Two months ago, I wrote about a new bill aiming to help New Yorkers find out where our nearest defibrillators are. Great news: the text of the bill has been revised and is a lot better, and the New York City Council's Committee on Health will vote on it tomorrow. So now's a good time to tell your councilmember you support Intro 814, "The HEART Act."

Some more details about what might happen tomorrow, how and why the bill is better than it was in March, and diving into data bureaucracy for the future!

Bill name and status, and what happens tomorrow

Intro # 0814-2022 has gotten a lot further through the legislative process at the New York City Council. The next public thing to happen will be a Health Committee vote at 9:30am on Thursday, June 8th. If you have opinions on the bill, now's a great time to mention them to your local councilmember.

Tomorrow after the committee hearing there may be a public press event where I say a few words. [Edited later June 7th to say: this was postponed because of air quality.] And then, starting at 1:30pm that day, the full City Council will meet in a stated session. Assuming that Intro 814 passes the Health Committee vote, the full Council may well vote on the bill in that session.

The bill's primary sponsor is my representative, Councilmember Shekar Krishnan. Today's Gotham Gazette includes a piece by him and New York Senator Chuck Schumer about expanding access to AEDs; Schumer is supporting the proposed federal Access to AEDs Act (H.R.2370).

Hamlin's cardiac arrest makes a clear call to action: we must enact legislation to expand access to even more communities. At the federal level, Leader Schumer’s bipartisan Access to AEDs Act will provide grants to place devices in elementary and secondary schools nationwide, and will fund the development of school-based cardiac emergency response plans. And in New York City, Council Member Krishnan’s HEART (Harihareswara Expand Access to Rapid Treatment) Act will require the Department of Health to publish online the precise locations and quantities of all publicly-accessible AEDs within the five boroughs, whether in schools, firehouses, or government office buildings. If loaded into common mapping apps on our smartphones, this data can make AEDs as quick and easy to find as a cup of coffee.

(Yes, the unofficial name of the NYC bill is "The HEART Act" and the H stands for my last name, in honor of my dad who died of a heart attack in 2010, and in honor of me for advocating on this issue. In New York City, bills don't formally get titles like "The X Act," so this is informal.)

How and why the bill is now better

Originally, the bill would just have caused a city department to issue reports about where AEDs are in general, but probably without street addresses, and probably not as data others could easily reuse.

In April, I submitted written testimony on the bill (PDF), and I thank the several of you who also submitted written testimony to support the bill and ask for improvements (PDF). Councilmember Krishnan's office was really receptive to hearing my thoughts on how to improve the bill, so I sent them a followup email to share thoughts on questions New Yorkers can answer with this data, comparable cities (including DC), and what fields I thought would be useful in the dataset.

[A refresher on where the data comes from & who needs to open it: The short version: it's held by New York City Regional Emergency Medical Services Council (NYC REMSCO), and the Fire Department of NYC definitely has access to it, and the Department of Health and Mental Hygiene of NYC (DOHMH) may have access as well. (Beyond the explanation in my blog post above, I further detailed the who-what-where in the relevant section of my written testimony: section III, "We have this data and can open it to the public", subsection A, "Who holds it and has access to it".)]

The Councilmember's office followed up on this, and on conversations with the relevant folks in the City administration, and revised the bill (which is technically now Intro 814-A; see the Attachments at the bill's Legistar page for the old and new versions of the bill). It now specifies:

No later than January 1, 2024, and every 6 months thereafter, the department shall submit to the mayor and the speaker of the council and post on the department’s website a report indicating the quantities and locations of automated external defibrillators placed in public places pursuant to subdivision b of this section. Such report shall include:

1. Location data for each such device;

2. Location names;

3. Location addresses; and

4. Where practicable, specific information describing the placement of each such device at a location, such as the floor, room, or stairwell.

Such report shall additionally specify whether each such automated external defibrillator is equipped with pediatric-attenuated pads or otherwise equipped with child-appropriate functionality. The department shall submit to the speaker of the council in a machine-readable format all raw data upon which the report required by this subdivision is based.

This is good! Now, if this bill passes, then the specific data fields we want from the Public Access Defibrillator registry will be submitted to the "speaker of the council in a machine-readable format" which, in the usual course of things, means the department will publish them as an NYC Open Data dataset!

Data bureaucracy nerdery

(This bit is super skippable unless you are very into the labyrinthine flows of data from contributor to steward to publisher, or unless you want to use Intro 814 as model legislation for a similar bill in your area.)

I figure questions New Yorkers will use this data to answer are:

• where are the AED units nearest a particular location or route, and are they available 24/7?
• what neighborhoods have very few AEDs, or AEDs that are often unavailable (because the building is often closed)?
• where should we put a new AED for maximum utility?
• which public access AEDs are not properly maintained?

