AMC MCQ Ethics and Population Health: Consent to Biostatistics

AMC ethics questions look easy until you sit them. The stem is short, every option sounds reasonable, and the best answer depends on Australian law and the Medical Board's code of conduct rather than on what you learnt at home. Add a population health AMC MCQ that asks for a positive predictive value in 84 seconds, and this 12.5% of the exam suddenly feels much bigger.

This guide sets out the rules the AMC expects you to apply. It then works through the biostatistics and screening principles step by step, and every calculation has been checked.

How AMC ethics questions and population health items are examined

The AMC's MCQ specifications describe this patient group in four parts:

• Population health: strategies that improve the health of targeted populations, including screening.
• Ethics: professional behaviour, including boundaries, confidentiality and informed consent.
• Law: the regulations that govern Australian health care.
• Across settings: population health concepts can come up in any clinical setting.

Most stems end with "What is the most appropriate action?" and two or three of the options are usually lawful. The best answer protects the patient, respects their autonomy and takes the least drastic lawful path.

Consent: the rules behind most AMC ethics questions

Good medical practice: a code of conduct for doctors in Australia requires informed consent before any examination, investigation or treatment, except in an emergency. It also requires consent before patients are involved in teaching or research. You must give the information in a form the patient can understand, and that includes your fees.

Valid consent has four elements: capacity, voluntariness, information and specificity. Specificity means consent to a diagnostic laparoscopy does not cover a hysterectomy you find convenient to do at the same time.

The Australian standard for what to disclose comes from Rogers v Whitaker (1992). A risk is material if a reasonable person in the patient's position would attach significance to it, or if you know this particular patient would.

Capacity and substitute decisions

Capacity is presumed, and it applies to one decision at one time. The patient must be able to:

• understand the relevant information
• retain it
• weigh it
• communicate a choice.

An unwise decision does not prove incapacity. If capacity fluctuates, for example in delirium or hypoglycaemia, treat the cause and reassess.

If the patient lacks capacity, consent comes from a substitute decision-maker, who must take any advance care directive into account. Who that person is depends on state guardianship law. It is usually an appointed guardian, then a spouse or partner, a carer, or a close relative or friend. In a genuine emergency with no directive and no substitute available, give only the treatment needed to save life or prevent serious harm.

Minors: Gillick competence and the mature minor in Australia

In Marion's case (1992), the High Court adopted the Gillick principle. A minor can consent once they have enough understanding and intelligence to fully understand what is proposed. The code asks you to put the young person's interests first, consider their capacity, and encourage them to involve their parents where appropriate.

• South Australia: under the Consent to Medical Treatment and Palliative Care Act 1995, a person aged 16 or over can make decisions about their own treatment as an adult. A child under 16 can consent if you believe they understand the nature, consequences and risks, and the treatment is in their best interests. Another doctor who has examined the child must support that opinion in writing.
• New South Wales: the NSW Health consent manual sets out three bands.
  • At 16–17, the young person's own consent is usually enough.
  • At 14–15, you should also seek a parent's consent unless the young person objects.
  • At 13 and under, you need a parent's consent.
• Refusal: a court can override a minor's refusal of life-saving treatment. Being competent to consent does not always mean being competent to refuse.
• Emergencies: state law lets you treat a child without consent to save their life or prevent serious damage to their health. That includes giving blood when the parents object.
• Procedures outside parental power: some procedures, such as non-therapeutic sterilisation, need court or tribunal approval. That was the issue in Marion's case.

Confidentiality and its exceptions

The code says you must treat patient information as confidential unless release is required or permitted by law. You should access a record only when you have a legitimate need, so looking up a neighbour's record out of curiosity is a breach.

The serious-threat exception comes from the Privacy Act. You must reasonably believe the disclosure is necessary to lessen or prevent a serious threat to life, health or safety, and getting consent must be unreasonable or impracticable. The threat does not have to be imminent, but disclosure that is merely helpful or convenient is not allowed.

Parents of adult children and spouses have no automatic right to a patient's information. You can always listen to their concerns, but you can't disclose anything without the patient's consent.

Use qualified interpreters rather than family. Private practitioners can use TIS National's free service for Medicare-rebateable care by calling the priority line on 1300 131 450.

Under Austroads' Assessing Fitness to Drive standards, the driver is responsible for reporting a condition to the licensing authority. After a first seizure, your duty is to advise the patient not to drive, explain the standard and document your advice. Consider notifying the authority only if they keep driving against your advice.

Mandatory reporting: children at risk and impaired practitioners

Child abuse and neglect

Doctors are mandatory reporters in every state and territory, and in the Northern Territory the duty applies to everyone. The legal threshold differs by jurisdiction:

• Reasonable suspicion: NSW, Queensland and South Australia.
• Belief on reasonable grounds: the ACT, the Northern Territory, Victoria and Western Australia.
• Either: Tasmania.

No jurisdiction requires you to be certain. The types of abuse covered also vary. Victoria covers physical injury and sexual abuse, while NSW and Tasmania also cover neglect, emotional harm and exposure to family violence.

Ahpra mandatory notifications about practitioners

Since 1 March 2020 the rules have been designed so that practitioners can seek help for their own health without fear. Ahpra notes that a health issue rarely needs a mandatory notification.

You need a reasonable belief before you notify, not rumour, and you must notify as soon as practicable. Treating practitioners in Western Australia are exempt.

Boundaries, open disclosure and end-of-life care

Professional boundaries

Never use your professional position to pursue a sexual or exploitative relationship with a patient.

• Avoid treating family members or close friends.
• Decline any gift from industry that is more than trivial in value.
• You may hold a conscientious objection, but you must tell the patient and must not use it to block their access to lawful care, such as contraception.
• Culturally safe care for Aboriginal and Torres Strait Islander patients is part of good medical practice.

