Medical Research in Pakistan: Where to Start

To start medical research in Pakistan, begin with a focused, locally relevant question, find a dependable mentor, review the existing evidence, write a realistic protocol, obtain institutional ethics approval before collecting identifiable human data, and plan the analysis and publication pathway in advance. A small, well-designed study completed transparently is more valuable than an ambitious project that cannot be finished.

Why Medical Research in Pakistan Matters

Pakistan faces a wide range of clinical and public-health problems, but evidence generated in other countries may not always reflect local disease patterns, health-seeking behaviour, resource constraints, cultural factors, or healthcare delivery systems. Locally conducted research can help clinicians, hospitals, universities, and policymakers understand what is happening in their own populations.

For students and early-career doctors, research also develops practical skills in critical appraisal, scientific writing, teamwork, data management, and ethical decision-making. Publication can support an academic portfolio, but the primary goal should be to answer a meaningful question using sound methods.

Medical Research in Pakistan: Where to Start

1. Choose a Manageable Research Area

Start with a clinical department, patient group, public-health issue, or research method that genuinely interests you. Good beginner topics often arise from recurring questions during ward rounds, outpatient clinics, community placements, audits, journal clubs, or discussions with supervisors.

A suitable first project should meet four conditions:

• The question is important to patients, clinicians, or health services.
• The required participants or data are realistically accessible.
• The project can be completed with the available time, skills, and budget.
• The study adds something beyond what is already known.

Avoid selecting a topic only because it appears easy to publish. A vague or duplicated survey may consume months without producing useful evidence.

2. Turn the Topic into a Focused Question

A broad topic such as diabetes, depression, antibiotic use, or maternal health is not yet a research question. Define the population, exposure or intervention, comparator where relevant, and outcome.

For example, instead of asking, “What is the burden of depression?”, a more workable question might be: “What proportion of final-year medical students at selected teaching hospitals in Lahore have moderate-to-severe depressive symptoms, and which academic or lifestyle factors are associated with those symptoms?”

For intervention questions, the PICO framework can help:

• Population: Who is being studied?
• Intervention: What treatment, programme, or exposure is being assessed?
• Comparator: What is it being compared with?
• Outcome: What will be measured?

For observational, qualitative, diagnostic, or health-services research, adapt the framework rather than forcing every question into PICO.

3. Search the Literature Before Designing the Study

Search PubMed, Google Scholar, the Cochrane Library, and any databases available through your university. The Higher Education Commission’s digital-library services may provide participating institutions with access to additional scholarly resources.

Your initial review should answer:

• Has the same question already been answered in Pakistan?
• Which study designs and measurement tools have previous researchers used?
• What limitations remain unresolved?
• Which outcomes are clinically meaningful?
• Is a validated Urdu or regional-language version of the instrument available?

Keep a search log containing databases, keywords, filters, and dates. Use a reference manager such as Zotero, Mendeley, or EndNote from the beginning rather than organising citations manually at the manuscript stage.

4. Find the Right Mentor and Team

A useful mentor does more than add a senior name to the paper. The mentor should help refine the question, assess feasibility, strengthen the methods, navigate ethics requirements, and review the final manuscript.

When approaching a potential supervisor, send a brief concept note containing the proposed question, why it matters, the intended design, the likely study setting, and the support you need. Demonstrating preparation makes it easier for a faculty member to assess the proposal.

Build a small team with clearly defined responsibilities. Depending on the project, this may include a clinician, epidemiologist, statistician, laboratory specialist, qualitative researcher, or data manager. Discuss authorship and contribution expectations early. International Committee of Medical Journal Editors guidance links authorship with substantial intellectual contribution, manuscript involvement, final approval, and accountability—not merely data collection, seniority, or general supervision.

Selecting a Suitable First Study Design

For a first project, a carefully planned audit, chart review, cross-sectional study, qualitative study, or systematic review may be more achievable than a multicentre cohort or clinical trial. Simpler does not mean methodologically weak; it means matching the design to the question and available resources.

Write the Protocol Before Collecting Data

A protocol is the project’s operating plan. Writing it before data collection reduces arbitrary decisions and makes ethics review, collaboration, and analysis more consistent.

A basic protocol should include:

• Title and background
• Research question and objectives
• Study design and setting
• Eligibility criteria
• Sampling method and sample-size justification
• Variables, outcomes, and measurement tools
• Data-collection procedure
• Statistical or qualitative analysis plan
• Ethical considerations
• Data-security arrangements
• Timeline, responsibilities, and budget
• Dissemination plan

Ask a statistician or methodologist to review the protocol before data collection, not after the dataset has already been created. Sample size should be based on the primary objective, expected effect or prevalence, precision, study design, and anticipated missing data—not a convenient round number.

Ethics Approval and Participant Protection

Research involving human participants, identifiable records, biological material, or sensitive personal information should be submitted to the relevant institutional review board or ethics review committee before recruitment or data extraction begins. Approval requirements vary by institution, so follow the written procedures of the hospital, university, or collaborating organisation.

The World Medical Association’s Declaration of Helsinki provides ethical principles for medical research involving human participants, including respect for autonomy, scientific validity, risk minimisation, privacy, confidentiality, and informed consent where applicable.

Common ethics submissions include the protocol, participant information sheet, consent form, questionnaire or interview guide, data-collection form, investigator credentials, permission letters, and plans for confidentiality and data storage.

