Two Carriers in Kyzylorda: Counselling After a Diagnosis
In Kyzylorda, a couple's first child, a boy, was diagnosed with phenylketonuria (PKU) through newborn screening at 10 days of age. PKU is an autosomal recessive disorder caused by deficiency of the enzyme phenylalanine hydroxylase; if a low-phenylalanine diet is started in the first weeks of life, intellectual disability is largely prevented.
Both parents are healthy and have no symptoms. Genetic testing confirms that both are carriers of a pathogenic variant in the PAH gene. They have one other child, a healthy 4-year-old daughter who has never been tested. The mother's cousin says: 'The first child was affected, so the next ones will probably be fine - and anyway, one in four means that out of four children exactly one will be ill.'
The couple wants more children and asks the paediatrician what the risks are and whether the daughter should be tested.
1. Explain why two healthy parents can have an affected child, and state the probabilities for the next pregnancy.
Read the worked answer
Why healthy parents can have an affected child. PKU is autosomal recessive. Each parent carries one pathogenic PAH variant and one working copy. With one working copy the enzyme level is sufficient, so neither parent shows symptoms - they are carriers, not patients.
The gametes. Each parent produces gametes carrying the pathogenic variant with probability 1/2 and the working copy with probability 1/2. Because the two parents' gametes combine independently, the four equally likely outcomes for a child are:
- variant from both parents → affected: probability 1/2 × 1/2 = 1/4 (25%);
- variant from one parent only → healthy carrier: 2 × (1/2 × 1/2) = 1/2 (50%);
- working copy from both → healthy non-carrier: 1/4 (25%).
The affected first child does not change these odds for a future pregnancy: each conception is an independent draw. What the affected child changed was the parents' information - before it, their carrier status was unknown; now it is certain, which is why the 25% figure can be quoted confidently.
A useful counselling point: 3 of 4 outcomes produce a healthy child, but the 50% carrier figure matters for the child's own future reproductive decisions.
2. The couple plans three more children. Calculate the probability that at least one of the three is affected, and correct the cousin's two statements.
Read the worked answer
Calculation. For one child, the probability of being unaffected is 3/4. For three independent children, the probability that all three are unaffected is (3/4)³ = 27/64 = 0.4219.
Therefore P(at least one affected among three) = 1 − 0.4219 = 0.578, about 58%.
That is the number that matters for family planning: with three further pregnancies, an affected child is more likely than not - a very different impression from the reassuring 'one in four'.
Correcting the cousin.
'The first was affected, so the next will probably be fine.' This is the gambler's fallacy. Conceptions are independent; the previous outcome has no memory. The risk stays at 25% for every pregnancy, whether the previous child was affected or not.
'One in four means exactly one ill child out of four.' A probability describes a distribution over many families, not a rule inside one family. For four children the number of affected children follows a binomial distribution: P(exactly one) = 4 × (1/4) × (3/4)³ ≈ 0.42; P(none) ≈ 0.32; P(two or more) ≈ 0.26. So 'exactly one in four' happens less than half the time. The 25% figure is a per-child risk, not a quota.
3. Should the healthy 4-year-old daughter be tested? Give the probability she is a carrier and a reasoned recommendation that distinguishes medical action from information.
Read the worked answer
Her probability. Both parents are confirmed carriers, so the daughter's genotype is one of the three unaffected outcomes, each with prior probability 1/4: carrier (1/4) or non-carrier (1/4). Given that she is unaffected, the total probability of 'unaffected' is 3/4, so:
P(carrier | unaffected) = (1/4) / (3/4) = 2/3 ≈ 67%.
What testing would and would not change. It would not change anything medically now: a carrier of PKU is healthy, needs no diet, and requires no monitoring - the newborn screening programme is designed to detect affected infants, not carriers, and that is why the daughter was never flagged. So there is no urgent clinical reason to test a 4-year-old.
Where the information does have value. At reproductive age, knowing her carrier status converts a 2/3 chance into certainty and allows her partner to be tested first. If he is not a carrier, the risk to her children is negligible; if he is, the couple knows it before pregnancy and can use prenatal or preimplantation options. Testing now versus at 18 makes no difference to the result, so the reasonable course is to defer it and to document the family history so she can be offered testing and counselling when it becomes relevant.
The judgement being assessed is the difference between a test that changes management and a test that only changes information - and the recognition that the second kind is legitimate but can wait for the person whose life it concerns to consent.