Types of Health Research and Their Applications
Health research may be classified by the kind of data collected, purpose of the inquiry, degree of investigator control and timing of observation. Selecting the approach must begin with the research question—not with a preferred software package or method.
Learning objectives
- Differentiate quantitative, qualitative and mixed-methods research.
- Classify research by purpose and study design.
- Match research approaches to clinical and public-health questions.
- Explain strengths, limitations and complementary uses of each approach.
Quantitative research
Quantitative research collects numerical data and uses statistical analysis to estimate frequency, compare groups, test hypotheses or model relationships. It usually uses structured tools and predefined variables.
Descriptive
Describes frequency or distribution: for example, prevalence of hypertension among adults attending a clinic.
Analytical
Examines associations or causes: for example, comparing tobacco exposure among lung-cancer cases and controls.
Experimental
Investigator assigns an intervention: for example, a randomised trial comparing two treatments.
Qualitative research
Qualitative research explores meaning, experience, behaviour and social context using non-numerical data such as interview transcripts, observations and documents. Sampling is often purposive, collection continues toward information sufficiency, and analysis identifies patterns or themes.
- Phenomenology: explores lived experience.
- Ethnography: studies culture and shared practices.
- Grounded theory: develops theory from systematically analysed data.
- Case study: investigates a bounded person, programme, facility or event in depth.
- Narrative inquiry: examines stories and how people make sense of experience.
Mixed-methods research
Mixed methods intentionally integrates quantitative and qualitative evidence. A convergent design collects both during a similar period; an explanatory sequential design uses qualitative work to explain quantitative results; an exploratory sequential design uses initial qualitative findings to develop a later quantitative study.
| Approach | Best answers | Typical data | Example |
|---|---|---|---|
| Quantitative | How many? How much? Is there an association? | Counts, measurements, scores | Effect of a drug on mean blood pressure |
| Qualitative | How and why? What is the experience? | Words, observations, images | Reasons patients miss clinic visits |
| Mixed methods | What is happening and why? | Integrated numerical and textual evidence | Coverage survey plus interviews on vaccine hesitancy |
Classification by purpose
- Basic research: expands fundamental knowledge without an immediate practical goal.
- Applied research: addresses a practical clinical or health-system problem.
- Exploratory research: clarifies a poorly understood issue and generates ideas.
- Descriptive research: documents characteristics, frequency or patterns.
- Explanatory research: investigates mechanisms or causal relationships.
- Evaluation research: assesses programme relevance, implementation, effectiveness, efficiency or impact.
- Implementation research: studies how to adopt and sustain proven interventions in real settings.
Observational and interventional studies
| Design | Direction | Typical measure/use |
|---|---|---|
| Cross-sectional | Exposure and outcome measured at one time | Prevalence |
| Case-control | Start with outcome, look back for exposure | Odds ratio; useful for rare diseases |
| Cohort | Start with exposure, follow to outcome | Incidence and risk ratio |
| Randomised controlled trial | Assign intervention and compare outcomes | Treatment efficacy/effectiveness |
Choosing an approach
Consider the question, current evidence, ethical acceptability, required data, time, resources, expertise and intended use. A strong design provides enough evidence to answer the question while protecting participants and minimising bias.
Applications in the health system
- Estimate disease burden and service coverage.
- Identify risk and protective factors.
- Evaluate diagnostics, medicines and prevention programmes.
- Understand patient and provider experiences.
- Improve quality, policy implementation and resource allocation.
- Generate locally relevant evidence for guidelines and advocacy.
Scenario 1: choosing a quantitative design
A district wants to know the proportion of pregnant women who complete at least eight antenatal contacts and which characteristics are associated with completion. A cross-sectional survey can estimate coverage and examine associations. It is efficient, but because exposure and outcome may be measured together, temporal sequence and causality can be uncertain.
Scenario 2: choosing a qualitative design
Clinic records show low uptake of cervical-cancer screening, but the reasons are unclear. In-depth interviews with eligible women can explore knowledge, fear, privacy, partner influence, cost and previous experiences. Focus groups may reveal community norms, whereas private interviews may be better for sensitive personal experiences.
Scenario 3: mixed-methods integration
A hospital survey finds that 38% of caregivers discontinue childhood epilepsy treatment. Interviews then explore explanations such as medicine shortages, stigma, traditional beliefs and adverse effects. Integrating prevalence and lived experience produces a stronger intervention plan than either dataset alone.
Detailed design comparison
| Design | Best feature | Major limitation | Applied example |
|---|---|---|---|
| Ecological | Compares groups or populations | Group association may not apply to individuals | District sanitation coverage versus cholera rate |
| Cross-sectional | Rapid prevalence estimate | Weak temporal evidence | Current depression among health workers |
| Case-control | Efficient for rare outcomes | Recall and selection bias | Risk factors for neonatal tetanus |
| Cohort | Establishes exposure before outcome | Time, cost and loss to follow-up | HIV-exposed infants followed to outcome |
| Randomised trial | Strong control of confounding | Ethical and feasibility limits | New adherence reminder versus usual care |
| Quasi-experimental | Evaluates real-world interventions without randomisation | Groups or time periods may differ | Hand-hygiene policy before-and-after study |
Qualitative sampling and data collection
- Purposive sampling: deliberately selects participants with relevant experience.
- Maximum-variation sampling: includes diverse cases to explore differing perspectives.
- Snowball sampling: participants refer others in difficult-to-reach populations.
- Key-informant interview: gains specialised knowledge from informed individuals.
- Focus-group discussion: explores shared beliefs and group interaction.
- Observation: records practices and context rather than relying only on reports.
