DISCUSSION

Per protocol analysis revealed a positive effect of the pharmaceutical interventions practiced on chronic patients. The proportion of patients reaching the therapeutic goals set by the physician almost doubled at three months (33.3% to 61.7%), consistently to some previous studies[32,37,40]. The lost to follow up population did not differ from the rest of the patients included, confirming the suitability of a per protocol analysis. This difference was also retrieved in the sensitivity analysis, logically higher and significant in the “best case scenario”. In the “worst case scenario”, the difference was no more significant but still suggested a positive effect of the intervention. Thus, if the number of patients included had reached the number statistically estimated as necessary, the difference might also get significant in this more pessimistic analysis. This study therefore demonstrated the direct clinical effect of a single pharmacist intervention on chronic patients with hypertension, type 2 diabetes and hypercholesterolemia.
According to a previous study, the benefit of a pharmacist intervention was greater in polypharmacy patients[54]. The number of patients achieving therapeutic goals decreased conversely to the total number of drugs prescribed in the control group. This result is consistent with the literature, which reports a decrease in compliance with the increase in the number of daily doses[55,56], and an increase in the risk of drugs interaction and side effects in polypharmacy patients[57]. Hence, polypharmacy patients appeared to be a preferential population to benefit from this kind of intervention.
Age and the number of drugs prescribed appeared to have similar effects on the achievement of therapeutic objectives. However, in our population, the number of drugs prescribed was correlated with the age of patients, as previously described[58]. After adjusting for the number of drugs, age did no longer appear as a factor influencing the results of pharmaceutical intervention. Therefore, the age-related effect of pharmaceutical intervention was probably due to the higher number of prescribed drugs in elderly[58], thus revealing age as a confounding factor[59]. Age then appeared as a less important criterion than polypharmacy in selecting a target population for pharmaceutical interventions.
Surprisingly, the intervention was also more effective on patients with a high educational level. It is well established that a low socioeconomic level is related to a bigger morbidity, especially for cardiovascular diseases[60,61] so we could have expected a more important effect on low educational level patients. However, this result should be interpreted with caution, as an inconsistent relationship between the socioeconomic level and the adherence to medication has also been reported[19,62]. Thus, we need further studies to confirm the results related to educational levels and to find a way to reach this population.
Pharmaceutical intervention was effective for all the studied pathologies, which is consistent with the literature and corroborates previous studies concerning hypertension[33,40], hypercholesterolemia[40] and type 2 diabetes[36,37,63]. However, the rate of patients achieving therapeutic objectives, regardless of the group, was lower for type 2 diabetes patients, than for the 2 other diseases. This could be related to the low adherence of type 2 diabetic patients found in the literature[64] and it highlights the need for these interventions in these patients.
Thirty-five minutes were required to discuss all topics with the patient. In France, the average duration of a consultation with a general practitioner consultation is 10 to 16 minutes[65]. Unfortunately, prescribers cannot afford to spend an additional 20 minutes with all their patients to discuss drug topics in detail in addition to their regular consultations. This implies the necessity to have these interventions performed by a practitioner other than the physician, with a solid knowledge of medicines, whose activity would be at least partially devoted to these interventions. This finding highlighted the difficulty to offer this kind of intervention to all patients. It is therefore necessary to target these pharmaceutical interventions to specific populations of patients who need them most.