Results of hepatitis B lucky draw announced
Analysis:
The first part is an overview.
First, meaning.
Experimental research is an important research type of clinical research methods. According to the principle of random distribution, the subjects were divided into study group and control group. Its biggest purpose is to strongly test various hypotheses. Usually divided into field trials and clinical trials.
Second, the characteristics
(1) A control group must be established.
(2) Subjects must be randomly divided into groups.
(3) artificially imposed intervention measures.
(4) The research direction is forward-looking, that is, from "cause" to "effect".
(5) Blind method is often used to collect data.
Third, advantages and disadvantages
(A) Advantages
(1) The researcher can standardize the conditions, exposure, intervention measures and result analysis of the selected research object according to the research design.
(2) By random allocation, the subjects were randomly divided into the experimental group and the control group, and the known and unknown confounding factors in the experimental group and the control group were balanced, so as to improve the comparability of the two groups.
(3) Generally speaking, the sample size of the experimental group and the control group is roughly equal, so it has high statistical validity.
(4) Because the study time of the experimental group and the control group is synchronous, the interference of external factors acts on both groups at the same time, so it has little influence on the knot.
(5) Because the results of the experimental group and the control group can be observed at the end of the study, the complications or adverse reactions of the intervention measures can be determined.
(2) Disadvantages
The design and implementation of (1) is complicated, which is sometimes difficult to achieve in practical work.
(2) The relevant conditions of the research object are controlled too strictly.
(3) Medical ethics should be considered when implementing intervention measures with unknown effects.
Section 2 Field Test
Field trials, also known as crowd trials or intervention trials, take the community as the main research object and can be divided into two types according to design principles: randomized controlled trials and quasi-trials.
I. Design Principles and Scope of Use
design philosophy
Field test refers to randomly selecting a certain number of people from a community as the research object, and randomly dividing the research object into experimental group and control group according to the principle of random pairing.
(ii) Scope of application
(1) is used to evaluate the effect of a preventive measure in the population, such as vaccination, drug prevention effect and health education effect evaluation.
(2) It is used to study the causes of diseases, such as observing and comparing the occurrence of certain diseases after artificially eliminating or controlling some suspicious pathogenic factors.
Second, the design and implementation of field testing
(A) the choice of research sites
(1) The population of the unit or community to be tested should be relatively stable, with low mobility, and the population number and characteristics can meet the research requirements.
(2) When evaluating the effect of preventive measures, we should choose areas with high and stable incidence to ensure that there are enough patients in the experimental group at the end of the trial, which is convenient for epidemiological effect evaluation.
(3) When evaluating the vaccine effect, we should choose areas where the disease has not been prevalent recently.
(4) The site should have certain medical conditions, sound medical and health institutions and perfect routine registration records.
(5) Local leaders attach importance to support and the masses are highly compliant.
(B) the choice of research objects
The research object refers to the samples randomly selected from the people who meet the research requirements.
Follow the following principles:
1. Select the subjects from the population whose intervention may be effective.
2. Select the population with higher expected incidence as the research object.
3. People whose intervention measures are known to be harmful cannot be taken as research objects.
4. Choose people with high compliance as the research object.
(3) Estimation of sample size
In the field investigation, because it involves the field population and has a heavy workload, it is necessary to estimate the sample size. If the sample size is too small, it will reduce the effectiveness of the study and get no research results. Excessive sample size will lead to unnecessary waste of manpower, material resources and time.
The main determinants of estimating the sample size are: ① the incidence of a disease in the general population; ② Differences in some characteristics between the experimental group and the control group; ③ significance level and test validity (1-p); ④ Unilateral or bilateral inspection is recommended.
1. Sample size value of counting data.
The formula for estimating the sample size of counting data is:
Among them, pc is the incidence of the control group, and p 1 is the incidence of the experimental group.
For example, the annual incidence rate of the control group is 20%, and that of the experimental group is 15%. Observation per subject 1 year. α=0.05,β=0. 1. This study is a bilateral test and the sample size is calculated.
View the quantile table of normal distribution:
Zα is the standard normal difference of α level, zβ is the standard normal difference corresponding to 1-β level, and n is the number of calculated samples.
That is, the sample size of each group is about 12 10 people.
2. Sample size of measurement data
δ is the standard deviation of estimation, and d is the difference between two continuous variables.
Example: A drug can lower the serum cholesterol level of the experimental group than that of the control group 10mg/d 1. According to the literature, δ is about 50mg/dl. This study is a bilateral test with α=0.05 and β=0. 1. Try to calculate the number of samples in each group.
