Who Is Most at Risk of Depression at University? Insights From Our New BMC Psychology Study

Who is most at risk of depression at university? Our newly published BMC Psychology study suggests that the answer is more complex than simply identifying one supposedly high-pressure academic discipline. Among 602 university students in Lahore, Pakistan, non-medical students reported higher depressive symptoms than medical students after adjustment, female students reported higher symptom scores across multiple mental-health domains, and longer sleep duration was associated with lower PHQ-9 depression scores. These findings challenge common assumptions about which students deserve the greatest attention and support.

By Taimoor Asghar

The central message is not that every non-medical student, woman, or short sleeper will experience depression. Nor does the study establish that these characteristics cause depression. Instead, the results identify patterns within this particular student sample that may help universities think more carefully about screening, prevention, research, and access to mental-health support.

What Did Our BMC Psychology Study Examine?

Our study, published in BMC Psychology in August 2026, examined mental-health symptoms and the measurement performance of two widely used questionnaires among undergraduate university students in Pakistan. The study included 602 students from universities in Lahore: 424 medical students and 178 non-medical students.

Participants completed the Patient Health Questionnaire-9, or PHQ-9, together with the 21-item Depression Anxiety Stress Scales, commonly known as the DASS-21. Rather than relying on a single statistical comparison, the analysis included descriptive statistics, group comparisons, multivariable regression models, confirmatory factor analysis, item response theory and symptom network analysis.

Readers interested in the complete methods, statistical analyses and results can read the BMC Psychology study on PHQ-9 and DASS-21 symptom profiles among Pakistani university students.

The study was cross-sectional, meaning that students were assessed at one point in time. This distinction matters throughout the interpretation of the findings: associations can identify potentially important groups and variables, but they cannot demonstrate that one factor caused another.

Who Appeared to Have the Highest Depression Risk?

There was no single characteristic that could define a universally “high-risk university student.” However, several findings stood out. Non-medical students had higher depressive symptom scores than medical students, female gender was associated with higher scores across the measured mental-health domains, and greater sleep duration was associated with lower PHQ-9 scores.

These patterns are particularly useful because they move the discussion away from stereotypes. University mental-health programmes are often built around assumptions about which courses are most stressful. Our findings indicate that student vulnerability may not follow those assumptions.

1. Non-Medical Students Reported Higher Depressive Symptoms

One of the most striking findings involved academic discipline. Medical students are frequently portrayed as an especially vulnerable university population because their training combines demanding coursework, examinations, clinical responsibilities and competitive environments. These concerns are legitimate, and extensive research has documented psychological distress among medical students.

However, in our sample, non-medical students reported higher PHQ-9 depression scores. Their median PHQ-9 score was 10 compared with 9 among medical students.

The difference remained after statistical adjustment. In the multivariable analysis, academic discipline was significantly associated with PHQ-9 scores, with non-medical students having approximately 1.43 points higher scores than medical students after accounting for variables included in the model.

A similar pattern appeared when depression was measured through the DASS-21. Non-medical students also had higher adjusted DASS depression scores. They additionally reported higher adjusted DASS anxiety scores, whereas the adjusted difference in DASS stress scores between the academic groups was not statistically significant.

This does not mean that medical students are protected from depression. A one-point difference in group medians also should not be treated as a clinical diagnosis or interpreted as proof that every non-medical student is worse off than every medical student. The distributions of symptoms overlap considerably.

What the finding does show is that mental-health vulnerability cannot safely be inferred from degree programme alone.

Why Might Non-Medical Students Be Vulnerable?

Our cross-sectional data cannot establish why non-medical students reported higher symptom scores. Any explanation therefore needs to remain a hypothesis rather than a conclusion from the study.

Nevertheless, the result raises several important questions for future research. Different university disciplines may expose students to different combinations of academic uncertainty, employment concerns, financial pressures, family expectations, competition and access to institutional support. Some programmes may have established student-support structures because their mental-health pressures are already widely recognized, while distress in other disciplines can attract less attention.

The broad category “non-medical students” is also heterogeneous. It can include students studying subjects with very different workloads, career pathways and socioeconomic contexts. Future studies should therefore compare individual faculties and programmes rather than treating all non-medical education as a single exposure.

Most importantly, universities should avoid designing mental-health services on the assumption that medical students are automatically the group with the greatest need. Support should remain accessible across the university.

Female Students Reported Higher Symptoms Across All Domains

Gender was another important finding. In the adjusted analyses, female gender predicted higher scores on the PHQ-9 as well as the DASS-21 depression, anxiety and stress domains. The associations were statistically significant across all four outcomes.

This pattern deserves attention because it appeared across different symptom measures rather than in a single isolated comparison.

