作者: Llorenc Quinto , John J. Aponte , Clara Menendez , Jahit Sacarlal , Pedro Aide
DOI: 10.1111/J.1365-3156.2006.01679.X
关键词: Cross-sectional study 、 Linear regression 、 Anemia 、 Age distribution 、 Hematocrit 、 Internal medicine 、 Rural health care 、 Medical electronics 、 Hemoglobin 、 Medicine
摘要: Summary Introduction Anaemia is the most frequent haematological disorder in childhood. The notion that defines naemia does not change throughout life, although parameters used for its evaluation show significant variations during Haematocrit (Hct) (%) usually defined as three times value of haemoglobin (Hgb) (g/dl), while clinical definition anaemia related to either an abnormal Hct or Hgb value. Objective To evaluate agreement between and values anaemia, relationship these two their age-dependence. Methods paired from children aged 2–18 months Ifakara (Tanzania) 1–4 years Manhica (Mozambique) were analysed. Haematological determinations samples done using a KX-21N cell counter (Kobe, Japan) analysed semiautomatic (Sysmex F800 microcell counter, TOA Medical Electronics, Kobe, Japan). κ-statistic was calculate definitions each group. Crude multivariate levels by linear regression model estimation. age-dependence crude ratio (Hct/Hgb) models fractional polynomials. Results prevalences mild moderate group 61% 6%, respectively, 41% 2% Hct. In 74% 10%, 42% 3% Hct, respectively. Agreement made up fair moderate. decreased with age high levels, whereas they increased low levels. classification cases improved when higher age-related cut-off are used. significantly different 3, this modified age. showed non-linear asymptotic trend 3. Conclusions Measurement haematocrit count easy can be performed rural health care centres. However, corresponding cannot derived acceptable accuracy 3 conversion factor. Furthermore, commonly assumed ‘equivalent’ points need re-evaluated.