Observational study- Week 4 Flashcards
Observational study
In observational studies, the investigator measures or records events but does not intervene. Observational studies can answer basic who, what, when and where questions. Observational studies can be divided into descriptive or analytical studies.
Good to generate a hypothesis
Descriptive study designs
Descriptive studies include ecological and case study designs. As the name suggests descriptive study designs describe the data but do not make any inferences about the relationships between the data (e.g. cannot establish association or causation).
Analytical study designs
Analytical study designs will include data analysis that investigates the relationships or associations between independent (exposure) and dependent (outcome) variables. There are three major types of observational analytical study designs:
Cross-sectional studies
Case-control studies
Cohort studies
What is an ecological study?
An ecological study is an observational study in which at least one variable, either exposure or the outcome, is measured at the group (not individual) level. This type of study is appropriate for the initial investigation of the causal hypothesis, although the level of strength is low.
Examples of group-level measures include average UV exposure at a specific geographic location or even mean blood glucose level of patients in a medical clinic
Strengths of ecological study
Useful for generating hypotheses
Compare populations in different places at the same time
Compare the same population at different times
Fast, easy and cheap
Limitation of ecological study
Link between exposure & effect at the individual level can’t be made – may confuse characteristics of group for characteristics of individuals
Be careful of ecological fallacy
Highly susceptible to bias and confounding
Associations not causation
ecological fallacy
An ecological fallacy is seen when inappropriate conclusions are made about individuals from the aggregated data – hence any association observed at a group level does not necessarily mean that the same association exists for a single individual from that group.
What is a case study?
A case study, or case report, is one that reports observations on a single individual. Cases usually include descriptions of events for a person seeking healthcare such as:
unexpected complications unexpected success failure of treatment conditions that present very rarely A case series is an aggregate of individual case studies, usually all receiving the same intervention, with no control group.
Strength and limitation of case study
strength: Identification of rare diseases/exposures which leads to further rigorous studies
Limitation:Selection bias
Confounding
What is a cross-sectional study?
In a cross-sectional study, a defined population is observed at a single point in time or time interval and exposure and outcome are determined simultaneously. Data can be used to assess the prevalence of acute or chronic conditions in a population. It may not always be possible to distinguish whether the exposure proceeded or followed the disease.
Strength of cross sectional study
Measures prevalence (existing cases) of disease
Exposure & Outcome measured at the same time
Data on all variables is only collected once
Can investigate multiple exposures at once
Cheap, inexpensive and easy
Useful for understanding disease etiology and for the generation of hypotheses
LIMITATION
Unable to measure the incidence of disease
Which came first, exposure or effect? Unable to determine
Only a snapshot – the situation may provide differing results if another time-frame had been chosen
Not suitable for studying rare diseases
Interpret associations with caution – difficult to make causal inference
Case-control study
two samples are selected where the patients in one sample have the disease (cases) and the patients in another sample do not have the disease (control).
Then we look back and determine how many from each group have/had the risk factor.
In a case-control study since the risk factors are investigated retrospectively, the study is often called a retrospective study.
Controls = from the same population as cases, but no disease
Often age/gender match cases and control
OR
used to estimate association between exposure and disease in a case control study
Odds ration = odds of being exposed (Case)/ Odds of being exposed (control)
OR>1 the risk factor can be possible cause of the disease
OR<1 Protective factor
OR= 1 no association
Strength of case control study
cheap, easy, fast
Investigates causes of disease – especially rare disease (e.g. Thalidomide)
LIMITATION of case control study
Identifying and enrolling control subjects can be difficult
not good to study rare exposure
Retrospective design: increased probability of recall bias, selection bias and measurement error
What is a cohort study?
A cohort is a group of people who share a common characteristic or experience within a defined time period. Comparison groups may be another cohort of persons thought to have had little or no exposure to the substance under investigation, but otherwise similar. Also, subgroups within the cohort may be compared with each other.
Prospective cohort study
Investigators conceive and design the study, recruit subjects, and collect baseline exposure data on all subjects before any of the subjects have developed any of the outcomes of interest.