Substituting into the formula, we get:
n = 2×( 1.96+ 1.282)2×(50)2/( 10)2 = 525
The sample size of each group is about 525 people, and the total sample size is 1050 people.
(D) Randomly select the research object.
In the field experiment, the concept of randomization includes two aspects: one is random sampling, that is, each individual has an equal opportunity to be selected as the research object; The second is random grouping, that is, all subjects are equally likely to be divided into experimental group or control group.
There are three commonly used random grouping methods:
(1) simple random grouping
The selected subjects can be randomly divided into experimental group and control group by drawing lots or using random number table.
(2) Hierarchical random grouping
Firstly, according to the characteristics of the subjects, such as age, gender, disease, course of disease and other factors that may have mixed effects, the subjects are stratified. Then, on each floor, the subjects were randomly divided into experimental group and control group.
(3) random grouping of clusters
Families, schools, hospitals, villages and streets were randomly divided into experimental units and control units.
(5) Blind method
Blind method means that the research object or researcher does not know the grouping situation of the research, that is, he does not know what kind of intervention a research object has received.
1.
Also known as open trial, that is, both researchers and subjects know the grouping situation. For example, when comparing the effects of surgical treatment and drug treatment, the control group can't use fake surgery to cover it up. The advantages are easy to realize, easy to find problems and deal with them in time. The disadvantage is that it is easy to produce observation deviation.
2. Blindness in one eye
In other words, one of the researchers or subjects did not know the grouping of the study. Its advantages are easy implementation and effective avoidance of observation deviation.
3. Double blind
That is, neither the subjects nor the researchers know the grouping of each subject. The advantage is that it can avoid the deviation caused by researchers and research objects. The disadvantage is that this method is complicated and difficult to implement.
4. Three blindness
It is an extension of double-blind trial, which means that the subjects, researchers and data analysts can evaluate the response variables more objectively without understanding the research groups. The disadvantage is that it is difficult to deal with it in time.
(VI) Use of placebo
One is a negative placebo, which is similar to the tested drug in appearance and shape, but has no effect on the study. The other is a positive placebo, also called a standard drug, which refers to a kind of treatment or intervention with clear and recognized curative effect.
(seven) the requirements of human intervention measures in the study.
Field testing needs special care because it involves a wide range and many participants. It is generally believed that manual intervention measures should meet the following conditions.
1. Safe
Field test is an experimental study, and intervention measures must be given to the research group. This measure must first ensure that it is safe and harmless to people. Safety evaluation can be based on animal experiments, individual human experiments and small-scale crowd experiments, and can also be determined according to the pre-test situation.
Ethical
The research object of field experiment is human, so ethical issues must be considered. International ethical issues related to medical research have been expounded in the Hersingen Declaration.
Including content:
(1) The research must conform to scientific principles, with good laboratory conditions, animal experiment foundation and sufficient scientific literature knowledge.
(2) The research plan must be submitted to the Ethics Committee for review.
(3) Research benefits patients more than society and science.
(4) Patients have the right to agree or refuse to participate in the study.
(5) In the research, once the disadvantages are found to outweigh the advantages, the research should be stopped immediately.
3. Acceptability
We should consider the acceptance of the masses.
4. Simple and easy to operate
It is not easy to carry out complex testing work.
5. Economy
Get more benefits with less cost.
(8) Determination of observation indexes
Observation indicators shall meet the following requirements:
1. feature
2. There is a time limit
3. Testability
repeatability
(9) Indicators commonly used in field test
1. Main indicators for evaluating curative effect
(1) Effective rate
Effective rate = number of effective cases treated ÷ total cases treated * 100%.
(2) Cure rate
Cure rate = number of cured cases * 100%
(3)N-year survival rate
N-year survival rate = number of cases surviving for n years ÷ number of cases followed up for n years * 100%.
2. Main indicators for evaluating the effect of preventive measures
(1) protection rate.
Protection rate = (control group morbidity-experimental group morbidity) ÷ control group morbidity * 100%
(2) Effect index.
Effect index = incidence of control group ÷ incidence of experimental group * 100%.
(x) Standardization of research
65438+
The physical, chemical and biological detection methods used in field experiments need to be clearly defined and explained in detail, and the questionnaires used should be standardized as much as possible.
2. Standardization of diagnostic criteria and observation indicators
In field experiments, the diagnostic criteria of diseases should be as clear and standardized as possible. It is best to use unified classification standards or diagnostic standards at home and abroad in order to compare with other people's research results. All kinds of observation indicators should be objective and unified.
