Critical Evaluation Of Epidemiological Study Designs

Observational Studies

Discuss about the Critical Evaluation of Epidemiological Study Designs.

Epidemiology deals with assessing the extent of occurrence of a disease in a population or different population. Often it is a part of such studies to evaluate the outcome of an exposure and its association with a disease. There are various studies that can be chosen based on the design of the experiment and hypothesis. Broadly, there are two ways an epidemiological studies ca be designed; experimental and observational. The experimental design is based on laboratory evaluations which can be carried out on experimental animals and models by exposure to factors that may influence the disease (Checkoway, Pearce, & Kriebel, 2007; Savovic et al., 2012). Since this type of study design is not practical with human subjects, due to ethical reasons, we may design an experiment to observe the effect of removal or withdrawal of factors and record changes.

As noted earlier, due to ethical issues design of experiments using human subjects is limited and hence most epidemiological studies are based on observation of disease occurrence in subjects already divided into groups. As for example, a data of people suffering from hepatitis can be recorded to check the possible association with liver cancer. However, there are certain limitations with observational studies as there may be a wide difference in characters and lifestyle of individuals that may attribute to the disease under investigation (Checkoway et al., 2007; Szklo & Nieto, 2012; Woodward, 2013). Therefore, in broad term, epidemiological studies can be classified as intervention studies and observational studies. Intervention studies includes clinical trials and field trials whereas observational studies includes cohort studies, case-control studies, cross-sectional surveys and routine data-based studies (Von Elm et al., 2014).

The essay will investigate the pros and cons of observational study designs so as to enable us to understand various epidemiological study methods and help in choosing the correct design during investigation of a problem in epidemiology research. The essay will introduce the three methods of observational studies; case-control, cross-sectional and cohort studies followed by analysis of the advantages and disadvantages of each studies. Appropriate examples will be used to further corroborate the type of research methods and strengthen the argument. At the end of the essay we expect a better understanding of study designs of epidemiology research based on observations.

In cohort studies, which is an observational design, the starting point is a careful choice of study population also known as a cohort. In the population the number of individuals exposed to the factor and those unexposed are observed. The study is carried out for a long period of time and finally outcome of the exposure to the factor is compared with unexposed group in the population. Cohort studies closely resembles intervention studies with the exception that the lifestyle and choice of subjects cannot be controlled by the investigator (Song & Chung, 2010).

Cohort Studies

In such studies it is of importance that the study groups are as closely similar to all factors associated with the disease. Since in cohort studies the investigator has no control over the exposure to the factor under investigation it is very likely that factors that can influence the disease other than the one being investigated may differ and hence careful means of analysis should be employed to ensure most possible even distribution of the factor under study (Szklo & Nieto, 2012).

One of the examples of a cohort study is the evaluation of relapse of epilepsy after surgical intervention in which 615 individuals who underwent surgery were followed up for an average of 8 years (range 1-19 years). It was found that 52% of patients remained seizure free even after 5 years of surgery and the figure went down to 47% after 10 years (de Tisi et al., 2011). The study indicated a need for devising a much effective mode of intervention. Here, a cohort study is the right choice as a follow-up of the patients were required after an intervention. In another example a risk of exposure to radiation during CT scan in children was studied for a long time and found that exposure of cumulative dose of 50Gy or more increases the risk of brain tumor and leukemia three times (Pearce et al., 2012).Cross-sectional studies or case-control studies would not have proved the result as efficient as a cohort study in the two cases.

Case-control studies, in contrast to cohort studies, investigates cases of diseases and controls of it in a population and backtracks history of past exposure to the factor under investigation. This type of study design is best suited for study of cancer and rare diseases. In case an investigator used cohort for investigations of rare disease one have to wait for a significant period of time to observe a statistically significant number of rare-disease cases. Case-control studies are also suitable for investigation of diseases that has a very long incubation or induction period which may not be feasible with cohort studies (Song & Chung, 2010; Szklo & Nieto, 2012). This is because a significant number of subjects with the disease can be selected with appropriate controls and there is no need for a long wait for the disease to manifest as in cohort studies.

The advantage of case-control studies is that a relatively small sample size can be selected for the studies compared to a large sample size and long period of study in case of cohort studies. This also makes it a preferable choice when expenditure is taken into account. On the other hand, the disadvantage of the case-control study is that the results are relatively harder to interpret. The choice of controls can lead to bias in the result as a thoroughly matched controls need to be selected. Besides the investigator cannot be sure of the prevalence of the disease before the exposure to the factor (Gallo et al., 2012).

Case-Control Studies

One of the examples of case-control study is an evaluation of burden and aetiology of diarrhoea in developing countries.in the study children seeking medical intervention at community hospitals due to diarrhoea were selected and compared with age and sex matched controls of the same population who do not suffer from the disease (Kotloff et al., 2013). The study found few common causative agents, control of which could reduce burden of the disease. A selection of case-control study correctly answered the research question in such cases.

