Understanding the factors that influence the over-the-counter (OTC) medicine use can provide important information on guiding the proper use of OTC medicines and reducing national medical care expenditure. This study investigates the status of self-medication with OTC medicines for colds and coughs in Japan after COVID-19 pandemic and explores the associated factors.

This study is an online cross-sectional survey conducted from April 25 to June 26, 2024. The status of self-medication behaviors against colds and coughs in Japan and covariates of social background and psychological scales were collected. Associations between them were analyzed using multivariate logistic regression. Subgroup and sensitivity analyses were conducted to validate the robustness of the findings.

This study included 1,086 participants. 43.6% of the participants would take OTC medicines from the onset of colds and coughs. The proportion of seeking healthcare after symptoms lasted one week was 61.7%. Over 80% of the participants would strictly follow the usage instructions. Factors associated with seeking healthcare within one week included age, living area, education level, marital status, insurance type, having an underlying disease, regular doctor visits, and extraversion. When considering dosage adherence, the agreeableness trait was determined to be a positive factor, whereas having a child or children was a negative factor. For the awareness of expiration dates for OTC medicines, eHEALS, which indicated internet literacy for searching health-related information, was found to be a significant and robust positive factor.

A high proportion of Japanese were found to use OTC medicines for colds and coughs. Most participants demonstrated a strong awareness of proper OTC use. To further promote OTC medicine, it is important to address the key factors found in this study.

Self-medication is common in individual care when people experience symptoms and minor diseases without consulting a physician [1, 2] and is typically regarded as the use of over-the-counter (OTC) medicines to treat these conditions. The WHO also recommends self-medication with OTC medicines to treat minor diseases [1]. Proper and responsible self-medication can significantly enhance national health, reduce national medical costs, and save personal time required for medical treatments [2, 3].

Although the Japanese government also recommends self-medication, some Japanese people prefer hospital/clinic visits for minor diseases. One reason is that clinics and hospitals are readily available, and people may feel more at ease with hospital or clinic visits [2, 4]. In addition, Japan has achieved universal health insurance coverage in its healthcare system, and residents are required to pay only 30% of their medical costs in clinics or hospitals [5], with the option to purchase private insurance further to cover their remaining medical expenses.

However, hospital or clinic visits for minor diseases may result in the ineffective use of medical resources and increased health insurance costs [2]. The Japanese government encourages responsible self-medication to reduce pressure on healthcare facilities, particularly for minor diseases. It has introduced several policies, including a tax exemption scheme for the use of OTC medicine and promotions for the use of "switch OTC medicines." Nowadays, OTC medicines treating a variety of diseases are available in Japan, and among them, cold and cough medicines are frequently used and commonly found in pharmacies. In Japan, the cold and influenza season typically peaks during the winter months, which may lead to an increase in the consumption of OTC cold and cough medicines.

The COVID-19 pandemic has significantly influenced public health behaviors, particularly in relation to respiratory symptoms. There has been a marked increase in awareness about these symptoms and their potential implications. This heightened consciousness has led to greater precautionary measures such as mask-wearing, handwashing, gargling, and prompt medical consultations when experiencing cold or flu-like symptoms. A previous study has shown that outpatient prescriptions for medications related to obstructive airway diseases and cough and cold treatments decreased by up to 20% in 2020 during the pandemic [5]. This shift suggests a change in how people manage respiratory symptoms, potentially reflecting an increase in self-medication practices. A recent survey conducted by Daiichi Sankyo Healthcare Co., Ltd. revealed that over 40% of participants have turned to OTC medications more frequently to address colds [6]. However, there is limited research detailing the specific patterns of Japanese OTC cold and cough medicine use and the factors related to self-medication behaviors. Additionally, the appropriateness of using these medications depends on the progression of symptoms. While mild symptoms are generally well-suited for OTC treatments, persistent or severe ones warrant medical attention and evaluation. These observations motivate us to investigate the current status of self-medication practices for colds and coughs in Japan.

