Medical Students Weren’t the Most Depressed Group in Our Study—Here’s What We Found

Medical students were not the group with the highest depressive symptom scores in our study. Among 602 university students in Lahore, Pakistan, non-medical students had a higher median PHQ-9 score than medical students, and the difference remained statistically significant after adjustment for several relevant factors. Non-medical students also scored higher on the DASS-21 depression and anxiety domains. The finding does not mean medical students are protected from mental-health difficulties. Instead, it challenges the assumption that academic discipline alone tells us who is most psychologically vulnerable.

By Taimoor Asghar

Why We Expected Medical Students to Be at Higher Risk

Medical education has a well-established reputation for psychological strain. Students may face demanding examinations, dense curricula, long study hours, clinical responsibilities, exposure to illness and death, competitive environments, and uncertainty about future training. Because of these pressures, research and public discussion about university mental health often focus heavily on medical students.

That focus is understandable, but it can create an unintended assumption: if medical education is stressful, medical students must necessarily have worse mental health than students studying other subjects.

That is an empirical question rather than a conclusion we should take for granted.

Students outside medicine face their own combinations of academic pressure, financial uncertainty, family expectations, employment concerns, social stressors, sleep disruption, and personal mental-health histories. Those pressures may be less visible in the research literature without being less important to the people experiencing them.

Our study therefore directly compared medical and non-medical university students rather than examining medical students in isolation.

What Our Study Examined

The study included 602 undergraduate students from universities in Lahore. Of these, 424 were medical students and 178 were non-medical students. Participants completed two widely used psychological symptom measures: the Patient Health Questionnaire-9, or PHQ-9, and the 21-item Depression Anxiety Stress Scales, or DASS-21.

The PHQ-9 assesses nine depressive symptoms experienced during the previous two weeks. The DASS-21 assesses three related symptom domains: depression, anxiety, and stress.

We did more than compare average questionnaire scores. The analysis also examined reliability, factor structure, measurement invariance, item-level performance, symptom networks, and demographic or behavioural variables associated with symptom severity. The full peer-reviewed article, Psychometric properties and symptom profiles of the PHQ-9 and DASS-21 among medical and non-medical university students, was published in BMC Psychology in August 2026.

Medical Students Weren’t the Most Depressed Group

The most immediately surprising finding concerned PHQ-9 scores.

Medical students had a median PHQ-9 score of 9, whereas non-medical students had a median score of 10. On a questionnaire with a possible range of 0 to 27, a one-point difference should not be exaggerated as though the two populations were psychologically distinct groups. Nevertheless, the statistical comparison indicated that the difference was unlikely to be explained by random sampling variation alone.

More importantly, the pattern did not disappear after adjustment.

In the multivariable PHQ-9 model, academic discipline remained associated with depressive symptom severity. With the study’s coding of discipline, the coefficient corresponded to non-medical students scoring approximately 1.43 points higher than medical students after adjustment, with a p-value of 0.004.

The DASS-21 results pointed in the same direction. Non-medical students scored higher on the depression domain and the anxiety domain in adjusted models. The estimated differences were approximately 1.82 points for DASS depression and 2.31 points for DASS anxiety. Both were statistically significant.

Stress was different. Although the estimated DASS stress score was also higher among non-medical students, that comparison did not meet the conventional statistical significance threshold in the adjusted model.

What the Finding Does—and Does Not—Mean

The result is easy to oversimplify, so several distinctions matter.

It does not mean medical students have good mental health

A comparison between two groups is relative. Finding higher depressive symptoms among non-medical students does not imply that medical students were psychologically well or that mental-health services for medical trainees are unnecessary.

Both groups contained students experiencing meaningful symptoms. A university mental-health strategy should therefore not replace concern for medical students with concern for non-medical students. It should expand the frame to include the entire student population.

It does not prove that studying a non-medical subject causes depression

This was a cross-sectional study. Academic discipline and psychological symptoms were measured at the same general period rather than followed prospectively over time.

As a result, the analysis can identify associations but cannot establish a causal pathway in which enrolling in a particular discipline produces higher depression or anxiety.

There may also be unmeasured differences between groups. Financial circumstances, employment expectations, family pressures, academic workload, social support, personality, socioeconomic background, physical health, previous psychiatric symptoms, institutional characteristics, and other variables could potentially contribute to group differences.

It does challenge a simplistic risk hierarchy

The finding does provide evidence against treating the label “medical student” as a sufficient indicator of who is most distressed.

