Chapter 4 Cause Of Disease

Cause of Disease 杨海燕

Yang Haiyan

Epidemiology is a discipline to study distribution (or occurrence and development) of diseases and health conditions in human populations and the determines influencing the distribution as well as to study the measures for the prevention and control of disease and promotion of health.

Population Distribution Cause Prevention

• Definition and Models of Cause of Disease • Classification of Cause of Diseases • Process of Study on Cause of Disease • Evaluation of Presence of A Valid Statistical Association • Criteria in the Judgment of Cause of Disease

Section 1 Models of

Definition and Cause of

Disease

• Definition of Cause of Disease

Cause of disease is the factors which can increase the incidence rate of disease. And if one or more of these factors don’t exist, the incidence rate of population will decrease.

In epidemiology, we can also describe these factors as risk factors. That means these factors can be indirect causes of disease . For example, smoking--?--?--?--?---lung cancer  

• Models of Cause of Disease (1)    Organism Human Disease only suitable for infectious diseases. For example, tubercle germ--tuberculosis （ TB ） However, the organism alone is an essential factor, but not a sufficient factor for infectious diseases.

(2)    Triangle Model Host

Pathogen

Environment

For example, influenza virus

influenza.

In fact, this model is also suitable for infectious diseases. But, in this model, environments are attended. For example: tubercle + low level of immunity

tuberculosis

(3)    The Wheel Model Social Environment Biologic Environment

Host G

Genetic Core

Physical Environment

The relative size of the different components of the wheel depend upon the specific disease. For hereditary diseases, the genetic core would be relatively large. For a disease like measles, the genetic core would be smaller, the state of immunity of the host and biological sector of environment would be larger.

Section 2 Classification of Cause of Diseases Host Factor (Intrinsic) Genetic Factor Immunity Personality

(1) Genetic Factor Most diseases have relationship with genetic factor , but the degree is different. Some diseases only due to genetic factors, such as color blinding.

Some diseases partly due to genetic factors,partly due to environment factors. For example, ABO blood type Type A gastric cancer Type O duodenal ulcer

(2) Immunity low immunity---tuberculosis (3) Personality Type A: CHD (coronary heart disease) Type B: Cancer

Environmental Factors (Extrinsic) (1)  Biological Environment such as bacteria, virus (2)   Social Environment In Japan stomach cancer

(3)    Physical Environment Including heat, light, air, water, radiation, atmospheric pressure, and so on. Air pollution chronic respiratory disease, such as chronic bronchitis and lung cancer (4)    Chemical Environment Such as heavy metal, trace element.

Section 3 Study on

Process of Cause of

Disease

1 Descriptive study

to understand the distribution of disease formulate etiologic hypotheses

For example: cancer

smoking

and

lung

For example: lung cancer At the begin,we don’t know the cause of lung cancer. Through descriptive study, we can know the characters of distribution of this disease. Population: incidence rate in male>female in older>younger

Place: consumption of tobacco is more,incidence rate is higher. Time: current year> before time Raise clue or hypotheses of cause: Smoking? True or not?

Analytic Study

2

Analysis study and experimental Epidemiology test the hypotheses Patients with lung case-controlcancer study: Two groups

survey

inpatients

Frequency of smoking in group 1 Relationship exists

＞

group 2

But, We can not estimate causeeffect relationship. Why? The direction of this kind study is retrospective, cause effect. If we want to know whether there is cause—effect relationship, we must do cohort study.

• Cohort study Two groups

smoking

Follow up

Non-smoking Incidence rate of lung cancer in smoking group > Non-smoking group

Cause-effect relationship exits Because the direction of cohort study is prospective, cause effect

Experimental Epidemiology Clinical trails Intervention test Filed trail

Example of intervention test: Lung cancer ----- smoking Smoking people

Experimental group Control group

Give up smoking

Follow up

Continue smoking If group 1 ＜

group 2

Cause-effect relationship

Comparing incidence rate

3    Judgment Relationship

of

A

Cause-effect

In this chapter, we mainly talk about judgment of cause-effect relationship. But ,before this we must do a work firstly,that is:

Section 4 Evaluation of Presence of A Valid Statistical Association

In epidemiology study ,we can find relationship between cause and disease, but it is just a statistical association. That means it’s may be the true effect of an exposure on the development of disease, and it also possible has other explanation: ① it’s maybe spurious association ② it’s maybe indirect association

1 Spurious Association

In the course of epidemiology such as design, selecting subjects, dividing groups,collecting data and analysis data, if there is difference between this two groups, the result we obtained is error, we can also call the relationship as spurious association.

For example: smoking --- lung cancer In case-control study male female Patient group: 80% 20% Control group: 50% 50% Result : smoking frequency patient group > control group

2 Indirect associationA

C

B Indirect association

Example:

Stomach ulcer

smoking

Lung cancer

Section 5 Criteria in the Judgment of Cause of Disease There are about 6 criteria in the judgment of cause of disease. After epidemiological study has been done we shall use these criteria to deduce the cause of disease.

1         Time Sequence The exposure occurs in former and the disease occurs in later, this is an essential condition to judge a cause-effect relationship.

It is easy to judge time sequence in prospective cohort study , or randomized trail. But in case-control study or cross-sectional study, it is difficult.

2 Strength of the Association We usually express strength of the association by use of risk ratio (RR) . RR is more, the possibility of association between exposure and disease is larger. incidence rate of exposure group RR= incidence rate of non exposure group

RR of smoking and lung cancer is 9 RR of smoking and CHD is only 2 It is showed that smoking is more likely the cause of lung cancer than CHD. In general, if the design and analysis are all correct and RR is more than 3 or 4, we can believe that the exposure is the cause of this disease.

3

Biologic Credibility If there is a known or postulated mechanism which supports a cause and effect relationship, it can enhance the relationship.

For example, smoking ---lung cancer There were so many chemical carcinogens in tar, such as arsenic, carbon monoxide, and so on. Mean while , let dog inhale the cigarette smoking ,the dog occurred lung cancer.

4         Reeligible of Association To a cause-effect relationship, if different researchers use different methods in different time or different place among different populations, they all show similar or same results, this can support association much more.

For example, Doll and Hill smoking and lung cancer more than 7 times cohort study more than 35 times casecontrol study

5         Dose-Response Relationship When the degree of exposure has changed, if the morbidity of disease changed corresponding, the possibility of a causeeffect relationship is more large. For example, the more amount population smoke every day, the higher mortality rate they dead from lung cancer.

6 If eliminating the exposure,the morbidity rate of disease will decrease. For example, after having got rid of smoking ,we found incidence rate and mortality rate of lung cancer decrease.This enhance the persuasiveness to cause-effect association. In fact, this list also corresponding the second sentence in definition of cause of disease.

These 6 are criteria when we judge a cause-effect relationship . But that is not to say, judging a relationship must accord with all of this 6 criteria. Time sequence is essential. For the other 5criteria, the more criteria it accords with, the higher possibility the hypotheses establishes.

Summary • Epidemiologists are concerned with discovering the causes of disease in the environment, nutrition, life-style, and genes of individuals and populations-------that is, the causes or factors that when removed or modified will be followed by a reduction in the disease burde

• Several criteria must be met to establish a causal relationship between a factor and a disease

Important contents: • Definition of cause of disease • Process of Study on Cause of Disease • Criteria in the Judgment of Cause of Disease