The role of conceptual frameworks in epidemiological analysis: a hierarchical approach.
01 Feb 1997-International Journal of Epidemiology (Oxford University Press)-Vol. 26, Iss: 1, pp 224-227
TL;DR: Conceptual frameworks provide guidance for the use of multivariate techniques and aid the interpretation of their results in the light of social and biological knowledge.
Abstract: Background This paper discusses appropriate strategies for multivariate data analysis in epidemiological studies. Methods In studies where determinants of disease are sought, it is suggested that the complex hierarchical inter-relationships between these determinants are best managed through the use of conceptual frameworks. Failure to take these aspects into consideration is common in the epidemiological literature and leads to underestimation of the effects of distal determinants. Results An example of this analytical approach, which is not based purely on statistical associations, is given for assessing determinants of mortality due to diarrhoea in children. Conclusions Conceptual frameworks provide guidance for the use of multivariate techniques and aid the interpretation of their results in the light of social and biological knowledge.
Citations
More filters
Book•
18 Aug 2014
2,305 citations
Cites background or methods from "The role of conceptual frameworks i..."
...6 shows data from a matched case–control study of risk factors for infant death from diarrhoea in Brazil [Victora et al. (1987) Lancet ii: 319–322], in which an attempt was made to ascertain all infant deaths from diarrhoea occurring over a one-year period in two cities in southern Brazil, by means of weekly visits to all hospitals, coroners’ services and death registries in the cities....
[...]
...Appropriate methods for this situation are reviewed by White et al. (2001). 3 For a binary outcome variable, if the sensitivity and specificity with which it was measured are known then estimated odds ratios from logistic regression may be corrected, as described by Magder and Hughes (1997)....
[...]
...Taking into account the need to combine external knowledge with statistical associations, we recommend the following strategy for choosing confounders: 1 Formulating a conceptual, hierarchical framework for the relationships between the different variables and the disease outcome is strongly recommended, as described by Victora et al. (1997) in the context of determinants of childhood diarrhoea mortality....
[...]
TL;DR: It is argued that including a process evaluation would improve the science of many randomised controlled trials.
Abstract: Most randomised controlled trials focus on outcomes, not on the processes involved in implementing an intervention. Using an example from school based health promotion, this paper argues that including a process evaluation would improve the science of many randomised controlled trials
1,091 citations
01 Jan 2007
TL;DR: Evidence on the long-term effects of breastfeeding : systematic review and meta-analyses / Bernardo L. Horta ... [et al].
Abstract: Evidence on the long-term effects of breastfeeding : systematic review and meta-analyses / Bernardo L. Horta ... [et al.]. permission to reproduce or translate WHO publications – whether for sale or for noncommercial distribution – should be addressed to WHO Press, at the above address (fax: +41 22 791 4806; e-mail: permissions@who.int). The designations employed and the presentation of the material in this publication do not imply the expression of any opinion whatsoever on the part of the World Health Organization concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. Dotted lines on maps represent approximate border lines for which there may not yet be full agreement. The mention of specific companies or of certain manufacturers' products does not imply that they are endorsed or recommended by the World Health Organization in preference to others of a similar nature that are not mentioned. Errors and omissions excepted, the names of proprietary products are distinguished by initial capital letters. All reasonable precautions have been taken by the World Health Organization to verify the information contained in this publication. However, the published material is being distributed without warranty of any kind, either expressed or implied. The responsibility for the interpretation and use of the material lies with the reader. In no event shall the World Health Organization be liable for damages arising from its use.
603 citations
Cites background from "The role of conceptual frameworks i..."
...The “adjusted” estimate will reflect the residual effect of breastfeeding which is not mediated by current weight (23)....
[...]
TL;DR: The prevalence of physical inactivity in this Brazilian adult population is high, even though lower than reported in studies of leisure-time activity alone in other populations, and studies in developing countries may be seriously biased if activities during labor, transportation, and housework are not assessed.
Abstract: HALLAL, P C, C G VICTORA, J C K WELLS, and R C LIMA Physical Inactivity: Prevalence and Associated Variables in Brazilian Adults Med Sci Sports Exerc, Vol 35, No 11, pp 1894–1900, 2003PurposeTo measure the prevalence of physical inactivity, and variables associated with it,
584 citations
Cites background or methods from "The role of conceptual frameworks i..."
