INTRODUCTION
Aging and the global increase in life expectancy of populations lead to a greater number of health problems and, consequently, to a higher number of prescribed medicines, especially in the older population and in patients diagnosed with chronic diseases. This situation conduces to an increased risk of drug adverse events and potential drug-drug interactions.1,2,3
Patients’ medication knowledge (MK) may influence the process of how the medicines are taken, potentially leading to incorrect use and lower effectiveness. Thus, MK may be a relevant factor to bear in mind, when assessing the use of medicines by the patients and their adherence to the prescribed therapy, which, consequently, have a substantial impact on health outcomes.4
According to García Delgado et al.5, the knowledge that a patient has about the medication can be defined as the amount of information that the patient acquires regarding that medication, which is required to use medicines properly, including the proper process of use (regimen, dosage, duration of treatment, and route of administration), the therapeutic objectives (indication and effectiveness), the security (adverse effects, precautions, contraindications, and interactions), and the conservation.
Frequent changes in medication (prescriptions, doses or new medicines) and the inability of patients may have to denominate the medicines used may contribute to reduce MK.6,7,8 Also, medicines, which have only recently become available on the market, seem to lead to more knowledge gaps, as suggested by a medication review study in patients taking non-vitamin K antagonist oral anticoagulants.7 Additionally, low MK has been correlated with non-adherence to the medication and the consequent risk of relapse as reported in a prospective study held in patients diagnosed with inflammatory bowel disease.9 Also, a lack of MK has been identified as a factor contributing to poor disease management for chronic diseases.10
It is not frequent to assess the MK during a patient care process nor during the evaluation of medication use. Despite MK not being a systematically systematically assessed parameter in daily practice, some studies, which included this assessment, revealed a high prevalence of patients with low levels of MK.8,11,12,13 An increased rate of inadequate MK (72%) was identified among Spanish community pharmacies costumers in a cross-sectional study. The lowest knowledge score was determined in the “medication safety” item (12.6% and 15.3% for “contraindications” and “side effects,” respectively).8
There are a few tools available to assess patients´ MK, but the most commonly used tools, which are structured interviews and specific questionnaires, are only available in English.14 However, studies that analyse their validity and reliability are still missing so that they can be used with confidence.6,13,15,16
In 2009, García Delgado et al.5 proposed a questionnaire to assess patients’ MK, which includes four dimensions (therapeutic goal, medicines use process, safety, and conservation). This tool is reported as reliable, presenting a Cronbach´s alfa value of 0.68.8 This questionnaire was applied to a group of patients followed up in a Spanish community pharmacy, taking one or more medicines. Some predictive factors for low patient MK score were identified and included the use of several associated medicines, unqualified workers and caregivers, and the inability to identify the name of the medicines.8 This questionnaire was used as a starting point to produce a cultural adaptation to the European Portuguese language, designated as Patient Knowledge about their Medications, Conocimiento del Paciente sobre sus Medicamentos; in Portugal; CPM-ES-ES. However, the authors mentioned that further studies are needed to demonstrate the equivalence of the psychometric properties (reliability and validity) of the Portuguese version, so it could be used in pharmaceutical care research projects in Portugal.17
Given the inexistence of a range of tools properly certified to be applied to the Portuguese population directly in clinical practice, the aim of this study was to perform out a pilot study employing a new tool to perform the assessment of older patients’ MK in a daily clinical practice.
MATERIALS AND METHODS
Study design
We conducted a descriptive, cross-sectional study, in an outpatient diabetes clinic in the municipality of Faro (Algarve, Portugal). This clinic is integrated into a regional association of patients with diabetes (in Portuguese, AEDMADA - Associação para o Estudo da Diabetes Mellitus e Apoio ao Diabético do Algarve).
Patient selection
Patients were recruited for the study during their routine, previously scheduled, and consultation with a physician. Before the consultation, following the usual procedure at the clinic, the date of the consultation was confirmed with each patient by phone, and patients were specifically asked to bring with them all medicines and food supplements they were taking at date.
