Understanding and measuring pregnancy-related anxiety in low- and middle-income contexts: A pilot study in northern Ghana

Background
Women's anxiety during pregnancy is widespread and has been associated with poor mental and physical health outcomes for both mothers and their children. One type of anxiety which seems to be particularly problematic is anxiety about the pregnancy itself. We know little about this type of anxiety in low- and middle-income countries (LMICs). However, what we do know suggests that pregnancy-related anxiety may differ for women in LMICs as compared to women in high-income countries: women in LMICs are more concerned about healthcare access, disease, and maternal mortality. Despite these differences in pregnancy-related anxiety between high-income and LMIC contexts, no measures of this construct have been validated in any LMIC. Together, this suggests that: 1) maternal mortality may be more important to the pregnancy-related anxiety of women in LMICs, 2) there is variation in domains of pregnancy-related anxiety in different contexts, and 3) the scales which are developed to measure pregnancy-related anxiety in pregnant women in high-income countries may not be effective in measuring this anxiety in women in LMICs because they may not be capturing all locally-relevant components of this anxiety.

Objectives
The proposed project will address this through the following research questions:
1. How do levels of pregnancy-related anxiety differ by region and by maternal and neonatal mortality rates?
2. How is pregnancy-related anxiety conceptualized in the northern Ghanaian context?
3. How can we effectively measure pregnancy-related anxiety in the Northern Region of Ghana?

To answer these questions, we will first systematically review research on prenatal anxiety. This will allow us to get a clear overview of levels of pregnancy-related anxiety, including clarifying geographic gaps in our knowledge and understanding its links with mortality. We will then apply a bottom-up approach, using focus group discussions with 150 women to investigate the domains of pregnancy-related anxiety in Ghanaian women. We will use this enhanced understanding to inform the adaptation of the Pregnancy-Related Anxiety Scale. We will adapt this measure to context and then validate it using a survey study of 575 pregnant women in northern Ghana. This culturally-adapted measure will allow for more accurate and reliable screening for pregnancy-related anxiety in Ghanaian women.

We will use Northern Region of Ghana as an exemplar in addressing Objectives 2 and 3. We have chosen this location because:
1) Symptoms of anxiety are common among pregnant Ghanaian women, with 11% having a generalized anxiety disorder.
2) Rates of maternal and neonatal mortality are high in this Region. Understanding - and ultimately, addressing - their mental health during pregnancy has the potential lead to reductions in these rates.
3) 97.7% of women in the Northern Region receive antenatal care. Interventions to reduce pregnancy-related anxiety are commonly conducted in healthcare settings, and so this Region is well-suited for future intervention work.
4) Our study team have a number of collaborations together in this Region, and thus are well-placed to deliver this project.

Significance
The outcome of this project will be a clearer understanding what underlies pregnancy-related anxiety and how to accurately measure it in LMIC contexts. Without a clear understanding of what underlies this anxiety, we cannot develop effective interventions, and without good measures of this anxiety, we cannot test whether those interventions are effective. This project directly addresses both of those issues and thus will pave the way for future work promoting mental health and wellbeing during pregnancy and reducing associated maternal mortality and morbidity. It is clearly aligned with the UN Sustainable Development Goal #3 of promoting health and wellbeing.

Women's anxiety during pregnancy is widespread and has been associated with poor mental and physical health outcomes for both mothers and their children. One type of anxiety which seems to be particularly problematic is anxiety about the pregnancy itself. We know little about pregnancy-related anxiety in LMICs and there are no measures of pregnancy-related anxiety which have been validated for use in any LMIC.

To address this gap, we will conduct a systematic review to get a clear overview of levels of pregnancy-related anxiety and relations with maternal and neonatal mortality. We will also conduct a mixed-methods study to measure how pregnancy-related anxiety is conceptualized and to adapt and validate the Pregnancy Related Anxiety Scale among pregnant women in the Northern Region of Ghana. To do this, we will conduct 15 focus groups (n = 150) and collect survey data from 575 pregnant women attending antenatal care in Mion, Savelugu, and Tamale Metropolitan Districts. We will hire one PDRA and 5 RAs to assist with the systematic review and qualitative and quantitative data collection.

We will use inductive thematic analysis to analyse the focus group data. We will use exploratory factor analysis, correlations, and t-tests to establish construct validity, convergent validity, and test-retest reliability of the adapted scale. We will also consult with a local advisory committee to ensure the use of appropriate methods and measures, and to establish face validity of our adapted measure.

This project is directly linked with UN Sustainable Development Goal #3 (good health and wellbeing). Without a clear understanding of what underlies pregnancy-related anxiety and how to measure it, we cannot develop and implement effective interventions nor can we evaluate whether any implemented interventions are effective. This project will pave the way for future work promoting mental health during pregnancy and reducing associated maternal mortality and morbidity.

The proposed research will of interest to at least three different and distinct user groups:

1. Researchers. The written results of this project, the adapted measure, and the datasets which will arise will be of broad interest to multiple disciplines within the scientific community, including midwifery, family medicine, public health, and psychology. With regards to the written results, the systematic review will provide important new insights to researchers regarding the levels of pregnancy-related anxiety worldwide, gaps in knowledge, and links with maternal and neonatal mortality. This will also be one of the few studies on pregnancy-related anxiety in a low- or middle-income country; what underlies pregnancy-related anxiety in this context will be of broad interest. With regards to the adapted measure, we will make the adapted and validated Pregnancy-Related Anxiety Scale publicly available at the end of this project. This measure will be useful to researchers in Ghana and potentially in other similar contexts. With regards to the datasets, we will archive two de-identified datasets with the UK Data Archive by the end of project: one qualitative and one quantitative dataset. The qualitative dataset will include all 15 focus groups. Researchers interested in pregnancy, mental health, or health care access may find this rich dataset of up to 150 pregnant women to be useful. The quantitative dataset will include the 575 participants in the survey study, and will have information on a range of mental health problems (pregnancy-related anxiety, generalized anxiety, fear of childbirth, distress, depressive symptoms, etc) as well as demographic information on pregnant women in the Northern Region. These datasets will be of considerable interest to a broad range of researchers doing secondary data analyses.

2. Policy makers and non-govnernmental organizations. The systematic review will highlight what is known about levels pregnancy-related anxiety generally and specifically about its links to mortality. This information will be relevant to policy makers worldwide. Further, the qualitative and quantitative study findings will be of great interest to policy makers and administrators in the Northern Region and throughout Ghana. These policy makers will benefit from the findings by gaining a better understanding of the mental health needs of pregnant women in the Northern Region. There is relatively poor provision for mental health difficulties in this Region, with a small number of mental health nurses providing the bulk of the care. If we find that mental health problems are prevalent in our sample, this would provide policy makers with evidence to request more mental health provision. If our systematic review shows that pregnancy-related anxiety is related to maternal or child mortality, this would strengthen that request further. We will ensure that policy makers and administrators learn of our findings, through written study reports, through one-to-one meetings with Abiiro and Hadfield, and through the dissemination workshop at the end of the project.

3. General public. The study findings will also be of interest to the general public, particularly in Ghana. Besides providing more insight into issues relating to mental health of pregnant women generally, the general public will benefit from a clearer understanding of what pregnancy-related anxiety entails. We will ensure that the general public has the opportunity to learn of our study findings by posting the brief study reports on the Ghana Medical Help website and on social media. We will also create press releases for the resultant papers. Finally, we will invite Ghanaian press to the dissemination workshop at the end of the project, with the goal of getting press attention to our findings in Ghana and the surrounding region.