Baseline adjustment for statistical efficiency on clinical controlled trial

In a clinical controlled trial involving repeated measures of continuous outcomes such as quality of life, distress, pain, activity level at baseline and after treatment, the possibilities of analyzing these outcomes can be numerous with quite varied findings. This paper examined four methods of statistical analysis using data from an outcome study of a clinical controlled trial to contrast the statistical power on those with baseline adjustment. In this study, data from a CCT with women with breast cancer were utilized. The experiment (n=67) and control (n=74) were about equal ratio. Four method of analysis were utilized, two using ANOVA for repeated measures and two using ANCOVA. The multivariate between subjects of the combined dependents variables and the univariate between subjects test were examined to make a judgement of the statistical power of each method. The results showed that ANCOVA has the highest statistical power. ANOVA using raw data is the least power and is the worst method with no evidence of an intervention effect even when the treatment by time interaction is statistically significant. In conclusion, ANOVA using raw data is the worst method with the least power whilst ANCOVA using baseline as covariate has the highest statistical power to detect a treatment effect other than method. The second best method as shown in this study was in using change scores of the repeated measures.

Angling sport fishing in Lobo-Broa reservoir (Itirapina, SP, Brazil) / As pescarias esportivas no reservatório Lobo-Broa (Itirapina, SP, Brasil)

The objective of this paper was to study angling from September, 2002 to September, 2004 in the Lobo-Broa Reservoir, Itirapina, SP (22° 15' S and 47° 49' W). Interviews (total 1,027) with sport fishers were accomplished in the three main fishing sites (Horto, Píer and Praia). This fishing was practiced with a simple fishing rod and reel, mainly in Horto, where the catches and fishing effort were higher. The catches were mainly composed of Cichlidae (Geophagus brasiliensis, Oreochromis niloticus, Tilapia rendalli and Cichla monoculus). We tried to determine which factors (fishing sites, type of baits and season) and the covariate fishing effort, expressed in number of fishing rods multiplied by fishing time, would affect catches, using a 3 way-ANCOVA. The final model showed that only fishing sites and effort determined the captures of sporting fishing in the reservoir. Some measures for managing fishing practices are discussed.

O objetivo deste trabalho foi estudar as pescarias esportivas, entre setembro de 2002 e setembro de 2004 na Represa do Lobo-Broa, Itirapina, SP (22° 15' S e 47° 49' W). Foram realizadas 1.027 entrevistas junto aos pescadores esportivos nos três principais locais (Horto, Píer e Praia). Estas pescarias foram praticadas com vara de pescar simples e com molinete, principalmente no Horto, onde as capturas e esforço pesqueiro foram maiores. A captura foi composta principalmente de Ciclídeos (Geophagus brasiliensis, Oreochromis niloticus, Tilapia rendalli e Cichla monoculus). Buscou-se detectar que fatores (locais de pescarias, tipos de isca e sazonalidade) e se a covariável esforço de pesca, expresso em termos de número de varas de pesca multiplicado pelo tempo de pesca, afetavam as capturas, usando uma ANCOVA trifatorial. O modelo final mostrou que apenas os locais de pesca e o esforço de pesca determinaram as capturas das pescarias esportivas na represa. Algumas medidas de manejo para as pescarias foram discutidas.

Associações entre as biomassas de peixes Sciaenidae (Teleostei: Perciformes) e de camarões Penaeoidea (Decapoda: Dendrobranchiata) no litoral norte do Estado de São Paulo / Relationships among Sciaenidae fish (Teleostei: Perciformes) and Penaeoidea shrimp (Decapoda: Dendrobranchiata) biomass from the north coast of São Paulo State, Brazil

Visando analisar as associações entre as biomassas de peixes Sciaenidae e de camarões Penaeoidea, foram feitas amostragens com barco de arrasto nas quatro estações do ano de 2001, em um gradiente de profundidade, nas regiões de Ubatuba, Caraguatatuba e São Sebastião. Em cada transecto, foram obtidas amostras de água e de sedimento. As hipóteses de que os fatores ambientais foram diferentes entre as estações do ano e profundidades foram testadas por ANOVAs. A dependência estatística entre as biomassas de peixes (variável resposta) e as biomassas de camarões (covariável) nas três regiões, foi testada por um modelo de ANCOVA. Para quantificar as relações entre as biomassas de cada espécie foi utilizada a CCA. A importância das estações do ano e das regiões no resultado da CCA foi avaliada pelo teste de Kruskal-Wallis e da profundidade pelo teste de Spearman. Nas estações do ano, as temperaturas e salinidades foram significativamente diferentes, o que não ocorreu com matéria orgânica (MO) e o diâmetro médio do sedimento (phi = fi). Foram amostrados 378.618 g de peixes e 79.338 g de camarões, com maiores capturas no inverno e nas menores profundidades. Em Caraguatatuba foram coletadas as maiores biomassas de ambos os grupos e a correlação entre as capturas de peixes e de camarões foi negativa. Já nas demais regiões, a correlação foi positiva. Assim, os dados de Caraguatatuba foram retirados do modelo final da ANCOVA, onde 47 por cento da variabilidade na biomassa dos peixes Sciaenidae pôde ser explicada pela presença dos camarões Penaeoidea. Xiphopenaeus kroyeri foi o único fator importante na distribuição de Sciaenidae no eixo 1, enquanto o eixo 2 evidenciou indiretamente o gradiente de profundidade. Os resultados sugerem que, possivelmente, muitas espécies de peixes busquem as mesmas áreas de ocorrência dos camarões em função dos fatores abióticos ou para utilizá-los como recurso alimentar.

To study the relationships among Sciaenidae fish and Penaeoidea shrimp biomass, fieldwork was carried out in the north coast of São Paulo state, Brazil, during the four seasons of 2001. Samples were collected in transects from different depths at Ubatuba, Caraguatatuba, and São Sebastião cities. Water and sediment were collected from each transect. The hypotheses that the environmental factors were different among the seasons and depths were tested by ANOVAs. The statistical dependence among the biomasses of fishes (response variable) and the biomasses of shrimps (covariable) in the three sections was tested by an ANCOVA model. CCA was used to quantify the relationships among each species' biomass. The seasons and sections important at the CCA results were evaluated by Kruskal-Wallis test and depth by Spearman's test. The temperatures and salinities were significantly different among seasons but organic matter and phi were not. A total of 378,618 g of fish and 79,338 g of shrimp were collected and the majority was captured during the winter and in shallow transects. The greatest biomass of both groups was captured at Caraguatatuba and the correlation between fish and shrimp captures was negative. A positive correlation was found in the other areas. Therefore, data from Caraguatatuba was removed from the final ANCOVA model, in which 47 percent of Sciaenidae biomass variability was explained by the presence of Penaeoidea shrimps. Xiphopenaeus kroyeri was the only important factor in Sciaenidae distribution in the axis 1, while axis 2 indirectly evidenced the depth gradient. The results suggest that many fish species search for the same areas where shrimp occurs, related to abiotic factors, and use them as food resource.