Aim-To describe a new inexpensive method (the WHO Colour Scale) for es
timating haemoglobin concentration from a drop of blood by means of a
colour scale, and to compare its reliability with a standard laborator
y method of measuring haemoglobin, and its clinical usefulness in fiel
d trials. Methods-The new colour scale method was used to measure haem
oglobin concentration in 1213 random venous blood samples from routine
work in four laboratories (one each in the UK, South Africa, Thailand
, and Switzerland). Limited field trials of the method for assessing c
linical usefulness were done in a rural hospital (in South Africa) sta
ffed by nurses, at two blood donor sessions (one each in South Africa
and Thailand), and by non-laboratory personnel in malaria clinics (in
Thailand), following training and a short practice session. Results-In
the laboratory based comparability study the presence of anaemia was
reliably detected using the new method with 91% sensitivity and 86% sp
ecificity. Clinically relevant levels of anaemia (mild to moderate, pr
onounced, and severe) were graded and serious anaemia (< 8 g/dl) was i
dentified with an efficiency of 89%. The clinical trials showed the ea
se and reliability with which the colour scale could be used by non-la
boratory persons after brief training. The blood donor trials showed i
t to be at least as reliable as the copper sulphate method with the ad
vantage of being more convenient. Conclusions-The preliminary studies
have shown that the WHO Colour Scale is a reliable screening method fo
r detecting anaemia, especially for diagnosing serious anaemia. Follow
ing a brief training session health workers found it simple to use and
, at a cost of about 1/10th that for traditional photometric analysis,
it should be of value in ''countries in need'' for primary health cen
tres, obstetrical management, paediatric clinics, tropical disease con
trol programmes, blood transfusion donor selection, as well as for ind
ustrial health checks and epidemiological studies.