The results of the present study show a high prevalence of PPI use in a large population cohort throughout a 14-year observation period. Despite the insistence on the need to reduce the use of these drugs, only a slight decrease in the consumption of certain types of PPI was observed in 2011; otherwise, there was a marked increase from 2002 to 2015, with a particularly high prevalence of use among the elderly population.
According to the latest report on the use of antiulcer drugs in Spain, from 2002 to 2012 the use of these drugs increased from 33.3 DHD (DDD/1000 inhabitants) in 2000 to 136.8 DHD in 2012, which represents an increase of 310.4%; this increase is partly explained by the increase in the use of PPIs (> 500%). Of the PPIs, the most commonly used was omeprazole, with a DHD of 18.1 DHD in 2000 and 104.0 in 2012. The use of other PPIs (esomeprazole, lansoprazole, pantoprazole, and rabeprazole) has also increased during this period, although to a lesser extent than omeprazole. In absolute terms .
Our results are in line with those observed in different European and non-European countries [23, 24]. In France, where there are more studies on the use of PPIs, the prevalence fluctuates between 19.5 and 33%. In general, the use of PPIs seems to be higher in France than in other European countries, which report prevalence ranging from 7 to 18%. [7, 25,26,27].
In contrast, in Denmark, the prevalence of PPI use increased fourfold between 2002 and 2013, reaching 7.4% in 2014; however, even this peak prevalence is significantly lower than the prevalence observed in our study.  and in other studies of similar populations, such as the Icelandic population, which also experienced an increase in PPI consumption between 2003 and 2015 (from 8.5 to 15.5%), although it was slightly lower increases observed in our study . In Switzerland, an increase in PPI consumption from 19.7% to 23.0% was observed between 2012 and 2017, an increase of 4.8% compared to 6.4% .
In population-based studies conducted in countries less comparable to ours, the prevalence of PPI use in the Australian population was 12.6% in 2016 and it was 20-37% in hospitalized populations in China and Thailand [31, 32].
It appears that the prevalence of consumption increased significantly with the age of the patients, reaching prevalences of 19.91% and 54.64% in individuals between 45 and 64 years old and those over 65 years old, respectively. The Danish study also found that the prevalence increased significantly with age, reaching 20% in people over 80. . In the Australian study, the prevalence increased with age, especially after 65 years (33.4%), reaching 42.2% in people aged 75 to 84 and 42.8% in people older than 85 years old. This increase in the dispensing of PPIs with age is observed in both men and women.  and was particularly noticeable in people over the age of 75 .
In terms of gender, we observed that the prevalence of PPI use was higher in women (20.43%) than in men (15.69%). Most articles in European and non-European populations presented similar data [23, 28, 29]although in some these differences were not observed [24, 30].
In general, the duration of PPI treatment recommended in clinical guidelines is 12 weeks . Several definitions of long-term treatment are used in different studies . Like some studies, such as the Australian study which defined long-term treatment as 3 months, we used a value of 180 DDD, which was based on 3 months of PPI use. In our study, we found that 25% of patients consumed more than 180 DDDs. This proportion was higher in elderly patients (93.9%) and lower in young people (
This excessive use of PPIs, often off-label, can be explained by the perception of PPIs as benign treatments with few adverse effects or because they are prescribed according to the clinical picture in patients (especially elderly) with symptoms suggestive of digestive pathology that require treatment but are not confirmed by endoscopy. This can also be explained by the increased use of antiplatelet agents in primary prevention, which observational studies have shown to increase the risk of bleeding. [35, 36]. However, as different studies have shown, primary prophylaxis associated with NSAID use is often performed incorrectly in populations without bleeding risk factors associated with NSAID use. .
In our study, we did not have access to information regarding the reasons for PPI use nor to data regarding the prevalence of gastroesophageal reflux disease or peptic ulcer disease to allow a discussion of these factors.
This study has a number of limitations. The main one is the lack of data on the specific clinical indications for the use of PPIs and on whether PPIs were correctly prescribed in the study population. Second, prevalence data refer to the dispensing of drugs by the public health system and not to their actual use. Although there are studies that have shown that the dispensing of drugs correlates well with their consumption and provides better results than using prescription data, the limitations of using dispensing data must be taken into account. . Third, consumption was estimated using the DDD. The DDD values established by the WHO have additional limitations, as there may be differences between them and the actual doses used in clinical practice. However, this technical unit of measurement allows the comparison of consumption data between different countries. Fourth, actual consumption of these drugs may have been higher than reflected in this study, since private providers and patients taking PPIs without a prescription were excluded. However, the denominator took into account the population of the RSL, which was somewhat larger than the population that could obtain drugs from the public health system. Finally, although the population included in the study is representative of the general population, it was not possible to ensure that the prescribing habits of LHR family physicians are representative of the prescribing habits of all physicians. family in the country.