During the consultation, patients were invited to be a part of the study. Those who accepted filled and signed all necessary informed consent forms. Data collection occurred at the end of the consultation and was registered anonymously. Patients with alterations in cognitive abilities that could hinder understanding the study aim were excluded.
Study recruitment was performed for three months (January-March 2018), using a convenience sample, according to patient acceptance.
Approval for this study was previously obtained from the clinic’s administration board.
This study only included patients with 65 or more years old.
Medication knowledge assessment
The assessment of patient MK was carried out through a structured interview using an algorithm specifically developed for this purpose, that considers six different parameters; the name of the medicines, the strength of the drugs, the therapeutic indication, the timing of administration, dosing intervals, and the storage conditions (Table 1). All six parameters were evaluated individually for each drug used by the patients.
For each patient, the average percentage of the six parameters considered in the MK assessment was determined and then used to obtain the score for medication knowledge (SMK) according to the following formula:

The patient was considered as having “lack of MK”, when the obtained score was less than 50%.
Furthermore, for each patient, data regarding socio-demographic characteristics, the clinical profile, and the number of physicians following the patient at the moment of the interview were collected using an appropriate form created for this purpose. Health literacy was assessed throughout the Short Assessment of Health Literacy - Portuguese language (SAHL-PT),18 using a tool where patients were flagged as “low health literacy” when reached a score ≤14 (out of 18). WHO ATC index was used to perform medication classification (https://www. whocc.no/atc_ddd_index/). Treatment adherence was also assessed using the Haynes Sackett method.19 In this test, patients with scores between 80% and 100% were considered “adherents to the therapeutic”.
Ethical consideration
Ethical approval of this study was obtained from the Cranfield University Research Ethics Committee (reference: CURES/840/2016). All data were collected anonymously without any identification of the participants.
Statistical analysis
IBM-SPSS software, version 26.0 (SPSS Inc, Chicago, Illinois), and AMOS 24.0 (SPSS Inc, Chicago, Illinois) were used to analyse all the collected data. Quantitative variables are presented as minimum, maximum, mean, standard deviation, and median. The qualitative data are described by counts (n) and the respective percentages (%). Adherence to normal distribution was assessed with Kolmogorov-Smirnov’s test. Parametric (Student’s t-test, Pearson correlation coefficient) or non-parametric (chi-square, Mann-Whitney’s U test, Spearman’s correlation coefficient) procedures were used for analysing associations or group differences. Statistical significance for all procedures was considered when p<0.05.
RESULTS
A sample of 49 patients was included in this study, of whom 27 (55.1%) were male. The mean age was 73.22 ± 5.72 years, with 16 (32.7%) over 75 years old. Patients were mainly retired (n: 44; 89.8%) and only 4 (8.1%) were living alone. More than half of the patients (n: 27; 55.1%) only concluded up to 6 years of schooling (Table 2).
Patients were taking a mean of 6.9 ± 2.8 medicines and 7.7 ± 3.3 daily units per day, and 346 medicines. Most of them were using 5 or more medicines per day (81.6%), which can be considered polymedication (Table 2). Only 6 patients (12.2%) were using food supplements.
Medicines most frequently taken were those acting on the alimentary tract and metabolism (group A), cardiovascular system (group C) and nervous system (group N) (Table 3).
Low health literacy was flagged in 27 (55.1%) patients with an average score of 13.3 ± 3.6, and a prevalence of non-adherence to treatment in 9 patients (18.4%).
Regarding the use of medicines, 15 patients (30.6%) showed a lack of MK (score <50%), but an average prevalence of 58.5 ± 15.2% for correct information about the medicines currently used was determined.
Despite most patients (n: 46; 94%) stated being able to read the packaging, the name of the medicine was identified only in less than half of the medicines used (n: 21; 41.9%) (Table 4).
