Skip to main content
International Journal of Women's Health logoLink to International Journal of Women's Health
. 2026 Nov 4;18:628836. doi: 10.2147/IJWH.S628836

Factors Associated with Self-Medication Among Pregnant Women Attending Maternal and Child Health Centers in Peru: A Cross-Sectional Study

Stefany Y Ynoñan Lopez 1, Yosef Aroste-Paniora 1, Miguel A Arce-Huamani 1,✉
PMCID: PMC13744729  PMID: 43001177

Abstract

Objective

To determine factors associated with self-medication among pregnant women attending maternal and child health centers in Lima, Peru.

Methods

An analytical cross-sectional study was conducted among 206 pregnant women receiving obstetric or prenatal care at CMI Virgen del Carmen and CMI Buenos Aires de Villa from August 1 to September 30, 2025. Self-medication was defined as medication use without a medical prescription or professional indication. Sociodemographic, obstetric, clinical, and access-related variables were assessed using a structured questionnaire. Crude and adjusted prevalence ratios (PRs) with 95% confidence intervals (CIs) were estimated using Poisson regression with robust variance.

Results

Self-medication prevalence was 35.4%, similar to estimates reported in antenatal-care populations in Brazil and Morocco. Difficulty accessing medical consultation was associated with a higher prevalence of self-medication (adjusted PR = 2.15; 95% CI: 1.50–3.08; p < 0.001). Frequent or intense symptoms during the first months of pregnancy (adjusted PR = 1.62; 95% CI: 1.14–2.29; p = 0.007) and age 30–34 years (adjusted PR = 1.85; 95% CI: 1.13–3.03; p = 0.014) were also associated with self-medication. Travel time and number of prenatal visits were not independently associated.

Conclusion

Self-medication was common in this two-center population and was associated with barriers to medical consultation, symptom burden, and age 30–34 years. These findings support further evaluation of timely consultation pathways and medication-safety counselling in similar maternal healthcare settings.

Keywords: self-medication, pregnancy, prenatal care, maternal health

Introduction

Medication use during pregnancy lies at the intersection of maternal symptom relief, fetal safety, and clinical uncertainty, because medication-safety evidence for pregnant and breastfeeding women remains fragmented and requires stronger systems for monitoring and communication.1,2 The World Health Organization defines self-medication as the selection and use of medicines by individuals for the treatment of self-recognized illnesses or symptoms.3 Self-medication adds a critical layer to this challenge, as global evidence shows that its frequency among pregnant women varies widely across settings, income contexts, and study designs.4 In antenatal-care populations, self-medication may include pharmaceutical products, herbal remedies, or their combined use, often motivated by perceived safety, symptom burden, cost, and easy access to medicines or advice outside formal consultation.5 Recent evidence from institutional antenatal-care settings also emphasizes that unsupervised medication use may expose women to avoidable risks related to drug interactions, inappropriate gestational timing, and insufficient counseling.6 Self-medication during pregnancy is therefore not simply an individual decision, but a measurable expression of how medication safety is understood and communicated within routine maternal care.7

Studies from Brazil and Jordan indicate that self-medication may persist among pregnant women who are already receiving prenatal or outpatient care, particularly when common symptoms are perceived as manageable without professional evaluation.8,9 Evidence from Ethiopia has associated this practice with previous medication use, perceived minor illness, easy access to medicines, informal advice, and difficulties obtaining timely professional care.10–12 A multicenter study from Sierra Leone further emphasized the importance of assessing the specific medicines used and their safety profile,13 whereas recent Ethiopian studies reinforced the need to identify locally relevant factors associated with self-medication within antenatal-care settings.14,15 Taken together, these findings suggest that self-medication during pregnancy is shaped by both individual experience and healthcare-access conditions.

Despite the growing international literature, evidence from Peru remains limited regarding how sociodemographic, obstetric, clinical, and healthcare-access factors are jointly associated with self-medication among pregnant women receiving routine care at reference maternal and child health centers. This gap is relevant because medicine availability, service organization, and access to timely medical consultation may differ across healthcare systems and local settings. Accordingly, this study aimed to determine the factors associated with self-medication among pregnant women attending two reference maternal and child health centers within DIRIS Lima Sur, Peru.

Materials and Methods

Study Design and Setting

We conducted a quantitative, observational, analytical, cross-sectional study among pregnant women attending maternal and child health centers in Lima, Peru. Data collection was conducted from August 1 to September 30, 2025. The study was designed to identify sociodemographic, obstetric, clinical, and access-related factors associated with self-medication during pregnancy.

The study was conducted at two Maternal and Child Health Centers under the jurisdiction of the Dirección de Redes Integradas de Salud Lima Sur: CMI Virgen del Carmen and CMI Buenos Aires de Villa. These facilities were selected because they serve as reference maternal and child health centers within DIRIS Lima Sur and provide obstetric and prenatal care to pregnant women from a broad urban and peri-urban catchment area in southern Lima. Their inclusion allowed the study to capture women receiving care across different sectors of the health network. Participant selection was conducted using proportionate stratified random sampling, with each center treated as a stratum and simple random selection applied within each stratum.

Study Population, Sample Size, and Sampling

The sample size was calculated using OpenEpi for an analytical cross-sectional study, assuming a 95% confidence level, 80% statistical power, and a 1:1 ratio between exposed and unexposed participants. Secondary education was used as the reference exposure based on data reported by Pereira et al.9 The expected prevalence of self-medication was 39% among exposed women and 20% among unexposed women, corresponding to an anticipated odds ratio of approximately 2.5. Using the Fleiss method with continuity correction, the minimum required sample size was 206 pregnant women. To account for potential non-response and incomplete questionnaires, 224 women were randomly selected and invited to participate. Of these, eight declined to participate, six did not attend the scheduled survey, and four submitted incomplete questionnaires, resulting in a final analytical sample of 206 participants (Figure 1).

Figure 1.

Flow diagram showing 412 eligible pregnant women: 246 at CMI Virgen del Carmen and 166 at CMI Buenos Aires de Villa. Of these, 224 were randomly selected and invited to participate: 134 and 90 from the respective centers. Eighteen women were not included: eight declined participation, six did not attend the scheduled survey, and four submitted incomplete questionnaires. The final analytical sample included 206 women: 123 from CMI Virgen del Carmen and 83 from CMI Buenos Aires de Villa.

Participant recruitment flow diagram. Of 412 eligible pregnant women registered at the two participating centers, 224 were randomly selected and invited to participate. Eighteen women were not included in the final analysis: eight declined to participate, six did not attend the scheduled survey, and four submitted incomplete questionnaires. The final analytical sample comprised 206 pregnant women.

Notes: The questionnaire was pilot-tested in 20 pregnant women with characteristics similar to those of the study population. Participants in the pilot test were not included in the final analytical sample.

The sampling frame was constructed separately for each participating center using lists provided by their obstetrics departments. These lists included 412 eligible pregnant women who attended or were scheduled to attend during the study period: 246 at CMI Virgen del Carmen and 166 at CMI Buenos Aires de Villa. The 224 women invited to participate were allocated proportionally between the two centers, corresponding to 134 women from CMI Virgen del Carmen and 90 from CMI Buenos Aires de Villa.

Within each center, every eligible woman was assigned a sequential identification number and selected without replacement using the random-number generator in Microsoft Excel. Therefore, the procedure corresponded to proportionate stratified random sampling, with each center treated as a stratum and simple random selection applied within each stratum. The final analytical sample included 123 women from CMI Virgen del Carmen and 83 from CMI Buenos Aires de Villa.

Eligibility Criteria

Pregnant women were eligible if they were 18 years of age or older, attended obstetric or prenatal care at one of the selected Maternal and Child Health Centers during the study period, agreed to participate voluntarily, signed the informed consent form, and had sufficient oral communication and comprehension capacity to answer the questionnaire.

Women were excluded if they had a diagnosed mental health condition that could interfere with questionnaire completion, were hospitalized due to severe obstetric complications at the time of data collection, voluntarily withdrew from the study, or had incomplete information for the outcome or main explanatory variables.

Variables and Definitions

The main outcome variable was self-medication during pregnancy. Self-medication was defined as the use of medications without medical prescription or professional indication during the current pregnancy. It was assessed as a dichotomous variable and categorized as yes or no.

The explanatory variables were grouped into three domains: sociodemographic factors, obstetric factors, and clinical and access-related factors. Sociodemographic variables included age group, educational level, marital status, occupation, health insurance, monthly family income, and travel time to the health center. Age was categorized as 18–24, 25–29, 30–34, and 35 years or older. Educational level was categorized as primary/no education, secondary, or higher education. Marital status was categorized as single, married/cohabiting, or separated/divorced/widowed. Occupation was categorized as housewife, formal employment, informal employment, or unemployed. Health insurance was categorized as none, Seguro Integral de Salud, EsSalud, or private insurance. Monthly family income was categorized as less than S/1130, S/1130 to S/2000, and more than S/2000. Travel time to the health center was categorized as less than 30 minutes, 30–59 minutes, or 60 minutes or more.

Obstetric variables included gestational trimester, number of pregnancies, and number of prenatal care visits completed at the time of the survey. Gestational trimester was categorized as first, second, or third trimester. Number of pregnancies was categorized as first pregnancy, second pregnancy, or three or more pregnancies. Prenatal care visits were categorized as none, 1–5 visits, or 6 or more visits.

Clinical and access-related variables included use of home remedies, awareness of the risks of self-medication, previous knowledge or use of medications, difficulty accessing medical consultation, frequent or intense symptoms during pregnancy, and perceived safety of over-the-counter medications during pregnancy. Home remedies were defined as non-pharmaceutical preparations made at home from herbal or food-based ingredients and used to relieve pregnancy-related symptoms. Their use was recorded as a dichotomous variable (yes/no); the questionnaire did not collect information on the specific ingredients or formulations used. Frequent or intense symptoms were assessed using a single dichotomous item asking whether the participant had experienced frequent or intense symptoms during the first months of pregnancy, with response options of yes or no. The questionnaire did not identify the specific symptoms experienced or assess their frequency and intensity separately. Awareness of risks, previous knowledge or use of medications, and difficulty accessing consultation were also assessed as dichotomous variables. Perceived safety of over-the-counter medications was categorized as no, yes, or unsure.

Data Collection Procedures

Data were collected through structured, face-to-face surveys administered individually to pregnant women during their attendance at the selected Maternal and Child Health Centers. The questionnaires were administered directly by two members of the research team, Stefany Y Ynoñan Lopez and Yosef Aroste-Paniora. Before applying the questionnaire, each participant received information about the study objective, procedures, voluntary nature of participation, confidentiality safeguards, and absence of direct clinical risk. Written informed consent was obtained before data collection.

The study instrument was a structured questionnaire adapted from a previous study on the prevalence and factors associated with self-medication during the first trimester of pregnancy among women attending health centers in Lima Este.16 Because the source questionnaire was developed in Spanish and applied to pregnant women in Lima, translation or formal cross-cultural adaptation was not required. Nevertheless, the wording of the items was reviewed for clarity and contextual appropriateness for the participating maternal and child health centers. The questionnaire included 16 closed-ended items distributed across three domains: sociodemographic factors, obstetric factors, and clinical and access-related factors. Self-medication during pregnancy was recorded as a dichotomous outcome. The questionnaire did not collect the names, therapeutic classes, doses, routes of administration, or trimester-specific exposure of the medications used because the prespecified objective was to estimate the prevalence of self-medication and examine its associated factors.

The adapted questionnaire underwent content validation by three experts with backgrounds in medicine, public health, and gynecology. Each expert independently assessed the pertinence, relevance, and clarity of the items, and the recommended revisions were incorporated before data collection. Internal consistency was evaluated for the applicable seven-item subscale using a standardized Cronbach’s alpha based on Spearman correlations. The coefficient was 0.933 (95% CI: 0.890–0.964). Formal construct validity was not assessed because the questionnaire was designed primarily as a structured data-collection instrument comprising distinct sociodemographic, obstetric, clinical, and access-related variables rather than as a psychometric scale measuring a single latent construct. Measurement validity was therefore evaluated through expert-based content review and pilot testing. Before the main data collection, the questionnaire was pilot-tested in 20 pregnant women with characteristics similar to those of the study population. The pilot test assessed item comprehension, the time required to complete the questionnaire, and potential difficulties during its administration. Women who participated in the pilot test were not included in the final analytical sample of 206 participants. Completed questionnaires were reviewed by the two researchers for completeness before data entry. The database was then coded, cleaned, and checked for inconsistencies prior to statistical analysis.

Statistical Analysis

Categorical variables were summarized using absolute and relative frequencies. The prevalence of self-medication was estimated as the proportion of pregnant women who reported self-medication during pregnancy among the total analytical sample. Missing data were handled using complete-case analysis. Participants with incomplete information for the outcome or the main explanatory variables were excluded before analysis. All 206 women included in the final analytical sample had complete data for the variables reported in the descriptive, bivariate, and multivariable analyses; therefore, no missing-data imputation was performed.

Bivariate analyses were performed to evaluate the association between self-medication and the explanatory variables. Categorical variables were compared using Pearson’s chi-square test or Fisher’s exact test when expected cell frequencies were low. Variables for the adjusted model were selected using a conceptual epidemiological framework informed by prior evidence and directed acyclic graph principles, rather than solely on the basis of statistical significance in the bivariate analysis.17–19 Age was included as a sociodemographic factor; travel time to the health center and number of prenatal visits represented geographic access and healthcare utilization, respectively; difficulty accessing medical consultation represented perceived organizational access; and frequent or intense symptoms during the first months of pregnancy represented clinical need. All selected variables were entered simultaneously into the adjusted model. Interaction terms were not examined because no effect-modification hypotheses had been prespecified and the available sample size, together with the small number of participants in some categories, could have produced unstable interaction estimates. Multicollinearity was assessed using generalized variance inflation factors adjusted for the degrees of freedom of each model term. Bivariate p values were reported descriptively and were not used as the sole criterion for variable inclusion.

Crude and adjusted prevalence ratios with 95% confidence intervals were estimated using Poisson regression models with robust variance. Self-medication during pregnancy was considered the dependent variable. The adjusted model included age group, travel time to the health center, number of prenatal care visits, difficulty accessing medical consultation, and frequent or intense symptoms during the first months of pregnancy. Reference categories were 18–24 years for age, less than 30 minutes for travel time, no prenatal care visits, no difficulty accessing consultation, and absence of frequent or intense symptoms during the first months of pregnancy.

A two-sided p value < 0.05 was considered statistically significant. Goodness of fit of the final adjusted model was evaluated using the deviance and Pearson chi-square statistics relative to their degrees of freedom, together with the Akaike information criterion and Bayesian information criterion. Ratios of the deviance and Pearson chi-square statistics to their respective degrees of freedom were examined to assess potential overdispersion. All statistical analyses were performed using IBM SPSS Statistics version 26.

Ethical Considerations

The study was conducted in accordance with the ethical principles of the Declaration of Helsinki and applicable Peruvian regulations for health research. The protocol was reviewed and approved by the Institutional Committee of Ethics and Scientific Integrity of Universidad Privada Norbert Wiener, Lima, Peru, under approval file No. 1743–2025. The approved protocol was titled “Factores asociados a la automedicación en gestantes atendidas en Centros Materno Infantiles Peruanos, 2025”. The study protocol was not published or registered in a publicly accessible repository.

Authorization for fieldwork was also granted by the Dirección de Redes Integradas de Salud Lima Sur, which approved the development of the study at CMI Virgen del Carmen and CMI Buenos Aires de Villa. All participants provided written informed consent before completing the questionnaire. Participation was voluntary, and women were informed that they could withdraw at any time without consequences for their medical care.

The study involved minimal risk because no intervention was performed and data were obtained through questionnaires. Confidentiality was maintained throughout the research process. Each participant was assigned a coded identifier, and no directly identifying personal information was included in the analytical database. Access to the database was restricted to the research team, and results were reported only in aggregate form.

Results

Characteristics of the Study Sample

A total of 206 pregnant women were included in the analysis, of whom 73 (35.4%) reported self-medication during pregnancy. Most participants were aged 18–24 years, were married or cohabiting, were homemakers, and were covered by Comprehensive Health Insurance. More than half had completed six or more prenatal visits. Difficulty accessing medical consultation was reported by 80 women (38.8%), while 99 (48.1%) reported frequent or intense symptoms during the first months of pregnancy. The specific symptoms experienced were not recorded in the questionnaire. Additional sociodemographic, obstetric, clinical, and access-related characteristics are presented in Table 1.

Table 1.

Characteristics of the Study Sample

Variable N (%)
Self-medication
 No 133 (64.6)
 Yes 73 (35.4)
Sociodemographic factors
 Age, years
  18–24 69 (33.5)
  25–29 52 (25.2)
  30–34 42 (20.4)
  35 or older 43 (20.9)
 Educational level
  Primary/no formal education 34 (16.5)
  Secondary 97 (47.1)
  Higher education 75 (36.4)
 Marital status
  Single 55 (26.7)
  Married/cohabiting 134 (65.0)
  Separated/divorced/widowed 17 (8.3)
 Occupation
  Homemaker 95 (46.1)
  Formal employment 46 (22.3)
  Informal employment 47 (22.8)
  Unemployed 18 (8.7)
 Health insurance
  None 45 (21.8)
  Comprehensive Health Insurance (SIS) 106 (51.5)
  EsSalud 49 (23.8)
  Private insurance 6 (2.9)
 Family income
  Less than PEN 1130 79 (38.3)
  PEN 1130 to PEN 2000 81 (39.3)
  More than PEN 2000 46 (22.3)
 Travel time, minutes
  Less than 30 108 (52.4)
  30–59 71 (34.5)
  60 or more 27 (13.1)
Obstetric factors
 Gestational trimester
  First trimester 50 (24.3)
  Second trimester 76 (36.9)
  Third trimester 80 (38.8)
 Number of pregnancies
  First 68 (33.0)
  Second 54 (26.2)
  Third or more 84 (40.8)
 Prenatal visits to date
  None 11 (5.3)
  1 to 5 77 (37.4)
  6 or more 118 (57.3)
Clinical and access-related factors
 Use of home remedies
  No 134 (65.0)
  Yes 72 (35.0)
 Awareness of the risks of self-medication
  No 59 (28.6)
  Yes 147 (71.4)
 Knowledge/prior use of medications
  No 121 (58.7)
  Yes 85 (41.3)
 Difficulty accessing medical consultation
  No 126 (61.2)
  Yes 80 (38.8)
 Frequent/intense symptoms during pregnancy
  No 107 (51.9)
  Yes 99 (48.1)
 Believes that OTC medicines are safe during pregnancy
  No 99 (48.1)
  Yes 61 (29.6)
  Unsure 46 (22.3)

Notes: Data are presented as frequencies and percentages. Percentages may not sum to 100% because of rounding. PEN refers to Peruvian soles.

Abbreviations: SIS, Comprehensive Health Insurance; OTC, over-the-counter.

Bivariate Analysis of Self-Medication by Subgroup

In the bivariate analysis, self-medication differed significantly by age group (p = 0.025). Among women who self-medicated, 21 (28.8%) were aged 30–34 years and 19 (26.0%) were aged 35 years or older, compared with 21 (15.8%) and 24 (18.0%), respectively, among women who did not self-medicate. Difficulty accessing medical consultation was reported by 43 women (58.9%) in the self-medication group and 37 (27.8%) in the non-self-medication group (p < 0.001). Frequent or intense symptoms during the first months of pregnancy were reported by 43 women (58.9%) and 56 (42.1%), respectively (p = 0.021). No statistically significant differences were observed for the remaining sociodemographic, obstetric, clinical, or access-related variables (Table 2).

Table 2.

Characteristics of Participants by Self-Medication Status (Bivariate Analysis)

Variable No Self-Medication
n = 133, n (%)
Self-Medication
n = 73, n (%)
p-value
Sociodemographic factors
 Age, years 0.025
  18–24 52 (39.1) 17 (23.3)
  25–29 36 (27.1) 16 (21.9)
  30–34 21 (15.8) 21 (28.8)
  35 or older 24 (18.0) 19 (26.0)
 Educational level 0.876
  Primary/no formal education 22 (16.5) 12 (16.4)
  Secondary 61 (45.9) 36 (49.3)
  Higher education 50 (37.6) 25 (34.2)
 Marital status 0.873
  Single 36 (27.1) 19 (26.0)
  Married/cohabiting 87 (65.4) 47 (64.4)
  Separated/divorced/widowed 10 (7.5) 7 (9.6)
 Occupation 0.905
  Homemaker 60 (45.1) 35 (47.9)
  Formal employment 30 (22.6) 16 (21.9)
  Informal employment 30 (22.6) 17 (23.3)
  Unemployed 13 (9.8) 5 (6.8)
 Health insurance 0.308*
  None 29 (21.8) 16 (21.9)
  Comprehensive Health Insurance (SIS) 63 (47.4) 43 (58.9)
  EsSalud 36 (27.1) 13 (17.8)
  Private insurance 5 (3.8) 1 (1.4)
 Family income 0.435
  Less than PEN 1130 55 (41.4) 24 (32.9)
  PEN 1130 to PEN 2000 51 (38.3) 30 (41.1)
  More than PEN 2000 27 (20.3) 19 (26.0)
 Travel time, minutes 0.168
  Less than 30 71 (53.4) 37 (50.7)
  30–59 41 (30.8) 30 (41.1)
  60 or more 21 (15.8) 6 (8.2)
Obstetric factors
 Gestational trimester 0.232
  First trimester 37 (27.8) 13 (17.8)
  Second trimester 45 (33.8) 31 (42.5)
  Third trimester 51 (38.3) 29 (39.7)
 Number of pregnancies 0.512
  First 41 (30.8) 27 (37.0)
  Second 34 (25.6) 20 (27.4)
  Third or more 58 (43.6) 26 (35.6)
 Prenatal visits to date 0.132
  None 6 (4.5) 5 (6.8)
  1 to 5 44 (33.1) 33 (45.2)
  6 or more 83 (62.4) 35 (47.9)
Clinical and access-related factors
 Use of home remedies 0.650
  No 88 (66.2) 46 (63.0)
  Yes 45 (33.8) 27 (37.0)
 Awareness of the risks of self-medication 0.208
  No 42 (31.6) 17 (23.3)
  Yes 91 (68.4) 56 (76.7)
 Knowledge/prior use of medications 0.251
  No 82 (61.7) 39 (53.4)
  Yes 51 (38.3) 34 (46.6)
 Difficulty accessing medical consultation <0.001
  No 96 (72.2) 30 (41.1)
  Yes 37 (27.8) 43 (58.9)
 Frequent/intense symptoms during pregnancy 0.021
  No 77 (57.9) 30 (41.1)
  Yes 56 (42.1) 43 (58.9)
 Believes that OTC medicines are safe during pregnancy 0.646
  No 67 (50.4) 32 (43.8)
  Yes 37 (27.8) 24 (32.9)
  Unsure 29 (21.8) 17 (23.3)

Notes: Data are presented as frequencies and column percentages according to self-medication status. P-values were estimated using Pearson’s chi-square test, except for health insurance (*), for which Fisher’s exact test was used. Percentages may not sum to 100% because of rounding.

Abbreviations: SIS, Comprehensive Health Insurance; OTC, over-the-counter.

Factors Associated with Self-Medication in the Multivariable Model

In the adjusted Poisson regression model, difficulty accessing medical consultation was associated with a higher prevalence of self-medication (adjusted PR: 2.15; 95% CI: 1.50–3.08; p < 0.001). Frequent or intense symptoms during the first months of pregnancy were also associated with self-medication (adjusted PR: 1.62; 95% CI: 1.14–2.29; p = 0.007). Compared with women aged 18–24 years, those aged 30–34 years had a higher prevalence of self-medication (adjusted PR: 1.85; 95% CI: 1.13–3.03; p = 0.014). No statistically significant associations were observed for the other age groups, travel time to the health center, or number of prenatal visits (Table 3).

Table 3.

Factors Associated with Self-Medication (Multivariable Analysis): Crude and Adjusted Prevalence Ratios

Variables Crude PR (95% CI) p-value Adjusted PR (95% CI) p-value
Sociodemographic factors
 Age, years
  18–24 Reference Reference
  25–29 1.25 (0.70–2.23) 0.454 1.15 (0.67–1.98) 0.612
  30–34 2.03 (1.22–3.39) 0.007 1.85 (1.13–3.03) 0.014
  35 or older 1.79 (1.05–3.06) 0.032 1.54 (0.95–2.52) 0.083
 Travel time, minutes
  Less than 30 Reference Reference
  30–59 1.23 (0.85–1.80) 0.277 1.17 (0.82–1.67) 0.389
  60 or more 0.65 (0.31–1.38) 0.261 0.80 (0.41–1.57) 0.513
 Prenatal visits to date
  None Reference Reference
  1 to 5 0.94 (0.47–1.90) 0.869 0.96 (0.52–1.77) 0.896
  6 or more 0.65 (0.32–1.32) 0.236 0.74 (0.39–1.40) 0.357
 Difficulty accessing medical consultation
  No Reference Reference
  Yes 2.26 (1.55–3.28) <0.001 2.15 (1.50–3.08) <0.001
 Frequent/intense symptoms during pregnancy
  No Reference Reference
  Yes 1.55 (1.06–2.26) 0.023 1.62 (1.14–2.29) 0.007

Notes: Crude and adjusted prevalence ratios were estimated using Poisson regression models with robust variance. The 18–24 years age group, travel time of less than 30 minutes, absence of prenatal visits, no difficulty accessing medical consultation, and absence of frequent/intense symptoms during pregnancy were used as reference categories, as applicable. Adjusted estimates include all variables shown in the multivariable model. Statistically significant associations were defined as p < 0.05. Boldface indicates statistically significant adjusted prevalence ratios (p < 0.05).

Abbreviations: PR, prevalence ratio; CI, confidence interval.

The final adjusted model yielded a deviance of 128.497 with 196 degrees of freedom (deviance/df = 0.656) and a Pearson chi-square value of 133.504 with 196 degrees of freedom (Pearson chi-square/df = 0.681). The Akaike information criterion was 294.497 and the Bayesian information criterion was 327.776. Both dispersion ratios were below 1, indicating no evidence of overdispersion. No evidence of problematic multicollinearity was observed. The adjusted generalized variance inflation factors ranged from 1.02 to 1.04, with a maximum value of 1.04 for number of prenatal visits.

Discussion

In this cross-sectional study, more than one-third of the pregnant women reported self-medication. After adjustment, difficulty accessing medical consultation, frequent or intense symptoms during the first months of pregnancy, and age 30–34 years were associated with a higher prevalence of self-medication. In contrast, travel time to the health center and number of prenatal visits were not independently associated with the outcome. These findings should be interpreted as associations and do not establish temporal or causal relationships. Moreover, because the study included 206 pregnant women attending two reference maternal and child health centers within DIRIS Lima Sur, the findings should be interpreted primarily within similar urban and peri-urban primary-care settings and should not be extrapolated to all pregnant women in Peru.

The prevalence observed in our study (35.4%) was close to that reported in Brazil (36.0%)9 and Morocco (32.0%),20 but lower than the global pooled estimate of 44.5%4 and the 74.1% reported in Ghana.21 Conversely, it exceeded estimates from a Peruvian national maternal institute (10.5%)22 and Indonesian primary care services (11.7%).23 These differences are unlikely to reflect a single underlying factor. Similarities with Brazil and Morocco may partly relate to the inclusion of facility-based antenatal populations, whereas discrepancies with Ghana, Indonesia, and the earlier Peruvian study may reflect differences in outcome definitions, recall periods, inclusion of herbal products or supplements, pharmacy regulation, and healthcare setting. In particular, the previous Peruvian study assessed self-prescribing rather than the broader construct of self-medication used in the present study. Accordingly, direct numerical comparisons should be made cautiously, and our estimate should be interpreted within the organization of maternal care in southern Lima.

The most pronounced association observed in our study was between difficulty accessing medical consultation and self-medication. This finding is consistent with evidence from Ethiopia, where Niriayo et al10 reported that easy access to medicines, perceiving an illness as minor, and saving time were common reasons for self-medication. Similarly, Adama et al21 found that access to medicines without a prescription was associated with self-medication in Ghana, while Tujuba et al24 reported that living close to a healthcare facility was associated with lower odds of self-medication. In Nigeria, Edet et al25 identified cost, distance, medicine unavailability, and delays in receiving care as reported reasons for this practice. Collectively, these findings indicate that barriers to timely professional care may be relevant to self-medication decisions. However, the cross-sectional design of our study does not establish that difficulty accessing consultation caused self-medication. The observed association supports evaluating timely maternal consultation pathways together with medication-safety counselling in prospective studies.

Frequent or intense symptoms during the first months of pregnancy were associated with a higher prevalence of self-medication in our study. However, the questionnaire did not identify the specific symptoms experienced. Previous studies provide relevant context: Pereira et al9 identified headache as the most frequent symptom among pregnant women who self-medicated, while Niriayo et al10 reported morning sickness, headache, and upper respiratory tract infections as common indications. Chergaoui et al20 identified nausea or vomiting, headache, heartburn, and back or leg pain, whereas Tujuba et al24 reported pregnancy-related discomforts as a frequent reason for self-medication. Although these findings suggest that common pregnancy-related symptoms may contribute to unsupervised medication use, our data do not allow us to determine which specific symptoms were associated with self-medication. Future studies should assess symptom type, frequency, and intensity separately.

