How to choose a public health surveillance tool

"Public health surveillance" is not one category of software so much as five different jobs that happen to share a mission: notice a disease trend before it shows up in clinical case counts, or organize the data once it has. A wastewater lab, a genomic data-sharing initiative, a news-mining outbreak map and a survey-building tool for field epidemiologists solve completely different problems, and none of them substitutes for another. If you are a state or local health department, a hospital system, a ministry of health, a field epidemiology training program or a newsroom trying to track an emerging outbreak, the first decision is not which product is best — it is which kind of signal you actually need.

This is also a category where most of the leading tools are not commercial products in the usual sense. A university consortium, a CDC program and a philanthropically funded research network sit alongside a venture-backed lab-services company. That changes how you shop: price is rarely the differentiator, and "who governs the data and who gets attributed for it" often matters more than any feature.

Decide what kind of signal you need

• Biological signal from a population, with no individual testing. Wastewater-based epidemiology tests composite sewage samples from a treatment plant to estimate how much of a pathogen — or a substance like an opioid — is circulating in that sewershed, without identifying anyone. Biobot Analytics and WastewaterSCAN both do this; see Biobot Analytics vs WastewaterSCAN for how they differ.
• Genomic sequence data on a specific pathogen. If you need to know which variant of a virus is circulating, or to track how a pathogen is mutating across countries, you need a sequence database, not a case-count dashboard. GISAID is the field's standard for this, covering influenza, SARS-CoV-2, RSV, mpox and several arboviruses.
• Early, informal signal from open text. HealthMap mines news reports, official bulletins and informal accounts in nine languages to plot emerging outbreaks in near real time. It is fast and broad but reflects reporting quality in a region, not laboratory confirmation.
• Structured data collection for an outbreak response or a registry. Epi Info and REDCap both let epidemiologists build forms, capture line-list data and run statistics, but they are not signal sources themselves — they are how you organize whatever cases you already know about. See Epi Info vs REDCap.

Most surveillance programs of any size end up combining at least two of these: a wastewater or genomic feed for early, population-level warning, and a data-collection tool for managing the confirmed-case side once an investigation starts.

Understand the access and governance model

Because so much of this category is public-good infrastructure, "how do I get access" is often a more useful question than "what does it cost."

• Open and free, no registration. HealthMap's map and app are simply public.
• Free, but access-controlled and attributed. GISAID requires identity verification and agreement to a data-sharing etiquette: you can use any sequence in the database, but you must credit the originating and submitting laboratories. This exists to keep labs willing to share during an outbreak, and it is not optional — build the attribution step into your workflow rather than treating it as paperwork.
• Free to a defined institutional community. Epi Info is published directly by the CDC with no membership required. REDCap is different: it is distributed through a consortium, and only non-profit and government institutions that join can obtain the codebase and self-host it. If your organization is not consortium-eligible, REDCap is not available to you regardless of budget.
• Paid service, not software. Biobot sells sampling kits, lab analysis and reporting as a contracted service with quoted pricing; there is no self-serve tier. WastewaterSCAN, doing similar work, is philanthropically funded and publishes its sewershed-level results publicly at no charge — the same underlying method, two different funding and access models.

Check which of these your organization actually qualifies for before you evaluate features. A tool that is technically excellent but requires a consortium membership you cannot obtain, or a contract your budget cycle cannot accommodate this year, is not a real option.

Weigh latency against specificity

Every source in this category trades speed for confidence in a different place. Wastewater data can show a rising signal days before people seek clinical care, but it tells you about a sewershed, not a diagnosis — you get trend direction and rough magnitude, not a confirmed case count. News-mining tools like HealthMap can surface an unusual cluster of pneumonia cases within hours of an early report, before any lab has sequenced anything, but the same speed means noise and false leads are part of the deal. Genomic databases are the slowest and most precise: a sequence takes time to generate and submit, but once it is in GISAID you know exactly what pathogen and variant you are looking at. None of this makes one source "better" — a mature surveillance program uses the fast, noisy sources to know where to look, and the slow, precise ones to confirm what it found.

Check deployment and data residency

Deployment varies more in this category than the pricing pattern suggests. Biobot, GISAID, HealthMap and WastewaterSCAN are all cloud services you access as a user. Epi Info ships desktop, mobile and web/cloud versions, so it can run disconnected in the field. REDCap is the outlier: it is self-hosted only, run and supported independently by each member institution, which means your institution's IT team — not a vendor — is responsible for uptime, patching and the HIPAA-relevant security controls around it. If your team does not have that capacity, factor the hosting burden into the decision alongside the (zero) license cost.

A shortlist by situation

• If you are a health department wanting an early-warning signal for a specific sewershed and can fund a contracted service, Biobot Analytics gives you dedicated sampling and lab turnaround on your timeline.
• If you want the same wastewater method without a contract, check whether your area is already covered by WastewaterSCAN's public network before building your own.
• If you need to track which variant of a pathogen is spreading, GISAID is close to mandatory; budget time for the credentialing and attribution process, not money.
• If you want broad, fast, informal situational awareness across many diseases and regions at once, HealthMap is free and requires no setup.
• If you are running an outbreak investigation or a registry and need to build forms fast, Epi Info needs no institutional membership; REDCap is stronger for longitudinal, multi-site research if your institution already belongs to the consortium.

Questions to ask before you commit

1. Who is legally allowed to access this data, and does our institution meet that bar today?
2. What attribution or acknowledgment obligations come with using the data, and who enforces them?
3. If this is a paid service, what happens to historical trend data if we stop the contract?
4. Who hosts and secures this system, and does that meet our compliance requirements (HIPAA or equivalent)?
5. How is a false or delayed signal in this source likely to show up, and how would we catch it?

Common mistakes

Treating a fast, informal source like HealthMap as if it were laboratory-confirmed data, and briefing leadership on it as though it were. Assuming REDCap is simply "free software" without accounting for the hosting and support burden that lands on your own institution. Skipping the GISAID attribution step because it feels like bureaucracy, which risks access for your whole institution, not just you. And building a surveillance program around a single source when the entire value of this category comes from combining an early, noisy signal with a slower, precise one.

Browse every tool in this category for the full list.