Although OTC medicines are encouraged, awareness of potential risks is crucial to ensure the safe and effective use of these medicines. Several issues in using OTC medicines should be addressed, such as adverse effects and risky practices, including misuse, overuse, and abuse [3, 7,8,9,10,11]. These behaviors may prolong the disease, cause adverse effects, and lead to drug addiction and dependence [3, 7]. Therefore, adherence to instructions and choosing the proper medicines are considered crucial. Abuse and overuse of OTC medicines have been frequently reported [7, 8, 11]; however, Japanese consciousness of the safe use of OTC medicines and the related factors are less reported.

This study conducted an online cross-sectional survey to provide a comprehensive understanding of the status of self-medication for colds and coughs, as well as the proper use of OTC cold and cough medicines in Japan. We were also interested in resources for information acquisition when people are unsure about the choice of OTC medicines and the extent of their knowledge regarding safe use. Furthermore, this study explored the potential factors associated with self-medication behaviors.

The questionnaire contained the investigation of self-medication behaviors against colds and coughs at different progresses and situations, measurements of psychological factors related to behaviors, and the demographic information of participants (Table S1).

For the demographic information, sex, age, living in an urban area or not, regional classification in Japan, education level, marital status, having children or not, the status of underlying diseases, the kind of insurance, perceived health, occupation, having pharmaceuticals-related training or not, regularly visiting a doctor or not, and the annual household income, were collected (Table S1 Part 1).

Outcomes

To ascertain outcomes, nine questions were initially developed to elucidate patterns of over-the-counter (OTC) medicine use (Table S1 Part 2). Questions 1–3 were designed to gather data regarding initial self-treatment practices for common cold or cough symptoms, practices after one week of symptom progression, and approaches to self-treatment while traveling abroad. Questions 4–7 focused on identifying behaviors employed when participants were uncertain about which medication to select. These behaviors encompass seeking recommendations from peers, consulting healthcare professionals (including physicians, pharmacists, nurses, and registered sales personnel), seeking advice from non-medical contacts (such as family and friends), and utilizing online information resources. Finally, Questions 8–9 were designed to evaluate safe medication practices, specifically adherence to dosage instructions and awareness of expiration dates.

From the perspective of safe OTC medication use, practices such as seeking medical attention when symptoms persist, consulting healthcare professionals when uncertain about appropriate medication selection, strictly adhering to dosage instructions, and being aware of expiration dates can be considered indicative of responsible self-treatment. Furthermore, given the increasing accessibility of information via the Internet and artificial intelligence (AI), the extent to which participants utilize and rely on online sources is also of interest in this study. Consequently, the underlying causes and contributing factors influencing these individual behaviors were investigated.

Psychological factors and other scales

The psychological factors were measured based on several psychological scales [12,13,14,15,16].

Personality traits were measured using the Japanese Ten-Item Personality Inventory (TIPI-J) [12, 13] (Table S1 Part 3). This widely utilized instrument in psychological research is designed to assess respondent personality characteristics, including extraversion, agreeableness, conscientiousness, neuroticism, and openness. Previous research has demonstrated a relationship between these personality traits and health-related behaviors [17,18,19]. Consequently, we sought to determine whether these factors also influence the observed OTC medicine-related behaviors within the Japanese context, leading to the adoption of this scale.

Optimistic self-beliefs regarding the ability to cope with challenging life demands were measured using the Japanese Adaptation of the General Self-Efficacy Scale (JGSE) [20, 21] (Table S1 Part 6). The JGSE, a 10-item scale, yields higher scores indicative of a stronger belief in the efficacy of personal agency in achieving success. Self-efficacy has been established as an important psychosocial factor in healthcare [22,23,24] and has also been linked to preventative behaviors related to infectious diseases, such as COVID-19 [25]. Whether JGSE influences OTC medicine-related behaviors was examined in this study.

