Tc. Aitchison et al., PROGNOSTIC TREES TO AID PROGNOSIS IN PATIENTS WITH CUTANEOUS MALIGNANT-MELANOMA, BMJ. British medical journal, 311(7019), 1995, pp. 1536-1539
Objectives-To design user friendly guides to prognosis for patients wh
o have had invasive primary cutaneous malignant melanomas surgically e
xcised. Design--Adaptation of the classification tree method was used
to derive prognostic trees for four different subgroups of malignant m
elanoma patients in whom known and possible prognostic variables inter
acted in different ways. Setting-Scotland. Subjects-Statistical modell
ing for prognostic trees was based on 1978 patients whose primary mali
gnant melanoma was first diagnosed in 1979-86 for whom five year follo
w up and all relevant clinical pathological data were available, The r
esultant model was validated with 300 patients first diagnosed in 1987
for whom the same information was available. Main outcome measures--A
ctual and predicted rate of survival after diagnosis of primary cutane
ous malignant melanoma. Results--The four subgroups of patients were m
en and women with ulcerated and non-ulcerated cutaneous primary melano
mas. Validation of the model showed excellent agreement between actual
status of patients in the relevant subgroups and their status as pred
icted by the model. Conclusions--The prognostic trees are simple to us
e and give more accurate prognosis for individual patients than is cur
rently available from tumour thickness alone.