Abstract
To evaluate the prevalence, influencing factors, and behavior rules of self-medication in children. Articles on self-medication in children from various electronic databases (PubMed, Cochrane Library, Web of Science, the WHO website (https://www.who.int/), ABI, CNKI, and Wanfang), were searched to August 2022. The single-group meta-analyses of the prevalence, influencing factors, and behavior rules of self-medication in children were performed using Revman 5.3 and Stata 16.0. The overall pooled prevalence of self-medication in children was 57% (95% CI: 0.39-0.75, I² = 100%, P < .00001 Z = 6.22). The pooled prevalence for main influencing factors, in terms of caregivers, was: 73% (95% CI: 0.72-0.75, I² = 100%, P < .00001, Z = 111.18) for those in rural areas; 55% (95% CI: 0.51-0.59, P = .04, Z = 26.92, I² = 68%, P < .00001) for females; 75% (95% CI: 0.74-0.76, I² = 68%, P < .00001, Z = 106.66) for those whose income was less than 716 dollars; 77% (95% CI: 0.75-0.79, I² = 99%, P < .000001, Z = 92.59) for the middle-aged and elderly; and 72% (95% CI: 0.58-87, I² = 99%, P < .00001, Z = 9.82) for those with a degree below bachelor. In the process of self-medication for children, 19% (95% CI: 0.06-0.32, I² = 99%, P < .00001, Z = 2.82) of the caregivers did not read the instructions, 28% (95% CI: −0.03-0.60, I² = 100%, P < .000001, Z = 1.77) neglected adverse effects, 49% (95% CI: 0.11-0.87, I² = 100%, P = .01, Z = 2.51) spontaneously increased or decreased the dosages, 49% (95% CI: 0.48-0.55, I² = 65%, P < .00001, Z = 16.51) had an awareness of over-the-counter (OTC) drugs, and 41% (95% CI: 0.18-0.64, I² = 99%, P < .00001, Z = 3.49) misrecognized the antibiotics. Self-medication for children was common, although the overall prevalence was not very high. The prevalence of self-medication in children was relatively higher among those caregivers who were female, rural, had low-income, were elder, or had a degree below bachelor. Common behaviors during self-medication in children included spontaneous dose increase or decrease, a lack of awareness of OTC drugs, and misconception of antibiotics. Government departments should formulate corresponding policies to provide quality health education resources for the caregivers of children.
The prevalence, influencing factors, and behavior rules of self-medication in children vary among primary studies.
We performed a systematic review and meta-analysis on the prevalence, influencing factors, and behavior rules of self-medication in children to provide a reference for the regulation of self-medication in children, improving attitudes toward taking medicine, habits intervention when taking medicine, understanding the economic burden, and other related problems.
The prevalence of self-medication in children was relatively higher among those caregivers who were female, rural, low-income, age, or education degree below bachelor. We also systematically evaluated the self-medication behavior of caregivers who had self-medication of their children. Some polices should be formulated to provide the quality health education and better medical resource.
Introduction
Self-medication is the self-determined use of non-prescription drugs to relieve mild and short-term discomfort or to treat mild medical conditions without the guidance of doctors or other health professionals.1,2 It has been suggested that arbitrarily increasing or decreasing the dose of a prescription drug under the guidance of doctors is also considered self-medication.3,4 A survey across 25 countries showed that the proportion of individuals who self-medicated themselves was up to 50%. 5 In Poland, 45.6% of people self-medicated during the COVID-19 pandemic. 6 Lawan et al. 7 found that 78.95% of the individuals had experienced self-medication behaviors in the past. In the United States, patients with symptoms (such as flu) were more likely to self-medicate 8 . In Brazil, self-medication mainly involves over-the-counter (OTC) drugs. 9 It was predicted that 2.5 billion euros would be saved per year if 10% of German patients self-medicated with OTC drugs. 10 Nearly 50% of Chinese residents above the age of 15 have exhibited self-treatment behaviors, 11 while 43.5% had self-medication within a 6-month period. 12 Most drugs used for self-medication (66.2%) were obtained from retail pharmacies. 13 The prevalence of self-medication was higher in both urban and rural women, 14 and the mean prevalence of self-medication in middle-aged and elderly individuals was found to be 45.52%. 15 Another study found that 78.7% of college students self-medicated when they felt unwell. 16
There are 253.38 million children are aged 0 to 14 years in China. The prevalence, influencing factors, and behavior rules of self-medication in children vary greatly across studies. Some studies have reported that the prevalence of self-medication in children is as high as 91%. 4 Conversely, the results of other studies also showed that only 25% of children experienced self-medication. 17 Moreover, some studies found that the prevalence of women who self-medicated their children was 50.2%, 18 which was slightly higher than that of men. In contrast, other studies have reported a higher prevalence in men. 19 The prevalence, influencing factors, and code of self-medication in children vary among studies. Therefore, we performed a systematic review and meta-analysis on the prevalence, influencing factors, and behavior rules of self-medication in children to provide a reference for the regulation of self-medication in children, improving attitudes toward taking medicine, habit intervention when taking medicine, understanding the economic burden, and other related problems.
