The Medical Student Mental Health Assumption May Be Too Simple

The idea that medical students are automatically the most psychologically distressed group on a university campus is appealing because it sounds plausible: medical training is demanding, competitive, and emotionally intense. But evidence from a recent study of 602 university students in Lahore, Pakistan suggests that the comparison is more complicated. In this sample, non-medical students reported higher depressive and anxiety symptoms than medical students, while academic discipline itself explained only a small fraction of the differences between individuals.

By Taimoor Asghar

That finding does not mean medical students have good mental health, nor does it imply that concerns about medical training are misplaced. Instead, it highlights a more useful principle: student mental health cannot be understood reliably from academic labels alone. Gender, sleep, previous mental-health treatment, individual circumstances, and overlapping symptom patterns may tell us more than whether someone studies medicine.

Why the medical student mental health assumption is so persuasive

Medical education has several characteristics that make psychological distress a reasonable concern. Students face large volumes of material, frequent examinations, competitive environments, clinical responsibility, exposure to illness and death, and pressure to perform professionally. These features have understandably made medical students a prominent population in mental-health research.

The problem arises when a legitimate concern becomes an untested hierarchy: medical students are stressed, therefore they must be more depressed or anxious than students studying other subjects.

Those are different propositions.

Demonstrating substantial psychological symptoms within a medical-school population does not demonstrate that the symptoms are higher than in an appropriate comparison group. To answer the comparative question, researchers need medical and non-medical students assessed under reasonably similar conditions using the same instruments and analytical framework.

That distinction matters because university students outside medicine also encounter significant pressures. Financial insecurity, employment concerns, family expectations, uncertain career pathways, academic competition, relationship difficulties, housing problems, disrupted sleep, social isolation, and limited access to mental-health care are not exclusive to medical schools. Some of these pressures may even differ systematically between academic disciplines.

What our Pakistani university study found

Our study, published in BMC Psychology in August 2026, examined 602 undergraduate students from universities in Lahore, Pakistan. The sample included 424 medical students and 178 non-medical students. Participants completed the Patient Health Questionnaire-9, or PHQ-9, and the Depression Anxiety Stress Scales-21, commonly known as the DASS-21.

The full study, Psychometric properties and symptom profiles of the PHQ-9 and DASS-21 among medical and non-medical university students, did more than compare average questionnaire scores. We also examined adjusted associations, reliability, confirmatory factor structure, item response characteristics, measurement invariance, and symptom networks.

The most immediately striking comparison was that non-medical students had a higher median PHQ-9 score than medical students: 10 compared with 9.

After adjustment for measured covariates, the same general pattern remained. Non-medical students had higher scores on the PHQ-9, DASS-21 Depression, and DASS-21 Anxiety domains. The difference in DASS-21 Stress scores, however, was not statistically significant.

This is important because it prevents an overly simple interpretation. The data did not indicate that every dimension of psychological distress was uniformly worse among non-medical students. Instead, differences depended partly on what was being measured.

Academic discipline explained relatively little

Perhaps the more important result was not which group had the higher adjusted score. The regression models explained only modest amounts of the overall variation in symptoms, with reported R-squared values ranging from approximately 0.029 to 0.041.

In practical terms, knowing whether a participant belonged to the medical or non-medical group was far from sufficient to explain that person’s mental-health profile.

This distinction between statistical association and explanatory importance is essential. A group difference can be statistically detectable while still accounting for only a small proportion of the enormous variation that exists between individual students.

That is one reason headlines such as “medical students are more depressed” or “non-medical students are more depressed” can both obscure what the data actually show. Population averages are useful for identifying patterns, but they are poor substitutes for understanding an individual.

The better question is not simply who has the higher score

Comparative studies often encourage a ranking mentality. Which faculty has the worst depression? Which discipline is most anxious? Which students are under the greatest stress?

Those questions can be useful for resource planning, but they can also distract from more informative questions:

• Which student characteristics are consistently associated with greater symptoms?
• Which symptoms carry the greatest measurement information?
• Do common screening instruments behave similarly across different academic groups?
• Are depression, anxiety, and stress genuinely distinct in the population being studied?
• Which students are likely to be overlooked when support programs focus on one high-profile discipline?

Our findings suggest that these questions deserve at least as much attention as the medical-versus-non-medical comparison.

