Epidemiology an introduction rothman pdf


Request PDF on ResearchGate | Epidemiology. An introduction: K J Rothman | The aim of this book is clearly stated by K J Rothman in the preface: “ to. KWH. [PDF] Epidemiology: An Introduction Unlimited. Detail ○ ○ ○ ○ ○ ○. Author: Kenneth J. Rothman Pages: pages Publisher. BOOK REVIEW Epidemiology: An Introduction Kenneth J. Rothman, Oxford University Press, , pages, $ Kenneth Rothman wrote in the preface of.

Language:English, Spanish, Indonesian
Genre:Politics & Laws
Published (Last):17.05.2016
Distribution:Free* [*Register to download]
Uploaded by: BRETT

74920 downloads 155710 Views 36.73MB PDF Size Report

Epidemiology An Introduction Rothman Pdf

Rothman, Kenneth J. published by Oxford University. Press, USA () PDF by aa: Epidemiology: An Introduction 2nd (second) Edition by. Rothman, Kenneth. Editorial Reviews. Review. "In summary, Epidemiology: An Introduction is a superb addition to . Rothman's Epidemiology is a must for everybody interested in the discipline. He is a stubborn challenger of the common understanding of basic. PDF. Epidemiology. An introduction. Free. Loading. S Márquez-Calderón The aim of this book is clearly stated by K J Rothman in the preface: “ to present a Chapter 1 is an introduction to epidemiological thinking, based on the concept.

Toon meer Toon minder Samenvatting Across the last forty years, epidemiology has developed into a vibrant scientific discipline that brings together the social and biological sciences, incorporating everything from statistics to the philosophy of science in its aim to study and track the distribution and determinants of health events. A now-classic text, the second edition of this essential introduction to epidemiology presents the core concepts in a unified approach that aims to cut through the fog and elucidate the fundamental concepts. Rather than focusing on formulas or dogma, the book presents basic epidemiologic principles and concepts in a coherent and straightforward exposition. By emphasizing a unifying set of ideas, students will develop a strong foundation for understanding the principles of epidemiologic research. Recensie s Students and scholars who pursue epidemiology as a career should find this book a useful addition to their library The text focuses on concepts, not on mathematics, and discusses statistical techniques in the context of the real problems they can solve. This text bridges the gap between what is taught in an introductory statistics text and what you need to be an effective researcher and analyst. Frequently introductory texts strive for brevity through superficiality of coverage. Rothman chose a different approach: selectivity of subject matter. As a result, Epidemiology: An Introduction provides an exceptionally lucid overview of analytic epidemiology

The text focuses on concepts, not on mathematics, and discusses statistical techniques in the context of the real problems they can solve. This text bridges the gap between what is taught in an introductory statistics text and what you need to be an effective researcher and analyst.

Frequently introductory texts strive for brevity through superficiality of coverage. Rothman chose a different approach: selectivity of subject matter.

As a result, Epidemiology: An Introduction provides an exceptionally lucid overview of analytic epidemiology The mix of text, graphics, tables, and sidebars that is used throughout the book proves very useful in explicating important concepts such as confounding, interaction, study design, and biases, which may be challenging to the student in an introductory course The clarity of expression in this book should be an inspiration to other authors and teachers of epidemiology.

The author has achieved the stated goal of providing a coherent overview of epidemiologic principles and concepts. In a little over pages Rothman covers all the essential topics for an introductory graduate level course in epidemiology. In contrast to the many available introductory books several of them written in a naive way , this book does not trivialize these concepts.

Join Kobo & start eReading today

It perfectly prepares the reader for the reference book of epidemiologic methods entitled Modern Epidemiology by Rothman et al Along with this growing appreciation for specific causal relations comes the general concept that some events or conditions can be considered causes of other events or conditions.

Thus, our first appreciation of the concept of causation is based on our own observations. These observations typically involve causes with effects that is immediately apparent. For example, when one changes the position of a light switch on the wall, one can see the instant effect of the light going on or off. Suppose the electric lines to the building are down in a Storm. Turning on the switch will have no effect. Suppose the bulb is Burned out.

Epidemiology : An Introduction: Kenneth J. Rothman: Telegraph bookshop

Again, the switch will have no effect. One cause of the light Going on is having the switch in the proper place, but along with it we must include a supply of power to the circuit, a working bulb, and wiring. When all other factors are already in place, turning the switch will Cause the light to go on, but if one or more of the other factors is not playing its causal role, the light will not go on when the switch is turned.

There is a tendency to consider the switch to be the unique cause of turning on the light, but in reality we can define a more intricate causal mechanism, in which the switch is one component of several. The tendency to identify the switch as the unique cause stems from its usual role as the final factor that acts in the causal mechanism.

