Introduction

The aim of pharmacotherapy is the achievement of definite therapeutic outcomes that recover the patient’s quality of life while minimizing patient risk (1). Drugs show their efficacy only if administered in appropriate quantity with proper combination of drugs and foods and at suitable times (2). A drug interaction is a situation in which a substance affects the activity of a drug in a way that the effects are increased or decreased, or produces a new effect. When discussing drug interaction, what first occurs to mind is typically drug-drug interaction. However, interactions may also exist between drugs and foods, as well as between drugs and herbs/supplements (2).

Foods may alter the effects of drugs by interfering with pharmacodynamic mechanisms through exerting synergistic, additive, or antagonistic impacts or with pharmacokinetics process, such as absorption, metabolism, and excretion, resulting in decreased drug efficacy or increased drug toxicity (3). They may also produce a new effect that might not produce on its own (4). A medication can also change the body's ability to utilize particular foods or nutrients, which are important in supporting human health (2).

The interaction of food and drug is a common hidden problem encountered in clinical practice that can cause pharmacotherapeutic complications through prolonged hospitalization or readmission to hospital, prescription cascades, and deterioration of the patient's health condition (5). Therefore, the determination of the factors associated with the occurrence of food-drug interactions are important for both clinical and economic reasons. Although it is difficult to identify the pharmacotherapeutic complications caused by food-drug interactions (6), it is possible to determine the factors exposing the patients to such complications (5).

To the best of our knowledge, there is no study investigating this issue in Iran. Even though numerous food-drug interactions have been reported, the scale of this negative phenomenon has not been assessed in the population of patients in Iran. With this background in mind, this study aimed to investigate the main factors that can expose the patients to the risk of potential food-drug interactions among inpatients in Northeast Iran. This study was conducted in the Department of Internal Medicine of a teaching hospital in Mashhad, Iran, to determine the frequency and also the factors associated with the risk of food-drug interactions.

Methods

This cross-sectional, descriptive study was conducted on 400 patients admitted to the Department of Internal Medicine of a teaching hospital (a 1000-bed tertiary general grade A hospital), affiliated to the Mashhad University of Medical Sciences, Mashhad, Iran, from 20 March 2013 to 20 April 2013. This center is a medical education research unit and a training center for students at professional and super professional levels.

The study protocol was approved by the Institutional Review Board and the Local Ethics Committee (IR.MUMS.REC.1391.754). Informed consent was obtained from all patients prior to their inclusion. The investigator was trained to achieve adequate qualification prior to performing the study assessments. The patients who received at least one of the above mentioned drugs were screened for eligibility to enter the study. The subjects with more than 48 h hospital stay and those taking oral medication were included in the study. On the other hands, the patients who were not able to communicate verbally or those who could not take a drug orally were excluded from the study. Nutritional supplements and vitamins were not in the scope of this study.

To evaluate the potential food-drug interactions, 19 commonly prescribed medications with tough guidelines regarding the time of consumption in regard to meals were selected. The undesirable potential food-drug interactions (i.e., the improper use of drugs in relation to meals) for each drug were checked with the United States pharmacopeia dispensing information (also called drug information for the health care provider) or/and patient information leaflets and expressed in percentage as the ratio of the number of cases with potential interaction to all cases.

All demographic information (e.g., age, gender, education level) and other relevant data, including duration of disease, drug name/dosage form/dosage, and timing of the administration, were recorded for each patient. In addition, the presence of a companion during hospitalization, as well as the physician, nurse, and patient’s knowledge for the proper use of drugs were considered. All the above required data were collected from the patients’ medical records, treatment charts, and/or patient interview by the investigator. The sample size was determined as 384 case rounded up to 400 patients with a margin of error of 5% and a confidence level of 95%.

A descriptive analysis was established for summarizing the study variables (i.e., potential food-drug interactions, gender, level of education, number of patients accompanied by someone, and number of patients educated to consume their medicines at an appropriate time with respect to meals). The non-normally distributed quantitative data were presented as median and interquartile range (IQR). Furthermore, the qualitative data were expressed in terms of frequency and percentage. There were no missing data.

Multiple logistic regression (forward selection) was used in order to identify the factors associated with food-drug interaction, including those factors previously defined as statistically significant according to the univariate regression analysis. Additionally, 95% confidence intervals were calculated for the most interesting variables (i.e., gender, age, duration of disease, number of medications, education level, companionship). 

