The Edinburgh Postnatal Depression Scale (EPDS) is a ten-item self-report questionnaire that screens for depressive symptoms in the perinatal period. Each item asks how the respondent has felt over the past seven days, covering anhedonia, self-blame, anxiety, panic, feeling overwhelmed, sleep difficulty linked to unhappiness, sadness, tearfulness and thoughts of self-harm. Each item is scored 0-3, giving a total of 0-30, with higher scores indicating more depressive symptomatology.
The EPDS was deliberately designed for new mothers, so it avoids somatic items - fatigue, appetite change, sleep disruption - that are near-universal after childbirth and inflate scores on general depression measures. Three items (3, 4 and 5) tap anxiety symptoms, which are common in perinatal presentations, and item 10 asks directly about thoughts of self-harm. It is a screening instrument: a high score signals that a diagnostic assessment is warranted, not that depression is present.
02 - Origin & purpose
Where it comes from.
The scale was developed by John Cox, Jenifer Holden and Ruth Sagovsky at health centres in Livingston and Edinburgh, and published in the British Journal of Psychiatry in 1987. Existing depression questionnaires performed poorly after childbirth precisely because of their somatic content, and health visitors needed a brief, acceptable tool for routine use. In the original validation against Research Diagnostic Criteria diagnoses in 84 mothers, a threshold of 12/13 identified depressed mothers with a sensitivity of 86% and a specificity of 78%.
The EPDS has since become the most widely used perinatal depression screen in the world, translated into dozens of languages and embedded in routine antenatal and postnatal care pathways, including NICE-recommended perinatal mental health assessment in the UK. Validation work has extended it to pregnancy (Murray & Cox, 1990) and to fathers and partners (Matthey et al., 2001), though optimal cutoffs differ in those groups.
03 - Scoring & cutoffs
How scoring works.
Each of the ten items is scored 0-3 according to the response anchors; note that items 1, 2 and 4 are scored in the reverse direction to the others, so use the printed item values rather than assuming a uniform order. The total is a simple sum, range 0-30. Scores of 10-12 suggest possible depression and warrant monitoring and repeat screening within two to four weeks; scores of 13 or more indicate probable depression and should prompt a full clinical assessment. A large individual-participant-data meta-analysis found that a cutoff of >=11 gave the best balance of sensitivity and specificity, while >=13 trades sensitivity for higher specificity - a reasonable choice where assessment capacity is limited. Any score above 0 on item 10 (thoughts of self-harm) requires immediate risk assessment regardless of the total score.
Score
Severity
Interpretation
0–9
Low likelihood
Depression is unlikely on this screen.
10–12
Possible depression
Possible depression. Monitor and consider repeating in two weeks.
13–30
Likely depression
Likely depression. Further assessment is recommended. Any positive response on the self-harm item needs immediate attention.
04 - Validation evidence
How well it performs.
The EPDS has one of the deepest evidence bases of any screening instrument in psychiatry. The definitive accuracy estimate comes from Levis and colleagues' 2020 individual-participant-data meta-analysis in the BMJ, pooling 58 studies and 15,557 pregnant and postpartum women, 2,069 of whom had major depression on a validated diagnostic interview. Against semi-structured interviews, a cutoff of >=11 gave sensitivity 0.81 (95% CI 0.75-0.87) and specificity 0.88 (0.85-0.91); the widely used >=13 cutoff gave sensitivity 0.66 (0.58-0.74) and specificity 0.95 (0.92-0.96). Accuracy was similar in pregnant and postpartum women. Internal consistency is good: the original study reported split-half reliability of 0.88, and validation during pregnancy found Cronbach's alpha of 0.82-0.84 across trimesters (Bergink et al., 2011). The scale is sensitive to change in depression severity over time, and a reliable-change threshold has been formally derived (see Confidence & precision).
0.81
Sensitivity at cutoff >=11 against semi-structured diagnostic interview (Levis et al., 2020)
0.95
Specificity at cutoff >=13; the stricter threshold trades sensitivity for fewer false positives (Levis et al., 2020)
0.82-0.84
Cronbach's alpha across trimesters of pregnancy (Bergink et al., 2011)
15,557
Participants in the individual-participant-data meta-analysis underpinning the accuracy estimates (Levis et al., 2020)
05 - How it compares
How it compares to the alternatives.
Instrument
Items
Time
When to reach for it
EPDS
10 items
3-5 min
First-line depression screen during pregnancy and the postnatal year; avoids somatic items confounded by childbirth.
Epidemiological research on depressive symptoms in community samples.
06 - When to use it
Right tool, wrong tool.
Reach for it when
-Routine depression screening at booking, later in pregnancy, and at postnatal checks (e.g. 6-8 weeks)
-Monitoring symptom change across the perinatal period, using the 4-point reliable-change threshold
-Screening fathers and partners, with the caveat that a lower cutoff (5/6) has been proposed for men
-Settings where somatic symptoms of new parenthood would confound a general depression measure
Reach for something else when
-Diagnosing depression - a positive screen needs a clinical interview; the EPDS is not a diagnostic instrument
-Standalone suicide-risk assessment - item 10 flags thoughts of self-harm but does not quantify risk
-Detecting the full range of perinatal mental illness - it will miss psychosis, PTSD and most anxiety disorders beyond the three anxiety-toned items
-General adult depression screening outside the perinatal context, where the PHQ-9 is better validated
07 - Confidence & precision
Reading the score with care.
Like all brief scales, the EPDS carries measurement error, so single-point differences should not be over-interpreted. Matthey (2004) formally derived a reliable-change threshold from the scale's reliability: a change of at least 4 points between administrations is needed for 95% confidence that mood has genuinely changed rather than the score simply fluctuating within measurement error. In practice, treat a fall from 16 to 14 as noise and a fall from 16 to 11 as meaningful improvement. Scores sitting exactly on a cutoff deserve a repeat administration two weeks later rather than a definitive judgement.
08 - Limitations
What it cannot tell you.
The EPDS is a screen, not a diagnosis: at >=13, a third of true cases score below threshold, and at >=11 around one in eight non-depressed women screen positive, so both false reassurance and unnecessary referral occur at scale. Optimal cutoffs vary across languages and cultures, and translated versions should use locally validated thresholds where available. The scale performs differently in men, requiring a lower cutoff. It does not assess symptom duration or functional impairment, both required for diagnosis. Item 10 detects thoughts of self-harm but is not a substitute for structured risk assessment. Finally, although three items carry anxiety content, the EPDS should not be relied on to detect perinatal anxiety disorders.
See how Aisel removes friction where it costs most. A 20-minute walkthrough tailored to your clinic.
We value your privacy
We use cookies to analyse site usage and improve your experience. Analytics and embedded media (e.g. YouTube) only load if you accept. Read our cookie policy.