Table 1.
Comparing the ZERO-G method to available methods for adjusting passive surveillance data.
| Input data | Output estimates | Advantages | Disadvantages | ||||
|---|---|---|---|---|---|---|---|
| Data source | Frequency | Spatial scale | Temporal resolution | Spatial resolution | |||
| Standard indirect estimators (e.g. WHO malaria report) | Passive surveillance data for focal disease | Annual | Subnational (Regional) | Annual | Regional |
· Straightforward adjustment method · Directly accounts for health-seeking behaviors |
· Only available at regional or national scales · Requires adequate coverage of DHS surveys · Limited to annual estimates · Not appropriate for rare diseases |
| Survey data of health-seeking behavior (e.g. DHS) | Multi-annual | Subnational (Regional) | |||||
| Ecological downscaling45 | Prevalence survey | Once or Multi-Annual | Subnational (Point data) | Annual | 5 × 5 km | · Avoids bias in passive surveillance data |
· Requires environmental and socio-economic variables · Requires prevalence data with adequate spatial coverage |
| Environmental Variables (e.g. Bioclim) | Annual to Long-term Average | 5 × 5 km | |||||
| Socio-economic variables | Multi-annual to annual | Regional | |||||
| ZERO-G estimator | Passive surveillance data for focal disease | Monthly | Community | Monthly | Community |
· Relies solely on health system data commonly available to Ministries of Health · Provides continuous, real-time estimates of incidence · Corrects for missing data due to data quality issues |
· Requires passive surveillance data at the community level · Only appropriate for diseases with regular incidence and reporting |
| All-cause consultation rates | Monthly | Community | |||||
| Health facility characteristics | Monthly to annual | Subnational (Facility·level) | |||||
All methods require basic administrative data, such as geographic boundaries of administrative zones and population, which are not mentioned here.