NSSP — Emergency Department Visits¶

The nssp module queries the NSSP BioSense Platform (National Syndromic Surveillance Program) via the CMU Delphi Epidata API (covidcast endpoint). It tracks the percentage of ED visits attributed to COVID-19, influenza, RSV, or a combined respiratory illness index.

No API key required. Data updates weekly.

Signals¶

Signal	What it measures
`covid`	% ED visits with COVID-19 diagnosis
`influenza`	% ED visits with influenza diagnosis
`rsv`	% ED visits with RSV diagnosis
`combined`	Combined respiratory illness (COVID + flu + RSV)

Geographic levels¶

Level	`--geo` value	Notes
National	`nation`	Single aggregate value
State	`state` (default)	2-letter USPS code e.g. `ca`
County	`county`	5-digit FIPS e.g. `06037`
HHS Region	`hhs`	HHS region number 1–10
Metro area	`msa`
HRR	`hrr`	Hospital Referral Region

Commands¶

`query` — flexible single-signal query¶

# National COVID-19 ED visits, last 52 weeks (default)
uv run python -m nssp query covid --geo nation -f table

# Influenza by state, last 52 weeks
uv run python -m nssp query influenza --geo state --region '*' -f csv

# RSV in California
uv run python -m nssp query rsv --geo state --region ca -f table

# Specific HHS region
uv run python -m nssp query influenza --geo hhs --region 4 -f table

# Custom date range (epiweek format YYYYWW)
uv run python -m nssp query covid --geo state --region ny \
  --start 202201 --end 202360 -f csv

`national` — all four signals at once¶

# All four signals, national, last 52 weeks
uv run python -m nssp national -f table

# Custom range
uv run python -m nssp national --start 202201 --end 202360 -f csv

Sample output:

signal       epiweek   value
covid        202501    3.2
influenza    202501    2.1
rsv          202501    0.8
combined     202501    5.9

`hhs` — by HHS region¶

# Influenza across all 10 HHS regions
uv run python -m nssp hhs influenza -f table

# Specific region
uv run python -m nssp hhs influenza --region 4

# COVID by region, custom range
uv run python -m nssp hhs covid --start 202340 --end 202560 -f csv

Python SDK¶

from nssp.client import NSSPClient

client = NSSPClient()

# National COVID, last year
rows = client.fetch(
    signal="covid",
    geo_type="nation",
    geo_value="us",
)

# All states, influenza
rows = client.fetch(
    signal="influenza",
    geo_type="state",
    geo_value="*",
)

# Custom epiweek range
rows = client.fetch(
    signal="rsv",
    geo_type="state",
    geo_value="ca",
    time_values="202240-202560",
)

Response fields¶

Field	Description
`signal`	Signal name
`geo_type`	Geographic level
`geo_value`	Geographic code
`time_value`	Epiweek (YYYYWW)
`value`	% of ED visits
`stderr`	Standard error
`sample_size`	Number of ED visits in the denominator
`direction`	Trend direction vs prior week

Multi-year collection¶

NSSP data via Delphi goes back to 2020 for COVID, 2021 for influenza/RSV. Collecting all available history:

from nssp.client import NSSPClient

client = NSSPClient()

# Full history for all signals
for signal in ["covid", "influenza", "rsv", "combined"]:
    rows = client.fetch(
        signal=signal,
        geo_type="state",
        geo_value="*",
        time_values="202001-202660",  # epiweeks 2020 week 1 → 2026 week 60
    )
    # Save to CSV
    ...

No rate limit — the Delphi API doesn't enforce 15-second waits like WONDER. Fetches are typically < 2 seconds and results are cached for 24 hours.

Combining with GRASP FluView¶

NSSP and GRASP FluView measure different things for influenza:

	NSSP	GRASP FluView ILINet
Metric	% ED visits with flu dx	% outpatient visits with ILI symptoms
Coverage	2020–present	1997–present
Geography	State, county, HHS	National, HHS, state
Lag	~1 week	~1 week
Notes	Confirmed diagnosis required	Symptom-based (not lab-confirmed)

Use ILINet for historical trend data; use NSSP for current season monitoring with confirmed diagnoses.