3. Standardization of measurement methods
In the same study, it is required to use reagents from the same manufacturer, brand and batch number as much as possible. Before the study, all kinds of testing instruments should be calibrated and the samples should be tested under the same or similar conditions as far as possible. At the same time, it is necessary to unify the training of researchers and unify the investigation methods and judgment standards.
Third, common prejudice.
The most common biases are non-response bias and lost-visit bias.
Fourth, research examples
Population trial of sodium fluoride in preventing dental caries
In order to finally determine the relationship between sodium fluoride and dental caries, and to explore the effect and safety of artificially adding 1* 10-6 sodium fluoride to drinking water on dental caries, Ast and his colleagues conducted a population experiment of 10 in new york.
Newburg and Kingston are two market towns in new york. The population of each town is around 30,000, and their age, gender and skin color are basically the same. Only 30 miles apart, they are both located in the Hudson River, with similar geographical conditions such as climate, water and soil. The drinking water for residents comes from Hudson River, and the fluoride concentration in the river water is below 1 * 10-7.
Before the experiment, Ast organized experienced doctors to check the teeth and health status of children in two towns. The subjects were children aged 6- 12, and 3200 children in each town were examined.
Physical examination targets are infants to 9-year-old children, with about 500 people in each town. After that, newborns will be examined and registered year by year. There is no significant difference in the results of dental examination and health examination between the two towns. See Table 5-2.
Ast set Newburg Town as the experimental area and Kingston Town as the control area. Sodium fluoride was added to the experimental area from May 1945 through the town's water supply system, and kept at 1* 10-6. The control area continued to drink river water with fluorine content lower than l* 10-7 without taking any measures.
From June 65438 to June 0955, the dental diseases and health status of children in two towns were reviewed. The results showed that the number of permanent teeth in newberg town was 98.4 per 100 children, while that in Kingston town was 233.70. N lower than k town.
In the age groups of 10 ~ 12, 13 ~ 14, 16, the number of dental caries in N town is 53%, 47.9% and 40.9% lower than that in K town.
According to the intervention experiment lasting 10 years, Ast thinks that there is sufficient evidence that fluoride is an important factor to inhibit the occurrence of dental caries, and drinking water containing 1* 10-6 sodium fluoride can effectively reduce the incidence of dental caries.
Five-level test
A complete field study should have four basic characteristics of experimental research, namely, setting control, random grouping, human intervention and prospective follow-up. If an experimental study lacks any of these characteristics, it is called quasi-experiment.
According to whether it is controlled or not, class testing can be divided into two categories:
(1) Uncontrolled Class Testing
① Complete self-control. It refers to the comparison of the same subject before and after receiving intervention measures.
② Compared with the known results without taking measures.
For example, the average probability of mother-to-child transmission of hepatitis B virus among HBsAg positive mothers in China is 40%-50%. At this stage, if you want to observe the effect of hepatitis B vaccine in blocking mother's transmission, you can not set up a control group. You can compare the experimental results with the known and recognized rate and draw a conclusion.
(2) Non-random control tests.
For example, drinking water in place A is fluoridated and drinking water in place B is not fluoridated; A school injects hepatitis B vaccine, while B school does not; Compare the two results.
The hypothesis of this study is that large-scale health education can reduce the risk factors of cardiovascular disease in the experimental population by 20% compared with the control group. O The researchers hope that through the implementation of this plan, some health promotion institutions will be established among residents to monitor the trend of cardiovascular disease incidence and mortality and explore its causes.
In this study, five cities in Northern California were selected. Each city has a population of over 30,000 and a total population of over 300,000. These five cities have similar demographic characteristics and are geographically isolated from each other. Among them, two are experimental groups and three are control groups.
The propaganda tools used in the experiment are television, radio and various printed materials. For example, in radio programs, special broadcasts about heart health are set up, columns are set up in newspapers, and they are widely publicized to the masses through health departments, hospitals, schools and other health institutions.
The evaluation method is to collect the data of residents' health behavior and diet through questionnaire survey, test the respondents' height, weight, blood pressure, hematuria and exercise test, and master the data of cardiovascular disease incidence and death through epidemiological monitoring.
After the end of the observation period, the quantitative differences of cardiovascular diseases and risk factors between the experimental group and the control group were compared, and t test was carried out. The results show that health education has a good effect on reducing the risk factors of cardiovascular diseases, and the benefits are high.