Data and Methods
Literature Search
Chinese articles on the status quo and factors influencing self-medication in children were identified in the China National Knowledge Infrastructure and Wanfang databases. The following search terms were used: “self-medication,” “children’s self-medication,” “self-medication-related policies,” “self-care,” and “influencing factors of children’s self-treatment.” We accessed all studies published up to August 2022. PubMed, Cochrane Library, Web of Science, the WHO website (https://www.who.int/), and ABI full-text databases were searched for articles on the current status, influencing factors, strategies, and economics using the following search terms: “self-medication,” “self-care,” “self-treatment,” “children’s self-medication,” “influencing factors of children’s self-treatment,” “prevalence of children’s self-medication,” and “attitude and strategies of children’s self-treatment.” We accessed all studies published up to August 2022.
Inclusion and Exclusion Criteria
Inclusion criteria was as follows:
(1) Publicly published articles with data integrity, including articles on self-medication for child-related diseases.
(2) Articles on the current situation of self-medication in China and around the world.
(3) Articles analyzing the influencing factors and strategies of self-medication in China and worldwide
(4) Articles written in Chinese or English.
Exclusion criteria:
(1) Duplicate publications.
(2) Conference reports, reviews, and dissertation-type articles.
(3) Articles with incomplete data, not quantifying self-medication data, including data only for self-medication-related services, or including data on influencing factors based on hierarchical evaluation, rather than the original study of a certain factor.
Data Analysis
The effect index and 95% confidence interval were calculated using RevMan 5.3 and Stata 16.0. Heterogeneity was tested using I² values. A random-effects model was used when I² > 50%, which indicated heterogeneity across studies, while a fixed-effects model was used when I² < 50%. Subgroup analysis was conducted to compare the differences considering various factors, and funnel plot analysis was conducted to determine publication bias in the published literature. A sensitivity analysis was conducted, and a P value <.05 was considered statistically significant.
Literature Screening Process and Characteristics of the Included Studies
See Figures 1 to 3, and Tables 1 to 4 for details of the literature search and the risk of deviation for the final articles selected.
Flowchart of selection of studies for the systematic review and meta-analysis.
Diagram depicting the risk of deviation of each article.
Diagram depicting the risk of deviation.
Descriptive Summary of Studies Included in the Systematic Review and Meta-Analysis 1.
Inclusion indicators were: (1) The total prevalence of self-medication among children. (2) Prevalence of self-medication among urban children. (3) Prevalence of self-medication among rural children. (4) Prevalence of self-medication among female caregivers. (5) Prevalence of male caregivers self-medicating children. (6) Prevalence of self-medication in children with an income higher than 5000 yuan. (7) Prevalence of self-medication in children with an income lower than 5000 yuan. (8) Prevalence of self-medication in children with middle-aged or older caregivers. (9) Prevalence of self-medication among children with young caregivers. (10) Prevalence of self-medication in children with a bachelor’s degree or above. (11) Prevalence of self-medication in children with a degree below bachelor degree. (12) Prevalence of individuals who did not read the instructions. (13) Prevalence of individuals who ignored adverse reactions. (14) Prevalence of individuals who increased or decreased the dose of medicine on their own. (15) Prevalence of awareness of OTC drugs. (16) Prevalence of individuals with misconceptions regarding antibiotics.
Descriptive Summary of Studies Included in the Systematic Review and Meta-Analysis 2.
