Health Informatics and Research Methods
Core concepts
Understand the research cycle: question, literature, design, sampling, data collection, analysis, interpretation, dissemination. Master study designs, sampling methods (probability vs non-probability), validity and reliability, and ethical principles (respect for persons, beneficence, justice; informed consent; ethics review). Know data types, descriptive and inferential statistics, and how to present results. For informatics, understand electronic health records, health information systems, data standards (ICD, HL7, SNOMED), data quality dimensions (accuracy, completeness, timeliness), and privacy/confidentiality principles.
Traps
- Confusing reliability (consistency) with validity (accuracy).
- Assuming random sampling eliminates all bias-it addresses selection but not information bias.
- Forgetting that informed consent requires capacity, information, and voluntariness.
- Mixing data standards: ICD for diagnoses, HL7 for messaging, SNOMED for clinical terminology.
Worked example / rule of thumb
Sample size for a proportion: n = Z²·p(1−p)/d². For p=0.5, Z=1.96, d=0.05: n = 3.84×0.25/0.0025 ≈ 384. Rule of thumb: a 5% margin of error at 95% confidence needs roughly 400 respondents for a proportion near 50%.