Medicine Dissertation Topics for UK Medical Students and Postgraduates

Before you commit to one of these medicine dissertation topics, it helps to settle a few things about your own project:

• Can I realistically reach the patients, staff or dataset this question needs within my university’s timetable?
• Does my medical school treat this as a systematic review, a secondary data analysis, an audit, a survey or an intercalated laboratory project?
• Will my project need NHS ethics review, or only university approval?
• Is the question narrow enough to answer in a single dissertation?

This page focuses on cross-specialty and general medicine topics: questions about access, workforce, prescribing, ethics, digital care and inequalities that fit an MBBS or MBChB student selected component, an intercalated BSc, or a Master’s in medical sciences or clinical research. If you already know your specialty, the single-specialty pages for cardiology research ideas, neurology project topics and endocrinology titles will take you deeper. Position correct as of October 2026 for the live policy debates mentioned below, so check each one again before you finalise a proposal.

Medicine Dissertation Topics on Access, Waiting Times and Service Performance

These topics suit students who prefer working with published data over recruiting patients. NHS England publishes monthly performance statistics, so you can build a project without ethics review beyond your university’s own process. Check this with your supervisor first, because rules differ between medical schools.

1. Regional variation in the 18-week referral-to-treatment standard for one elective specialty, using the last five years of NHS England waiting list statistics. Pick a single specialty such as trauma and orthopaedics or ophthalmology so the dataset stays manageable. A descriptive time-series analysis in Excel or R is enough at undergraduate level. A Master’s student could add a regression comparing trusts by size, rurality and deprivation.

2. Whether Emergency Department four-hour performance tracks winter bed occupancy across a group of acute trusts. The question is whether occupancy explains performance better than attendance numbers do. Use publicly available monthly data and keep the trust sample small (10 to 15) so you can check each trust’s data quality yourself. A common pitfall is comparing trusts that report different site types, so state your inclusion rules clearly.

3. Do patients who wait longer for a first outpatient appointment show higher rates of emergency admission for the same condition? This needs linked data such as Hospital Episode Statistics, which usually means a supervisor-led application and several months of lead time. It is better suited to a Master’s or PhD project than a short undergraduate dissertation.

4. Patient-reported access to a GP appointment and its association with deprivation in a defined group of practices. The GP Patient Survey is open and published at practice level, which makes this realistic for a first research project. You could pair it with Fingertips deprivation indices and test the association with a simple correlation or multilevel model.

5. The accuracy of NEWS2 early warning scores in detecting deterioration among older patients on acute medical wards. This is a retrospective notes review and needs local approval as a service evaluation or audit. A realistic scope is one ward over one winter, with sensitivity and specificity as outcomes. Define “deterioration” before collecting any data, or the results will be hard to defend.

Workforce, Training and Medical Education Research Topics

Medical education is one of the most accessible areas for MBBS students because your own cohort is a ready participant group. Surveys and short interviews are common, and many medical schools have a dedicated ethics route for education research.

6. How prepared do final-year students feel for the UK Medical Licensing Assessment, and which parts of the curriculum do they say contributed least? A cross-sectional survey of one cohort with free-text responses analysed thematically works well. Keep the questionnaire short and pilot it on 5 or 6 peers first.

7. Why resident doctors leave UK training, or consider leaving, after the Foundation Programme. You could draw on the GMC’s National Training Survey results and published BMA surveys as secondary sources, or interview a small group of foundation doctors. The question is sensitive because the topic overlaps with ongoing pay and contract disputes, so keep the framing analytical and avoid advocacy language.

8. The scope-of-practice debate on physician associates following the Leng review. The review reshaped the policy conversation around the role, including how it is named and supervised. This suits a Master’s-level policy analysis or a structured literature review of reports, regulator documents and professional-body statements. Read the current GMC guidance first, because the position has moved more than once.

9. Differential attainment between demographic groups in UK medical school assessments. The main difficulty is data access, since school-level data are rarely open. A realistic undergraduate version is a systematic review of published UK studies on attainment gaps. A Master’s student with institutional permission could analyse anonymised assessment outcomes from one school.

