Health Information Systems and Data Analytics
Health Information Systems and Data Analytics
Understand the EHR ecosystem: clinical documentation, computerized provider order entry (CPOE), clinical decision support (CDS), e-prescribing, and interoperability standards (HL7, FHIR). Know the difference between an EHR, a data warehouse, and a health information exchange (HIE).
Data governance and quality: completeness, accuracy, timeliness, and consistency. Understand structured versus unstructured data, coding systems (ICD, CPT, SNOMED, LOINC), and how data quality problems propagate into analytics and quality reporting.
Analytics levels: descriptive (what happened), diagnostic (why), predictive (what will happen), and prescriptive (what to do). Know basic measures-sensitivity, specificity, positive predictive value-and why PPV depends on prevalence. Recognize confounding and selection bias in observational data.
Security: HIPAA Security Rule (administrative, physical, technical safeguards), access controls, audit logs, encryption, and breach response.
Traps: confusing sensitivity with PPV; assuming more data means better data; treating a predictive model's output as a diagnosis; forgetting that dashboards are only as good as underlying data quality; overlooking that CDS alerts can cause alert fatigue.
Worked rule of thumb: PPV = true positives ÷ (true positives + false positives). A highly sensitive screening tool in a low-prevalence population yields many false positives-so confirmatory testing matters.
Connect analytics to action: the goal is decision support, not data accumulation.