Introduction

Given the significant progress of mankind in various scientific fields, we have now more access to various drugs, so that this easy access has become as a social harmful phenomenon due to the misuse and overuse of drugs (1).
Therefore, the uncontrolled use of drugs and generally self-medication has become a big social, health and economic problem of societies and can cause problems such as an increase in per capita drug consumption, drug resistance, lack of optimal treatment, poisoning, and unwanted drug side effects (1, 2).
According to studies, 65% of diseases are due to non-compliance with proper prescription patterns by physicians and irrational use of drugs (2).
The general self-medication can be done industrial or handmade, and through access to drugs without physician's prescription, using or distributing drugs prescribed for someone among family members and acquaintances, using the additional drugs in the house or self-medication by intake of main drug, either by additional use of drug or without any physician's prescription  or non-full consumption of drug (3.(
According to studies, the prevalence of self-medication has been reported 68% in European countries, 77% in the US, 92% in Kuwait, 31% in India, and 59% in Nepal. (4)
Similar studies indicated that the prevalence of self-medication varied from 12% to 90% in Iran (5); and estimate indicated that the rate of self-medication in Iran is three times more than the global average (6).
Iran is also among the top 20 countries in terms of drug use and is ranked second in Asia after China (7).
Studies on this issue indicate that every Iranian takes 339 drugs annually, that is 2 to 4 times higher than the world standard. The per capita use of injectable drugs is also 4.11 in Iran that is 4 times more than the per capita use of other countries (1).
On the other hand, the self-medication frequency and pattern vary in different countries and regions, depending on socioeconomic status, demographic factors, level of access to medications, advertising of medicinal products, medical history of people, medicines available at home (8) as well as gender (9), especially female sex (10).
In this regard, it is important to pay attention to the population of women because of being in critical times such as pregnancy, lactation, as well as greater contact with family members and being a role model for them (2).
Studies have indicated that women have special tendency to self-medication and usually use medications to treat problems like dysmenorrhea, symptoms of menopause, menstrual disorders, mood disorders, prevention of osteoporosis as well as problems during pregnancy and lactation. All of these factors may underlie self-medication in women (3).
Previous studies indicated the greater self-medication in women, especially young women and those with certain conditions such as pregnancy, single person, illness or chronic health problems, and psychological and emotional problems (11).
Studies also indicate that the self-medication in women accounts for more than 13% of congenital malformations, and increases the probability of preterm labor by 2 to 3 times in pregnant women or leads to abortion and fetal malformations (3).
Given the high prevalence of self-medication in women, and because of their vulnerability, especially at the reproductive age, investigating the self-medication burden in women and identifying its relevant factors can help these individuals and healthcare providers to modify risky behaviors and promote women's health. The present study aimed to determine the frequency and some associated factors of self-medication in women of reproductive age in Gorgan.
 
 
 
 

Methods

This cross-sectional study was conducted in 2018-2019. Inclusion criteria were as follows: women's age of 15 to 49 years, living in Gorgan during the last 6 months, and consent to participate in the study. The incomplete questionnaires were excluded from the study. In the study, the self-medication referred to consumption of drugs without a prescription by a physician or therapist during the last three months, but the use of herbal remedies or traditional products, as well as chemical medications, which were consumed for more than three months after data collection, were not considered as the self-medication.
Sample size was estimated using the following formula. In this formula, the sample size was calculated to be 755 assuming that P (prevalence of self-medication in women) was 0.77 according to previous studies (11), z (confidence level) was 95% and d (error) was equal to 0.03. Due to the power of study, sample size was eventually 800.