10. Does simulation-based communication training change how students handle breaking bad news, compared with observation alone? A small quasi-experimental design with a validated rating tool (such as the SEGUE framework or an equivalent your school already uses) fits an intercalated year. Be clear about what you can and cannot claim from a short follow-up.

Prescribing, Pharmacology and Obesity Medicine Topics

Prescribing research has some of the best open data in UK medicine. OpenPrescribing publishes practice-level prescribing data, and the MHRA Yellow Card scheme publishes suspected adverse reaction summaries. Both let a student build a quantitative project without recruiting anyone. For purely pharmacological ideas, the pharmacology dissertation topics page goes further.

11. Uptake of tirzepatide and semaglutide for weight management across English practices after NICE guidance and the phased NHS rollout. OpenPrescribing may not capture every route of supply (private prescribing sits outside it), so state that limit early. A descriptive analysis by region and deprivation is feasible for undergraduates, and a Master’s student could add an interrupted time-series model around the guidance dates.

12. Deprescribing in older patients: how consistently do GPs and pharmacists apply STOPP/START criteria during structured medication reviews? An audit of anonymised reviews or a short interview study could work. Check whether your host practice will give you access before building the design around it.

13. Variation in antibiotic prescribing for respiratory infections between practices, and how far it tracks local resistance patterns. You can combine OpenPrescribing with the UKHSA’s antimicrobial resistance reporting. The risk is drawing causal conclusions from ecological data, so frame findings as associations.

14. Patterns in Yellow Card reports for a newer drug class during its first two years of wide use. Spontaneous reports cannot give incidence rates, which is an important limitation to discuss. This is a good test of whether you understand pharmacovigilance rather than just describing numbers.

Medical Ethics, Law and End-of-Life Care Dissertation Topics

Ethics and law projects need no patient recruitment, which makes them practical under tight deadlines. They do need a clear method (doctrinal, comparative or normative argument), and a strong version defends a position rather than listing viewpoints. If law is your main interest, the medical law dissertation topics page has more.

15. How should safeguards for assessing capacity and coercion be designed if assisted dying becomes lawful in England and Wales? The Terminally Ill Adults (End of Life) Bill has drawn sustained parliamentary debate. Check the current stage and the latest text before you commit. A comparative study of safeguards in Oregon, Canada and Victoria, Australia, set against the Bill’s own provisions, gives a clear structure at Master’s level.

16. Whether the Mental Capacity Act 2005 is applied consistently when consent is assessed in acute hospital settings. A doctrinal analysis of case law, combined with published audit and Care Quality Commission material, is feasible without patient contact. A pitfall is treating capacity as a single yes or no decision when the Act requires it to be decision-specific.

17. The duty of candour and how junior doctors experience it in practice. This could be a short qualitative study with 8 to 12 interviews, or a literature-based project at undergraduate level. Ethics approval is likely to be needed for interviews, so build that time into your plan.

18. Fair allocation of scarce NHS resources: how do NICE’s quality-adjusted life year thresholds sit alongside ideas of fairness for rare diseases? This is a philosophical project for students with a taste for argument. It pairs well with the health economics dissertation topics if you want to add a cost-effectiveness element.

Digital Health, AI and Data in Medicine Research Ideas

19. Clinicians’ trust in ambient voice technology for documentation in outpatient clinics. NHS England has issued guidance on ambient scribing, so a qualitative or mixed-methods study of early adopters is timely. Because pilots change quickly, cut the scope to a single trust or practice group and ask what they have actually used.

20. Skin-tone bias in pulse oximetry and what it means for treatment thresholds. The Independent Review of Equity in Medical Devices raised this as a patient safety issue. A systematic review of published accuracy studies is realistic for a Master’s student. Undergraduates can write a critical appraisal of three or four key papers.