Two additional non-psychological scales included in this study were the eHealth Literacy Scale (eHEALS) [14, 15] and the Media Use Scale (MUS) [16]. These scales were employed to assess participants' IT literacy regarding health information and their frequency of social media use. Specifically, the eHEALS [14, 15] was utilized to evaluate respondents' perceived skills in locating, evaluating, and applying electronic health information to address health concerns. This scale comprises eight items and yields a unidimensional construct. The MUS [16] measured individual media use across six distinct purposes: social communication, self-presentation, social action (including advocacy and promotion of justice), leisure and entertainment, information acquisition, and commercial transactions. These factors were hypothesized to be associated with behaviors, particularly reliance on online sources when individuals are uncertain about which OTC medication to select.

Prior to their use as indicators, the reliability and validity of all scales were assessed via factor analysis. With the exception of the MUS, all scales demonstrated acceptable psychometric properties and were utilized as originally reported. Due to our inability to establish a satisfactory single-factor structure for the MUS, its individual items were treated as independent variables in subsequent analyses rather than being summed. (See Supplemental Information A. Results of Instrument Validation in Detail and Tables S2-S3).

A previous report calculated the necessary number based on the following formula [18].

Where p denotes the self-medication rate,\(q = 1 - p,\)\(\alpha\mathit=\mathit0\mathit.\mathit{05}\mathit,\mathit\;Z_{\mathit\alpha\mathit/\mathit2}^{\mathit2}\mathit\;\mathit=\mathit1\mathit.\mathit{96}\mathit\;\mathit\approx\mathit2\mathit,\mathit\;\delta\) is the margin of error δ = 0.1 ×p.Here, p = 0.325 is used, a reported global self-medication rate. Considering some responses may be invalid, an additional 20% is deemed necessary; thus, the final essential sample size was 1039.

Therefore, we follow this result and consider the sample size at least 1039.

The inclusion criteria were as follows: (1) age over 18 years, (2) Japanese nationality, and (3) participation in the study and provision of informed consent.

A cross-sectional, anonymous questionnaire survey was conducted in Japan from April 25 to June 26, 2024, using voluntary sampling with monetary incentives [26]. The participants were recruited using CrowdWorks (https://crowdworks.jp) and asked to complete a questionnaire on Google Forms. Those who completed the questionnaire were paid 160 Japanese Yen.

Descriptive statistics

Descriptive statistics summarized the demographics of participants and the status of the use of OTC cold and cough medicines.

Logistic regression

An approach was undertaken to investigate the factors influencing OTC use-related behaviors. The study examined five specific behaviors:

Each analysis was conducted in two stages: first, using univariate logistic regression (LR) to identify potential significant factors, and second, employing multivariate LR to determine the influential predictors. To ensure the robustness of the findings, subgroup analyses were performed by removing individuals with potential confounding backgrounds that could influence their OTC behavior. Furthermore, sensitivity analyses were conducted following over-sampling techniques to address highly imbalanced datasets. The detailed methodology and results are subsequently outlined. The response and explanatory variables used in multivariate logistic regression (LR) were listed in Table S4.

Univariate logistic regression

The univariate LR was used to explore the significant factors associated with self-medication behavior. The significance levels (α) used to select variables were 0.1.

Multivariate logistic regression

Based on the significant factors obtained from univariate LR, multivariate LR analyses with stepwise selection were conducted. Variables were selected according to the Akaike information criterion (AIC) [27, 28]. The significance levels (α) were 0.05.

Subgroup analysis was conducted using stepwise variable selection in multivariate LR. However, this analysis was restricted to respondents meeting all of the following criteria: 1) self-identified as male or female, 2) had not received training in pharmaceutical sciences, 3) held public and/or private health insurance, 4) reported no underlying medical conditions, and 5) did not regularly consult a physician. As a result, this subgroup can be considered suitable for providing more accurate analyses for the following reasons.

Please note that there is no discrimination against those who are not selected for this subgroup.

