Complete AI Training
Sign inGet my AI kit

Your job's AI kit

Get your AI kit

Tell us who you are and what you do. We show you your kit right away and email you the link: skills, prompts, AI agents, MCP servers and courses for your job.

500+ jobs ready, and we make a kit for any other job. No payment needed to look.

Share

AI agent for epidemiologists

Vaccine Coverage Gap Mapping Agent

A map of real coverage gaps, with reporting artifacts identified and removed.

Vaccine Coverage Gap Mapping Agent: what goes in, what the agent does and what you get

What it does

Coverage looks fine at the district level while a few neighborhoods sit well below target, and sometimes the low number is just a data problem. This agent combines registry and population data, checks data quality, and calculates coverage by small area. It flags the low areas, then tests whether each gap is real or an artifact, such as a wrong population denominator, missing registry submissions, or a clinic not reporting. It rechecks flagged areas with a second source, such as school records or clinic counts. It builds a map and a list. You approve the map before it is shared. Edge case: small areas with fewer than a set number of people use pooled estimates to avoid false alarms.

How it works

Follow the arrows from top to bottom. The orange dashed arrow is the loop: when a check fails, the agent goes back and tries again.

Start and resultWhat it doesA check on its own workWaits for your OKGoes back and retries
Yes, continueYes, continueApprovedNoNo 1 STARTS WHEN Registry data loaded 2 USES A TOOL Load registry and population data and check basicquality 3 DOES Calculate coverage for each small area 4 DOES Flag areas under the target 5 CHECKS THE RESULT Does a second source confirm each flagged area istruly low? If not: check denominators, missing clinic reports andduplicates, then recalculate. Back to step 4. 6 DOES Pool or smooth very small areas 7 CHECKS THE RESULT Do the flagged areas remain under target afterpooling? If not: remove them from the list and note theuncertainty. Back to step 4. 8 DOES Build the map and the list of confirmed gaps withcauses 9 YOU APPROVE Epidemiologist approves the map 10 RESULT Coverage gap map and list
Read the steps as a list
  1. Registry data loaded
  2. Load registry and population data and check basic quality
  3. Calculate coverage for each small area
  4. Flag areas under the target
  5. Does a second source confirm each flagged area is truly low?If not: check denominators, missing clinic reports and duplicates, then recalculate. Back to step 4.
  6. Pool or smooth very small areas
  7. Do the flagged areas remain under target after pooling?If not: remove them from the list and note the uncertainty. Back to step 4.
  8. Build the map and the list of confirmed gaps with causes
  9. Epidemiologist approves the mapThe agent waits here for your OK.
  10. Coverage gap map and list

How it decides

An area is low if coverage is under target with enough population. It is real when a second source or a data check confirms it.

  • Area population under 100: use pooled estimates
  • Clinic with no submissions this month: possible artifact, ask for data
  • Coverage under the target by over 10 points and confirmed: priority area
  • Denominator differs from census by over 20 percent: recheck before use

Make it yours

Every agent is a starting point. You choose these settings for your own situation.

  • Target coverage (default 90 percent)
  • Small area size rule
  • Second sources used
  • Geography level
  • Update schedule

What keeps you in control

It always asks you first

  • The map before sharing
  • Outreach plans based on it

Hard limits

  • Never share maps with individual-level data
  • Never present a gap without a data check

It stops when

  • Done: map approved
  • Stop: registry data incomplete

Set it up

We guide you through the set-up, step by step

Members get the full set-up guide for this agent. No technical skills needed: you copy, paste and upload.

10 minto set it up in your AI
5 AIsChatGPT, Claude, Copilot, Gemini, Grok
  • One set of instructions to paste into your AI, with the clicks for ChatGPT, Claude, Microsoft 365 Copilot, Gemini and Grok
  • The agent then walks you through connecting your own data, one source at a time
  • A downloadable copy with the flow chart, the rules and the full guide
Get access to this agent

An example run

What happensIn June, 14 tracts fell under 80 percent. The agent found that tract 112 had a 31 percent denominator mismatch, and a clinic in tract 118 had not submitted since March. The check failed for both. After correction, tract 112 rose to 84 percent and 118 to 76. Nine tracts stayed low, three pooled. The epidemiologist approved a map of 9 confirmed areas.

More agents for epidemiologists