For 279 (80.6%) medicines, patients could not indicate the drug’s strength, and for 83 (24.0%) medicines, patients did not know either their therapeutic indication or presented the wrong one. Problems related to the administration time were identified in 32 (9.2%) medicines, and in 25 (7.2%) of the analyzed medicines, difficulties related to the number of units were also pointed. Incorrect storage conditions were mentioned in relation to 207 (59.8%) medicines (Table 4).
The “administration time” and “number of units” were the items with the highest knowledge rate, followed by the item “therapeutic indication.” On the opposite, the items presenting the lowest knowledge rate were “drug´s strength” and “storage conditions” (Table 4).
Older patients, presenting fewer years of schooling, revealed significantly lower MK than younger patients (p=0.049) (Table 5). Patients with 12 years of schooling had significantly more MK than those with 4 years of schooling (72.5 ± 13.6 vs. 55.4 ± 12.4; p=0.036).
No significant differences were observed for MK score in relation to the remain socio-demographic variables, number of medicines used per patient and patient´s medication adherence (p>0.05).
Polymedicated patients (5 or more medicines) exhibited an increased score of MK (mean of 60.4 ± 14.7 vs. 50.3 ± 15.0), although differences were statistically non-significant (p>0.05) in relation to patients taking less than 5 medicines.
An increased score for health literacy (p<0.05) and treatment adherence (p>0.05) was achieved for patients showing a higher MK (Table 5).
A lower MK (score <50%) was determined for the medicines belonging to the C group (drugs acting in the cardiovascular system) and A groups (drugs acting on alimentary tract and metabolism) compared to other therapeutic groups, but no significant differences were observed between all different groups (p>0.05).
Additionally, patients being followed by 2 or more physicians presented a higher but not statistically significant (p>0.05) MK score.
DISCUSSION
Some methodologies are emerging for evaluating MK, although each one includes different evaluation parameters, making it difficult to compare the obtained results. By the time this study was conducted, for the Portuguese population, only one validated tool was available to assess MK. However, that tool only considers one drug per patient and the results of the psychometric tests were shown to be inadequate possibly due to the sample size.20
Using the methodology presented in the current study, about a third of the patients enrolled (n: 15; 30.6%) showed a lack of MK while another cross-sectional study that applied the questionnaire (PKM-PT-PT) and that was carried out in community pharmacies in the Lisbon Metropolitan area (Portugal) identified 65.9% of patients without appropriate MK. Although, in the current study, patients revealed a higher rate of MK, items such as “therapeutic goal” (70.9%) and “process of use” (36.7%) were the items with higher scores and results were similar in both studies. Likewise, the item with the lowest rate was “storage conditions”, which also had similar rates of knowledge in both studies. However, note that the patients’ MK was assessed only for a medicine they were taking,9 while in the present study, all medicines prescribed were considered for each patient and so the comparisons between the obtained results should be interpreted with caution.
The results obtained in the current study indicate a lower prevalence (n: 15; 30.6%) of inappropriate MK compared to the score determined by Romero-Sanchez et al.8, in a study enrolling community pharmacy users, where an inadequate MK was determined in 71.9% of the patients, with 65.7% having no MK and 6.2% insufficient MK. The interpretation of these results should take in consider that different tools were used to assess MK, using different weighting factors for the various parameters analyzed, which may contribute to the different results reported.
One factor previously correlated with inappropriate MK is the lack of knowledge that patients revealed in relation to the name of the medicine they use,8 which was also verified in about 60% of the medicines used by the patients in this study.
Although medicines included in groups A and C are the most used by this group of patients, a particular lack of MK associated with these groups was observed, which is a surprising and worrying situation as we could expect that patients know more about the medicines they use regularly, highlighting the need of improving patients’ knowledge.
