Causal inference accounting for unobserved confounding after outcome regression and doubly robust estimation.

Causal inference with observational data can be performed under an assumption of no unobserved confounders (unconfoundedness assumption). There is, however, seldom clear subject-matter or empirical evidence for such an assumption. We therefore develop uncertainty intervals for average causal effects based on outcome regression estimators and doubly robust estimators, which provide inference taking into account both sampling variability and uncertainty due to unobserved confounders. In contrast with sampling variation, uncertainty due to unobserved confounding does not decrease with increasing sample size. The intervals introduced are obtained by modeling the treatment assignment mechanism and its correlation with the outcome given the observed confounders, allowing us to derive the bias of the estimators due to unobserved confounders. We are thus also able to contrast the size of the bias due to violation of the unconfoundedness assumption, with bias due to misspecification of the models used to explain potential outcomes. This is illustrated through numerical experiments where bias due to moderate unobserved confounding dominates misspecification bias for typical situations in terms of sample size and modeling assumptions. We also study the empirical coverage of the uncertainty intervals introduced and apply the results to a study of the effect of regular food intake on health. An R-package implementing the inference proposed is available.

Análise de frações de fibra alimentar em cultivares de feijão cultivadas em dois ambientes / Analysis of dietary fiber fractions in common bean cultivars in different locations

O objetivo deste trabalho foi determinar as frações de fibra alimentar em cultivares de feijão obtidas em diferentes locais, com vistas à avaliação dos efeitos da interação cultivares x locais. A fibra alimentar foi determinada pelo método enzimático gravimétrico em grãos de 19 cultivares de feijão, obtidos na safra de 2004/2005, em dois municípios do Estado do Rio Grande do Sul. Interações cultivares x locais significativas foram observadas em relação aos teores de fibra alimentar total, à fibra insolúvel, à fibra solúvel e ao rendimento de grãos. A fibra alimentar total variou de 20,85 por cento ("Macotaço", em Santa Maria) a 31,35 por cento ("Macanudo", em Pelotas). As cultivares "Macanudo" e "Guateian 6662" apresentam elevados teores de fibra alimentar total, insolúvel e solúvel, e a utilização dessas cultivares como alimento funcional nas dietas deverá ser investigada.

The objective of this study was to estimate the dietary fiber fractions in common bean cultivars obtained in different locations, for evaluation of the genotype x environment interaction effects. The dietary fiber was determined by enzymatic gravimetric method in grains of the nineteen common bean cultivars, obtained in 2004/2005 crop season, in two locations of Rio Grande do Sul State, Brazil. Significant genotype x environment interactions were detected for total, insoluble and soluble dietary fiber content and grain yield. The total dietary fiber content varied from 20.85 percent ('Macotaço', in Santa Maria) to 31.35 percent ('Macanudo', in Pelotas). The 'Macanudo' and 'Guateian 6662' cultivars showed high total, insoluble and soluble dietary fiber content and its utilization as a regular food in diets can be studied.

Teores de fibra e rendimento de grãos em populações de feijão / Fiber content and yield grain in common bean populations

O desenvolvimento de cultivares de feijão com alto potencial de produtividade e maior teor de fibra é desejável, por isso objetivou-se neste trabalho investigar a presença de variabilidade genética para o teor de fibra e para o rendimento de grãos em populações segregantes de feijão do grupo preto. Os cruzamentos possíveis entre os genitores CNFP 8100, FT 96-1282, Varre-Sai e BRS Valente foram realizados em casa-de-vegetação. As populações obtidas (F1, F1 recíprocos, F2 e RC) e os genitores foram avaliados a campo durante a primavera/verão de 2003/04, em delineamento de blocos ao acaso. Apesar de ter-se utilizado genitores contrastantes para teor de fibra bruta, definidos em experimento prévio, não foram detectadas diferenças significativas para fibra bruta, fibra solúvel, fibra insolúvel, fibra alimentar total e rendimento de grãos. Assim, não foi possível obter populações segregantes com variabilidade genética, a partir dos cruzamentos entre os genitores CNFP 8100, FT 96-1282, BRS Valente e Varre-Sai. As populações obtidas apresentaram altos teores de fibra alimentar solúvel, insolúvel e total.

The development of bean cultivars with high grain yield and high dietary fiber content is desirable, thus objective of this study was to evaluate genetic variability for fiber content and grain yield in segregant populations of the black common bean. Crossings were performed inside a greenhouse among the parents CNFP 8100, FT 96-1282, Varre Sai and BRS Valente. The populations obtained (F1, F1 reciprocals, F2 and RC) and parents were assessed in the field during spring/summer 2003/04, using a complete randomized blocks experimental design. Significant differences were not obtained for crude fiber, soluble fiber, insoluble fiber, dietary fiber and grain yield, although contrast parents were used. In this go, it was not possible to obtain segregant populations with genetic variability of the crossing among the parents CNFP 8100, FT 96-1282, Varre Sai and BRS Valente. The populations obtained showed high soluble, insoluble and total dietary fiber content.