According to this study:

* Providing instruction on sputum sample submission significantly improved tuberculosis detection in women in low-income countries.

Women in developing countries with suspected tuberculosis are less likely than men to test positive for tuberculosis on a sputum smear test. Testing requires the patient to submit a sputum sample one day and an early-morning sample the following day. This study examined whether teaching patients how to provide good-quality sputum samples increased the rate of "smear-positive case detection" in women.

Patients suspected of having tuberculosis were enrolled over a three month period in 2005 at an outpatient tuberculosis hospital in Pakistan; 1,494 women and 1,561 men were randomized. A female health care worker instructed those in the intervention group on how to produce a good-quality sample of sputum-as opposed to saliva-for the test. She also told the intervention group how much sputum the test required and that it was important to bring a second sample of sputum, which was to be expectorated after waking the following morning. These instructions took about two minutes. Patients in the control group received no instruction, as was standard protocol.

Smear tests were positive at a significantly higher rate for women in the intervention group than in the control group. Women who received instruction were more likely to submit good-quality sputum samples, and significantly more of them (88%) submitted a second sample than did women in the control group (84%).

Men who received the instruction intervention also had greater test positivity and improved sample quality than those in the control group, but the differences between the groups were not significant.