OSHA Severe-Injury Reports Scraper
DeprecatedPricing
from $5.00 / 1,000 results
OSHA Severe-Injury Reports Scraper
DeprecatedOne clean item per OSHA severe-injury report: employer, city, state, industry code, and whether a hospitalisation or amputation was reported. Firms only, no worker names. Pay per run.
Pricing
from $5.00 / 1,000 results
Rating
0.0
(0)
Developer
US Tech Automations
Maintained by CommunityActor stats
0
Bookmarked
2
Total users
1
Monthly active users
16 hours ago
Last modified
Categories
Share
One clean row per OSHA severe-injury report. You give it a state, a date range, an industry-code prefix and/or a keyword; it hands back a table.
What each row is
event_date, employer, city, state, naics (industry code),
hospitalized, amputation, nature, body_part, and the source_url it came
from. The subject is always the employer (a firm). No injured worker is
named — the public dataset carries no such name and this actor keeps none. No
street address is kept; town and state only.
Source
OSHA's public severe-injury dataset (the "Get the data" CSV behind
https://www.osha.gov/severeinjury). Public-domain US Government data.
Heads up: from some networks osha.gov answers automated fetches with
HTTP 403. A real run on Apify fetches live; local apify run from a blocked
network uses the clearly-labelled synthetic fixture in fixtures/ (every value
is marked SAMPLE so it can never be mistaken for a real report). See
../../SOURCES.md.
Price (pay-per-event, billed by Apify)
- $0.50 to start a run (
run-start) - $0.005 for each report returned (
result-item)
So a run that returns 200 reports costs about $1.50, billed by Apify to the buyer's Apify account.
Compute cost vs price
Timed locally, parsing + filtering is ~0.4 ms per 200 rows — the run's cost is dominated by the one CSV download. Estimating a generous 10-second run at the 512 MB default memory and Apify's ~$0.40 per GB-hour:
| value | |
|---|---|
| Run wall time (fetch + parse, est.) | ~10 s |
| Memory | 0.5 GB |
| Compute cost of the run | ~$0.0006 |
| Revenue of a 200-item run | $1.50 |
| Compute ÷ revenue | ~0.04% |
Well under the 30% ceiling.
Run it
maxItems is capped hard at 1000 and there is a run timeout, so a run cannot run
away. main.collect(inp, rows=None) is a plain function you can import and test
offline.
Not affiliated with OSHA. Not legal, tax or professional advice.