Open disclosure

The Australian Open Disclosure Framework applies when an adverse event harms a patient. It has five elements:

1. an apology or expression of regret that includes the words "I am sorry" or "we are sorry"
2. a factual explanation of what happened
3. an opportunity for the patient, their family and carers to relate their experience
4. a discussion of the possible consequences
5. the steps being taken to manage the event and prevent it happening again.

Every Australian jurisdiction has laws that protect apologies in certain legal settings. Disclosure starts early with the facts you know. Don't wait for the incident review.

End of life

The code says you have no duty to prolong life at all cost, but you must relieve distress. Patients may refuse treatment or ask for it to be withdrawn.

• Advance care directives: follow a valid directive that clearly applies, even if the family disagrees.
• Withholding and withdrawing treatment: these are ethically equivalent.
• Double effect: opioids titrated to relieve symptoms are lawful even if they may shorten life.
• Voluntary assisted dying: each state and territory has its own Act with strict criteria. The ACT's Act, for example, requires the person to be 18 or older and to have decision-making capacity at every stage, and mental illness alone does not qualify.

Population health AMC MCQ topics: study designs and key terms

• Prevalence ≈ incidence × duration. So a treatment that prolongs life without curing the disease raises prevalence.
• Confounding: a third factor linked to both the exposure and the outcome. You control it with randomisation, matching, stratification or multivariable analysis. A bigger sample doesn't fix bias.
• Type I error (α): a false positive, usually set at 0.05.
• Type II error (β): a false negative. Power = 1 − β, often 80%.
• 95% confidence intervals: a ratio is not significant if its interval includes 1, and a difference is not significant if its interval includes 0.

Biostatistics with worked calculations

Draw the 2×2 table the same way every time, with disease status across the top and the test result down the side. That gives you true positives (a), false positives (b), false negatives (c) and true negatives (d).

• Sensitivity = a ÷ (a + c). A sensitive test, when negative, rules the disease out (SnNout).
• Specificity = d ÷ (b + d). A specific test, when positive, rules the disease in (SpPin).
• PPV = a ÷ (a + b).
• NPV = d ÷ (c + d).
• LR+ = sensitivity ÷ (1 − specificity).
• LR− = (1 − sensitivity) ÷ specificity.

Worked example: prevalence changes PPV

A test with 90% sensitivity and 80% specificity is used on 1,000 people, and the prevalence is 10%.

• Of the 100 people with the disease, 90 test positive (true positives) and 10 test negative (false negatives).
• Of the 900 people without the disease, 720 test negative (true negatives) and 180 test positive (false positives).
• PPV = 90 ÷ 270 = 33.3%.
• NPV = 720 ÷ 730 = 98.6%.

Now use the same test at 1% prevalence. You get 9 true positives and 198 false positives, so the PPV falls to 9 ÷ 207 = 4.3%, and the NPV rises to 99.9%. The likelihood ratios stay the same at both prevalences: LR+ = 0.90 ÷ 0.20 = 4.5 and LR− = 0.10 ÷ 0.80 = 0.125.

Risk, NNT and NNH

Suppose a stroke occurs in 10% of controls (the control event rate, CER) and 6% of treated patients (the experimental event rate, EER).

• RR = 0.06 ÷ 0.10 = 0.6
• ARR = 10% − 6% = 4%
• RRR = 0.04 ÷ 0.10 = 40%
• NNT = 1 ÷ 0.04 = 25

If bleeding occurs in 4% of treated patients and 2% of controls, then NNH = 1 ÷ 0.02 = 50. Round an NNT up to the next whole person.

In a case–control study, suppose 60 of 100 cases and 30 of 100 controls were exposed. Then OR = (60 × 70) ÷ (40 × 30) = 3.5. You can't calculate a relative risk from a case–control design.

Screening principles: Wilson–Jungner and Australian programs

The Wilson and Jungner criteria (WHO, 1968) set out when screening is justified:

1. The condition is an important health problem.
2. There is an accepted treatment.
3. Facilities for diagnosis and treatment are available.
4. There is a recognisable latent or early stage.
5. A suitable test exists.
6. The test is acceptable to the population.
7. The natural history of the condition is understood.
8. There is an agreed policy on whom to treat.
9. The cost is economically balanced.
10. Case-finding is a continuing process.

Australia's Population Based Screening Framework adapts these criteria and adds two more: high-quality evidence that benefits outweigh harms, and program requirements such as a register, informed choice and equity of access.

Watch for the biases that flatter screening:

• Lead-time bias: survival looks longer only because the diagnosis came earlier.
• Length-time bias: screening picks up more slow-growing disease.
• Overdiagnosis: screening finds disease that would never have caused harm.

The proof that a screening program works is a fall in disease-specific mortality.

Australia has no national prostate screening program. If an asymptomatic man asks for a PSA test, discuss the benefits and harms with him before testing.

Common traps in AMC ethics questions, and how to practise

• Applying another country's law, such as the UK Mental Capacity Act, or assuming the family can consent for an adult.
• Breaking confidentiality before you have talked to the patient.
• Treating a refusal you disagree with as evidence of incapacity.
• Notifying Ahpra about a colleague who is in treatment and has already changed how they practise.
• Using the relative risk reduction to calculate an NNT.

Because our bank is blueprint-weighted, about one in eight questions is population health and ethics. Every explanation covers all five options.

• Do short timed blocks at 84 seconds per question in the question bank.
• Write a one-line rule for each mistake.
• Sketch a 2×2 table on the exam whiteboard every time you get a test-accuracy question.

Pair this guide with:

• Australian guidelines for the AMC MCQ
• the population health and ethics MCQ hub
• AMC MCQ high yield topics
• our AMC MCQ study plan
• the AMC MCQ exam guide.