Study designs for medical theses — at a glance
| Design | Best for | Example thesis title |
|---|---|---|
| Descriptive cross-sectional | Prevalence, patterns, KAP studies | Prevalence of anaemia among antenatal women attending a tertiary care hospital |
| Analytical cross-sectional | Exposure–outcome association, single visit | Association between screen time and dry eye disease in IT professionals |
| Case-control | Rare outcomes; quick, retrospective | Risk factors for surgical site infection following emergency laparotomy: a case-control study |
| Cohort (prospective/retrospective) | Rare exposures, temporality, incidence | Maternal and foetal outcomes in pregnancies with gestational diabetes: a prospective cohort |
| Randomised controlled trial | Comparing interventions | Bupivacaine vs ropivacaine for postoperative analgesia: a randomised controlled trial |
| Quasi-experimental | Interventions without randomisation | Effect of a structured education module on insulin technique among nurses |
| Diagnostic accuracy | New test vs gold standard | Diagnostic accuracy of GeneXpert versus culture in pulmonary tuberculosis |
Examiner tip: Name your design precisely in the protocol — "hospital-based analytical cross-sectional study" beats just "observational study". Vague design statements are a classic reason for protocol queries.
Frequently asked questions
Which study design is easiest to finish within residency?
Cross-sectional studies — one visit per participant, no follow-up, modest sample size. Case-control studies are also practical for rare outcomes. RCTs and prospective cohorts are stronger evidence but need time and resources a 3-year residency rarely allows.
Case-control vs cohort — what's the difference?
Case-control starts from the outcome and looks backward at exposures (good for rare diseases). Cohort starts from the exposure and follows forward to the outcome (good for rare exposures, proves temporality).
Can my thesis be a retrospective record-based study?
Yes — retrospective cohort or case-control designs using hospital records are accepted by most universities, provided records are complete and ethics permission covers record access. They're often the fastest option for super-speciality theses.
Related free tools
- Sample Size Calculator — your design decides the formula; calculate it next
- Statistical Test Chooser — pick the right test for your data
- How to Write a Thesis Protocol — full guide