Sensitivity analysis was conducted for the variable with extremely imbalanced data, specifically "being aware of expiration dates." To address this imbalance, we employed the Synthetic Minority Oversampling Technique (SMOTE). This method involved over-sampling the minority class and under-sampling the majority class. The over-sampling percentage was set to 200%, meaning each minority instance was duplicated twice, while the under-sampling percentage was set to 150%, reducing the majority instances by half. This process created balanced artificial data, which was then used to validate the robustness of our findings from the multivariate logistic regression model. In this analysis, we focused on the variable-selected model, ensuring that our results were consistent and reliable across different sampling conditions.

All analyses were performed using R software (Ver. 4.4.1). The major functions and packages used are glm in stats (Ver. 4.4.1), lavaan (Ver. 0.6.18), and psych (Ver. 2.4.3).

The study involved 1,086 participants (Fig. 1). Of these participants, 53.9% were male, 45.7% female, and 0.5% identified as other genders. Participants ranged in age from 19 to 75 years old (Mean ± SD: 42.0 ± 10.1). Residency details indicate that 43.5% lived in urban areas, while the remaining participants were in non-urban areas (Table 1). The respondents came from the main regions of Japan (Fig. 2a). However, the regional distribution of respondents did not follow that of the Japanese population [29] (chi-squared = 50.44, p-value < 0.001).

Flowchart

Full size image

Respondent distribution and answer distribution

Full size image

In terms of demographics, 50.4% of the respondents were married, and 42.4% had children. Educational backgrounds varied: 64.0% held a bachelor's degree or higher, 30.1% were high school graduates, 4.8% had attained a master's degree or higher, and 1.1% reported having less than a high school education. Regarding health status, 16.9% of participants reported having underlying diseases, while 35.5% indicated regular visits to a doctor. Occupational distribution showed that 56.0% were employed in full-time positions (including self-employed individuals), 23.3% were unemployed, 18.1% worked part-time or irregular hours, 1.4% were pensioners, and 1.2% were students.

Annual household incomes ranged from less than ¥3,000,000 to over ¥15,000,000. Professional backgrounds included 3.7% who identified as professionals and 2.0% with training in pharmaceutical science-related fields. Lastly, participants rated their perceived health on a scale from "very poor" to "excellent" (Table 1).

At the onset of colds and coughs, 43.6% of the respondents stated that they would initially take OTC medicines, followed by performing self-care (35.1%), doing nothing (18.0%), and visiting a hospital/clinic (3.4%) (Fig. 3a). If the symptoms lasted for more than one week, most respondents chose to visit a hospital/clinic (61.7%). Moreover, when we suppose that Japanese residents went overseas, the distributions of proportions of behaviors showed high similarities compared to when they were in Japan. However, more people preferred clinic visits (6.7% overseas vs. 3.4% in Japan, χ² = 11.73, p < 0.001), and fewer people would perform self-care and doing nothing (41.1% overseas vs. 53.0% in Japan, χ² = 30.75, p < 0.001).

Answers to Q1 to Q9

Full size image

The behaviors related to the use of OTC medicines were surveyed (Fig. 3b). Regarding the choice of OTC medicines, 55.2% of the respondents selected OTC medicines based on their perceived knowledge or experience. Thus, there may be a latent risk in choosing the wrong medications. When uncertain about their OTC medication choice, 56.3% indicated they would consult medical professionals. In comparison, 45.1% responded to seeking advice from non-medical personnel, such as family members and friends. Additionally, 76.2% answered that they would refer to online information.

This study focused on two aspects of proper use: adherence to dosage and usage instructions and awareness of expiration dates. Per our survey, 80.5% of the respondents replied that they would strictly follow the dosage and usage instructions, and 90.1% answered that they were aware of the expiration dates of the OTC medicines (Fig. 3b). However, although low in proportion, 19.5% of the respondents might not always strictly follow the dosage and usage instructions for OTC medicines.

