Assessment of Psychometric Properties of the Persian Version of the Medication Literacy Measure Authors: Mona Kargar, Soha Namazi, Ava Mansouri, Mohadeseh Balvardi, Kheirollah Gholami Published: 2026 ABSTRACT Objective:Given the association between medication literacy (ML) and its correct use, assessing the status of ML is essential in the community. This study aimed to validate a Persian-language ML assessment instrument, addressing the lack of a previously validated Persian tool.Methods:This was a cross-sectional study conducted in Tehran. A database search was conducted to identify current tools. The selected instrument was translated and then back-translated into English. Questions regarding medication information-seeking behaviors and attitudes toward pharmacist guidance were added. The instrument was presented to the Academy of Medical Sciences, and their feedback was incorporated. The content validity index (CVI), content validity ratio (CVR), and comprehensiveness were assessed based on the experts. Reliability was evaluated through test–retest (intraclass correlation coefficient [ICC]) and internal consistency (Kuder–Richardson [KR] coefficient).Findings:Two systematic reviews that evaluated 52 ML tools were assessed. ML measure (MLM-17) was selected. Four questions on attitudes and behaviors were added. Following feedback from the academy, the number of these questions increased to 7, while one ML question was replaced. Following the experts’ comments, a CVI of 0.77 was calculated, and questions with a CVR <0.99 were revised or removed. Accordingly, two questions were removed from the attitude and behavior sections, and one question was added to the ML section. The assessment of reliability showed ICC = 0.67 and KR = 0.71.Conclusions:The psychometric evaluation confirmed the tool’s validity and reliability. Therefore, it is a suitable instrument for assessing ML in Persian-speaking populations. INTRODUCTION As a component of health literacy, medication literacy (ML) was first introduced in 2005.[1] It encompasses the cognitive and social skills required for adhering to medication information, extending beyond merely reading labels or recalling drug names.[2] In most healthcare settings, medications are the most widely used interventions and the most frequent cause of harm.[3] Insufficient details in prescriptions or misunderstandings of instructions, mostly due to limited information and inadequate knowledge about medication usage, can lead to medication errors and reduced effectiveness of treatment.[4] In each healthcare experience, understanding the written health information materials, like notes on medication packaging, is a crucial part. Healthcare providers, comprising pharmacists, frequently overestimate patients’ capacity to read prescription information, health forms, and other associated documents. However, many patients with limited literacy skills struggle to comprehend written information, like medication labels. Pharmacists anticipate that patients will properly take their prescribed medications and adhere to their treatment plans. To do so, patients need sufficient information as well as the necessary skills in reading, calculation, and self-management to apply this information properly.[5] ML is defined as “the degree to which individuals can obtain, comprehend, communicate, calculate, and process patient-specific information about their medications to make informed medication and health decisions to safely and effectively use their medications, regardless of the mode by which the content is delivered (e.g., written, oral, and visual).”[6] In fact, ML is health literacy in the medication use domain. Usually, individuals with a higher level of ML demonstrate improved adherence to medications. Conversely, for individuals with limited ML, comprehending drug information and adhering to instructions for managing their medication can be challenging. This is much of a notice when considering that medication misuse may result in a lower quality of life for those patients.[1] Literacy depends not only on general reading and writing abilities but also on the specific setting and subject matter. Even those with strong overall literacy may have challenges when they need to apply their skills in new or specialized areas that demand particular knowledge, such as healthcare situations. For this reason, health literacy has developed as a distinct type of literacy focused on understanding and using information in a medical context.[7] Using nonprescription medications is common for managing minor illnesses worldwide. Research indicates that patients frequently become confused by the medication information supplied by pharmaceutical companies and healthcare providers. Misunderstanding medication details (including warning labels, package labels, and leaflets) can lead to inappropriate utilization and nonadherence, and the misuse of these medications can result in major health consequences. In fact, misunderstanding dosage instructions on medication packaging has been identified as a significant factor contributing to medication errors and side effects.[8] In numerous investigations focusing on the assessment of patients’ ML, researchers have predominantly employed general health literacy assessment tools, such as the Rapid Estimate of Adult Literacy in Medicine and the Test of Functional Health Literacy in Adults.[9] Other tools that have been developed to assess ML levels include the 14-item MedLitRxSE, the 20-item Numeracy Understanding in Medicine Instrument, the 21-item Montana State University Complementary and Alternative Medicine Health Literacy tool, and the 6-item Medication Health Literacy Measure.[8] Nevertheless, the general instruments are not tailored to address the unique competencies necessary for managing and comprehending medications. Consequently, their use in clinical settings may lead to concerns regarding the accuracy of measurement, as relying on inappropriate tools can result in misguided assessments and affect patient care.