The Brier Score as the quadratic scoring rule was also determined calculating the mean squared differences between the predicted and actual outcomes. The Brier Score ranges from 0 for a perfect model to 0.25 for a model with 50% incidence of the outcome. The area under the receiver operating characteristic curve (AUC) is reported as a measure of accuracy (an area of 1 indicates an accurate model). All statistical analyses were performed using SPSS (version 16.0; SPSS Inc., Chicago, IL, USA). P-value equal to or less than 0.05 was considered statistically significant.

Results

Out of the 400 patients, 44.5% of them were male. The median age of the patients was 47 years (IQR=11). Most of the hospitalized patients were suffering from cardiovascular and gastrointestinal diseases. Furthermore, 89.8% of the patients (n=359) often consumed their medicines at an inappropriate time with respect to meals. The prevalence rates of potential food-drug interactions regarding the 19 investigated drugs in the study participants are briefly summarized in Table 1. According to the results, nitroglycerin (100%) and metronidazole (100%) had the highest potential interactions among the investigated drugs, followed by captopril (86.7%), and ferrous sulfate (83.8%). However, no potential interaction was detected for aspirin and magnesium hydroxide (Table 1).

Table 1. Potential interactions between foods/nutrients and drugs prescribed for inpatients

Table 2 presents the demographic variables and nature of potential food-drug interactions. The results revealed that the patients would be less exposed to potential food-drug interactions as their age increased (Table 2). As shown in Table 2, the food-drug interactions were more frequent among the women. In terms of the patient education level, 49.2% of the participants had less than high school educational level, and 21% of them were highly educated. The food-drug interactions were most commonly observed in the poorly educated patients(45.1%). As indicted in Table 2, 84 (21%) patients had a high education level, and food-drug interactions observed in all these patients.

Only 33.2% of the patients were educated (only by nurses) on how to use their medications with respect to meal. However, the patients educated by nurses were also at a high risk of food-drug interactions. Despite the recommendations, all these patients consumed their medications at the inappropriate time with respect to meal. Due to the lack of dispersion, patient education was excluded from the further analysis.

Table 2. Prevalence of food-drug interactions based on patients’ characteristics

 * Variables are described in median (interquartile range)

† Variables are described in number (%)

The median number of drugs given on a regular basis per patient was 4 cases (IQR=2) per day. The large number of the concurrently used drugs placed the patients at higher risk for food-drug interactions (Table 2). The majority of the patients (54.8%) had no companion. Out of the 359 patients who had at least one potential food-drug interaction, 159 accompanied patients (44.3%) took their medicine at inappropriate time with respect to meal.

As the findings indicated, longer duration of the disease resulted in higher possibility of food-drug interaction (Table 2). It was also revealed that the number of potential food-drug interactions increased in parallel with the number of drugs used by the patients (Table 2). The results of the univariate and multiple logistic regression analyses are tabulated in Table 3.

According to the univariate analysis, the patient’s age, disease duration, number of medications, and education level were significantly associated with higher risk of food-drug interactions (P<0.001).

Table 3. Univariate and multivariate analysis of food-drug interaction in hospitalized patients admitted to the internal medicine ward (n=400)

^a Analysis by Fisher’s exact test.

^b Analysis by independent-samples t-test.

^c To adjust for multiple comparison, P2 for the final model is 0.68.

However, food-drug interaction showed no statistically significant association with gender (P=0.36) and having a companion (P=0.25). After adjustment for confounders, age [β=0.005, CI: 0.0 to 0.01, P=0.033, OR=0.88, (CI 0.78-0.99), P=0.048], number of medications [β=0.1, CI: 0.083 to 0.117, P<0.001, OR=1.52, (CI 1.19–1.93), P<0.001], and duration of disease [β=-0.037, CI: -0.05 to -0.023; P<0.001] remained statistically significant. The R² for the final model was 0.68, and the AUC revealed a good level of accuracy (AUC=0.83).

Discussion

According to the findings of the present study, the age, number of medications, and duration of disease were the most important factors significantly associated with higher risk for potential food-drug interactions in hospitalized patients admitted to the Department of Internal Medicine. In this regard, the older patients or those with longer duration of disease were at lower risk for exposure to potential food-drug interactions, and the patients were more likely to be exposed to potential food-drug interactions as the number of their medication increased.

