Central District
24 cases · 1 death
5.33× immediate baseline
Public Health Analytics & Decision Intelligence
Disease Surveillance & Outbreak Intelligence
Subnational epidemiological analysis for early detection and public-health decision-making.
Simulated data · Fictionalized geographyEpidemiological analysis · Decision support
01 / The question
Can routine data reveal an emerging signal early enough to act?
Look beyond the case count: check the data, locate the unusual increase, and decide what needs investigation first.
02 / The data
A simulated surveillance setting.
This analysis uses a simulated, de-identified surveillance dataset structured to reflect routine subnational disease-surveillance workflows. Geographic and facility identifiers are fictionalized. No patient-identifiable or confidential programme data are included.
- Analysis-ready records
- 696
- Epidemiological weeks
- 26
- Diseases analysed
- 4
All cases, deaths and signals are simulated scenarios, not actual events in a real jurisdiction.
03 / Data quality
Make the evidence usable first.
716raw records preserved
696analysis-ready records
- 8 duplicate case IDs identified16 affected rows; 8 duplicate copies excluded.
- 12 implausible agesFlagged and excluded from the core analysis. No imputation.
- 18 missing lab resultsKept distinct from pending and not-collected results.
20 records excluded in total. Raw records and quality flags remain available in the analytical workflow.
04 / Person, place & time
Burden and severity tell different stories.
Reported disease burden
Eligible cases · Observed case fatality ratio (CFR)
Measles366 cases · 1.1% CFR
Cholera217 cases · 4.6% CFR
Meningitis70 cases · 7.1% CFR
Lassa fever43 cases · 16.3% CFR
Age distribution
Years · Eligible cases
0–492
5–14129
15–24167
25–34154
35–4491
45–5444
55–6416
65+3
05 / The outbreak signal
A sharp increase at EW16.
EW15 is the first threshold exceedance. EW16 strengthens the signal, with 36 cases and 3 deaths.
Weekly casesRolling signal threshold
EW15First threshold exceedance
EW1636 cases · 3 deaths
EW15
12 cases0 deaths · Mean 5.50
Threshold 7.74
EW16
36 cases3 deaths · Mean 6.75
Threshold 12.97
EW22
10 cases1 deaths · Mean 4.75
Threshold 9.73
Method and all weekly values
The threshold is the preceding four complete weeks’ mean + 2 standard deviations. The current week is excluded. Values below retain the workbook’s reported precision. This transparent portfolio method does not replace disease-specific national thresholds.
The threshold rises after the peak because the rolling baseline incorporates it. Subsequent non-exceedance does not establish that risk has resolved.
| EW | Cases | Deaths | Prior mean | Threshold |
|---|---|---|---|---|
| 1 | 18 | 0 | — | — |
| 2 | 7 | 0 | — | — |
| 3 | 1 | 0 | — | — |
| 4 | 5 | 0 | — | — |
| 5 | 18 | 2 | 7.75 | 20.35 |
| 6 | 6 | 0 | 7.75 | 20.35 |
| 7 | 10 | 0 | 7.50 | 20.18 |
| 8 | 4 | 0 | 9.75 | 19.99 |
| 9 | 5 | 0 | 9.50 | 20.22 |
| 10 | 5 | 1 | 6.25 | 10.81 |
| 11 | 7 | 1 | 6.00 | 10.70 |
| 12 | 5 | 0 | 5.25 | 7.43 |
| 13 | 6 | 0 | 5.50 | 7.24 |
| 14 | 4 | 0 | 5.75 | 7.41 |
| 15 | 12 | 0 | 5.50 | 7.74 |
| 16 | 36 | 3 | 6.75 | 12.97 |
| 17 | 9 | 0 | 14.50 | 40.02 |
| 18 | 5 | 0 | 15.25 | 39.89 |
| 19 | 8 | 0 | 15.50 | 39.68 |
| 20 | 5 | 0 | 14.50 | 39.50 |
| 21 | 1 | 1 | 6.75 | 10.33 |
| 22 | 10 | 1 | 4.75 | 9.73 |
| 23 | 10 | 1 | 6.00 | 12.78 |
| 24 | 8 | 0 | 6.50 | 14.04 |
| 25 | 3 | 0 | 7.25 | 14.65 |
| 26 | 9 | 0 | 7.75 | 13.47 |
06 / Where should attention go?
Concentrate the investigation.
During EW15–17, Central and East Districts carry the highest operational priority.
East District
18 cases · 1 death
12.00× immediate baseline
Central carried the larger case burden, while East showed the sharper increase relative to baseline.
Fictional districts. Ratios compare the weekly mean in EW15–17 with EW11–14; they are not population incidence ratios.
07 / So what?
Each finding changes the next action.
- Unusual increase
- Verify cases
- Geographic concentration
- Prioritize investigation
- Deaths identified
- Review severity and case-management readiness
- Reporting delays
- Strengthen notification
- Unresolved laboratory status
- Follow specimens and results
08 / Recommended action
Verify. Investigate. Reassess.
Verify
0–24 hoursCheck cases, onset dates, locations and outcomes; coordinate through the surveillance chain.
Investigate
24–72 hoursAssess clustering and common exposures; review case-management readiness.
Lab follow-up
0–48 hoursResolve specimen status and reconcile pending results.
Active case search
0–48 hoursReview registers and seek additional cases in priority facilities and communities.
Reassess
Within 7 daysReview reporting delays and repeat signal assessment with verified and newly reported cases.
Activities overlap: laboratory follow-up and active case search begin alongside verification and investigation.
09 / Important interpretation
A signal is a reason to investigate.
It does not, by itself, confirm an outbreak, prove transmission or establish a common source.
Interpretation and limitations
- Case verification, epidemiological linkage, laboratory evidence where applicable and field investigation are needed before stronger conclusions.
- No district population denominators are available; district-specific incidence cannot be calculated.
- Incomplete laboratory information and absent exposure/WASH variables limit transmission hypotheses.
- The operational priority index combines burden, growth, deaths and reporting delay. It is not a validated prediction score.
- The simulated dataset demonstrates analytical reasoning, not a real surveillance event.
10 / Tools & capabilities
Analysis that supports a decision.
Analytical workflow developed in Excel, from preserved source records to quality assessment, epidemiological interpretation and a proportionate action pathway.
- Data quality assessment
- Descriptive epidemiology
- Person-place-time analysis
- CFR interpretation
- Signal detection
- Reporting-timeliness assessment
- Operational prioritization
- Public-health decision support