Descriptive Summary of Studies Included in the Systematic Review and Meta-Analysis 3.
Descriptive Summary of Studies Included in the Systematic Review and Meta-Analysis 4.
Risk criteria: 1. The source of information was defined; 2. Inclusion and exclusion criteria for exposed and unexposed subjects were listed, or a previous publication referred to; 3. The time period used to identify the patients was; 4. Indicated whether the subjects were consecutive, if not population-based; 5. Indicated if evaluators of the subjective components of the study were masked to other aspects of the status of participants; 6. The assessments undertaken for quality assurance purposes were as follows: 7. The exclusion of any patient from the analysis was explained; 8. Described how confounders was assessed or controlled; 9. If applicable, the handling of missing data in the analysis was explained as follows; 10. Patient response rates and completeness of data collection are summarized.
Results
Epidemic of Self-Medication in Children
Combined effects of different studies
Seven studies providing specific data on 24 290 surveyed caregivers were included in the analysis. The highest reported prevalence of self-medication was 91%, whereas the lowest was 25.2%. As shown in Figure 4, the combined results showed a prevalence of 57% (95% CI:0.39-0.75, P = .57, Z = 6.22) and a visible degree of heterogeneity (I² = 100%, P < .00001). During the analysis, we found that I² is more than 50%, so the random effect model was used to calculate the combined result.
Forest plot of the pooled prevalence of self-medication in children.
Impact of Specific Factors on Self-Medication in Children
Regional effects
As shown in Figure 5, the combined results of the 4 studies showed that the prevalence of rural caregivers who conducted self-medication for their children was 73% (95% CI:0.72-0.75, I² = 100%, P < .00001, Z = 111.18), which was higher than that of urban caregivers (61%) (95% CI:0.60-0.63, I² = 100%, P < .00001).
Forest plot of pooled estimated prevalence of self-medication provided by caregivers from different regions.
Effect of gender
As shown in Figure 6, the integrated results of the 3 studies showed that the prevalence of children self-medicated by female caregivers was 55% (95% CI:0.51-0.59, P = .04, Z = 26.92, I² = 68%, P < .00001), which was higher than that of male caregivers (45%) (95% CI:0.41-0.49, I² = 66%, P < .00001).
Forest plot of pooled estimated prevalence of self-medication provided by caregivers based on gender.
Effect of income
As shown in Figure 7, the combined results of 3 studies showed that the prevalence of caregivers earning less than 716 dollars who self-medicated their children was 75% (95% CI:0.74-0.76, I² = 68%, P < .00001, Z = 106.66), which was higher than that of caregivers earning more than 716 dollars (68%) (95% CI:0.66-0.70, I² = 99%, P < .00001).
Forest plot of pooled estimated prevalence of self-medication provided by caregivers with different incomes.
Effect of age
As shown in Figure 8, the combined results of the 4 studies showed that the prevalence in middle-aged and elderly caregivers who self-medicated their children was 77% (95% CI:0.75-0.79, I² = 99%, P < .000001, Z = 92.59), which was higher than the prevalence of 67% observed in younger caregivers (95% CI:0.65-0.68, I² = 100%, P < .00001).
Forest plot of pooled estimated prevalence of self-medication provided by caregivers at different ages.
Effect of education level
As shown in Figure 9, the combined results of the 3 studies showed that caregivers with a bachelor’s degree or above, self-medicated their children at a rate of 38% (95% CI:0.32-0.44, I² = 66%, P < .00001, Z = 12.13), while caregivers with a degree below bachelor self-medicated their children at a rate of 72% (95% CI:0.58-87, I² = 99%, P < .00001, Z = 9.82).
Forest plot of pooled estimated prevalence of self-medication provided by caregivers with different educational levels.
Self-Medication Behavior of Caregivers
We systematically evaluated the self-medication behavior of caregivers who had self-medication of their children.
Prevalence of caregivers who do not read the instructions
The combined results of the 4 studies showed that the prevalence of caregivers who did not read the instructions was 19% (95% CI:0.06-0.32, I² = 99%, P < .00001, Z = 2.82). However, the heterogeneity across studies was high, with the rate in one study below 1%, while in another study, nearly one-third of caregivers did not read the instructions for their children (Figure 10).
Forest plot of the prevalence of caregivers who do not read the instruction manual.
