LONDON (AP) -- An asthma treatment that targets the inflammation rather than just reacting to symptoms could be more effective in reducing severe attacks, new research suggests.

Conventional treatment, based on simple breathing tests and noting other symptoms, sometimes cannot head off an asthma attack.

Scientific advances may now provide an early warning. A study published this week in The Lancet medical journal indicates for the first time that looking for microscopic cells in phlegm led to fewer attacks and fewer admissions to the hospital.

Asthma, which is one of the most common chronic diseases worldwide, makes the sides of the airways in the lungs thick and swollen, causing recurring episodes of wheezing, breathlessness, chest tightness and coughing.

The World Health Organization estimates that asthma affects nearly 150 million people worldwide. Experts estimate that more than 17 million Americans have asthma. It is responsible for nearly 2 million emergency room visits a year there and kills about 5,000 people every year.

For many patients, medication must be taken every day to control symptoms and prevent attacks.

According to the new study, microscopic cells called eosinophils, which cause the inflammation in the airways that leads to asthma symptoms, can be found in phlegm several weeks before an asthma attack.

"This study shows that tailoring therapy to the degree of eosinophilic inflammation seems to be better than the rather blind approach which we use at the present time," said Dr. Martyn Partridge, chief medical adviser to the National Asthma Campaign in Britain, which has one of the world's highest rates of asthma.

Patridge, who was not involved in the study, said that until now doctors have "only been able to measure that inflammation by indirect means such as by monitoring symptoms or by monitoring lung function. If we had a better non-invasive marker of inflammation, it would enable us to tailor the medication more closely to that patient's degree of inflammation."

In the study by scientists at the Glenfield Hospital in Leicester, England, 74 patients with moderate to severe asthma who were already regularly coming to the hospital for asthma treatment were randomly allocated to either the traditional strategy of managing asthma or to an approach aimed at controlling the concentration of eosinophils in the phlegm they coughed up.

All patients were assessed nine times over 12 months. Patients in the phlegm analysis group had their medication adjusted in response to changes in their eosinophil levels.

There were 35 asthma attacks in the phlegm test group during the year, compared with 109 attacks in the group with the normal treatment strategy. Only one patient in the cell test group was hospitalized for asthma problems, compared with six in the comparison group.

"The current guidelines suggest that if you have no symptoms of your asthma for a couple of months, we would reduce the dose of your inhalers," said the study's leader Dr. Ruth Green of the Institute for Lung Health in Leicester.

"What we've shown is that even if you don't have symptoms, this inflammation in your airways builds up and then suddenly patients get an attack, so relying on symptoms and reducing treatment might be quite the wrong thing to do."

Copyright 2002 The Associated Press. All rights reserved.