The significant variables selected from univariate LR were further analyzed by the stepwise multivariate LR (Table 2). After controlling factors, compared to people aged less than 30 years, those aged from 40–49 years old were less likely to seek healthcare (OR [90% CI]: 0.48 [0.29, 0.78]), followed by those with 50–59 years old (0.36 [0.21, 0.62]), after one week of cold and cough symptoms. Compared with people in non-urban areas, those in urban areas were less likely to seek healthcare (0.73 [0.56, 0.96]). Compared to married people, non-married people were less prone to seeking healthcare (0.64 [0.483, 0.850]). Compared with those with both public and private insurance, those with only private or public insurance (0.48 [0.25, 0.91] and 0.63 [0.47, 0.83], respectively) were less likely to seek healthcare. In contrast, those with higher education levels were more likely to seek healthcare (high school graduates, 8.52 [2.04, 58.69]; bachelor's degree only,10.34 [2.49, 71.01]; and master's degree and higher, 6.48 [1.40, 47.30]). Having underlying diseases (1.66 [1.07, 2.60]) and regular visits to a doctor (2.19 [1.58, 3.06]) were also relevant. Moreover, the personal trait of extraversion (1.06 [1.00, 1.11]) was identified as a relevant factor, implying that people high in extraversion are more likely to seek healthcare.

The stepwise multivariate LR model (Table 3) shows that having children and media use for commercial transactions (0.68 [0.50, 0.94] and 0.86 [0.76, 0.97]) are negative factors for strictly following the dosage, whereas agreeableness (1.10 [1.03, 1.18]) is a positive factor.

Unawareness of expiration dates may be dangerous because the efficacy and stability of medicines cannot be guaranteed. Stepwise variable selection determined that sex, age, media use for social action, and the eHEALS were significant variables (Table 4). Compared to female participants, male participants responded less about the expiration date (0.45 [0.29, 0.69]). Compared to people aged less than 30 years, people aged 50–59 years responded more to knowing the expiration dates (3.09 [1.26, 8.14]). People who are social media users know less about the expiration date (0.77 [0.64, 0.93]). Conversely, participants with higher eHEALS scores knew more about the expiration date (1.06 [1.02, 1.09]).

Consulting medical personnel when unsure of medicine choice is considered good practice. The multivariate LR model was constructed, in which sex (0.72 [0.55, 0.92]), age (30–39 years old (0.53 [0.33, 0.85]) and 60 years old above (0.40 [0.20, 0.78])), type of insurance (only public (0.69 [0.53, 0.91])), regular doctor visits (1.43 [1.09, 1.87]), media use for social action (1.17 [1.04, 1.32]), extraversion (1.07 [1.02, 1.13]), and agreeableness (1.08 [1.03, 1.15]), were determined to be significant (Table 5). People who are high in extraversion and agreeableness, and those who often use social media for social actions, are willing to talk to others and have a high acceptance of suggestions from others. These psychological factors positively contribute to consultations with medical personnel when selecting medicines.

Univariate and stepwise multivariate LR (Table 6) were constructed. Training in pharmaceutical science (0.27 [0.14, 0.55]), neuroticism (1.10 [1.04, 1.16]), and eHEALS (1.11 [1.08, 1.14]) was also a significant factor. People who had been trained in pharmaceutical science did not search for Internet information. In contrast, neuroticism and eHEALS positively contributed to the action of referring to Internet information.

The sample number of subgroups was reduced to 596 due to a large number of regular visits to a doctor. Factors associated with seeking healthcare even after one week of symptoms (Table 2, results of Subgroup analysis), such as age (0.49 [0.24, 0.95] and 0.31 [0.14, 0.66]), marital status (0.53 [0.37, 0.76]), and kind of insurance (0.63 [0.64,0.89]), are robust and significant factors. Area, education level, and extraversion were not detected as significant factors.