If universities direct almost all student mental-health attention toward medicine because medical training is assumed to be uniquely harmful, they may overlook substantial needs elsewhere on campus.

Academic Discipline Explained Only a Small Part of the Picture

One of the most important findings was not simply which group scored higher. It was how little of the overall variation could be explained by the regression models.

The reported R-squared values were approximately 0.029 to 0.041 across the relevant models. In practical terms, the measured predictors explained only a small fraction of the variability in symptom scores.

This is a useful warning against turning a statistically significant group difference into a psychological stereotype.

Two students enrolled in the same degree can have dramatically different experiences. One may have strong social support, stable finances, adequate sleep, and no previous mental-health difficulties. Another may be struggling with family problems, financial pressure, insomnia, isolation, or a history of depression. Their degree title alone tells us relatively little about that complexity.

For student mental health, statistical significance and predictive importance are not the same thing. Academic discipline showed an association, but it was far from a complete explanation.

Female Students Reported Higher Symptoms Across the Measured Domains

Gender showed a broader pattern than discipline. Female gender was associated with higher scores across the PHQ-9 and all three DASS-21 domains in the adjusted analyses, with p-values below 0.001.

This should be interpreted carefully. The result identifies a pattern within this particular sample; it does not establish that gender itself biologically causes greater psychological distress. Gender-related differences in symptom reporting, social expectations, exposure to stressors, support systems, safety, discrimination, financial autonomy, and help-seeking may all influence observed questionnaire scores.

For universities, the practical implication is that subgroup analysis can reveal needs that disappear when only an overall student average is reported. At the same time, screening and support should remain individualized rather than assuming every person in a demographic category has the same level of risk.

Sleep Was Associated With Lower PHQ-9 Scores

Sleep duration was another meaningful variable in the analysis. Each additional reported hour of nightly sleep was associated with a 0.37-point lower PHQ-9 score after adjustment in the primary model.

This finding is compatible with the close relationship between sleep and psychological health, but the cross-sectional design again limits causal interpretation. Poor sleep can contribute to emotional difficulties, depressive symptoms can disrupt sleep, and both may arise from common influences such as academic workload, stress, irregular routines, physical illness, substance use, or social circumstances.

It would therefore be inappropriate to interpret the coefficient as evidence that simply increasing sleep by a particular number of hours will reduce an individual’s depression score by a predictable amount.

What the result does support is the value of treating sleep as part of the student mental-health conversation rather than considering depression purely as an academic-performance issue.

Previous Depression Treatment Was a Stronger Signal Than Academic Discipline

In the expanded PHQ-9 model, a history of previous depression treatment was associated with substantially higher current PHQ-9 scores. The estimated coefficient was 4.15 points, with a p-value below 0.001.

This association is not surprising: students who have previously required treatment for depression may have experienced persistent, recurrent, or more severe symptoms. However, the size of the association is useful when placed next to the much smaller discipline difference.

It reinforces a central message from the study: understanding student mental health requires looking beyond what someone studies.

Previous mental-health history, current symptoms, sleep, personal circumstances, social support, and other individual factors may be far more informative than categorizing a student simply as “medical” or “non-medical.”

The Questionnaires Themselves Also Mattered

A major purpose of the study was to evaluate how the PHQ-9 and DASS-21 performed in this population. Before comparing groups using a psychological scale, researchers should have reasonable confidence that the scale is functioning consistently and measuring similar constructs across the groups being compared.

Internal consistency was good across the instruments, with Cronbach’s alpha values ranging from 0.82 to 0.88.

The DASS-21 three-factor model also showed adequate confirmatory factor analysis fit. Reported indices included a comparative fit index of 0.954, Tucker-Lewis index of 0.948, and root mean square error of approximation of 0.052.

There was, however, an important complication. Correlations between the latent DASS-21 factors were extremely high. The depression-stress correlation was 0.939, while anxiety-stress was 0.949.

That means the three domains could be statistically modelled, but in this sample they were also very strongly intertwined. Researchers and clinicians should therefore be cautious about treating a small difference between DASS depression, anxiety, and stress scores as evidence of completely independent psychological processes.

Measurement Invariance Strengthened the Group Comparison

When two groups are compared using a questionnaire, an important question is whether the instrument operates similarly in both groups. If an item has a different psychological meaning for medical and non-medical students, differences in total scores might partly reflect measurement behaviour rather than true differences in symptoms.

