechanical ventilation and long-term oxygen therapy, although not always provided by the prescribing hospital 12, are important respiratory healthcare services with specific quality control protocols 13, 14. It is noticeable that quality control plays an important role, not only in optimising the healthcare provided to the population via services directly applied to each individual patient, but also in more general and indirect procedures. This is the case, for example, in clinical trials aimed at implementing guidelines or at investigating pharmaceutical drugs 15, 16, in the legal and administrative settings for publishing alerts on adverse incidents 12, in publishing research 17 or in issues concerning the use of diagnostic and therapeutic devices 18, 19.
Assessing the safety and effectiveness of medical devices is particularly relevant in respiratory medicine 20. One reason for this is the fast advance of technology 21, resulting in the marketing of considerably complex and intelligent devices to apply ventilatory support to patients with acute/chronic respiratory failure or sleep apnoea 22–24. Another reason is the central role that devices measuring lung function play in defining and classifying the degree of severity of respiratory diseases. In this regard, both the American Thoracic Society and the European Respiratory Society have published joint standardisation rules on quality control of the calibration, processing and interpretation of most lung function devices and tests 25–27. Moreover, given the central role that spirometry plays in the routine assessment of lung function, quality control studies have been carried out to investigate whether spirometry can be reliably performed in particularly difficult contexts, such as paediatrics 28, primary care centres 29, multicentre trials 30 and procedures using telemedicine tools 31.
In the present issue of the European Respiratory Journal, Jensen et al. 32 report the application of a quality control programme to the measurement of the single-breath diffusing capacity of the lung for carbon monoxide (DL,CO) in a multicentre clinical trial involving 125 lung function laboratories from most parts of the world. Even though specific standardisation rules for DL,CO measurements have been issued 27, the routine application of this technique exhibits remarkable variability, which makes it difficult to compare data obtained from different laboratories in clinical trials. One source of variability that can be reduced, but not eliminated, comes from the practical implementation of the measurements in each laboratory and patients' biological variability 33. However, DL,CO tests, like any other lung function measurement, are also affected by instrumental variability, which can and must be considerably reduced, and even eliminated, by quality assurance processes based on patient simulators. In fact, Jensen et al. 32 show that at the beginning of the trial as many as 25% of lung function laboratories did not pass the quality control tests on DL,CO equipment (a figure that was reduced to 1% after the initial control). Interestingly, the performance of the DL,CO equipment after the initial certification remained constant throughout the quality assurance process undertaken during the clinical trial 32.
The kind of results obtained by Jensen et al. 32 may seem trivial and predictable from a medical or scientific viewpoint that focuses exclusively on the clinical trial in which the equipment is used, and assuming that there are no methodological anomalies. However, as the proverb says, “the devil is in the details”. In any case, it seems clear that attention to quality control in all the specific tasks involved in respiratory healthcare will ultimately result in improvements in patient care.
Support statement
R. Farré and D. Navajas are supported by Ministerio de Ciencia e Innovación (SAF2008-02991) and Ministerio de Sanidad y Consumo (FIS2008-PI081908), respectively.
Statement of interest
None declared.
- Received January 2, 2009.
- Accepted January 7, 2009.
- © ERS Journals Ltd