Public health

Sample: Public health, Kampala: report

Project: Sample: Public health, Kampala
Goal: Describe the project Sample: Public health, Kampala: run the analyses that tell the most about its data, draw every result, and write a cited report.
Run: 12778c77-4c3f-4379-9aca-c9e85676282e (running, 19 agent run(s))
Generated: 2026-10-19T04:37:18Z, version 1
Model cost: USD 0.0000 over 0 model call(s)

Data used

Timeline

The timeline counts 4,741 dated events in Cases, Community health worker visits, Vaccination sessions and 2 more per week, from week of 2025-09-29 to week of 2026-09-28 123456. By dataset: Cases 2,290; Community health worker visits 1,000; Weekly cases by sub-district 936; Clinic stock by month 315; Vaccination sessions 200 17. The trend is +1.18 events per week; the second half of the weeks has 3,039 against 1,597 in the first (+90.3%) 1. Week over week, the largest rise was from week of 2026-04-20 to week of 2026-04-27 (119 to 183, +64) and the largest fall was from week of 2026-06-01 to week of 2026-06-08 (147 to 87, -60) 18. The count fell from 36 in week of 2026-09-21 to 18 in week of 2026-09-28 (-50.0%), beyond what chance would give (p = 0.0198) 910. Between the two, the largest rise was in Riverside sub-district (1 to 3, +2) and the largest fall in Canal Side sub-district (5 to 1, -4) 910. The count fell from 9 in week of 2026-06-29 to 2 in week of 2026-07-06 (-77.8%), within what chance would give (p = 0.0654) 1112. Between the two, the largest rise was in Canal Side sub-district (0 to 1, +1) and the largest fall in Fernhill sub-district (2 to 0, -2) 1112. The count fell from 136 in week of 2026-05-18 to 86 in week of 2026-05-25 (-36.8%), beyond what chance would give (p = 0.000963) 1314. Between the two, the largest rise was in Foundry Row sub-district (5 to 8, +3) and the largest fall in Parkside sub-district (77 to 39, -38) 1314.

Busiest weeks

• week of 2026-05-11: 220 events 1
• week of 2026-05-18: 200 events 1
• week of 2026-05-04: 190 events 1

Quietest weeks

• week of 2025-09-29: 16 events 1
• week of 2025-10-13: 38 events 1
• week of 2026-02-09: 40 events 1

Overview

Overview: the top 3 results in the report's order, each in one distinct color, over Health sub-districts as context 15.

Spatial structure

Health sub-districts are 18 areas of 5.004 km2 on average, covering 100 percent of their extent, with no overlap 16. Cases are clustered (nearest-neighbor ratio 0.90, z = -9.4; features lie 87.9 m apart on average against 98 m if random) 17. Cases are dated by 'onset_date' from 2025-10-01 to 2026-09-30 at a week grain 17. Over time the pattern changes: the earlier half is random (ratio 0.9714) and the later half clustered (ratio 0.8894); the center moved 783 m 17. Clinics are evenly spread (dispersed) (nearest-neighbor ratio 1.58, z = +6.6; features lie 1.115 km apart on average against 705 m if random) 18. Water points are spread at random (nearest-neighbor ratio 0.96, z = -0.9; features lie 359 m apart on average against 373 m if random) 19. Community health worker visits are clustered (nearest-neighbor ratio 0.95, z = -3.3; features lie 139 m apart on average against 147 m if random) 20. Outbreak response teams are evenly spread (dispersed) (nearest-neighbor ratio 1.55, z = +2.6; features lie 1.635 km apart on average against 1.052 km if random) 21. Vaccination sessions are spread at random (nearest-neighbor ratio 1.00, z = +0.1; features lie 327 m apart on average against 326 m if random) 22. Vaccination sessions are dated by 'held_on' from 2026-01-05 to 2026-09-29 at a week grain 22. Over time the pattern changes: the earlier half is random (ratio 1.0941) and the later half dispersed (ratio 1.122); the center moved 259 m 22. 100 percent of Cases lie inside Health sub-districts (2290 of 2290) 1617. 100 percent of Clinics lie inside Health sub-districts (35 of 35) 1618. 100 percent of Community health worker visits lie inside Health sub-districts (1000 of 1000) 1620. 100 percent of Outbreak response teams lie inside Health sub-districts (6 of 6) 1621. 100 percent of Vaccination sessions lie inside Health sub-districts (200 of 200) 1622. 100 percent of Water points lie inside Health sub-districts (150 of 150) 1619. Outbreak response teams lie 35.3 m from the nearest of Cases (median); 83 percent lie within 100 m 1721. Water points lie 71.9 m from the nearest of Cases (median); 71 percent lie within 100 m 1719. Community health worker visits lie 77.4 m from the nearest of Cases (median); 62 percent lie within 100 m 1720. Vaccination sessions lie 79.5 m from the nearest of Cases (median); 64 percent lie within 100 m 1722.

Findings

• Evidenced: Health sub-districts are 18 areas of 5.004 km2 on average, covering 100 percent of their extent, with no overlap 16.
• Evidenced: Cases are clustered (nearest-neighbor ratio 0.90, z = -9.4; features lie 87.9 m apart on average against 98 m if random) 17.
• Evidenced: Cases are dated by 'onset_date' from 2025-10-01 to 2026-09-30 at a week grain 17.
• Evidenced: Over time the pattern changes: the earlier half is random (ratio 0.9714) and the later half clustered (ratio 0.8894); the center moved 783 m 17.
• Evidenced: Clinics are evenly spread (dispersed) (nearest-neighbor ratio 1.58, z = +6.6; features lie 1.115 km apart on average against 705 m if random) 18.
• Evidenced: Water points are spread at random (nearest-neighbor ratio 0.96, z = -0.9; features lie 359 m apart on average against 373 m if random) 19.
• Evidenced: Community health worker visits are clustered (nearest-neighbor ratio 0.95, z = -3.3; features lie 139 m apart on average against 147 m if random) 20.
• Evidenced: Outbreak response teams are evenly spread (dispersed) (nearest-neighbor ratio 1.55, z = +2.6; features lie 1.635 km apart on average against 1.052 km if random) 21.
• Evidenced: Vaccination sessions are spread at random (nearest-neighbor ratio 1.00, z = +0.1; features lie 327 m apart on average against 326 m if random) 22.
• Evidenced: Vaccination sessions are dated by 'held_on' from 2026-01-05 to 2026-09-29 at a week grain 22.
• Evidenced: Over time the pattern changes: the earlier half is random (ratio 1.0941) and the later half dispersed (ratio 1.122); the center moved 259 m 22.
• Evidenced: 100 percent of Cases lie inside Health sub-districts (2290 of 2290) 1617.
• Evidenced: 100 percent of Clinics lie inside Health sub-districts (35 of 35) 1618.
• Evidenced: 100 percent of Community health worker visits lie inside Health sub-districts (1000 of 1000) 1620.
• Evidenced: 100 percent of Outbreak response teams lie inside Health sub-districts (6 of 6) 1621.
• Evidenced: 100 percent of Vaccination sessions lie inside Health sub-districts (200 of 200) 1622.
• Evidenced: 100 percent of Water points lie inside Health sub-districts (150 of 150) 1619.
• Evidenced: Outbreak response teams lie 35.3 m from the nearest of Cases (median); 83 percent lie within 100 m 1721.
• Evidenced: Water points lie 71.9 m from the nearest of Cases (median); 71 percent lie within 100 m 1719.
• Evidenced: Community health worker visits lie 77.4 m from the nearest of Cases (median); 62 percent lie within 100 m 1720.
• Evidenced: Vaccination sessions lie 79.5 m from the nearest of Cases (median); 64 percent lie within 100 m 1722.

Space time hot spots

What this analysis is. Emerging hot spots: Where hot spots are new, consecutive, intensifying, persistent, diminishing, sporadic, oscillating or historical over time 23. In ArcGIS it is Emerging Hot Spot Analysis or Create Space Time Cube 23.

Why it was run on this data. Community health worker visits carry dates in 'visited_at': emerging hot spot analysis finds new, intensifying and fading hot spots per week 24.

What it found. Space time hot spots on Community health worker visits: Emerging hot spots: 1 layer saved: Emerging hot spots: Community health worker visits (124 features) 2025. It reported result, count 124; result, elapsed ms 17; result, resolution 8; result, features used 1,000; result, time neighbors 0; billing, cost 0.15 25.

What the map shows. The map draws Emerging hot spots: community health worker visits by category in 2 classes: Sporadic hot spot, No pattern detected 26. The map draws Health sub-districts with one symbol 26. The map draws Emerging hot spots: cases with one symbol 26. It draws 267 features 26. It covers longitude 32.5 to 32.661 and latitude 0.252 to 0.372, at about 1:74,770 26.

Findings

• Evidenced: Space time hot spots on Community health worker visits: Emerging hot spots: 1 layer saved: Emerging hot spots: Community health worker visits (124 features) 2025.
• Evidenced: It reported result, count 124; result, elapsed ms 17; result, resolution 8; result, features used 1,000; result, time neighbors 0; billing, cost 0.15 25.

Tool gp_space_time_hot_spots; parameters {"project": "2612067c-badd-4310-9678-8f580d5e3d32", "features": {"layer": "Community health worker visits (4d70d1ce)"}, "interval": "week", "bin_shape": "h3", "resolution": 8, "time_field": "visited_at"}

Space time hot spots

What this analysis is. Emerging hot spots: Where hot spots are new, consecutive, intensifying, persistent, diminishing, sporadic, oscillating or historical over time 23. In ArcGIS it is Emerging Hot Spot Analysis or Create Space Time Cube 23.

Why it was run on this data. Vaccination sessions carry dates in 'held_on': emerging hot spot analysis finds new, intensifying and fading hot spots per week 27. Measured: the pattern of Vaccination sessions changes over time (earlier half random, later half dispersed; the center moved 259 m) 27.

What it found. The analysis of vaccination sessions using the gp_space_time_hot_spots tool resulted in the generation of 24 features, saving the output as "Emerging hot spots: Vaccination sessions" 28. This computation had an elapsed time of 5 milliseconds 28. The process utilized 200 features and had a resolution of 7 28. The total cost incurred for this operation was 0.11 28.

What the map shows. The map draws Emerging hot spots: Vaccination sessions by category in 2 classes: Sporadic hot spot, No pattern detected 29. The map draws Health sub-districts with one symbol 29. It draws 42 features 29. It covers longitude 32.489 to 32.673 and latitude 0.239 to 0.376, at about 1:85,295 29.

Findings

• Evidenced: Space time hot spots on Vaccination sessions: Emerging hot spots: 1 layer saved: Emerging hot spots: Vaccination sessions (24 features) 2228.
• Evidenced: It reported result, count 24; result, elapsed ms 5; result, resolution 7; result, features used 200; result, time neighbors 0; billing, cost 0.11 28.

Tool gp_space_time_hot_spots; parameters {"project": "2612067c-badd-4310-9678-8f580d5e3d32", "features": {"layer": "Vaccination sessions (6798c97c)"}, "interval": "week", "bin_shape": "h3", "resolution": 7, "time_field": "held_on"}

Space time hot spots

What this analysis is. Emerging hot spots: Where hot spots are new, consecutive, intensifying, persistent, diminishing, sporadic, oscillating or historical over time 23. In ArcGIS it is Emerging Hot Spot Analysis or Create Space Time Cube 23.

Why it was run on this data. Cases carry dates in 'onset_date': emerging hot spot analysis finds new, intensifying and fading hot spots per week 30. Measured: the pattern of Cases changes over time (earlier half random, later half clustered; the center moved 783 m) 30.

What it found. Space time hot spots on Cases: Emerging hot spots: 1 layer saved: Emerging hot spots: Cases (125 features) 1731. It reported result, count 125; result, elapsed ms 26; result, resolution 8; result, features used 2,290; result, time neighbors 0; billing, cost 0.2145 31.

What the map shows. Drawn with space time hot spots on the map Emerging hot spots: community health worker visits and cases 26.

Findings

• Evidenced: Space time hot spots on Cases: Emerging hot spots: 1 layer saved: Emerging hot spots: Cases (125 features) 1731.
• Evidenced: It reported result, count 125; result, elapsed ms 26; result, resolution 8; result, features used 2,290; result, time neighbors 0; billing, cost 0.2145 31.

