Sleep, Gender and Depression: What We Found Among Pakistani University Students

Among 602 undergraduate students in Lahore, our study found two particularly relevant patterns: female students reported higher symptom scores across depression, anxiety and stress measures, while each additional hour of reported sleep was associated with a lower PHQ-9 depression score. These findings do not prove that being female or sleeping less causes depression. They do, however, suggest that university mental-health strategies should look beyond academic discipline alone and pay closer attention to sleep health, gender-related differences and students with an existing history of mental-health treatment.

By Taimoor Asghar

Our recently published BMC Psychology study examined depressive symptoms among medical and non-medical university students in Pakistan while also evaluating the psychometric performance of two widely used mental-health measures: the Patient Health Questionnaire-9 (PHQ-9) and the Depression Anxiety Stress Scales-21 (DASS-21). Although the comparison between academic disciplines was one of the original questions, the adjusted analyses revealed a broader story. Sleep, gender and previous depression treatment were among the variables associated with students’ symptom scores.

This matters because university mental health is rarely explained by a single characteristic. A student’s degree programme may contribute to their experience, but so may sleep schedules, previous mental-health difficulties, social circumstances, academic demands and many other factors that cannot be reduced to a medical-versus-non-medical comparison.

The full research article, our BMC Psychology study of PHQ-9 and DASS-21 symptom profiles among Pakistani university students, was published on 12 August 2026. Here, I focus specifically on what the results tell us about sleep, gender and depressive symptoms—and, equally importantly, what they do not tell us.

What did our Pakistani university student study examine?

We conducted a cross-sectional survey involving 602 undergraduate students from universities in Lahore, Pakistan. The sample included 424 medical students and 178 non-medical students. Participants completed the PHQ-9 and DASS-21, allowing us to examine depressive symptoms using the PHQ-9 as well as depression, anxiety and stress symptom domains using the DASS-21.

The study went beyond simply comparing average questionnaire scores. The analyses included descriptive statistics, Wilcoxon rank-sum testing, multivariable linear regression, confirmatory factor analysis, item response theory and symptom network analysis. We also evaluated reliability and whether measurement properties were sufficiently comparable across academic disciplines.

That broader analytical approach is useful because a questionnaire score can tell us how many symptoms participants report, but psychometric analyses help address another question: how well are the instruments themselves functioning in this population?

For the present discussion, however, three findings deserve particular attention:

• Female gender was associated with higher scores across the PHQ-9 and all three DASS-21 symptom domains in adjusted analyses.
• More reported sleep was associated with lower PHQ-9 depressive symptom scores.
• A previous history of treatment for depression showed the strongest association with PHQ-9 scores among the reported predictors.

These findings need to be interpreted as associations rather than evidence of cause and effect.

Sleep and depression among Pakistani university students

In the adjusted PHQ-9 regression model, each additional reported hour of sleep was associated with a 0.37-point lower PHQ-9 score, with a p-value of 0.010. Put simply, students who reported sleeping for longer tended to report fewer depressive symptoms after accounting for the other variables included in the model.

It would be easy to turn this into the claim that an extra hour of sleep reduces depression by 0.37 points. That interpretation would be incorrect.

The study was cross-sectional. Sleep and depressive symptoms were measured within the same observational study rather than by experimentally changing sleep duration and following subsequent changes in depression. We therefore cannot determine the direction of the relationship from these data.

Why direction matters

At least three explanations are plausible.

First, insufficient or disrupted sleep may contribute to worse emotional functioning and greater depressive symptom burden. Sleep and mental health are closely interconnected, and broader research has repeatedly identified associations between sleep disturbance and depression.

Second, depression itself can change sleep. Sleep disturbance is a familiar feature of depressive illness, and people experiencing depression may sleep too little, have fragmented sleep, struggle with sleep onset, wake early or, in some cases, sleep substantially more than usual.

Third, both sleep and depressive symptoms may be influenced by other factors. Examination periods, financial pressure, family difficulties, physical illness, social isolation, irregular schedules, substance use, nighttime screen exposure, living arrangements and many additional circumstances could affect both variables.