However, it should be interpreted appropriately. The study measured symptoms rather than establishing psychiatric diagnoses, and an association between gender and questionnaire scores does not explain the mechanisms producing that difference.

Potential contributors to student mental health can exist at several levels, including academic pressure, social expectations, financial circumstances, safety concerns, family responsibilities, discrimination, exposure to stressful experiences, coping resources and willingness to report psychological symptoms. Our study was not designed to establish which of these mechanisms explains the observed gender difference.

The practical implication is therefore not that female students should be labelled as depressed. It is that universities and researchers should take gender-related differences seriously and investigate the conditions that may be producing them.

Sleep Was Associated With Depression Scores

Sleep provided another notable finding. Each additional hour of reported sleep was associated with a 0.37-point lower PHQ-9 score in the adjusted model.

This is an association, not evidence that simply adding an hour of sleep will reduce an individual student’s PHQ-9 score by a predictable amount. Sleep and depression have a complicated, potentially bidirectional relationship. Depressive symptoms can interfere with sleep, while inadequate or disrupted sleep can accompany poor psychological well-being.

There is also an important measurement issue: sleep disturbance itself is one of the symptoms assessed by the PHQ-9. Consequently, relationships between sleep behaviour and PHQ-9 scores require thoughtful interpretation.

Even with those limitations, sleep remains a useful variable for universities to consider because student routines frequently involve irregular schedules, late-night study, examinations, commuting, digital-media use and social commitments.

Rather than framing sleep as a personal failure, universities can consider whether timetables, assessment schedules, workload structures and student education support healthier routines.

Risk Factors Are Not the Same as Diagnoses

The phrase “at risk” needs careful handling in mental-health discussions. Finding that a group has a higher average questionnaire score does not mean that every member of that group has depression, and it does not mean that members of other groups are safe from depression.

For example, our results should not be translated into statements such as “non-medical students are depressed” or “women at university have depression.” The study instead found average differences in self-reported symptom scores.

Several distinctions are essential:

• Group-level association is not individual prediction. A statistical difference between groups cannot accurately determine the mental-health status of a specific student.
• Screening is not diagnosis. PHQ-9 and DASS-21 scores provide information about symptoms; a clinical diagnosis requires appropriate professional assessment.
• Association is not causation. Cross-sectional studies cannot establish whether academic discipline, gender or sleep duration caused the observed differences.
• Statistical significance is not automatically clinical significance. The size, context and practical meaning of an observed difference all matter.

These principles are particularly important when research findings are translated into university policy or public communication.

What the PHQ-9 and DASS-21 Tell Us

The study was not only a comparison of student groups. It also evaluated how the PHQ-9 and DASS-21 performed psychometrically in this sample.

The PHQ-9 is a nine-item questionnaire designed to assess depressive symptoms. The DASS-21 contains separate depression, anxiety and stress domains. Because psychological constructs cannot be measured as simply as blood pressure or body temperature, researchers need evidence that questionnaires behave appropriately in the populations in which they are being used.

Our analyses therefore included internal-consistency assessment, confirmatory factor analysis, item response theory and symptom-network methods. This measurement-focused component strengthens the interpretation of the symptom comparisons because it evaluates more than simply calculating questionnaire totals.

At the same time, good psychometric performance does not transform a questionnaire into a standalone diagnostic tool. Screening instruments can identify symptom burden and people who might benefit from further assessment, but they do not replace clinical evaluation.

The Findings Challenge the “Medical Students Are Most Depressed” Assumption

The most useful interpretation of our study may be that university mental health should not be approached as a competition between disciplines.

Medical education deserves serious attention. Medical students can experience intensive workloads, repeated examinations, clinical exposure and substantial professional expectations. But focusing almost exclusively on them can unintentionally obscure distress elsewhere on campus.

Our findings demonstrate exactly why comparative research matters. If researchers only recruit medical students, they can describe distress among medical students but cannot determine whether that distress is higher or lower than in other student populations recruited under comparable conditions.

In our study, the comparison produced a result that might not have been predicted from common narratives: non-medical students showed higher adjusted depressive and anxiety symptom scores.

This should encourage universities to ask a broader question: rather than deciding in advance which faculty is most vulnerable, can we build systems capable of identifying struggling students wherever they study?

What Universities Can Learn From These Results

Provide Mental-Health Support Across Faculties

Counselling, screening and referral pathways should not be concentrated only in programmes that are traditionally regarded as stressful. Students in business, social sciences, humanities, engineering and other disciplines may also experience substantial psychological distress.

Use Data Instead of Stereotypes

Universities can periodically assess student well-being using validated instruments and ethically designed surveys. Data can reveal whether particular faculties, years of study or demographic groups consistently report greater difficulties.

Make Services Easy to Access

A theoretically excellent mental-health service has limited value if students face stigma, long waiting periods, inconvenient locations, uncertainty about confidentiality or lack of awareness that support exists.

