Introduction to Health Data Research

Estimated time: 25 minutes

ON THE BOARD · Module 7 summary

Ask a question the available data can actually answer.

1. Ask a question the available data can actually answer.
2. A fall in attendance has several possible explanations.
3. Association is not causation. A third factor can move both figures.
4. A short brief states the finding, the limit, and the next check.

MODULE SUMMARY

Module 7 summary

Ask a question the available data can actually answer.

A fall in attendance has several possible explanations.

Association is not causation. A third factor can move both figures.

A short brief states the finding, the limit, and the next check.

The board writes the point she is saying. It changes as the lesson moves on. Keep the sound on.

READ COMPLETE NOTES

A hospital manager notices that outpatient attendance has declined over three months and wants to know why. Possible explanations include service availability, staff shortages, transport, demand, incomplete reporting, and a change in how attendance is recorded. The numbers alone may not explain what happened. You need a question, the available evidence, the limits of that evidence, and a careful way to report it.

7.1.1 What is health data research?

Health data research means examining information systematically to answer a question about health, healthcare services, health systems or population health. Sources can include hospital records, facility reports, public health surveys, programme monitoring data, community interviews, electronic health records, surveillance information, and administrative data.

Research does not always need complicated mathematics. A well-designed descriptive study can answer a practical question with simple calculations, careful comparisons and a clear interpretation.

7.1.2 Why is research important in healthcare?

Identify gaps in service delivery.
Understand patterns in health information.
Examine reporting problems.
Assess whether programme targets are being achieved.
Explore possible reasons for an observed change.
Inform resource allocation.
Identify questions that still need investigation.

If some departments keep submitting incomplete reports, a structured review can show how often information is missing, which fields are affected, and what operational reasons are worth checking. That can guide the form, the instructions, or the training. It cannot, by itself, prove why a field was left blank.

7.1.3 The basic research process

1. Problem
2. Question
3. Method
4. Data
5. Analysis
6. Interpretation
7. Report
8. Action

Identify a problem. Example: monthly service reports frequently contain missing fields.
Develop a question. Example: which required fields are most frequently missing?
Select a method. Example: review a defined sample of monthly reports and record whether each required field is complete.
Obtain data through approved records and a documented process.
Prepare the data. Check missing records, inconsistent definitions and other quality problems before you calculate.
Interpret which fields have the highest levels of missing information.
Communicate the findings and the limitations.
Recommend an action that the evidence can support.

7.1.4 Descriptive research versus other research

Approach	Main purpose	Example
Descriptive	Describe what is happening	What percentage of expected reports were received?
Exploratory	Investigate an issue that is not yet well understood	What challenges do staff report when completing digital reports?
Analytical	Examine relationships, differences or patterns	Is reporting completeness different across facility types?
Evaluative	Assess a programme or intervention	Did the programme meet its defined objectives?

These categories can overlap. The question and the design decide the approach.

Lesson 7.1 knowledge check

Choose an answer for each question, then check. The suggested answers stay hidden until you do.

1. What should generally happen before you analyse data?
2. Which is a descriptive research question?

Research starts with a defined problem and a suitable question, not with a calculation that has no purpose.