Tool gp_space_time_hot_spots; parameters {"project": "2612067c-badd-4310-9678-8f580d5e3d32", "features": {"layer": "Cases (03c8f72b)"}, "interval": "week", "bin_shape": "h3", "resolution": 8, "time_field": "onset_date"}

Hot spots

What this analysis is. Hot spot analysis: Getis-Ord Gi* hot and cold spots with z-scores, p-values and confidence bins 32. In ArcGIS it is Find Hot Spots or Hot Spot Analysis (Getis-Ord Gi*); in QGIS, qgis:hotspot (plugin) 32.

Why it was run on this data. 1000 points: Getis-Ord Gi* tests whether the concentrations in Community health worker visits are statistically significant 33. Measured: Community health worker visits are clustered (nearest-neighbor ratio 0.95, z = -3.3), median spacing 124 m 33. Cells of 500 m and a band of 1000 m from the measured spacing 33.

What it found. Hot spots on Community health worker visits: Hot spot analysis: 1 layer saved: Hot spot analysis: Community health worker visits (390 features) 2034. It reported result, bins 390; result, count 390; result, elapsed ms 17; result, cell size m 500; result, incidents used 1,000; result, distance band m 1,000 34.

What the map shows. The map draws Hot spot analysis: Community health worker visits by gi_bin in 7 classes: Hot spot - 99% confidence, Hot spot - 95% confidence, Hot spot - 90% confidence, Not significant, Cold spot - 90% confidence, Cold spot - 95% confidence and 1 more 35. The map draws Health sub-districts with one symbol 35. It draws 408 features 35. It covers longitude 32.51 to 32.652 and latitude 0.257 to 0.366, at about 1:66,000 35.

Findings

• Evidenced: Hot spots on Community health worker visits: Hot spot analysis: 1 layer saved: Hot spot analysis: Community health worker visits (390 features) 2034.
• Evidenced: It reported result, bins 390; result, count 390; result, elapsed ms 17; result, cell size m 500; result, incidents used 1,000; result, distance band m 1,000 34.

Tool gp_hot_spots; parameters {"project": "2612067c-badd-4310-9678-8f580d5e3d32", "features": {"layer": "Community health worker visits (4d70d1ce)"}, "aggregation": "hexagon", "cell_size_m": 500.0, "distance_band_m": 1000.0}

Hot spots

What this analysis is. Hot spot analysis: Getis-Ord Gi* hot and cold spots with z-scores, p-values and confidence bins 32. In ArcGIS it is Find Hot Spots or Hot Spot Analysis (Getis-Ord Gi*); in QGIS, qgis:hotspot (plugin) 32.

Why it was run on this data. 2290 points: Getis-Ord Gi* tests whether the concentrations in Cases are statistically significant 36. Measured: Cases are clustered (nearest-neighbor ratio 0.90, z = -9.4), median spacing 75.6 m 36. Cells of 200 m and a band of 500 m from the measured spacing 36.

What it found. Hot spots on Cases: Hot spot analysis: 1 layer saved: Hot spot analysis: Cases (2259 features) 1737. It reported result, bins 2,259; result, count 2,259; result, elapsed ms 75; result, cell size m 200; result, incidents used 2,290; result, distance band m 500 37.

What the map shows. The map draws Hot spot analysis: Cases by gi_bin in 7 classes: Hot spot - 99% confidence, Hot spot - 95% confidence, Hot spot - 90% confidence, Not significant, Cold spot - 90% confidence, Cold spot - 95% confidence and 1 more 38. The map draws Health sub-districts with one symbol 38. It draws 2,277 features 38. It covers longitude 32.512 to 32.649 and latitude 0.26 to 0.365, at about 1:63,311 38.

Findings

• Evidenced: Hot spots on Cases: Hot spot analysis: 1 layer saved: Hot spot analysis: Cases (2259 features) 1737.
• Evidenced: It reported result, bins 2,259; result, count 2,259; result, elapsed ms 75; result, cell size m 200; result, incidents used 2,290; result, distance band m 500 37.

Tool gp_hot_spots; parameters {"project": "2612067c-badd-4310-9678-8f580d5e3d32", "features": {"layer": "Cases (03c8f72b)"}, "aggregation": "hexagon", "cell_size_m": 200.0, "distance_band_m": 500.0}

Hot spots

What this analysis is. Hot spot analysis: Getis-Ord Gi* hot and cold spots with z-scores, p-values and confidence bins 32. In ArcGIS it is Find Hot Spots or Hot Spot Analysis (Getis-Ord Gi*); in QGIS, qgis:hotspot (plugin) 32.

Why it was run on this data. 200 points: Getis-Ord Gi* tests whether the concentrations in Vaccination sessions are statistically significant 39.

What it found. Hot spots on Vaccination sessions: Hot spot analysis: 1 layer saved: Hot spot analysis: Vaccination sessions (99 features) 2240. It reported result, bins 99; result, count 99; result, elapsed ms 5; result, cell size m 1,000; result, incidents used 200; result, distance band m 2,000 40.

What the map shows. The map draws Hot spot analysis: Vaccination sessions by gi_bin in 7 classes: Hot spot - 99% confidence, Hot spot - 95% confidence, Hot spot - 90% confidence, Not significant, Cold spot - 90% confidence, Cold spot - 95% confidence and 1 more 41. The map draws Health sub-districts with one symbol 41. It draws 117 features 41. It covers longitude 32.499 to 32.659 and latitude 0.251 to 0.373, at about 1:73,740 41.

Findings

• Evidenced: Hot spots on Vaccination sessions: Hot spot analysis: 1 layer saved: Hot spot analysis: Vaccination sessions (99 features) 2240.
• Evidenced: It reported result, bins 99; result, count 99; result, elapsed ms 5; result, cell size m 1,000; result, incidents used 200; result, distance band m 2,000 40.

Tool gp_hot_spots; parameters {"project": "2612067c-badd-4310-9678-8f580d5e3d32", "features": {"layer": "Vaccination sessions (6798c97c)"}, "aggregation": "hexagon", "cell_size_m": 1000.0, "distance_band_m": 2000.0}

Summarize within

What this analysis is. Summarize within: Count and aggregate the features inside each polygon (or bin), proportionally for lines and polygons 42. In ArcGIS it is Summarize Within or Aggregate Points; in QGIS, native:countpointsinpolygon or native:sumlinelengths 42.

Why it was run on this data. Count Community health worker visits inside each feature of Health sub-districts by purpose; the counts map as a choropleth (natural breaks, sequential colors) 43.

What it found. Summarize within on Community health worker visits, Health sub-districts: Summarize within: 1 layer saved: Summarize within: Health sub-districts (18 features) 162044. It reported result, count 18; result, elapsed ms 2; billing, cost 0.03036; billing, list price 0.03036; billing, upstream cost 0 44.

What the map shows. The map draws Summarize within: Health sub-districts by count in 5 classes: 14 to 28, 28 to 53, 53 to 75, 75 to 116, 116 to 150 45. It draws 18 features 45. It covers longitude 32.515 to 32.648 and latitude 0.261 to 0.363, at about 1:61,580 45.

Findings

• Evidenced: Summarize within on Community health worker visits, Health sub-districts: Summarize within: 1 layer saved: Summarize within: Health sub-districts (18 features) 162044.
• Evidenced: It reported result, count 18; result, elapsed ms 2; billing, cost 0.03036; billing, list price 0.03036; billing, upstream cost 0 44.

Tool gp_summarize_within; parameters {"stats": [], "project": "2612067c-badd-4310-9678-8f580d5e3d32", "features": {"layer": "Community health worker visits (4d70d1ce)"}, "group_by": "purpose", "polygons": {"layer": "Health sub-districts (433b88cc)"}}

Summarize within

What this analysis is. Summarize within: Count and aggregate the features inside each polygon (or bin), proportionally for lines and polygons 42. In ArcGIS it is Summarize Within or Aggregate Points; in QGIS, native:countpointsinpolygon or native:sumlinelengths 42.

Why it was run on this data. Count Community health worker visits inside each feature of Health sub-districts by purpose; the counts map as a choropleth (natural breaks, sequential colors) 43.

What it found. Summarize within on Vaccination sessions, Health sub-districts: Summarize within: 1 layer saved: Summarize within: Health sub-districts (18 features) 162246. It reported result, count 18; result, elapsed ms 0; billing, cost 0.01438; billing, list price 0.01438; billing, upstream cost 0 46.

What the map shows. The map draws Summarize within: Health sub-districts by count in 5 classes: 3 to 7, 7 to 11, 11 to 19, 19 to 27.5, 27.5 to 32 47. It draws 18 features 47. It covers longitude 32.515 to 32.648 and latitude 0.261 to 0.363, at about 1:61,580 47.

Findings

• Evidenced: Summarize within on Vaccination sessions, Health sub-districts: Summarize within: 1 layer saved: Summarize within: Health sub-districts (18 features) 162246.
• Evidenced: It reported result, count 18; result, elapsed ms 0; billing, cost 0.01438; billing, list price 0.01438; billing, upstream cost 0 46.

Tool gp_summarize_within; parameters {"stats": [{"stat": "sum", "field": "doses"}], "project": "2612067c-badd-4310-9678-8f580d5e3d32", "features": {"layer": "Vaccination sessions (6798c97c)"}, "group_by": "vaccine", "polygons": {"layer": "Health sub-districts (433b88cc)"}}

Summarize within

What this analysis is. Summarize within: Count and aggregate the features inside each polygon (or bin), proportionally for lines and polygons 42. In ArcGIS it is Summarize Within or Aggregate Points; in QGIS, native:countpointsinpolygon or native:sumlinelengths 42.

Why it was run on this data. Count Community health worker visits inside each feature of Health sub-districts by purpose; the counts map as a choropleth (natural breaks, sequential colors) 43.

What it found. The gp_summarize_within function summarized Health sub-districts, which contained 18 features 48.

What the map shows. The map draws Summarize within: Health sub-districts by count in 5 classes: 33 to 64, 64 to 108, 108 to 291, 291 to 402, 402 to 513 49. It draws 18 features 49. It covers longitude 32.515 to 32.648 and latitude 0.261 to 0.363, at about 1:61,580 49.

Findings

• Evidenced: Summarize within on Cases, Health sub-districts: Summarize within: 1 layer saved: Summarize within: Health sub-districts (18 features) 161748.
• Evidenced: It reported result, count 18; result, elapsed ms 3; billing, cost 0.05616; billing, list price 0.05616; billing, upstream cost 0 48.

Tool gp_summarize_within; parameters {"stats": [], "project": "2612067c-badd-4310-9678-8f580d5e3d32", "features": {"layer": "Cases (03c8f72b)"}, "group_by": "disease", "polygons": {"layer": "Health sub-districts (433b88cc)"}}

Period change

What this analysis is. Period change: Counts or sums per period (week over week, month over month, any two periods, or before and after a date), per area, per H3 cell or in total, with the places of largest rise and fall and a significance note 50. In ArcGIS it is Summarize Percent Change (Crime Analysis) or Change Point Detection 50.

Why it was run on this data. Community health worker visits carry dates in 'visited_at' over 26 weeks: the count per week, week over week, with the places of largest rise and fall per feature of Health sub-districts and whether the change is beyond chance 51.

What it found. Period change on Health sub-districts, Community health worker visits produced Period change: Community health worker visits (week of 2026-09-21 to week of 2026-09-28) (14 features) 9101620. Community health worker visits: count per week shows a trend of -0.0873 per week; the second half of its weeks has 488 against 472 in the first (+3.4%) 53. The chart Community health worker visits: week of 2026-09-21 to week of 2026-09-28 by place is drawn from 28 rows 54. The chart Period change: Community health worker visits (week of 2026-09-21 to week of 2026-09-28) is drawn from 5 rows 55.

What the map shows. The map draws Period change: Community health worker visits (week of 2026-09-21 to week of 2026-09-28) by change in 5 classes: < -1.5 Std; Dev, -1.5 to -0.5 Std; Dev, -0.5 to 0.5 Std; Dev, 0.5 to 1.5 Std; Dev, > 1.5 Std; Dev 52. It draws 14 features 52. It covers longitude 32.512 to 32.65 and latitude 0.261 to 0.364, at about 1:63,751 52.