Prevalence of caregivers who ignore adverse reactions
Across 4 studies, the combined prevalence of caregivers who ignored the adverse drug effects was 28% (95% CI:-0.03-0.60, I² = 100%, P < .000001, Z = 1.77). However, most caregivers were deeply concerned about the adverse effects of self-medication on children. In one study, 81% of caregivers still considered self-medicating their children, despite ignoring other adverse effects caused by drugs (Figure 11).
Forest plot of the prevalence of caregivers who ignore adverse effects.
Prevalence of spontaneously increasing or decreasing the dosage of drugs
Across the studies, the combined prevalence of caregivers who increased or decreased the dosage of a drug without medical consultation was 49% (95% CI:0.11-0.87, I² = 100%, P = .01, Z = 2.51). Although this rate was quite high, there was variability across studies, with a rate of up to 75% in one study but much lower rates in other studies, with the lowest being 8% (Figure 12).
Forest plot of the prevalence of caregivers spontaneously increasing or decreasing the dosage of drugs.
Prevalence of non-prescription drug awareness
The results showed that the prevalence of caregivers who knew and understood OTC drugs was 49% (95% CI:0.48-0.55, I² = 65%, P < .00001, Z = 16.51). Three studies showed a relatively low level of heterogeneity, indicating that only approximately half of the caregivers understood common OTC drugs (Figure 13).
Forest plot of the prevalence of OTC drug awareness among caregivers.
Misconception of antibiotics
The combined results of the 4 studies showed that the prevalence of caregivers who misrecognized antibiotics was 41% (95% CI:0.18-0.64, I² = 99%, P < .00001, Z = 3.49). Although in one study the prevalence was as low as 23%, in another study the prevalence was as high as 73% (Figure 14).
Forest plot of the prevalence of caregivers who misrecognize antibiotics.
Publication Bias and Sensitivity Analysis
Analysis of publication bias
The included studies were subjected to funnel plot analysis. The funnel plot did not show a completely symmetrical distribution, indicating relatively large publication bias (Figure 15).
Funnel plot of the pooled prevalence of self-medication in children.
Sensitivity analysis
The combined results were recalculated after removing each study separately and the point estimates were within the credible interval of the overall combined effects. The credible intervals of the 2 analyses partially overlapped, indicating that the studies had no decisive effects on the overall effects (Figure 16).
Sensitivity analysis for single study influence of self-medication prevalence in children.
Discussion
Self-medication is common among children. This can be affected by several factors. However, different studies have reported different results. This study systematically evaluated the status quo, influencing factors, and behaviors affecting self-medication in children, which provided a reference for further use of self-medication in children and the formulation of relevant policies.
In some studies,17,24 91% of the children experienced self-medication, and the prevalence was reported to be 78% in another study. 4 However, in 3 other studies,20,21,23 the prevalence rates were 28%, 25%, and 39%, respectively, which were much lower than the pooled prevalence. This may be due to the caregiver’s fear of diagnostic errors, 30 which may have resulted in a delayed illness.
The prevalence of children who experienced self-medication was associated with many factors, such as region, 31 gender, 32 degree of education, 33 age, 34 and income 35 of caregivers. The pooled prevalence of self-medication in children in rural areas was 73%, which was higher than that in children in urban areas (61%). This may be due to the living conditions in rural areas and the fact that it is not convenient for these residents to seek medical advice, resulting in a relatively high prevalence of self-medication. 36
The pooled prevalence of children who experienced self-medication among female caregivers was 55%, which was higher than that of male caregivers (45%). This may be because traditionally women took care of their children and thus accumulated more experience, thereby increasing the likelihood that they would self-medicate their children.
