Prevalence rates of HIV infection in KwaZulu-Natal are high, with a significant amount of those infected being women of reproductive age. A diagnosis of HIV infection has been associated with an increased risk for the development of depression. Antenatal depression is a serious health concern, having the potential to cause wide-reaching adverse consequences for mother and unborn child.
To compare depressive scores between newly diagnosed HIV-infected and HIV-uninfected pregnant women.
Antenatal clinics at two regional hospitals in KwaZulu-Natal, South Africa.
A cross-sectional questionnaire-based analysis of 102 newly HIV-tested black African pregnant women (HIV infected:
About 9.8% of women suffered from significant depressive symptoms, irrespective of HIV status. Prevalence rates of antenatal depressive symptoms did not differ significantly between HIV-infected and HIV-uninfected cohorts (
Prevalence rates of depressive symptoms were low. Knowledge of a new diagnosis of HIV infection at the first antenatal visit places women at an increased risk for the development of depression during pregnancy. Younger age and unemployment influence depression. This study provides an important step in documenting the need for screening for antenatal depression in HIV-associated pregnancies in a South African population group.
Women represent approximately half (51.0%) of adults living with HIV infection worldwide,
A diagnosis of HIV infection is associated with a surge in depressive symptoms.
A diagnosis of HIV infection is accompanied by psychiatric distress, with depression being a common complication.
The knowledge and timing of this diagnosis can also affect the development of depression. Studies found that women who are recently diagnosed with HIV infection are at a higher risk of depression, whereas women who know their HIV status before becoming pregnant were less likely to develop depressive symptoms.
However, it is controversial whether HIV infection impacts the frequency of depressive rates in pregnant women. As KZN is considered to be the global epicentre of the HIV pandemic,
A cross-sectional questionnaire-based analysis of depressive scores in newly HIV-tested pregnant women was conducted.
From August 2016 to October 2016, a sample of 102 pregnant women (62 HIV-uninfected and 40 HIV-infected) were recruited from antenatal clinics at two regional hospitals in KZN, South Africa. Patients were recruited based on the study’s inclusion and exclusion criteria. Given the study design, only newly HIV-tested women (prior to initiation of antiretroviral therapy) were included. Only Black South African women were included as there is a higher percentage of this population group attending these hospitals than other demographic groups. Additional inclusion criteria were being isiZulu speaking and period of gestation (either second or third trimester). Pregnant women previously tested for HIV, non-black South African patients and those that declined entry into the study were excluded. Furthermore, patients in the first trimester of pregnancy, those with medical and surgical complications, as well as individuals with a previous history of psychiatric illness, were excluded.
Whilst women waited for their routine antenatal appointment, the purpose and nature of the study was explained to the whole group by the research nurse. Research participants were then given individual patient information sheets in isiZulu and invited to participate. Socio-demographic and clinical data including maternal age, gestational age, parity, whether the pregnancy was planned or unplanned, area of residence, level of education, employment status, substance use, relationship status, HIV status and CD4 cell counts were recorded on a structured data sheet and participants were then assessed for depressive symptomology using the EDS. This 10-item self-report scale relates to an individual’s depressive symptoms during the past 7 days. Each item is scored on a 4-point scale with a total score range of 0–30, where a higher score indicates greater distress. It focuses on the individuals’ cognitive and affective depressive symptoms, whilst omitting somatic symptoms that could be confounded by normal pregnancy-related changes. It was originally established in 1987 and named the Edinburgh Postnatal Depression Scale, as it was developed as a screening tool for postpartum depression.
Data were analysed using GraphPad Prism 5.00 for Windows (GraphPad Software, San Diego, California, USA) and STATA version 12 (StataCorp LP, College Station, Texas, USA). Parametric tests were performed. Continuous variables were described in terms of means ± standard deviations and categorical variables using their frequencies and percentages.