Our finding therefore supports sleep as a potentially useful marker within student mental-health assessment, but it does not establish sleep duration as an independent treatment for depression.

Why sleep deserves attention on university campuses

The importance of sleep in university populations is not limited to our study. University life frequently creates conditions that can challenge healthy sleep: changing schedules, late-night studying, early classes, commuting, social activities, digital-media use and academic deadlines can all affect when and how students sleep.

A 2022 systematic review and meta-analysis by Chandler and colleagues examined psychological and behavioural sleep interventions in university students. Across the included evidence, sleep interventions improved sleep disturbance and were also associated with small improvements in depression and anxiety outcomes. That does not mean sleep interventions replace established depression treatment, but it strengthens the case for considering sleep within a broader university mental-health strategy.

Another systematic review and meta-analysis by Gardani and colleagues found moderate associations between poor sleep quality or insomnia symptoms and stress among undergraduate students. These studies reinforce an important principle: sleep is not merely a lifestyle detail disconnected from psychological health.

For universities, the practical implication is not that every student with low mood should simply be told to sleep more. A more useful approach is to recognize persistent sleep difficulty as something worth asking about when assessing student well-being.

Sleep duration is not the whole story

Our study used reported sleep hours, which is a relatively simple indicator. Sleep health is much more multidimensional. Duration does not capture every relevant characteristic of sleep.

Two students may both report seven hours of sleep while having very different experiences. One may sleep continuously on a regular schedule and wake refreshed. Another may spend the same number of hours asleep but experience repeated awakenings, irregular bedtimes or significant daytime fatigue.

Future studies in Pakistani university populations could therefore benefit from investigating sleep quality, insomnia symptoms, bedtime consistency, chronotype, daytime sleepiness and objectively measured sleep where feasible. Longitudinal research would be especially valuable because it could examine whether changes in sleep precede changes in depressive symptoms or vice versa.

Gender and depression: what did we find?

Gender was another notable predictor in our analyses. Female students had higher symptom scores across all four examined outcomes: PHQ-9 depression, DASS-21 depression, DASS-21 anxiety and DASS-21 stress. The associations were statistically significant at p<0.001 across these domains.

This consistency is noteworthy. The finding was not limited to one questionnaire or only to depressive symptoms. Female gender was associated with greater reported symptom burden across multiple measures of psychological distress.

However, the correct conclusion is that female students in this particular sample reported higher symptom scores—not that gender itself biologically determines who will develop depression.

Gender is not a simple causal variable

Gender-related differences in mental health can reflect a complex mixture of biological, psychological, social and cultural influences. Potential contributors can include differences in exposure to stressors, safety concerns, social expectations, caregiving responsibilities, discrimination, economic circumstances, willingness to disclose emotional symptoms and patterns of help-seeking.

Our study was not designed to determine which of these mechanisms explained the observed gender differences. We should therefore avoid attaching a causal explanation that the data did not test.

The result is better treated as a signal for further investigation and a reminder that student support services should be capable of recognizing groups that may report a greater burden of symptoms without assuming that every individual within that group has the same experience.

How do these findings fit with depression more broadly?

The World Health Organization describes depression as a common mental disorder and notes that it affects women more frequently than men globally. Depression can involve persistent low mood or loss of interest alongside symptoms affecting thinking, sleep, appetite, energy and everyday functioning.

Our results are therefore consistent with the broader observation that depressive symptom burden can differ by gender, but a single Pakistani university sample cannot establish whether the same magnitude of difference exists across Pakistan or among all university populations.

Cultural context is particularly important. Universities differ greatly in admission systems, socioeconomic composition, residence patterns, academic structures, student support services and gender norms. Findings from Lahore should not automatically be generalized to students in every province, university or educational setting.

Previous depression treatment was an especially strong marker

One of the most important findings can easily be overlooked when discussing gender or sleep. Previous depression treatment was the strongest reported predictor of PHQ-9 scores in the adjusted model, with an estimated coefficient of 4.15 and a p-value below 0.001.

