This was a prospective, cross-sectional, quantitative and observational study.

The research was conducted at the CHBAH antenatal clinic. The hospital is a public, academic and tertiary-level institution situated in Soweto, Johannesburg.

A sample size of 200 participants was used, which was calculated based on a confidence level of 95%, a prevalence rate of 15.2%, and an estimated population size of 25460. Pregnant women were recruited from the antenatal clinic at the hospital between March and December 2022. A convenience sampling method was used. Prospective participants in the waiting area were given an information sheet detailing all the aspects of the study. Once participants agreed to engage in the study, they were screened to ascertain if they met the relevant inclusion and exclusion criteria, and a signed consent was obtained. The hospital where the study was conducted, being a tertiary academic institution, runs separate high-risk clinics for those with previous perinatal complications and physical health conditions. Participants were sampled on different days of the week, not only on high-risk clinic dates.

Participants who were excluded from the study included those who were under the age of 18, who have not had a first antenatal booking, had significant intellectual impairment, were attending the maternal mental health clinic or could not communicate proficiently in English.

A structured self-reported questionnaire was used to collect the relevant data. Data collection was carried out in an interview style, on a paper-based questionnaire and investigator administered in a private room within the clinic. Medical and obstetric information was additionally obtained from viewing the patient’s clinical record at the time of the interview, if required.

The first part of the questionnaire aimed to acquire data regarding socio-demographic characteristics, obstetric factors, and psychiatric and medical co-morbidity. These variables were assessed as categorical data, with binary and nominal options and continuous data for age and gestation. The generalised anxiety disorder 7-item scale (GAD-7) was used for the second part of the research questionnaire to screen for the presence of anxiety symptoms. It is one of the identified recommended tools to screen for anxiety during pregnancy.²¹ It consists of seven items with Likert-responses and relevant scoring between 0 and 3, with a total range of 0 to 21 points.²² A cut-off score of 10 or more has been recommended for the general population, with higher scores reflecting increased severity of symptoms and less than 10 indicating no need for treatment.²² Use of the GAD-7 is freely available.^23,24 Participants with scores of 10 or more were referred to the maternal mental health clinic using a referral letter provided to the participant for further assessment and management.

The data were captured on Microsoft Excel^TM. All statistical analyses were conducted using R software (R version 3.4.2; https://www.r-project.org). Tests are two-tailed probability values and statistical significance is accepted when α ≤ 0.05. Data are presented in tables or figures, as appropriate. Continuous variables are reported as mean and standard deviation (s.d.). Categorical data are presented as counts and percentages. The association between categorical socio-demographic and clinical variables was analysed using Pearson’s Chi-squared tests or Fisher’s exact tests. The continuous variables, age of the patients and gestation were assessed for departures from normality using the Shapiro–Wilk test and subsequently analysed using a Mann–Whitney U test. Odds ratios are reported for significant Fisher’s p values.

An informed, written consent was obtained from each participant prior to participation. Participant confidentiality was preserved, as no identifying data were noted on questionnaires; instead, participant numbers were allocated. Participants screening high for anxiety were given a referral for further assessment, with contact information for emergency care if required. Ethical approval from the ethics committee of the University of the Witwatersrand was obtained prior to conducting data collection. The ethics approval number is M210601. Permissions from the head of the department of the Obstetrics unit, Psychiatric unit and the Chief Executive Officer (CEO) at the facility were obtained. The research is also registered on the National Health Research Database (NHRD). The NHRD reference number is GP_202103_072. All procedures were in accordance with the ethical standards of the 1964 Helsinki Declaration and its later amendments.

In this study, we found the prevalence of anxiety symptoms in pregnant women attending the antenatal clinic at CHBAH to be 33%, based on scoring of 10 or more on the GAD-7. Participants with anxiety were younger, employed and had lower perceived social support. Women with planned and wanted pregnancies had a lower prevalence of anxiety. There was no noted significant correlation between anxiety and level of education, relationship status, gestation and psychiatric and medical co-morbidity.

Previous research has shown a varying pattern of the emergence and severity of anxiety prevalence in pregnancy. The prevalence of anxiety ascertained in this study is noted to be higher than that of previous studies conducted in the same facility. A study conducted by Redinger et al. at CHBAH between 2014 and 2016 noted a prevalence of 15.2% for anxiety in the first trimester on the State-Trait Anxiety Inventory.⁶ Using a clinical interview, a prevalence rate of 14.5% for antenatal anxiety disorders was reported in another study carried out at Rahima Moosa Hospital antenatal clinic in Johannesburg between July 2015 and April 2016.²⁵ These prevalence rates are comparable with results found in other African, low-middle income countries at the time.⁶ Other studies reflected higher prevalence rates of antenatal anxiety. A cross-sectional study conducted at primary-level clinics in a low income setting in Cape Town between November 2011 and August 2012 revealed an overall prevalence for anxiety disorders in pregnancy of 23% using diagnostic interview techniques.²⁶ This was comparable to another study, where high levels of psychological distress, 26.5%, were noted in pregnant women in Nkangala district in Mpumalanga using the Kessler Psychological Distress scale.³ A systematic review study looking at the prevalence of antenatal and postnatal anxiety in studies published up to January 2016 reported anxiety symptoms to have an overall prevalence of 22.9% across all three trimesters.²⁷ Of note, in the above-mentioned studies, different methods of assessing anxiety were used, with some using different screening tools and others using clinical assessment. This may be attributed to the differences noted in prevalence.