21. Can machine-learning triage tools be fairly evaluated without race and sex data in training sets? This one sits at the edge of medicine and computing. You will need at least basic Python or R, and you should check your department allows the crossover. For more computing-led projects, the artificial intelligence dissertation topics page is a better starting point.

22. Virtual wards and hospital at home: which patient groups show the clearest evidence of benefit, and where is the evidence thin? A scoping review is a sound choice here because the literature is mixed in quality and design. Define your inclusion criteria and search strings in advance, and record them in a protocol.

Inequalities, Prevention and Multimorbidity Topics

23. Diagnostic delay in endometriosis and how it differs by ethnicity and region. Use published cohort data and national audit material to frame the problem, then ask what a feasible local audit could add. Be careful with ethnicity categories, which vary between datasets.

24. Multimorbidity and treatment burden: how do patients with three or more long-term conditions describe the work of managing their care? Semi-structured interviews with a small sample, analysed using thematic analysis, would fit a Master’s project. Recruitment through a GP practice needs NHS approval, so ask early whether your supervisor holds a relevant existing study you could join.

25. Uptake of NHS Health Checks in the most deprived areas of one integrated care board. Local authority public health teams publish uptake data. A student could compare uptake by neighbourhood deprivation and suggest what a targeted invitation strategy might involve.

26. Long COVID in primary care: how have recorded diagnoses changed since the coding framework was introduced? OpenSAFELY research is a possible route, though access to its data is controlled and it usually needs a supervisor with an approved project. A less demanding version is a literature synthesis on diagnostic criteria.

27. Does NHS Staff Survey data show a link between reported burnout in medical staff and patient-safety concerns at trust level? The survey’s trust-level results are public, which keeps this feasible for undergraduates. Ecological fallacy is the risk: findings describe trusts, not individuals.

Intercalated BSc and Small-Scale Clinical Project Ideas

Intercalating students often have a full year and a research supervisor, so a more ambitious design becomes possible. Even so, smaller is better than big and unfinished.

28. A systematic review and meta-analysis of a narrow clinical question. For example, whether a specific intervention reduces length of stay after a defined procedure. Register a protocol on PROSPERO, follow PRISMA, and agree the inclusion criteria with your supervisor before searching. This is often the safest design when ethics timelines are uncertain.

29. A closed-loop clinical audit against a NICE or Royal College standard. Choose a standard that can be measured from notes, such as time to antibiotics in suspected sepsis. Audit, re-audit after one change, and report what moved. Register it with the trust’s audit department so it counts as a service improvement project.

30. A biomarker validation study using stored samples and an existing consented biobank. This depends entirely on the lab you join. If your supervisor has a running project, ask for a well-defined subset, such as validating one marker in an existing cohort, rather than a new discovery study. For related ideas see pathology research topics and genetics project ideas.

31. A cross-sectional survey on patients’ understanding of a common consent form. A readability analysis of the form itself, plus a short comprehension questionnaire, is a compact design. Patient recruitment will need NHS review, so a version using lay volunteers and a university ethics route is often more realistic.

32. Case-series write-up with literature review for a rare presentation seen in your placement. This needs consent and local governance approval, and you cannot publish identifiable detail. It is best treated as a small, careful learning exercise rather than a headline project.

Worked Examples: Turning a Medicine Topic into an Aim and Objectives

Four sample frames show what a workable aim and objectives can look like. Adapt the wording to your own data and school, and the aims and objectives generator can help you test a draft.

From Topic 4: GP Appointment Access and Deprivation

Aim: To examine whether the proportion of patients reporting a good experience of making an appointment differs by area deprivation across GP practices in one integrated care board.

• Extract practice-level appointment-experience results from the most recent GP Patient Survey for the chosen area.
• Link each practice to its local deprivation decile using Fingertips.
• Test the association with Spearman correlation and report effect size with confidence intervals.

From Topic 11: Obesity Drug Prescribing in Primary Care

Aim: To describe how NHS primary care prescribing of newer weight-management drugs varies between regions in England after national guidance.