What the chart shows. The time series chart illustrates community health worker visits per week from March 30, 2026, to September 28, 2026 53. The overall trend is a minor decrease of -0.0873 per week, with the first half averaging 472 and the second half averaging 488, which is a +3.4% increase 53. The data shows fluctuations throughout the period, with the count reaching a peak of approximately 48 around the week of July 6 and a noticeable drop to 18 in the final week of September 28 53. The analysis highlights a specific fall from 36 in the week of September 21 to 18 in the week of September 28 (a -50.0% change), which is stated as unlikely to be chance (p = 0.0198) 53.

Inferred: read from the image by gemma4:e4b, not evidence.

Findings

• Evidenced: Period change on Health sub-districts, Community health worker visits produced Period change: Community health worker visits (week of 2026-09-21 to week of 2026-09-28) (14 features) 9101620.
• Evidenced: Community health worker visits: count per week shows a trend of -0.0873 per week; the second half of its weeks has 488 against 472 in the first (+3.4%) 53.
• Evidenced: The chart Community health worker visits: week of 2026-09-21 to week of 2026-09-28 by place is drawn from 28 rows 54.
• Evidenced: The chart Period change: Community health worker visits (week of 2026-09-21 to week of 2026-09-28) is drawn from 5 rows 55.

Tool gp_period_change; parameters {"by": "area", "areas": {"layer": "Health sub-districts (433b88cc)"}, "period": "week", "project": "2612067c-badd-4310-9678-8f580d5e3d32", "features": {"layer": "Community health worker visits (4d70d1ce)"}, "area_field": "name", "time_field": "visited_at"}

Period change

What this analysis is. Period change: Counts or sums per period (week over week, month over month, any two periods, or before and after a date), per area, per H3 cell or in total, with the places of largest rise and fall and a significance note 50. In ArcGIS it is Summarize Percent Change (Crime Analysis) or Change Point Detection 50.

Why it was run on this data. Community health worker visits carry dates in 'visited_at' over 26 weeks: the count per week, week over week, with the places of largest rise and fall per feature of Health sub-districts and whether the change is beyond chance 51.

What it found. Period change on Health sub-districts, Vaccination sessions produced Period change: Vaccination sessions (week of 2026-06-29 to week of 2026-07-06) (10 features) 11121622. Vaccination sessions: count per week shows a trend of -0.00587 per week; the second half of its weeks has 103 against 95 in the first (+8.4%) 57. The chart Vaccination sessions: week of 2026-06-29 to week of 2026-07-06 by place is drawn from 20 rows 58. The chart Period change: Vaccination sessions (week of 2026-06-29 to week of 2026-07-06) is drawn from 5 rows 59.

What the map shows. The map draws Period change: Vaccination sessions (week of 2026-06-29 to week of 2026-07-06) by change in 4 classes: < -0.5 Std; Dev, -0.5 to 0.5 Std; Dev, 0.5 to 1.5 Std; Dev, > 1.5 Std; Dev 56. It draws 10 features 56. It covers longitude 32.512 to 32.65 and latitude 0.261 to 0.364, at about 1:63,751 56.

What the chart shows. Overall, the vaccination sessions count fluctuates significantly over the 39-week period from January to September 20 57.

Inferred: read from the image by gemma4:e4b, not evidence.

Findings

• Evidenced: Period change on Health sub-districts, Vaccination sessions produced Period change: Vaccination sessions (week of 2026-06-29 to week of 2026-07-06) (10 features) 11121622.
• Evidenced: Vaccination sessions: count per week shows a trend of -0.00587 per week; the second half of its weeks has 103 against 95 in the first (+8.4%) 57.
• Evidenced: The chart Vaccination sessions: week of 2026-06-29 to week of 2026-07-06 by place is drawn from 20 rows 58.
• Evidenced: The chart Period change: Vaccination sessions (week of 2026-06-29 to week of 2026-07-06) is drawn from 5 rows 59.

Tool gp_period_change; parameters {"by": "area", "areas": {"layer": "Health sub-districts (433b88cc)"}, "period": "week", "project": "2612067c-badd-4310-9678-8f580d5e3d32", "features": {"layer": "Vaccination sessions (6798c97c)"}, "area_field": "name", "time_field": "held_on"}

Period change

What this analysis is. Period change: Counts or sums per period (week over week, month over month, any two periods, or before and after a date), per area, per H3 cell or in total, with the places of largest rise and fall and a significance note 50. In ArcGIS it is Summarize Percent Change (Crime Analysis) or Change Point Detection 50.

Why it was run on this data. Community health worker visits carry dates in 'visited_at' over 26 weeks: the count per week, week over week, with the places of largest rise and fall per feature of Health sub-districts and whether the change is beyond chance 51.

What it found. Period change on Health sub-districts, Cases produced Period change: Cases (week of 2026-05-18 to week of 2026-05-25) (18 features) 13141617. Cases: count per week shows a trend of -0.115 per week; the second half of its weeks has 1,289 against 980 in the first (+31.5%) 61. The chart Cases: week of 2026-05-18 to week of 2026-05-25 by place is drawn from 30 rows 62. The chart Period change: Cases (week of 2026-05-18 to week of 2026-05-25) is drawn from 6 rows 63.

What the map shows. The map draws Period change: Cases (week of 2026-05-18 to week of 2026-05-25) by change in 6 classes: < -3.5 Std; Dev, -3.5 to -2.5 Std; Dev, -2.5 to -1.5 Std; Dev, -1.5 to -0.5 Std; Dev, -0.5 to 0.5 Std; Dev, > 0.5 Std; Dev 60. It draws 18 features 60. It covers longitude 32.512 to 32.65 and latitude 0.261 to 0.364, at about 1:63,751 60.

Findings

• Evidenced: Period change on Health sub-districts, Cases produced Period change: Cases (week of 2026-05-18 to week of 2026-05-25) (18 features) 13141617.
• Evidenced: Cases: count per week shows a trend of -0.115 per week; the second half of its weeks has 1,289 against 980 in the first (+31.5%) 61.
• Evidenced: The chart Cases: week of 2026-05-18 to week of 2026-05-25 by place is drawn from 30 rows 62.
• Evidenced: The chart Period change: Cases (week of 2026-05-18 to week of 2026-05-25) is drawn from 6 rows 63.

Tool gp_period_change; parameters {"by": "area", "areas": {"layer": "Health sub-districts (433b88cc)"}, "period": "week", "project": "2612067c-badd-4310-9678-8f580d5e3d32", "features": {"layer": "Cases (03c8f72b)"}, "area_field": "name", "time_field": "onset_date"}

Calculate density

What this analysis is. Calculate density: Kernel density per square kilometer on a grid 64. In ArcGIS it is Calculate Density or Kernel Density; in QGIS, native:heatmapkerneldensityestimation 64.

Why it was run on this data. 1000 points in Community health worker visits: a density surface shows where they concentrate 65. Measured: Community health worker visits are clustered (nearest-neighbor ratio 0.95, z = -3.3), median spacing 124 m 65.

What it found. Calculate density on Community health worker visits: Calculate density: 1 layer saved: Calculate density: Community health worker visits (2452 features) 2066. It reported result, count 2,452; result, elapsed ms 43; result, bandwidth m 500; result, cell size m 200; result, excluded null 0; result, max density per km2 34.43 66.

What the map shows. The map draws Calculate density: Community health worker visits by density_per_km2 in 6 classes: 2.99 to 6.52, 6.52 to 9.98, 9.98 to 13.6, 13.6 to 18, 18 to 23.7, 23.7 to 34.4 67. The map draws Health sub-districts with one symbol 67. The map draws Dense areas of community health worker visits with one symbol 67. It draws 2,475 features 67. It covers longitude 32.509 to 32.654 and latitude 0.257 to 0.368, at about 1:66,921 67.

Findings

• Evidenced: Calculate density on Community health worker visits: Calculate density: 1 layer saved: Calculate density: Community health worker visits (2452 features) 2066.
• Evidenced: It reported result, count 2,452; result, elapsed ms 43; result, bandwidth m 500; result, cell size m 200; result, excluded null 0; result, max density per km2 34.43 66.

A model draft was discarded because a sentence did not cite a source of this section.

Tool gp_calculate_density; parameters {"kernel": "quartic", "project": "2612067c-badd-4310-9678-8f580d5e3d32", "features": {"layer": "Community health worker visits (4d70d1ce)"}, "bandwidth_m": 500.0, "cell_size_m": 200.0}

Calculate density

What this analysis is. Calculate density: Kernel density per square kilometer on a grid 64. In ArcGIS it is Calculate Density or Kernel Density; in QGIS, native:heatmapkerneldensityestimation 64.

Why it was run on this data. 2290 points in Cases: a density surface shows where they concentrate 68. Measured: Cases are clustered (nearest-neighbor ratio 0.90, z = -9.4), median spacing 75.6 m 68.

What it found. The density was calculated using the Cases input 17 and processed 2,508 features 69.

What the map shows. The map draws Calculate density: Cases by density_per_km2 in 6 classes: 9.49 to 19.7, 19.7 to 29.4, 29.4 to 57.9, 57.9 to 111, 111 to 189, 189 to 282 70. The map draws Health sub-districts with one symbol 70. It draws 2,526 features 70. It covers longitude 32.508 to 32.653 and latitude 0.257 to 0.367, at about 1:66,921 70.

Findings

• Evidenced: Calculate density on Cases: Calculate density: 1 layer saved: Calculate density: Cases (2508 features) 1769.
• Evidenced: It reported result, count 2,508; result, elapsed ms 49; result, bandwidth m 500; result, cell size m 200; result, excluded null 0; result, max density per km2 282 69.

Tool gp_calculate_density; parameters {"kernel": "quartic", "project": "2612067c-badd-4310-9678-8f580d5e3d32", "features": {"layer": "Cases (03c8f72b)"}, "bandwidth_m": 500.0, "cell_size_m": 200.0}

Calculate density

What this analysis is. Calculate density: Kernel density per square kilometer on a grid 64. In ArcGIS it is Calculate Density or Kernel Density; in QGIS, native:heatmapkerneldensityestimation 64.

Why it was run on this data. 200 points in Vaccination sessions: a density surface shows where they concentrate 71.

What it found. Calculate density on Vaccination sessions: Calculate density: 1 layer saved: Calculate density: Vaccination sessions (616 features) 2272. It reported result, count 616; result, elapsed ms 105; result, bandwidth m 2,000; result, cell size m 500; result, excluded null 0; result, max density per km2 5.525 72.

What the map shows. The map draws Calculate density: Vaccination sessions by density_per_km2 in 6 classes: 0.404 to 1.01, 1.01 to 1.59, 1.59 to 2.25, 2.25 to 3.27, 3.27 to 4.47, 4.47 to 5.53 73. The map draws Health sub-districts with one symbol 73. It draws 634 features 73. It covers longitude 32.49 to 32.674 and latitude 0.241 to 0.382, at about 1:85,295 73.

Findings

• Evidenced: Calculate density on Vaccination sessions: Calculate density: 1 layer saved: Calculate density: Vaccination sessions (616 features) 2272.
• Evidenced: It reported result, count 616; result, elapsed ms 105; result, bandwidth m 2,000; result, cell size m 500; result, excluded null 0; result, max density per km2 5.525 72.

Tool gp_calculate_density; parameters {"kernel": "quartic", "project": "2612067c-badd-4310-9678-8f580d5e3d32", "features": {"layer": "Vaccination sessions (6798c97c)"}, "bandwidth_m": 2000.0, "cell_size_m": 500.0}

Find point clusters

What this analysis is. Find point clusters: Density-based clusters (DBSCAN, HDBSCAN or OPTICS, optionally space-time) with hulls and noise 74. In ArcGIS it is Find Point Clusters or Density-based Clustering; in QGIS, native:dbscanclustering or native:stdbscanclustering 74.

Why it was run on this data. DBSCAN finds dense groups of Community health worker visits and separates noise 75. Measured: Community health worker visits are clustered (nearest-neighbor ratio 0.95, z = -3.3), median spacing 124 m 75. DBSCAN at 200 m (twice the measured spacing) 75.