The study’s measurement-invariance analyses supported the use of the scales across academic discipline. This does not make the comparison perfect or causal, but it increases confidence that the observed medical versus non-medical differences were not simply artifacts of fundamentally different questionnaire functioning between the two groups.

Looking Beyond Total Scores: Which Depression Symptoms Were Most Informative?

The study also used item response theory to investigate individual PHQ-9 items. Instead of asking only whether the total score is reliable, this approach examines how much information each symptom contributes at different levels of the underlying depression trait.

Items involving self-worth, concentration, feeling down, and appetite showed relatively high discrimination in the analysis. The loss-of-interest or anhedonia item showed the lowest discrimination parameter, reported as 0.468.

This does not mean anhedonia is clinically unimportant. It remains a core depressive symptom and is part of standard diagnostic assessment. Psychometric discrimination has a narrower meaning: it describes how effectively an item differentiated individuals at different levels of the latent trait within this particular dataset.

The distinction is essential. A symptom can be clinically significant even when it contributes less statistical information to a specific measurement model.

The Symptom Network Highlighted Self-Worth and Concentration

Network analysis provided another way of examining the data. Rather than treating depression, anxiety, and stress solely as hidden variables producing questionnaire responses, symptom-network models examine relationships among the individual symptoms themselves.

Self-worth, concentration, and downheartedness emerged as some of the most central nodes in the combined network.

This result is potentially interesting because it points toward symptoms that are highly connected to other reported difficulties. However, centrality must not be interpreted as proof that changing one symptom would automatically improve the entire network.

The reported centrality-stability coefficients ranged from 0.31 to 0.44, indicating that some network conclusions should be viewed with appropriate caution. Network analysis in cross-sectional data is exploratory and cannot establish the direction of causal relationships among symptoms.

Its strongest contribution here is not a treatment prescription. It is a richer description of student psychological distress than a single total score can provide.

Why Might Non-Medical Students Have Reported More Symptoms?

The study was not designed to establish the mechanism behind the discipline difference, so any explanation remains a hypothesis rather than a demonstrated finding.

Several possibilities deserve future investigation.

• Career uncertainty: students in some non-medical fields may perceive less predictable employment pathways or greater uncertainty about the economic return on their degree.
• Differences in institutional support: medical colleges may have different mentoring structures, peer networks, academic monitoring, or access to health-related resources.
• Selection factors: students entering medicine may differ from other students in socioeconomic background, academic history, family resources, coping style, or other characteristics that were not fully captured.
• Different types of academic pressure: medicine may involve intense workload, while other disciplines may produce different combinations of uncertainty, competition, financial concern, or perceived future instability.
• Unmeasured personal factors: family relationships, financial strain, loneliness, chronic illness, trauma, social support, commuting burden, housing conditions, and many other variables could influence symptoms.

These explanations should not be presented as findings from the study. They are questions generated by the result and should be tested directly in longitudinal and multi-institution research.

Why Cross-Disciplinary Mental-Health Research Matters

A large proportion of student mental-health literature focuses on populations that are easy to define or presumed to be high risk: medical students, nursing students, particular professional programmes, or students attending counselling services.

Such research is valuable, but studying one discipline in isolation cannot tell us whether its students are actually more distressed than their peers.

Suppose a survey finds a high mean depression score among medical students. Without a comparison group, several interpretations remain possible. Medical students may indeed be unusually distressed. The entire university-age population may be experiencing similar symptoms. Another discipline may be doing even worse. Or the apparent burden may partly reflect characteristics of the institution or sample rather than medical education itself.

Direct comparison changes the research question from “Are medical students distressed?” to “How does distress vary across students, and what actually explains those differences?”

The second question is more useful for designing campus-wide policy.

What Universities Can Learn From the Findings

The most defensible implication is not that resources should be moved away from medical students. It is that mental-health systems should avoid assuming that one faculty contains the students most in need.

A broader strategy could include:

• accessible mental-health screening and counselling across faculties rather than only within health-profession programmes;
• clear pathways for students with previous mental-health treatment or recurrent symptoms to obtain appropriate professional support;
• attention to sleep, workload, academic schedules, and other modifiable aspects of student life;
• faculty-specific needs assessments rather than assuming that one intervention fits every discipline;
• confidential systems that reduce barriers created by stigma or concerns about academic consequences;
• appropriate escalation procedures when screening identifies severe symptoms or thoughts of self-harm.

