Table 1 Studies included for the systematic review on reliability and validity of the Edinburgh Postnatal Depression Scale in low- and lower-middle-income countries
Authors/Year | Setting | Study type | Sample and recruitment | Outcome Variable | Outcome Measured | Psychometric properties of the LLV-EPDS |
|---|---|---|---|---|---|---|
1. Nepal et al., (1999) [24] | Specialised hospital for women and university teaching hospital, located in the Kathmandu (capital), Nepal. | Cross-sectional | • 132/149 postnatal women • Recruited from the maternity wards of respective hospitals (≥2 day’s post-delivery) and followed up after 4 weeks. | Psychometric properties | Cut-off point; Sensitivity (Se); Specificity (Sp); Positive predictive value (PPV); Negative predictive value (NPV) | Nepalese version • Cut-off point: 12/13 • Se: 68.4 • Sp: 93.8 • PPV: 65 • NPV: 94.6 |
2. Regmi et al., (2002) [40] | University teaching hospital, located in the Kathmandu (capital), Nepal. | Cross-sectional | • 100 postnatal women (2–3 months) • Recruited from the post-natal/child immunization clinic. It recruited 40 non-child bearing women as controls (mainly nurses & their friends). | Psychometric properties | Cut-off point; Sensitivity (Se); Specificity (Sp); Positive predictive value (PPV); Negative predictive value (NPV) | Nepalese version • Cut-off point: 12/13 • Se: 100 • Sp: 92.6 • PPV: 41.6 • NPV: 100 |
3. Patel et al., (2002) [28] | Hospital in the Northern town of Goa (Western State), India. | Prospective | • 270/297 pregnant women (≥30 week) • Recruited from the antenatal clinics, and then followed 6–8 weeks after delivery. | Psychometric properties PCMDs prevalence | - Cut-off point; Sensitivity (Se); Specificity (Sp) - PCMDs prevalence | Konkani version: • Cut-off point: 11/12 • Se: 92 • Sp: 85 |
4. Uwakwe (2003) [30] | University teaching hospital, located in the Eastern State, Nigeria. | Cross-sectional | • 225/292 postnatal women • Recruited from the maternity ward (≥7 day’s post-delivery) & then followed in the first postnatal clinic visit. | Psychometric properties | Cut-off point; Sensitivity (Se); Specificity (Sp); Positive predictive value (PPV); Negative predictive value (NPV) | Igbo version (Eastern) • Cut-off point: 8/9 • Se: 75 • Sp: 97 • PPV: 75 • NPV: 97 |
5. Fisher et al., (2004) [23] | Maternal and Child and Family Planning Centre, located in the Ho Chi Minh city, Viet Nam. | Cross-sectional | • Cross-sectional, 506 postnatal women (6–8 weeks) • Recruited from the infant health clinics or those came for medical review | PCMDs prevalence | PCMDs prevalence | Reason included for this Review Earlier Vietnamese version-EPDS (developed by Small et al., (1999) [43] for women living in Australia) was revised before assessment of prevalence |
6. Rahman et al., (2005) [39] | Rural sub-district of Rawalpindi, Pakistan. | Cross-sectional | • 541/570 postnatal women (10 to12 weeks) • Recruited from community. | Psychometric properties | Cut-off point; Sensitivity (Se); Specificity (Sp); Positive predictive value (PPV) | Urdu version • Cut-off point: 9/10a • Se: 81.5 • Sp: 73.5 • PPV: 52.6 |
7. Adewuya et al., (2006) [31] | 5 Health centres, located in a semi-urban town of the Western Nigeria. | Cross-sectional | 182 pregnant women (≥32 weeks) • Recruited from antenatal clinics. | Psychometric properties | Cut-off point; Sensitivity (Se); Specificity (Sp); Positive predictive value (PPV); Negative predictive value (NPV) | Yoruba version (Western) • Cut-off point: 9/10 • Se: 86.7 • Sp: 91.5 • PPV: 68.4 • NPV: 97 |
8. Pollock et al., (2006) [32] | Central Psychiatric Hospital, Mental Health and Necrology Centre, and 3 Primary Health Care Centres (PHCCs), located in the Ulaanbaatar (capital), Mongolia. | Cross-sectional | • 94/100 women (in reproductive age) - Recruited from psychiatric units (55) - Rest from PHCCs immunization clinics | Psychometric properties | Cut-off point; Sensitivity (Se); Specificity (Sp); Positive predictive value (PPV); Negative predictive value (NPV) | Mongolian version • Cut-off point: 12/13 • Se: 80.9 • Sp: 61.7 • PPV: 67.9 • NPV: 76.3 |
9. Gausia et al., (2007) [19] | Social and Behavioural Sciences Unit (SBSU) and Hospital, located in the Dhaka (capital), Bangladesh. | Cross-sectional | • 10 female employee from SBSU • 11 mothers (baby ≤ 1 year) attending immunization clinic • 4 women whose infants were admitted to the hospital | Cultural and operational equivalence of Bangla version EPDS | Correlation between Bangla and original English version Correlation between self-report and interview administration | • Bangla and original English version (0.981; p < 0.01) • Self-report and interview ((0.752; p = 0.01) |