What it found. Find point clusters on Community health worker visits: Find point clusters: 1 layer saved: Find point clusters: Community health worker visits (1000 features) 2076. It reported result, count 1,000; result, noise 1,000; result, distance m 200; result, elapsed ms 576; result, min points 20; billing, cost 0.07 76.

What the map shows. The map draws Find point clusters: Community health worker visits with one symbol 77. It draws 1,000 features 77. It covers longitude 32.516 to 32.647 and latitude 0.263 to 0.363, at about 1:60,312 77.

Findings

• Evidenced: Find point clusters on Community health worker visits: Find point clusters: 1 layer saved: Find point clusters: Community health worker visits (1000 features) 2076.
• Evidenced: It reported result, count 1,000; result, noise 1,000; result, distance m 200; result, elapsed ms 576; result, min points 20; billing, cost 0.07 76.

Tool gp_find_point_clusters; parameters {"method": "dbscan", "project": "2612067c-badd-4310-9678-8f580d5e3d32", "features": {"layer": "Community health worker visits (4d70d1ce)"}, "distance_m": 200.0, "min_points": 20}

Find point clusters

What this analysis is. Find point clusters: Density-based clusters (DBSCAN, HDBSCAN or OPTICS, optionally space-time) with hulls and noise 74. In ArcGIS it is Find Point Clusters or Density-based Clustering; in QGIS, native:dbscanclustering or native:stdbscanclustering 74.

Why it was run on this data. DBSCAN finds dense groups of Cases and separates noise 78. Measured: Cases are clustered (nearest-neighbor ratio 0.90, z = -9.4), median spacing 75.6 m 78. DBSCAN at 200 m (twice the measured spacing) 78.

What it found. Find point clusters on Cases: Find point clusters: 2 layers saved: Find point clusters: Cases (2290 features); Find point clusters: cluster hulls (1 feature) 1779. It reported result, count 2,290; result, noise 1,961; result, distance m 200; result, elapsed ms 24; result, min points 20; billing, cost 0.0958 79.

What the map shows. The map draws Find point clusters: cluster hulls with one symbol 80. The map draws Find point clusters: Cases by cluster_id in 2 classes: Cluster 0, Noise 80. It draws 2,291 features 80. It covers longitude 32.516 to 32.647 and latitude 0.262 to 0.363, at about 1:60,732 80.

Findings

• Evidenced: Find point clusters on Cases: Find point clusters: 2 layers saved: Find point clusters: Cases (2290 features); Find point clusters: cluster hulls (1 feature) 1779.
• Evidenced: It reported result, count 2,290; result, noise 1,961; result, distance m 200; result, elapsed ms 24; result, min points 20; billing, cost 0.0958 79.

Tool gp_find_point_clusters; parameters {"method": "dbscan", "project": "2612067c-badd-4310-9678-8f580d5e3d32", "features": {"layer": "Cases (03c8f72b)"}, "distance_m": 200.0, "min_points": 20}

Outliers

What this analysis is. Find outliers: Anselin local Moran’s I: HH, LL clusters and HL, LH outliers 81. In ArcGIS it is Find Outliers or Cluster and Outlier Analysis (Anselin Local Moran’s I) 81.

Why it was run on this data. Local Moran's I maps clusters and outliers of 'population' 82. Measured: Health sub-districts have 39 neighboring pairs (4.33 neighbors per area, 0 isolated), so neighbors can be compared 82.

What it found. The outlier analysis process used Health sub-districts as input data 16. The operation, which saved 1 layer for Health sub-districts, successfully analyzed 18 features 83.

What the map shows. The map draws Find outliers: Health sub-districts by category in 5 classes: High-high cluster, High-low outlier, Low-high outlier, Low-low cluster, Not significant 84. It draws 18 features 84. It covers longitude 32.515 to 32.648 and latitude 0.261 to 0.363, at about 1:61,580 84.

Findings

• Evidenced: Outliers on Health sub-districts: Find outliers: 1 layer saved: Find outliers: Health sub-districts (18 features) 1683.
• Evidenced: It reported result, count 18; result, elapsed ms 2; result, permutations 199; result, excluded null 0; result, distance band m 2,467; billing, cost 0.05054 83.

Tool gp_outliers; parameters {"field": "population", "project": "2612067c-badd-4310-9678-8f580d5e3d32", "features": {"layer": "Health sub-districts (433b88cc)"}, "aggregation": "hexagon"}

What follows from the results

Change beside the counts per health sub-district, from Period change: Cases (week of 2026-05-18 to week of 2026-05-25) and Summarize within: Health sub-districts: the busiest area, Parkside sub-district (32 in all), changed by -38 871346. Change of vaccination sessions where density is high, from Period change: Vaccination sessions (week of 2026-06-29 to week of 2026-07-06) and Calculate density: Community health worker visits: the largest rise (+1, Canal Side sub-district) is where density averages 10.88 per square kilometer 921166. Hot cells of cases per health sub-district, from Hot spot analysis: Cases: 8 of 18 areas hold hot spot cells; Parkside sub-district holds the most (103) 9737. Nearest outbreak response team to each cluster of cases, from Find point clusters: cluster hulls: the 1 clusters lie 0 to 0 m from the nearest facility (Rapid response team 1 and others) 10379. Dense areas of community health worker visits, from Calculate density: Community health worker visits: the densest of five classes joins 491 cells into one area 11066. Change of vaccination sessions per health sub-district, from Period change: Vaccination sessions (week of 2026-06-29 to week of 2026-07-06): Canal Side sub-district rose the most (+1) and Fernhill sub-district fell the most (-2) 11311. Outbreak response teams within 250 m of each cluster of cases, from Find point clusters: cluster hulls: 1 of 1 clusters have one within 250 m 11879. Density of community health worker visits and cases together, from Calculate density: Community health worker visits and Calculate density: Cases: 4960 features from 2 sources 1236669.

Findings

• Evidenced: Change beside the counts per health sub-district, from Period change: Cases (week of 2026-05-18 to week of 2026-05-25) and Summarize within: Health sub-districts: the busiest area, Parkside sub-district (32 in all), changed by -38 871346.
• Evidenced: Change of vaccination sessions where density is high, from Period change: Vaccination sessions (week of 2026-06-29 to week of 2026-07-06) and Calculate density: Community health worker visits: the largest rise (+1, Canal Side sub-district) is where density averages 10.88 per square kilometer 921166.
• Evidenced: Hot cells of cases per health sub-district, from Hot spot analysis: Cases: 8 of 18 areas hold hot spot cells; Parkside sub-district holds the most (103) 9737.
• Evidenced: Nearest outbreak response team to each cluster of cases, from Find point clusters: cluster hulls: the 1 clusters lie 0 to 0 m from the nearest facility (Rapid response team 1 and others) 10379.
• Evidenced: Dense areas of community health worker visits, from Calculate density: Community health worker visits: the densest of five classes joins 491 cells into one area 11066.
• Evidenced: Change of vaccination sessions per health sub-district, from Period change: Vaccination sessions (week of 2026-06-29 to week of 2026-07-06): Canal Side sub-district rose the most (+1) and Fernhill sub-district fell the most (-2) 11311.
• Evidenced: Outbreak response teams within 250 m of each cluster of cases, from Find point clusters: cluster hulls: 1 of 1 clusters have one within 250 m 11879.
• Evidenced: Density of community health worker visits and cases together, from Calculate density: Community health worker visits and Calculate density: Cases: 4960 features from 2 sources 1236669.

Change beside the counts per health sub-district

What this analysis is. Spatial join: Attach attributes (or aggregates) of matching features from another layer, by location, distance, time and attributes 85. In ArcGIS it is Join Features or Spatial Join; in QGIS, native:joinattributesbylocation or native:joinbylocationsummary 85.

Why it was run on this data. It was run toward the goal "Describe the project Sample: Public health, Kampala: run the analyses that tell the most about its data, draw every result, and write a cited report" 86.

What it found. Spatial join on Summarize within: Health sub-districts, Period change: Cases (week of 2026-05-18 to week of 2026-05-25) produced Change beside the counts per health sub-district (18 features) 14878889. It reported result, count 18; result, elapsed ms 1; billing, cost 0.01074; billing, list price 0.01074; billing, upstream cost 0 87. The chart Change beside the counts per health sub-district is drawn from 6 rows 91.

What the map shows. The map draws Change beside the counts per health sub-district by count in 5 classes: 3 to 7, 7 to 11, 11 to 19, 19 to 27.5, 27.5 to 32 90. It draws 18 features 90. It covers longitude 32.515 to 32.648 and latitude 0.261 to 0.363, at about 1:61,580 90.

Findings

• Evidenced: Spatial join on Summarize within: Health sub-districts, Period change: Cases (week of 2026-05-18 to week of 2026-05-25) produced Change beside the counts per health sub-district (18 features) 14878889.
• Evidenced: It reported result, count 18; result, elapsed ms 1; billing, cost 0.01074; billing, list price 0.01074; billing, upstream cost 0 87.
• Evidenced: The chart Change beside the counts per health sub-district is drawn from 6 rows 91.

Tool gp_spatial_join; parameters {"join": {"layer": "Summarize within: Health sub-districts (9220b574)"}, "fields": ["count"], "target": {"layer": "Period change: Cases (week of 2026-05-18 to week of 2026-05-25) (66f934f7)"}, "project": "2612067c-badd-4310-9678-8f580d5e3d32", "predicate": "largest_overlap", "output_name": "Change beside the counts per health sub-district"}

Change of vaccination sessions where density is high

What this analysis is. Spatial join: Attach attributes (or aggregates) of matching features from another layer, by location, distance, time and attributes 85. In ArcGIS it is Join Features or Spatial Join; in QGIS, native:joinattributesbylocation or native:joinbylocationsummary 85.

Why it was run on this data. It was run toward the goal "Describe the project Sample: Public health, Kampala: run the analyses that tell the most about its data, draw every result, and write a cited report" 86.

What it found. The analysis utilized a spatial join 92, integrating data derived from Calculate density: Community health worker visits 93 and the specific period change for Vaccination sessions (week of 2026-06-

What the map shows. The map draws Change of vaccination sessions where density is high by join_count in 5 classes: 75 to 126, 126 to 151, 151 to 218, 218 to 292, 292 to 366 95. The map draws Health sub-districts with one symbol 95. It draws 28 features 95. It covers longitude 32.515 to 32.648 and latitude 0.261 to 0.363, at about 1:61,580 95.

Findings

• Evidenced: Spatial join on Calculate density: Community health worker visits, Period change: Vaccination sessions (week of 2026-06-29 to week of 2026-07-06) produced Change of vaccination sessions where density is high (10 features) 12929394.
• Evidenced: It reported result, count 10; result, elapsed ms 3; billing, cost 0.05926; billing, list price 0.05926; billing, upstream cost 0 92.
• Evidenced: The chart Change of vaccination sessions where density is high is drawn from 5 rows 96.

Tool gp_spatial_join; parameters {"join": {"layer": "Calculate density: Community health worker visits (3aeb4280)"}, "stats": [{"agg": "mean", "field": "density_per_km2"}], "target": {"layer": "Period change: Vaccination sessions (week of 2026-06-29 to week of 2026-07-06) (3345972d)"}, "project": "2612067c-badd-4310-9678-8f580d5e3d32", "predicate": "intersects", "output_name": "Change of vaccination sessions where density is high

Hot cells of cases per health sub-district

What this analysis is. Summarize within: Count and aggregate the features inside each polygon (or bin), proportionally for lines and polygons 42. In ArcGIS it is Summarize Within or Aggregate Points; in QGIS, native:countpointsinpolygon or native:sumlinelengths 42.

Why it was run on this data. Count Community health worker visits inside each feature of Health sub-districts by purpose; the counts map as a choropleth (natural breaks, sequential colors) 43.

What it found. Summarize within on Hot spot analysis: Cases, Health sub-districts produced Hot cells of cases per health sub-district (18 features) 16979899. It reported result, count 18; result, elapsed ms 38; billing, cost 0.05554; billing, list price 0.05554; billing, upstream cost 0 97. The chart Hot cells of cases per health sub-district is drawn from 6 rows 101.