[9] Moreover, the application of the other above-mentioned tools is limited due to their narrow area of coverage domains. Furthermore, some of their content may not be applicable in everyday use in patient care or the community.[8] Furthermore, despite the impact of ML on patients’ safe use of medications, existing tools are primarily designed to measure general health literacy or for specific patient populations, and few are specifically dedicated to ML.[10] Given the lack of a valid Farsi tool for assessing ML, selecting, translating, and, if necessary, developing a valid tool in Farsi is of great importance. Using such a tool enables the assessment of the current situation, and based on the results, appropriate strategies and effective interventions can be designed. Therefore, the main objective of this study was to develop a valid and reliable tool for measuring ML in the Iranian population. METHODS This cross-sectional study was conducted in Tehran, Iran. Study instrument In the initial phase, a search of English-language literature on ML assessment tools was conducted. The identified instruments were then critically evaluated and compared. Following the selection of the optimal questionnaire, forward translation was performed into Persian by a researcher with expertise in both the content and the English language. Subsequently, the translated version was reviewed against the original by another bilingual researcher, with necessary revisions made to ensure linguistic accuracy and conceptual equivalence. Afterward, the Persian version was back-translated into English by an independent bilingual translator. Two researchers compared the original and back-translated versions to verify faithfulness to meaning and concepts. Additionally, supplementary items were added to assess respondents’ attitudes and practices regarding medication information seeking and pharmacist guidance. Because several items aimed to evaluate patients’ comprehension of information provided in leaflets, drug labels, and accompanying materials, which were in English in the original version, these sources were substituted with images and leaflets of medications available in Iran. Moreover, the alignment of response options for over-the-counter (OTC) precaution/warning items with Iranian drug leaflet content was made. At this stage, both the original questionnaire items and the translated/culturally adapted counterparts were presented by the researchers during a session with the Pharmaceutical Working Group of the Academy of Medical Sciences of the Islamic Republic of Iran. Written copies were also distributed to the members for detailed review and commentary. Subsequently, the questionnaire items were revised and refined in accordance with the feedback received. Psychometrics Assessment of validity Content validity To this end, the judgments of a panel of experts were systematically utilized. They were requested to evaluate each questionnaire item in terms of its relevance, simplicity/comprehensibility, and clarity using a four-point ordinal scale: 1 = undesirable, 2 = somewhat desirable, 3 = desirable, and 4 = fully desirable. In addition, they were invited to propose modifications or recommend the removal of items where deemed necessary. The necessity of each item was further assessed through categorical ratings of “essential,” “useful but not essential,” and “not essential.” The expert panelists were composed of specialists in clinical pharmacy, pharmacoeconomics, as well as individuals with experience in rational drug use and the drug information center. Validity was evaluated using the content validity index (CVI). Additionally, the content validity ratio (CVR) was calculated based on experts’ ratings of item necessity. According to Lawshe’s table for six experts, the minimum acceptable CVR threshold is 0.99. Therefore, items with values below this cutoff were revised or removed from the questionnaire. Comprehensiveness This refers to the tool’s ability to cover all domains related to the topic under investigation. To evaluate this index, experts were invited to propose any omitted items at the end of the questionnaire. Assessment of reliability Test–retest assessment In the reliability validation phase, the questionnaire was completed by volunteer participants on two occasions, 2 weeks apart. Given that the tool’s objective was to assess the ML among the general adult population, individuals aged over 18 years, excluding healthcare professionals (physicians, pharmacists, nurses, dentists, midwives, physiotherapists, and operating room technicians, etc.) were eligible. The intraclass correlation coefficient (ICC) was calculated between scores from the two administrations and interpreted as the test–retest reliability coefficient. Internal consistency method To determine the questionnaire’s reliability, internal consistency was also assessed by calculating the Kuder–Richardson (KR) coefficient. A coefficient between 0.7 and 0.8 was considered indicative of appropriate reliability. Analyses were performed using Excel software. Ethical considerations This study was approved by the Research Ethics Committee of the Institute of Pharmaceutical Sciences-Tehran University of Medical Sciences, with identification code IR.TUMS.TIPS.REC.1402.092. RESULTS Study instrument A preliminary search of electronic databases, including Google Scholar and PubMed, was conducted to identify established tools for measuring ML. The search revealed two comprehensive systematic reviews that had rigorously collected the studies. Therefore, these reviews served as the basis for evaluating the primary ML tools. The first study, by Pantuzza et al.,[11] systematically investigated definitions, tools, and associated factors of ML. The authors searched multiple databases, including Medline, Lilacs, Cochrane, CINAHL, PsycINFO, Embase, Scopus, Web of Science, as well as