STUDY DESIGN

OYETUNDE ADERONKE
OUTLINE

• Introduction/definition
• Definition of terms
• Types of designs
• Observational studies
• Experimental studies
• Conclusion
• Practice test
• References
INTRODUCTION

• Once a study or research question has been defined, the next step is to choose the study
design that most effectively and efficiently answers the question.
• A study design, therefore, is the methodology that is used to collect information and
answer a research question.
INTRODUCTION

• A study design is a specific plan or protocol for conducting the study, which allows the
investigator to translate the conceptual hypothesis into an operational one.
• It is a framework or a set of methods and procedures used to collect and analyze data on
variables specified in a particular research problem.
INTRODUCTION

• The goal of every research is to:

• To quantify (e.g. prevalence of a disease in community)
• To compare (which intervention is better?)
• To predict (who gets cancer?)
• To assess association
• To explore aetiology (exposure causing outcome)
INTRODUCTION

• The type of study design chosen depends on:

• What is the research question/ objective?
• Time available for study
• Resources available for the study
• Common/rare disease or production problem
• Type of outcome of interest
• Choosing an established design gives you a huge head start in design, analysis and eliminating
biases.
DEFINITION OF TERMS

• Research: It is a diligent enquiry or examination to seek or revise hypothesis, facts,

principles, theories or interventions.
• Hypothesis: is an educated guess about an association that is testable by a scientific
investigation. It tends to be broad initially, but is then refined to have a narrower focus.
DEFINITION OF TERMS

• Bias: A systematic error that introduces uncertainty in estimates of effect or association.

• Blinding: A state whereby a participant is unaware of their treatment status.
• Confounding: Complex relationships among variables that can distort relationships
between risk factors and the outcome.
• Matching: A process of organizing comparison groups with similar characteristics.
DEFINITION OF TERMS

• Placebo: An inert substance designed to look, feel and taste like the active experimental
treatment.
• Protocol: A step-by-step plan for a study that details every aspect of the study design and
data collection plan.
• Randomization: A process by which participants are assigned to receive different
treatments (usually based on probability schemes).
RESEARCH PROCESS
TYPES OF STUDY DESIGNS

• Observational vs Interventional
• Descriptive vs Analytical (Inferential)
• Cross-sectional vs Longitudinal
• Prospective vs Retrospective
• Primary vs Secondary
TYPES OF STUDY DESIGNS
TYPES OF STUDY DESIGNS

• Non intervention/observational studies: The researcher describes or analyzes researchable

objects or situations, but does not intervene.
• Intervention studies: The researcher manipulates researchable objects or situations, and
then measures the outcome of the manipulation.
STUDY DESIGNS
 DESCRIPTIVE STUDIES

• These studies involve the systematic collection, analysis and interpretation of data to give
a clear picture of a particular situation.
• The wealth of data obtained in most descriptive studies allows the generation of
hypothesis, which can then be tested by analytical experimental design.
• Both qualitative and quantitative techniques may be used.
DESCRIPTIVE STUDIES

• Descriptive epidemiology aims to answer the questions: Who, Where, and When?
• Hence, we organize and summarize data according to: Person, Place and Time.
• These three characteristics are sometimes called the Epidemiologic variables.
• This information provides important clues to the causes of the disease, and these clues
can be turned into testable hypotheses.
DESCRIPTIVE STUDIES

• Characterize who, where, or when in relation to what (outcome).

• Person: characteristics (age, sex, race, tribe, occupation) of the individuals affected by the
outcome.
• Place: geography (residence, work, hospital) of the affected individuals.
• Time: when events (diagnosis, reporting, testing) occurred.
ANALYTICAL STUDIES

• Analytical studies are used in epidemiological investigations to test specific hypotheses.

• The term ‘analytical’ implies that the study is designed to establish the cause of a disease
by looking for association between exposure to a risk factor and disease occurrence.
• Analytical studies test hypotheses about exposure-outcome relationships, by answering
the questions: Why and How?
BASIC QUESTIONS IN ANALYTICAL STUDIES

• Are exposure and disease linked?

• Look to link exposure and disease by answering the following questions:
• What is the exposure? (Define the Case)
• Who are the exposed? (Population at Risk)
• What are the potential health effects?
• What approach will you take to study the relationship between exposure
and effect? (Study Design)
DESCRIPTIVE VS ANALYTICAL
QUANTITATIVE VS QUALITATIVE STUDIES

• Quantitative studies are those in which the data can be analysed using conventional
statistical methods.
• Qualitative studies are used to gain insight into domains such as attitude or behaviours.
Information here cannot be analysed using conventional statistical methods and are useful
for generating hypotheses.
 QUALITATIVE STUDIES

• They are an investigation of meaning and processes.

• Ask opinions rather than ranking feelings on a scale.
• Study behaviour from subjects’ perspectives.
• Leads to a better understanding of how the subject think, feel or act.
• Can complement quantitative studies.
• Should aim, as far as possible, to study the subjects in their own environment.
• Example: Patients’ perceptions of doctors’ advice to quit smoking.
PRIMARY AND SECONDARY STUDIES

• Studies can be further categorized as primary or secondary research.