What the map shows. The map draws Hot cells of cases per health sub-district by count in 5 classes: 73 to 96, 96 to 151, 151 to 215, 215 to 290, 290 to 358 100. It draws 18 features 100. It covers longitude 32.515 to 32.648 and latitude 0.261 to 0.363, at about 1:61,580 100.

Findings

• Evidenced: Summarize within on Hot spot analysis: Cases, Health sub-districts produced Hot cells of cases per health sub-district (18 features) 16979899.
• Evidenced: It reported result, count 18; result, elapsed ms 38; billing, cost 0.05554; billing, list price 0.05554; billing, upstream cost 0 97.
• Evidenced: The chart Hot cells of cases per health sub-district is drawn from 6 rows 101.

Tool gp_summarize_within; parameters {"project": "2612067c-badd-4310-9678-8f580d5e3d32", "features": {"layer": "Hot spot analysis: Cases (7dc9eb55)"}, "group_by": "gi_bin", "polygons": {"layer": "Health sub-districts (433b88cc)"}, "output_name": "Hot cells of cases per health sub-district"}

Nearest outbreak response team to each cluster of cases

What this analysis is. Find nearest: The k nearest features for each input, with geodesic distance and bearing 102. In ArcGIS it is Find Nearest or Near; in QGIS, native:joinbynearest or native:nearestneighbouranalysis 102.

Why it was run on this data. It was run toward the goal "Describe the project Sample: Public health, Kampala: run the analyses that tell the most about its data, draw every result, and write a cited report" 86.

What it found. The gp_find_nearest tool was executed using the Outbreak response teams input data 21. This analysis used the cluster hulls from 104 to determine the nearest outbreak response team to each cluster of cases 105. The operation generated one result, completing in 2 103 milliseconds and incurring a cost of 0.01014 103. A chart detailing the nearest outbreak response team to each cluster of cases was generated, containing 1 row 107.

What the map shows. The map draws Nearest outbreak response team to each cluster of cases with one symbol 106. The map draws Health sub-districts with one symbol 106. It draws 19 features 106. It covers longitude 32.585 to 32.606 and latitude 0.289 to 0.305, at about 1:10,087 106.

Findings

• Evidenced: Find nearest on Outbreak response teams, Find point clusters: cluster hulls produced Nearest outbreak response team to each cluster of cases (1 features) 21103104105.
• Evidenced: It reported result, count 1; result, elapsed ms 2; billing, cost 0.01014; billing, list price 0.01014; billing, upstream cost 0 103.
• Evidenced: The chart Nearest outbreak response team to each cluster of cases is drawn from 1 rows 107.

Tool gp_find_nearest; parameters {"k": 1, "to": {"layer": "Outbreak response teams (1b8f0e74)"}, "from": {"layer": "Find point clusters: cluster hulls (53d9a3e2)"}, "project": "2612067c-badd-4310-9678-8f580d5e3d32", "output_name": "Nearest outbreak response team to each cluster of cases"}

Dense areas of community health worker visits

What this analysis is. Pipeline: Run up to 20 GIS operations in one engine call; a step's params may use "$id" (an earlier step's result) or "$id.key", and layer references 108. Every step's map outputs are saved as styled layers 108.

Why it was run on this data. Health sub-districts holds population in 'population': Community health worker visits per 10,000 people in each feature compares features of different size; mapped as a choropleth (natural breaks, sequential colors) 109.

What it found. A pipeline of 2 steps (classify_field > dissolve) was executed, saving 1 layer 110.

What the map shows. Drawn with calculate density on the map Density: community health worker visits 67.

Findings

• Evidenced: Pipeline on Calculate density: Community health worker visits produced Dense areas of community health worker visits (5 features) 93110111.
• Evidenced: It reported result, count 5; result, elapsed ms 46; billing, cost 0.06905; billing, list price 0.06905; billing, upstream cost 0; list price 0.06905 110.
• Evidenced: The chart Dense areas of community health worker visits is drawn from 5 rows 112.

Tool gp_pipeline; parameters {"save": "final", "steps": [{"id": "classes", "params": {"field": "density_per_km2", "method": "quantile", "classes": 5, "features": {"layer": "Calculate density: Community health worker visits (3aeb4280)"}}, "operation": "classify_field"}, {"id": "areas", "params": {"by": ["class"], "features": "$classes"}, "operation": "dissolve"}], "project": "2612067c-badd-4310-9678-8f580d5e3d32", "output_name

Change of vaccination sessions per health sub-district

What this analysis is. Summarize within: Count and aggregate the features inside each polygon (or bin), proportionally for lines and polygons 42. In ArcGIS it is Summarize Within or Aggregate Points; in QGIS, native:countpointsinpolygon or native:sumlinelengths 42.

Why it was run on this data. Count Community health worker visits inside each feature of Health sub-districts by purpose; the counts map as a choropleth (natural breaks, sequential colors) 43.

What it found. The analysis utilized the input of 'Health sub-districts' 16 and a 'Period change: Vaccination sessions (week of 2026-06-29 to week of 2026-07-06)' 12 to generate the 'Change of vaccination sessions per health sub-district' 114 using the gp_summarize_within tool 113. This resulted in a count of 18 features for the layer 113. Furthermore, the chart displaying the change in vaccination sessions is drawn from 6 rows 116.

What the map shows. The map draws Change of vaccination sessions per health sub-district by sum_change in 2 classes: -2 to -1, -1 to 1 115. It draws 18 features 115. It covers longitude 32.515 to 32.648 and latitude 0.261 to 0.363, at about 1:61,580 115.

Findings

• Evidenced: Summarize within on Period change: Vaccination sessions (week of 2026-06-29 to week of 2026-07-06), Health sub-districts produced Change of vaccination sessions per health sub-district (18 features) 1216113114.
• Evidenced: It reported result, count 18; result, elapsed ms 1; billing, cost 0.01058; billing, list price 0.01058; billing, upstream cost 0 113.
• Evidenced: The chart Change of vaccination sessions per health sub-district is drawn from 6 rows 116.

Tool gp_summarize_within; parameters {"stats": [{"agg": "sum", "field": "change"}], "project": "2612067c-badd-4310-9678-8f580d5e3d32", "features": {"layer": "Period change: Vaccination sessions (week of 2026-06-29 to week of 2026-07-06) (3345972d)"}, "polygons": {"layer": "Health sub-districts (433b88cc)"}, "output_name": "Change of vaccination sessions per health sub-district"}

Outbreak response teams within 250 m of each cluster of cases

What this analysis is. Pipeline: Run up to 20 GIS operations in one engine call; a step's params may use "$id" (an earlier step's result) or "$id.key", and layer references 108. Every step's map outputs are saved as styled layers 108.

Why it was run on this data. Health sub-districts holds population in 'population': Outbreak response teams per 10,000 people in each feature compares features of different size; mapped as a choropleth (natural breaks, sequential colors) 117.

What it found. Pipeline on Find point clusters: cluster hulls, Outbreak response teams produced Outbreak response teams within 250 m of each cluster of cases (1 features) 21104118119. It reported result, count 1; result, elapsed ms 0; billing, cost 0.02015; billing, list price 0.02015; billing, upstream cost 0; list price 0.02015 118. The chart Outbreak response teams within 250 m of each cluster of cases is drawn from 1 rows 121.

What the map shows. The map draws Outbreak response teams within 250 m of each cluster of cases with one symbol 120. The map draws Health sub-districts with one symbol 120. It draws 19 features 120. It covers longitude 32.581 to 32.61 and latitude 0.286 to 0.308, at about 1:13,495 120.

Findings

• Evidenced: Pipeline on Find point clusters: cluster hulls, Outbreak response teams produced Outbreak response teams within 250 m of each cluster of cases (1 features) 21104118119.
• Evidenced: It reported result, count 1; result, elapsed ms 0; billing, cost 0.02015; billing, list price 0.02015; billing, upstream cost 0; list price 0.02015 118.
• Evidenced: The chart Outbreak response teams within 250 m of each cluster of cases is drawn from 1 rows 121.

Tool gp_pipeline; parameters {"save": "final", "steps": [{"id": "buffer", "params": {"features": {"layer": "Find point clusters: cluster hulls (53d9a3e2)"}, "distance_m": 250}, "operation": "buffer"}, {"id": "count", "params": {"features": {"layer": "Outbreak response teams (1b8f0e74)"}, "polygons": "$buffer"}, "operation": "summarize_within"}], "project": "2612067c-badd-4310-9678-8f580d5e3d32", "output_name": "Outbreak respo

Density of community health worker visits and cases together

What this analysis is. Merge layers: Append several layers into one 122. In ArcGIS it is Merge Layers or Merge; in QGIS, native:mergevectorlayers 122.

Why it was run on this data. It was run toward the goal "Describe the project Sample: Public health, Kampala: run the analyses that tell the most about its data, draw every result, and write a cited report" 86.

What it found. The gp_merge_layers tool combined inputs for Community health worker visits 93 and Cases 124 to create the merged layer 123.

Findings

• Evidenced: Merge layers on Calculate density: Community health worker visits, Calculate density: Cases produced Density of community health worker visits and cases together (4,960 features) 93123124125.
• Evidenced: It reported result, count 4,960; result, elapsed ms 109; billing, cost 0.01; billing, list price 0.01; billing, upstream cost 0 123.
• Evidenced: The chart Density of community health worker visits and cases together is drawn from 14 rows 126.

Tool gp_merge_layers; parameters {"layers": [{"layer": "Calculate density: Community health worker visits (3aeb4280)"}, {"layer": "Calculate density: Cases (b48a2b54)"}], "project": "2612067c-badd-4310-9678-8f580d5e3d32", "output_name": "Density of community health worker visits and cases together", "source_field": "source"}

Maps and charts

The run also made 1 map image(s) 127 and 19 chart(s) 128129130131132133134135136137138139140141142143144145146.

More maps

1 more map(s), beyond the 24 the report body holds, in the composer's order: Density of community health worker visits and cases together 147.

Documents

What the documents name. Epidemiological situation report (1 page, 1 page(s) from the text layer) names 1 organization, 4 event terms, 6 places and 1 quantity 148. Organizations named in Epidemiological situation report: Malaria (page 1, line 5) 149. Event terms in Epidemiological situation report: cholera (page 1, line 3, 2 mentions), outbreak (page 1, line 3, 2 mentions) 149. Places named in Epidemiological situation report: Parkside sub-district (page 1, line 3, 2 mentions), 029731, 3259548 (page 1, line 3), Rapid response team 1 (page 1, line 11), Eastfield sub-district (page 1, line 12) 149.

Findings

• Evidenced: Epidemiological situation report (1 page, 1 page(s) from the text layer) names 1 organization, 4 event terms, 6 places and 1 quantity.
• Evidenced: Organizations named in Epidemiological situation report: Malaria (page 1, line 5).
• Evidenced: Event terms in Epidemiological situation report: cholera (page 1, line 3, 2 mentions), outbreak (page 1, line 3, 2 mentions).
• Evidenced: Places named in Epidemiological situation report: Parkside sub-district (page 1, line 3, 2 mentions), 029731, 3259548 (page 1, line 3), Rapid response team 1 (page 1, line 11), Eastfield sub-district (page 1, line 12).

Changes since the last run

89 number(s) changed since the report of 2026-10-19T04:03:52Z.