The PAD registry can help us answer these questions. It includes data from the following New York State forms:

1. Notice of Intent to Provide Public Access Defibrillation, DOH 4135
2. Collaborative Agreement (such as this sample from NYC REMSCO)
3. Quality Assurance Reports or usage reports (I think this is, for example, this "PAD Q.I. report" or the Notice of Use by PAD Agency in this packet)
4. possibly a PAD Provider Information Form but I can't find examples of this

and thus includes:

• the street addresses of buildings with AEDs
• the "agency" type ("agencies" including "Physician's Office", "Transportation Hub" "Stadium", "Business" and other categories)
• the number of trained providers and AED units at each location
• whether the units at a location are Pediatric Capable [in versions of the Notice of Intent from at least 2009 onwards, in form DOH-4135(4/09)]

For reference, I also shared examples from some other cities that put public AED locations in their open data portals: Kansas City, MO; Boise, Idaho; District of Columbia. They particularly found the DC example helpful -- thanks for helping me research and for finding those, Leonard!

So, that's where we'll be getting each of the data fields mentioned in Intro 814.

1. "Location data for each such device": This could include Type of Entity (the "agency" type from the Notice of Intent), and reuse the Number of AED Units from the Notice of Intent.
2. "Location names": This would probably be AED Owner (the organization in charge of the site: "Entity Providing PAD" on the Notice of Intent).
3. "Location addresses": The street address, including ZIP code, from the Notice of Intent.
4. "Where practicable, specific information describing the placement of each such device at a location, such as the floor, room, or stairwell.": In some other cities' datasets this is called AED Physical Location or Placement Details and includes a pointer to a more specific location, such as “On wall between shop and office”. There is no specific place for this in the Notice of Intent form, but I suspect people annotate the Address field and so it's in the registry.
5. Pediatric capability ("whether each such automated external defibrillator is equipped with pediatric-attenuated pads or otherwise equipped with child-appropriate functionality"): This data comes from the question "is the AED Pediatric Capable?" in the Notice of Intent; some older versions of the of Intent do not include this info. I noticed that the language of the Notice of Intent assumes that all the AEDs at a site are the same model, and thus doesn't allow for the possibility that some of the AEDs are Pediatric Capable and some aren't. So that's worth looking into.

In theory, the following data could also be published as separate fields in this dataset or a New York State dataset. And it'll be worth thinking about this in the future in case we realize that the published data doesn't quite help us answer the questions we want to answer, and we need a bit more.

• Data from the usage reports, which I think are also known as quality assurance/quality improvement (QA and QI) reports. Within a few days of the usage of a PAD AED, the PAD Provider needs to submit to the REMSCO a usage report that includes the name of the PAD Provider, date and time of incident, patient age and sex, estimated time from arrest to 1st AED shock, estimated time from arrest to CPR, number of shocks administered to the patient, name of transporting ambulance service, and patient outcome at incident site. Getting these stats could help us answer the question "where should we put a new AED for maximum utility?"
• Type of Entity: Some entities are open 24 hours a day and 7 days a week, such as police stations. Some, such as restaurants and doctor's offices, aren't. Some proportion of cardiac arrests happen on weekends and early in the morning. So some neighborhoods might only have public access AEDs that are often unavailable, because the building is often closed. This affects users who want to know where their nearest always-available AED unit is, or to analyze what neighborhoods have too few always-available AEDs, or to suggest where should we put a new AED for maximum utility. So we might want to break out "type of entity" separately from "Location data for each such device".
• Date of the last Collaborative Agreement (should be within the last two years). If we decide we need more data to help answer "which public access AEDs are not properly maintained?" then this field could come in handy.
• Date of the last NoI update (Notice of Intent to Provide Public Access Defibrillation update). Pretty similar to the previous item.
• Number of AED units at a specific location. I suspect each row of the new dataset will refer to a device, not a location, so if one location has three devices, we'll get three rows with identical names and addresses. It should be reasonably easy to work out from that data how many AED units sit at a specific location, but if it's not, then we might want that broken out into its own field.
• PAD training program: The PAD rules say that "The PAD Provider shall ensure that all persons designated to operate an AED successfully complete an AED training course which has been approved by the New York State Department of Health or the New York State Emergency Medical Services Council." It could conceivably be the case that some of these courses are more effective than others.
• Number of trained PAD Providers: Similar reasoning to the previous item.
• AED geocoded location (the street address, automatically translated into latitude and longitude for use in mapping software). I think we don't need this because it's pretty trivial to turn a street address into a lat/long.

Hope this passes!

If the HEART Act passes, and New York City implements it, correctly, then by January 1st we'll have a cool new opportunity to help educate New Yorkers so we can save lives. I'll be posting here again to ask for your help then to make maps and tools, and spread the word.

So I hope this passes!