• In primary research, new data are gathered from or about participants or their
environment.
• Secondary research involves the generation of new data from existing studies.
• The generation of evidence by combining results from different studies (as occurs in a
systematic review) is secondary research.
PRIMARY AND SECONDARY STUDIES
TIME AND STUDY DESIGN

• The time period over which a study is conducted is a fundamental issue in study design.
• Studies can be described as Cross‐sectional, so carried out at a point in time.
• Longitudinal studies which follow participants over time, either by looking backward
from the present time (retrospective) or looking forward (prospective).
• The issue of time is crucial, as cause and effect can only be inferred from longitudinal
studies
TIME AND STUDY DESIGN
CASE REPORT/SERIES

• This is the simplest type of research report, it is a description of a single case or series of
cases, typically of an unusual finding or alternative means of managing a condition.
• It is often accompanied by a brief literature review that describes similar cases reported
previously and a discussion of the implications of the case or cases reported.
CASE REPORT/SERIES

• Ghoneim and coworkers (2011) presented information in a series of 20 adult patients with
nocturnal enuresis. The authors wanted to investigate the correlation between the depth of
sleep and the degree of nocturnal enuresis. They concluded that the relationship between
the depth of sleep and the degree of nocturnal enuresis is very strong.
CASE REPORT/SERIES

• Small number
• 20 patients

• Description of interesting observation

• Nocturnal enuresis in adults.

• Generates hypothesis
• Correlation between degree of enuresis and depth of sleep.

• No control.
• Covers a short time.
ADVANTAGES OF CASE REPORTS

• Useful for describing unusual events or rarely encountered situations.

• Educational – may help those who subsequently encounter a similar case.
• May be the first report of a never before recognized condition. For example, hairy
leukoplakia, one of the oral manifestations of Acquired Immune Deficiency Syndrome
(AIDS), was initially reported in a series of case reports.
DISADVANTAGES OF CASE REPORTS

• They are reports of only one or a few cases and are anecdotal and subjective in nature.
• There is no control for bias and confounding.
• They cannot be used to draw conclusions about the cause, prevention or definitive
management of the condition described.
CROSS-SECTIONAL STUDIES (SURVEYS)

• Cross‐sectional/prevalence/epidemiologic studies or surveys are commonly used in dental

public health. They can be used to:
• Estimate the prevalence of a disease/condition/habit.
• Report the current health status of a defined group.
• Establish reference ranges.
• Determine the cut‐off points for a diagnostic test.
CROSS-SECTIONAL STUDIES (SURVEYS)

• Characters:
• Analyze data collected on a group of subjects at one time rather than over a period of time.

• Answer the question :

• “ What is happening right now?”

• It is best for determining the status of a disease or a condition such as HIV in a given
population.
CROSS-SECTIONAL STUDIES (SURVEYS)

• Can be used to determine how common a condition is – the prevalence.

• Can be used to determine access to services at a given point in time.
• Can be used to gather current opinion (opinion poll).
• Are easy to conduct compared to longitudinal studies.
DISADVANTAGES

• As they lack a temporal element, cross‐sectional studies cannot be used to make causal
inferences.
• Provide only a “snapshot in time” of a disease or process which may be misleading.
• Impractical for rare diseases.
• One can miss diseases still in latent phase.
• Recall of previous exposure may be difficult.
CASE CONTROL STUDY

• It is an observational study design that at the outset selects cases (individuals with the
disease or condition of interest) and matches these with controls (individuals who do not
have the disease or condition of interest).
• Cases and controls are matched for confounding factors such as age and gender.
• The researchers then examine the difference in exposure to suspected risk/aetiological
factors of both cases and controls. Case‐control studies are retrospective in nature.
CASE CONTROL STUDY

• Definition:
• The study contains 2 groups:
• Case group: contains the disease.
• Control group: dose not contain the disease.

• Characters:
• Always controlled
• Always retrospective ( Backward looking )
• Covers extended period of time (longitudinal)
• Answers the question: “ What happened?”
ADVANTAGES

• The quickest and least expensive.

• Ideal for investigators who need preliminary data prior to writing a proposal for a more
complete, expensive, and time – consuming study.
• Good choice for someone who needs to complete a clinical research project in a short
time.
COHORT STUDY

• This is a form of observational study in which the participants are followed over a period
of time from the starting point.
• A cohort study is prospective in nature.
• At the beginning of the study the participants are free of the disease/condition of interest,
but vary in their exposure to possible risk factors.
COHORT STUDY

• Over time, they are examined and separated into those who do and those who do not
develop the condition of interest.
• The differences in exposure to risk factors are observed and the relative risk of
developing the disease depending on exposure to the examined risk factors can be
calculated.
INDICATIONS

• Cohort studies are indicated when:

• There is good evidence of an association between exposure and disease as derived from:
• Clinical observations
• Supported by descriptive and case control studies
• The exposure is rare, but the incidence of disease is high among those exposed.
CONSIDERATIONS IN SELECTION OF COHORTS

