Ten years after they were introduced in Europe and North America, HIV/AIDS drugs are still effective but many patients are not being put on them soon enough, scientists said on Friday.

Experts had feared the AIDS virus would become resistant to the treatments and deaths would increase but research published ahead of an international AIDS conference in Toronto later this month has shown it has not happened.

The combinations of drugs reduce mortality and progression to AIDS by about 80-90 percent but tuberculosis (TB) has become a deadly co-infection in patients.

“Ten years on these treatments still work as well as they did initially (but) there is a change in terms of TB becoming more important,” said Professor Matthias Egger, of the University of Bern in Switzerland, said in an interview.

He said people should be diagnosed and start treatment earlier.

“If that would happen these drugs would even achieve more.”

The findings of the Antiretroviral Therapy Cohort Collaboration reported in the Lancet medical journal are based on an analysis of data on more than 22,200 HIV positive people in Europe and North America who started treatment between 1995 and 2003.

More than 40 million people worldwide are living with HIV/AIDS, the majority in sub-Saharan Africa. About 1.3 million people in low- and middle-income countries were receiving AIDS drugs by the end of 2005, according to the World Health Organization.

Although it varies, Egger, a co-author of the study, said there was widespread consensus that patients should start treatment when their CD4 cell count, a measure of immune system response, dropped below 350 or if the person was unwell or showed symptoms of illness.

The median CD4 cell count when starting treatment increased from 170 in 1995-1996 to 269 in 1998 but then dropped to around 200, according to the research.

People who start treatment with a CD4 count less than 200 have a higher risk of their illness progressing and of dying of AIDS than patients with a higher baseline count, the scientists added in the study.