Prior to the commencement of study activities, ethical approval (BE271/16) and permission to conduct the study at a large district and a tertiary referral hospital, was obtained. All participants signed a written informed consent form and were ensured confidentiality and anonymity through the use of research codes
Background characteristics of study population are depicted in
Background characteristics of study population.
Characteristics | HIV-infected ( |
HIV-uninfected ( |
|
---|---|---|---|
Age (years) | 28 ± 4.86 | 22 ± 2.83 | 0.0001 |
0.34 | |||
Yes | 8 (20) | 8 (12.90) | |
No | 32 (80) | 54 (87.10) | |
0.07 | |||
Urban | 40 (100) | 57 (91.94) | |
Rural | 5 (8.06) | ||
0.35 | |||
No formal education | 1 (2.5) | 0(0) | |
Primary school | 1 (2.5) | 0(0) | |
Secondary school | 25 (62.5) | 43 (69.4) | |
University | 13 (32.5) | 19 (30.6) | |
0.22 | |||
Employed | 12 (30) | 12 (19.4) | |
Unemployed | 28 (70) | 50 (80.6) | |
0.52 | |||
Single | 36 (90) | 58 (93.6) | |
Married | 4 (10) | 4 (6.4) |
Age is presented as mean ± s.d. (standard deviations). Planned pregnancy, area of residence, level of education, employment status and relationship status are presented as frequencies (
Associations between socio-demographic and clinical factors and total depressive scores on the EDS are represented in
Associations between socio-demographic and clinical factors and total depressive scores on the Edinburgh Depression Scale.
Characteristics | Total depressive scores mean ± s.d. | |
---|---|---|
0.0001 |
||
Infected | 10.43 ± 3.46 | |
Uninfected | 6.77 ± 3.13 | |
0.87 | ||
Yes | 8.06 ± 4.60 | |
No | 8.23 ± 3.55 | |
0.55 | ||
Secondary school | 8.06 ± 3.74 | |
University | 8.53 ± 3.67 | |
0.09 | ||
Employed | 8.55 ± 3.34 | |
Unemployed | 7.08 ± 4.61 | |
0.65 | ||
Single | 8.26 ± 3.65 | |
Married | 7.63 ± 4.53 |
s.d., standard deviation.
, Result significant at
As shown in
Prevalence of depressive symptoms amongst 102 newly diagnosed HIV-infected and HIV-uninfected pregnant women in KZN, South Africa. No depression: EDS scores of 0–9; significant distress: EDS scores of 10–12; probable depression: EDS scores of 13–30.
Analysis amongst women’s total depressive scores from the EDS, irrespective of depressive severity, showed a significant association between depression and HIV status and maternal age, which are further highlighted in
Depressive scores of HIV-infected versus HIV-uninfected pregnant women. Results are presented as mean ± standard deviations.
Correlation between maternal age and total depressive scores on the EDS. The red line represents the cut-off score for probable depression (13) on the EDS. *, Total depressive scores differed significantly with maternal age of women,
A new diagnosis of HIV infection during pregnancy can affect the severity of depressive scores. In this study, we used the EDS to evaluate the prevalence and severity of depressive scores in newly diagnosed HIV-infected and HIV-uninfected pregnant women in order to elucidate the role of a new diagnosis of HIV infection in the development of depression.
We report a low prevalence rate of depressive symptomology regardless of HIV status across the study population. Notably, the majority of women displaying scores consistent with significant depressive symptoms were from the HIV-infected cohort. However, when only considering women with probable depression (equivalent to a score of ≥ 13), no significant difference in the prevalence of depressive symptoms between the HIV-infected compared to the HIV-uninfected cohorts was found (
Moreover, our study reports that women’s total depressive scores, irrespective of depressive severity, were significantly different between HIV-infected and HIV-uninfected pregnant women. Higher depressive symptom scores were demonstrated in the HIV-infected cohort, highlighting that a new diagnosis of HIV infection predisposes a pregnant woman to a greater risk of developing depression. Interestingly, amongst the HIV-infected cohort of our study, the majority of women displayed symptoms of significant distress rather than that of probable or severe depression. This was also highlighted in other studies which found a diagnosis of HIV infection to increase the risk and severity of emotional distress in pregnant women.
