Abstract
Observational studies are fundamental to public health research but require transparent reporting to mitigate inherent biases. This study evaluates the reporting quality of 165 Master of Public Health (MPH) dissertations in China using the STROBE checklist. Results indicate a median compliance rate of 74.79%, with significant deficiencies in reporting methodological details such as missing data and sensitivity analyses. Factors such as funding and prior publication history were strong predictors of higher reporting quality.
Introduction
In the hierarchy of medical evidence, Randomized Controlled Trials (RCTs) are often prioritized, yet observational studies remain the backbone of public health research due to ethical and practical considerations. In China, the majority of Master of Public Health (MPH) dissertations utilize observational designs. However, these studies are prone to information bias, necessitating rigorous and transparent reporting. The Strengthening the Reporting of Observation Studies in Epidemiology (STROBE) statement provides an international standard for such reporting, but its application in Chinese graduate education has not been extensively studied.
Methods
A systematic review was conducted using the Wanfang database, targeting MPH dissertations from 2014 to 2019. The final sample included 165 original observational studies (61 cross-sectional, 66 case-control, and 38 cohort studies). Each dissertation was evaluated against the 22-item STROBE checklist by two independent investigators. Scoring was standardized: 1 point for full reporting, 0.5 for partial, and 0 for non-reporting. Multivariate logistic regression was employed to identify variables associated with high-quality reporting.
Results
The analysis revealed a mean STROBE score of 14.29 (SD 1.84) out of a possible 22. While introductory elements like study background and objectives were universally reported (100% compliance), critical methodological and result-oriented items were frequently omitted.
Key findings regarding reporting gaps include:
- Missing Data: Only 6.67% of studies explained how missing data were addressed.
- Sensitivity Analyses: Only 3.63% reported these analyses.
- Flow Diagrams: Only 15.15% utilized flow diagrams to describe participant progression.
- Absolute Risk: 0% of relevant studies reported absolute risk estimates.
Conversely, certain factors were associated with "high reporting quality" (defined as the top quartile of scores). Cohort studies (OR=3.41), funded research (OR=4.37), and students with two or more prior publications (OR=3.46) were significantly more likely to produce high-quality reports.
Discussion
The "Intro vs. Method" divide suggests that while students are well-trained in identifying research problems, they lack the technical expertise or guidance required for rigorous reporting of complex methodologies. The strong correlation between funding and quality implies that external oversight and the competitive nature of grant-seeking drive better documentation. Similarly, the "practice makes perfect" effect of prior publication highlights the value of the peer-review process in teaching reporting standards.
Conclusion
The reporting quality of observational studies in Chinese MPH dissertations is currently suboptimal. To improve scientific transparency, academic institutions should consider adopting the STROBE statement as a formal requirement for dissertation submission. Enhanced training in methodological reporting, particularly regarding data handling and bias assessment, is essential for the next generation of public health researchers.
Summary updated