Identify Risk Factors

From the slides:

The second step in the public health approach to solving health problem is to identify risk and protctive factors.

• Knowing the leading causes of deaths is a diagonosis, we have to look into what causes these conditions. How do we do that?

Epidemiology

Epidemiology is the study of the distribution and determinants of health-related states or events in specified populations and the application of this study to control health problems.

• Determinants are causes of health condition or things that increase one's risk for that health condition (called risk factors).

Q: How do we know if something increases your risk of disease or injury? A: Epidemiology, specifically Risk Studies

Risk Factor Studies asks "does X cause Y?", "Is X associated with increased risk of Y?". E.g., "does smoking cause lung cancer?", "Is smoking associated with increased risk of lung cancer"

Risk is the probability that harm will occur.

• Risk is measured by calculating the incident rate (proportion of the population of interest who develop a disease or injury during a specified period of time) of the health condition of interest in our population of interest
• incident is the occurence of new acses of a disease or injury. It counts all new cases in the population during that time period, regardless of how long the person lives with the disease or injury. If it began during the time collection period, it gets counted. '

Relative risk (RR) can be calculated by (incident rate in exposed) / (incident rate in no-exposed) where (incident rate in exposed) = (number of exposed with incident) / (number exposed).

Cohort studies: start with healthy people, compare the incident of disease/injury over some follow-up time period in exposed and non-exposed individuals, then calculate relative risk (RR).

Case-control studies: start by identifying the cases, then find healthy controls, compare past exposures in cases and controls, calculate odds ratio (OR) = (odds of exposure in cases)/(odds of exposure in controls).

How do we decide which factors to look at/what do we study? We can see some common patterns from step 1:

• does a certain group of people have higher rates of disease than others
• what do these people have in commonj
• do they liev near each other
• do you work in the same occupation
• do they eat the same things
• do they do the same activites.

Study Designs

Sullivan Study Designs.pdf

Study design is the methodology that is used to collcet the infomration to address the research question. Study design occurs after refining a study objective/research question.

Studies can be organized into two broad types: observational and randmoized.

• observational: observes a phenomenon. Also called descriptive, associational, nonrandomized, or historical studies. Used to alert the medical community or generate hypotheses.
• randomized: intervene and measure a response. Also called analytic or experimental. Used to test hypothesis or evaluate the effect of an intervention.

Observational

Case Report/Case Series

Case Report is a very detailed report of the specific features of a particular participant or case.

A case series is asystematic review of the interesting and common features of a small collection of cases.

Commonly used to generate hypotheses which are further tested because this type of observational study is easy to conduct but criticized as unplanned, uncontrolled, and not designed to answer a specific question.

Cross-Sectional Survey

Conducted at a single point in time. The group of participants are non-randomized and useful for estimating the prevelance of specific risf actors or prevalance of disease at a point of time.

Pros: Easy to conduct, large in size, and ethical, allowsfor estimates of risk factors, diseases, practices, or opinions. Limitatons: both the exposure/development of a risk factor and the outcome have occured. It is limited by the fact that the study is conducted at a singular point in time so you cannot measure temporal relationship. Non-response is also an issue because this introduced bias and limit generalizability.

Cohort Study

Involves a group of individuals who meet a set of inclusion criteria. This group followed and associateions are made between a risk factor and a disease.

Data about the risk factor/exposure can happen in two ways:

• a retrospective cohort study obtains the risk factors by lookign back in time
• a prospectve cohort study is more common and it allows investigators to assess temporal relationships to estimate the incident of a disease.

Confounding is a distortion of the effects of an exposure or risk factor on the outcome by other characteritsics. For examples, even though smokers coud be found to be more likely to develop cardiovascular disease, it could also be the case that smokers are less likely to exercise, have higher cholesterol, etc. The relationship is more complex.

Case-Control Study

Participants are identified based on their outcome status. We select a set of cases/persons, and then select a set of controls. Then, we assess exposure or risk factor status retrospectively. We hypothesize that the exposure or risk factor is related to the disease and evaluate this by comparing the cases and controls with respect to the proportions that are exposed.

Positive features: cost and time efficient. good if many harmful exposures are in play.

Challenges: common sources of biases

• misclassification bias: wrong classification of outcome or exposure status.
• Selection bias: selection of cases causes distorition of the association between exposure and outcome status.
• Recall bias: cases and controls differentially recall exposure status.

Nested Case-Control Study

Designed from cohort study and is a type of case-control study.

YRuna cohort study and then zoom into those that have the problem.

Randomized Study Designs

More resistant againstbias and confounding

Randomized Controlled Trial of Clinical Trial

RCT randomizes is participants to one of several cmopaprison treatments or groups. One group randomly gets an experimental drug and other gets a placebo (placebo-controlled trial). If a placebo is not ethical, then we give them a medication currently available and considered the standard of cacre (this is called an active-controlled trial).

The Crossover Trial

Each participant is assigned two orm ore treatments sequentially. Each participant receives both treatments. Wash-out period in between.

Everyone is their own control.