This should not be interpreted to mean that depression treatment increases depressive symptoms. A far more plausible interpretation is that a history of treatment identifies students who have previously experienced depression severe or persistent enough to come to clinical attention. Some may continue to experience symptoms or may be vulnerable to recurrence.

This distinction is essential. Observational regression coefficients describe relationships in the dataset; they do not automatically describe the causal effect of the variable itself.

From a campus-support perspective, the finding suggests that previous mental-health history deserves attention when students seek help. Continuity of care may be particularly important for people arriving at university with a previous history of clinically significant symptoms.

Sleep, gender and academic discipline should be considered together—not in isolation

Our original research also compared medical and non-medical students. Contrary to the common assumption that medical students necessarily have the highest depressive burden, non-medical students in our sample had higher PHQ-9 scores, with a median of 10 compared with 9 among medical students.

Adjusted analyses likewise found differences in PHQ-9 depression, DASS-21 depression and DASS-21 anxiety by academic discipline, while the adjusted DASS-21 stress difference was not statistically significant.

Yet the statistical models explained only a small proportion of the total variation in symptoms. Reported R-squared values ranged from 0.029 to 0.041. In practical terms, most of the variation in student mental-health scores remained unexplained by the predictors included in these models.

This is one of the most useful findings of the study because it prevents an overly simple narrative.

Depression among university students cannot be adequately explained by whether someone studies medicine, their gender or the number of hours they report sleeping. These variables may be associated with symptom levels, but a large amount of individual variation remains.

Future research should consider a wider range of possible influences, including financial strain, academic workload, social support, loneliness, chronic illness, family circumstances, substance use, physical activity, relationship stress, experiences of harassment or discrimination, living conditions and access to mental-health care.

What the PHQ-9 result does—and does not—mean

The PHQ-9 is a symptom questionnaire used extensively in clinical and research settings. It assesses nine symptoms associated with depressive disorders. It can be extremely useful for screening and tracking symptom severity, but questionnaire scores should not be confused with a complete psychiatric diagnosis.

A diagnosis requires clinical interpretation that considers symptom duration, functional impairment, differential diagnoses, medical conditions, medications, substance use, possible bipolar-spectrum symptoms and the wider clinical context.

This matters when discussing research on university students. Saying that a group had a higher average PHQ-9 score is not equivalent to saying that everyone in that group had clinical depression.

Our psychometric analyses also highlighted that individual symptoms do not contribute identically to measurement. Item response theory indicated that self-worth, concentration, feeling down and appetite-related items showed relatively high discrimination in our sample, whereas the interest or anhedonia item showed lower discrimination. In the symptom network analysis, self-worth, concentration and downheartedness were among the most central nodes.

These findings emphasize that depression is more than a total score. Understanding which symptoms are especially informative or interconnected may eventually help researchers develop more nuanced approaches to screening, although our network results require caution because stability coefficients were modest.

What universities in Pakistan can reasonably take from these findings

The study does not provide evidence for a single intervention, but it does support several practical priorities for university mental-health systems.

• Do not restrict mental-health attention to medical students. Non-medical students in our sample reported substantial symptoms and, on several measures, higher scores.
• Include sleep in student well-being conversations. Persistent short or disturbed sleep may accompany psychological difficulties and warrants appropriate assessment rather than dismissal as an inevitable part of university life.
• Recognize gender-related differences without stereotyping individuals. Female students reported greater symptom burden in our sample, but support should remain accessible to students of every gender.
• Ask about previous mental-health difficulties. Students with a history of depression treatment may represent an important group for continuity of support and early recognition of recurrent symptoms.
• Avoid relying on questionnaire scores alone. Screening tools can identify possible concerns, but they do not replace clinical evaluation where that is needed.

Universities could also consider mental-health literacy programmes, clearly signposted counselling pathways, staff training for recognizing students in difficulty and policies that reduce unnecessary barriers to seeking professional support.

Important limitations of our study

Responsible interpretation requires acknowledging what the study cannot establish.

First, it was cross-sectional. Because variables were assessed at one period rather than followed prospectively, temporal direction cannot be determined. We cannot conclude that less sleep caused greater depression, for example.

