Lecture 7: Research and Methods Flashcards
Differentiate between quantitative and qualitative research. Give examples of each.
1) Quantitative: Focus on numeric data and statistical analysis
-Surveys, experiments, observational studies
-The “how many”
2) Qualitative: Focus on non-numeric data to gain deeper insights into complex social phenomena
-Interviews, focus groups, participant observation
-The “why”
List the 9 basic steps of research
(just know the general flow)
1) ID the problem and formulate a research question
2) Review the recent literature
3) Revise the research question
4) Select the study design
5) Plan data/ analyze strategy
6) Develop and submit the project description to an IRB for approval
7) Collect data
8) Analyze data
9) Present results
Identifying a problem and formulating a research question:
What are the 3 types of research you can choose from? Describe the goals of each
1) Descriptive Research: Characteristics, behaviors or patterns within a given population “snapshot” of the situation without manipulating variables or establish causation.
2) Exploratory Research: Focused on investigating new and uncharted areas.
-Researchers seek to generate a hypothesis and ideas for further investigation.
3) Relational/Casual Research: Aim to explain the relationships between variables.
-Better understand the underlying mechanisms driving theses changes.
What are the two primary approaches to research?
1) EOP
2) PICOT
List + define each part of EOP research
1) Exposures: factors potentially linked to the prevention, development or enhancement of a health status
2) Outcome: A result tied (potentially) to specified exposures (morbidity/mortality)
3) Population: Case studies (individuals), communities, specific groups
`List each part of PICOT research
1) Patient, population or problem: Who/what is the patient/population to be addressed?
2) Intervention: What is the intervention that will be tested?
3) Comparison: What will the intervention be compared to?
4) Outcome: What is the desired or expected outcome of interest?
5) Time Frame: How long will it take to reach your desired outcome?
List some specific aims from the bottom up
1) Broad question abt the outcome or interest
2) Existing knowledge from the literature
3) Selection of population and region of interest
4) Potential sponsorship opportunities
5) Feasibility: time, access to population, partnerships, funding, clarity
6) Final research question
Describe a review of relevant literature (2 key components)
Surveyed to shape a research idea and identify gaps
1) Primary source utilized: Peer-reviewed articles published in reputable academic journals
2) Generalizability: Capacity of a study’s methods to produce comparable results across different populations
Study designs; describe:
1) Cross-sectional study
2) Cohort study (give an example too)
1) Cross-sectional study:
-Collects data at a single timepoint
2) Cohort study: A group of individuals is followed over time to observe the development of health outcomes
-Framingham Heart Study
Study designs; describe:
1) Randomized controlled trials
2) Phenomenology
3) Grounded theory
1) Experimental clinical studies that randomly assign participants to either an intervention or control group
2) Qualitative study design to uncover the essence of a phenomenon as experienced by individuals
3) Qualitative study design to develop theories grounded in the data collected, allowing new concepts and insights to immerge
What does sampling do?
Influences the generalizability and reliability
Sampling types; describe:
1) Random sampling
2) Single-blind study
3) Double-blind study
1) Helps minimize bias and ensures each member of the population has an equal chance
2) Either the participants or the researchers are unaware of certain critical information
3) Both the participants and the researchers are unaware of certain critical information
Sampling types; describe:
1) Purposeful Sampling
2) Saturation
1) Identifying individuals with varied experiences, perspectives or sociodemographic backgrounds
2) New data cease to provide additional insights or perspectives
1) What is informed consent?
2) What are the 4 things it requires?
1) Process in which researchers must obtain voluntary and informed consent from participants
2) Provide and explain purposes of the research
-Time the participant will be expected to participated
-Anticipated risks and discomforts/expected benefits
-Refusing to participate with involve no penalty or loss of benefits
List 4 ethical concerns in research
1) Informed consent
2) Privacy
3) Confidentiality
4) Vulnerable populations
Describe the following concerns in research:
1) Privacy
2) Confidentiality
3) Vulnerable populations
1) Protections to ensure that participants choose what information they want shared
2) Protection of personal information, identities are not disclosed to anyone unauthorized
3) Children, pregnant persons, incarcerated and socioeconomically disadvantaged individuals
1) What does IRB stand for?
