Objective: We aimed to assess the validity of maternal recall of exclusive breastfeeding (EBF) at 3 months obtained 12 months after childbirth. Design: A population-based birth cohort study. The gold standard is maternal report of EBF at the age of 3 months (yes or no) and age of introduction of other foods in the infant’s diet. EBF was considered when the mother reported that no liquid, semi-solid or solid food was introduced up to that moment. The variable to be validated was obtained at 12 months after childbirth when the mother was asked about the age of food introduction. The prevalence of EBF at 3 months, and sensitivity, specificity, positive (PPV) and negative predictive values (NPV), and accuracy of 12-month recall with 95 % CI were calculated. Setting: Pelotas, Brazil. Participants: 3700 mothers of participants of the Pelotas 2004 Birth Cohort. Results: The prevalence of EBF at 3 months was 27·8 % (95 % CI 26·4, 29·3) and 49·0 % (95 % CI 47·4, 50·6) according to gold standard and maternal recall, respectively. The sensitivity of maternal recall at 12 months was 98·3 % (95 % CI 97·4, 99·0), specificity 70·0 % (95 % CI 68·2, 71·7), PPV 55·8 % (95 % CI 53·4, 58·1), NPV 99·1 % (95 % CI 98·6, 99·5) and accuracy 77·9 % (95 % CI 76·6, 79·2). When the analyses were stratified by maternal and infant characteristics, the sensitivity remained around 98 %, and the specificity ranged from 64·4 to 81·8 %. Conclusions: EBF recalled at the end of the first year of infant’s life is a valid measure to be used in epidemiological investigations.

No abstract

Objectives: Asthma causes impairments in health-related quality of life (HRQL). The widely used generic HRQL tool, the EQ-5D-5L, is used to evaluate interventions improving HRQL. However, there is not enough information about its suitability in the asthma population. Methods: Data stems from the randomized controlled trial EPRA, offering pulmonary rehabilitation (PR) for the intervention group. We measured HRQL for 371 patients at four time points according to the intervention group: T0 (randomization), T1 (start of PR), T2 (end of PR), and T3 (three months after PR). We measured the generic EQ-5D-5L (utilities and the Visual Analog Scale (VAS)) and the disease specific Asthma Quality of Life Questionnaire (AQLQ). We calculated intraclass correlation (T0-T1), Cohen's d (T2, T3) and regression analysis to evaluate the responsiveness to changes in asthma control, measured with Asthma Control Test (ACT). Results: ICC was 0.82, 0.72 and 0.74 for AQLQ, utilities and VAS respectively. All HRQL tools could differentiate better between well-controlled asthma versus not wellcontrolled asthma, than between not well-controlled asthma and very poorly controlled asthma. AQLQ showed the highest Cohen's d in almost every case. ACT changes were detected by VAS in every period. AQLQ was more responsive for positive changes while the utilities did not appropriately react to changes. Conclusions: The utilities have difficulties to detect changes due to a change in asthma control. This could modify results of health economic evaluations. Therefore, we suggest the use of additional measures (e.g. AQLQ) to provide a comparison.

No abstract

No abstract