In this observational study we found that the proportion of patients with controlled asthma was significantly higher in France (41%) than in Spain (30%). In both countries, costs were higher and HRQL lower as level of asthma control decreased. The average asthma-related total health care costs over a three-month period were €85.4, €314.4 and €537.9 in France and €152.6, €241.2 and €556.8 in Spain for patients with controlled, partially controlled and uncontrolled asthma, respectively. The HRQL scores (EQ-5D-3L®
) were 0.88, 0.78 and 0.63 in France (p<0.0001) and 0.89, 0.82 and 0.69 (p<0.0001) in Spain, for patients with controlled, partially controlled and uncontrolled asthma, respectively. According to the GINA 2009 Guidelines
], the goal of asthma treatment is to achieve and maintain asthma control. The level of control is assessed based on symptoms, the use of reliever treatments, the adaptation of daily life, and measurement of peak expiratory flow or FEV1. However, this assessment of the level of asthma control may be conducted in different ways and the level of control can also vary over time.
The GINA 2009 criteria did not specify precisely the period over which asthma control should be assessed. Instead, any of the criteria observed over a given week may affect the level of asthma control for the whole period. Recently, for the first time, GINA 2010
] guidelines provided a time frame for the assessment of asthma control and recommended that asthma control must be assessed “preferably over 4 weeks”. In our study, asthma control was assessed using symptoms data on a three-month period but FEV1 was measured only at the end of that period of time and this may be considered as a limitation of our study.
Previous studies estimated that approximately 40% of patients in France
] have uncontrolled asthma. Similar figures were observed in Spain
]. These estimates are higher than those in our study in France (uncontrolled asthma=21.4%) but consistent with our results in Spain (uncontrolled asthma=36%). However, the differences between our results and previous ones may be the consequence of slightly different definitions of asthma control. The reasons for the differences in asthma control between the two neighbouring countries are not known. Possible explanations include differences in patient compliance with asthma treatments, as well as environmental and genetic factors. In addition, there may be measurement issues as physicians may vary in their assessment of GINA’s criteria.
Indeed, an important study found also great variability in the prevalence of uncontrolled asthma across European countries
In our study, the average total cost per patient over a 3-month period was higher when asthma was poorly controlled. Hospitalizations for asthma and emergency room visits associated costs were higher in patients with uncontrolled asthma and they represent a higher percentage of the total direct costs (16.7% and 18.6% of the total direct cost in France and Spain respectively versus
0% and 0.6% in patients with controlled asthma). This result is in line with previous European results
However, antiasthmatic drugs were the main driver of direct costs in both countries. In addition, the use of controller treatment was associated with the highest direct asthma-related costs regardless of the level of control and after adjustment for several potentially confounding factors (except in the subgroup of Spanish patients with uncontrolled asthma).
The EUCOAST study presents several limits. First, both the sampling frame and the sampling method used did not ensure the representativeness of the investigators even if their main characteristics were comparable to those of all French and Spanish GPs.
Secondly, as investigators had to recruit several patients (until 5 in Spain) there was a potential within-unit correlation in the data which was not taken into account in the statistical analysis.
Thirdly, due to a possible short term memory bias GPs and patients may have under-reported healthcare consumptions. Therefore, costs may be lower than those found in studies based on claims databases or systematic healthcare data. However, it is difficult to know whether, or to what extent, this might have biased our estimates of the effects associated with level of asthma control. Indeed, the existence of such bias would depend on whether under-reporting of costs was differential according to level of asthma control.
Costs of medications were based on prescription data. As some patients may not have been compliant, our estimates may be over-estimates of the true medication-related costs.
The response rate for the EQ-5D-3L® was lower in France as compare to Spain (73% vs. 97%). This was likely the consequence of differences in the HRQL data collection method in the two countries. In France, patients were asked to send back their questionnaires to the study center whereas in Spain most patients completed the questionnaire in the physician’s office. However, we did not find significant differences between responders and non-responders in their level of asthma control or overall costs.