Pulmonary Function: For children who were developmental able to provide a valid measurement, lung function was tested prior to and 5 min after the administration of two puffs of a P-adrenergic bronchodilator at baseline and at the 12-month follow-up visit. (Postbronchodilator testing was erroneously omitted at enrollment for 17 children because the laboratory initially continued its standard procedure of omitting this measurement when the initial value of the FEV1 was >80% of predicted.) Testing was performed by the VCH Pulmonary Function Laboratory using standardized methods and equipment that met American Thoracic Society requirements (Vmax series pneumo-tach, using a bi-directional flow-sensing mechanism; SensorMed-ics, Yorba Linda, CA). Normal (predicted) values were those of Polgar and Promadhat. buy amoxicillin online
Urine Cotinine: Assays for urine cotinine and creatinine were performed for both intervention and control children at enrollment and at the 6-month and 12-month follow-ups by the VCH Clinical Laboratory using a radioimmunoassay (Coti-Traq; International Diagnostic System Corp; St. Joseph, MI). Children in the intervention group also contributed urine samples at each intervention session.
Asthma Medications: This study was conducted through the Pediatric Pulmonary Service of VCH. Each child was examined at enrollment by a pediatric pulmonary specialist (R.S. or L.R.), and baseline pulmonary function testing was ordered. Most children had been receiving episodic urgent medical care and/or general medical care from primary-care physicians in the community. Many were on drug regimens that were clearly inadequate, as evidenced by their baseline lung function, symptoms, activity limitations, and prior acute health-care utilization (see “Results” section). In light of their poorly controlled asthma and high risk, it was considered medically and ethically requisite that an adequate regimen be prescribed for each child at the outset of the study, and from a design standpoint, this helped assure that there would be comparable management in both experimental groups. Where necessary, the medication regimen was adjusted on the basis of history, physical examination, and pulmonary function results such that all children were prescribed a regimen consistent with national guidelines.
Health-Care Utilization: For all eligible children, the dates on which the child was seen in the VCH Peds Plus clinic (pulmonary department or ED) or in the inpatient service for asthma, another respiratory condition, or a nonrespiratory condition were obtained from VCH records. Health-care utilization was measured by (1) the number of visits for asthma and (2) the number of times the child was hospitalized for asthma and the number of days of each hospitalization in the year prior to enrollment in the study (baseline) and in the follow-up year.
Randomization: Following baseline data collection, participants were randomized equally to the intervention group (n = 44) or control group (n = 43) using a randomization design with blocks of length four. This avoided long strings of assignments to either group while ensuring that the assignment of any participant was not predictable from the preceding assignments.