• The cohorts must be free from the disease under study

• Both study and control cohorts should be equally susceptible to the disease under study
• Both groups should be comparable in respect of all the possible variables that may
influence the frequency of disease, and
• The diagnostic and eligibility criteria for the disease should be defined beforehand
STAGES IN CONDUCTING A COHORT STUDY

• Selection of study subjects

• Collation of exposure data
• Selection of comparison groups/controls
• Follow up
• Analysis
ADVANTAGES

• The time sequence of exposure to the suspected risk factor and development
of the disease can be determined.
• More than one outcome can be considered.
• This design allows the risk of disease to be measured directly.
• It is possible, given sufficient resources, to measure information about
exposure to risk factors in detail and also to examine exposure to a range of
risk factors.
ADVANTAGES

• There is likely to be less chance of recall and selection bias than in case ‐control studies.
• Exposure is measured prior to outcome and so helps avoid bias
DISADVANTAGES

• Because the participants have to be followed up for a long period of time, cohort studies
are expensive to conduct.
• If the disease or condition is rare, it is necessary to follow up a large number of
individuals (hence case‐control studies are generally better for investigating rare
conditions).
DISADVANTAGES

• Loss of participants over time can introduce bias. This is particularly a problem if the loss
is linked to a risk factor.
• It is important to maintain consistency of measuring either exposure to the risk factor or
the disease of interest for the duration of the study (which may in some cases be years).
• Disease outcome or risk factors may themselves change over time.
RANDOMIZED CONTROLLED TRIAL (RCT)

• This is an interventional study and is prospective in nature.

• Study participants are identified according to predefined inclusion criteria.
• The participants are then assigned at random to either a test group or a control group.
• The test group then receives the intervention under test, while the control group receives
no intervention, a placebo intervention or an existing treatment.
RANDOMIZED CONTROLLED TRIAL (RCT)

• The participants are followed for a period of time and the outcome
measured at the end.
• The change from baseline in both test and control groups is measured. The
test and control groups are then compared to determine whether there is any
difference in outcome in the test and control groups.
RANDOMIZED CONTROLLED TRIAL (RCT)

• In this way, a conclusion can be reached on whether the intervention or treatment applied
to the test group had a more desirable outcome than the intervention received by the
control group.
ISSUES RELATED TO RCTS

• Randomization
• Stratification
• Blinding
• Open label
• Single blind
• Double blind
• Triple blind

• Trial arm
• Training and calibration
ADVANTAGES

• Are considered the ‘gold standard’ of study design, largely because of the
degree to which they can control for bias and confounding factors.
• Allow randomization and can therefore minimize the effect of confounding.
• Can be ‘blind’ and therefore minimize the effect of bias.
• Are often multicentre – that is, participants are recruited from more than
one site.
• Participants in RCTs get excellent care (even if in the control group).
DISADVANTAGES

• Expensive to conduct.
• Require specialist skills to organize and conduct.
• Recruitment of sufficient numbers of study participants in a defined period of time can be
a challenge.
• Participants can be lost to follow‐up.
• RCTs can only be used to examine interventions with a positive outcome for the
participants.
RANDOMIZED CONTROL TRIAL.

• In addition to the conventional parallel, two‐arm randomized controlled trial, there are
two other forms of interventional study design that are encountered in dental research.
• These are both forms of controlled clinical trials in which the participants act as both test
and control:
• A split mouth trial
• A cross‐over trial.
SYSTEMATIC REVIEW

• A systematic review is a form Secondary research which involves the generation of new
data from existing studies and obtaining evidence by combining results from different
studies.
• Efforts are made to identify not only information published in standard academic
journals, but also that from the ‘grey literature’, not controlled by commercial publishers.
Data are extracted from individual studies.
SYSTEMATIC REVIEW

• Provided that a common outcome measure has been used and that there are sufficient
studies, data can be combined statistically (Meta‐analysis) to give an overall effect.
ADVANTAGES OF SYSTEMATIC REVIEWS

• Provide an overall summary of the effectiveness of an intervention or contribution of a

risk factor to a disease/condition.
• Overcome bias inherent in traditional narrative reviews, where the search strategy,
inclusion and exclusion criteria and method for reaching a conclusion are not transparent.
LIMITATIONS OF SYSTEMATIC REVIEWS

• Are reliant on a sufficient number of studies having been conducted with a similar
outcome measure.
• The proper conduct of a systematic review requires researchers with a range of skills, e.g.
literature search skills, as well as statisticians, research methodologists and clinicians.
CONCLUSION

In designing a study and appraising the results of studies conducted by other research
groups; it is important to recognize the strengths and the limitations of the different types of
study design that can be used so as to choose the most appropriate design.
REFERENCES

• Community Dentistry Manual (2016-2018) by Prof E.O Ogunbodede and Dr M.A

Adedigba, Dept of Community Dentistry, OAUTHC
• Update topic on Principles and Guidelines of Evidence-based Dentistry, by Dr W.L
Adeyemo.
• A Practical Manual of Public health Dentistry by CM Marya:
• Dental Public Health at a glance by Ivor G. Chestnutt
• Descriptive and Analytic studies by Department of Health and Human Services (CDC)