Objective-To audit the care that had been provided to couples before t
he birth of a child with cystic fibrosis where a sibling had been prev
iously diagnosed. Design-Retrospective review of case notes. Sample-Fa
milies where at least one affected child had been born between 1 Janua
ry 1991 and 30 June 1995 and the diagnosis in the first child was made
before the second affected pregnancy reached 20 weeks. The combinatio
n of information on these families with data from the prenatal diagnos
is register allowed the reconstruction of a cohort of pregnancies in w
omen with a previous affected child. Main results-Forty six eligible f
amilies with a second affected child were identified. Details from the
paediatrician who had diagnosed the first affected child were obtaine
d in 43 cases: all 43 couples were offered genetic counselling, but wh
ere provided by a paediatrician this was difficult to assess as no cou
ple was sent a summary letter. Details were obtained from the obstetri
cian in the subsequent affected pregnancy in 42 cases: prenatal diagno
sis was not offered in 10 (24%), offered and declined in 24 (57%), off
ered and accepted but termination declined in eight (19%). In the over
all cohort of at risk pregnancies, the estimated rate of prenatal diag
nosis offer was 97%, prenatal diagnosis uptake 86%, false negative pre
natal diagnosis rate 0%, and uptake of termination 95%. Conclusions-(1
) Parental choice was an important determinant of second affected birt
hs. (2) Despite widespread availability, prenatal diagnosis was not of
fered in an estimated 3% of at risk pregnancies. (3) There were shortc
omings in counselling documentation, in particular failure to send a s
ummary letter to counselled couples.