Similarly, a higher prevalence of children experienced self-medication from middle-aged and older caregivers, which was likely because these caregivers were more inclined to choose self-medication owing to the accumulation of experience. 37
The pooled prevalence of self-medication in children with a bachelor’s degree or above was 38%, which was lower than that of caregivers with a degree below bachelor’s degree. This may be due to a higher level of knowledge in more educated caregivers who took a more comprehensive and cautious approach to their children’s diseases; thus, they chose professional medical treatment rather than self-medication. This might also be explained by sampling bias, as the sample size for caregivers with a bachelor’s degree or above was rather small. 14
In terms of income, the pooled prevalence of self-medication by caregivers earning more than 5000 yuan was 68%, which was lower than that of caregivers earning less than 716 dollar, indicating that caregivers with a lower income may want to reduce costs by using self-treatment rather than going to the hospital. 38
The prevalence rates of incorrect drug dosage, not understanding the prescribed drugs, and misconceptions about antibiotics in caregivers were 49%, 51%, and 41%, respectively. It is important for caregivers to realize that dosages used for adults should not be given to children. In addition, it was not likely to be related to the ingredients of the children’s drugs. These issues should not be resolved by reducing the dosage, as drugs should be used in strict accordance with the instructions and relevant regulations without arbitrarily increasing or decreasing the dosage, which may reduce efficacy or produce adverse side effects. 39 Any adverse effects arising from the use of medication should be considered. 40
Surveillance data reports from the United States showed that more than a third of poisoning incidents in children under the age of 5 were caused by drug misuse. 41 Further incorrect use is the primary cause of drug poisoning in children. 42 Unreasonable self-medication behaviors may result in health risks, such as delaying optimal treatment opportunities, creating antagonism between drugs, or drug abuse. 43 A survey in Cameroon found that 60.7% of the antibiotics purchased by adult patients were given to children. 44 Although antibiotics work quickly, they may lead to drug resistance, which lowers the efficacy of subsequent treatments. Therefore, antibiotics must not be abused. 45 Some researchers believe that the public’s knowledge of antibiotics is very low, 46 and that the abuse of antibiotics is relatively common. 47 It has been suggested that public knowledge on the rational use of antibiotics should be strengthened through education 48 to improve self-medication safety for common diseases in children currently and in the future.49,50
Since guides on the self-medication of children are rarely reported in China, it has been suggested that targeted health education is required in the community, and relevant management systems should be improved. 51 According to findings of this study, caregivers with low-level education and economic status, from rural areas, and of female gender, have a high rate of self-medication. We suggest that governments should carry out more self-medication guidance and health education in rural areas, and increase drug guidance training for women, in regions with low education level and low income. We also found some articles which reported that the prevalence of self-medication is high in rural areas and in regions with low education levels. 52
In addition, some studies have suggested that QR code technology should be used to make “patient medication guidance labels” or “medication guidelines,” 53 and residents could have access to pharmacists 54 from home 55 for professional guidance. Because self-medication in children in China is not standardized, relevant departments should conduct targeted health education in different regions and for different groups. 41 Various forms of education should be combined to improve caregivers’ awareness of self-medication behaviors in children to reduce health risks for children. 56
Although the above results definitely illustrate the influencing factors and level of behavior rules of children’s self-medication, there are still the following limitations in this study. Firstly, we only conducted a literature analysis, so there is a lack of original research. Secondly, although the article focuses on children’s self-medication, the subjects investigated in the published literature are all children’s caregivers, and memory bias will occur in the process of answering the questionnaire. Thirdly, all of the literature included was cross-sectional studies, and there may be a non-response bias in reflecting the proportion of children’s self-medication.
All in all, it was found that it is common to self-medicate children when needed, although the overall prevalence is not very high. The prevalence of self-medication in children was relatively higher among those caregivers who were female, rural, had low-income, were elder, or had a degree below bachelor level. Common behaviors during self-medication in children included spontaneous dose increase or decrease, a lack of awareness of OTC drugs, and misconception of antibiotics. Government departments should formulate corresponding policies to provide quality health education resources for the caregivers of children.
Footnotes
Author Contributions
Bi BQ: Design, analyze data and write the manuscript.
Li SG, Zhang YC: Design, quality control and review of manuscript.
Qin JM, Zhang LF, Lin CM: Participated in data collection, literature evaluation and data analysis.
Acknowledgements (For all contributors,except authors)
No
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Beijing High level Public Health Technical Talents Training Plan (Grant No. 2023-3-005) CNHDRC-KJ-L-2022-22-04410.
Ethics
This paper is a secondary literature analysis based on other original articles. The original articles we analyzed have passed the ethical review, but the present analysis in this paper does not need ethical review again.
Consent
This paper is a secondary literature analysis based on other original articles. The original articles we analyzed have passed the consent of the patients or patients’ carer, but the present analysis in this paper does not need the consent again.
