To produce a scale useful for individual clinical decision making, the Marks Asthma Quality of Life Questionnaire (AQLQ-M) was modified to a 22-item scale using a 7-point Likert response scale, and the validity of the new instrument was assessed. Adult asthma subjects with moderate to severe disease, recruited from two hospitals in Adelaide, Australia, were surveyed at baseline (n = 293), and at 3-month follow-up (n = 234). Cronbach's alpha for the Total scale of the modified AQLQ-M (MAQLQ-M) was 0.97 and all subscale values exceeded 0.90. Test-retest reliability values for all scales were between 0.88 and 0.93. All correlations between disease reference measures were statistically significant to at least the p < 0.01 level. Stronger associations were seen with symptom and self-rating scales than for lung function, medication usage, and health service utilization measures of outcome. The MAQLQ-M showed good discriminative ability for all asthma symptom categories and for different FEV1 values. Moderate, statistically significant associations were seen between changes in MAQLQ-M scores and clinical measures. Higher baseline MAQLQ-M scores were associated with lower risks over 12 months for hospital admissions (odds ratio, OR = 0.58) and repeated emergency department visits (OR = 0.47). The MAQLQ-M is a highly valid measure of asthma-related quality of life.