Introducing Statistical Methods in Epidemiology

Prof. Haroun O. Isah

MBBS, MPH, FWACP, FIMC, CMC, Cert. Stat. Epidemiol, Cert. APM

Provost

College of Medicine & Health Sciences

Bingham University, New Karu

Nasarawa State

​

​

Introduction

• We're quite familiar with definition of Biostatistics:
• The study and analysis of the distribution, patterns and determinants of health and disease conditions in defined populations
• The application of statistical principles to questions and problems in medicine, public health or biology
• Statistical Epidemiology:
• An emerging branch of the disciplines of Epidemiology and Biostatistics that aims to:
• Bring more statistical rigour to bear in the field of Epidemiology
• Recognize the importance of applied statistics, especially with respect to the context in which statistical methods are appropriate and inappropriate
• Aid and improve our interpretation of observations
• Role of statistical epidemiology:
• To bring the most appropriate methods available to bear on study from medical research
• Creates a broad appreciation of the underpinning methods and their context of applicability and interpretation

What Statistical Methods in Epidemiology Seeks to Achieve

Appropriate use of statistical methods in the analysis and interpretation of epidemiological studies, including life-table analysis

​

​

Epidemiological Methods in the Investigation of Etiology

Epidemiologic Study Design: Classification

Epidemiologic Study Design: Classification

Epidemiologic Study Design: Classification (contd.)

CORRELATIONAL

Observational vs Experimental Studies

• Observational studies: Allow nature to take its cause; the investigator measures but does not intervene
• Descriptive study: focuses on the description of the occurrence of a disease in a population
• Analytical study analyses relationships between health status and other variables
• Experimental or interventional studies: involve an active attempt to change a disease determinant (e.g an exposure or a behaviour) or the progress of a disaese (through treatment)
• The studies are based on a grp which has had the experience compared with control grp which has not had the experience.

Purpose of Descriptive Epidemiology

• To generate hypothesis
• To permit evaluation of trends in health & disease and comparisons among countries and subgroups within countries.
• To provide a basis for planning, provision and evaluation of health services
• To identify problems to be studied by analytical methods and to suggest areas that may be fruitful for investigation

Case Studies (Case Series)

• Case reports: documents unusual medical occurrence and can represent the first clues to the formulation of hypothesis, generally report a new or unique findings and previous undescribed disease.
• Case series: collection of individual case reports which may occur within a fairly short time, and experience of a group of patients with similar diagnosis.
• Advantages
• Useful for hypothesis generation
• Informative for very rare disease with few established risk factors
• Usually of short duration.
• Disadvantages
• Cannot study cause and effect relationships
• Cannot assess disease frequency

​

​

Cross-Sectional Study

• It is also called epidemiologic study or prevalence study
• It analyses (describes)data collected on a group of subjects at one point in time rather than over a period of time i.e. they survey exposure and disease at a single point in time.
• Both exposure and outcome variables are been evaluated at the same point in time(without any inbuilt directionality)
• Most sophisticated descriptive study
• It answers the question “WHAT IS HAPPENING RIGHT NOW?”

​

Cross-Sectional Studies

• Advantages
• Best for determining the status quo(prevalence)
• Quick
• Relatively inexpensive
• Disadvantages
• Only a snapshot at a time leading to a misinformation
• Response rate may be low ,with result not representative of the population
• Weakest observational design, (it measures prevalence, not incidence of disease). Prevalent cases are survivors
• The temporal sequence of exposure and effect may be difficult or impossible to determine
• Usually don’t know when disease occurred
• Rare events a problem. Quickly emerging diseases a problem

​

​

14

Correlational Study

• A study comparing incidence/prevalence of one event against another on a global scale
• Measures that represent characteristics of entire populations are used to describe the disease in relation to some factor of interest (such as age, calendar time, food consumption, drug use and utilization of health services)
• Advantage
• Compares events among nations
• Disadvantage
• Doesn’t compare individuals, so it might lead to overgeneralization.

​

​

Analytical Studies

• Two basic designs:
• Case – control or retrospective study
• Cohort or prospective

​

• NOTE
• There must be a comparison group
• No control No conclusion(NCNC)

Case Control or Case History Study

• A group of affected people is compared to unaffected people(the control)
• It’s a LONGITUDNAL STUDY (like cohort study) because it’s a study over a period of time.
• Subjects are selected based on a particular outcome and a study backwards in time to try to detect the causes or risk factors that may have earlier been reported in a descriptive study
• Subjects are then matched and assigned into the two groups. Subject selected on the basis of disease (e.g lung cancer).
• Sometimes called a retrospective study because of the direction of study

​

Case Control or Case History Study

Advantages & Disadvantages of Case Control

• Advantages
• It is relatively easy to carry out bcos we go back to existing records in the hospital
• It is also rapid and inexpensive
• It requires comparatively few subjects
• It can assist one in studying different etiological factors
• One does not need an ethical clearance
• There is no risk to the subject
• Disadvantages
• It introduces bias
• To select an appropriate control could be difficult
• It may be difficult to distinguish between the cause of a disease and an associated factor

​

Cohort Study

• A cohort is a grp of people who have something in common and remain part of a group over an extended time
• A group of people exposed to a suspected etiological agent are compared with a matched control who have not been similarly exposed. Subject selected on the basis of exposure (aetiological factor; cigarette smoking)
• Follow-up over a period to compare the outcome
• Also a longitudinal study or prospective study

​

Advantages and Disadvantages of Cohort Study

• Advantages
• There is no bias
• The risk can be calculated bcos the incidence can be calculated
• It is effective for studying rare exposures
• It allows the study of the natural history of the disease
• It assists in determining the temporal relationship between the etiological factor & the disease
• Disadvantages
• It takes a long time
• It is expensive
• Large no of subjects are needed
• There could be changes in the standard methods or diagnostic criteria

​

​

Experimental Studies

• Studies in which 1 group is deliberately subjected to an experience compared with a control group with no similar experience
• The gold standard in medicine because it proves causality
• Can be controlled or uncontrolled

Uncontrolled Experimental Studies

• Intervention is not compared with a control
• The aim is to confirm that the Intervention made a difference

Controlled Experimental Studies

• In this study, a drug or procedure is compared to:
1. Another drug
2. Procedure
3. Placebo
4. Previously accepted tx
• The aim is to prove the difference due to treatment
• Blind trial-single or double
• Control could be:
• METHODOLOGY
2. Concurrent or parallel: randomized or non- randomized(quasi)
3. Sequential control: self controlled or cross over
4. External control

B. STUDY POPULATION

• Clinical trials
• Field trials
• Community trials

​

Experimental Studies

Advantages

• Best study type
• Greatest prove of causality
• Gold standard for other design
• Least bias
• Proves best treatment or procedure efficacy

Disadvantages

• Greatest expense
• Long duration
• Unproven facts adopted by community can hinder study acceptance

Overview

End