Methods for handling missing data in serially sampled sputum specimens for mycobacterial culture conversion calculation

Table 1 Hypothetical example of missing data patterns and influence on estimated conversion week

+=positive culture, − = negative culture, X = culture result missing, *=conversion week
^a Culture conversion estimate is delayed to week 10 from missing cultures for weeks 7–9
^b Culture conversion is not achieved with last observed culture as negative
^c Culture conversion is not achieved with last observed culture as positive
^d Missing cultures occur after conversion and do not affect estimated conversion week
^e Conversion week defined as first of two consecutive cultures without Mycobacterium tuberculosis present

ISSN: 1471-2288