# Conceptual problems in the definition and interpretation of attributable fractions.

@article{Greenland1988ConceptualPI, title={Conceptual problems in the definition and interpretation of attributable fractions.}, author={S. Greenland and J. Robins}, journal={American journal of epidemiology}, year={1988}, volume={128 6}, pages={ 1185-97 } }

We have argued that the concept of attributable fraction requires separation into the concepts of excess fraction, etiologic fraction, and incidence-density fraction. These quantities do not necessarily approximate one another, and the etiologic fraction is not generally estimable without strong biologic assumptions. For these reasons, care is needed in deciding which (if any) of the concepts is appropriate for a particular application. It appears that the excess fraction (like incidence… Expand

#### Topics from this paper

#### 385 Citations

The International Journal of Biostatistics CAUSAL INFERENCE Attributable Fractions for Sufficient Cause Interactions

- 2011

A number of results concerning attributable fractions for sufficient cause interactions are given. Results are given both for etiologic fractions (i.e. the proportion of the disease due to a… Expand

Attributable Fractions for Sufficient Cause Interactions

- Mathematics, Medicine
- The international journal of biostatistics
- 2010

A method is described to estimate the lower bounds on attributable fractions using marginal structural models and identification is discussed in settings in which time-dependent confounding may be present. Expand

What Does the Population Attributable Fraction Mean?

- Medicine
- Preventing chronic disease
- 2007

The limitations to the common interpretations of the attributable fraction are discussed and it is argued that these limitations cannot be overcome merely by better statistical modeling or by use of better data sets and must be addressed through discussion of specific interventions and the hypothesized causal consequences of such specified interventions. Expand

Estimability and estimation of excess and etiologic fractions.

- Medicine
- Statistics in medicine
- 1989

Conditions under which epidemiologic data can provide estimates of the excess fraction and the etiologic fraction are described, which have implications for injury compensation in lawsuits involving the probability of causation. Expand

On the relations between excess fraction, attributable fraction, and etiologic fraction.

- Psychology, Medicine
- American journal of epidemiology
- 2012

The authors thoroughly clarify the conceptual relations between excess, attributable, and etiologic fractions by explicating the correspondence between these 2 models and clarify the relation between 3 distinct assumptions-positive monotonicity, no preventive action (or sufficient-cause positive Monotonicity), and no preventive sequence. Expand

Concepts and pitfalls in measuring and interpreting attributable fractions, prevented fractions, and causation probabilities

- Psychology
- 2015

Measures of causal attribution and preventive potential appear deceptively simple to define, yet have many subtle variations and are subject to numerous pitfalls in conceptualization, interpretation,… Expand

Concepts and pitfalls in measuring and interpreting attributable fractions, prevented fractions, and causation probabilities.

- Mathematics, Medicine
- Annals of epidemiology
- 2015

Basic concepts, measures, and problems of epidemiologic attribution and preventive potential are reviewed to serve as an introduction to more detailed literature. Expand

Attributable Fraction and Probability of Causation

- Medicine
- 2014

The concepts and definitions associated with terms such as attributable fraction, attributable risk, and probability of causation, which are not estimable without strong assumptions about biologic mechanisms of disease are discussed. Expand

A heuristic approach to the formulas for population attributable fraction

- Medicine
- Journal of epidemiology and community health
- 2001

A heuristic approach, coupled with pictorial representations, is offered to help understand and interconnect the structures behind the Levin and Miettinen formulas. Expand

Use and misuse of population attributable fractions.

- Medicine
- American journal of public health
- 1998

Computational and conceptual issues relevant to population attributable fraction estimation that are infrequently discussed elsewhere are considered, with illustrations from the breast cancer literature. Expand

#### References

SHOWING 1-10 OF 32 REFERENCES

Estimability and estimation of excess and etiologic fractions.

- Medicine
- Statistics in medicine
- 1989

Conditions under which epidemiologic data can provide estimates of the excess fraction and the etiologic fraction are described, which have implications for injury compensation in lawsuits involving the probability of causation. Expand

Estimating the population attributable risk for multiple risk factors using case-control data.

- Medicine
- American journal of epidemiology
- 1985

A straightforward and unified approach is presented for the calculation of the population attributable risk per cent in the general multivariate setting, with emphasis on using data from case-control studies, so that risks need not be estimated separately in a large number of strata. Expand

Statistical methods for estimating attributable risk from retrospective data.

- Mathematics, Medicine
- Statistics in medicine
- 1982

Levin's measure of attributable risk is extended to adjust for confounding by aetiologic factors other than the exposure of interest and there appears no advantage in using the log-based interval suggested by Walter which is always longer than the simpler symmetric interval. Expand

Statistical issues in the estimation of assigned shares for carcinogenesis liability.

- Engineering, Medicine
- Risk analysis : an official publication of the Society for Risk Analysis
- 1987

The proposed formula to assign shares in cancer causation to specific doses of radiation, for use in establishing liability and compensation awards, is incorrect as an expression for "probability and causation," that it implies hidden, debatable policy judgments in its treatment of factor interactions and uncertainties, and that it can not in general be quantified with sufficient precision to be useful. Expand

Invariants and noninvariants in the concept of interdependent effects.

- Psychology, Medicine
- Scandinavian journal of work, environment & health
- 1988

This paper shows that the identification of interdependence using Miettinen's definitions varies with the choice of the reference categories for the exposures, and in particular, MietTinen's concepts of synergism and antagonism are not invariant under exposure recoding. Expand

A graphical approach to the identification and estimation of causal parameters in mortality studies with sustained exposure periods.

- Medicine
- Journal of chronic diseases
- 1987

The analytic approach introduced in this paper may be necessary to control bias in any epidemiologic study in which there exists a risk factor which both determines subsequent exposure and is determined by previous exposure to the agent under study. Expand

The probability of causation under a stochastic model for individual risk.

- Mathematics, Medicine
- Biometrics
- 1989

A mathematical definition for the probability of causation is offered that formalizes the legal and ordinary-language meaning of the term and shows that, under this definition, even the average probability of correlation among exposed cases is not identifiable from epidemiologic data. Expand

On the measurement of susceptibility in epidemiologic studies.

- Medicine
- American journal of epidemiology
- 1989

The authors derive simple estimates of the proportion of susceptibles in a population based on relative risk, and disease and exposure frequencies, which increases with increasing disease frequency and relative risk but declines at high exposure frequency. Expand

The estimation and interpretation of attributable risk in health research.

- Mathematics, Medicine
- Biometrics
- 1976

Methods to analyse confounded, polytomous and interacting risk factors are proposed and it shown that there is a simple relationship between two distinct estimators previously suggested for use with deleterious and beneficial (or preventive) factors. Expand

Interval estimation of the attributable risk for multiple exposure levels in case-control studies.

- Mathematics, Medicine
- Biometrics
- 1983

Estimation for the 2 X 2 table ( case-control status versus dichotomous exposure) and the 2X k table (case- control status versus exposure at several levels) are reviewed, and a method for finding confidence intervals for attributable risk in the2 X k table is proposed. Expand