Application of Measures of Morbidity in Epidemiology
Morbidity refers to disease, injury, disability or departure from physical or mental wellbeing in a population. Morbidity measures quantify how frequently health conditions occur, how many people are affected, how severe the burden is and how health services should respond.
Learning objectives
- Define morbidity and distinguish it from mortality.
- Identify common morbidity measures and their data requirements.
- Select suitable measures for outbreaks, chronic disease and service planning.
- Interpret numerator, denominator, time and population at risk.
- Explain limitations and sources of bias in morbidity data.
Why measure morbidity?
Describe burden
Shows the amount and distribution of illness, including conditions that rarely cause death.
Plan services
Guides staffing, medicines, diagnostics, beds, rehabilitation and outreach.
Detect change
Identifies outbreaks, emerging conditions and changes in risk over time.
Evaluate programmes
Measures whether prevention, treatment and control reduce disease occurrence or complications.
Main morbidity measures
| Measure | What it describes | Typical use |
|---|---|---|
| Incidence proportion (risk) | New cases during a period among people initially at risk | Probability of developing disease |
| Incidence rate | New cases per person-time at risk | Speed of disease occurrence |
| Point prevalence | Existing cases at a specified point in time | Current service burden |
| Period prevalence | People who had the condition at any time during a period | Burden over weeks, months or years |
| Attack rate | New cases among an exposed population during an outbreak | Acute outbreak investigation |
| Secondary attack rate | New cases among susceptible contacts after exposure to primary cases | Household or institutional spread |
| Notification rate | Reported cases relative to population | Routine surveillance trends |
| Hospital admission/attendance rate | Service episodes relative to population or catchment | Health-service utilisation |
Essential parts of every measure
- Numerator: the health events or people counted.
- Denominator: the population from which the numerator arose.
- Population at risk: people capable of developing the condition.
- Time: the point, period or person-time covered.
- Multiplier: such as 100, 1,000 or 100,000 to make the result understandable.
Denominator rule
The numerator must be contained within, and logically linked to, the denominator. Do not include people who cannot experience the outcome. For example, when measuring risk of a first episode, people who already have the disease at the start may not belong in the at-risk denominator.
Ratios, proportions and rates
| Term | Structure | Example |
|---|---|---|
| Ratio | One quantity divided by another; numerator need not be part of denominator | Male-to-female cases |
| Proportion | Numerator is part of denominator; ranges from 0 to 1 or 0% to 100% | Proportion of patients with controlled blood pressure |
| Rate | Measures occurrence over time; the denominator includes time or a defined time period | New tuberculosis cases per 100,000 population per year |
Applications in public health
1. Community diagnosis
Age-, sex- and location-specific measures reveal which groups carry the greatest burden. Mapping morbidity can identify clusters associated with water sources, housing, workplaces or poor access to care.
2. Outbreak investigation
Attack rates compare illness among exposed and unexposed groups, identify likely vehicles and prioritise control measures. Epidemic curves show the timing and pattern of exposure.
3. Chronic disease planning
Prevalence estimates how many people require continuing medicines, monitoring and complication prevention. Incidence shows whether new disease is increasing or prevention is succeeding.
4. Programme evaluation
Comparing measures before and after an intervention can suggest impact, but changes in diagnosis, reporting, migration or population structure must be considered.
5. Resource allocation
High-prevalence districts may need more long-term care, while high-incidence areas may require stronger prevention and case detection.
Worked applications
Example 1: clinic planning
A district has many people living with diabetes but relatively few new diagnoses each month. High prevalence indicates substantial continuing need for medicines, laboratory monitoring and complication screening. Incidence is still needed to judge prevention trends.
Example 2: school outbreak
After a shared meal, 60 of 200 students develop vomiting and diarrhoea. The overall attack rate is 60 ÷ 200 × 100 = 30%. Food-specific attack rates can help identify the likely vehicle.
Sources of morbidity data
- Outpatient, inpatient and emergency registers
- Notifiable-disease surveillance
- Laboratory reports
- Disease registries
- Household and health surveys
- Screening programmes
- Insurance, pharmacy and electronic health records
- Community-based reporting and special studies
Limitations and bias
| Problem | Effect | Response |
|---|---|---|
| Under-diagnosis | True burden is underestimated | Improve access, testing and community surveillance |
| Under-reporting | Recorded cases fall below diagnosed cases | Simplify tools, supervise and give feedback |
| Changing case definition | Apparent trend may not reflect real change | Document definitions and compare like with like |
| Referral bias | Facility data overrepresent severe disease | Use population-based sources where possible |
| Duplicate records | Events or people are counted more than once | Use identifiers and data-quality checks |
| Screening intensity | More testing increases detected cases | Interpret alongside testing volume and coverage |
| Population change | Crude counts become misleading | Use rates and current population estimates |
Comparing populations fairly
Crude measures may differ because populations have different age or sex structures. Use specific rates or standardisation where appropriate. Always report confidence intervals when estimating from a sample and disaggregate data to reveal inequities.
Morbidity and disability burden
Case frequency alone does not capture severity or duration. Additional measures include days lost from work or school, functional limitation, quality of life and years lived with disability. Composite measures such as disability-adjusted life years combine fatal and non-fatal burden for population comparison.
Interpretation checklist
- What is the exact case definition?
- Are new cases or all existing cases counted?
- Who is in the denominator and are they truly at risk?
- What time period is covered?
- Is the measure crude or specific?
- How complete and accurate are the data?
- Could access, testing or reporting have changed?
- What action should the result trigger?
Exam-focused summary
- Morbidity measures quantify non-fatal illness, injury and disability.
- Incidence measures new occurrence; prevalence measures existing burden.
- Attack rates are mainly used in outbreaks.
- Service data reflect both disease and healthcare access/utilisation.
- Correct numerators, denominators, time and case definitions are essential.
Review questions
- Define morbidity and state five reasons for measuring it.
- Differentiate a ratio, proportion and rate.
- List six sources of morbidity data.
- Explain how incidence and prevalence guide different decisions.
- Describe five limitations of routine morbidity data.
- State the steps for interpreting a morbidity measure.
References and further reading
- CDC Field Epidemiology Manual: Describing epidemiologic data.
- WHO Global Health Observatory indicator metadata.
- Clinical Medicine Year 2 curriculum.
Educational use only. Use current Uganda reporting definitions, denominators and national health-information guidance.