Factors associated with strictly following the dosage (Table 3, results of Subgroup analysis), such as age (3.77 [1.50, 9.81] and 9.44 [1.67, 178.59]), having children (0.63 [0.40, 0.98]), and conscientiousness (1.12 [1.04, 1.21]), were determined as significant factors. Instead of agreeableness, conscientiousness was detected in subgroup analysis. Highly conscientious people are prone to strictly following the dosage.

For the factors associated with awareness of expiration dates (Table 4, results of Subgroup analysis), age (5.04 [1.35–18.88]), area (0.54 [0.32–0.91]), occupation (3.30 [1.11, 9.79]), and eHEALS (1.05 [1.01, 1.10]) were detected as significant factors. Both age and eHEALS were confirmed robust. Living areas (urban vs. non-urban) showed a p < 0.1 in the earlier analysis. This factor has become significant when considering this subgroup. On the contrary, sex lost its significance. Notable: All respondents in the age category of greater than 60 or pensioners in this subgroup answered that they knew the expiration dates. Thus, this makes the CI incalculable. However, this would not influence the significance of other variables.

For factors associated with consulting medical personnel (Table 5, results of Subgroup analysis)), the factors except sex (0.67 [0.47, 0.94]) did not retain their significance. In this subgroup, social media use for communication (1.18 [1.01, 1.38]) and JGSE (1.04 [1.01, 1.07]) were determined to be positive contributors.

For factors associated with referring to internet information (Table 6, results of Subgroup analysis), neuroticism (1.18 [1.09, 1.28]) and eHEALS (1.13 [1.09, 1.17]) were also significant factors in this subgroup, suggesting robustness.

Sensitivity analysis for the question on awareness of expiration dates showed that sex (0.55 [0.39, 0.76]), age categories (1.73 [1.01–2.99], and 2.62 [1.40, 4.98]), media use for social action (0.72 [0.61, 0.84]), and eHEALS (1.06 [1.03, 1.09]) were found robust (Table 4, results of sensitivity analysis).

This study identified key findings as follows:

This research highlights important insights into self-medication behaviors, personality traits influencing healthcare decisions, and the role of health literacy in safe and appropriate OTC medicine use among Japanese individuals.

This study investigated the status of self-medication and the use of OTC cold and cough medicines after the COVID-19 pandemic. Previously, the global prevalence of OTC medicines was estimated to range from 11.2 to 93.7%, depending on the country and area [30]. It was reported that the use of OTC medicines in Japan is less popular in other countries because of the high coverage of the Japanese national health insurance system and the easy access to healthcare [4]. Per this study, 43.6% of the respondents replied that they would use OTC medicines at the onset of a cold and cough. It was reported in previous research that people felt reassured and believed that they could quickly recover when examined by a physician [2]. In contrast, our results showed that compared to visiting a hospital/clinic for colds and coughs, individuals were more likely to conduct self-medication, especially using OTC medicines.

To our knowledge, there is no report on the selection of healthcare versus OTC medicines when Japanese people travel overseas. This study found that the behavior could change when they travel overseas. Although most Japanese still chose OTC medicine use, approximately double the number of people would choose to visit a hospital/clinic if they went overseas (6.7% overseas vs. 3.4% in Japan). One possible reason could be the language barrier faced by Japanese travelers when purchasing OTC medicines in foreign pharmacies. Other factors, such as the perceived severity away from home and changes in the availability and trust of pharmacies and OTC medicines, may also exist.

Early seeking healthcare may help speed up recovery time and identify diseases masked by symptoms. When we supposed that colds and coughs lasted longer than one week, 61.7% of the respondents would visit a hospital/clinic, indicating good practice among Japanese residents.

For all the statuses of OTC medicine use, we found differences in respondents from different regions (Fig. 2b). However, as our data were not regional population-based, further investigation is needed.

