Research story · accepted
Psychometric properties and symptom profiles of the PHQ-9 and DASS-21 among medical and non-medical university students: a cross-sectional study in Pakistan
Non-medical students reported slightly higher depression and anxiety symptoms, but academic discipline explained little of the overall difference.
01
Why this matters
University mental health support is often built around the assumption that medical students carry the greatest burden. This study tests that assumption in Pakistan and asks whether the PHQ-9 and DASS-21 perform reliably enough to support comparisons between medical and non-medical students. The central message is not that one discipline is healthy and the other is unwell. Gender, sleep, and previous depression treatment were more informative than discipline alone.
02
Study snapshot
- Design
- Cross-sectional online survey
- Sample
- 602
- Population
- 424 medical and 178 non-medical university students; mean age 21.62 years
- Setting
- Universities in Lahore, Pakistan
- Measures
- PHQ-9 and DASS-21, plus demographic, discipline, treatment-history, substance-use, and sleep variables
- Analysis
- Robust linear regression, reliability analysis, confirmatory factor analysis, measurement invariance, item response theory, and symptom network analysis
Adjusted PHQ-9 point difference for medical versus non-medical students; 95% CI -2.40 to -0.45. The difference was statistically detectable but small.
03
Results, without the jargon
This cross-sectional study compared depressive, anxiety, and stress symptoms among 602 medical and non-medical university students and evaluated the psychometric performance of the PHQ-9 and DASS-21. Non-medical students reported slightly higher depression and anxiety scores after adjustment, while academic discipline explained only a small proportion of the variance. Female gender, shorter sleep duration, and prior depression treatment were associated with higher symptom scores. Both instruments showed good internal consistency. Confirmatory factor analysis supported the DASS-21 structure, although very high correlations among its latent factors may limit subscale distinction in this population.
Median PHQ-9 score
Non-medical median 10, IQR 7 to 14; medical median 9, IQR 5 to 13.
Source: Accepted manuscript, Table 2aAdjusted discipline difference
Medical students scored 1.43 points lower; 95% CI -2.40 to -0.45, p = .004. Model R squared was 3.4%.
Source: Accepted manuscript, Table 3aPrevious depression treatment
Adjusted PHQ-9 difference; 95% CI 3.15 to 5.15, p < .001.
Source: Accepted manuscript, Table 3aFemale gender
Adjusted PHQ-9 difference; 95% CI 0.72 to 2.65, p < .001.
Source: Accepted manuscript, Table 3aEach additional sleep hour
Adjusted PHQ-9 difference per average hour; 95% CI -0.61 to -0.05, p = .021.
Source: Accepted manuscript, Table 3aPHQ-9 reliability
Cronbach alpha. DASS subscale alphas ranged from 0.854 to 0.879.
Source: Accepted manuscript, Table 5aDASS-21 model fit
TLI 0.948, RMSEA 0.052, SRMR 0.031. Overall fit was good.
Source: Accepted manuscript, Tables 4a to 4cAnxiety-stress factor correlation
Near-unity overlap questions how distinct the subscales are in this sample.
Source: Accepted manuscript, Table 4c04
What this could mean
Campus mental health services should serve the whole university rather than assume only medical students need support. Previous treatment history and current symptom burden may be more useful for outreach than discipline. Sleep support is a reasonable wellness target, but this study cannot establish that more sleep caused lower scores. The DASS-21 total distress score may deserve display alongside its subscales in Pakistani student research.
05
Read this with the limitations
Cross-sectional data cannot determine cause. Convenience sampling and a Lahore-only sample limit generalizability. All outcomes were self-reported. Examination timing, socioeconomic status, and academic year were not measured. English-only administration may have affected participation. Measurement invariance across sex was not tested. Two PHQ-9 items showed model misfit. The exploratory network was only moderately stable. Low adjusted R squared values mean about 96% to 97% of symptom variation remained unexplained.
06
How I contributed
I led this study across the research lifecycle, from shaping the question and study design to coordinating the team, guiding student researchers, analysing and interpreting the findings, developing the manuscript, and supporting the publication process. The work reflects my wider focus on research mentorship, rigorous methods, and making complex evidence easier to understand.
07
Explore and cite the paper
Asghar T, Hassan A, Sahar I, Tahir M, Shahid B, Komal K. Psychometric properties and symptom profiles of the PHQ-9 and DASS-21 among medical and non-medical university students: a cross-sectional study in Pakistan. BMC Psychology. 2026. doi:10.1186/s40359-026-05332-5.
This is a plain-language research summary, not personal medical advice. Interpret findings in the context of the design and limitations.