thesis databases and reference lists of included studies. Observational, experimental, or methodological studies were considered eligible. From this process, 35 studies were selected and critically appraised based on conceptual dimensions and terminology. These 35 studies were ultimately reviewed in detail by the present research team. The second systematic review, by Gentizon et al.,[9] critically evaluated, compared, and summarized the properties of the ML measurement tools in adult care recipients. The authors performed structured searches across Embase, Medline, CINAHL, APA PsycINFO, and Web of Science Core Collection. Supplementary searches were conducted in these indexed in ProQuest, DART Europe, Google Scholar, and the reference lists of eligible studies. Methodological and measurement quality were rigorously evaluated using the “Consensus-based standards for the selection of health measurement instruments.” Out of the initial 5035 citations, 17 studies were ultimately selected, which were subsequently reviewed in depth by the present research team to guide instrument selection and adaptation. In total, the 52 studies identified from the two aforementioned systematic reviews served as the initial pool for the present study [Figure 1]. Of these, 12 duplicate studies were removed. Since the present study aimed to develop a tool for measuring ML in the general population, studies conducted in specific populations (14 studies), studies focused on specific medications (2 studies), nonoriginal studies (2 studies), and studies with no access to the questionnaire (1 study) were excluded. This process yielded 21 eligible studies. Among the remaining studies, 6 used researcher-constructed questionnaires, and 15 studies used 8 distinct, validated tools. These tools were retained for further comparative evaluation. The process of reviewing and selecting medication literacy assessment tools using studies included in two systematic reviews. ChMLM = Chinese medication literacy measurement, MLT = Medication literacy tool, MSU CAM HLS: Montana State University Complementary and Alternative Medicine Health Literacy Scale, PTHL-SR = Pharmacotherapy literacy questionnaire in Serbian, NSAID = Nonsteroidal anti-inflammatory drug, RALPH = Recognition and Addressing of Limited Pharmaceutical Literacy Subsequently, data from all identified tools were extracted into a dual-sheet Excel database. Sheet 1 documented instrument-level metadata, including tool name, author(s), publication year, country of origin, target audience, number of items, type of medication studied, factors influencing responses, administration mode (in-person/remote), administration method, and concurrent instruments. Sheet 2 captured item-level details for each question, encompassing source tool, author(s), publication year, type of medication (prescription or OTC), specific medication referenced, ML domain, estimation of the question’s simplicity for the general Iranian audience (5 being easiest, 1 being hardest), reliance on prior knowledge versus provided material, and type of supplementary content provided alongside the questionnaire, etc. This structured extraction encompassed 186 items across the tools. Following a comprehensive review, the Medication Literacy Measure (MLM-17) was selected for adaptation in the present study. The MLM comprises 17 items categorized into four sections titled medication vocabulary (five questions with true/false/don’t know answers), OTC medication labels (five questions), supplement advertisements (three questions), and prescription medication labels (four questions).[8] Based on the original developers, each question in MLM is given one point for correct answers. High ML is applied when the total score of the tool is 13 or higher. To enhance comprehensiveness, additional items were developed to assess respondents’ reliance on pharmacists and sources of medication information. Following forward and back translation, feedback from the academy was obtained and incorporated into the tool. Refinements included expansion of attitude and practice items from 4 to 7, removal of one vocabulary item on fixed-dose combinations due to low relevance in Iranian clinical and community contexts, addition of a practical item on medication administration after a meal, and elimination of all brand names and manufacturer identifiers from questions and accompanying images to comply with ethical and regulatory standards. Despite national prohibitions on public media advertising of supplements and medications, items evaluating the interpretation of supplement advertisements were retained, due to the prevalence of advertisements for these products on social media, to assess respondents’ interpretation of the proposed scenario. Psychometrics Assessment of validity Content validity The calculated CVI was 0.77. All items with a CVI <0.79 and a CVR <0.99 were edited or removed if necessary. Based on the received feedback, the attitude and practice questions were reduced to five. Additionally, one question was added to the ML section (OTC label). The final questionnaire, with a total of 23 items, was used for the next phase. Assessment of reliability Test–retest assessment For the test–retest reliability assessment, a convenience sample of 19 participants was recruited. Each participant completed the self-administered questionnaire on two separate occasions, with a 2-week interval, using either a paper-based or electronic format. It was determined that the ICC = 0.67 with a confidence interval (0.31–0.86), which is considered moderate. The Bland–Altman plot showing the mean difference between test and retest scores is presented in Figure 2. Bland–Altman plot showing the difference between the mean scores of the test–retest translation. SD = Standard deviation Internal consistency method The internal reliability of the items was examined by calculating the KR reliability coefficient, which was 0.71 for the present questionnaire. Structure