...Multivariate analysis was conducted using a hierarchical approach (28) with the model incorporating three levels: the distal, including sex, age, and skin color; the intermediate, including social status and partner status; and the proximal, including BMI and self-reported health status....
[...]
...analysis (28) adjusts the effect of each variable for those in the same level or above in the model....
[...]
TL;DR: The impact of the three main sources of potential bias in the traditional approach to mediation analyses are reviewed and discussed: (i) mediator-outcome confounding; (ii) exposure-mediator interaction and (iii) mediATOR- outcome confounding affected by the exposure.
Abstract: In epidemiological studies it is often necessary to disentangle the pathways that link an exposure to an outcome. Typically the aim is to identify the total effect of the exposure on the outcome, the effect of the exposure that acts through a given set of mediators of interest (indirect effect) and the effect of the exposure unexplained by those same mediators (direct effect). The traditional approach to mediation analysis is based on adjusting for the mediator in standard regression models to estimate the direct effect. However, several methodological papers have shown that under a number of circumstances this traditional approach may produce flawed conclusions. Through a better understanding of the causal structure of the variables involved in the analysis, with a formal definition of direct and indirect effects in a counterfactual framework, alternative analytical methods have been introduced to improve the validity and interpretation of mediation analysis. In this paper, we review and discuss the impact of the three main sources of potential bias in the traditional approach to mediation analyses: (i) mediator-outcome confounding;(ii) exposure-mediator interaction and (iii) mediator-outcome confounding affected by the exposure. We provide examples and discuss the impact these sources have in terms of bias.
564 citations
References
More filters
1,196 citations
Journal Article•
TL;DR: The incidence of radiologically confirmed pneumonia was associated with low paternal education, the number of persons in the household, young maternal age, attendance at day-care centers, low birth weight and weight-for-age, lack of breast-feeding and of non-milk supplements, and a history of previous pneumonia or wheezing.
Abstract: Objective. To investigate risk factors for pneumonia for infants Design. Hospital-based, case-control study with neighborhood control subjects. Setting. Urban area in southern Brazil. Subjects. Five hundred ten infants with radiologically confirmed pneumonia who were admitted to a pediatric hospital. One age-matched neighborhood control subject was selected for each case. Results. Multiple conditional regression modeling was used to control for confounding, taking into account the hierarchical relationships between risk factors. The incidence of radiologically confirmed pneumonia was associated with low paternal education, the number of persons in the household, young maternal age, attendance at day-care centers, low birth weight and weight-for-age, lack of breast-feeding and of non-milk supplements, and a history of previous pneumonia or wheezing. Day-care center attendance showed the highest risk, with an adjusted odds ratio of 11.75. Conclusions. In addition to continued efforts toward appropriate case management, actions directed against the above risk factors may help prevent the major cause of deaths of children younger than 5 years.
232 citations
Journal Article•
TL;DR: Children who had suffered from previous episodes of wheezing or been hospitalized for pneumonia had a greater than threefold increased risk of contracting the disease, and attendance at a day care centre was associated with a high odds ratio.
Abstract: Reported are the results of a case-control study carried out between July 1989 and June 1990 in Fortaleza city, Ceara State, Brazil, to determine the factors that place young children living in urban slum conditions at increased risk of contracting pneumonia. Cases were 650 under-2-year-olds with a radiological diagnosis of pneumonia who were recruited at the main paediatric hospital in the city over a full calendar year. Age-matched controls were recruited from the neighbourhood where the cases lived. Cases and controls were compared with respect to a variety of sociodemographic, environmental, reproductive, nutritional, and morbidity factors, and a risk factor questionnaire was administered to the mother of each child or to the child's normal guardian. Cases and controls were also weighed and measured. Malnutrition was the most important risk factor for childhood pneumonia in the study population, with weight-for-age, height-for-age, and weight-for-height also being important risk factors. In view of the high prevalence of stunting in the study population, there is an urgent need to reduce the level of malnutrition as a priority. Attendance at a day care centre was also associated with a high odds ratio. In view of the growing numbers of children attending day care centres in both developing and developed countries, it is essential that ways be identified to improve the design and management of such centres in order to minimize the risk of pneumonia. Increased risks of childhood pneumonia were also associated with low birth weight, non-breast-feeding, crowding, high parity, and incomplete vaccination status, but not with socioeconomic status or environmental variables. Finally, children who had suffered from previous episodes of wheezing or been hospitalized for pneumonia had a greater than threefold increased risk of contracting the disease.
120 citations
Journal Article•
25 citations