People with high levels of extraversion personality were more likely to visit hospitals/clinics to seek healthcare. Extraversion and openness were reported to have significant effects on cognitive flexibility [31], which is related to high willpower and motivation as well as low impulse inhibition. These could lead to the behavior of calmly going to the hospital when symptoms persisted for more than a week. As suggested in previous studies conducted in other countries [32, 33], extraversion was associated with openness to seeking treatment or preventive healthcare and was a protective factor. However, in our subgroup analysis, none of the personalities were significant. The stringent selection criteria for subjects in the subgroup analysis may substantially impact the characteristics of the population. For instance, individuals with underlying diseases were excluded from the subgroup analysis to minimize potential confounding effects. However, these conditions may influence personality traits such as extraversion. It is well-established that the relationship between personality traits and health is bidirectional and intricate [34, 35]. Extraversion may offer a protective role against certain diseases [36]; conversely, chronic conditions can lead to changes in personality traits, including a reduction in extraversion [37]. Consequently, these factors were not significant within the subgroup but attained significance when analyzing the entire group. Notably, the interaction between special groups and changes in personality traits was not thoroughly examined in this study. Given the complex interplay involved, additional research is warranted to explore this area further.

To evaluate the proper use of OTC cold and cough medicines, we investigated whether the respondents obeyed the dosage instructions and were aware of the expiration dates. Previous studies in other countries have shown that some caregivers do not read the instructions and spontaneously change the dosages when practicing self-medication for children [38]; leftover medicines from earlier treatment are also used [10]. In this study, 80.5% of the respondents answered that they would strictly obey the dosage instructions, whereas 90.1% responded that they knew the expiration dates, indicating relatively high adherence to OTC medicines and awareness of their proper use in Japan. Respondents with high agreeableness were more likely to follow the dosage instructions. In contrast, those with children and those using social media for commercial transactions were less likely to follow them strictly (Table 3). Previous studies conducted in other countries have also shown that low levels of agreeableness are related to higher barriers to medication adherence in many diseases [39,40,41]. In contrast, having children may increase the risk of spontaneously changing dosages, as aforementioned. However, in our subgroup analysis, removing participants with potentially heterogeneous backgrounds revealed that, instead of agreeableness, conscientiousness became a significant factor, which is consistent with a previous meta-analysis reporting that a higher level of conscientiousness was associated with better medication adherence [42]. For the awareness of the expiration dates, eHEALS was a positive significant factor in the overall data, subgroup, and sensitivity analyses. High eHEALS scores indicated high literacy in obtaining health information on the Internet [15]. This demonstrated the importance of obtaining the necessary information effectively and appropriately. Female participants were found to be significantly more common in the entire group and sensitivity analysis but lost significance in the subgroup analysis. Previous studies have shown that female participants are more likely to be interested in reading leaflets and checking expiration dates [43]. While sex did not show significant differences overall in this study, it should still be considered as a potential factor in promoting safe OTC medicine use. It appears that males might exhibit less concern or attention regarding safe OTC medicine use. Therefore, promoting safe OTC medicine use should involve personalized approaches tailored to individual personalities.

Regarding whether appropriate selection of OTC medicines was conducted, this study revealed that 52.2% of Japanese respondents were unlikely to consult with others when selecting medicines. Even when they needed clarification about the medicine to choose, only 56.3% of the respondents consulted medical personnel. In a similar survey in China, 86.2% of respondents considered advice from medical staff important when selecting OTC medicines [18]. Our focus on OTC medicines for the treatment of colds and coughs may have caused these differences. However, low consultation rates may also arise from national character or personal traits, as it was found that people with higher extraversion and agreeableness were more likely to consult medical personnel. Similar to previous reports [44, 45], this finding suggests that considering personal traits is crucial when promoting proper OTC medicine use.

When evaluating the Internet searches for information when confused about OTC medicine selection, high levels of neuroticism were detected to be a positive factor. People with high levels of neuroticism are reported to be more likely to experience anxiety and uncertainty about their actions [46]. The higher anxiety and uncertainty may cause a higher level of behavior.

