Risk perception increase due to COVID-19 impacted antenatal care utilization among women in an indigenous community

Abstract

Background

Risk perception varies greatly among individuals, affecting their behavior and decision-making in risky situations. The COVID-19 pandemic affected worldwide, but the role of risk perception related to COVID-19 in ethnic minorities in Mexico is unclear. This study quantifies the impact of COVID-related risk perception (susceptibility and severity) and perceived fear on the utilization of antenatal care services among indigenous women in San Cristobal de las Casas, Chiapas, Mexico.

Methods

We conducted a retrospective crossover study between June and December 2021, interviewing 98 women from San Cristóbal de las Casas, Chiapas. In a crossover design, each subject acts as their own control, so we required the participants to have a previous pregnancy experience. A logistic model was used to calculate the odds ratio for the outcome of having an adequate number of antenatal care visits. The analysis considered the period (during or before the pandemic) as well as perceived severity and susceptibility levels as independent variables.

Results

COVID-19 reduced antenatal care utilization by 50%. During the pandemic, the adjusted odds ratio for attending health antenatal care services was 0.83 (95% CI: 4.8, 14.5) compared to pre pandemics. Adjusted for fear of contagion, the mother’s perception of severity was associated with an increased likelihood of an insufficient number of antenatal visits. OR = 0.25 (95% CI: 0.10, 0.65).

Conclusion

The risk perception for COVID-19 decreased the likelihood of receiving an adequate number of antenatal care visits.

Background

Risk perception refers to the cognitive response and evaluation of perceived threat, as stated by Han et al. [1] Two theories of risk perception, namely the Protection Motivation Theory (PMT) by Rogers [2] and the Health Beliefs Model (HBM) by Becker [3] distinguish between perceived susceptibility and perceived severity. The former represents the individual subjective appraisal of the likelihood of developing a health problem. At the same time, the latter pertains to the level of concern regarding the seriousness of the related consequences, according to Rosi et al. [4] Both risk perception models prioritize an individual’s subjective perception of reality rather than the objective truth, as Prentice-Dunn and Rogers highlighted [5].

Sanitary measures and acknowledgment of the possibility of contagion can result in fear, anxiety, and stress, affecting emotional and mental health. High levels of threat and fear characterized the pandemic, greatly influencing the cognitive process of human decision-making. [6] Consequently, it is necessary to assess the levels of anxiety and fear as motivators that could influence the adoption of protective behaviors. [7] Risk perception and fear during the pandemic might have induced a motivational prioritization of the COVID-19 perceived threat over other needs, [1] such as good use of antenatal care services. This phenomenon is not unique to COVID-19; similar behavioral shifts were observed during the Zika virus outbreak in 2016, [8] where the fear of birth defects such as microcephaly significantly influenced the decisions of pregnant women regarding their prenatal care and protective measures. Understanding risk perceptions of a public health crisis and protective behaviors are critical for disease prevention and control [9].

Antenatal care reduces adverse pregnancy outcomes and maternal and fetal mortality rates. It is thus considered a critical and essential service that must be provided to women throughout pregnancy. [10] Improving access to and providing high-quality antenatal care is a priority for global health initiatives. However, despite the importance of these services, the World Health Organization (WHO) reported that only 64% of women worldwide received adequate antenatal care in 2015, leading to 2.7 million infant and 303,000 maternal deaths within the first 28 days postpartum [11].

In Mexico, the utilization of antenatal care services is widespread, with most women attending at least one antenatal care visit. Nevertheless, significant disparities persist based on ethnicity, wealth, and education, particularly affecting marginalized communities. The challenges of accessing these services have been further exacerbated by the COVID-19 pandemic, which has disrupted health systems and impeded care access for women. The pandemic has resulted in a significant increase in maternal mortality rates compared to 2019, [12, 13] underscoring the detrimental impact of reduced antenatal care coverage.

The pandemic has disproportionately affected socially disadvantaged groups, exacerbated pre-existing health disparities and widening the gap between privileged and marginalized groups [14].

This study aims to quantify the impact of COVID-related risk perception (susceptibility and severity) and perceived fear on the utilization of antenatal care services among indigenous women in San Cristobal de las Casas, Chiapas, Mexico.

Methods

Design and settings

We used a retrospective case-crossover design. Each subject’s data is treated as if they were from a matched pair, switching between periods of different risk (pre-pandemic vs. pandemic period) [15, 16]. The study was conducted with pregnant women from the Tzotzil ethnic group, residents of Chiapas, Mexico. Between June and December 2021, women who had experienced a pregnancy both before and during the COVID-19 pandemic were invited to be volunteer participants.

Pilot

Eighteen women in the study population were pilot-tested with the questionnaire. As a result of the pilot test, three questions were modified. The supplementary material contains the final instrument.

To estimate the sample size, we applied the case-control study design formula, using one control for each case. This approach was appropriate because our retrospective case-crossover study utilized a previous pandemic pregnancy as a control measure considering that each participant served both as a case and as a control. The sample size calculation was based on the following parameters: a control exposure frequency of 50%, an odds ratio (OR) of 2.5 to detect the minimum risk, a significance level of 0.05, and a test power of 0.80. As a result, we determined that a total sample size of 40 cases was needed [17, 18].

Standardized personnel administered a questionnaire during one of their visits to one of two health centers where we had permission to conduct the study in San Cristobal de las Casas, Mexico. Both health centers belonged to state services of the Mexican Ministry of Health. One center provided first-level care, and the other provided second-level care. The questionnaire was administered in a private room provided by the authorities of these health centers and was conducted entirely in Spanish. The questionnaire was not translated into Tzotzil (the most spoken indigenous language in the San Cristóbal de las Casas community); however, a translator proficient in both Spanish and Tzotzil was available during the data collection process.

Participants were asked to provide information about their antenatal care utilization during their previous and current pregnancy. They were also asked several questions to determine their perception of COVID-related risks and fear of contagion during their most recent pregnancy, which occurred during the pandemic.

Informed consent

was obtained verbally, during which participants were provided detailed information about the study’s objectives and the procedures they were required to follow. It was emphasized that participation was entirely voluntary and that they had the freedom to withdraw at any time without consequences. Subsequently, when participants lacked literacy skills, the designated administrator presented the questionnaire orally, and at least one literate witness signed the informed consent form.

Inclusion criteria

Participants included in this cross-over study were: (1) Women with recurrent pregnancy, whose previous pregnancy occurred up to three years before the health contingency and the current or most recent pregnancy during the COVID-19 pandemic, (2) Attending San Cristóbal de las Casas health centers for antenatal care, and (3) Agreed to participate in the study voluntarily and signed the informed consent when the participant was literacy, when the woman wanted to participate but was illiteracy, informed consent was obtained orally and one literate witness signed the informed consent form.

Exclusion criteria

Women were excluded if they were unable to respond to the questionnaire or interview because of a mental or cognitive apparent disability.

Risk perception

We evaluated two components from the Health Belief Model: perceived susceptibility and severity. The items that made up both scales resulted from a discussion among the group of researchers based on the definition of both constructs. As Conner and Norman have pointed out “The HBM had the advantage of specifying a discrete set of common-sense beliefs”. [19] Based on the characteristics of the study population, we opted for items expressed in the simplest and most parsimonious way. The susceptibility component was evaluated with three questions: (1) I could get COVID by going to pregnancy checkups, (2) People around me do not follow the COVID safety protocol, and (3) My baby could get COVID if I go to pregnancy checkups; a likert scale format was used with four response options: (1) It is something I’m concerned about, (2) No, it doesn’t concern me, (3) It’s not something I had thought about, (4) I really don’t think it could happen.

The severity component was assessed with four questions: (1) Illness from COVID-19 is more likely to be more aggressive for someone like me, (2) It would be more difficult for me to recuperate from COVID-19 than for the general population, (3) I could have premature labor due to COVID-19, and (4) I could have an abortion due to COVID-19. A Likert scale format was used with 6 response options to each question: (1) strongly disagree, (2) disagree, (3) kind of disagree, (4) kind of agree, (5) agree, and (6) strongly agree.

In order to determine the total susceptibility variable, a value of zero was assigned to each of the three susceptibility questions if the response score was 2 (No, it doesn’t concern me) or higher. If the response selected was 1 (It is something I’m concerned about), the value of one was assigned. These values were then added together to obtain the total susceptibility score. To calculate the total severity variable, each of the four severity questions was assigned a value of one if the response was 4 (kind of agree) or higher, and a value of zero if it was lower. These variables were then added together to obtain the total risk score.

Fear of contagion

Fear was evaluated with the question: Were you afraid of getting infected? (Were you afraid of catching it? ) which could be answered with yes or no.

Antenatal care utilization

To consider the antenatal care utilization as adequate a minimum of six visits was required, while attending less than that number was considered inadequate. Based on the Mexican Official Regulation called “NOM-007-SSA2-2016, for the Care of Women during Pregnancy, Childbirth, and Puerperium, and of the Newborn”, it is recommended that pregnant women with low risk attend 8 antenatal care visits and a minimum of 5 visits [20]. However, to prevent the spread of SARS-CoV-2, it was recommended that the frequency of check-ups be spaced out and the number of patients reduced by day, thus reducing the recommended antenatal care visits from 8 to 6 [21].

Ethical considerations

The Ethics and Research Committee of the Faculty of Medicine of the National Autonomous University of Mexico approved the protocol of the present study under Research Protocol: FM/DI/115/2019. Participation in this study was voluntary. Before completion, participants were informed of their rights as outlined in the Helsinki Declaration. [22] All participants were informed about the study’s objective, their research rights, that there would be no consequences if they chose not to participate, and the confidential nature of their participation. When a woman was illiterate but wanted to participate, informed consent was obtained orally, and one literate witness signed the informed consent form.

Statistical analysis

The analysis considered data from the questionnaire regarding sociodemographic information, antenatal care utilization, as well as perceptions of severity, susceptibility, and fear of contagion. We described categorical variables expressed as numbers and percentages (%), and Chi-square tests were used to test differences between the groups of data before and during the pandemic. A p-value of less than 0.05 was considered significant.

To test the concordance between vulnerability and severity items, Gamma and Kendall’s coefficients were applied to the whole sample based on the ordinal nature of each item (Likert scale from 1 to 5).

Using a logistic regression model, we evaluated the changes in the utilization of antenatal care before and during the pandemic. Afterward, additional adjusted logistic regression models were employed to estimate the association between adequate antenatal care utilization and severity and susceptibility indices. Variables with a p-value of less than 0.20 in the bivariate analyses were used to adjust the model. Goodness-of-fit analysis was performed with the Hosmer-Lemeshow test. The data were analyzed with Stata v. 18. (Stata Corp, College Station, Texas, USA).

Results

Ninety-eight pregnant women participated in the study, with an average age of 21.9 years. 12% were younger than 20 years old. Most of them spoke an indigenous language (96%) and were literate (92%). It was not necessary to use Tzotzil translators as all participants understood Spanish. Fewer than 9% had acquired a high school education or higher. Most of them were housewives (50%) and lived with a partner (99%). (Table 1)

Table 1 Population characteristics of the Reproductive Health Study San Cristóbal De Las Casas, Chiapas. México. Year 2021

The data showed a significant decrease in the number of antenatal care visits during the COVID-19 pandemic. The percentage of women who received an adequate number of visits went from 73.5% before the pandemic to 21.4% during (p < 0.05). The percentage of women who had their first antenatal care visit during the first trimester of pregnancy was the same for both periods (90.8%). In both periods, Seguro Popular/INSABI (Spanish acronym for Health Institute for Wellbeing) was the most frequently used health institution (92.9% and 94.5%). Furthermore, during the pandemic, participants experienced longer wait times for appointments and delays in being seen upon arrival at the health center; 24.0% of participants waited for more than an hour to be seen by a medical professional after arriving at the health center compared to 3.2% before the pandemic (Table 2).

Table 2 Characteristics associated with pregnancy and COVID-19 San Cristóbal De Las Casas, Chiapas. México. Year 2021

Less than half of the participants perceived themselves as susceptible. Just 44.9% of participants reported being concerned about catching COVID-19 during their antenatal care visits. The same percentage of women were concerned about their babies contracting COVID-19 during their antenatal care visits. Slightly more than half the participants perceived COVID-19 as a more aggressive disease in pregnant women than in the general population (52.0%), and 51.0% perceived that it would be more difficult for pregnant women to recover if they got infected. Nearly half of participants did not perceive a risk of abortion if they became infected (48.0%). (Table 3).

Table 3 Susceptibility and severity associated with pregnancy and COVID-19 San Cristóbal De Las Casas, Chiapas. México. Year 2021

Compared to pre-pandemic pregnancies, the odds ratio for not receiving an adequate number of antenatal care visits during the pandemic increased by more than 8-fold, adjOR: 8.3 (95% CI: 4.8, 14.5).

In both the severity and vulnerability assessments, the concordance coefficient was greater than 0.35, which suggests consistency. The Gamma test indicated statistical significance at 0.05. Regarding the association between an adequate number of antenatal care visits and risk perception, the perception of susceptibility was not significantly associated with the utilization of antenatal care, OR: 0.48 (95% CI: 0.12, 1.97; p value: 0.306). While perceived severity reduced the odds of an adequate number of antenatal care visits adjOR: 0.25 (95% CI: 0.10, 0.65; p < 0.05). Figure 1 shows the predicted probability of susceptibility and the OR for adequate antenatal care, adjusted for fear of contracting COVID-19 and pandemic indicator (pre-pandemic or during the pandemic).

Fig. 1

Predicted probability for adequate ANC during the pandemic by severity perception and fear contagion

Discussion

The purpose of this study was to quantify the impact of COVID-related risk perception (susceptibility and severity) and perceived fear on the utilization of antenatal care services among indigenous women in San Cristobal de las Casas, Chiapas, Mexico.

We found a decrease in the percentage of participants who received an adequate number of antenatal care visits during the pandemic. This finding is consistent with previous research that also reported a significant decrease in antenatal care utilization due to the COVID-19 outbreak. [23,24,25]

According to UNICEF(2016), 90% of pregnant women in the Latin America and Caribbean region receive at least four antenatal care visits. However, it is crucial to recognize that these national averages conceal significant disparities based on ethnicity, wealth, and educational background. [26] Addressing and understanding these differences is of paramount importance to ensuring equitable and effective reproductive health care for all communities in the region. Additionally, it has been reported that the poorest communities receive at least one antenatal care visit from any type of provider, including most women in Chiapas; however, indigenous women living in these communities are less likely to receive skilled, timely, sufficient, and adequate antenatal care compared to women living in poverty. [26] (Dansereau et al., 2018).

In addition, we found that among those who perceived COVID-19-related fear of contagion, the decrease in their odds of receiving adequate antenatal care visits as the severity score increased was significantly greater compared to those who did not perceive fear. This finding is consistent with Anggraeni et al. who found that pregnant women often prefer to delay their antenatal care visit because they are afraid of being infected with COVID-19. [27]

As part of our results regarding risk perception, we found that roughly half of the participants felt vulnerable to COVID-19 contagion and perceived as severe the potential complications of the disease. Considering susceptibility in particular, 44.9% of pregnant women reported feeling afraid of being infected by COVID19 during their antenatal care visits. As expected, the susceptibility component was inversely associated with the possibility of receiving an adequate number of antenatal care visits, however, our investigation yielded no statistically significant correlation. This lack of statistical significance can be attributed to two plausible factors: the relatively modest sample size and the intricate intersectorial ramifications of the pandemic. The latter introduces a gendered lens to the pandemic’s impact. [28]

On the other hand, it was found that the perception of severity, adjusted for the mother’s fear of contagion, was associated with a higher likelihood of receiving inadequate antenatal care adjOR: 0.25 (95% CI: 0.10, 0.65). This result suggests that as women perceive greater severity in the risks associated with COVID-19, the likelihood of avoiding necessary antenatal care increases, which is consistent with a cross-sectional study conducted in Northeast Ethiopia where fear of COVID-19 infection was found to reduce overall service utilization by 87% (AOR = 0.13, 95% CI = 0–0.31) [29].

Limitations

Although our results are based on primary data, we acknowledge some limitations to our study. An important limitation is the sample size of 98 women. However, despite the limited number of participants from a specific community, it may be relevant to other communities that share comparable characteristics; in Mexico, there are numerous Indigenous groups that have their own distinct languages and reside in a variety of communities with diverse beliefs and perspectives, including those related to reproductive health. Another limitation was that only the perceived threat component of the health belief model was used, excluding the perceived benefits and barriers components. In addition, the authors recognize the potential for participants to have had incomplete comprehension of the questionnaire, despite the efforts to make it simple and unambiguous. Consequently, it is plausible that some individuals may not have answered all questions or expressed uncertainty in their responses. A retrospective case-crossover design utilizes participants as their own controls. In our study, women were asked about their current and previous pregnancies, acknowledging the potential for recall bias. However, the time between the previous pregnancy and the current one was 29.5 months on average; this short time could diminish the potential recall bias. Finally, the authors recognize a limitation in the formal validation of the questionnaire and the items about vulnerability and severity; however, the results from the concordance test performed showed that both concepts could be considered consistent.

Conclusions

The findings of this study align with the predictive capacity of the severity perception component within the Health Belief Model, demonstrating its relevance in guiding health-related behavioral decision-making. The significant impact of perceived fear of COVID-19 contagion on the utilization of appropriate antenatal care visits underscores the role of psychological factors in shaping healthcare-seeking behaviors. However, it’s important to acknowledge that the observed reduction in care utilization may also be attributed to broader contextual factors, such as the diminished availability of healthcare services in Mexico.