New findings:

• Calculate density on Vaccination sessions: Calculate density: 1 layer saved: Calculate density: Vaccination sessions (616 features).
• Change beside the counts per health sub-district, from Period change: Cases (week of 2026-05-18 to week of 2026-05-25) and Summarize within: Health sub-districts: the busiest area, Parkside sub-district (32 in all), changed by -38 871346.
• Change of vaccination sessions per health sub-district, from Period change: Vaccination sessions (week of 2026-06-29 to week of 2026-07-06): Canal Side sub-district rose the most (+1) and Fernhill sub-district fell the most (-2) 11311.
• Change of vaccination sessions where density is high, from Period change: Vaccination sessions (week of 2026-06-29 to week of 2026-07-06) and Calculate density: Community health worker visits: the largest rise (+1, Canal Side sub-district) is where density averages 10.88 per square kilometer 921166.
• Dense areas of community health worker visits, from Calculate density: Community health worker visits: the densest of five classes joins 491 cells into one area 11066.
• Density of community health worker visits and cases together, from Calculate density: Community health worker visits and Calculate density: Cases: 4960 features from 2 sources 1236669.
• Hot spots on Vaccination sessions: Hot spot analysis: 1 layer saved: Hot spot analysis: Vaccination sessions (99 features).
• It reported result, bins 2,259; result, count 2,259; result, elapsed ms 75; result, cell size m 200; result, incidents used 2,290; result, distance band m 500.
• It reported result, bins 99; result, count 99; result, elapsed ms 5; result, cell size m 1,000; result, incidents used 200; result, distance band m 2,000.
• It reported result, count 1,000; result, noise 1,000; result, distance m 200; result, elapsed ms 576; result, min points 20; billing, cost 0.07.
• It reported result, count 10; result, elapsed ms 3; billing, cost 0.05926; billing, list price 0.05926; billing, upstream cost 0.
• It reported result, count 125; result, elapsed ms 26; result, resolution 8; result, features used 2,290; result, time neighbors 0; billing, cost 0.2145.
• It reported result, count 18; result, elapsed ms 0; billing, cost 0.01438; billing, list price 0.01438; billing, upstream cost 0.
• It reported result, count 18; result, elapsed ms 38; billing, cost 0.05554; billing, list price 0.05554; billing, upstream cost 0.
• It reported result, count 2,290; result, noise 1,961; result, distance m 200; result, elapsed ms 24; result, min points 20; billing, cost 0.0958.
• It reported result, count 2,452; result, elapsed ms 43; result, bandwidth m 500; result, cell size m 200; result, excluded null 0; result, max density per km2 34.43.
• It reported result, count 2,508; result, elapsed ms 49; result, bandwidth m 500; result, cell size m 200; result, excluded null 0; result, max density per km2 282.
• It reported result, count 24; result, elapsed ms 5; result, resolution 7; result, features used 200; result, time neighbors 0; billing, cost 0.11.
• It reported result, count 4,960; result, elapsed ms 109; billing, cost 0.01; billing, list price 0.01; billing, upstream cost 0.
• It reported result, count 5; result, elapsed ms 46; billing, cost 0.06905; billing, list price 0.06905; billing, upstream cost 0; list price 0.06905.
• It reported result, count 616; result, elapsed ms 105; result, bandwidth m 2,000; result, cell size m 500; result, excluded null 0; result, max density per km2 5.525.
• Merge layers on Calculate density: Community health worker visits, Calculate density: Cases produced Density of community health worker visits and cases together (4,960 features).
• Outbreak response teams within 250 m of each cluster of cases, from Find point clusters: cluster hulls: 1 of 1 clusters have one within 250 m 11879.
• Pipeline on Calculate density: Community health worker visits produced Dense areas of community health worker visits (5 features).
• Spatial join on Calculate density: Community health worker visits, Period change: Vaccination sessions (week of 2026-06-29 to week of 2026-07-06) produced Change of vaccination sessions where density is high (10 features).
• Summarize within on Vaccination sessions, Health sub-districts: Summarize within: 1 layer saved: Summarize within: Health sub-districts (18 features).
• The chart Dense areas of community health worker visits is drawn from 5 rows.
• The chart Density of community health worker visits and cases together is drawn from 14 rows.