The association observed in women aged 30–34 years suggests that self-medication may increase among women with greater reproductive or medication-use experience. This interpretation is supported by Chergaoui et al,20 who found that women older than 30 years were more likely to self-medicate (AOR = 4.19), and by Atmadani et al,23 who reported that older pregnant women aged 28–45 years had higher odds of self-medication (aOR = 2.14). The Moroccan study also found a strong association with multiparity (AOR = 6.03), suggesting that previous pregnancy experience may create confidence in managing symptoms without professional consultation. However, this age pattern is not universal. Edet et al25 reported higher self-medication among women younger than 25 years in Nigeria, indicating that age may operate differently depending on health literacy, service access, and prior interactions with the health system. In our setting, the association with women in their early thirties may reflect accumulated experience with pregnancy symptoms and medicines, but also possible normalization of self-care when consultations are difficult. This finding highlights the need for counselling strategies that address not only younger pregnant women, but also adult women who may perceive themselves as experienced enough to self-manage symptoms.

Travel time was not independently associated with self-medication after adjustment in our study. This contrasts with findings from rural Ethiopia, where Tujuba et al24 reported that living close to health facilities substantially reduced the likelihood of self-medication (AOR = 0.23), and with the Nigerian study by Edet et al,25 where distance to health facilities was described as one of the reasons for self-medication. This difference may be explained by the urban or peri-urban nature of maternal and child health centers in our sample, where physical distance may be less decisive than administrative or organizational barriers. In this context, the relevant barrier may not be reaching the facility, but obtaining timely attention once symptoms appear. This interpretation is consistent with the stronger role of difficulty accessing consultation in our study. Therefore, the absence of an adjusted association with travel time does not weaken the access hypothesis; rather, it refines it by suggesting that service availability, appointment responsiveness, and waiting times may matter more than geographic distance alone.

The number of prenatal controls was not significantly associated with self-medication after adjustment. This finding should be interpreted carefully, because evidence from other settings suggests that antenatal care can reduce self-medication only when it includes effective counselling. In Morocco, Chergaoui et al20 found that previous antenatal follow-up was associated with lower odds of self-medication in adjusted models, while Ahmed et al6 reported that women who did not receive health education about self-medication during antenatal care were more likely to self-medicate (AOR = 1.50). Atmadani et al23 also showed that knowledge of possible medication-related risks was associated with lower self-medication (aOR = 0.29). These findings suggest that the protective value of antenatal care depends less on the number of visits and more on the quality, specificity, and recall of counselling received. In our population, more prenatal controls may not necessarily have been translated into clearer guidance on how to respond to headache, nausea, pain, urinary symptoms, or respiratory symptoms. This result supports the need to integrate structured medication-safety counselling into routine prenatal care.

Education, marital status, occupation, income, and insurance were not significantly associated with self-medication in our study. This differs from several studies where socioeconomic variables were relevant. Pereira et al9 found higher odds of self-medication among women with high-school and university education, whereas Edet et al25 reported that marital status and education were significant predictors in Nigeria. Niriayo et al10 also found that absence of health insurance was associated with self-medication (AOR = 2.75). However, Moroccan data from Chergaoui et al20 showed no significant association with education in bivariate analysis, suggesting that socioeconomic patterns are not consistent across settings. In our sample, the lack of association may indicate that self-medication crossed conventional social categories and was more strongly shaped by immediate clinical need and difficulty obtaining consultation. This interpretation is important because it prevents reducing self-medication to a problem of low education or poverty alone. From a public-health perspective, interventions should be universal within prenatal care, while still being sensitive to groups with specific access barriers.

Use of home remedies, awareness of risks, previous knowledge or use of medications, and beliefs about the safety of over-the-counter medications were not associated with self-medication in the bivariate analysis. This contrasts with Atmadani et al,23 who found that higher general knowledge of over-the-counter medicines increased self-medication (aOR = 2.14), while correct knowledge of possible risks reduced it (aOR = 0.29). In Croatia, Jovanović and Vulić7 reported a high prevalence of self-medication (69.5%) despite most women recognizing that pregnant women should consult a doctor before using over-the-counter medicines and that such medicines may pose fetal risk. Similarly, Miní et al22 found that 64.5% of Peruvian pregnant women believed self-prescribing could cause congenital malformations, yet self-prescribing still occurred, and all women who self-prescribed had done so before pregnancy. These findings suggest that awareness of risk does not always translate into safer behavior when symptoms are frequent and access to consultation is difficult. Thus, medication-safety education must be paired with accessible, rapid, and trusted maternal-care pathways.

From a maternal healthcare perspective, the observed associations support evaluating specific interventions within similar primary-care settings. These may include brief medication-safety counselling during antenatal visits, pharmacist-led guidance for pregnant women seeking over-the-counter medicines, rapid consultation pathways for common pregnancy-related symptoms, and community-based educational activities addressing the risks of unsupervised medication use. Coordination between maternal health services and community pharmacies may also facilitate timely referral when pregnant women seek medicines without professional indication. However, the effectiveness of these strategies cannot be established from the present cross-sectional study and should be assessed in prospective or intervention-based research.

The main strength of this study is its analytical approach to self-medication during pregnancy, moving beyond prevalence estimation to identify potentially modifiable factors that can be addressed within maternal health services. The use of adjusted prevalence ratios allowed a more clinically interpretable estimation of associations in a cross-sectional design, while the inclusion of access-related, symptomatic, sociodemographic, prenatal-care, and knowledge-related variables provided a broad assessment of potentially associated factors.

Nevertheless, several limitations should be acknowledged. The cross-sectional design does not allow temporal or causal inferences, and self-medication was assessed through self-report, which may be affected by recall or social desirability bias. In addition, formal construct validity was not evaluated, as the questionnaire primarily comprised separate categorical variables rather than a psychometric scale. The closed-ended questionnaire assessed frequent or intense symptoms and difficulty accessing medical consultation using single dichotomous items. It did not identify the specific symptoms experienced, evaluate symptom frequency and intensity separately, or distinguish the different dimensions of healthcare access. Residual confounding also remains possible because parity, previous pregnancy complications, chronic comorbidities, maternal health literacy, medication availability, and pharmacy accessibility were not assessed. In addition, the sample size and the small number of participants in some categories, particularly those with private insurance or no prenatal visits, may have limited the precision of several estimates.

