Population Studies
Introduction
Epidemiological research deals with the study of diseases at the population level. This method investigates the health of individuals in 'real life situations', and thus reflects the research hypothesis we wish to test (for instance, whether the use of mobile phones increases the risk of developing brain tumors in humans). Conversely, this level of research is very complex to conduct, and is hampered by many methodological difficulties in choosing the research methods. In general, this type of research is conducted through observation and comparison of suspected risk factors and morbidity between two population groups. The main limitation of this type of research lies in the difficulty of controlling the many differences (personal and environmental) that exist between individuals, each of which may influence the findings of the study.
Epidemiological research deals with:
|
|
|
Analytical epidemiological research
Analytical epidemiological studies are conducted to detect a possible association between a risk factor and the development of a disease in humans. The following is a brief description of the different types of analytical epidemiological studies:
Cohort study (= Follow up study)
A cohort study is a prospective study which follows a population group exposed to the studied risk factor for a certain condition or disease and an unexposed control population for comparison. Following the collection of data on the two populations, the proportion (number of individuals in a defined unit of population, e.g. percentage) of individuals who developed the disease in the two groups is compared. The index for evaluation of the result in such a study is the relative risk, which expresses the difference between the exposed and non-exposed groups regarding the risk of developing the disease.
Relative risk clarification:
- When the relative risk = 1, there is no difference in the rate of developing the disease between the exposed and non-exposed groups regarding the studied risk factor, i.e. there is no association between the studied risk factor and the disease.
- When the relative risk > 1, the exposed group is at a greater risk of developing the disease (e.g. if the relative risk = 1.2, the risk of developing the disease is 20% greater, or 1.2-fold greater in the exposed group than in the non-exposed).
- When the relative risk < 1, the exposed group is at a lesser risk of developing the disease than the non-exposed (e.g. if the relative risk = 0.7, the risk of developing the disease is 30% lower in the exposed group than in the non-exposed).
In order to examine the possible influence of mobile phones on the development of brain tumors in children using this type of study, a group of children who have never used mobile phones and a group who use them regularly should be identified. The development of disease in the two groups is followed up over a certain period of time, e.g. 30 years, to assess whether the exposure of the children to non-ionizing radiation resulting from the use of mobile phones had any influence on the risk of developing specific diseases e.g. brain tumors.
From this example one can understand the main problems involved in this type of research:
- The number of children who are not exposed to mobile phones today is decreasing, and the number of non-exposed adults is negligible.
- People's habits in the use of mobile phones are constantly changing; consequently, over the years of the study meticulous follow-up is required and the data analysis is complex because it must take into account all the variations of mobile phone use and the changes in use patterns that have occurred over the years.
- The complexity of the data processing has increased even more in view of the advances in mobile phones technologies that bring about changes in the level of exposure and frequencies used.
- Most of the relevant diseases, especially cancer, develop slowly, and long periods may elapse until they are manifested or diagnosed. The follow-up period must therefore be very long (up to 20-40 or even 50 years).
Case control study
In this type of research a group of individuals who developed a certain disease is selected (study group) and compared with a group that did not develop the disease (control group). The study involves the collection of data on the past exposure of the two groups to the risk factor being investigated (e.g. use of mobile phones). This information may be gathered by interview (questionnaire) or through other sources (such as computerized data). After the information on the two groups is collected, the number of individuals who were exposed to the risk factor and developed the disease is compared with the number of exposed individuals who did not develop the disease. The index for evaluation of the results of a case control study is the odds ratio.
Odds ratio clarification;
- If the odds ratio = 1, the tested variable does not constitute a risk factor for developing the disease.
- If the odds ratio > 1, the tested variable constitutes a risk factor; hence individuals exposed to this factor are at a greater risk of developing the disease compared to the non-exposed.
- If the odds ratio < 1, the tested variable constitutes a protective factor; therefore individuals exposed to this factor are at a reduced risk of developing the disease.
As an example, in studies assessing the possible association between the use of mobile phones and the development of brain tumors, the patterns of use of individuals diagnosed with such tumors are compared with those of individuals who did not develop the disease in question.
The results of case control studies are based on the ability of research subjects to recall their patterns of use of mobile phones in the past, and to report them. Thus, the problem of recalling such precise details (e.g. the subject's inability to recall when the mobile phone was first used), will lead to a lack of precision in the research results. One of the main limitations of the case control study is recall bias, which comes into play when the faulty memory is systematic; for example, if mobile phone users systematically report a use inferior to the actual use (under-reporting). Recall bias affects the research results particularly if it differs between the study group and control group (differential bias). In other words, if the disease (such as a brain tumor) affects memory and causes under-reporting of the use of mobile phones by study subjects compared with controls, erroneous results would be obtained, showing no association between the use of mobile phones and an increased risk of developing the disease.
Ecological research
In this type of research the data on exposure of a certain population to a potential risk factor is compared with the incidence of a specific disease in the same population. An example of such a study in the field of non-ionizing radiation is a comparison between the percentage of individuals who had a mobile phone during certain years (data on the number of subscribers to service providers), and the rate of individuals who developed brain tumors during those years (data from the Ministry of Health or Cancer Registries).
An advantage of this type of study is the availability of data, but it has several limitations because it is conducted at the population level and not at the individual level. It is thus not possible to know whether those exposed to the potential risk factor (e.g. use of a mobile phone) are those who developed the disease. Also, increased exposure to a specific risk factor in a certain year will not necessarily be evidence of an increased incidence of this disease in the same year, because with chronic diseases, especially cancer, there is a long delay period (years) between exposure to the risk factor and development of the disease (latency period). For these reasons, ecological research is considered imprecise and cannot provide evidence of a causal association between an exposure factor and the development of a disease. However, such research is useful for testing initial hypotheses, and its results may support those of other studies conducted by more precise methods.
Besides the types of studies described above, many types of studies that are more complex and complicated may combine the above methods and/or additional methods.
27.10.15