Sources

1. 1Rendering: Chart: Timeline: events per week by dataset. 265 rows id 9ed316de-c13f-4c96-9941-d0ab254813c6
2. 2Dataset: Cases. vector, 2290 rows id 59cb2ed5-709d-4ca6-94aa-fd8eb621eb37
3. 3Dataset: Community health worker visits. vector, 1000 rows id 44c21bd2-0729-457b-ba05-d7a36b9885ed
4. 4Dataset: Vaccination sessions. vector, 200 rows id 46c023c6-ea3b-4a1d-9a6c-130a334b6790
5. 5Dataset: Weekly cases by sub-district. table, 936 rows id 1ce22bbc-4659-446b-a2ab-9d6f1bfe6418
6. 6Dataset: Clinic stock by month. table, 315 rows id 930b3eee-3ebe-4aff-8fa0-a2c38a029a36
7. 7Rendering: Chart: Timeline lanes: events per week by dataset. 265 rows id dc67edf4-bf01-4d54-a054-9cccf4719256
8. 8Rendering: Chart: Timeline: change from one week to the next. 52 rows id b08e07a6-2857-423d-9faf-2edf4b6db093
9. 9Tool result: gp_period_change, step 0. The count fell from 36 in week of 2026-09-21 to 18 in week of 2026-09-28 (-50.0%); a change this large is unlikely to be chance (p = 0.0198). id da0d02de-5858-490e-a2a5-d4daf11b4a77:0
10. 10Layer: Period change: Community health worker visits (week of 2026-09-21 to week of 2026-09-28) id 3bec1b01-80c0-4802-8f6f-e05ccd607485
11. 11Tool result: gp_period_change, step 1. The count fell from 9 in week of 2026-06-29 to 2 in week of 2026-07-06 (-77.8%); a change this small is within what chance would give (p = 0.0654). id da0d02de-5858-490e-a2a5-d4daf11b4a77:1
12. 12Layer: Period change: Vaccination sessions (week of 2026-06-29 to week of 2026-07-06) id 3345972d-58bd-4b02-b103-633f6c7a1c21
13. 13Tool result: gp_period_change, step 2. The count fell from 136 in week of 2026-05-18 to 86 in week of 2026-05-25 (-36.8%); a change this large is unlikely to be chance (p = 0.000963). id da0d02de-5858-490e-a2a5-d4daf11b4a77:2
14. 14Layer: Period change: Cases (week of 2026-05-18 to week of 2026-05-25) id 66f934f7-4d7f-4b71-bfe5-db9cf372b64d
15. 15Rendering: Map: Overview: the top 3 results. view longitude 32.48887 to 32.67295, latitude 0.23939 to 0.37643 id 56c16576-ebcf-441b-9376-fc19268f64e4
16. 16Layer: Health sub-districts. spatial profile id 433b88cc-cf57-4d48-b158-220cc8b77697
17. 17Layer: Cases. spatial profile id 03c8f72b-99c1-4176-8f56-aabff1d062b7
18. 18Layer: Clinics. spatial profile id ac5b28b9-59c1-40e0-b7bd-1b0014ea13e0
19. 19Layer: Water points. spatial profile id c63bde31-7c18-420e-894d-3e824a409c2d
20. 20Layer: Community health worker visits. spatial profile id 4d70d1ce-8162-42eb-b3d7-d864edf2721e
21. 21Layer: Outbreak response teams. spatial profile id 1b8f0e74-8ce3-4ca6-bbf8-e76348e17d6e
22. 22Layer: Vaccination sessions. spatial profile id 6798c97c-ff1d-462f-b9b0-5baab95708a5
23. 23Catalog: Emerging hot spots (operation catalog, gp_space_time_hot_spots). Where hot spots are new, consecutive, intensifying, persistent, diminishing, sporadic, oscillating or historical over time. ArcGIS: Emerging Hot Spot Analysis, Create Space Time Cube. id gp_space_time_hot_spots
24. 24Plan: Space time hot spots of Community health worker visits (the Start plan's candidate c_70590e6d2eeb). Community health worker visits carry dates in 'visited_at': emerging hot spot analysis finds new, intensifying and fading hot spots per week. id 502cfe77-e27d-4781-888d-59ecc8241e24:c_70590e6d2eeb
25. 25Tool result: gp_space_time_hot_spots, step 0. Emerging hot spots: 1 layer saved: Emerging hot spots: Community health worker visits (124 features). id ca57d695-3247-4efe-bf1e-0e4a458c66f4:0
26. 26Rendering: Map: Emerging hot spots: community health worker visits and cases. view longitude 32.49952 to 32.66088, latitude 0.25211 to 0.37225 id 7e95523d-3393-400e-8c04-5e9bb1751ad0
27. 27Plan: Space time hot spots of Vaccination sessions (the Start plan's candidate c_136a5b017678). Vaccination sessions carry dates in 'held_on': emerging hot spot analysis finds new, intensifying and fading hot spots per week. Measured: the pattern of Vaccination sessions changes over time (earlier half random, later half dispersed; the center moved 259 m). id 502cfe77-e27d-4781-888d-59ecc8241e24:c_136a5b017678
28. 28Tool result: gp_space_time_hot_spots, step 1. Emerging hot spots: 1 layer saved: Emerging hot spots: Vaccination sessions (24 features). id ca57d695-3247-4efe-bf1e-0e4a458c66f4:1
29. 29Rendering: Map: Emerging hot spots: vaccination sessions. view longitude 32.48887 to 32.67295, latitude 0.23939 to 0.37643 id 56aebece-6125-4fb6-b449-86d97ba94702
30. 30Plan: Space time hot spots of Cases (the Start plan's candidate c_24319662e98f). Cases carry dates in 'onset_date': emerging hot spot analysis finds new, intensifying and fading hot spots per week. Measured: the pattern of Cases changes over time (earlier half random, later half clustered; the center moved 783 m). id 502cfe77-e27d-4781-888d-59ecc8241e24:c_24319662e98f
31. 31Tool result: gp_space_time_hot_spots, step 2. Emerging hot spots: 1 layer saved: Emerging hot spots: Cases (125 features). id ca57d695-3247-4efe-bf1e-0e4a458c66f4:2
32. 32Catalog: Hot spot analysis (operation catalog, gp_hot_spots). Getis-Ord Gi* hot and cold spots with z-scores, p-values and confidence bins. ArcGIS: Find Hot Spots, Hot Spot Analysis (Getis-Ord Gi*). QGIS: qgis:hotspot (plugin). id gp_hot_spots
33. 33Plan: Hot spots of Community health worker visits (the Start plan's candidate c_6211f60c8ad6). 1000 points: Getis-Ord Gi* tests whether the concentrations in Community health worker visits are statistically significant. Measured: Community health worker visits are clustered (nearest-neighbor ratio 0.95, z = -3.3), median spacing 124 m. Cells of 500 m and a band of 1000 m from the measured spacing. id 502cfe77-e27d-4781-888d-59ecc8241e24:c_6211f60c8ad6
34. 34Tool result: gp_hot_spots, step 0. Hot spot analysis: 1 layer saved: Hot spot analysis: Community health worker visits (390 features). id c2d5c745-9aa1-4cd9-a0a8-2ff3ab77a62a:0
35. 35Rendering: Map: Hot spots: community health worker visits. view longitude 32.50994 to 32.65238, latitude 0.25713 to 0.36631 id 98052aef-9d4d-4f54-ba5f-c8b6a1168787
36. 36Plan: Hot spots of Cases (the Start plan's candidate c_9fb20ef66d1e). 2290 points: Getis-Ord Gi* tests whether the concentrations in Cases are statistically significant. Measured: Cases are clustered (nearest-neighbor ratio 0.90, z = -9.4), median spacing 75.6 m. Cells of 200 m and a band of 500 m from the measured spacing. id 502cfe77-e27d-4781-888d-59ecc8241e24:c_9fb20ef66d1e
37. 37Tool result: gp_hot_spots, step 1. Hot spot analysis: 1 layer saved: Hot spot analysis: Cases (2259 features). id c2d5c745-9aa1-4cd9-a0a8-2ff3ab77a62a:1
38. 38Rendering: Map: Hot spots: cases. view longitude 32.5125 to 32.64913, latitude 0.25979 to 0.36453 id a56eeb41-2edc-432a-88c7-a24c0f5a72b3
39. 39Plan: Hot spots of Vaccination sessions (the Start plan's candidate c_e1e27b62d2c9). 200 points: Getis-Ord Gi* tests whether the concentrations in Vaccination sessions are statistically significant. id 502cfe77-e27d-4781-888d-59ecc8241e24:c_e1e27b62d2c9
40. 40Tool result: gp_hot_spots, step 2. Hot spot analysis: 1 layer saved: Hot spot analysis: Vaccination sessions (99 features). id c2d5c745-9aa1-4cd9-a0a8-2ff3ab77a62a:2
41. 41Rendering: Map: Hot spots: vaccination sessions. view longitude 32.49938 to 32.65852, latitude 0.25127 to 0.37325 id 68ed8d3d-fda0-438d-87ed-17c5d6d517d1
42. 42Catalog: Summarize within (operation catalog, gp_summarize_within). Count and aggregate the features inside each polygon (or bin), proportionally for lines and polygons. ArcGIS: Summarize Within, Aggregate Points. QGIS: native:countpointsinpolygon, native:sumlinelengths. id gp_summarize_within
43. 43Plan: Community health worker visits per feature of Health sub-districts (the Start plan's candidate c_f0b17dbbc715). Count Community health worker visits inside each feature of Health sub-districts by purpose; the counts map as a choropleth (natural breaks, sequential colors). id 502cfe77-e27d-4781-888d-59ecc8241e24:c_f0b17dbbc715
44. 44Tool result: gp_summarize_within, step 0. Summarize within: 1 layer saved: Summarize within: Health sub-districts (18 features). id a98160db-9f77-4a98-acfd-a879b8f97de9:0
45. 45Rendering: Map: Counts: community health worker visits, by health sub-district. view longitude 32.51464 to 32.64754, latitude 0.26149 to 0.36336 id 3b9b93aa-6cb9-420c-beb3-f36bcbea491e
46. 46Tool result: gp_summarize_within, step 1. Summarize within: 1 layer saved: Summarize within: Health sub-districts (18 features). id a98160db-9f77-4a98-acfd-a879b8f97de9:1
47. 47Rendering: Map: Counts: vaccination sessions, by health sub-district. view longitude 32.51464 to 32.64754, latitude 0.26149 to 0.36336 id 72daafcb-45d3-43a7-bc54-d22b2cc6e843
48. 48Tool result: gp_summarize_within, step 2. Summarize within: 1 layer saved: Summarize within: Health sub-districts (18 features). id a98160db-9f77-4a98-acfd-a879b8f97de9:2
49. 49Rendering: Map: Counts: cases, by health sub-district. view longitude 32.51464 to 32.64754, latitude 0.26149 to 0.36336 id 2135f78f-d2ac-433f-9d8a-6e3d45ba6e33
50. 50Catalog: Period change (analysis catalog, gp_period_change). Counts or sums per period (week over week, month over month, any two periods, or before and after a date), per area, per H3 cell or in total, with the places of largest rise and fall and a significance note. ArcGIS: Summarize Percent Change (Crime Analysis), Change Point Detection. id gp_period_change
51. 51Plan: Community health worker visits: change per week per feature of Health sub-districts (the Start plan's candidate c_a2fcbe1fcdab). Community health worker visits carry dates in 'visited_at' over 26 weeks: the count per week, week over week, with the places of largest rise and fall per feature of Health sub-districts and whether the change is beyond chance. id 502cfe77-e27d-4781-888d-59ecc8241e24:c_a2fcbe1fcdab
52. 52Rendering: Map: Period change: community health worker visits, by health sub-district. view longitude 32.51235 to 32.64994, latitude 0.26121 to 0.36364 id 1ba857b6-25f9-4a1e-9809-7dabd1dba73f
53. 53Rendering: Chart: Community health worker visits: count per week. 27 rows id 38a7f1b5-f586-4a93-acb8-4706829d7084
54. 54Rendering: Chart: Community health worker visits: week of 2026-09-21 to week of 2026-09-28 by place. 28 rows id cc02c36d-d7ba-4b30-b17f-b4d59c59a182
55. 55Rendering: Chart: Period change: Community health worker visits (week of 2026-09-21 to week of 2026-09-28). 5 rows id 5a1e5aac-b54b-49dd-82a1-6de54c6f280a
56. 56Rendering: Map: Period change: vaccination sessions, by health sub-district. view longitude 32.51235 to 32.64994, latitude 0.26121 to 0.36364 id 4021c7ee-ffb3-49da-9404-41221d96a7a2
57. 57Rendering: Chart: Vaccination sessions: count per week. 39 rows id f56fc17e-709a-4ed7-a297-d2fd6e33d067
58. 58Rendering: Chart: Vaccination sessions: week of 2026-06-29 to week of 2026-07-06 by place. 20 rows id 293fe134-01c3-4c24-b991-e34d182657c7
59. 59Rendering: Chart: Period change: Vaccination sessions (week of 2026-06-29 to week of 2026-07-06). 5 rows id a076831a-6ae4-47aa-8be1-9a5d36f509fa
60. 60Rendering: Map: Period change: cases, by health sub-district. view longitude 32.51235 to 32.64994, latitude 0.26121 to 0.36364 id c02c2da5-7128-4cbe-bc4c-1af8118de764
61. 61Rendering: Chart: Cases: count per week. 53 rows id a232e499-3702-4315-b7d3-da9d00e71897
62. 62Rendering: Chart: Cases: week of 2026-05-18 to week of 2026-05-25 by place. 30 rows id c86a0c4a-b1af-4354-b292-8496363ee353
63. 63Rendering: Chart: Period change: Cases (week of 2026-05-18 to week of 2026-05-25). 6 rows id 916d1401-e116-42e0-b939-a730c5fff396
64. 64Catalog: Calculate density (operation catalog, gp_calculate_density). Kernel density per square kilometer on a grid. ArcGIS: Calculate Density, Kernel Density. QGIS: native:heatmapkerneldensityestimation. id gp_calculate_density
65. 65Plan: Kernel density of Community health worker visits (the Start plan's candidate c_224143430356). 1000 points in Community health worker visits: a density surface shows where they concentrate. Measured: Community health worker visits are clustered (nearest-neighbor ratio 0.95, z = -3.3), median spacing 124 m. id 502cfe77-e27d-4781-888d-59ecc8241e24:c_224143430356
66. 66Tool result: gp_calculate_density, step 0. Calculate density: 1 layer saved: Calculate density: Community health worker visits (2452 features). id 12155ea1-761a-4dfb-bea9-c367f910d998:0
67. 67Rendering: Map: Density: community health worker visits. view longitude 32.50938 to 32.65381, latitude 0.25717 to 0.36787 id c4802079-26fd-4077-ab26-93b7925b6482
68. 68Plan: Kernel density of Cases (the Start plan's candidate c_070749f5aead). 2290 points in Cases: a density surface shows where they concentrate. Measured: Cases are clustered (nearest-neighbor ratio 0.90, z = -9.4), median spacing 75.6 m. id 502cfe77-e27d-4781-888d-59ecc8241e24:c_070749f5aead
69. 69Tool result: gp_calculate_density, step 1. Calculate density: 1 layer saved: Calculate density: Cases (2508 features). id 12155ea1-761a-4dfb-bea9-c367f910d998:1
70. 70Rendering: Map: Density: cases. view longitude 32.50843 to 32.65285, latitude 0.25669 to 0.3674 id 06af517c-f53b-4797-a57c-3a39a0f27a0d
71. 71Plan: Kernel density of Vaccination sessions (the Start plan's candidate c_e0476445813b). 200 points in Vaccination sessions: a density surface shows where they concentrate. id 502cfe77-e27d-4781-888d-59ecc8241e24:c_e0476445813b
72. 72Tool result: gp_calculate_density, step 2. Calculate density: 1 layer saved: Calculate density: Vaccination sessions (616 features). id 12155ea1-761a-4dfb-bea9-c367f910d998:2
73. 73Rendering: Map: Density: vaccination sessions. view longitude 32.49029 to 32.67437, latitude 0.24081 to 0.3819 id 8e8a277d-768e-4e3d-b59e-c64d53beeb67
74. 74Catalog: Find point clusters (operation catalog, gp_find_point_clusters). Density-based clusters (DBSCAN, HDBSCAN or OPTICS, optionally space-time) with hulls and noise. ArcGIS: Find Point Clusters, Density-based Clustering. QGIS: native:dbscanclustering, native:stdbscanclustering. id gp_find_point_clusters