• Download monthly prescribing counts from OpenPrescribing for the relevant products across a defined period.
• Compare regional rates per 1,000 registered patients using list-size denominators.
• Discuss what the dataset cannot show, including private supply.

From Topic 20: Pulse Oximetry and Skin Tone

Aim: To appraise the quality of evidence on pulse oximeter accuracy across skin tones and its implications for clinical thresholds.

• Search three databases using a documented strategy and screen results against pre-set criteria.
• Assess risk of bias in included studies using a recognised tool.
• Synthesise the direction and size of reported error, and state where the evidence is inconsistent.

From Topic 29: Sepsis Antibiotic Timing Audit

Aim: To establish whether one acute medical unit meets the local standard for time to first antibiotic in suspected sepsis and whether one intervention improves it.

• Collect time-to-antibiotic data for a defined consecutive sample of cases.
• Agree and implement one change, such as a prompt in the admission pathway.
• Re-audit after a set interval and compare compliance.

Choosing Between Medicine Research Designs

Your design should follow your access, not the other way around. If you cannot reach patients in time, a systematic review, a secondary data analysis or a policy and ethics project will serve you better than a half-finished survey. If you can reach a cohort, such as your own year group, a short survey with a clear sampling plan is acceptable.

For quantitative work, decide your outcome measure and analysis before collecting data. The statistical test selector and the sample size calculator help check whether the plan holds up. For qualitative work, agree your sampling logic and analysis method (thematic, framework or content analysis) before interviews begin. A methodology chapter guide can help you justify these decisions in writing.

Ethics Approval and Data Access for Medicine Dissertations

The route depends on who you involve. Research that recruits NHS patients usually needs Health Research Authority approval and NHS ethics review. Research that uses only public data, or that involves students and staff outside patient care, often needs only university ethics. Student projects at undergraduate and Master’s level are treated differently in some cases, so read the Health Research Authority’s current guidance on research and ethics review and confirm the route with your supervisor before you design anything.

Open datasets are the fastest option. NHS England publishes its monthly performance statistics without application, and OpenPrescribing, Fingertips and the GP Patient Survey are similar. Controlled sources such as CPRD, Hospital Episode Statistics, OpenSAFELY and UK Biobank all require an application, usually through a supervisor, and can take months. Check current access rules and fees directly with each provider before you plan around them.

How to Narrow a Broad Medicine Idea into a Researchable Question

Start with a problem, then add four limits: a population, a setting, a time period and an outcome. “Obesity” becomes “primary care prescribing of newer weight-loss drugs in one region over two years.” Then write the question in one sentence and check whether you could answer it with data you can reach. Our guide on how to write a research question walks through the process, and the guide to choosing research topics covers the earlier step of picking a direction. The dissertation topic finder can also suggest alternatives if an idea fails the access test.

Common Mistakes in Medicine Dissertations

• Choosing a project that depends on an ethics decision you have not yet asked about.
• Claiming causation from observational or ecological data.
• Reporting statistical significance without effect size or confidence intervals.
• Reviewing a topic so broad that no search strategy can cover it.
• Using a policy debate as a topic without checking its current status.
• Leaving the limitations section to the last night, when it should shape the design from the start.

Getting Support With Your Medicine Dissertation

If you are stuck on structure, method or analysis, tailored support from medical dissertation specialists can help you work out what to prioritise. For earlier-stage planning, a research proposal service gives you a clear starting document, and good literature review support keeps the evidence section focused.

Frequently Asked Questions About Medicine Dissertation Topics

Can I do a medicine dissertation without recruiting patients?

Yes. Systematic reviews, secondary data analyses of public datasets, ethics and law projects and education studies within your own school all avoid patient recruitment. Each is a legitimate way to demonstrate research skills.

How many topics should I shortlist before choosing?

Three is usually enough. Check each against data access, ethics route, supervisor interest and your timetable, then drop the weakest.

Is a systematic review acceptable for an MBBS student selected component?

It depends on the school. Many accept them, but some require primary data or a particular word limit, so check your module handbook before you register a protocol.