Presentation on theme: "Biostatistics course Part 2 Types of studies in epidemiology Dr. en C. Nicolas Padilla Raygoza Departrment of Nursing and Obstetrics Division of Health."— Presentation transcript:
Biostatistics course Part 2 Types of studies in epidemiology Dr. en C. Nicolas Padilla Raygoza Departrment of Nursing and Obstetrics Division of Health Sciences and Engineering University of Guanajuato Mexico
Biosketch Medical Doctor by University Autonomous of Guadalajara. Pediatrician by the Mexican Council of Certification on Pediatrics. Postgraduate Diploma on Epidemiology, London School of Hygine and Tropical Medicine, University of London. Master Sciences with aim in Epidemiology, Atlantic International University. Doctorate Sciences with aim in Epidemiology, Atlantic International University. Professor Titular A, Full time, University of Guanajuato.
Competencies The reader will differentiate between observational and experimental studies. The reader will know advantages and disadvantages of both types of studies.
Types of studies Types of epidemiological studies Observational Cases and controls CohortCross sectional Cases or report of cases Ecologic Experimental Quasi- Experi- mental Randomized Controlled studies
Cases in series Advantages: They are easy to write. The observations are useful to other researchers. Disadvantages: There are a lot of bias.
Cases and controls studies Exposed Non-exposed Exposed Non-exposed Time Beginning of study Direction of research Cases Controls
Cases and controls studies Advantages: They are adequate to study rare outcomes. They are adequate to outcomes with long latency period. They are cheap and easy to apply. It is not necessary to wait to present outcome. Disadvantages: A lot of bias. They depend on the quality of registries. Control group should be adequately selected, because they represent the population without the outcome.
Nested cases and controls studies Cohort selected to study Non- exposed Exposed With outcome Without outcome Cases Sample of controls With outcome Without outcome Cases Sample of controls Beginning of study Time
Cohort studies Selection of a cohort for study Non-exposed Exposed With outcome Without outcome With outcome Without outcome Beginning of study Time
Cohort studies Subjects are selected because do not have the outcome and they are classified if have or not have the risk factor (exposure). We follow up to prove if they develop the outcome. The cohort study can be prospective if the follow up is forward in the time or it can be retrospective (historic), if it go back in the time.
Cohort studies Advantages: They are adequate to know the causes of an outcome. To know the natural history of disease. They adequate when the exposure is rare. They are useful when we study two or more outcome at the same time. Disadvantages: They take a long time. They are expensive. Subjects can be lost in the follow up. They are not adequate for study rare outcomes.
Cross sectional studies Subjects selected to study Beginning of study Exposed with outcome Exposed without outcome Non-exposed with outcome Non-exposed without outcome
Cross sectional studies Analyze data of a subjects group in a point of time. Describe a disease and its importance for the population. Define the needed on health. They can be classified in: Descriptive Analytic
Cross-sectional studies Advantages: They are useful to know the burden of a disease in a group. Useful to evaluate diagnostic procedures. To study common risk factors. To study common outcomes. Disadvantages: Populations little willing to collaborate. The sample can not be representative from the population. It is not useful to search causes of the outcome.
Experimental studies Classification Randomized clinical trials. Quasi experimental. With historic controls.
Experimental studies Outcome Subjects that participate Outcome Controls Beginning of study Intervention Time Without outcome Exposed
Experimental studies They are called clinical trials. It is administrated an intervention to a group, randomize selected and we do not know what is receiving (blind). The group that does not receive the intervention, it is a control group. The allocation of subjects in experimental or control group is given by chance. By ethics reasons, only it is permitted beneficial interventions.
Experimental studies Blind single is when the subjects do not know what intervention are receiving. Double blind is when neither subjects nor researcher know what intervention are receiving each subject.
Experimental studies Cross-design There are clinical trials with auto controls. The same group work as control group.
Experimental studies There are cross design where it is administrated an intervention (1) to experimental group and another (2) in a control group. After, interventions are suspended, and left a space (wash out period) without it, then the intervention 1 is administrated to control group and intervention 2 is administrated to experimental group.
Experimental studies Experimental group Subjects that participate Without outcome Controls Without outcome Outcome Experimental group Outcome Without outcome Outcome Controls Beginning of study Intervention Time
Experimental studies There are clinical trials with external controls. We compare the results with the results of another researcher or with the results of a previous study. Also, they are called historic controls.
Experimental studies Subjects Without outcome With outcome Results of a previous study Without outcome Beginning Intervention only in subjects Time of study With outcome
Experimental studies Advantages: Give strong evidence of causality. There are less bias. Historic controls are used in preliminary studies. Disadvantages: Inappropriate use of historic controls lead a severe mistakes. Expensive. They need time.
Ecologic studies Compare exposure and the outcome between groups. Measure the exposure and outcome, in the group as all. They are only studies that offer to study differences between groups.
Ecologic studies Advantages: Fast Cheap Use routinely data Disadvantages: They did not take into account to the individual. They depend on the quality of routinely data They are difficult to interpret.
Bibliography 1.- Gordis L. Epidemiology. Phialdelphia, W.B. Saunders Company, Songer T. Study designs in epidemiologic research. Supercourse, 2005 (http://www.pitt.edu/~super1/lecture/lec19101 /index.htm) (Accesed October 2008).http://www.pitt.edu/~super1/lecture/lec19101 /index.htm 3.- Hennekens CH, Buring J, Mayrent SL. Epidemiology in Medicine. Boston, Little Brown and Company, 1987.