75. 75Plan: Cluster points of Community health worker visits (the Start plan's candidate c_b85a20a2f1a4). DBSCAN finds dense groups of Community health worker visits and separates noise. Measured: Community health worker visits are clustered (nearest-neighbor ratio 0.95, z = -3.3), median spacing 124 m. DBSCAN at 200 m (twice the measured spacing). id 502cfe77-e27d-4781-888d-59ecc8241e24:c_b85a20a2f1a4
76. 76Tool result: gp_find_point_clusters, step 0. Find point clusters: 1 layer saved: Find point clusters: Community health worker visits (1000 features). id 77bbbd8a-81ae-4bc9-b0c3-23aaf5f1fa0f:0
77. 77Rendering: Map: Clusters: community health worker visits. view longitude 32.51641 to 32.64657, latitude 0.26283 to 0.3626 id 8bb4310d-659d-4524-a471-aaae6fdb36c0
78. 78Plan: Cluster points of Cases (the Start plan's candidate c_c6803ff2f7da). DBSCAN finds dense groups of Cases and separates noise. Measured: Cases are clustered (nearest-neighbor ratio 0.90, z = -9.4), median spacing 75.6 m. DBSCAN at 200 m (twice the measured spacing). id 502cfe77-e27d-4781-888d-59ecc8241e24:c_c6803ff2f7da
79. 79Tool result: gp_find_point_clusters, step 1. Find point clusters: 2 layers saved: Find point clusters: Cases (2290 features); Find point clusters: cluster hulls (1 feature). id 77bbbd8a-81ae-4bc9-b0c3-23aaf5f1fa0f:1
80. 80Rendering: Map: Clusters: cases. view longitude 32.51557 to 32.64664, latitude 0.26229 to 0.36275 id 48c33ecb-0323-4a65-a670-4a31e60946af
81. 81Catalog: Find outliers (operation catalog, gp_outliers). Anselin local Moran’s I: HH, LL clusters and HL, LH outliers. ArcGIS: Find Outliers, Cluster and Outlier Analysis (Anselin Local Moran’s I). id gp_outliers
82. 82Plan: Local morans i of Health sub-districts (the Start plan's candidate c_39aa394b7122). Local Moran's I maps clusters and outliers of 'population'. Measured: Health sub-districts have 39 neighboring pairs (4.33 neighbors per area, 0 isolated), so neighbors can be compared. id 502cfe77-e27d-4781-888d-59ecc8241e24:c_39aa394b7122
83. 83Tool result: gp_outliers, step 0. Find outliers: 1 layer saved: Find outliers: Health sub-districts (18 features). id 97423338-c177-4702-bebb-f1beb57bfaff:0
84. 84Rendering: Map: Outliers: health sub-districts. view longitude 32.51464 to 32.64754, latitude 0.26149 to 0.36336 id b07c4b22-dbfb-4024-b401-a4a40ee5d0c0
85. 85Catalog: Spatial join (operation catalog, gp_spatial_join). Attach attributes (or aggregates) of matching features from another layer, by location, distance, time and attributes. ArcGIS: Join Features, Spatial Join. QGIS: native:joinattributesbylocation, native:joinbylocationsummary, native:joinbynearest. id gp_spatial_join
86. 86Plan: The run's goal. Describe the project Sample: Public health, Kampala: run the analyses that tell the most about its data, draw every result, and write a cited report. id 12778c77-4c3f-4379-9aca-c9e85676282e:goal
87. 87Tool result: gp_spatial_join, step 0. Spatial join: 1 layer saved: Change beside the counts per health sub-district (18 features). id a1498fe0-e6b0-47ea-b178-eea9d9489b4e:0
88. 88Layer: Summarize within: Health sub-districts. 18 features id 9220b574-337c-44ac-8245-9bddea6ca9e7
89. 89Layer: Change beside the counts per health sub-district. 18 features id 329bcab2-5a7e-47ff-9e3a-70ab2bfb87e9
90. 90Rendering: Map: Change beside the counts per health sub-district. view longitude 32.51464 to 32.64754, latitude 0.26149 to 0.36336 id 1a975715-964d-4cea-a9d4-e1f36f7402f9
91. 91Rendering: Chart: Change beside the counts per health sub-district. 6 rows id d8232fd5-0620-44d0-a014-35f7ff7c1ccb
92. 92Tool result: gp_spatial_join, step 0. Spatial join: 1 layer saved: Change of vaccination sessions where density is high (10 features). id dcf7086a-4b87-4442-bf12-c3721d1867e7:0
93. 93Layer: Calculate density: Community health worker visits. 2,452 features id 3aeb4280-7dc5-445b-bbe5-0f612cf31e62
94. 94Layer: Change of vaccination sessions where density is high. 10 features id 9cf2f4e1-0c98-47a0-8eda-017184bee6f6
95. 95Rendering: Map: Change of vaccination sessions where density is high. view longitude 32.51464 to 32.64754, latitude 0.26149 to 0.36336 id 2e0eef15-a319-4b25-8746-f0ab62d6c9ad
96. 96Rendering: Chart: Change of vaccination sessions where density is high. 5 rows id 339db613-2509-4201-bb54-b9cf3018c42e
97. 97Tool result: gp_summarize_within, step 0. Summarize within: 1 layer saved: Hot cells of cases per health sub-district (18 features). id cbaa2d66-7e86-4062-a268-8ceebb34226f:0
98. 98Layer: Hot spot analysis: Cases. 2,259 features id 7dc9eb55-076a-4811-89ea-633c6eba1ad4
99. 99Layer: Hot cells of cases per health sub-district. 18 features id c90fa521-67fb-47d1-b4be-789822f3c247
100. 100Rendering: Map: Hot cells of cases per health sub-district. view longitude 32.51464 to 32.64754, latitude 0.26149 to 0.36336 id c1fd9863-16f7-454b-b2e4-46156c47d93d
101. 101Rendering: Chart: Hot cells of cases per health sub-district. 6 rows id d428d2a4-23ca-4584-9e0e-18e6da6305ec
102. 102Catalog: Find nearest (operation catalog, gp_find_nearest). The k nearest features for each input, with geodesic distance and bearing. ArcGIS: Find Nearest, Near, Generate Near Table. QGIS: native:joinbynearest, native:nearestneighbouranalysis. id gp_find_nearest
103. 103Tool result: gp_find_nearest, step 0. Find nearest: 1 layer saved: Nearest outbreak response team to each cluster of cases (1 feature). id bf6d2133-da41-4a20-84bf-87e492f1da5b:0
104. 104Layer: Find point clusters: cluster hulls. 1 features id 53d9a3e2-2347-4fc0-bb22-fd75844c6cc8
105. 105Layer: Nearest outbreak response team to each cluster of cases. 1 features id b7c1c4f9-ecec-44aa-bf5a-a330316e549f
106. 106Rendering: Map: Nearest outbreak response team to each cluster of cases. view longitude 32.5846 to 32.60637, latitude 0.28861 to 0.3053 id 561b2c9d-1c47-4350-a8f1-2ace1078c213
107. 107Rendering: Chart: Nearest outbreak response team to each cluster of cases. 1 rows id 3762f5c3-d6af-4f5c-847a-dcb0fcf4fbe1
108. 108Catalog: Pipeline (tool description, gp_pipeline). Run up to 20 GIS operations in one engine call; a step's params may use "$id" (an earlier step's result) or "$id.key", and layer references. Every step's map outputs are saved as styled layers. Never calls Mapbox. id gp_pipeline
109. 109Plan: Community health worker visits per 10,000 people per feature of Health sub-districts (the Start plan's candidate c_9273635ebf24). Health sub-districts holds population in 'population': Community health worker visits per 10,000 people in each feature compares features of different size; mapped as a choropleth (natural breaks, sequential colors). id 502cfe77-e27d-4781-888d-59ecc8241e24:c_9273635ebf24
110. 110Tool result: gp_pipeline, step 0. Pipeline of 2 steps (classify_field > dissolve): 1 layer saved. id 228e478a-4ba9-4d74-976e-e45efd5229c7:0
111. 111Layer: Dense areas of community health worker visits. 5 features id 3c9ccc18-20ed-47f2-ac66-234751a0679d
112. 112Rendering: Chart: Dense areas of community health worker visits. 5 rows id 42cf1971-6f22-441e-aa1a-a1326ec273d8
113. 113Tool result: gp_summarize_within, step 0. Summarize within: 1 layer saved: Change of vaccination sessions per health sub-district (18 features). id 17676f2a-994e-4660-9745-0a3db803f8a8:0
114. 114Layer: Change of vaccination sessions per health sub-district. 18 features id f4ad8c88-5019-4a59-a2a0-054c153f32be
115. 115Rendering: Map: Change of vaccination sessions per health sub-district. view longitude 32.51464 to 32.64754, latitude 0.26149 to 0.36336 id 5b65b748-4632-4299-9ce1-4f8eb9a0301e
116. 116Rendering: Chart: Change of vaccination sessions per health sub-district. 6 rows id 7d5f7fc8-f8d9-41e4-be37-e2cea1f883b7
117. 117Plan: Outbreak response teams per 10,000 people per feature of Health sub-districts (the Start plan's candidate c_9a76ebe60c76). Health sub-districts holds population in 'population': Outbreak response teams per 10,000 people in each feature compares features of different size; mapped as a choropleth (natural breaks, sequential colors). id 502cfe77-e27d-4781-888d-59ecc8241e24:c_9a76ebe60c76
118. 118Tool result: gp_pipeline, step 0. Pipeline of 2 steps (buffer > summarize_within): 1 layer saved. id 8ab3e530-5833-406a-81af-1fb4eae27c2f:0
119. 119Layer: Outbreak response teams within 250 m of each cluster of cases. 1 features id b7e12b85-5a4c-4d51-b141-8bb6248100da
120. 120Rendering: Map: Outbreak response teams within 250 m of each cluster of cases. view longitude 32.58093 to 32.61005, latitude 0.28573 to 0.30805 id 2a7e6b94-090d-4ceb-9543-bf0e2eaad33c
121. 121Rendering: Chart: Outbreak response teams within 250 m of each cluster of cases. 1 rows id 8d19bda1-77d2-4fd7-9777-e6693a0dd643
122. 122Catalog: Merge layers (operation catalog, gp_merge_layers). Append several layers into one. ArcGIS: Merge Layers, Merge. QGIS: native:mergevectorlayers. id gp_merge_layers
123. 123Tool result: gp_merge_layers, step 0. Merge layers: 1 layer saved: Density of community health worker visits and cases together (4960 features). id 1e12b18a-4e3f-4064-82dd-a2dfd4e27abc:0
124. 124Layer: Calculate density: Cases. 2,508 features id b48a2b54-e6f8-464b-bc6d-b3f70448006d
125. 125Layer: Density of community health worker visits and cases together. 4,960 features id 39d58f96-c415-42fe-bd1d-212d8c37be6e
126. 126Rendering: Chart: Density of community health worker visits and cases together. 14 rows id 2adbb49c-d53c-48e7-a48f-ad1f6f209cf5
127. 127Rendering: Map: Health sub-districts by population id 5ef70e51-0bfd-4073-a856-905f5169fec3
128. 128Rendering: Chart: Cases over time (onset_date). 159 rows id 2767bc98-56c4-42c3-b1d6-bd12a2987bec
129. 129Rendering: Chart: Clinic stock by month over time (month). 36 rows id f8cfe58c-5392-477f-95a8-16f1c3223752
130. 130Rendering: Chart: Emerging hot spots: Community health worker visits. 8 rows id d847a912-2ac6-45dd-9ff7-e69e54f5091b
131. 131Rendering: Chart: Emerging hot spots: Vaccination sessions. 6 rows id e41e6ccc-7233-4910-b880-fa8e491bc99b
132. 132Rendering: Chart: Emerging hot spots: Cases. 8 rows id 8d0dd7bd-fa8d-46d4-a605-510c07477368
133. 133Rendering: Chart: Hot spot analysis: Community health worker visits. 21 rows id 2f9efce8-7875-4dac-b005-38853d079029
134. 134Rendering: Chart: Hot spot analysis: Cases. 53 rows id fa55dd0e-666e-4751-b2b2-9e1f0b859733
135. 135Rendering: Chart: Hot spot analysis: Vaccination sessions. 14 rows id b1e62dd9-e001-4d7a-b5e6-ab1dc1f840cc
136. 136Rendering: Chart: Summarize within: Health sub-districts. 6 rows id 9f7e951b-937a-4c91-a27e-78f52dcc8881
137. 137Rendering: Chart: Summarize within: Health sub-districts. 6 rows id c229b3e5-c6e5-4952-a2f7-694aa2209b3f
138. 138Rendering: Chart: Summarize within: Health sub-districts. 6 rows id 5e8cdc54-af5b-4a64-8630-d8cc680b09a0
139. 139Rendering: Chart: Calculate density: Community health worker visits. 13 rows id fde12f1f-a5aa-4aeb-bf82-9263e56dca6f
140. 140Rendering: Chart: Calculate density: Cases. 13 rows id d3819a7e-df6e-410c-a08e-5dc99421b7b4
141. 141Rendering: Chart: Calculate density: Vaccination sessions. 11 rows id e67f7ebe-6edb-44cb-992b-568211369e6d
142. 142Rendering: Chart: Find point clusters: Cases. 2 rows id 9446f731-3ec5-4204-a55f-680b73096a9b
143. 143Rendering: Chart: Find point clusters: cluster hulls. 1 rows id 47f73bfc-cea9-49b5-a8c8-0daf229197af
144. 144Rendering: Chart: Find outliers: Health sub-districts. 18 rows id 80bc5037-fe96-47f6-9f2d-09b7e0b01cee
145. 145Rendering: Chart: Find point clusters: Cases over time. 106 rows id eda4bec9-eaa5-497f-855f-0e9893628d6b
146. 146Rendering: Chart: Find point clusters: Community health worker visits over time. 108 rows id b48c3ed5-295d-4fff-a7f2-8068c4ce4536
147. 147Rendering: Map: Density of community health worker visits and cases together. view longitude 32.50837 to 32.6538, latitude 0.25641 to 0.36789 id 88c9d95c-3e1b-4a51-8f51-669e638423ad
148. 148Document: Epidemiological situation report. 1 page(s), 1 page(s) from the text layer id cb6fd9c4-f91e-4a89-a40d-68bf28554744
149. 149Document page: Epidemiological situation report, page 1. Malaria cases rose after both rainy seasons, three weeks after the rain peaks. id cb6fd9c4-f91e-4a89-a40d-68bf28554744:1
150. 150Dataset: Health sub-districts. vector, 18 rows id 9bb4e797-2cc2-4974-b5ec-02ddc0a96670
151. 151Dataset: Clinics. vector, 35 rows id 668e992e-3b51-4493-9704-d33893bb3dfc
152. 152Dataset: Water points. vector, 150 rows id 899479cd-15fb-482f-8652-6656bfa10bfb
153. 153Dataset: Outbreak response teams. vector, 6 rows id af26ec39-903c-46e8-8314-2dea85af0098
154. 154Dataset: Epidemiological situation report. document, ? rows id 33a668e1-f7f4-454b-ac70-576cd982e98f
155. 155Record: Sample: Public health, Kampala id 2612067c-badd-4310-9678-8f580d5e3d32
156. 156Layer: Emerging hot spots: Community health worker visits. 124 features id 32cd91ad-60ea-4d7e-a96f-48417549e7e1
157. 157Layer: Emerging hot spots: Vaccination sessions. 24 features id 99a03e93-7113-48b6-b8ad-aa357667356c
158. 158Layer: Emerging hot spots: Cases. 125 features id fd743779-7b6c-495f-9c0d-b1ac9a8dafa4
159. 159Layer: Hot spot analysis: Community health worker visits. 390 features id e0bdc602-6847-43ba-83b2-d2832f74b413
160. 160Layer: Hot spot analysis: Vaccination sessions. 99 features id dc4def90-dea3-4ca0-88dd-5778b14b932b
161. 161Layer: Summarize within: Health sub-districts. 18 features id def8ac79-eb53-495e-af55-828f67a5b5a2
162. 162Layer: Summarize within: Health sub-districts. 18 features id f4259241-f5e4-42fa-9481-0cf7434aa6d5
163. 163Layer: Calculate density: Vaccination sessions. 616 features id c49f160e-369f-44fa-b91b-296dc8ea18ee
164. 164Layer: Find point clusters: Community health worker visits. 1,000 features id 97d77c9e-c58d-4c5a-af88-3874d12ef032
165. 165Layer: Find point clusters: Cases. 2,290 features id f97a47bc-a875-42e7-891d-ba3d7809245f
166. 166Layer: Find outliers: Health sub-districts. 18 features id 2d5379dc-e698-4b04-b2db-6504a49d353d