The wiring can be considered part of the causal mechanism, but once it is put in place, it seldom warrants further attention. What Is Causation? One way this concept is expressed is by the strength of a causal effect. Thus, we say that smoking has a strong effect on lung cancer risk because smokers have about 10 times the risk of lung cancer as nonsmokers.

On the other hand, we say that smoking has a weaker effect on myocardial infarction because the risk of a heart attack is only about twice as great in smokers as in nonsmokers.

With respect to an individual case of disease, however, every component cause that played a role in bringing that case into existence was necessary to the occurrence of that case. According to the causal pie model, for a given case of disease, there is no such thing as a strong cause or a weak cause.

There is only a distinction between factors that were causes and factors that were not causes. To understand what epidemiologists mean by strength of a cause, we need to shift from thinking about an individual case to thinking about the total burden of cases occurring in a population.

We can then define a strong cause to be a component cause that plays a causal role in a large proportion of cases, whereas a weak cause would be a causal component in a small proportion of cases. Because smoking plays a causal role in a high proportion of the lung cancer cases, we call it a strong cause of lung cancer.

For a given case of lung cancer, smoking is no more important than any of the other component causes for that case; but on the population level, it is considered a strong cause of lung cancer because it causes such a large proportion of cases.

The strength of a cause, defined in this way, necessarily depends on the prevalence of other causal factors that produce disease. As a result, the concept of a strong or weak cause cannot be a universally accurate description of any cause. For example, suppose we say that smoking is a strong cause of lung cancer because it plays a causal role in a large proportion of cases.

Epidemiology: An Introduction

Exposure to ambient radon gas, in contrast, is a weaker cause because it has a causal role in a much smaller proportion of lung cancer cases.

Now imagine that society eventually succeeds in eliminating tobacco smoking, with a consequent reduction in smoking- related cases of lung cancer. One result is that a much larger proportion of the lung cancer cases that continue to occur will be caused by exposure to radon gas. It would appear that eliminating smoking has strengthened the causal effect of radon gas on lung cancer. This example illustrates that what we mean by strength of effect is not a biologically stable characteristic of a factor.

From the biologic perspective, the causal role of a factor in producing disease is neither strong nor weak: the biology of causation corresponds simply to the identity of the component causes in a causal mechanism. The proportion of the population burden of disease that factor causes, which we use to define the strength of a cause, can change from population to population and over time if there are changes pdfMachine Is a pdf writer that produces quality PDF files with ease!

In short, the strength of a cause does not equate with the biology of causation. Interaction between Causes The causal pie model posits that several causal components act in concert to produce an effect. Consider the example above of the person who sustained trauma to the head that resulted in an equilibrium disturbance, which led years later to a fall on an icy path. The earlier head trauma played a causal role in the later hip fracture, as did the weather conditions on the day of the fracture.

If both of these factors played a causal role in the hip fracture, then they interacted with one another to cause the fracture, despite the fact that their time of action was many years apart. We would say that any and all of the factors in the same causal mechanism for disease interact with one another to cause disease.

Thus, the head trauma interacted with the weather conditions as well as with the other component causes, such as the type of footwear, the absence of a handhold, and any other conditions that were necessary to the causal mechanism of the fall and the broken hip that resulted.

One can view each causal pie as a set of interacting causal components. This model provides a biologic basis for the concept of interaction that differs from the more traditional statistical view of interaction. We discuss the implication of this difference later, in Chapter9. Sum of Attributable Fractions Consider the data in Table 2—1, which shows the rate of head-and-neck cancer according to smoking status and alcohol exposure. Suppose that the differences in the rates reflect causal effects.

Among those who are smokers and alcohol drinkers, what proportion of cases of head and neck cancer that occur is attributable to the effect of smoking? We know that the rate for these people is 12 cases per 10, person-years. If these same people were not smokers, we can infer that their rate of head-and- neck cancer would be 3 cases per 10, person-years.

The answer is yes, because when we do so, some cases are counted more than once as a result of the interaction between smoking and alcohol. These cases are attributable both to smoking and to alcohol drinking, because both factors played a causal role in producing those cases.

One consequence of inter- role in producing those cases. Epidemiology: An introduction In an individual instance, we would not be able to learn the exact length of an induction period, since we cannot be sure of the causal mechanism that produces disease in an individual instance or when all of the relevant component causes in that mechanism acted. With research data, however, we can learn enough to characterize the induction period that relates the action of a single component cause to the occurrence of disease in general.

A clear example of a lengthy induction time is the cause—effect relation between exposure of a female fetus to diethyistilbestrol DES and the subsequent development of adenocarcinoma of the vagina. The cancer is usually diagnosed between the ages of 15 and 30 years. Since the causal exposure to DES occurs during gestation, there is an induction time of about 15 to 30 years for its carcinogenic action. During this time, other causes presumably operate; some evidence suggests that hormonal action during adolescence may be part of the mechanism.