Although the participating facilities are reference maternal and child health centers serving a broad catchment area within DIRIS Lima Sur, the study was limited to two centers in southern Lima, which restricts the generalizability of the findings to other populations and healthcare settings. Finally, because the questionnaire was designed to estimate the prevalence of self-medication and examine its associated factors, it did not collect the names, therapeutic classes, doses, routes of administration, or trimester-specific exposure of the medications used. Consequently, medication-specific safety patterns and clinical outcomes could not be evaluated.

Conclusion

Self-medication was reported by 35.4% of pregnant women attending two reference maternal and child health centers in southern Lima. Difficulty accessing medical consultation, frequent or intense symptoms during the first months of pregnancy, and age 30–34 years were independently associated with a higher prevalence of self-medication. Travel time to the health center and number of prenatal visits were not independently associated with the outcome. Given the cross-sectional design and the sample of 206 women from two centers, these findings do not establish causal relationships and should be interpreted within similar urban and peri-urban maternal healthcare settings. The results support further evaluation of timely consultation pathways and medication-safety counselling during prenatal care. Prospective multicenter studies are needed to confirm these associations and assess medication-specific exposures and maternal or fetal outcomes.

Acknowledgments

The authors thank the Dirección de Redes Integradas de Salud Lima Sur, CMI Virgen del Carmen, and CMI Buenos Aires de Villa for facilitating the conduct of this study. The authors also thank the pregnant women who voluntarily participated and the research team members involved in data collection, database review, and quality control.

Funding Statement

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Patient and Public Involvement

Patients and/or the public were not involved in the design, conduct, reporting, or dissemination plans of this study. However, pregnant women attending obstetric or prenatal care at the selected maternal and child health centers participated voluntarily as study respondents after providing written informed consent.

Use of AI and AI-Assisted Tools

ChatGPT (GPT-5.6 SoL; OpenAI) was used after completion of the initial manuscript draft solely to assist with English-language editing, grammar, clarity, stylistic refinement, and the wording of responses to editorial and reviewer comments. The tool was not used to design the study, collect data, perform statistical analyses, or independently determine the scientific interpretation of the findings. All manuscript content, statistical results, citations, and references were independently reviewed and verified by the authors. The authors confirm the originality and accuracy of the final manuscript and take full responsibility for the integrity and accuracy of its entire content, including all references. The authors reviewed the applicable terms of use for ChatGPT and confirm that its use was appropriate for this publication.

Abbreviations

CI, Confidence interval; CIEIC-UPNW, Institutional Committee of Ethics and Scientific Integrity of Universidad Privada Norbert Wiener; CMI, Maternal and Child Health Center; DIRIS Lima Sur, Directorate of Integrated Health Networks of South Lima; IQR, Interquartile range; OTC, Over-the-counter; PEN, Peruvian sol; PR, Prevalence ratio; SD, Standard deviation; SIS, Comprehensive Health Insurance; SPSS, Statistical Package for the Social Sciences.

Data Sharing Statement

The de-identified dataset and analytic syntax are available from the corresponding author upon reasonable request, subject to institutional approval and applicable ethical restrictions.

Ethics Approval and Informed Consent

This study was conducted in accordance with the ethical principles of the Declaration of Helsinki and applicable Peruvian regulations for health research. The study protocol was reviewed and approved by the Institutional Committee of Ethics and Scientific Integrity of Universidad Privada Norbert Wiener (CIEIC-UPNW), Lima, Peru, under approval file No. 1743–2025. The approved protocol was titled “Factores asociados a la automedicación en gestantes atendidas en Centros Materno Infantiles Peruanos, 2025”. Authorization for fieldwork was also granted by the Dirección de Redes Integradas de Salud Lima Sur, which approved the conduct of the study at CMI Virgen del Carmen and CMI Buenos Aires de Villa. All participants received information about the study objectives, procedures, voluntary nature of participation, confidentiality safeguards, absence of direct clinical risk, and their right to withdraw at any time without affecting their medical care. Written informed consent was obtained from all participants before questionnaire administration. The study involved minimal risk because no intervention was performed and data were collected through structured questionnaires during routine obstetric or prenatal care. Participant confidentiality was protected by assigning coded identifiers, excluding direct personal identifiers from the analytical database, restricting access to the research team, and reporting all findings only in aggregate form.

Consent for Publication

Not applicable. This manuscript does not contain any individual person’s identifiable data, images, clinical details, or personal information.

Disclosure

The authors declare that they have no conflicts of interest in this work.