Sampling was performed using a randomized multi-stage clustering method. Therefore, Gorgan city was first divided into 50 geographical regions based on the population and 16 individuals were selected through health records from each region. Households were randomly selected in each region, and then all eligible women in the household were included in the study. If one or more eligible households were not present during the data collection at home, data of present individuals were collected and we referred for absent ones at another time (up to 2 times). Due to the uneven number of people living in households, the survey of households in that area was stopped whenever the required 16 individuals were completed in each neighborhood.
Data were collected through home visits and interviews. A checklist with 19 questions was considered as the data collection tool. The first part consisted of 10 questions about demographic characteristics (age, sex, marital status, occupation, household size, number of children, household income, health insurance, and type of insurance) and the second part consisted of 9 questions containing information about using the medication over the past three months, self-medication over the past three months, frequency of self-medication, illnesses or problems leading to self-medication, medication category, causes of self-medication, forms of medication, self-medication information source, and medication supply source.
The checklist was designed according to similar papers (1-6); and its validity was approved by experts in the field of pharmacology, toxicology and social medicine.
Data collection was performed by two graduates of toxicology and pharmacology. It took between 15 to 20 minutes to complete each checklist; and due to easier access and reduced household absences, checklists were completed in the evenings.
Before conducting the study, the protocol was approved in ethics committee of Golestan University of Medical Sciences (IR.GOUMS.REC.1396.271). The participants was ensured about their confidentiality, and then they sign the written consent form before participating in the study.
 Data were analyzed in SPSS statistics for windows, version 16.0 (SPSS Inc., Chicago, III., USA) using descriptive statistics such as frequency, percentage, standard deviation and range of variation. Pearson Chi-square test was used to examine relationships of variables. The significance level of statistical tests was less than 0.05.
 
Results
In the present study, the majority of women were at the age range of 30-39 (47.9%), married (69.6%), high school degree (35.9%), housewives (39.5%), and with one or two children (85.6%). Findings indicated that 68.9% of women were self-medicated during the past three months. The highest rate of self-medication by women at the reproductive age was seen in the age group of 40-49 years (78.5%), divorced women (85.7%), married (73.8%), those with high school degrees (74.6%), and the employee (86.1%). Furthermore, self-medication was more prevalent in women with a household size of 5 and higher (72.3%); and it was more prevalent (86.7%) among women with 4 children and more. On the other hand, women with a monthly household income of more than 3 million RLS (about $214) had higher self-medication (84.6%) than women with lower incomes. Moreover, women who had health insurance had higher self-medication than women without insurance, but the increase was not significant (P=0.179). In the field of self-medication in women of childbearing age according to type of insurance, social security insurance, which had the highest coverage in the statistical population, had the least frequency of self-medication than other insurances (66.4%).There were statistically significant relationships between self-medication and variables such as age, marital status, education, occupation, household size, number of children, monthly household income, and type of health insurance (P<0.05) (Table 1).

Regarding the frequency of self-medication in the past three months, the mean frequency was 3.45±5.17 times; and 11.4% had once self-medication, 27.8% twice, 18% three times, 47 individuals (8.8%) four times, and 189 individuals (34.3%) had five or more Among drugs commonly used in the self-time of self-medication. The headache was the most common cause of illness leading to self-medication (68.6%) in the women, and then the cold was the second leading cause (57.7%) (Table 2).

In terms of causes, according to women's views in the statistical population, 307 cases had self-medication due to previous experience of drug use; and 266 cases had self-medication due to the existence of drugs at home (Figure 1).