In the cross-sectional studies, survey is the main approach. In this a sample of subjects are selected from a target population and surveyed at a particular point of time to assess exposure and outcome. The cross-sectional study selects samples that is representative of the target study population (Szklo & Nieto, 2012). A careful selection is necessary in order to avoid bias of results (selection bias). In order to avoid this a random sampling approach is best suited for such studies.

The cross-sectional study is employed to enumerate a common condition for a considerable duration of time (Yang et al., 2010). This study design is not suitable for cases like cancer as a large population survey may be required. However, such studies are employed to study association of HPV and cancer as there is a known risk of the virus towards cancer.

Although the cross-sectional studies are easier to conduct and can be concluded in a short duration of time as estimation of factor and cause are evaluated at the same point of time it does not clearly indicate if the outcome is an actual result of the causative agent or factor as information of both is taken at the same time by a survey(Checkoway et al., 2007; Von Elm et al., 2014).

In a cross-sectional investigation to correlate bone density and vertebral fracture images of CT scan and data on trabecular bone density of respective patients were used to determine relation between the two (Zebaze et al., 2010). The result indicated a statistically significant relation between trabecular bone density and risk of vertebral fracture. The choice of a cross-sectional study was appropriate for the case as the data of osteoporosis and fracture were available at the same time. A cohort and case-control study would have been possible but would require a long duration of study in case of cohort and a considerable sample size in case of a case-control evaluation.

Cross-Sectional Studies

Conclusion

Epidemiology involves study of different diseases in populations and may demand selective study designs for a reasonable conclusion. We have analysed different study designs employed in epidemiological studies and came to the conclusion that the choice of study design is based on the study type, hypothesis of the experiment and the nature of the disease. If a long term effect of a factor is to be evaluated a cohort study would be beneficial to generate the right answer. A case-control evaluation would determine effect of the factor on the disease under investigation in a correctly chosen sample whereas a cross-sectional study can be used when a data of effect and effector are taken at the same time. Every method of study has its pros and cons and should be considered before starting a study. 

References

Checkoway, H., Pearce, N., & Kriebel, D. (2007). Selecting appropriate study designs to address specific research questions in occupational epidemiology. Occupational and environmental medicine, 64(9), 633-638.

de Tisi, J., Bell, G. S., Peacock, J. L., McEvoy, A. W., Harkness, W. F., Sander, J. W., & Duncan, J. S. (2011). The long-term outcome of adult epilepsy surgery, patterns of seizure remission, and relapse: a cohort study. The lancet, 378(9800), 1388-1395.

Gallo, V., Egger, M., McCormack, V., Farmer, P. B., Ioannidis, J., Kirsch-Volders, M., . . . Stromberg, U. (2012). STrengthening the Reporting of OBservational studies in Epidemiology–Molecular Epidemiology (STROBE-ME): An extension of the STROBE statement. European journal of clinical investigation, 42(1), 1-16.

Kotloff, K. L., Nataro, J. P., Blackwelder, W. C., Nasrin, D., Farag, T. H., Panchalingam, S., . . . Breiman, R. F. (2013). Burden and aetiology of diarrhoeal disease in infants and young children in developing countries (the Global Enteric Multicenter Study, GEMS): a prospective, case-control study. The lancet, 382(9888), 209-222.

Pearce, M. S., Salotti, J. A., Little, M. P., McHugh, K., Lee, C., Kim, K. P., . . . Craft, A. W. (2012). Radiation exposure from CT scans in childhood and subsequent risk of leukaemia and brain tumours: a retrospective cohort study. The lancet, 380(9840), 499-505.

Savovic, J., Jones, H. E., Altman, D. G., Harris, R. J., Jüni, P., Pildal, J., . . . Gluud, L. L. (2012). Influence of reported study design characteristics on intervention effect estimates from randomized, controlled trials. Annals of Internal Medicine, 157(6), 429-438.

Song, J. W., & Chung, K. C. (2010). Observational studies: cohort and case-control studies. Plastic and reconstructive surgery, 126(6), 2234.

Szklo, M., & Nieto, J. (2012). Epidemiology: Jones & Bartlett Publishers.

Von Elm, E., Altman, D. G., Egger, M., Pocock, S. J., Gøtzsche, P. C., Vandenbroucke, J. P., & Initiative, S. (2014). The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: guidelines for reporting observational studies. International Journal of Surgery, 12(12), 1495-1499.

Woodward, M. (2013). Epidemiology: study design and data analysis: CRC press.

Yang, W., Zilov, A., Soewondo, P., Bech, O. M., Sekkal, F., & Home, P. D. (2010). Observational studies: going beyond the boundaries of randomized controlled trials. Diabetes research and clinical practice, 88, S3-S9.

Zebaze, R. M., Ghasem-Zadeh, A., Bohte, A., Iuliano-Burns, S., Mirams, M., Price, R. I., . . . Seeman, E. (2010). Intracortical remodelling and porosity in the distal radius and post-mortem femurs of women: a cross-sectional study. The lancet, 375(9727), 1729-1736.