BIOL 437 Week Three Flashcards
study design
- program that directs the researcher along the path of systematically collecting, analyzing and intrepreting data
- allows for descriptive assessment of events and for statistical inference
study types
-experimental
-observational
>descriptive
>analytic
descriptive epidemiology
- examining the distribution of disease in a population
- observing the basic features of it’s distribution
- when, where, who
analytic epidemiology
- testing a hypothesis about the cause of disease by studying how exposures relate to the disease
- observational or experimental
analytic epi-studies require info to
- know where to look
- know what to control for
- develop viable hypotheses
analytic epi- is built around
-the analysis of the relationship between two items
>exposures
>effects (disease)
-looking for determinants or possibe causes of disease
-useful for hypthesis testing
descriptive statistics
-can take on various forms >tables >graphs >numerical summary measures -application of statistical methods makes it possible to effectively describe the public health problem
helpfulness of descriptive epidemiology
- info about a disease
- clues to identify a new disease
- identifies the extent of the public health problem
- obtains a description of the problem that can be easily communicated
- identifies the population at greatest risk
- assists in planning and resource allocation
- identifies avenues for future research
4 types of descriptive studies
- Ecologic studies: population level
- Case reports
- Case series
- Cross-sectional surveys
ecological study
-involves aggregated data on the population level
-ecological fallacy
>when population level are used for individual level
-advantage: annoymous
-disadvantage: can’t have individual correlation
case report
-involves a profile of a single individual
case series
-involves a small group of patients with a similar diagnosis
-provide evidence for longer scale studies
>hypotheis gathering
cross-sectional survey
- prevelance survey
- short period of time
- no follow-up
- all variables measured at a point in time
- no distinction between potential risk factors and outcomes
- good for examining associations between factors
strengths of cross-sectional
- several associations at once
- relatively inexpensive
- short period of time
- control
- biases due to observation and loss to follow up does not exist
weakness of cross-sectional
-unable to establish sequence of events >association NOT causation -influenced by response bias -no follow-up -no incidence or relative risk data
serial surveys
-cross-sectional surveys that are routinely conducted
>regular frequencies
cohort study
-of persons who have been exposed and are followed over time with selected health outcomes
case-control study
-of persons who have been exposed and are followed over time with selected health outcomes
case-control study
- grouping people as cases and controls
- are cases more/less likely than contorls to have had exposures/behaviours
2 methods for age-adjusted rates
- Direct
2. Indirect
direct method
-assume males and females have same rate
indirect method
-age-specific rates are unstable
-standardized morbididty ratio (SMR)
>know total counts
>use rate from other gender
SMR=1
-events observed were same as expected
SMR>1
-more events observed than expected
SMR<1
-fewer events observed than expected
4 types of data
- Nominal
- Ordinal
- Discrete
- Continuous
nominal
- categorical: unordered
- dichotomous: 2 levels
- multichotomous: more than 2 levels
ordinal
-categorical: ordered, scale data
discrete
-quantitative
-differ by fixed amounts
Ex. number of sick kids
continous
-quantitative
-values on a continuum
-interval and ratio
Ex. age, weight
interval
- units of equal magnitude without an absolute 0
- ex. temperature
ratio
-units of equal magnitude, absolute 0
>starting point is 0
Ex. HR, BP, distance
attack rate
- cumulative incidence rate
- diseases or events that affect a larger proportion of the population than the conventional incidence rate
measures of central tendency
- mean
- median
- mode
measures of depression
- range
- interquartile range
- variance
- SD
- coefficient of variation
frequency distributions
- values of a variable and records with each value
- rapid visual assessment
spread
- variation of dispersion
- distribution from a central value
ex. range and SD
shape
- symmetrical vs. assymmetrical/skewed
- “tail” of the distribution-skewed
positively skewed
- tail to right
- central location is to the left
negatively skewed
- tail to left
- central location to right
mode
- value that occurs most in data
- could have more than one
- used for ‘descriptive’ measure
median
- middle value of data set in rank order
- 50th percentile of the distribution
- descriptive
arithmetic mean
-value closest to all the other values in a distribution
-often used in statisical analyses
-best descriptive measure for normally distributed data
>not great for skewed data
geometric mean
-mean of data on a log scale
>normally distributed
-ex. serial dilutions or environmental sampling
standard deviation
- measure of spread
- often used with the arithmetic mean
- ‘variability in a data set’
standard error (SE) of the mean
- variability expected of repeated samples from the same population
- quantifies variation in sample means
- used to determine confidence intervals