Second, the data were self-reported. Sleep duration and psychological symptoms were based on participants’ reports rather than objective sleep monitoring or diagnostic interviews.

Third, the sample was geographically specific. Participants were university students in Lahore. Results should not automatically be generalized to all Pakistani young adults or all university students nationally.

Fourth, residual confounding is likely. The relatively low R-squared values show that the regression models captured only a small fraction of the differences in symptom scores between individuals. Many potentially relevant social, behavioural and clinical variables were outside the scope of the analysis.

Finally, statistical significance should not be confused with clinical importance. A statistically detectable association may still be modest at the individual level. The estimated 0.37-point difference in PHQ-9 score per additional reported hour of sleep, for example, should be interpreted in the context of the full scale, the observational design and individual clinical circumstances.

What research should come next?

The next step should be to move beyond cross-sectional associations. Longitudinal studies could follow students across semesters to examine whether changes in sleep predict later changes in depressive symptoms. Repeated measurements around examinations and academic transitions could help distinguish short-term stress responses from persistent mental-health problems.

Studies could also examine whether gender differences remain after more detailed measurement of socioeconomic circumstances, social support, academic pressures, harassment, family expectations, menstrual and reproductive health factors where relevant, and access to care.

Sleep should ideally be measured more comprehensively than simply asking about duration. Validated sleep-quality and insomnia instruments, sleep diaries and objective approaches such as actigraphy could provide a clearer picture.

Intervention studies would then be needed to answer the question that our observational data cannot: whether improving sleep produces meaningful improvements in depressive symptoms in Pakistani university students.

The broader message: student depression is multifactorial

The most useful lesson from our findings may be that university mental health should not be organized around a single stereotype.

Medical education is stressful, but medical students are not the only students vulnerable to psychological distress. Female students in our sample reported higher symptoms across multiple domains, but gender alone does not explain depression. Students who slept longer reported lower PHQ-9 scores on average, but sleep and depression can influence each other and share common causes. Previous depression treatment was strongly associated with current symptom scores, but that does not imply treatment caused worse outcomes.

Each result represents one part of a much larger picture.

For researchers, this means designing studies that capture the complexity of student life rather than searching for a single universal predictor. For universities, it means making mental-health support broadly accessible while identifying groups and circumstances that may justify additional attention. For students, it means recognizing that persistent changes in mood, sleep, concentration, energy or functioning deserve to be taken seriously rather than dismissed simply as a normal consequence of university pressure.

Frequently asked questions

Did female students have higher depression scores in the study?

Yes. Female gender was associated with higher PHQ-9 scores and higher DASS-21 depression, anxiety and stress scores in adjusted analyses. The study does not establish why this difference occurred or prove that gender itself caused the higher symptom burden.

Did sleeping more reduce depression?

The study cannot establish that. Each additional reported hour of sleep was associated with a 0.37-point lower PHQ-9 score, but the cross-sectional design means we cannot determine whether sleep influenced depressive symptoms, depression influenced sleep, or other factors influenced both.

Were medical students the most depressed group?

No. In this sample, non-medical students had a higher median PHQ-9 score than medical students, and adjusted analyses also showed higher depressive and anxiety symptom scores among non-medical students on several outcomes.

Does a high PHQ-9 score mean someone definitely has depression?

No. The PHQ-9 is a validated symptom and screening instrument, but a questionnaire alone does not replace a clinical assessment or automatically establish a psychiatric diagnosis.

What should universities learn from the findings?

The results support mental-health strategies that include the whole student population rather than focusing exclusively on particular academic disciplines. Sleep health, previous mental-health history and possible gender-related differences are also reasonable areas for attention.

Medical disclaimer: This article is for educational and research communication purposes only. It does not provide a diagnosis or individual medical advice. Persistent depressive symptoms, significant sleep problems, impaired daily functioning or concerns about mental health should be discussed with an appropriately qualified healthcare professional. Anyone experiencing an immediate mental-health crisis or risk of harm should seek urgent professional assistance through appropriate local emergency or crisis services.