With the highlights of large language models, daily Internet and AI use has become increasingly frequent. This study revealed that eHEALS is associated with referring to Internet information when people do not know the best OTC medicine to choose. A previous study showed the correlations between OTC medicine sales and tweet numbers of symptoms [47]. More recently, a Japanese group conducted comparisons between generative AI and package inserts of medicines and concluded that consumers should not consult generative AI [48]. Therefore, the credibility and risk of incorrect information must be considered. Improvements in the eHEALS may pave the way for the future promotion of OTC medicine.

This study had some limitations. Firstly, data were collected nationwide but not regional-population-based, making it difficult to discuss the regional differences. Secondly, we used ClowdWorks for rapid data collection, limiting respondents to platform users and introducing potential selection biases related to IT literacy and educational levels. Additionally, this self-selection process has included individuals more inclined to respond or possess heightened health awareness and/or increased extroversion, potentially leading to over-optimistic results on safe OTC use. Since no explicit assumption on the reasons for selection bias could be made, it is difficult to evaluate the potential selection bias on the results. Interpretations should consider these limitations. Thirdly, although we limited the reply time to one, duplicate respondents, such as the same person using different user IDs, could not be detected because the survey was anonymous. Fourthly, although cold remedies (oral use), cold remedies (external use), antipyretic analgesic agents, antitussives, and expectorants are generally categorized as cold and cough medicines, our survey did not provide a clear definition of OTC cold and cough medicine. Consequently, data collection relied on individual interpretations, which may have introduced slight variability in how participants understood the term. Additionally, our study did not employ direct measures of disease severity or accessibility to OTC medications. Instead, we utilized symptom timing (onset and duration of one week) and location (Japan vs. overseas) as proxies for assessing disease severity. We also considered prefecture-level data to account for variations in OTC medicine access. While these indirect assessments are important, they may influence the results.

In future research, we aim to conduct surveys that detail various categories of OTC cold and cough medicines to identify potential differences among them. Additionally, we plan to explore regional distinctions through population-based sampling within specific geographic areas to understand how access and usage vary across different geographical areas. Furthermore, considering direct measures of disease severity and accessibility could offer deeper insights into the factors influencing OTC medical use. These investigations are being planned for future research.

eHEALS:

eHealth Literacy Scale

AIC:

Akaike Information Criterion

CI:

Confidence Interval

JGSE:

Japanese Adaptation of the General Self-Efficacy Scale

LR:

Logistic Regression

MUS:

Media Use Scale

OR:

Odds Ratio

OTC:

Over-the-counter

TIPI-J:

Japanese Ten-Item Personality Inventory

Download references

Yu-Shi Tian: Writing – review & editing, Writing – original draft, Methodology, Formal analysis, Data curation, Conceptualization. Xinhua Mao: Writing – review & editing, Writing – original draft, Methodology, Conceptualization. Yi Zhou: Writing – review & editing, Methodology, Formal analysis. Kaori Fukuzawa: Writing – review & editing, Conceptualization. Kenji Ikeda: Writing – review & editing, Conceptualization. Asuka Hatabu: Writing – review & editing, Writing – original draft, Methodology, Conceptualization.

Correspondence to Yu-Shi Tian.

The research protocol was reviewed and approved by the Ethics Review Committee of the Graduate School of Pharmaceutical Sciences at Osaka University (Yakuhito23-15). The study was conducted in accordance with ethical standards outlined in the Ethical Guidelines for Medical and Health Research Involving Human Subjects (Minister of Health, Labour and Welfare, Japan). Given the nature of the online questionnaire survey, informed consent to participate in the study was obtained from all participants as follows: At the top of the online questionnaire, participants were informed that submitting their responses indicated consent to participate in the study. The survey was anonymous, and results were aggregated to ensure confidentiality. Additionally, the purpose of the study, details of the questions, time required for completion, and measures taken to maintain confidentiality were clearly documented in the online instructions.

Not applicable.

The authors declare no competing interests.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions