You are the first responder

Most cardiac arrests outside hospital happen at home: 73.4% of them in the US, and 70 to 80% in the UK.¹² So the first person there is usually not a paramedic or a stranger. It's a partner, a son, a flatmate.

Two terms carry this whole course. In cardiac arrest the heart stops pumping blood: as the British Heart Foundation puts it, "When a person is in cardiac arrest their heart stops pumping blood, which means their brain gets no oxygen."[30] The person collapses, does not respond and is not breathing normally, and lesson 2 teaches you to recognise it. It is not the same thing as a heart attack, and lesson 12 separates the two. CPR, cardiopulmonary resuscitation, is pushing hard and fast on the centre of the chest, sometimes with rescue breaths, to keep some blood moving until help arrives. A defibrillator, or AED (automated external defibrillator), is a machine the public can use that reads the heart's rhythm and gives a shock only when it detects the kind that needs one.¹⁵ Lesson 5 is about it.

Ask people why they'd hold back, and the answers come from a short list: someone else will do it, I'll be sued, I'll make it worse, I'm not qualified, I don't know what to do first. This lesson takes each worry in turn and sets it against the evidence, so that by the end you know which ones hold, which don't, and what the course you're starting will and won't give you.

That box opens every lesson in this course, word for word, because someone may arrive at the choking lesson from a search engine with a child coughing in the next room and never read this one. Every clause in it is explained somewhere in the course: the numbers and the speakerphone in lesson 2, why a page cannot teach your hands in this lesson and in lesson 16.

What this course is, and what it is not

It teaches what the main resuscitation and first aid bodies currently tell a bystander to do in the first minutes of an emergency, and why. Those bodies are the American Heart Association (AHA), the American Red Cross, the European Resuscitation Council (ERC), the Resuscitation Council UK (RCUK), and the International Liaison Committee on Resuscitation (ILCOR), which runs the evidence reviews the others build on. The AHA, the ERC and RCUK published new guidelines in October 2025, built on ILCOR's evidence reviews from the same month.

It is not a certificate, and it cannot become one. Every body that issues a workplace first aid certificate wants your hands checked, not just your reading. The AHA's online Heartsaver route still requires that "students must complete hands-on skills session with an AHA BLS or Heartsaver Instructor", and the UK's Health and Safety Executive (HSE) says that "HSE strongly recommends that elements of the syllabus requiring practical demonstration of first-aid administration should be assessed by direct observation".³⁴ In a class that happens on a manikin, the life-size practice dummy. That is why the objectives in this course say state, explain, decide, sequence and recognise, and never perform. A page cannot show that you have pushed deep enough.

What a page can do is make you more likely to try. In one randomised trial of 336 people, tested in a simulated arrest on a manikin that measures CPR, 23.5% of the untrained group did not even attempt it, against 0.7% of those who had watched a short video (the videos ran from one minute to eight).⁵ Knowing what to do changed whether people moved at all, at least in a simulation. The course is built on that finding, and on the plain fact that most people who need CPR won't have anyone better beside them.

Some things are left out, so you know before you go looking:

• Anything a professional does. Pulse checks, oxygen, airway devices, rescue medicines, neck collars, splinting beyond improvised support.
• Wilderness and disaster care, which belongs to Emergency Preparedness and Wilderness Survival, much later on the Core. Lesson 15 covers heat, cold and bites for a bystander who can call an ambulance.
• Childbirth, and chest injuries such as an open chest wound. This course did not research either, so neither is taught.
• Mental health beyond first response. Lesson 16 gives the first aid response to someone with suicidal thoughts and points to Mental Fitness for the rest.
• Addiction. Lesson 7 teaches recognising an opioid overdose and giving naloxone, and nothing about drug use itself.
• Fire. Cooking Fundamentals lesson 2 covered the pan fire; this course picks up at the burn.
• Legal advice. This lesson and lesson 5 describe the law as general information only.

Two countries, and which one to follow

Where the US and UK/European bodies agree, the course says so in one sentence and names both. Where they differ, you will get a short table or an "In the US / In the UK and Europe" pair. Most of the differences are judgement calls on thin evidence, made differently for good reasons, and the course calls neither side wrong.

Follow the guidance of the country you're in, because that is what your call handler and your class will use. If you live somewhere else, follow your own national resuscitation council. The course shows these two because their guidelines are the ones it read.

How to read the label

Every instruction from lesson 2 on carries a short label in brackets, like this: "All lay rescuers should provide chest compressions for adults with presumed cardiac arrest." (AHA 2025, COR 1).⁶ The label tells you three things: who says it, in what year, and how strongly. "COR 1" is the AHA's top class of recommendation. ILCOR's labels read differently, as "strong" or "weak" with a certainty from high to very low.

If that sounds familiar, it should. Sleep lesson 6 pulled apart the two halves of a guideline grade, how strongly a panel recommends something and how good the evidence behind it is. The same two halves run through every label here, and lesson 4 teaches this field's version properly. For now, one thing: a lot of first aid advice carries a weak grade, and that almost never means "it doesn't matter what you do". It usually means the evidence cannot separate two ways of acting, while the evidence that acting beats doing nothing is much stronger.

The year is there because guidance moves. The October 2025 guidelines changed how a US rescuer treats a choking adult and how a UK rescuer times the call, and some public web pages still teach the old versions. When you meet an instruction anywhere, this course or elsewhere, the question to ask is: which guideline, which year, and how strongly?

Who will be there

The AHA's full breakdown for the US is "homes/residences (73.4%), followed by public settings (16.3%), and nursing homes (10.3%)".¹ The collapse in a shopping centre with a crowd around it is the minority case.

That matters for the worries, because some of them are about strangers. In a mailed survey of 755 adults in Arizona, "More than 50% of respondents reported being willing to perform CPR on a stranger and over 80% reported being willing to perform CPR on a family member."⁷ The course read only that survey's abstract, and a survey measures what people say rather than what they do. The gap says people report being readier to help family, and the person on your kitchen floor is more likely to be family than a stranger.

The story everyone knows

Almost everyone who has heard of the bystander effect heard it through one story. In 1964 a woman named Kitty Genovese was murdered, and the New York Times reported that 38 people watched and did nothing.

The correction does not make the bystander effect false. It means the most famous evidence for it was not evidence. The real evidence is in the laboratory studies and, more recently, in what cameras record.

What the bystander research does show

In 2011 Peter Fischer and colleagues pooled the laboratory studies: over 7,700 participants and 105 effect sizes. Overall, the presence of other people reduced an individual's chance of helping, with a standardised effect of g = -0.35.¹⁰ That is the kind of number Mental Fitness lesson 2 taught you to read, negative here because helping went down. On the normal-curve conversion that lesson used, which is its own arithmetic and not Fischer's, 0.35 puts the average person tested with others behind about 64 percent of those tested alone: a real effect, and a modest one. People helped less with others present. They did not stop helping.

The more useful part of their abstract is where the effect shrank. "The bystander effect was attenuated when situations were perceived as dangerous", and also when a perpetrator was present, when the cost of helping was physical, and when the other bystanders were a source of support, for instance because they were not strangers. The authors don't claim to have shown why. They say the pattern is "consistent with the arousal-cost-reward model", which proposes that dangerous emergencies are "recognized faster and more clearly as real emergencies", and so stir people up more and draw more help.¹⁰ That is a model the findings fit, not a finding. The course read only the abstract, which says nothing about responsibility or about who takes charge, so that is all it can say about why.

This course takes one practical point from it, and it's the course's inference rather than anything Fischer tested: if part of the freeze is not being sure this is a real emergency, then saying out loud that it is may help.

So the two lines of evidence fit together. Each individual is somewhat less likely to act when others are present, but a crowd as a whole usually produces someone who does. Philpot and colleagues draw that reconciliation themselves: more people may lower each person's chance of helping, but they also give a larger pool from which a helper can come.¹¹ That is why the footage does not overturn the laboratory.

Now ask the question Memory taught you to ask of any study: who was studied? Philpot's 219 clips are public conflicts: aggression, arguments, fights. Not one of them is a collapse. The study shows that people step into a fight far more often than the Genovese story implies. It does not show that anyone in a crowd will kneel down and start CPR, which is a different act needing different knowledge.

Why people say they wouldn't

When people are asked directly, fear of causing harm comes up again and again. In a 2025 YouGov poll for the British Heart Foundation (BHF), "Around half of people (51 per cent) who live with their partner said the fear of causing harm would make them reluctant to give cardiopulmonary resuscitation", and 57% said they wouldn't feel confident using a defibrillator.¹² In the charity's 2024 poll, "Nearly half of UK adults (43 per cent) have never learnt CPR" and "only 40 per cent of respondents could pick the correct first step of CPR".¹² These are polls commissioned by a charity with a campaign to run, so read them as a signal of what people feel, not as population estimates.

Two other sources add detail. A systematic review of barriers in deprived communities found them "more acute in deprived communities: the risk to personal safety in administering CPR; the fear of legal consequences; and the lack of community cohesion and other cultural barriers", with little evidence that willingness itself was lower.¹³ And an online survey of 548 Americans, asked why women receive bystander CPR less often than men in public, gave reasons including "fears about inappropriate touching, accusations of sexual assault, and fear of causing injury".¹⁴ Those are the public's guesses about why, not measured causes, and lesson 4 reports the figures on who actually receives CPR.

So the list is: I'll hurt them, I'll be sued, I'm not trained, I shouldn't touch them, it's not safe. Safety is real, and lesson 2 opens with it. The rest can be taken one at a time.

The kitchen floor

Start with the common case. You come downstairs and find your father on the kitchen floor. He does not respond when you shake his shoulders and shout.

Run the worries against it. Someone else will help? There's nobody else. The bystander effect needs bystanders, and you're alone with him. I'll be sued? The legal sections below show how rarely anyone has claimed that giving CPR itself did harm. I shouldn't touch him? He's your father, and he cannot answer; the consent section below shows how UK law covers helping someone who cannot answer. I'll make it worse? If his heart has stopped, RCUK's law booklet puts it bluntly: making things worse "would be virtually impossible, since without intervention death is inevitable."¹⁵

Each fear in the polls has an answer here, and what's left is the real question: what do I do first? The answer is to call, on speaker, and lesson 2 takes it from there, second by second.

That is the point of starting with the home. Some of the fears, being sued and touching a stranger, belong mostly to public places, and most arrests aren't there. The fear of doing harm comes home with you: the BHF's 51% were people who live with a partner. For the public case, you need the law.

Being sued: the United States

Every US state has some form of Good Samaritan law. The paper this section relies on speaks of "'Good Samaritan' statutes in all 50 states".¹⁶ They differ in detail, so here's one in full, from California's Health and Safety Code, section 1799.102:

"no person who in good faith, and not for compensation, renders emergency medical or nonmedical care or assistance at the scene of an emergency shall be liable for civil damages resulting from any act or omission other than an act or omission constituting gross negligence or willful or wanton misconduct."¹⁷

Three conditions are packed into that sentence, and they're conditions many such laws share: you acted in good faith, you were not paid, and what you did was not grossly negligent. A bystander who does their honest best with CPR meets all three. Lesson 5 adds a federal law that protects people who use a defibrillator where their state has no such law of its own.

The 170 CPR cases Murphy and colleagues found in thirty years of US court records, drawn to one scale.¹⁶ The chart is this course's drawing of the counts in the paper's abstract.

Read that carefully, because it's easy to take the wrong lesson from it. The three were claims that giving CPR did harm. The 167 claimed the CPR was inadequate or came too late, and inadequate CPR is CPR somebody gave, so some of those defendants had given it. The abstract does not say who the defendants were, or whether any were passers-by, and the course read only the abstract. So the chart does not show that strangers get sued for walking past, and it does not show that giving CPR keeps you out of court. What it does show is that suits claiming the CPR itself did harm were very rare: three found in thirty years across the whole country. The authors' own conclusion is that "The likelihood of litigation is significantly higher in cases with bystander CPR absent or delayed".¹⁶

Three states that require you to help

A few US states go further and put a duty on everyone. Vermont's statute, from 1968, says that a person who knows another "is exposed to grave physical harm shall, to the extent that the same can be rendered without danger or peril to himself or herself or without interference with important duties owed to others, give reasonable assistance", unless others are already providing it. Breaking it wilfully "shall be fined not more than $100.00".¹⁸ Minnesota has the same duty, and spells out what it asks: "Reasonable assistance may include obtaining or attempting to obtain aid from law enforcement or medical personnel." Breaking it is "a petty misdemeanor".¹⁸ Rhode Island requires a person to "give reasonable assistance", with a petty misdemeanour carrying up to six months or $500.¹⁸

None of the three asks you to put yourself in danger: each limits the duty to what you can do without danger to yourself.¹⁸ Minnesota's says in so many words that getting aid from the police or medical staff can count. Vermont's and Rhode Island's don't say what counts as "reasonable assistance", and this course reads a call from somewhere safe as the obvious minimum there too, which is a reading and not a court's ruling. Other states have narrower duties, such as reporting certain crimes, and the course did not survey them.

Being sued: the United Kingdom

The UK has no general Good Samaritan statute. It relies on the common law, the law built up from judges' past decisions, and RCUK's booklet Cardiopulmonary resuscitation, automated defibrillators and the law, reviewed in 2017 by one of its original authors and a legal specialist, sums up how the courts treat helpers: "The courts have always looked benevolently on those who have gone to the assistance of others."¹⁵

On lawsuits the booklet, in its April 2018 edition, is precise: "Although there have been a few cases in the UK when a claim for damages has been brought against a member of the public or a first aider who has attempted resuscitation, there have been no reported cases in which someone has successfully sued anyone who came to help them in an emergency situation."¹⁵

You may meet a stronger version. RCUK's own FAQ page says: "No such action has ever been brought against someone who performed CPR".¹⁹ The two do not quite agree: the booklet says a few claims have been brought and none succeeded, the FAQ that none has been brought at all. This course follows the booklet, the fuller and more careful text. It is the same move Personal Finance asked of you when two findings disagree: work out what each one counts. One counts claims brought, the other counts claims won, and "none won" is the one both support.

Three more points from the UK law:

• There's no duty to rescue. "In the UK, there is no legal obligation for others to help a person in need of resuscitation, provided they were not the cause of the person needing help." But "once a bystander volunteers to help, they are then considered to have a duty of care to assist the person as far as they are able."¹⁵ That means doing your honest best, not doing it perfectly. The booklet names exceptions to the first half: some professionals, people with "a particular relationship with the collapsed person", and a workplace first aider who has taken the role on as part of the job may already owe a duty of care.¹⁵
• Liability needs you to have made things worse, and in cardiac arrest the booklet says that "would be virtually impossible, since without intervention death is inevitable."¹⁵
• England and Wales have a statute; Scotland and Northern Ireland do not. The Social Action, Responsibility and Heroism Act 2015 tells a court deciding a negligence claim that it "must have regard to whether the alleged negligence or breach of statutory duty occurred when the person was acting heroically by intervening in an emergency to assist an individual in danger." Section 5 says: "This Act extends to England and Wales only."²⁰ RCUK notes that "it has been criticised by prominent members of the legal profession who state that it adds nothing to the protection already provided by existing common law and that it may, in fact, erode the protection that already exists".¹⁵

One more line from the booklet cuts the other way. It says it wouldn't be reasonable for an unqualified person to carry on alone if "a professional rescuer were present or arrived at the scene and offered to help", though you may keep helping under their direction.¹⁵

Should the law make you help?

Behind the three US statutes is a question the evidence cannot settle, because it's a question about what a law ought to do. The course describes it and does not answer it.

The case for a duty, as its supporters make it, runs like this. A minimal duty, limited to what is safe and satisfied by a phone call, states a moral floor that almost everyone already accepts, may nudge a hesitant bystander into acting, and costs nothing to anyone who would have helped anyway. Vermont, Minnesota and Rhode Island have written such duties into law.

The case against, in the common-law tradition that England and most US states follow, holds that the law shouldn't compel a stranger to act unless they caused the danger. Its defenders argue that virtue the law compels is not virtue, that words like "reasonable assistance" are vague and such laws rarely enforced, and that the threat of prosecution could crowd a scene rather than help it. UK critics of the 2015 Act make a narrower point about a protective statute, not a duty: the common law already protects helpers, so the Act adds little and may weaken that protection.

The ERC's 2025 book for the public takes no side here; it says people should be told about both "local 'Duty to Help' regulations and 'Good Samaritan' laws".²¹

The course read the statutes and RCUK's summary of the UK position, not a full argument from a defender of either view, so the case for and the case against above are the course's summary of each side.

What does not depend on which you prefer: where one of these duties exists, none asks you into danger and a call from somewhere safe is the obvious minimum, and where there's no duty, you're free to help and well protected when you do.

Consent

For a person who can answer, you ask. The American Red Cross's three steps, Check, Call, Care, begin: "First, CHECK the scene for safety. Keeping yourself safe will help you take care of others. Then, obtain consent and check the person to determine the nature of their illness or injury."²² In practice that is one sentence, and the wording is this course's, not the Red Cross's: tell them your name and ask if you can help.

For a person who cannot answer, you cannot ask, and the law does not expect you to. For the UK, RCUK's booklet names the defences as "implied consent" (the assumption that the person would agree if they could) and "necessity" ("that the treatment is given in the best interests of the patient"). It adds a qualifier that matters to you: "the defence of implied consent may not be as clear-cut if the rescuer isn't medically qualified", but "the defence of necessity may be available, provided the rescuer acts reasonably in the circumstances", and it gives CPR and an AED when no health professional is there as an example of acting reasonably. In England and Wales the Mental Capacity Act 2005 "may also add weight", for people aged 16 and over. In Scotland the booklet says an emergency falls under the same common-law defences, since the procedure in the Adults with Incapacity (Scotland) Act 2000 is unlikely to be practical in a resuscitation. For Northern Ireland, the booklet was written before that country's own 2016 Act came into force and says common-law necessity applies.¹⁵ The American Red Cross's long course lists "implied consent" in its glossary too, though this course did not read its definition there, and it did not survey US consent law.

One narrower point, because a distressed relative or a stranger may raise it. The booklet says, in a section on legally binding advance decisions in England and Wales, that it's very unlikely a rescuer in an emergency would know of one. If it did become known that a formal decision refusing treatment, or a DNACPR (do not attempt CPR) recommendation, "were in place, then it should be respected".¹⁵ The booklet says nothing about what to do when you're unsure. This course's own reading, drawn from that passage and not from any guideline, is that being told "he has a DNR" by someone at the scene is not the same as the decision being known to be in place, so unless you have been shown it, carry on as you would for anyone else: call, and start CPR if it's needed, and tell the call handler what you have been told. Lesson 16 returns to this.

What if a responsive adult says no? The sources this course read go no further than "obtain consent". The course's own reading is that consent covers touching and treating them, so if they refuse, don't. Calling the emergency number and staying nearby is a different act, and it does not touch anyone. And if they then stop responding, they're a person who can't answer, and what covers helping someone who can't answer applies.

"I'll make it worse"

This fear has two versions, and they have different answers.

The first is that you will hurt someone who is in cardiac arrest, by breaking ribs. That can happen, and lesson 3 gives the numbers and why they do not mean you did it wrong. The answer to the fear is the booklet's: without CPR, a person in cardiac arrest dies. RCUK's 2025 guideline says it without softening: "No greater harm can occur than failing to act when someone requires CPR and defibrillation."[23]

The second version is sharper: what if they were not in arrest at all, and I pushed on the chest of someone whose heart was beating? Studies of exactly that case, people given compressions who turned out not to be in arrest, found injuries in about 2% or fewer, and lesson 3 goes through them.[24] The AHA's 2025 guideline weighs the two risks directly: "The benefit of providing CPR for someone in cardiac arrest far outweighs any risk associated with providing chest compressions to someone who is not."⁶

And the benefit is real. RCUK's guideline says: "Bystander CPR and use of an AED increase the chances of survival by two to four-fold."[23] Lesson 4 looks at how that estimate is made and how far it can be trusted.

"I'm not qualified"

A certificate is a real thing with a real job, and the job is employment. In the US, the rule from the Occupational Safety and Health Administration (OSHA) for workplaces without a clinic near by says that "a person or persons shall be adequately trained to render first aid".[25] OSHA does not name a course; training providers sell "OSHA-compliant" cards, and the American Red Cross describes its own as "a two-year certification".[26] In the UK, HSE's guidance runs an Emergency First Aid at Work course of one day and a First Aid at Work course of three, and "All first-aid training certificates, whether FAW, EFAW or some other appropriate training, are valid for three years."⁴

None of that decides who may help. The AHA's 2025 guideline does not limit compressions to the certified: "All lay rescuers should provide chest compressions for adults with presumed cardiac arrest." (AHA 2025, COR 1).⁶ RCUK's law booklet says an untrained person "would be justified in using one [an AED] in an emergency when a more qualified person is not available".¹⁵

Learning without a classroom is possible, within limits. ILCOR's 2025 review of training suggests either instructor-led or self-directed digital training for lay adults (ILCOR 2025, weak recommendation, very low certainty), and suggests "self-directed digital training be used when instructor-led training is not accessible, or when quantity over quality of CPR training is needed". It also says "There was insufficient evidence to make a recommendation on game-in-film, virtual reality, computer programs, online tutorials or app-based training", and its justification carries the caution that training's benefits in simulation "may not translate to real-life situations".[27] So a free online course is a real start, on thin evidence. It is not the same as having your hands corrected by an instructor, which is why the box at the top of every lesson says what it says.

The bus stop

Now the harder case. A woman in her fifties collapses at a bus stop on a busy street. Six people are nearby. You're one of them. Nobody moves.

Once you have checked it's safe to reach her (lesson 2 starts there, with traffic first), take the worries in the order they arrive.

"Someone else will." Fischer's review says each of the six is somewhat less likely to act because the others are there. Philpot's footage says someone usually steps in, but that was fights. Neither promises anyone acts in the next minute, and in cardiac arrest the next minute matters. Fischer's authors say the effect's shrinking in dangerous emergencies fits a model in which those are recognised faster and more clearly, so, on this course's reading of that, the first useful thing is to say out loud what's happening: she's collapsed and she needs help.

"I'll be sued." In the US, if this is California, the statute above covers you: you're acting in good faith, unpaid, and doing your honest best is nowhere near gross negligence. In England, no reported claim against a helper has succeeded. Across the US, suits claiming the CPR itself did harm turned up three times in thirty years. (General information, not legal advice.)

"I'll hurt her." If she's in arrest, she dies without help. If she is not, the injury rate from compressions in studies of that case was about 2% or less.

That leaves the bystander effect itself, and what you can do about it. The guidelines' own sequences hand tasks to other people: St John Ambulance says "Ask a helper to find and bring a defibrillator (AED)", the American Red Cross has you call 9-1-1 and get equipment "or tell someone to do so", and the AHA notes that "a second lay rescuer can be instructed to call 911."[28][26]⁶ So don't ask the crowd whether someone could help. Point at one person: "You in the grey coat, call 999 and put it on speaker" (911 in the US). Point at another: "You, find a defibrillator." A named job makes it unmistakable who is doing what. That this beats the bystander effect is the course's own inference, from the guidelines' habit of handing out tasks and from the model Fischer's authors cite, in which clearer emergencies draw more help. Nobody the course read has tested the tactic itself.

What people get wrong

"Someone else will help." Each person is less likely to act in a crowd, and the footage showing someone usually does was of fights. Don't wait to find out: give people jobs.

"38 people watched and did nothing." The New York Times itself called its 1964 portrayal "erroneous".

"You'll be sued." Suits claiming CPR itself did harm: three in thirty years of US court records. In the UK, no reported successful claim against a helper. (General information, not legal advice.)

"You need a certificate to help." Certificates are for employers, and the AHA says all lay rescuers should give compressions.

"You might make it worse." In cardiac arrest, RCUK says that is "virtually impossible".