Quality criteria across approaches
| Quantitative emphasis | Qualitative counterpart | Practical strategy |
|---|---|---|
| Internal validity | Credibility | Appropriate design, triangulation, member checking |
| External validity | Transferability | Representative sampling or thick contextual description |
| Reliability | Dependability | Standardised tools or an audit trail |
| Objectivity | Confirmability | Reflexivity, transparent decisions and independent review |
Decision exercise
Match each question to a design and justify your answer:
- What proportion of surgical patients develop a wound infection?
- Why do adolescents avoid mental-health services?
- Does a new treatment reduce mortality compared with standard care?
- Are previous home deliveries associated with postpartum complications?
- How well was a national vaccination programme implemented, and how did communities experience it?
Frequent errors in design selection
- Calling every questionnaire study “qualitative”.
- Using a cross-sectional study to claim that an exposure caused an outcome.
- Selecting cases and controls on the basis of exposure rather than outcome.
- Failing to define the mixed-methods point of integration.
- Choosing a method before establishing the research question.
- Assuming qualitative research requires no systematic sampling or analysis.
Review questions
- Differentiate quantitative and qualitative research.
- When is a mixed-methods design useful?
- Compare cross-sectional, case-control and cohort studies.
- Distinguish basic from applied research.
Return to the Clinical Medicine Year 2 curriculum
References
- WHO. Health Research Methodology.
- Kothari CR, Garg G. Research Methodology: Methods and Techniques.
- UAHEB. Guidelines on Research Proposal and Report Writing.
Educational content only. Study design should be reviewed by qualified supervisors and appropriate ethics bodies.
Types of Health Research and Their Applications
Health research may be classified by the kind of data collected, purpose of the inquiry, degree of investigator control and timing of observation. Selecting the approach must begin with the research question—not with a preferred software package or method.
Learning objectives
- Differentiate quantitative, qualitative and mixed-methods research.
- Classify research by purpose and study design.
- Match research approaches to clinical and public-health questions.
- Explain strengths, limitations and complementary uses of each approach.
Quantitative research
Quantitative research collects numerical data and uses statistical analysis to estimate frequency, compare groups, test hypotheses or model relationships. It usually uses structured tools and predefined variables.
Descriptive
Describes frequency or distribution: for example, prevalence of hypertension among adults attending a clinic.
Analytical
Examines associations or causes: for example, comparing tobacco exposure among lung-cancer cases and controls.
Experimental
Investigator assigns an intervention: for example, a randomised trial comparing two treatments.
Qualitative research
Qualitative research explores meaning, experience, behaviour and social context using non-numerical data such as interview transcripts, observations and documents. Sampling is often purposive, collection continues toward information sufficiency, and analysis identifies patterns or themes.
- Phenomenology: explores lived experience.
- Ethnography: studies culture and shared practices.
- Grounded theory: develops theory from systematically analysed data.
- Case study: investigates a bounded person, programme, facility or event in depth.
- Narrative inquiry: examines stories and how people make sense of experience.
Mixed-methods research
Mixed methods intentionally integrates quantitative and qualitative evidence. A convergent design collects both during a similar period; an explanatory sequential design uses qualitative work to explain quantitative results; an exploratory sequential design uses initial qualitative findings to develop a later quantitative study.
| Approach | Best answers | Typical data | Example |
|---|---|---|---|
| Quantitative | How many? How much? Is there an association? | Counts, measurements, scores | Effect of a drug on mean blood pressure |
| Qualitative | How and why? What is the experience? | Words, observations, images | Reasons patients miss clinic visits |
| Mixed methods | What is happening and why? | Integrated numerical and textual evidence | Coverage survey plus interviews on vaccine hesitancy |
Classification by purpose
- Basic research: expands fundamental knowledge without an immediate practical goal.
- Applied research: addresses a practical clinical or health-system problem.
- Exploratory research: clarifies a poorly understood issue and generates ideas.
- Descriptive research: documents characteristics, frequency or patterns.
- Explanatory research: investigates mechanisms or causal relationships.
- Evaluation research: assesses programme relevance, implementation, effectiveness, efficiency or impact.
- Implementation research: studies how to adopt and sustain proven interventions in real settings.
Observational and interventional studies
| Design | Direction | Typical measure/use |
|---|---|---|
| Cross-sectional | Exposure and outcome measured at one time | Prevalence |
| Case-control | Start with outcome, look back for exposure | Odds ratio; useful for rare diseases |
| Cohort | Start with exposure, follow to outcome | Incidence and risk ratio |
| Randomised controlled trial | Assign intervention and compare outcomes | Treatment efficacy/effectiveness |
Choosing an approach
Consider the question, current evidence, ethical acceptability, required data, time, resources, expertise and intended use. A strong design provides enough evidence to answer the question while protecting participants and minimising bias.
Applications in the health system
- Estimate disease burden and service coverage.
- Identify risk and protective factors.
- Evaluate diagnostics, medicines and prevention programmes.
- Understand patient and provider experiences.
- Improve quality, policy implementation and resource allocation.
- Generate locally relevant evidence for guidelines and advocacy.
Review questions
- Differentiate quantitative and qualitative research.
- When is a mixed-methods design useful?
- Compare cross-sectional, case-control and cohort studies.
- Distinguish basic from applied research.
Return to the Clinical Medicine Year 2 curriculum
References
- WHO. Health Research Methodology.
- Kothari CR, Garg G. Research Methodology: Methods and Techniques.
- UAHEB. Guidelines on Research Proposal and Report Writing.
Educational content only. Study design should be reviewed by qualified supervisors and appropriate ethics bodies.
