BACKGROUND Atopic dermatitis (AD) can be traumatizing to family life. Little is known about the relationship between quality of life in AD and disease severity. OBJECTIVE To document family quality of life and relate this to severity of AD in children, for a 6-month period from a given point in time. STUDY DESIGN These data are part of a longitudinal study conducted in two parts of the UK to investigate risk factors for AD severity and its impact on quality of life. SUBJECTS and methods Thetargetedpopulation comprised children with AD aged 5-10 years in a primary-care setting. The general practitioners identified potential subjects and the UK diagnostic criteria for AD were used to verify the diagnosis. Both the children and their parents were interviewed. Eczema severity was assessed using a modified form of the SCORAD (SCORe Atopic Dermatitis) Index (SCORAD-D) from which parents' score of itching and sleep loss were excluded. The quality of family life was quantified by the Dermatitis Family Impact (DFI) questionnaire. These two parameters were evaluated on two occasions 6 months apart. ANALYSIS Multiple regression analysis was used to investigate the relationship between the quality of family life and the severity of the AD in the children, at a specific point in time and over the following 6-month period. RESULTS Of the 116 children attending the first visit, mean age 8 years, 106 attended the second visit (91%) and were included in the analysis. Quality of family life was shown to be significantly affected in 48 (45%) cases at the first visit and 38 (36%) cases at the second visit. The initial means of the DFI and SCORAD-D were 2.4 and 8.2, respectively. Six months later the mean final DFI and SCORAD-D were 1.9 and 7.7, respectively. Using multiple regression on the first and second visits, each unit increase in SCORAD-D was associated with 0.21 [95% confidence interval (CI) 0.06-0.37 P = 0.008] and 0.37 (95% CI 0.15-0.59, P = 0.001) units increase in quality of family life, respectively. This relationship remained significant even after adjustment for potential confounders (black skin, social class, sex, child's age, family size and location) each unit increase in SCORAD-D led to a 0.25 unit (95% CI 0.11-0.4, P = 0.001) and 0.23 unit (95% CI 0.05-0.42, P = 0.014) increase in DFI on the first and second visits, respectively. Changes in the DFI scores were significantly related to changes in the SCORAD-D scores (regression coefficient; 0.17 (95% CI 0.06-0.29, P = 0.002). CONCLUSIONS We show that quality of family life is related to the severity of AD in children. This confirms the importance of parental assessment of the impact of the disease in the management of AD, because the disease affects the entire family. Also, these results show the response of DFI to change predictably with disease severity. This may imply that the DFI questionnaire could be used as an extra measure of outcome in everyday clinical practice as well as in research studies.