"In the UK you must help by law." There's no general duty to rescue in the UK. The duties this lesson describes are three US states', and none asks you into danger.

Practice

Where to learn with your hands

Lesson 16 has the full list of classes, with what each option does and does not give you. Two free things you can start now, at home: the BHF's RevivR, which takes "just 15 minutes" and uses a cushion and your phone's camera to give feedback on your compressions, and RCUK's Lifesaver, an interactive film.[29] Neither is a hands-on class with an instructor, and neither replaces one.

Connections

Lesson 2 picks up at the kitchen floor: checking for danger, the call and what the call handler asks, and how to recognise cardiac arrest, including the gasping that fools people. The bus-stop jobs come back there with their exact words. Lesson 4 teaches the labels properly and reports who receives bystander CPR. Lesson 5 has the law on defibrillators, and lesson 16 comes back to certificates, classes and how guidelines change.

From earlier on the Core, one link was not made in the body: Logic and Argument would spot the unstated premise in "someone else will help", that someone else knows what to do, and will do it now.

Go deeper

• Resuscitation Council UK, CPR, AEDs and the law, 2018, free. Pages 1 to 13 read closely here, the rest skimmed. A clear statement of the UK position on helping, consent and liability, in plain English.
• Philpot and colleagues, postprint of "Would I be helped?", American Psychologist, 2020, free. Read here in full. Good on how a study of real behaviour is built from footage, and on its own limits.
• California Health and Safety Code §1799.102 and Minnesota Statutes §604A.01, free. Read here in full. Two short statutes that show, side by side, a law protecting helpers and a law requiring help.