10. Gausia et al., (2007) [25] | Hospital, located in the Dhaka (capital), Bangladesh. | Cross-sectional | • 100 /126 postnatal women (6–8 weeks) • Recruited from child immunization clinic. | Psychometric properties | Cut-off point; Sensitivity (Se); Specificity (Sp); Positive predictive value (PPV); Negative predictive value (NPV) | Bangla version • Cut-off point: 9/10 • Se: 88.9 • Sp: 86.8 • PPV: 40 • NPV: 98.6 |
11. Rowel et al., (2008) [37] | Field polyclinics in Kolonnawa, Western part of the Colombo (capital), Sri Lanka. | Cross-sectional | • 465 perinatal women recruited: - 265 pregnant (≥34 weeks) attending antenatal clinics. - 204 postpartum women (≥6 weeks) attending family planning or child wellbeing clinics. | Psychometric properties | Cut-off point; Sensitivity (Se); Specificity (Sp) | Sinhalese version • Cut-off point: 8/9 Pregnant • Se: 90.7 • Sp: 86.8 Postnatal • Se: 89.9 • Sp: 78.9 |
12. Hanlon et al., (2008) [33] | Butajra (rural region) located 130 km South of the Addis Abba (capital), Ethiopia. | Cross-sectional | • 101 postnatal women (median 5 Months) • Recruited from the community. | Psychometric properties | Cut-off point; Sensitivity (Se); Specificity (Sp) | Amharic version 1. Cut-off point: 5/6 • Sp: 76.5 • Se: 36.1 |
13. Weobong et al., (2009) [38] | Brong-Ahafo region (South part), Ghana. | Cross-sectional | • 160 pregnant women (5–11 week) • Identified from the database of 1/6 districts where vitamin A trial initiated. | Psychometric properties | Cut-off point; Sensitivity (Se); Specificity (Sp); Positive predictive value (PPV); Negative predictive value (NPV) | Twi version • Cut-off point: 10/11 • Se: 78 • Sp: 73 • PPV: 22 • NPV: 97 |
14. Tesfaye et al., (2010) [35] | 2 Primary Health Care Centres, located in peri-urban area of the Addis Ababa (capital), Nigeria. | Cross-sectional | • 100/102 postnatal women (6 to 14 weeks) • Recruited from child immunization and postnatal clinics. | Psychometric properties | Cut-off point; Sensitivity (Se); Specificity (Sp); Positive predictive value (PPV); Negative predictive value (NPV) | Amharic version • Cut-off point: 6/7 • Se: 78.9 • Sp: 75.3 • PPV: 42.9 • NPV: 93.8 |
15. Chibanda et al., (2010) [36] | 2 Primary Health Care Centres located in peri-urban area of the Harare (capital), Zimbabwe. | Cross-sectional | • 210/223 postnatal women (6–7 weeks) • Identified by computer generated randomization of clinic review cards, • Recruited from the 2 primary health care centres | Psychometric properties | Cut-off point; Sensitivity (Se); Specificity (Sp); Positive predictive value (PPV); Negative predictive value (NPV) | Shona version • Cut-off point: 10/11a • Se: 88 • Sp: 87 • PPV: 74 NPV: 94 |
16. Fernandes et al. (2010) [29] | Missionary hospital located in rural part of Karnataka State (South), India. | Cross-sectional | • 194/196 pregnant (32 – 38 weeks) women • Recruited from the antenatal clinic. | Psychometric properties | Cut-off point; Sensitivity (Se); Specificity (Sp); Positive predictive value (PPV); Negative predictive value (NPV) | Kannada version • Cut-off point: 12/13 • Se: 100 • Sp: 84.9 • PPV: 52 • NPV: 99 |
17. Tran et al., (2011) [27] | Randomly selected Commune Health Centres (CMCs) from the Hanoi (capital) and Ha Nam province, Vietnam. | Cross-sectional | • 364/392 perinatal women - 199 were ≥ 28 weeks pregnant - Rest were 4–6 weeks postpartum) • Mostly were recruited from the CMCs and in Ha Nam province, also house visit for some. | Psychometric properties | Cut-off point; Sensitivity (Se); Specificity (Sp); Positive predictive value (PPV); Negative predictive value (NPV) | Vietnamese version • Cut-off point: 3/4a • Se: 69.9 • Sp: 72.9 • PPV: 69.7 • NPV: 72 |
18. Husain et al., (2013) [26] | An urban slum in the Karachi (capital), Pakistan. | Cross-sectional | • 601/664 postnatal women (0–36 months) • Recruited from the slum. | Psychometric properties | Cut-off point; Sensitivity (Se); Specificity (Sp); Positive predictive value (PPV); Negative predictive value (NPV) | Urdu version • Cut-off point: 13/14a • Se: 79 • Sp: 74 • PPV: 82 • NPV: 70 |
19. Stewart et al., (2013) [34] | District hospital, Mangochi, (Southern township), Malawi. | Cross-sectional | • 224 pregnant women (2nd trimester) • Recruited from antenatal clinic. | Psychometric properties | Cut-off point; Sensitivity (Se); Specificity (Sp); Positive predictive value (PPV); Negative predictive value (NPV) | Chichewa version • Cut-off point: 4/5a • Se: 68.7 • Sp: 88.2 • PPV: 35.8 • NPV: 97.4 |