Table 2

Characteristics of indicators commonly used to justify health programmes.

Established validity as measure of health impact
Indicator
General ease of acquiring data to show health effects

Highest
• Crude Mortality, <5 mortality
Difficult in rural/diffuse settings, easier in camps

• Case fatality rate

High
• Nutritional status of children
Easy at the clinic data level, difficult but more valid with population surveys

• Disease rates


• Immunisation status of children


• Patient-specific mental health evaluations
Logistically easy, requires skill on part of evaluator

• Safety of blood supply

Moderate
• Food-basket evaluations
Easy in camps, more difficult in more diffuse populations

• Water and sanitation availability


• Reduction in measles, mumps and rubella through reproductive health services
Very difficult to measure even though benefits are likely to be occurring

• Improved patient outcomes via referrals


• Impregnated bednets distributed


• Comprehensive, timely health information system
Nearly impossible. These are difficult to measure, and all require a series of events to induce a health benefit

• Good coordination between sectors


• Knowledge & attitudes about services available


• Population practices

Low
• People given seeds and tools, shelter, or other materials
Easy to measure. Links to health are likely to be mediated via many steps.

• Drainage, fly control activities or tasks


• Number of clinic visits


• Distance to facilities, health workers per capita


• Trainings conducted, numbers trained
Easy to measure. May produce no effects on health.

• Change in knowledge without documented change in behaviour


• Messages/curricula developed


Roberts and Hofmann Emerging Themes in Epidemiology 2004 1:3   doi:10.1186/1742-7622-1-3