The induction time can be conceptualized only in relation to a specific component cause.

Thus, we say that the induction time relating DES exposure to clear cell carcinoma of the vagina is 15 to 30 years, but we cannot say that 15 to 30 years is the induction time for clear cell carcinoma in general.

Since each component cause in any causal mechanism can act at a time different from the other component causes, each will have its own induction time. For the component cause that acts last, the induction time always equals 0. If another component cause of clear cell carcinoma of the vagina that acts during adolescence were identified, it would have a much shorter induction time than DES.

Thus, induction time characterizes a specific cause—effect pair rather than just the effect. In carcinogenesis, the terms initiator and promoter are used to refer to component causes of cancer that act early and late, respectively, in the causal mechanism.

Cancer itself has often been characterized as a disease process with a long induction time. This characterization is a misconception, however, because any late-acting component in the causal process, such as a promoter, will have a short induction time and, by definition, the induction time will always be 0 for the last component cause to act.

After disease occurs, its presence is not always immediately apparent. If it becomes apparent later, the time interval between disease occurrence and its subsequent detection, whether by medical testing or by the emergence of symptoms, is termed the latent period. The induction period, however, cannot be reduced by early detection of disease, because there is no disease to detect until after the induction period is over.

Practically, it may be difficult to distinguish between the induction period and the latent period, because there may be no way to pdfMachine Is a pdf writer that produces quality PDF files with ease! This question leads directly to the philosophy of science, a topic that goes well beyond the scope of this book. Nevertheless, it is worthwhile to summarize two of the major philosophical doctrines that have influenced modern science.

Induction Since the rise of modern science in the seventeenth century, scientists and philosophers alike have puzzled over the question of how to determine the truth about assertions that deal with the empirical world. From the time of the ancient Greeks, deductive methods have been used to prove the validity of mathematical propositions.

These methods enable us to draw airtight conclusions because they are self-contained, starting with a limited set of definitions and axioms and applying rules of logic that guarantee the validity of the method. Empirical science is different, however.

Assertions about the real world do not start from arbitrary axioms, and they involve observations on nature that are fallible and incomplete. These stark differences from deductive logic led early modern empiricists, such as Francis Bacon, to promote what they considered a new type of logic, which they called induction not to be confused with the concept of induction period, discussed above.

Induction was an indirect method used to gain insight into what has been metaphorically described as the fabric of nature. The method of induction starts with observations on nature. To the extent that the observations fall into a pattern, the observations are said to induce in the mind of the observer a suggestion of a more general statement about nature. The general statement could range from a simple hypothesis to a more profound natural law or natural relation.

The statement about nature will be either reinforced by further observations or refuted by contradictory observations.

The induction itself involves an inference beyond the observations to a general statement that describes the nature of boiling water. As induction became popular, it was seen to differ considerably from deduction.

Although not as well understood as deduction, the approach was considered a new type of logic, inductive logic. Although induction, with its emphasis on observation, represented an important advance over the appeal to faith and authority that characterized medieval scholasticism, it was not long before the validity of the new logic was questioned.

The sharpest criticism came from the philosophical skeptic David Hume, who pointed out that induction had no pdfMachine Is a pdf writer that produces quality PDF files with ease! Rather, it amounted to an assumption that what had been observed in the past would continue to occur in the future.

When supporters of induction argued for the validity of the process because it had been seen to work on numerous occasions, Hume countered that the argument was an example of circular reasoning that relied on induction to justify itself.

Hume was so profoundly skeptical that he distrusted any inference based on observation, for the simple reason that observations depend on sense perceptions and are therefore subject to error. Perhaps the most influential reply to Hume was offered by Karl Popper. On the other hand, Popper asserted that statements about nature can be refuted by deductive logic.

To grasp the point, consider the example above regarding the boiling point of water. This single contrary observation carries more weight regarding the hypothesis about the boiling point of water than thousands of repetitions of the initial experiment at sea level. The asymmetrical implications of a refuting observation, on the one hand, and supporting observations, on the other hand, are the essence of the refutationist view.

This school of thought encourages scientists to subject a new hypothesis to rigorous tests that might falsify the hypothesis, in preference to repetitions of the initial observations that add little beyond the weak corroboration that replication can supply.

The implication for the method of science is that hypotheses should be evaluated by subjecting them to crucial tests. If a test refutes a hypothesis, then a new hypothesis needs to be formulated, which can then be subjected to further tests. This process describes an endless cycle of conjecture and refutation.

The conjecture, or hypothesis, is the product of scientific insight and imagination. It requires little justification except that it can account for existing observations. A useful approach is to pose competing hypotheses to explain existing observations and to test them against one pdfMachine Is a pdf writer that produces quality PDF files with ease! The refutations philosophy postulates that all scientific knowledge is tentative in that it may one day need to be refined or even discarded.