South African studies utilising the GAD-7 as a tool to screen for anxiety in the antenatal period are noted to be limited at this time. A study carried out by Hoque et al. used the GAD-7 to assess the level of anxiety as one of the objectives of the study in pregnant women during the coronavirus disease 2019 (COVID-19) epidemic in South Africa. The study was conducted at a primary health care centre in Durban and found an anxiety prevalence of 35.7% using the GAD-7.²⁸

The exposure to disease outbreaks may further exacerbate anxiety and negative emotional states during pregnancy.²⁹ On 11 March 2020, the World Health Organization (WHO) declared the novel COVID-19 outbreak a global pandemic.³⁰ Tomfohr-Madsen et al. conducted a rapid review and meta-analysis on depression and anxiety in pregnancy during COVID-19. A pooled prevalence of 25.6% was noted for depression and 30.5% for anxiety.³¹ This higher prevalence is in keeping with the prevalence noted in this study, and therefore, although data collection was conducted in 2022, the higher prevalence rate of anxiety may be an effect of the long-term repercussions of the COVID-19 pandemic. The WHO declared an end to COVID-19 as a public health emergency on 05 May 2023.³² In South Africa, the national state of disaster in relation to COVID-19 was lifted in April 2022.³³

The finding of increased anxiety with younger age correlates with other studies.^6,34,35 Lower level of education has been noted as a risk factor for anxiety in pregnancy; however, in this study, no significant correlation was noted between anxiety and level of education. A correlation was noted between employment, and specifically formal employment and higher levels of anxiety, which contrasts with other literature that notes unemployment as a risk for anxiety.³⁶ However, a study conducted in Pakistan noted that mothers working outside the household had a higher prevalence of anxiety.³⁷ The hospital in which this study was conducted services largely a lower socio-economic demographic, where often there is only one income sustaining multiple immediate and extended family members. Factors concerning employment and pregnancy such as maternity leave, childcare and meeting work demands may add to the burden of anxiety in pregnant working women.

Perceived social support was noted to increase the likelihood of not having significant anxiety. This is comparable to other studies. Lower perceived social support, high perceived stress, insufficient emotional and practical support, a history of abuse, adverse life events and having hostile in-laws have been noted to be associated with anxiety and other mental health conditions during pregnancy.^{7,34,35,38,39}

Specific gestation and primigravity were not associated with anxiety in this study. There are inconsistent findings regarding the association between gravidity and the presence of antenatal anxiety.⁴⁰ Some studies showed that being multigravida or having more children under the age of five in the household increased the risk of anxiety.^26,35 Other studies linked primiparity and bearing the first child to be associated with increased anxiety.³⁴

Having a planned and wanted pregnancy was associated with having no significant anxiety in this study. This was similar to other studies, where it was noted that maternal desire for pregnancy, including unplanned and unwanted pregnancies, were identified as predictors of anxiety.^{3,26,35,36,38}

The presence of previous pregnancy complications and physical health conditions has been associated with anxiety during pregnancy.^7,39,40 However, in this study, no significant association was noted.

Likewise in this study, there was no noted significant association between the history of past or present mental health illness and anxiety. This is a contradiction to other research that shows a history of mental illness to be a strong predictor of anxiety in pregnancy.^26,35,38,39 However, this study sample excluded patients already attending the maternal mental health clinic and therefore may have excluded many participants with known mental health conditions at the time. We also did not explore if participants were on treatment or not for mental health conditions.

This study was conducted in a low-middle income (LMIC) setting, and detailed biographical data were collected. A higher prevalence of anxiety has been noted in this study. This may reflect the impact of the COVID-19 pandemic and warrants re-looking at antenatal anxiety prevalence.

The study population sampled was from a tertiary institution; therefore, it may represent a group with overall increased risk factors for anxiety, leading to the higher prevalence of anxiety that was noted in this study. However, this may highlight the importance of routine screening for anxiety and other mental health conditions in this high-risk population.

Teenagers were excluded from the sample population because of ethical concerns and issues surrounding consent. They however represent a high-risk group for mental illness during pregnancy. The interviews were conducted exclusively in English, thus excluding participants who were not fluent in the language.

The use of self-report measures may lead to biases, possibly inflated prevalence rates and it is based on the participant’s ability to recall information and cannot be verified. However, self-report measures, like the GAD-7, have practical relevance and utility as it can be used in primary health care settings and maternal clinics where clinical staff may not have expertise in psychiatric training.

Broad questions were used to assess previous and current pregnancy complications, psychiatric illness and medical illness. Asking specific questions may be more insightful regarding the specific details of these factors.

This being a cross-sectional study, it does not account for the emergence of anxiety symptoms that may possibly occur at different points of pregnancy. Further longitudinal studies exploring patterns of anxiety emergence, specific anxiety disorders, risk factors and the impact of anxiety disorders on obstetric outcomes may be beneficial. In addition, there is a need for studies that include a diverse group of participants, such as teenagers and individuals who are not proficient in English. These studies should also encompass different health care settings.