In terms of the drug form used in self-medication, most women experienced using pills (95.6%). In this regard, capsule (26.5%), syrup (19.6%), suppository (13.2%), ointment (12.3%), ampoule (11.8%), drops (4.2%) were put in the next ranks; and 7 individuals experienced drug in other forms (1.3%).
Concerning the information sources of women who participated in the study for self-medication, previous drug use experience (55.7%) was the first cause of self-medication. In this regard, the experience obtained from the previous prescription of physician (26.3%), information obtained from pharmacies (25.4%), friends and acquaintances (16.7%), health personnel (14.2%), family (12.9%), internet (5.6%), and other cases (1.5%) were other causes.
In the present study, women who had self-medication during the past three months, were more likely to supply their medications from pharmacies in the city (454 women, 82.4%), and medicines available at home (181 women, 32.8%), pharmacies in hospitals and medical centers (30 women, 5.4%), and other sources (27 women, 4.9%).
Discussion
The present study which aimed to determine the frequency and factors affecting self-medication of women at the reproductive age in Gorgan revealed that 68.9% of women had self-medication during the past three months. The percentage was higher than the results of some similar studies such as research in Yasuj in 2016, Hamadan in 2013 (13), Shahrud in 2016 (14), and Poland in 2013 (15). According to these results and despite the high prevalence of self-medication in some studies such as similar studies in Tehran in 2017 (11), Ahvaz in 2013 (16), and Bandar Abbas in 2016 (1) compared to the present study, the frequency of self-medication in the population of reproductive age in Gorgan can be assumed to be high. A reason for high prevalence of self-medication maybe is that the target population of study was only women and they showed a greater tendency to self-medication (12).
Regarding the age of women with self-medication, the highest prevalence was in those aged 40-49 years. In a study in Tehran, the highest prevalence occurred among women at the age of 31 to 40 years. Another similar study in the north of Iran (17) reported more self-medication over the age of 40 years. In a study in the south of Iran (16), the highest frequency was seen at the age of 15 to 25. Regarding the increasing burden of self-medication at an older age, it is worth noting that the increasing burden of disease and its resulting problems increases the need for medication in people that is fulfilled by self-medication in some cases.
Consistent with the results of a similar study in Tehran, the prevalence of self-medication in women was higher in those with high school and academic degrees. A research in Yasuj in 2016 also found that the individuals were more likely to have high school or academic degrees. Studies in Mazandaran province in 2014 (17) and Nigeria in 2012 (18) also indicated that women with self-medication were more likely to have academic degrees. In general, women with higher education are likely to be able to obtain sufficient information from drug sources or diagnose a disease for the next time after a few prescriptions and take medication using the previously prescribed drugs (12).
Among women with children, the highest self-medication occurred in those with 4 or more children. In a study in Semnan (14) and Markazi (1) provinces, the results were consistent with the present study. Since the higher number of children can make problems for some families to pay for their care, the self-medication has increased among these families. On the other hand, as the number of children increases and subsequently the mother's experience and awareness of diseases and medications increase, families may apply the self-medication (19).
In the present study, it was found that the self-medication was higher among households with higher incomes. A similar study in Ahvaz (16) also confirmed this issue, and it can be argued that either the tendency to use medication is so high that individuals may self-medicate regardless of their income level or the cost of medicines is low in Iran and people can easily afford and use them (1).
An important result of the present study was to identify problems or diseases leading to the self-medication, among which headache and cold were the most important ones. The results were consistent with studies in Rasht (16) and Isfahan (3). Similar study in Sabzevar also indicated that cold was the most important cause of self-medication and then the headache was the second most common cause (20). According to authors in this paper, attention to time and season of research could be also important. For instance, given the higher prevalence of cold and respiratory infections in winter, conducting research in this time range can affect increasing the burden of self-medication due to cold.
Analgesics and then non-steroidal anti-inflammatory drugs have been the most widely used drug class used for self-medication in women of reproductive age. A study in Tabriz (16) also indicated that analgesics were the most important drug category, and then antibiotics were in the second category. In a study in Pakistan (8), 61% of participants were using analgesics for self-medication.
As the leading causes of self-medication from participants' viewpoints, the previous drug use experience and consequently drug availability at home were the most important issues. In Yasuj, the first two causes of self-medication were consistent with the present study (12). In Isfahan, the high cost of visits and previous drug use experience were the most important causes of self-medication (3). In Tabriz, high costs and no feeling of need to visit a physician were the main reasons for participants' self-medication (16). In a study in Mexico, mistrust in physicians and the effectiveness of Traditional medicines were noted by individuals (21).
Limitations of the present study included the self-reported data collection, which might be due to participants' inadequate awareness or lack of confidence in the confidentiality of information that might lead to bias of data collection.
 
Conclusion
The present study indicated that the frequency of self-medication was higher in women of reproductive age in Gorgan than many studies. Furthermore, the prevalence of self-medication was significantly correlated with variables such as age, job, education, marital status, household size, household income level, and type of insurance. Headache and cold were the most important medical causes of self-medication. Analgesics and non-steroidal anti-inflammatory drugs were the most important class of drugs used in the self-medication. In the present study, having previous drug use experience and existence of drugs at home were the most important causes of self-medication from the participants' perspective.
 
Acknowledgments
The present paper was resulted from a research project approved by Golestan University of Medical Sciences with a code 961103263. The authors are grateful to the Health Management and Social Development Research Center and all participants. The authors also declared no conflict of interest among them.