Satnam Kaur1, V Singh, A K Dutta, J Chandra. 1. Department of Pediatrics, Kalawati Saran Childrens Hospital and Lady Hardinge Medical College, New Delhi, India.
Abstract
OBJECTIVE: To check the validity of Integrated Management of Neonatal and Childhood Illness (IMNCI) algorithm for young infants (0-2 months). DESIGN: Prospective observational study. SETTING: The outpatient department and emergency room of a medical college attached hospital. METHODS: 419 infants (176 between 0-7 days, 243 between 7 days to 2 months) underwent a detailed diagnostic assessment and treatment as per the standard protocol of treating unit. These infants also underwent assessment, classification and identification of treatment as per IMNCI algorithm. The diagnostic and therapeutic agreement between standard protocol and IMNCI was computed to assess the validity of IMNCI algorithm. RESULTS: The IMNCI algorithm performed well in identifying sick young infants with sensitivity of 97%, 94% and 95%, and specificity of 85%, 87% and 87% in 0-7 days, 7 days to 2 months and 0-2 months age groups, respectively. The algorithm covered majority (80%) of recorded diagnoses, and could identify bacterial infection with 88.5% sensitivity and 57.4% specificity. Complete diagnostic agreement with gold standard was seen in 50%; overdiagnosis and under diagnosis was seen in 13% and 19%, respectively. Low birthweight and upper respiratory infection were the main reasons for overdiagnosis whereas surgical conditions resulted in under diagnoses in majority. CONCLUSION: IMNCI algorithm for evaluation and management of young infants has good sensitivity and specificity for referring cases with severe illness.
OBJECTIVE: To check the validity of Integrated Management of Neonatal and Childhood Illness (IMNCI) algorithm for young infants (0-2 months). DESIGN: Prospective observational study. SETTING: The outpatient department and emergency room of a medical college attached hospital. METHODS: 419 infants (176 between 0-7 days, 243 between 7 days to 2 months) underwent a detailed diagnostic assessment and treatment as per the standard protocol of treating unit. These infants also underwent assessment, classification and identification of treatment as per IMNCI algorithm. The diagnostic and therapeutic agreement between standard protocol and IMNCI was computed to assess the validity of IMNCI algorithm. RESULTS: The IMNCI algorithm performed well in identifying sick young infants with sensitivity of 97%, 94% and 95%, and specificity of 85%, 87% and 87% in 0-7 days, 7 days to 2 months and 0-2 months age groups, respectively. The algorithm covered majority (80%) of recorded diagnoses, and could identify bacterial infection with 88.5% sensitivity and 57.4% specificity. Complete diagnostic agreement with gold standard was seen in 50%; overdiagnosis and under diagnosis was seen in 13% and 19%, respectively. Low birthweight and upper respiratory infection were the main reasons for overdiagnosis whereas surgical conditions resulted in under diagnoses in majority. CONCLUSION: IMNCI algorithm for evaluation and management of young infants has good sensitivity and specificity for referring cases with severe illness.
Authors: Anne C C Lee; Aruna Chandran; Hadley K Herbert; Naoko Kozuki; Perry Markell; Rashed Shah; Harry Campbell; Igor Rudan; Abdullah H Baqui Journal: PLoS Med Date: 2014-10-14 Impact factor: 11.069