Subjects are then followed into the future in order to record the development of any of the outcomes of interest
Start with people free from disease, who are exposed/unexposed to different factors
See how many people develop the disease in each group
STRENGTH of cohort study
Identify the natural history of a disease/disorder: subjects are enrolled and have baseline data collected before any of them develop any of the outcomes of interest
Identify the temporal sequence between cause and outcome
Good for rare exposures and common outcomes
Very rigorous epidemiological design:
Data analysis cannot take place until enough ‘events’ or ‘outcomes’ have occurred
LIMITATION of cohort study
Selection bias Loss to follow-up (exposure status) Insufficient to study rare diseases Expensive and time-consuming Exposed and non-exposed subjects differ on important predictors of outcome
Defining a cohort
Need to be accessible
• Need to be outcome free prior to exposure
• Needs to be some within the group who will potentially be exposed and
thus may present with the outcome
• Need to be some within the group who don’t present with the outcome
Cohort studies do not have to include people who are typical of a
population.
Follow-up of a cohort
• Length needed varies depending on: • Objectives of the study • Outcome(s) being assessed • The longer they go for, the potential for more information but have the problem of loss to follow-up
• Advantages of a cohort study
- Identify the natural history of a disorder
- Identify the temporal sequence between cause and outcome
- Good for rare exposures and common outcomes
- Very rigorous epidemiological design
Limitations of a cohort study
- Selection bias
- Exposed and non-exposed subjects differ on important predictors of outcome
- Insufficient to study rare diseases
- Loss to follow-up (exposure status)
- Expensive and time consuming
An analysis for prospective and retrospective cohort studies is always done retrospectively, because a span of time has to have elapsed before you can compare incidence.
true or false
True
what is the common measurement in Cohort study?
Relative risk (RR) = risk of disease (exposed)/ Risk of disease (not exposed)
RR>1 increased risk of the disease
RR<1 lower risk
RR=1 risk the same
British doctors study
Does smoking cause cancer?
• Who: Sir Richard Doll & Sir Austin Bradford Hill
• When: Began 1951 follow-up 1957, 1966, 1971, 1978, 1991, 1998, 2001. Total 50-years
of results.
• How: World’s 1st large prospective cohort study.
• Questionnaire on smoking habits to all registered British Doctors n= 59,600.
• 40,701 completed responses (34,494 males & 6,207 females)
• Limited sample size and limited tobacco consumption females were excluded
• What: Assessed the lifelong effects of smoking and smoking cessation
• Impact: Demonstrated link between tobacco smoking and cause specific mortality:
• risk of death from lung cancer (1954),
• myocardial infarction and chronic obstructive pulmonary disease (1956)
• Those who continued to smoke on average lost 10 years of life.
• Stopping at age: 60 years gained 3 years, 50 years gained 6 years, 40 years gained 9 years & 30 years gained 10 years
Retrospective cohort studies
nvestigators looks back in time at archived or self-report data to examine whether the risk of disease was different
between exposed and non-exposed patients
• Retrospective cohort studies classify study participants on the basis of whether or not they were exposed to the
factor under investigation (same as prospective cohort)
• BUT, the research is started after both the exposure and the outcome (e.g. disease) have already occurred.
Retrospective cohort studies
• Direction of the study is still forwards in time
• Still start with the exposure (not the disease!)
• Identify a population with an exposure defined at a time in the past and follow them
through to document the outcome
• Need unbiased documentation of an exposure from the past, such as employment
records from a particular industry (for example the military)
• Such detailed, unbiased documentation is uncommon
A case-control study may have a particular advantage over a cohort study when the disease in question is: Select one: a. Rare b. Infectious c. Fatal d. Unknown e. Virulent
a. RARE
A research team wishes to investigate a possible association between smokeless (chewing) tobacco and mouth lesions in professional baseball players. At a training camp, they ask each player about current and past use of smokeless tobacco, cigarettes and alcohol, and a dentist notes the type and extent of lesions in the mouth.
What type of study design is this?
Select one:
a. Cross sectional study
b. Cohort study
c. Case series
d. Ecological study
e. Case control study
a. Cross sectional
In a study that begun in 1965, a group of 3000 adults in Ballarat were asked about alcohol consumption. The occurrence of cases of cancer was studied in this group between 1981 and 1995.
What type of study design is this?
