Evaluation of the Haemoglobin Colour Scale
Editor - The article by J.J. Paddle in this issue of the Bulletin suggests that a benevolent and inappropriate analysis of the performance of the Haemoglobin Colour Scale resulted in its unjustified promotion by WHO (1). However, as discussed below, the arguments to support this view are not corroborated by the evidence.
The focus in Paddle's paper is mainly on the inappropriateness of the coefficients of correlation as a performance indicator. However, in previous evaluations of the Colour Scale (25) the correlation coefficient was only one of the indicators considered in addition to sensitivity, specificity, positive predictive value, negative predictive value, precision within ±1 g/dl and ±2 g/dl haemoglobin, and Student's t-test on differences in paired results. This is reflected by the observation that the "innovative" analysis conducted by Paddle gives results that are compatible with those obtained by other studies (78% of readings within ±2 g/dl haemoglobin by the reference method).
The main difference between the view of Paddle and that of the WHO group involved in evaluating the Colour Scale is in the interpretation of such performance indicator results. It is widely recognized, and not a new observation, that a digital haemoglobinometer (HemoCue), is one of the more accurate tools to measure haemoglobin under field research conditions. However, the HemoCue is an expensive instrument each test uses a costly disposable cuvette and a power supply is needed. It is therefore, in our view, not appropriate for routine activities in the peripheral health services of least developed countries. Accordingly, WHO recommends clinical examination (i.e. pallor of mucosa, conjunctiva, and nail bed) as the routine approach for haemoglobin evaluation at peripheral level health centres (6). The Colour Scale is a clinical tool to supplement clinical examination and is not intended as an alternative to a digital haemoglobinometer, where such an instrument is available. This has been stated in several publications, but regrettably is not mentioned in the paper by Paddle (24).
Thus, the final paragraph: "The wide limits of agreement on the BlandAltman plot highlight the shortcomings of the Colour Scale, particularly in view of the good agreement demonstrated in the present study and elsewhere for devices such as the HemoCue" sounds like the suggestion made by Marie-Antoinette, Queen of France. Having heard from her advisers that the people of Paris were demonstrating because they had no bread, she asked: "Why don't they eat cake?" (7).
In conclusion, we believe that the accuracy of Haemoglobin Colour Scale is far superior to clinical assessment, currently the only alternative method to evaluate haemoglobin in least developed countries.
Conflicts of interest: none declared.
1. Paddle JJ. Evaluation of the Haemoglobin Colour Scale and comparison with the HemoCue haemoglobin assay. Bulletin of the World Health Organization 2002;80:813-6.
2. Montresor A, Albonico M, Khalfan N, Stoltzfus RJ, Tielsch JM, Chwaya HM, et al. Field trial of a haemoglobin colour scale: an effective tool to detect anaemia in preschool children. Tropical Medicine and International Health 2000;5: 129-35.
3. Ingram CF, Lewis SM. Clinical use of WHO haemoglobin colour scale: validation and critique. Journal of Clinical Pathology 2000;53: 33-7
4. Münster M, Lewis SM, Erasmus LK, Mendelow BV. Field evaluation of a novel haemoglobin measuring device designed for use in a rural setting. South African Medical Journal 1997;87:1522-6.
5. Van den Broek NR, Ntonya C, Mhango E, White SA. Diagnosing anaemia in pregnancy in rural clinics: assessing the potential of the Haemoglobin Colour Scale. Bulletin of the World Health Organization 1999;77:15-21.
6. WHO and UNICEF (1995). Management of childhood illness. Assess and classify sick child age 2 months up to 5 years. Geneva, World Health Organization. WHO Document WHO/CDR/95.14B. pp. 107-8.
7. Duby G. Histoire de la France, volume 2. Dynasties et révolutions, de 1348 à 1852. Paris: Larousse; 1987.
1 Medical Officer, Blood Safety and Clinical Technology, World Health Organization, Geneva, Switzerland.
2 Director, Blood Safety and Clinical Technology, World Health Organization, Geneva, Switzerland.
3 Senior Research Fellow, Department of Haematology, Imperial College, London, England.
4 Medical Officer, Strategy Development and Monitoring for Parasitic Diseases and Vector Control, World Health Organization, 1211 Geneva 27, Switzerland (email: firstname.lastname@example.org). Correspondence should be addressed to this author.
5 Coordinator, Addressing RTIs/STIs, Reproductive Health and Research, World Health Organization, Geneva, Switzerland.
6 Coordinator, Strategy Development and Monitoring for Parasitic Diseases and Vector Control, World Health Organization, Geneva, Switzerland.
7 Passagay, 74500 Saint Paul en Chablais, France.
Ref. No. 02-0320