The lower frequency of possible interactions in older individuals is inconsistent with the reports in the literature (7-9). This finding can be due to the fact that the older patients are more concerned about the control of their health; therefore, they often share their personal health information and decision-making with caregivers when needed. Moreover, they may be also more willing to ask questions about how and when to take their medications in relation to food and drugs. In addition, older patients with chronic health problems receive more attention from caregivers, especially nurses. Therefore, they were at lower risk of exposure to potential food-drug interactions.  

In the present study, it was concluded that the patients with a long history of disease faced with lower possibility of food-drug interactions. Paying more attention to the details of pharmacotherapy over a long period of time may explain this. In the current study, it was observed that the frequency of potential food-drug interactions increased in parallel with the number of drugs used by the patients. This is in line with the findings obtained by Mayur et al. (10) and Rasheed et. al. (11) showing a significant relationship between the number of medications a patient consumed and the number of drug-nutrient interactions for which a patient was at risk.

In this respect, the easiest way to reduce the frequency of potential food-drug interactions is to decrease the number of medicines prescribed, which is not rational for patients with multiple chronic conditions; therefore, a careful selection of therapeutic alternatives is recommended (1, 12).

In this study, no association was obtained between lower knowledge about the proper use of medications with regard to meal and a higher possibility of food-drug interactions. Similar to a study conducted by Jarosz and Wolnicka (2, 5), in our study, the majority of the patients (66.8%) were not educated about the factors (like meals) that might affect the medicinal impact. The patients are needed to be aware of the protective measures to avoid potential food-drug interactions (3, 13).

Accordingly, in the previous studies, the implementation of educational programs in this regard was reported to decrease the rate of drug-food interactions in hospital (3, 4, 13, 14). Many of potential food-drug interactions can be avoided by advising the patient to take their medicines at the appropriate time with respect to meals. It is emphasized that the nurses as the final link in the treatment chain are able to play a key role in food-drug interaction education to patients (5, 6, 15, 16). Knowledge and assessment skills are required to improve the patient safety (2, 5, 7, 8, 17, 18). Moreover, the clinical pharmacists can play an important role in nurse training as an effective method to reduce food-drug interactions in hospitals (4, 14).

However, we found that the patients who were educated by the nurses were also at high risk of potential food-drug interactions. It means that sometimes the patients do not pay enough attention to nurses’ recommendations. Nonetheless, they should be informed that communicating with caregivers (i.e., nurses, physicians, and pharmaceutical consultants) is an effective way to prevent food-drug interactions (18). Moreover, nurses need to have adequate knowledge of food-drug interactions to be able to educate the patients in this regard. Therefore, the health professionals and pharmacists are recommended to inform the nurses and patients for the signs and risk of possible side effects (12).

One of the limitations of this study was the different social and psychological conditions of the patients, which might have affected the results. In addition, we did not evaluate the actual pharmacotherapeutic complications due to the potential food-drug interactions. This study was conducted in only one inpatient teaching hospital, and the results may differ from the results of the studies carried out in other centres. Furthermore, this study was conducted in the Internal Subdivision; as a result, the findings may not be much comparable with those of other divisions due to the use of different medications in different divisions (especially specialized departments and ICUs where patients receive more drugs).

Implications for Practice

As findings of the present study revealed, the number of medication was associated with higher risk of potential food-drug interactions. Therefore, to lower the frequency of potential interactions among the inpatients, it could be necessary to decrease the number of medicines prescribed or make a careful selection of therapeutic alternatives. It is also suggested to consider patient’s knowledge of food-drug interaction to avoid impairment in the treatment process.

Future studies are warranted to further evaluate the real outcomes of food-drug interactions. Moreover, future research is recommended to determine the knowledge of physicians, pharmacists, nutritionists, and nurses about the food-drug interactions. The establishment of nurse-patient communication seems to be an effective way to prevent food-drug interactions. To improve the performance of medication administration, the pharmacists are suggested to implement training courses on food-drug interactions for the physicians, nutritionists, and nurses.

Acknowledgments

The present study was derived from a dissertation submitted by Mostafa Abdollahi in partial fulfillment of a requirement for the degree of Master of Science in nursing. Hereby, we extend our appreciation to all hospital staff, patients, and their families for the collaboration with this study. The study protocol was approved by the Institutional Review Board and the Local Ethics Committee (IR .MUMS.REC.1391.754). 

Conflicts of Interest

The authors declare that they have no conflicts of interest.