Coinfeciton with HIV and Mycobacterium tuberculosis is common in sub-Saharan Africa and each infection accelerates the progression of the other. Isoniazid prophylaxis is known to be effective in HIV-infected adults and has now been shown to be effective in children. In Cape Town (BMJ 2007;334:136–9; see also editorial, ibid: 105–6) a randomised trial of isoniazid against placebo for HIV-infected children taking co-trimoxazole was stopped early because of better results in the isoniazid group. A total of 263 children (median age 25 months) had been followed up for an average of 6 months. Mortality and tuberculosis incidence were both significantly lower in the isoniazid group (8% vs 16% and 4% vs 10% respectively). There were no culture-confirmed cases of tuberculosis in the isoniazid group.

In 2001 WHO and UNICEF agreed a plan to halve global measles mortality by the end of 2005, using 1999 as the baseline year. The plan depended on four strategies: high immunisation coverage (>90%) including a second dose when possible, case surveillance, monitoring of immunisation coverage, and appropriate clinical case management including provision of vitamin A. A recent report (Lancet 2007;369:191–200; see also Comment, ibid: 165–6) suggests that the aim has been achieved. The figures for measles mortality are based …