Accessible pathways might include student counselling, clearly communicated referral options, faculty awareness, digital appointment systems and procedures for responding appropriately when students report severe symptoms.

Consider Sleep and Academic Structure

The observed relationship between sleep duration and depression scores does not prove causation, but it reinforces the value of studying student routines alongside psychological symptoms. Assessment clustering, early schedules following late academic activities, excessive workload and chronically irregular timetables are institutional factors worth examining.

Avoid Treating Screening Scores as Labels

If universities use questionnaires such as the PHQ-9, they need clear procedures for interpreting results, maintaining confidentiality and directing students toward appropriate assessment. A numerical score should open a pathway to support when necessary rather than becoming a permanent label.

What Students Should Take From the Study

For students themselves, the findings offer an important message: experiencing depression or anxiety symptoms is not something that can be predicted reliably from what you study.

A student in a programme perceived as “less stressful” can still experience substantial distress. Conversely, being enrolled in medicine does not mean poor mental health is inevitable.

Students may benefit from paying attention to persistent changes in mood, interest, concentration, energy, sleep and everyday functioning rather than comparing their struggles with those of students in supposedly more demanding courses.

If symptoms become persistent, severe, interfere with academic or daily functioning, or cause significant distress, seeking professional assessment is appropriate. Thoughts of self-harm or suicide require urgent professional attention through appropriate local emergency or crisis services.

Important Limitations of the Study

Several limitations affect how broadly the findings can be interpreted.

First, the study was cross-sectional. Academic discipline, gender, sleep and symptom scores were measured within the same general study period, which prevents conclusions about the direction of cause and effect.

Second, the study included undergraduate students from universities in Lahore, Pakistan. Student experiences can vary substantially across cities, provinces, institutions, socioeconomic environments and countries. The results should therefore not automatically be generalized to every Pakistani university or to university systems internationally.

Third, the medical and non-medical groups were unequal in size, with 424 medical and 178 non-medical students. Future research with larger and more balanced samples across individual academic disciplines could produce more detailed comparisons.

Fourth, PHQ-9 and DASS-21 responses are self-reported. They measure reported symptoms rather than independently confirmed psychiatric diagnoses.

Finally, although multivariable models can adjust for measured variables, observational studies can still be affected by factors that were not measured or fully captured. Residual confounding is therefore possible.

These limitations do not invalidate the findings. They define what the study can and cannot establish.

What Research Should Come Next?

The results raise several questions that could be addressed through future research.

Longitudinal studies would be particularly valuable because following students over time could clarify whether sleep patterns, academic transitions and other exposures precede changes in depressive symptoms. Multi-university studies could determine whether the differences observed in Lahore are reproduced elsewhere in Pakistan.

Researchers could also examine specific non-medical disciplines separately. Combining many faculties into one non-medical group is useful for an initial comparison, but it cannot show whether certain programmes account for more of the observed difference.

Additional work should investigate potentially modifiable factors such as financial pressure, academic workload, perceived career uncertainty, social support, physical activity, commuting, housing conditions, help-seeking attitudes and access to mental-health services.

Qualitative research could add another layer by asking students directly how they understand the pressures affecting them. Numbers can identify patterns, but interviews and focus groups can help explain why those patterns exist.

Rethinking Who Needs University Mental-Health Support

So, who is most at risk of depression at university? Our study does not provide a universal profile, but it offers several evidence-based clues within this Pakistani student sample.

Non-medical students reported higher depressive symptoms than medical students and retained higher PHQ-9 and DASS depression scores after adjustment. Female students reported higher scores across depression, anxiety and stress outcomes. Greater sleep duration was associated with lower PHQ-9 scores.

The broader lesson is arguably more important than any individual coefficient: mental-health support should not depend on stereotypes about which students are expected to struggle.

Universities need inclusive systems that recognize psychological distress across disciplines, identify potentially vulnerable groups through evidence, and make professional support accessible without turning questionnaire results into diagnoses.

For researchers, the study also illustrates the importance of including comparison groups. We cannot know whether one academic population is relatively more vulnerable if we study it in isolation.

For students, the message is equally straightforward. Your degree programme does not determine whether your mental-health concerns are legitimate. Persistent depression, anxiety or stress symptoms deserve attention regardless of whether you study medicine, engineering, business, humanities or another subject.

Medical disclaimer: This article discusses population-level research findings for educational purposes only. PHQ-9 and DASS-21 scores are not substitutes for an individualized clinical assessment. If you are experiencing persistent or severe psychological symptoms, significant impairment, thoughts of self-harm or thoughts of suicide, seek assessment from an appropriate qualified healthcare professional or urgent local services when immediate safety is a concern.