Select one:
a. Case control study
b. Retrospective cohort study
c. Cross sectional study
d. Prospective cohort study
e. Ecological study
d. Prospective cohort study
To determine if vitamin D (the “sunshine” vitamin) is associated with reduced levels of ovarian cancer, an epidemiologist identified 10 geographic regions in Sweden and obtained data on annual days of sunlight and ovarian cancer rates for each region. The findings indicated that those regions with the most days of sunlight per year had significantly lower rates of ovarian cancer than those areas with the fewest days of sunlight per year.
What type of study design is this?
Select one:
a. Case series
b. Cross sectional study
c. Ecological study
d. Case control study
e. Cohort study
c. Ecological study
why? Study was done on different population at different time
In a small pilot study, 12 women with endometrial cancer and 12 women with no apparent disease were asked if they had ever used oestrogen. Each woman with cancer was matched by age, race, weight and parity (number of children) to a woman without disease.
What type of study is this?
Select one:
a. Ecological study
b. Case control study
c. Cross sectional study
d. Case series
e. Prospective cohort study
b. Case control study
why? case was identified (women with endometrial cancer) and was asked if they were exposed to Risk factor (Oestrogen) and was compared with control ( no endometrial cancer)
A total of 825 insulation workers employed between 1941 and 1944 was identified from the personnel records of three large insulation manufacturing plants in the Western Australia in 1975. During the period 1941-75, 26 deaths from lung cancer were discovered among the workers. Only six lung cancer deaths, however, were reported among a comparable group of 700 co-workers who did not work with insulation during the same time period. The investigators postulated that exposure to the insulation material increased the risk of lung cancer.
What type of study is this?
Select one:
a. Retrospective cohort study
b. Prospective cohort study
c. Cross sectional study
d. Case series
e. Case control study
a. Retrospective cohort study
Retrospective cohort studies are a type of observational research in which the investigator looks back in time at archived or self-report data to examine whether the risk of disease was different between exposed and non-exposed patients
A total of 5 patients with thyroid cancer are identified and surveyed by patient interviews regarding past exposure to radiation.
What type of study is this?
Select one:
a. Prospective cohort study
b. Case series
c. Case control study
d. Ecological study
e. Cross sectional study
B. Case series
why? small sample, no control group
True or false, cross-sectional studies are better suited to generating hypotheses about exposure-disease relationships than to testing such relationships?
Select one:
True
False
True
why? cross sectional study cannot give directional relationship e.g. did the risk factor or exposure caused the disease or vice versa
True or false, cross-sectional studies can examine multiple exposures and multiple diseases.
Select one:
True
False
True
True or false, ecological studies can yield estimates of an individuals risk of disease or death.
Select one:
True
False
false
- You may remember that three years ago there was a multistate outbreak of illnesses caused by a specific and unusual strain of Listeria monocytogenes. As part of the investigation of this outbreak, CDC workers checked the food histories of 20 patients infected with the outbreak strain and compared them with the food histories of 20 patients infected with other Listeria strains. This study design is best described as which one of the following:
a. Analytical, experimental
b. Analytical, observational, case-control
c. Analytical, observational, cohort
d. Descriptive
b.Analytical, observational, case-control
2. A published study follows a large group of women with untreated dysplasia of the uterine cervix, documenting the number who improve, stay unchanged, or progress into cervical cancer. This study design is best described as which one of the following: Analytic, experimental Analytic, observational, cohort Analytic, observational, case/control Descriptive, observational
Descriptive, observational
3. A community assesses a random sample of its residents by telephone questionnaire. Obesity is strongly associated with diagnosed diabetes. This study design is best described as which one of the following: Case-control Cohort Cross-sectional Experimental
Cross-sectional
4. Based on a list of residents from election rolls, 2/3 of men in a large city are invited (including repeated educational urgings) and 1/3 of men are not invited to be screened by PSA blood test for prostate cancer. Over the next 10 years the two groups are compared as to the rate of death from prostate cancer. This study design is best described as which one of the following: Case-control Cohort Cross-sectional Experimental
Experimental
- The initial studies establishing maternal diethylstilbesterol (DES) intake as a cause of vaginal adenocarcinoma in female offspring were case-control studies. This was probably largely because:
A couple of decades ago cohort studies hadn’t been invented.
A woman taking DES was always rare.
The disease outcome is rare.
The investigators had probably just happened to have a number of cases in their practices.
The disease outcome is rare.