Write a comment

MAJOR ISSUES

The study on the impact of unhealthy eating behaviors on sleep quality among university students, while providing meaningful associations, has several acknowledged limitations and potential issues:

Methodological Limitations (Cross-Sectional Design)

• Inability to Establish Causality: The most significant weakness is the study's cross-sectional methodology. While it can explore associations between unhealthy eating behaviors and poor sleep quality, it cannot establish causality. It is impossible to determine whether poor sleep leads to unhealthy eating habits or vice versa.

• Reliance on Self-Reported Data: The study relied on a self-administered questionnaire, which introduces the potential for response and recall bias. This could compromise the accuracy of reported dietary habits, meal timing, and sleep quality outcomes.

Scope and Generalizability Issues

• Restricted Generalizability: The inclusion of only university students limits the findings' generalizability to non-student populations.

• Focus on International Students: While targeting international students adds unique insight, their distinct challenges (cultural adaptation, language barriers, isolation) mean the findings may not fully apply to domestic student populations.

Unexamined Confounding Factors

• Lack of Detailed Dietary Analysis: The study did not provide a detailed analysis of diet quality or quantity beyond specific behaviors (e.g., heavy meals, snacking). This misses the potential confounding effect of nutrient intake on sleep.

• Mental Health and Environmental Factors: The study explicitly did not examine other factors that could significantly impact sleep quality, such as environmental factors or mental health assessments (e.g., anxiety, depression), which are known to be higher among female students and affect sleep.

Statistical Considerations

• Lack of Significance for Some Hypothesized Behaviors: While the study hypothesized that behaviors like skipping breakfast and late-night snacking would be associated with poor overall sleep quality, the adjusted multiple logistic regression analysis found they were not statistically significant independent predictors of overall poor sleep quality.

In summary, the key issues are the cross-sectional design precluding causal inference, the reliance on self-reported data, and the failure to account for critical unexamined confounders like diet quality and mental health.

MINOR ISSUES

• In addition to the major limitations, the study has a few minor issues and areas where detail or discussion could be improved:

1. Convenience Sampling:

The use of convenience sampling (a technique selected for its practicality and cost-effectiveness) introduces potential selection bias, as the participants were not randomly selected. While practical for this specific student cohort, it slightly weakens the representativeness of the sample compared to probability sampling methods.

2. Definition of Unhealthy Behaviors:

While the study defines key unhealthy behaviors, the assessment relies on the students' self-perception and reporting frequency rather than objective measurement (e.g., diet diaries or physiological markers). For example, the definition of an "adequate" meal-to-bedtime interval (three hours or more) is based on existing research but is a binary cut-off, which may not capture the full range of effects.

3. Ambiguity in "Other" Marital Status:

In Table 1, 13.5% of students are categorized under "Other" for marital status. The specific definitions included in this category are not detailed, which leaves a significant portion of the sample's marital context ambiguous.

4. Discrepancy in Authorship Listing:

The first page of the preprint lists Maha Al-Jawarneh as the sole author, while the second page lists four authors: Shalini Chauhan, Maha Al-Jawarneh, Ildikó Csölle, and Szimonetta Lohner. This discrepancy in the initial presentation of authorship could be considered a minor editorial or formatting issue.

5. Focus on the Sample:

Although the large sample size is listed as a strength, the sample is overwhelmingly female (68.8%). While sex-based variations are discussed, this imbalance might affect the generalization of certain findings, particularly those where males showed poorer outcomes (e.g., sleep efficiency).

Competing interests

The author declares that they have no competing interests.

Use of Artificial Intelligence (AI)

The author declares that they did not use generative AI to come up with new ideas for their review.