1 - Types of Studies

What is epidemiology

• The study of the distribution and determinants of health-related states or events and the application of this study to the control of disease and other health problems

• Involves data collection, analysis, interpretation and dissemination of data (sharing)

• This data is used to describe disease patterns, identify the cause of disease and aids in public health interventions and policies

Survaillance

Ongoing, systematic collection, analysis and interpretation of health data

Public health

• Preventing disease, prolonging life

• Promoting physical health

• Education of individuals

• Organisation of medical services for the early diagnosis and preventative treatment of disease

Scope of epidemiology

• Descriptive epidemiology - describes disease patterns present by person (age, underlying disease), place (ward, school) and time (hours, days, seasonal)

• Host, environment and ages must be identified

• Analytical epidemiology - investigates causes and risk factors, tests hypotheses e.g which exposures are associated with a certain disease

• Outbreak investigation - detecting, confirming and controlling outbreaks

• Screening and surveillance - national screening programmes, mandatory surveillance for diseases

• Measurement of health - incidence, prevalence, mortality, morbidity

• Epidemiology study designs - tools to explore associations and test hypothesis e.g. cohort studies

• Public health interventions - using data to guide public health policies e.g. vaccination programmes

Definitions

• Epidemic - disease occurrence among a population that is in excess of what is expected in a given time and place

• Cluster - group of cases in a specific time and place that might be more than expected

• Endemic - disease or condition present among a population at all times

• Pandemic - disease or condition that spreads across regions

Why carry out epidemiological studies

• Clarifies clinical and demographic characteristics

• Identifies who is at risk

• Provides clues to the cause of disease

• Guides preventative measures and interventions

Descriptive studies

• Describes specific characteristics in a population of interest rather than investigating cause of disease using existing data e.g surveillance data

• Measure prevalence and incidence to identify patterns and trends, and generate a hypothesis

• What - symptoms, lab results

• Who - age, race, gender

• Where - where illness began, exposure occurred

• When - time, patterns

Types of descriptive studies

• Case reports - individual, usually unusual presentation, describes person, place and time

• Case series - group of patient with unusual symptoms, can help identify patterns

• Incidence study - number of new cases identified during a specific time in a specific population

What is an analytical study

• Investigates how and why a health-related event occurs by comparing groups of people to determine weather an exposure if associated with an outcome

• Uses statistics to test hypotheses and quantify association between exposure and outcomes

• Aims to identify determinants (risk factors) relating to agents, hosts and environment

Types of analytical studies

• Experimental - lab or individual trial (seeing is new drug kills bacteria), community or field trial (effectiveness of a drug preventing influenza in the community), randomised control trials

• Observational - cohort, case control, cross sectional, ecologic

Experimental studies

• Investigators can control the exposure / intervention

• Aims to determine the effect of a particular treatment or intervention on an outcome

• Patients who receive no treatment, a placebo or different treatment are the control group

• All participants followed over time to determine outcome

Randomised control trials

• An experimental study in which participants are randomly assigned treatment and control groups which helps reduce the influence of other factors (cofounding variables) on the outcome

• For example if only younger or healthier individuals were chosen their outcomes could be better, randomisation helps distribute characteristics

• Considered the gold standard of research design, particularly for testing medications and treatments

Advantages of RCT

• Provide strong evidence for a casual relationship

• Randomisation reduces influence of cofounding factors

• They allow researchers to directly compare the effects of different interventions

• Particularly useful in determining weather a treatment actually causes an improvement in an outcome

Limitations of RCTs

• Can be expensive

• Participants may differ from general population

• Usually only address a specific and narrow research question

• Ethical limitations e.g. researchers cannot deliberately assign participants to a harmful exposure

Observational studies

• Researcher does not assign or control the exposure, observes exposure and outcomes as they naturally occur

• For example comparing incidence of lung cancer in those who smoke and who do not smoke, without tellin

Cross-sectional studies

• Measures the exposure and outcome at the same point in time which provides a snapshot of a population at a particular time

• Used to determine the prevalence of a disease, weather disease is more common among people who are exposed compared to those not exposed

• Can be difficult to establish if the exposure or outcome came first, limited in their ability to establish causation

• E.g. asking 100 college students weather they currently smoke and have asthma

Cohort studies

• Start with the exposure status, two groups, exposed and unexposed

• Groups are observed to determine if they develop the outcome of interest

• Can determine incidence as participants are followed to see who develops the disease

• Prospective - exposure status noted, followed forward in time to see if outcome occurs

• Retrospective - exposure and outcome have already occurred, researchers use existing data to reconstruct what happens overtime

Strengths and limitations of cohort studies

• Benefits - can establish a temporal relationship between exposure and outcome, can investigate more than one outcome, can measure incidence and calculate relative risk

• Limitations - expensive and time-consuming especially if follow-up is required, inefficient for rare diseases as a large number of people need to be followed to find enough cases

Case control studies

• Start with an outcome/disease rather than exposure, identify people who have the disease (case) and compare them with those who don’t (control)

• Look backwards in time to determine weather the two groups differed in exposure to potential risk factors

Advantages of case control studies

• Useful for studying rare diseases as researchers identify people who already have it

• Also diseases with long latency period as they do not have to wait to see who develops the disease

• Can investigate multiple possible exposures

• Quick and inexpensive

Limitations of case control studies

• Subject to bias, especially recall

• Generally focused on one outcome

• Cannot estimate incidence, not ideal for rare exposures

• Can be difficult to establish if an exposure occurred before the disease

• Finding an appropriate control group can be challenging

Ecological studies

• Compares two groups or populations rather than individual people

• Exposure and outcome measured at the population level

• E.g comparing alcohol consumption and rate of CHD between countries

• Cannot assume that an association observed between populations also applied to individuals within these populations (dk who was exposed and who had outcome)

• E.g cannot conclude that individuals who drank more alcohol were the individuals who developed CHD

Recall bias

• A bias is a systematic problem in how a study is designed, conducted or analysed that can distort the results

• Recall bias occurs when participants remember or report previous exposures differently

• People with a disease may be more motivated to remember possible risk factors than people without the disease

Selection bias

• Occurs when the people selected for a study are not representative of the population they are intended to represent

• E.g. control groups in case control studies not accurately matching the population from which the cases arose

• Can also occur if certain types of cases are more likely to be selected than others

Confounding

• A confounding variable is a factor that is associated with both the exposure and the outcome and can therefore distort the apparent relationship between them

• It can appear as though an exposure causes an outcome when the observed association can be explained by another factor

• E.g. hot weather, ice cream sales and sunburn, just because two things are correlated does not mean one thing causes the other

• Hot weather is the cofounding factor as it influences both