Medication adherence has been considered as a factor which may limit the improvement of chronic disease (e.g. diabetes, hypertension) control. Participants in the current study presented a 18.4% prevalence for medication non-adherence, adding to the observed lack of MK, will contribute to aggravate health outcomes with negative impacts on disease control.21,22,23
Education level has been considered a relevant factor to achieve positive health results, given its ability to influence treatment adherence and patient self-care.19,24 In a study conducted in Portugal with university students, the level of MK identified was low for self-medication.25 The results obtained in this study have shown that patients with lower schooling levels tend to present low MK scores, as described in the previous section, which reinforces the need to pay additional attention to those patients to help them achieve the desired positive outcomes with the prescribed medication.
In this study, a failure in the identification of drugs (name and drug strength) was clearly observed, which may represent a risk in case they should provide information about the medication they are taking (for example, to the other physicians or in any health service). Therefore, it could be useful, in the future, to draw up the list of medicines or to develop appropriate tools (information technology tools or others) for each patient, considering information from different sources (e.g., hospital, community pharmacies), so they can carry it with them, whenever they access health services or whenever they must identify their medicines.26,27
The signaling of patients with low MK scores may allow the identification of patients, who could benefit from participating in therapeutic education programs. An improvement in MK was achieved for patients under polymedication through the provision of a counseling session after being referred by physicians to medication review service, provided by pharmacists, due of signaling problems with therapeutic adherence or knowledge.28
Farsaei et al.29 have revealed that progress could be achieved in disease management, including a significant metabolic improvement, through the implementation of an educational program (interventional approach, conducted by pharmacists), which addressed several topics, such as medicines, therapeutic adherence, daily self-care records, and pill box usage.24
A more recent systematic review and meta-analysis, which evaluated patient-centred outcomes reported in studies searching for interventions to increase treatment adherence, has shown that MK assessment may be an important tool to be used in clinical services such as medication reviews.23
Also, in the management of chronic diseases, such as Parkinson’s disease, MK can have a relevant influence on physical and social performance, impacting the health-related quality of life.30
Study limitations
The current study has some strengths and limitations. The main strength is that the assessment of MK was applied to all medicines used by patients. Although this methodology may be time-consuming and exhaustive, it allows to specifically identify the items in which the patients have more difficulties regarding their MK.
The obtained results show that for each patient, the knowledge varies with the medicines they are taking, which may compromise their full benefits.
One limitation of MK assessment carried out in this study is that the questionnaire did not include questions related to drug safety (side effects, interactions, and contraindications). In the future, this MK tool should include questions related with this item, using a perceptible format that allows to assess patients’ knowledge about the safety of the medication.
Our sample size was low, as we were unable to recruit additional patients, and the methodology we used to assess MK was not previously validated. Additionally, the descriptive nature of this work does not allow us to establish causation, but despite these limitations and shortcomings, the use of a novel tool in a population that is scarcely studied with regard to the subject of this paper leads us to assert that the data we gathered was an important starting point, which will soon lead us to conduct further research in this matter.
CONCLUSION
The results obtained in the current study gave us access to detailed information about MK of each medicine used by the enrolled patients, which seems to be useful in the future for scheduling pharmaceutical interventions and customizing the needs for each patient according to inappropriate points in the use of each medicine.
ACNOWLEDGMENTS
The authors acknowledge the AEDMADA clinic for allowing the conduct of this study in their facilities and all subjects, who accepted to participate in this study.
Ethics
Ethics Committee Approval: Ethical approval of this study was obtained from the Cranfield University Research Ethics Committee (reference: CURES/840/2016).
Informed Consent: Those who accepted filled and signed all necessary informed consent forms.
Peer-review: Externally peer-reviewed.
Authorship Contributions
Concept: M.E-S., Design: M.E-S., T.N., Data Collection or Processing: M.E-S., Analysis or Interpretation: E.P., M.D.E., Literature Search: M.E-S., Writing: M.E-S., M.D.E.
Conflict of Interest: No conflict of interest was declared by the authors.
Financial Disclosure: The authors declared that this study received no financial support.