A significant difference in maternal age was noted between the study cohorts. HIV-infected pregnant women were much older than HIV-uninfected pregnant women. Additionally, we found that as maternal age increased, the level of significant depressive symptoms decreased, demonstrating that younger women were more inclined to develop depression.
Also as expected, unemployed pregnant women were more depressed than employed women. Although only borderline significant (
Our study demonstrates that 9.8% of pregnant women displayed symptoms of depression, irrespective of HIV status. This rate of depressive symptoms is lower than expected and contradicts previous literature in which higher prevalence rates of depression were found in similar settings: 38.5% in urban KZN and 47.0% in rural KZN.
We report that total depressive scores on the EDS, irrespective of depressive severity, differed significantly between the HIV-infected and HIV-uninfected cohort. This finding is not novel, and evidence of elevated depressive scores amongst HIV-infected pregnant women has been previously reported.
However, we did not find a difference in prevalence rates of depressive symptoms between the HIV-infected and HIV-uninfected cohorts. Similar findings were reported in HIV-infected and HIV-uninfected pregnant women in South Africa and the US.
Maternal age is a predictor of depression, indicating that younger women are at a greater risk of developing depression than older women.
Unemployed women were at a greater risk of developing depression than employed women in this study. This may be elucidated by the increased financial and social support one receives when having a permanent job. This finding has been demonstrated previously
In contrast to previous reports, we were unable to demonstrate correlations between level of education, planned pregnancy, relationship status and CD4 cell counts.
This study reports and compares depressive scores of newly diagnosed HIV-infected and HIV-uninfected pregnant women in KZN, South Africa. It provides valuable information on the prevalence, as well as the role played by HIV infection and other socio-demographic and clinical data as risk factors for the development of depressive symptoms in pregnant women, in a province that is the foremost contributor to the global HIV pandemic. The EDS also demonstrated good internal reliability, with a Cronbach’s alpha of 0.76, indicating that women in this study were consistent in their response.
Our study has limitations. One is the relatively small sample size, accompanied by the unequal sample size between cohorts. The study population was limited to Black African women; therefore, it is not known whether the results can be generalisable to other populations. Additionally, this study did not examine antenatal depression across all three trimesters of pregnancy, which could possibly impact the prevalence rate. It is also possible that women were not truly ‘newly diagnosed’, which could have affected the prevalence rate in this study. Furthermore, certain variables noted in previous studies that were found to be associated with depressive symptoms were not assessed in this study. Finally, the cross-sectional design of the study limits us to mere associations and prevents actual causal inferences.
Given the likelihood of the burden of antenatal depression on the quality of life of both mother and infant, future studies conducting research using a clinical diagnostic instrument for depression on a larger sample, involving all three trimesters of pregnancy and following delivery, are recommended.
We report a low prevalence of significant depressive symptoms in pregnancy, irrespective of HIV status, which may be attributed to the improved perception of HIV infection in South Africa. Additionally, we report that a new diagnosis of HIV infection in pregnancy exacerbates the risk for the development of depressive symptomology. Moreover, we demonstrate that younger maternal age and unemployment increase the development of depressive symptoms. Furthermore, we report similar rates of depressive symptoms between HIV-infected and HIV-uninfected cohorts. This study provides an important step in documenting the need for screening for antenatal depression across all three trimesters and following delivery in HIV-associated pregnancies.
The authors declare that they have no financial or personal relationships which may have inappropriately influenced them in writing this article.
All authors have contributed significantly to this article. P.N., T.N. and J.M. were all involved in the conception and design of the study. P.N. was involved in carrying out data collection, analysis and writing up of work. T.N. and J.M. were involved in the critical review and editing of the manuscript.