Sources

1. American Heart Association, "CPR Facts and Stats", Internet Archive capture 20260919101819 of cpr.heart.org/en/resources/cpr-facts-and-stats. Read: Wayback, direct. Location of out-of-hospital cardiac arrests.
2. Resuscitation Council UK, 2025 Resuscitation Guidelines: Epidemiology of cardiac arrest, published 27 October 2025. Read: full, direct. The share of UK arrests at home.
3. American Heart Association, "Heartsaver First Aid CPR AED Course Options", Internet Archive capture 20260422003426 of cpr.heart.org. Read: Wayback, direct. The hands-on requirement.
4. Health and Safety Executive, First aid at work, L74, 3rd edition 2013, amended 2018 and 2024. Read: full, direct (PDF). Paragraph 62 (direct observation) and paragraph 74 (three-year certificates); course lengths from HSE's first aid at work FAQs, read: full, direct.
5. B. J. Bobrow and colleagues, "The effectiveness of ultrabrief and brief educational videos for training lay responders in hands-only CPR", Circulation: Cardiovascular Quality and Outcomes 4(2), 2011, pp. 220 to 226. A randomised trial, n = 336. Read: abstract only.
6. M. E. Kleinman and colleagues, "Part 7: Adult Basic Life Support: 2025 AHA Guidelines for CPR and ECC", Circulation 152 (suppl 2), 2025, doi 10.1161/CIR.0000000000001369. Internet Archive capture 20251026225858 of the AHA's web version. Read: Wayback, full, direct. Recommendation 1 of the lay rescuer initiation table (COR 1, LOE B-NR), and the supporting text on the benefit of CPR outweighing the risk and on instructing a second rescuer to call.
7. S. J. Coons and M. C. Guy, "Performing bystander CPR for sudden cardiac arrest: behavioral intentions among the general adult population in Arizona", Resuscitation 80(3), 2009, pp. 334 to 340. Read: abstract only.
8. R. D. McFadden, "Winston Moseley, 81, Killer of Kitty Genovese, Dies in Prison", New York Times, 4 April 2016, Internet Archive capture 20161212182133. Read: Wayback, direct.
9. R. Manning, M. Levine and A. Collins, "The Kitty Genovese murder and the social psychology of helping: the parable of the 38 witnesses", American Psychologist 62(6), 2007, pp. 555 to 562, doi 10.1037/0003-066X.62.6.555. Read: abstract only.
10. P. Fischer and colleagues, "The bystander-effect: a meta-analytic review on bystander intervention in dangerous and non-dangerous emergencies", Psychological Bulletin 137(4), 2011, pp. 517 to 537, doi 10.1037/a0023304. Read: abstract only.
11. R. Philpot and colleagues, "Would I be helped? Cross-national CCTV footage shows that intervention is the norm in public conflicts", American Psychologist 75(1), 2020, pp. 66 to 75, doi 10.1037/amp0000469. Read: full, direct (postprint).
12. British Heart Foundation, 2025 Heart Month press release (YouGov) and 2024 Heart Month press release (Censuswide). Read: full, direct. Charity-commissioned polls, treated as indicative.
13. I. Uny, K. Angus, E. Duncan and F. Dobbie, "Barriers and facilitators to delivering bystander CPR in deprived communities: a systematic review", Perspectives in Public Health 143(1), 2023, pp. 43 to 53. Read: abstract only.
14. S. M. Perman and colleagues, "Public perceptions on why women receive less bystander CPR than men in out-of-hospital cardiac arrest", Circulation 139(8), 2019, pp. 1060 to 1068. Read: abstract only.
15. Resuscitation Council UK, Cardiopulmonary resuscitation, automated defibrillators and the law, April 2018 (first published 2000; reviewed 2017). Read: full, direct (pp. 1 to 13 closely; remainder skimmed).
16. T. W. Murphy and colleagues, "Risk and ROSC: legal implications of bystander CPR", Resuscitation 151, 2020, pp. 99 to 102, doi 10.1016/j.resuscitation.2020.03.017. Read: abstract only. The full paper was not obtained, so the lesson gives the abstract's counts and no dollar figures, and says nothing about who the defendants were, which the abstract does not report.
17. California Health and Safety Code §1799.102, as amended 2009. Read: full, direct.
18. Vermont, 12 V.S.A. §519; Minnesota, Minn. Stat. §604A.01; Rhode Island, R.I. Gen. Laws §11-56-1. Read: full, direct.
19. Resuscitation Council UK, FAQs: Basic Life Support (CPR). Read: full, direct. Quoted to show its disagreement with the booklet.
20. Social Action, Responsibility and Heroism Act 2015, c. 3, as enacted. Read: full, direct (ss. 1 to 5).
21. European Resuscitation Council, The ERC Guidelines 2025 on Resuscitation for Everyone, December
    2025. Read: the ethics, education and systems sections, direct.
22. American Red Cross, "Check, Call, Care: 3 Emergency Actions Steps". Read: full, direct.
23. Resuscitation Council UK, 2025 Resuscitation Guidelines: Adult basic life support, published 27 October 2025. Read: full, direct.
24. L. White and colleagues, "Dispatcher-assisted CPR: risks for patients not in cardiac arrest", Circulation 121, 2010, pp. 91 to 97; K. B. Haley and colleagues, "Frequency and consequences of bystander CPR on patients not in arrest", Prehospital Emergency Care 15, 2011, pp. 282 to 287; F. Williamson and colleagues, "Does delivering chest compressions to patients not in cardiac arrest cause unintentional injury? A systematic review", Resuscitation Plus 20, 2024, 100828. Read: abstracts only. Taught in lesson 3.
25. US Occupational Safety and Health Administration, 29 CFR 1910.151, paragraph (b). Read: full, direct.
26. American Red Cross, "First Aid Steps". Read: full, direct.
27. R. Greif and colleagues, "Education, Implementation, and Teams: 2025 ILCOR CoSTR", Circulation 152 (suppl 1), 2025, pp. S205 to S249, doi 10.1161/CIR.0000000000001359. Internet Archive capture 20260430180745. Read: Wayback, direct (recommendation sections read; evidence tables skimmed). Recommendation EIT 6406.
28. St John Ambulance, "How to do the primary survey (DR ABC)", clinically reviewed 28 April 2025. Read: full, direct.
29. British Heart Foundation, RevivR, read: full, direct; Resuscitation Council UK, Lifesaver page, read: full, direct.
30. British Heart Foundation, "How to do CPR", undated. Read: full, direct. What happens in cardiac arrest.