References

  • 1.Thurin NH, Pajouheshnia R, Roberto G, et al. From inception to ConcePTION: genesis of a network to support better monitoring and communication of medication safety during pregnancy and breastfeeding. Clin Pharmacol Ther. 2022;111(1):321–14. doi: 10.1002/cpt.2476 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Winterfeld U, Hodson K, Berlin M, et al. Communicating medication risks in pregnancy: towards shared decision making. Front Drug Saf Regul. 2025;5:1712216. doi: 10.3389/fdsfr.2025.1712216 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.World Health Organization. Guidelines for the regulatory assessment of medicinal products for use in self-medication. 2000. Available from: https://iris.who.int/items/468d79f6-9788-4296-9bc9-68014166373d. Accessed July 30, 2026.
  • 4.Bouqoufi A, Laila L, Boujraf S, et al. Prevalence and associated factors of self-medication in worldwide pregnant women: systematic review and meta-analysis. BMC Public Health. 2024;24(1):308. doi: 10.1186/s12889-023-17195-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Almubarak MM, Ibrahim EEM, Ali SA, et al. Prevalence and determinants of self-medication among pregnant women at an antenatal clinic at Soba Teaching Hospital, Sudan: a cross-sectional analysis. Med Sci Monit. 2024;30:e945711. doi: 10.12659/MSM.945711 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Ahmed HN, Damtew BS, Bezabih WA, Wakwoya EB, Abdi HB, Tiruye G. Self-medication practices and associated factors among pregnant women in Oromia, Ethiopia: institutional based cross-sectional study. Front Med. 2025;12:1489990. doi: 10.3389/fmed.2025.1489990 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Jovanović Ž, Vulić P. Self-medication and knowledge of pregnant women about the use of medication during pregnancy in the cities of Rijeka and Zadar, Croatia. Front Pharmacol. 2025;16:1536050. doi: 10.3389/fphar.2025.1536050 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Alsous MM, Al-Azzam SI, Nusair MB, Alnahar SA, Obeidat NA. Self-medication among pregnant women attending outpatients’ clinics in northern Jordan-a cross-sectional study. Pharmacol Res Perspect. 2021;9(2):e00735. doi: 10.1002/prp2.735 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Pereira G, Surita FG, Ferracini AC, Madeira CDS, Oliveira LS, Mazzola PG. Self-medication among pregnant women: prevalence and associated factors. Front Pharmacol. 2021;12:659503. doi: 10.3389/fphar.2021.659503 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Niriayo YL, Mohammed K, Asgedom SW, Demoz GT, Wahdey S, Gidey K. Self-medication practice and contributing factors among pregnant women. PLoS One. 2021;16(5):e0251725. doi: 10.1371/journal.pone.0251725 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Tuha A, Faris AG, Mohammed SA, Gobezie MY. Self-medication and associated factors among pregnant women attending antenatal care at Kemisie general hospital, North East Ethiopia. Patient Prefer Adherence. 2020;14:1969–1978. doi: 10.2147/PPA.S277098 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Girmaw F, Sendekie AK, Mesfin B, Kassaw AT. Self-medication practices among pregnant women in Ethiopia. J Pharm Policy Pract. 2023;16(1):74. doi: 10.1186/s40545-023-00584-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Abiri OT, Lawal S, Coker J, et al. Assessment of self-medication practices and safety profile of medicines utilisation among pregnant women attending antenatal clinics in Freetown, Sierra Leone: a multicentre cross-sectional study. J Pharm Policy Pract. 2024;17(1):2380874. doi: 10.1080/20523211.2024.2380874 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Abebe M, Gashaw S, Getahun D, Wassu W, Menshaw T, Lakew S. Prevalence and associated factors of self-medication among pregnant women in Sodo Town, Southern Ethiopia. Front Med. 2024;11:1379706. doi: 10.3389/fmed.2024.1379706 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Demis A, Altaye BM, Emiru M, Tefera M. Prevalence of self-medication practice and associated factors among pregnant women who attended antenatal care at public hospitals of North Shewa Zone, amhara region, Ethiopia. Adv Pharmacol Pharm Sci. 2024;2024(1):6668480. doi: 10.1155/2024/6668480 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Mendoza Orcon KB, Saldaña Cunibertti WP. Prevalencia y factores asociados a automedicación durante el primer trimestre del embarazo en gestantes atendidas en tres centros de salud de Lima Este, setiembre 2024 [undergraduate thesis]. Lima, Peru: Universidad Ricardo Palma; 2025. [Google Scholar]
  • 17.Tennant PWG, Murray EJ, Arnold KF, et al. Use of directed acyclic graphs (DAGs) to identify confounders in applied health research: review and recommendations. Int J Epidemiol. 2021;50(2):620–632. doi: 10.1093/ije/dyaa213 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Digitale JC, Martin JN, Glymour MM. Tutorial on directed acyclic graphs. J Clin Epidemiol. 2022;142:264–267. doi: 10.1016/j.jclinepi.2021.08.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Correia LCL, Mascarenhas RF, De Menezes FSC, et al. Confounder selection in observational studies in high-impact medical and epidemiological journals. JAMA Netw Open. 2025;8(7):e2524176. doi: 10.1001/jamanetworkopen.2025.24176 [DOI] [PubMed] [Google Scholar]
  • 20.Chergaoui S, Changuiti O, Marfak A, Saad E, Hilali A, Youlyouz Marfak I. Modern drug self-medication and associated factors among pregnant women at settat city, Morocco. Front Pharmacol. 2022;13:812060. doi: 10.3389/fphar.2022.812060 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Adama S, Wallace LJ, Arthur J, Kwakye S, Adongo PB. Self-medication practices of pregnant women attending antenatal clinic in northern Ghana: an analytical cross-sectional study. Afr J Reprod Health. 2021;25(4):89–98. doi: 10.29063/ajrh2021/v25i4.10 [DOI] [PubMed] [Google Scholar]
  • 22.Miní E, Varas R, Vicuña Y, et al. Automedicación en gestantes que acuden al Instituto Nacional Materno Perinatal, Perú 2011. Rev Peru Med Exp Salud Pública. 2012;29(2):212–217. doi: 10.17843/rpmesp.2012.292.343 [DOI] [PubMed] [Google Scholar]
  • 23.Atmadani RN, Nkoka O, Yunita SL, Chen YH. Self-medication and knowledge among pregnant women attending primary healthcare services in Malang, Indonesia: a cross-sectional study. BMC Pregnancy Childbirth. 2020;20(1):42. doi: 10.1186/s12884-020-2736-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Tujuba TD, Chilo D, Abebe E, Zenu S. Self-medication and associated factors among pregnant women in rural Ethiopia: the importance of husband Education in ensuring a safe pregnancy. Drug Healthc Patient Saf. 2023;15:39–50. doi: 10.2147/DHPS.S394346 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Edet CK, Wegbom AI, Samuel KG, et al. Prevalence and factors influencing self-medication among pregnant women attending antenatal clinics in urban Tertiary hospitals in Nigeria. BioMed. 2023;3(1):166–176. doi: 10.3390/biomed3010014 [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

The de-identified dataset and analytic syntax are available from the corresponding author upon reasonable request, subject to institutional approval and applicable ethical restrictions.


Articles from International Journal of Women's Health are provided here courtesy of Dove Press

RESOURCES