Do not assume that removing names alone makes data anonymous. Hospital numbers, dates, rare diagnoses, locations, or combinations of variables may still identify a person. Collect only the information needed for the research question, restrict access, use coded identifiers, and define how long data will be retained.

Register the Study When Appropriate

Interventional clinical trials should be registered prospectively in a registry accepted by the World Health Organization registry network or the International Committee of Medical Journal Editors. Registration should occur before the first participant is enrolled. Ethics approval and trial registration are separate requirements; one does not replace the other.

Systematic reviews can often be registered prospectively in an appropriate protocol registry when eligible. Observational studies may also benefit from a publicly accessible protocol or registration, even when journal policies do not require it, because prespecification improves transparency.

Collect High-Quality Data

Pilot the Tools

Test the questionnaire, case-report form, interview guide, or electronic database on a small number of records or volunteers before formal data collection. Piloting can reveal unclear wording, missing response options, calculation errors, excessive completion time, and problems with skip logic.

Use Validated Measures

When measuring symptoms, quality of life, knowledge, attitudes, or psychological constructs, use an instrument with evidence of validity and reliability for the intended population and language. Permission or licensing may be required. Translating a scale informally does not establish that it measures the same construct accurately.

Create a Data Dictionary

Define every variable, code, unit, permitted value, and missing-data label. For example, specify whether age is recorded in completed years, whether blood pressure is the first reading or an average, and how comorbidities are confirmed. A data dictionary prevents team members from interpreting fields differently.

Protect Data Quality

Train data collectors, use consistent procedures, check a sample of entries against source documents, restrict free-text fields where possible, and maintain a record of corrections. Never fabricate, alter, or delete observations merely to obtain statistically significant results.

Plan the Analysis Before Opening the Final Dataset

Specify the primary outcome and main analysis in advance. Distinguish descriptive analysis from hypothesis testing, and select methods that match the variable type, distribution, sampling strategy, and study design.

A responsible analysis plan should address missing data, outliers, confounding, multiple comparisons, subgroup analyses, and sensitivity analyses where relevant. Report effect estimates and confidence intervals rather than treating a p-value as the sole measure of importance. Statistical significance does not automatically imply clinical importance.

Beginners can use software such as Jamovi, JASP, SPSS, R, Stata, or Python, but software cannot correct a poorly framed question or biased design. Seek statistical advice when the analysis involves regression, repeated measurements, clustered data, survival analysis, diagnostic accuracy, psychometrics, meta-analysis, or causal inference.

Write Using the Correct Reporting Guideline

Reporting guidelines provide structured checklists of information readers need to understand and appraise a study. The EQUATOR Network maintains a searchable library of guidelines. Common examples include:

• CONSORT for randomised trials
• STROBE for observational studies
• PRISMA for systematic reviews
• STARD for diagnostic-accuracy studies
• CARE for case reports
• COREQ or SRQR for qualitative research
• TRIPOD for prediction-model studies

Choose the guideline while designing the study, not only after writing the manuscript. A checklist cannot repair missing outcomes, inadequate sampling, or unrecorded methodological details after the project is complete.

Choose a Journal Carefully

Select a journal whose scope, readership, study types, and author instructions match the manuscript. Review recent articles from the journal and verify its editorial information, peer-review process, indexing claims, fees, and publication policies.

Be cautious when a journal guarantees acceptance, promises unrealistically rapid publication, hides charges, uses deceptive indexing claims, or sends repeated unsolicited invitations unrelated to your field. Before submission, check the current journal-recognition requirements relevant to your institution and career pathway, including applicable Higher Education Commission policies.

Submit the manuscript to one journal at a time. Disclose conflicts of interest, funding, ethics approval, trial registration, prior dissemination, and any permitted use of automated or generative tools according to the target journal’s current policy.

Common Mistakes Made by New Researchers

• Collecting data before the protocol and ethics approval are complete
• Choosing a topic without checking whether it has already been studied locally
• Using a questionnaire without validation, permission, or cultural adaptation
• Calculating sample size after seeing the results
• Adding authors who did not meet authorship criteria
• Running many statistical tests until a significant result appears
• Ignoring missing data, selection bias, and confounding
• Writing the manuscript without an appropriate reporting checklist
• Submitting to a journal solely because it promises fast acceptance

A Practical 12-Week Starting Plan

1. Weeks 1–2: Identify a research area, meet potential mentors, and list three feasible questions.
2. Weeks 3–4: Search the literature, select one question, and prepare a one-page concept note.
3. Weeks 5–6: Choose the design, define outcomes, draft the protocol, and consult a statistician or methodologist.
4. Weeks 7–8: Prepare ethics documents, data tools, authorship expectations, and the project timeline.
5. Weeks 9–10: Submit for ethics and administrative approval; revise documents in response to feedback.
6. Weeks 11–12: Pilot the data tools, finalise the database and analysis plan, and begin only after all required approvals are documented.

Approval timelines may extend beyond 12 weeks. Use this period to improve the protocol, practise data management, refine the literature review, and prepare manuscript sections that do not depend on results.

Final Perspective

The best entry point into medical research in Pakistan is not the largest project or the most fashionable topic. It is a clear question, a credible mentor, an ethical and feasible protocol, careful data handling, and transparent reporting. Completing one rigorous project from question to dissemination creates a stronger foundation than joining several poorly organised studies.

This article is educational and does not replace the requirements of an institutional ethics committee, research office, journal, regulator, or qualified methodological adviser.