Gender was more consistently associated with symptoms

Female gender predicted higher scores across the PHQ-9 and all three DASS-21 domains in the adjusted analyses, with statistically significant associations across the models.

This does not mean gender alone determines mental health, and a cross-sectional study cannot establish why such differences appeared. It does show, however, that academic discipline was not the only meaningful grouping variable in the dataset.

For universities designing mental-health programs, that distinction has practical consequences. If support is built primarily around the assumption that medical students constitute the main high-risk population, substantial distress in other student groups may receive less attention.

A more defensible strategy would combine universal mental-health support with targeted services informed by locally observed patterns rather than by professional stereotypes.

Sleep may matter across academic boundaries

Another notable finding involved sleep. Each additional reported hour of sleep was associated with a lower PHQ-9 score in the adjusted model.

Because the study was cross-sectional, this association should not be interpreted as proof that increasing sleep by a particular amount will directly reduce someone’s depression score. The relationship between sleep and depressive symptoms can be bidirectional and influenced by many other factors. Depression may disrupt sleep, insufficient sleep may worsen emotional functioning, and both may reflect additional social or health circumstances.

Nevertheless, the result illustrates why student mental health should not be reduced to faculty membership. Sleep schedules can be affected by examinations, commuting, employment, social routines, digital-device use, household responsibilities, and many other aspects of student life.

Universities interested in mental health therefore need to consider the broader environment in which students live, not only the academic curriculum they follow.

Previous depression treatment was a particularly strong marker

Prior treatment for depression was the strongest predictor reported in the PHQ-9 model, with an adjusted coefficient of 4.15.

This finding should be interpreted carefully. Previous treatment does not cause higher current symptom scores. More plausibly, it identifies students with a history of clinically meaningful psychological difficulty or recurrent vulnerability.

From a research perspective, this illustrates another weakness of simplistic group comparisons. Two students from different faculties may share far more clinically relevant characteristics than two students from the same faculty. Mental-health history, current life circumstances, social support, sleep, financial pressures, and other factors may cut across disciplinary boundaries.

PHQ-9 and DASS-21 results also revealed measurement complexity

The study was designed not only to compare groups but also to examine how two widely used mental-health questionnaires performed in this student population.

Internal consistency was acceptable to good across the assessed scales, with Cronbach’s alpha values ranging from 0.82 to 0.88. Measurement invariance across academic discipline was also supported, which is useful because group comparisons become difficult to interpret if an instrument functions fundamentally differently in the groups being compared.

At the same time, the DASS-21 results raised an interesting psychometric question. Confirmatory factor analysis produced adequate global model fit, but correlations among the latent Depression, Anxiety, and Stress factors were extremely high. The reported Depression-Stress correlation was 0.939, while the Anxiety-Stress correlation reached 0.949.

These values suggest that although the conventional three-factor structure could be fitted to the data, the underlying constructs were very strongly overlapping in this sample.

Why overlapping symptoms matter

Depression, anxiety, and stress are conceptually distinguishable, but real psychological experiences rarely remain inside neat questionnaire categories. Difficulty relaxing, low mood, worry, fatigue, sleep disruption, concentration problems, and physiological arousal may coexist.

Very high correlations between latent factors therefore deserve attention. They do not automatically invalidate the DASS-21, but they caution against assuming that every subscale represents a completely independent psychological dimension in every population.

This is another example of why the medical student mental health assumption may be too simple. Even before comparing academic groups, researchers have to consider what exactly their scales are measuring and how distinct those constructs appear in the population under study.

Individual symptoms may provide information hidden by total scores

A total depression score is useful because it condenses several responses into a single interpretable number. But two students with the same total score can have very different symptom profiles.

Our item response analysis found that PHQ-9 items concerning self-worth, concentration, feeling down, and appetite showed relatively high discrimination. The loss-of-interest or anhedonia item showed the lowest discrimination parameter in the reported analysis.

Network analysis added another perspective. Self-worth, concentration, and downheartedness emerged among the most central symptoms in the estimated network.

Network centrality should not be interpreted as proof that treating one symptom will automatically improve all the others. The stability coefficients in the analysis were moderate rather than perfect, and network results from cross-sectional data are primarily descriptive. Even so, the convergence of several analytical methods on certain cognitive and self-evaluative symptoms is noteworthy.

For researchers, it shows why examining item-level information can reveal patterns that an overall mean score cannot.