Screening should also be connected to support. Administering a questionnaire without having a pathway for assessment, counselling, referral, or urgent care can identify distress without helping the student who disclosed it.

Important Limitations of Our Study

The results should be interpreted within the boundaries of the study design.

First, the research was cross-sectional. It cannot determine whether academic discipline, sleep patterns, or other variables caused the psychological symptoms observed.

Second, participants were recruited using convenience sampling rather than probability sampling. Students who chose to participate may differ from students who did not respond.

Third, the medical and non-medical groups were unequal in size, with substantially more medical students in the sample.

Fourth, the study was conducted among university students in Lahore. Results should not automatically be generalized to every university in Pakistan, other countries, postgraduate trainees, or people who are not university students.

Fifth, the PHQ-9 and DASS-21 are self-report symptom measures. They are valuable screening and research instruments, but they are not equivalent to a structured clinical diagnosis.

Finally, the regression models explained only a small portion of the variability in scores. Many determinants of student mental health were therefore either unmeasured or not captured adequately by the variables included.

The Bigger Message: Ask Who Is Struggling Instead of Assuming

The headline result—that non-medical students reported higher depressive symptoms than medical students—is interesting because it goes against a familiar narrative. But the deeper lesson is not about deciding which discipline “wins” an unfortunate competition for the highest depression score.

The more important message is that psychological distress does not respect faculty boundaries.

Medical training can unquestionably be demanding. So can life as an engineering student, business student, social-science student, computer-science student, or a young adult trying to complete any degree while navigating financial, family, social, and personal pressures.

Our data suggest that academic discipline was associated with symptoms, but it explained only a small part of the differences between individuals. Female students, sleep duration, and particularly previous depression treatment provided additional signals, while much of the variation remained unexplained.

That should encourage universities and researchers to replace assumptions with measurement. Rather than starting from the belief that one discipline must be the highest-risk group, we should collect comparable data across faculties, investigate the mechanisms behind differences, and build support systems that students can access regardless of what appears on their degree programme.

Frequently Asked Questions

Were non-medical students more depressed than medical students in this study?

Non-medical students reported higher depressive symptom scores. Their median PHQ-9 score was 10 compared with 9 among medical students, and the discipline difference remained statistically significant in adjusted regression analysis. This describes symptom scores in the study sample and does not establish that non-medical education causes depression.

Did non-medical students also have more anxiety?

Yes. In the adjusted DASS-21 model, non-medical students had higher anxiety scores. They also had higher DASS depression scores. The adjusted difference in DASS stress did not reach the conventional threshold for statistical significance.

Does this mean medical students are not at risk of depression?

No. The study compared relative symptom levels between groups. Medical students still reported depressive, anxiety, and stress symptoms, and the findings do not justify reducing mental-health support for them.

Can the PHQ-9 diagnose depression?

The PHQ-9 is a validated symptom questionnaire commonly used for screening and severity assessment. A questionnaire score alone should not be treated as a complete psychiatric diagnosis. Clinical diagnosis requires appropriate professional assessment and consideration of the individual’s history and circumstances.

What was one of the strongest predictors of PHQ-9 score?

A reported history of depression treatment showed a substantial association with current PHQ-9 scores in the expanded regression model. This was a stronger statistical signal than academic discipline, although the cross-sectional analysis cannot determine causal direction.

What should future research investigate?

Future studies should include multiple institutions, larger and more balanced disciplinary groups, longitudinal follow-up, and detailed measures of financial strain, social support, academic workload, career uncertainty, sleep, previous mental health, and other potential explanations for differences between students.

Conclusion

In our sample of 602 university students in Lahore, the assumption that medical students would have the highest depressive symptom burden did not hold. Non-medical students reported higher PHQ-9 scores and higher DASS-21 depression and anxiety scores, including after statistical adjustment. Yet academic discipline explained only a small fraction of overall symptom variation.

The finding argues for a broader approach to university mental health: continue supporting medical students, but do not assume that students elsewhere on campus are doing better simply because their degree is considered less psychologically demanding. Mental-health assessment should follow evidence, individual circumstances, and actual symptoms rather than stereotypes about a field of study.

Medical disclaimer: This article is for educational and research communication purposes only. The PHQ-9, DASS-21, and findings discussed here should not be used to diagnose an individual or replace assessment by a qualified healthcare professional. Anyone experiencing significant psychological distress, thoughts of self-harm, or concerns about their mental health should seek appropriate professional evaluation and urgent assistance when necessary.