Study design and study setting
An institution-based cross-sectional study was conducted to investigate the prevalence and associated factors of PPD among mothers who visited Expanded Program of Immunization (EPI) clinics in public health facilities of the Awsi Rasu Zone of the Afar Region from May to June 2021. The Afar National Regional State (ANRS) is geographically located in the Great East African Rift Valley (GEARV) and shares a border with Eritrea (Northeast), Tigray (Northwest), Oromia (South), Somali (Southeast), Amhara (West) and Djibouti (East) [37]. The ANRS has an estimated population size of almost two million with more than 12% of the population aged less than five years, and the livelihood of more than 85% of the Afar population depends on livestock production (including camel, goat and cow) [38]. The arid or semi-arid and low rainfall climate makes the ANRS vulnerable to adverse health problems including malnutrition, infection, and poor maternal and child survival [37].
The Awsi Rasu Zone is one of the five administrative zones (i.e., zone is the second administrative unit in the Ethiopia context) with the highest estimated population size of 498,873. In the Awsi Rasu Zone, there are three public hospitals, one private maternity hospital, and 25 health centres that provide preventive, curative and rehabilitative health services in the Awsi Rasu Zone [39] (Fig. 1).
Source and study population
All postpartum women who visited EPI clinics of health facilities in the Awsi Rasu Zone were taken as a source population and those randomly selected to be included in the study were the sample population. Postpartum women who were seriously sick during the data collection period and unable to hear were excluded.
Sample size determination and sampling procedure
A single population proportion formula considering a 95% confidence interval (CI), 0.05 margin of error, 23.3% proportion of PPD [40], and 10% non-response rate was used to calculate the required sample size. A total of 302 postpartum mothers were selected from six health centres (including Logia Health centre, Mile Health centre, Eliwuha Health centre, Afambo Health centre, Addar Health center and Aysaita Health centre) and two hospitals (including Ayssaita Primary Hospital and Dubti Referral Hospital). A systematic random sampling technique was used to select postpartum mothers using previously registered case flow of patients in each health facility. During the survey, the first study participant was selected using a lottery method.
Outcome variable
The main outcome variable of the current study was PPD measured using the Edinburgh Postnatal Depression Scale (EPDS). The EPDS is a set of 10 screening questions to check the mother for symptoms of emotional distress during postpartum. The EPDS is not a diagnostic tool rather it is developed to assist healthcare workers in screening PPD in conjunction with routine clinical assessment. The EPDS consists of 10 short statements with four possible answers on how the mother has felt during the prior week. Based on the magnitude of symptoms, the mother responded as 0, 1, 2 or 3, and the total score is calculated by adding together the scores for each of the 10 items [41]. Based on previously published studies in Ethiopia [42], we used a score of 13 as a cut-off point to define potential postpartum depression [43, 44]. The EPDS scale is validated in Ethiopia [45].
Independent variables
The independent variables were broadly classified as sociodemographic, obstetric and health service, and psychosocial factors. Sociodemographic factors included women’s age, women’s education, women’s occupation, husband’s education, marital status, ethnicity, sex of the baby, and preference for the gender of the baby. Obstetric and health service factors included the number of pregnancies, number of alive children, desire for the pregnancy, history of abortion, ANC follow-up, place of birth, birth assistance, mode of birth, duration of labour, birth complications, the difficulty of infant feeding, and history of infant illness.
Psychosocial factors included IPV (physical, sexual and emotional violence) and social support (emotional and practical support from husband, mother, father, friend, and other closed ones).
Psychosocial factors are defined as follows:
-
Physical violence includes, but is not limited to: scratching, pushing, shoving, throwing, grabbing, biting, choking, shaking, hair-pulling, slapping, punching, hitting, burning, use of a weapon (gun, knife, or other objects), and use of restraints or one’s body, size, or strength against another person [46].
-
Sexual violence was defined as forcing or attempting to force a person to take part in a sex act, sexual touching, or a non-physical sexual event (e.g., sexting) when the person does not or cannot consent [46].
-
Emotional violence was defined as the use of verbal and non-verbal communication with the intent to harm another person mentally or emotionally and/or to exert control over another person [46].
-
Social Support was measured by six questions which had 5 ordinal choices and with a minimum score of 6 and a maximum of 30. Postpartum women who scored above the mean for social support questions were grouped as ‘average support’, otherwise grouped as ‘below average support’[47].
Data collection and data quality procedures: Given the low level of women’s literacy in Ethiopia, we used an interviewer-administered questionnaire for the data collection process. The questionnaire included four parts: sociodemographic, obstetric and health service and psychosocial factors, and the EPDS scale to measure PPD. The questionnaire was initially prepared in English and translated into the local languages (AfarAf and Amharic), and then translated back to English to confirm its consistency. The tool was pre-tested on 15 postpartum women in other than selected health facilities and amendments were made based on the feedback from the pre-test. Appropriate training was provided for the data collectors (8 diploma nurses) and supervisors (bachelor of midwifery holders) on the main aims of the research, participant recruitment, data collection instrument and procedures, and interview techniques. The data were checked for completeness, consistency and accuracy every day, and the interview was performed in a private room or alone to maintain privacy. The data were entered and cleaned by the principal investigator before analysis.
Data management and statistical analysis
Data were cleaned, coded, and entered into Epi info version 4.2, and exported to Statistical Package for Social Science (SPSS) version 25 for the final analysis [48]. Multivariable logistic regression modelling was used to investigate the relationship between explanatory variables (including sociodemographic, obstetric and health service factors, and psychosocial factors) and PPD. Frequencies, percentages, mean, and standard deviation were used to describe the study participants. Odds ratios (ORs) and 95% CI were used to report the findings from the regression modelling, and P-value < 0.05 was used to declare statistical significance.