Under this philosophy, what we call scientific knowledge is a body of as yet unrefuted hypotheses that appear to explain existing observations.

How would an epidemiologist apply refutationist thinking to his or her work? If causal mechanisms are stated specifically, an epidemiologist can construct crucial tests of competing hypotheses. For example, when toxic shock syndrome was first studied, there were two competing hypotheses about the origin of the toxin.

Under one hypothesis, the toxin responsible for the disease was a chemical in the tampon, so women using tampons were exposed to the toxin directly from the tampon. Under the other hypothesis, the tampon acted as a culture medium for staphylococci that produced the toxin. Both hypotheses explained the relation of toxic shock occurrence to tampon use. The two hypotheses, however, led to opposite predictions about the relation between the frequency of changing tampons and the risk of toxic shock.

Under the hypothesis of a chemical intoxication, more frequent changing of the tampon would lead to more exposure to the toxin and possible absorption of a greater overall dose. This hypothesis predicted that women who changed tampons more frequently would have a higher risk of toxic shock syndrome than women who changed tampons infrequently. The culture-medium hypothesis predicts that the women who changed tampons frequently would have a lower risk than those who left the tampon in for longer periods, because a short duration of use for each tampon would prevent the staphylococci from multiplying enough to produce a damaging dose of toxin.

Thus, epidemiologic research, which showed that infrequent changing of tampons was associated with greater risk of toxic shock, refuted the chemical theory. Causal Criteria Earlier, we said that there is no simple checklist that can determine whether an observed relation is causal. Nevertheless, attempts at such checklists have appeared and merit comment here. Most of these lists stem from the canons of inference described by John Stuart Mill.

Although Hill did not propose these criteria as a checklist for evaluating whether a reported association might be interpreted as causal, many others have applied them in that way. Admittedly, the process of causal inference as described above is difficult. Unfortunately, this checklist, like pdfMachine Is a pdf writer that produces quality PDF files with ease! The only criterion on the list that is truly a causal criterion is temporality which implies that the cause comes before the effect.

This criterion, which is part of the definition of a cause, is useful to keep in mind, although it may be difficult to establish the proper time sequence for cause and effect. For example, does stress lead to overeating or does overeating lead to stress?

Join Kobo & start eReading today

In general, it is better to avoid a checklist approach to causal inference and instead to consider approaches such as conjecture and refutation. Checklists lend a deceptive and mindless authority to an inherently imperfect and creative process. In contrast, causal inference based on conjecture and refutation fosters a highly desirable critical scrutiny.

Generalization in Epidemiology A useful way to think of scientific generalization is to consider a generalization to be the elaboration of a scientific theory. A given study may test the viability of one or more theories.

Theories that survive such tests can be viewed as general statements about nature that tell us what to expect in people or settings that were not studied. Because theories can be incorrect, scientific generalization is not a perfect process. Formulating a theory is not a mathematical or statistical process, so generalization should not be considered a statistical exercise.

It is really no more nor less than the process of causal inference itself. It is curious that many people believe that generalizing from an epidemiologic study involves a mechanical process of making an inference about a target population of which the study population is considered a sample. This type of generalization does exist, in the field of survey sampling.

In survey sampling, researchers draw samples from a larger population to avoid the expense of studying the entire population. In survey sampling, the statistical representative ness of the sample is the main concern for generalizing to the source population. Nevertheless, while survey sampling is an important tool for characterizing a population efficiently, it does not always share the same goals as science. Survey sampling is useful for problems such as trying to predict how a population will vote in an election or what type of laundry soap the people in a region prefer.

These are characteristics that depend on attitudes and for which there is little coherent biologic theory on which to base a scientific generalization. For this reason, survey results may be quickly outdated election polls may be repeated weekly or even daily and do not apply outside of the populations from which the surveys were conducted. Disclaimer: I am not saying that social science is not science or that we cannot develop theories about social behavior.

I am saying only that surveys about the current attitudes of a specific group of people are not the same as social theories. What is Causation? A study conducted in Chicago that shows that exposure to ionizing radiation causes cancer does not need to be repeated in Houston to see if ionizing radiation also causes cancer in people living in Houston.

Generalization about ionizing radiation and cancer is based on an understanding of the underlying biology rather than on statistical sampling. It may be helpful to consider the problem of scientific generalization about causes of cancer from the viewpoint of a biologist studying carcinogenesis in mice. Most researchers study cancer, whether it be in mice, rats, rabbits, hamsters, or humans, because they would like to understand better the causes of human cancer.

But if scientific generalization depended on having studied a statistically representative sample of the target population, researchers using mice would have nothing to contribute to the understanding of human cancer.

Similar files:

Copyright © 2019 ruthenpress.info. All rights reserved.
DMCA |Contact Us