2) What is it responsible for?
3) What does it evaluate and why?
1) Institutional Review Boards
2) Responsible for providing protection for human participants
3) Evaluates and approves research
Minimize potential harm, ensures research is benefits are disrupted fairly across diverse populations
Data analysis:
1) What examines geographic patterns for patterns “hot spots”?
2) What involves repeated readings to identify patterns, themes and nuances?
3) What involves labeling and categorizing segments of data based on recurring themes
1) Spatial analysis
2) Data Immersion
3) Coding
Presenting Research Results:
1) Give some examples of journal manuscripts
2) What does it help do?
1) Peer-reviewed journals, validation and scrutiny of research methodologies and results
2) Maintain quality/integrity of the scientific literature
List 2 ways other than journals to present research results
1) Paper and Poster Presentations
-Platform for researchers to showcase their work, engage with pears
-Provides visual and interactive medium for presenting research
2) Reporting in General Media Outlets
-Broader audience beyond the academic and scientific communities
-Potential pitfalls of oversimplification or sensationalism
1) What does epidemiology rely on?
2) Who was the father of epidemiology?
1) Data and statistics
2) John Snow; research on cholera outbreaks 1854
Data and statistics are not absolutes. What are some sources of error?
Bias and confounding variables
Give an example of “science changing”
Breast cancer screenings: originally 50, then 45, now 40 every 2 years
List some aspects of probability
P-value, confidence interval, false negatives & false positives
1) Define P-value
2) Define confidence interval
1) Probability that the observed result could have occurred by chance alone
-0.05 or less is statistically significant
2) Confidence interval: Range of values within the true result
-The more narrow=less likely random
Differentiate between false-negatives and false-positives
1) False-negative: Study finds no effect when there is one
2) False-Positive: Study finds effect that is not real
1) Define secondary prevention
2) Give an example
1) Early detection and Tx
2) Mammograms and BC now starting at 40
Screening:
1) Who should you be screening?
2) No screening test is perfect, but what is ideal?
1) The population at risk to detect the disease early
2) Ideally want high sensitivity (can lead to overdiagnosis)
1) Define rates
2) Define crude rates
3) Define adjusted rates
1) Rates: Proportion of some disease or condition in a group
2) Crude: The actual rates of events (birth, deaths, cause of injury, etc…) in a population without adjustment
3) Adjusted: compares two groups that differ in some important variable by mathematically eliminating the effect of that variable (e.g., age)
What gives greater weight to deaths of young people?
Years of Potential Life Lost (YPLL)
1) Who is a big source of data for public health?
2) What is this a part of?
3) What do they do?
1) National Center for Health Statistics (NCHS)
2) Part of the Centers for Disease Control and Prevention (CDC)
3) Collects data from the states and periodic surveys
1) Providers and nurses sent to conduct physical and dental examinations of selected individuals describes what survey?
2) What does the National Health Care Survey gather?
1) The National Health and Nutrition Examination Survey
2) Data on operations of providers’ offices, community health centers, hospitals, etc
Behavioral Risk Factor Surveillance Survey (BRFSS):
1) What does it obtain?
2) What is notable abt it?
1) Information on health-related behaviors
Diet, physical activity, seat-belt use etc…
2) Information is self-reported
Give some potential uses of data
1) Assessment of the health of a community
2) Raw material for Research
3) Identification of special risk groups
4) Detection of New Health Threats
5) Planning of Public Health Programs and evaluation of their success
6) Preparation of Government Budgets
Describe the collection of data
1) Local Records are sources of data:
Birth certificates
Death certificates
Other vital statistics
2) Data is transmitted:
From local governments to states
From States to National Center for Health Statistics (NCHS is part of CDC)
3) Surveys are sources of data
List some examples of vital statistics
1) Birth and deaths are recorded and filed with the health department
2) Marriages/ Divorces
3) Spontaneous fetal deaths
4) Abortions
1) Birth and deaths are recorded and filed with the health department; then where do they go?