A group average is not a diagnosis

One of the easiest mistakes in mental-health reporting is to move too quickly from questionnaire scores to clinical conclusions.

The PHQ-9 and DASS-21 are valuable research and screening instruments, but a questionnaire score is not equivalent to an individualized psychiatric diagnosis. Screening tools estimate symptom burden using standardized questions. Clinical assessment incorporates duration, functional impairment, history, differential diagnoses, medical conditions, substance use, risk, context, and professional judgment.

Likewise, an average difference between medical and non-medical students cannot tell us whether a particular medical student is doing well or whether a particular non-medical student requires support.

The most defensible use of these results is at the population level: they challenge assumptions about where psychological distress is concentrated and encourage broader student support.

What the study does not prove

The findings should not be interpreted beyond the design of the research.

It does not prove that non-medical education causes depression

The study was cross-sectional, meaning exposure and symptom information were observed within the same general time frame. It cannot establish that studying a non-medical subject causes greater depressive or anxiety symptoms.

It does not prove that medical students are psychologically healthier

The comparison concerns average questionnaire scores in one sample. Medical students may still experience substantial psychological distress, and the results should not be used to minimize the pressures associated with medical education.

It does not establish a universal Pakistani ranking

Participants were university students in Lahore, and the two academic groups were not equal in size. Educational environments vary substantially between institutions, cities, subjects, socioeconomic groups, and countries.

It does not identify the mechanisms behind the difference

Although the regression analyses included measured covariates, the models explained only a small portion of the variation in symptoms. Numerous unmeasured or incompletely measured factors could contribute to the observed differences.

Future longitudinal and multi-institutional studies would be better suited to exploring how symptoms change over time and whether specific educational, social, financial, or personal exposures account for cross-disciplinary differences.

What universities can learn from the findings

The practical lesson is not to transfer attention from medical students to non-medical students. It is to avoid constructing campus mental-health systems around the assumption that one faculty owns the problem.

A broader approach could include:

• mental-health services accessible to students across all disciplines;
• screening and referral pathways that do not depend on faculty membership;
• attention to sleep, academic schedules, financial strain, and other modifiable environmental pressures;
• targeted outreach where local data identify groups with greater symptom burden;
• clear pathways for students with previous mental-health treatment or recurrent difficulties;
• training for faculty and student-support staff to recognize distress without attempting to diagnose it;
• repeated evaluation of whether commonly used screening instruments perform appropriately in the local population.

Most importantly, universities should collect their own data when possible. The risk profile of students at one institution cannot automatically be imported from another university, country, or discipline.

Why this changes the way we should discuss medical student mental health

There is a major difference between saying “medical students face important mental-health risks” and saying “medical students necessarily have worse mental health than everyone else.” The first statement can be supported without requiring a comparison. The second is an empirical claim that needs comparative evidence.

Our study provides one example in which the expected ranking did not appear. Non-medical students reported higher depressive and anxiety symptoms after adjustment, yet discipline explained little of the total individual variation.

That combination of findings is more informative than either group comparison alone. It tells us that the stereotype can fail, but it also tells us not to replace it with a new stereotype.

The appropriate conclusion is not that non-medical students are the new universal high-risk group. The conclusion is that mental health is heterogeneous, multidimensional, and shaped by factors that extend beyond a student’s academic title.

The larger lesson: measure rather than assume

Student mental-health research is most useful when it tests familiar beliefs rather than reproducing them. Medical students deserve serious attention because their training contains genuine psychological pressures. But other students deserve the same evidence-based consideration.

When we compare groups directly, examine individual predictors, test whether our questionnaires actually behave as intended, and look beyond total scores to symptom patterns, a more complicated picture emerges.

That complexity is not a weakness. It is a better representation of how mental health actually works.

Instead of asking which academic discipline should carry the label of “most distressed,” universities and researchers may gain more by asking which students are struggling, what factors accompany that struggle, whether our measurement tools capture it adequately, and how support can reach people before an academic stereotype determines who receives attention.

Medical disclaimer: This article is for educational and research communication purposes only. PHQ-9 and DASS-21 scores are not substitutes for an individualized clinical assessment or diagnosis. Anyone experiencing persistent psychological distress, functional impairment, thoughts of self-harm, or other concerning symptoms should seek evaluation from an appropriately qualified healthcare professional or urgent local services when necessary.