**8. Discussion**

One in every two people who take one of the considered medications is at risk of a potential interaction. Specifically, older people and the population with a higher educational level represent a profile at risk of a potential interaction between medications and nutritional supplements. Older people are also more likely to use both drugs and supplements because of a higher potential to ge<sup>t</sup> sick [10]. This is an important issue from a public health perspective, as there are population groups, such as the older population, who are high consumers of these two kinds of health products. This study is one of the first to examine factors associated with specific medication-dietary supplement interactions [11,12].

Qato et al. already showed that more than two-thirds of older adults used prescription medication with OTC medication or dietary supplements [4]. This is in line with our results indicating that the older population has greater probability of suffering a potential interaction. According to Kantor et al. supplement use in the US showed a downward trend among young adults aged 20 to 39 years, stable use among middle-aged adults aged 40–64 years, and an increase among adults over 65 years of age; this last population has a greater probability of being under pharmacotherapeutic treatment due to their clinical status [2]. In this population, the intake of both products is essential. Polypharmacy is a well-known phenomenon that is mainly observed in older populations [13]. The appropriateness of these medications is questioned in some cases [14]; however, in other situations such as antihypertensive or diuretic medication, the need is beyond doubt as cardiovascular illnesses are among the most important causes of death and disability in the US [15]. A similar scenario is observed with dietary supplements, where the effect of the supplementation may reduce the risk of several chronic diseases [16,17].

People with a primary level of education showed lower likelihood of being at risk of a potential interaction. This group also has a lower likelihood of taking dietary supplements. An explanation for this might be that people with a higher educational level might be over-concerned due to a flood of health information about health, and they might be taking dietary supplements when they do not need them. However, this result may also indicate a higher concern regarding the health of people with a higher education; higher education mediates the impact on health outcomes through health literacy [18]. Furthermore, people with a higher education level, which are likely to have a higher socioeconomic status, may have more resources to access dietary products.

These results are especially important for healthcare professionals. Determining whether patients at risk of suffering a potential interaction are taking a dietary supplement that might interact with their prescribed medication is important. In the case of tetracyclines, the effect of the drug may be decreased and the patient may be uncovered for a potential infection; in the case of thiazides, patients may suffer metabolic alkalosis; and in the case of angiotensin II receptor blockers, cardiac function may be affected. This information should be considered at the time of prescription or dispensation. A secondary assessment could be made to analyze whether this supplement is really needed, and if so, to try to adapt it to the pharmacologic treatment. In addition, the assistance of a specialist, such as a nutritionist, is recommendable.

In this line, public policies designed to inform healthcare professionals how to detect potential interactions and tools to help them identify them are highly recommended. These tools could be incorporated in the respective electronic tools of prescription and dispensing, and they might not only remind healthcare professionals to ask about the use of those dietary supplements that may generate an interaction, but also offer alternatives to avoid the interaction in the event that the supplement is recommended. In addition, it is important to inform citizens that, before taking a supplement, it is desirable to consult their doctors to verify that the supplement is necessary and safe taking into account their prescription drugs.

This study has several strengths. It is among the first to evaluate medication use with specific dietary supplements in a representative sample of the American population, providing useful information for targeted public health planning. In the NHANES, medication and dietary supplement use were assessed through in-home interviews, and boxes were seen by the interviewers in most participants. This reduces the recall bias, which is especially notable in medication [1]. However, the present study also has several limitations. First, there is no certainty that the interactions detected occurred. Furthermore, some of these interactions were dose-dependent, and information on dose was not available. Second, information only showed self-reported recent consumption, so it was impossible to discern whether the consumption was concurrent. Finally, other important medication groups that might generate serious interactions, such as quinolones, were not assessed due to their limited representation in the sample. These limitations are not restricted to the present study, as the

methodology and analysis strategy followed were similar to those of previous studies based on the NHANES database [4].
