The weaknesses of the testing programme were evident at Tran’s hospital in HCM City.
Since late 2003, 67 patients have been admitted to the bird flu isolation ward. All were initially identified as having the disease based on their symptoms and contact with sick birds. Later testing, however, indicated that only 13 actually had it. Then the Japanese tests showed that some of those who tested negative for the disease were indeed infected.
In response, Farrar has created a pilot project in clinics around HCM City to test patients with widely varying complaints, hoping to develop a more complete picture of the extent of infection.
The data on false negatives reported yesterday indicate only 1 in 5 H5N1 cases are detected with the PCR test used in southern Vietnam. Remarkably, the data cited above for the most likely cases, based on clinical symptoms and exposure to sick birds, indicate 1 in 5 tested registered positive. These data suggest that virtually all of the 67 patients with bird flu symptoms and sick bird exposure are H5N1 positive, and the clinical picture for these cases is far more accurate for identifying H5N1 infected patients than the lab testing.
If serum samples were kept on these 67 patients it may be possible to H5N1 confirm those who tested negative on the PCR test. It would also be useful to test the contacts and relatives of the 67 patients for H5N1 antibody. The same can be said for patients in the north, including the clusters in Thai Binh that extend to health care workers.
Understanding the number of patients with varying presentations would be useful. However, the extent of human-to-human transmission with or without clinical signs would also help control the spread of the virus.
The large number of false negatives is seriously interfering with the monitoring of H5N1 in contacts of patients who are likely to have been infected. Without such human-to-human transmission data in the most obvious cases, control effects cannot be accurately planned or predicted.