of the final questionnaire Considering the edits made, the final questionnaire consisted of a total of 23 item, as follows: attitude and practice (5 items) and ML (18 items) comprising medication vocabulary (5 items), OTC medication labels (6 items – one question on the ingredients of an OTC medication was added based on suggestions, and the question related to expiration date was edited), dietary supplement advertisements (3 items), and prescription medication labels (4 items) [Supplementary File 1]. DISCUSSION Evaluating the current state of ML within the general population is crucial, as establishing an accurate baseline is a fundamental prerequisite for defining target outcomes and formulating effective intervention strategies. Therefore, the present study was conducted to develop a standardized questionnaire based on validity and reliability indicators, to provide a valid and reliable tool for measuring ML in Persian. The standard psychometric steps for questionnaires developed in another language are essential and have been performed in numerous cases.[12] Based on the review of existing literature, we found that the sole previous investigation in Iran on ML is a study conducted by Peiravian et al.[13] The study employed a single-item health literacy screening tool, which was derived from the methodology of Morris et al., to evaluate health literacy with a specific focus on medications. The question asked was: “How often do you need to have someone help you when you read instructions, pamphlets, or other written material from your doctor or pharmacy?” Participants responded using a five-point scale ranging from 1 (never) to 5 (always), and scores exceeding 2 indicated inadequate literacy levels. This cross-sectional study was conducted on 1104 individuals ≥15 years old with reading ability from six counties in Qazvin province. Results demonstrated that the prevalence of inadequate ML was 30.3%. Furthermore, logistic regression analyses identified that age, educational attainment, county, residential area, and family socioeconomic status were significantly associated with ML.[13] In general, ML assessment tools commonly contain questions about medications that might not belong to respondents personally and may include both nonprescription and prescription drugs.[814] Respondents are expected to accurately read, comprehend usage instructions, and process based on the information provided with the medications.[10] The current study applied MLM-17, which is the revised and adapted English version of the Chinese Medication Literacy Measurement (ChMLM-17).[8] In the development of ChMLM, the following definition of ML as “the ability of individuals to safely and appropriately acquire, understand, and act upon basic medication information” was used. In Taiwan, where the questionnaire was developed, medication information is typically disseminated through patient education sheets, medication labels, and pharmaceutical advertisements. The developers emphasized the criticality of patients’ ability to read, deal with, comprehend, and assess information from such sources, proposing this questionnaire as an effective evaluation framework.[8] In the present study, a question relating to fixed-dose combination drugs, in the section of the vocabulary of medication, which primarily encompassed the literacy domain, was removed following expert recommendations and replaced with an item addressing medication intake after meals, underscoring the importance of culturally tailoring health-related questionnaires.[15] The former expression might even be unfamiliar to some healthcare providers. Sections such as those concerning OTC medications, prescription medication labels, and dietary supplement advertisements primarily targeted the information-seeking domain, with some items being more complex, involving overlapping subdomains like literacy, information seeking, and evaluation.[8] Additionally, the expert panel suggested including a question about tablet composition in the OTC medication labels section. Psychometric assessment revealed moderate validity with a calculated CVI and acceptable reliability metrics as indicated by the ICC and KR coefficient. Limitations This study represents the first attempt to evaluate the psychometric properties of an ML measurement tool in the Persian language. Despite this contribution, several limitations should be acknowleWdged. First, while the validation was conducted in Persian (as the national official language), the sociolinguistic variations – encompassing diverse ethnic groups such as Kurdish and Turkish speakers – were not explicitly accounted for. This may introduce a degree of measurement bias, as the conceptual interpretation of items could vary across different cultural and linguistic contexts. Second, the study was conducted in Tehran. Regional variations across the country in critical factors, including medication access, baseline literacy levels, and standards of pharmaceutical counseling, may influence both individual scores and the instrument’s overall psychometric performance. Finally, the sample size for the test–retest reliability assessment, though methodologically justified, remains a limitation for the study. CONCLUSIONS Adapting ML questionnaires to cultural contexts is essential, as the structure and presentation of pharmaceutical information differ across countries. Accordingly, the current questionnaire – developed based on the dosage forms and drug leaflets commonly used in Iran and achieved acceptable validity for reliability – offers a suitable instrument for researchers and academics working in this domain. AUTHORS’ CONTRIBUTIONS Conceptualization Kh.Gh, Investigation: M.K, A.M, S.N, Data Curation: M.K, Formal Analysis: M.B, Methodology: S.N, M.B, M.K, Project Administration: Kh.Gh, M.K, Writing – Original Draft: M.K, Writing – Review & Editing: Kh.Gh, M.B, A.M, S.N. Conflicts of interest There are no conflicts of interest.