Lecture 8- cohort studies

cohort

a well-defined group of individuals who share a common characteristic/experience

cohort study

analytical study that compares the risk/incidence of outcome between exposed vs. unexposed, first study where risk/incidence can be calculated

key feature of cohort study

participants are selected based on their exposure status: exposed vs. not exposed, everyone is free of disease, everyone is at risk of disease 

cohort studies- objective 

can calculate incidence of disease, risk= incidence= probability of an event occurring, people without the disease are followed over time into the future to assess disease incidence

cohort studies- design layout

exposure is not assigned by the researcher, at recruitment participants are free of disease of interest, in order to calculate risk you need to be at risk, recruitment based on exposure status

population-based cohort 

select a population based on geographic area, membership of an organization, then assess who has the exposure and who does not have the exposure

exposure-based cohort

selection is based on exposure, such as occupational exposure to pesticides- farmers, compares cohorts with or without different exposures, happen with unique exposure that gen pop don’t usually have 

prospective cohort study

participants are classified according to their exposure status and are observed over time to document the occurrence of new cases of disease, at recruitment participants are free of disease of interest, recruitment is based on exposure status

retrospective cohort study

conceived after exposure and outcome have occurred, exposure has already happened in the past, at the time of exposure, participants do not have the disease of interest, outcome data also recorded

cohort studies- measure of association 

compare the risk/incidence of disease between the exposed and non-exposed

relative risk

risk in exposed/risk in unexposed; quantifies a population’s risk of disease from a particular exposure

relative risk formula

Risk in exposed= A/(A+B)/Risk in unexposed= C/(C+D)

measure of association- wording

The risk of (disease) is (x) times (higher/lower) among (exposed) people compared to (unexposed) people

RR > 1 (if statistically significant)

risk of outcome is (x times) higher among exposed compared to unexposed; positive association, exposure is a risk factor for disease

RR < 1 (if statistically significant)

risk of outcome is (x times) lower among exposed compared to unexposed; inverse association, exposure is protective

RR = 1

risk of outcome is the same among exposed and unexposed, no association between exposure and outcome

cohort studies- advantages

permit direct observation of risk of disease, can study rare exposure, the temporal relationship between exposure and outcome is known and exposure factor well defined (prospective cohort)

cohort studies- disadvantages

expensive and time consuming, participants may be lost to follow-up, exposures can be misclassified in retrospective cohort studies

attributable risk

helps to determine how much of an outcome may be attributable to a particular risk factor in a population exposed to that factor

attributable risk formula

AR = Incidence in exposed a/(a+b) - Incidence in unexposed c/(c+d); per 1,000

attributable risk- wording

among (exposed), (x) cases of (disease) per 1,000 are attributable to (disease), assuming a causal relationship between (exposure) and (disease)

population attributable risk (%)

proportion or percentage of cases in the entire study population that can be attributed to the exposure, gives an idea of the amount of disease that could be prevented in the population if the exposure were to be eliminated, assumes a causal relationship between the exposure and outcome

population attributable risk (%) formula

PAR%= Prev_exp (RR – 1)/[Prev_exp (RR – 1) + 1];

Prev_exp= (a+b)/(a+b+c+d)= x; RR= a(a+b)/c(c+d)= y; 

PAR%= (x)(y-1)/[x(y-1)+1] * 100= answer%