1) Transmitted to the NCHS
Much of the information is confidential
2) Spontaneous fetal deaths: Many are incomplete, especially early
3) Abortions: Many hesitant to report due to confidentiality
What should you evaluate about health information on the internet?
1) Overall site quality (are ads separate from info, etc)
2) Authors
3) Information (sources, etc)
4) Relevance
5) Timeliness (can you tell when published, recency)
6) Links
7) Privacy (is privacy protected?)
1) The US census is mandated by what?
2) What does it serve as? Give examples
3) How often is it conducted?
1) U.S. Constitution
2) The denominator for most public health data:
Age, sex, race, ethnicity
3) Every 10 years
1) What does the census determine?
2) What were 2 unique things abt the 2020 census?
3) What is done between censuses and what’s it abt?
1) Political composition of the U.S. Congress
2) The citizenship status question excluded from 2020 census
-New household categories included for 2020 census
3) American Community Survey is done on an ongoing basis between censuses: Education, housing, health insurance
Why is so much public health data needed?
1) Used for Developing Surveillance Systems:
ex: Youth Risk Behavior Surveillance System (YRBSS)
-Diet, sleep behavior, etc.
2) Used by federal, state and local Agencies
3) Public Health Problems are defined according to available data: Epidemics
Describe the accuracy and availability of data
1) Imperfect: Students drop out of school. Self-reporting can lead to exaggerated data
2) Census is most accurate: Still overcounts and undercounts
3) Information technology increases accuracy and availability
CDC (and most other federal and state public health agencies) make information available over the Internet
Allows prompt response to new outbreaks
Governments have safeguards to protect information on individuals, but what may involve removal of information identifying individuals?
Use of data
Who can deny access to data?
Institutional Review Boards (IRBs)
What is an exception to confidentiality?
Being notified that they have been exposed to a communicable disease
What cause a lot of the privacy rules abt health data to be implemented?
AIDS Epidemic: Fear of discrimination by employers meant many didn’t get tested, so they allowed anonymous testing
Describe the 3 basic steps of public health research
1) Observe a possible association between exposure and illness
2) Develop a hypothesis on the cause-and-effect relationship
3) Test the hypothesis through a formal epidemiological study
“Most rigorous form of intervention study & difficult to control behavior” describes the problems of what kind of study?
Randomized controlled trials (RCTs)
What are 2 issues with cohort studies?
1) Does not require people to change behavior
2) Might lead to false assumptions
Case-control study: describe the problems with it
1) Patients report their behaviors, so might not have accurate reporting
2) Recall bias
List 3 sources of error
Random variation, confounding variables, and bias
Sources of error:
1) Which source is a consistent pattern that can happen as a result of day-to-day variation?
2) What are confounding variables?
3) What are 2 kinds of bias?
1) Random variation
2) Explanations other than the one being studied that may affect a result
3) Selection and recall bias
1) Define selection bias
2) Define recall bias
1) Treatment and control groups of a study are sufficient different due to the way the control-group subjects were selected
2) Systematic error that can happen when participants to not recall previous events or experiences
Give 5 examples of unethical epidemiologic activities
1) Mengele Nazi Experiments
2) Tuskegee Syphilis Study
3) Conflict of Interest: a situation in which the concerns of two different parties are incompatible
4) U.S. Food and Drug Administration (FDA): Requires randomized controlled trials; New Drug Submitted for publication
5) Publication Bias: Likelihood of a study being published based on the study findings
-Positive results were more likely published than negative ones
