Class Notes (1,100,000)
US (450,000)
UMD (8,000)
EPIB (7)
Dyer (7)
Lecture 5

EPIB 301 Lecture 5: Data Sources


Department
Epidemiology and Biostatistics
Course Code
EPIB 301
Professor
Dyer
Lecture
5

This preview shows page 1. to view the full 5 pages of the document.
EPIB 301: Data Sources/Surveillances 2/14/17
Quality of Epidemiologic Data:
Quality:
o Quality determined by sources used to obtain data and how completely the data covers
the reference population
o Affects the way data is used/types of analyses performed
Factors:
o Nature of data?
Sources/content
o Availability?
Personal medical records w identifying info cannot be released w/out consent
o How complete is the population coverage?
representativeness
Representativeness: external validity
o Refers to generalizailit of fidigs to the populatio it’s supposed to epeset
External validity of data from HMOs?
Voluntary campus surveys/course evals?
Has the data been collected at other time points/places/people?
Quality Assurance:
o Data security/confidentiality: protect subjects and secure data
o Data timeliness:
o Data completeness: missing data? Complete data set?
o Data accuracy: was everyone paying attention?
o Case detection:
Sources used in epi research
o U.S. Census: provides data, can be used to define denominator in rates with respect to
official estimates of total population size and subdivisions of population by geographical
data
Updated every 10 years
Fact finder
o Vital registration system/vital events: vital events are known as deaths, births,
marriages, divorces, fetal deaths
U.S. collects info on these events routinely
Laws of each state provide continuous and permanent registration system
Each system depends on efforts of physicians, hospital staff, funeral
staff, coroners, medical examiners, etc for original death records
Deaths: mortality data almost complete
Unlikely to go unrecorded
National Center for Health Statistics (NCHS) compiles/publishes
mortality rates
Certificates include demographics, date and place of death, and cause
of death including contributing factors
o Cause of death may not be entirely accurate
find more resources at oneclass.com
find more resources at oneclass.com
You're Reading a Preview

Unlock to view full version

Only page 1 are available for preview. Some parts have been intentionally blurred.

o Diagnostic criteria may lack consistent standards.
o Some conditions may be unreported due to stigma.
Births: include live births and fetal deaths
Presumed to be nearly complete, used to calculate birth rates
Help understand birth defects, length of pregnancy, birth weight,
demographic background of mother
Affeted  othe’s failue to eall pega ees (e: ifetios,
etc)
May miss conditions not detected at birth (ex: congenital anomalies)
o Data from PH surveillance programs: refers to systematic and continuous gathering of
info about occurrence of diseases and other health outcomes
Communicable and infectious diseases
Non infectious diseases
Risk factors for chronic diseases
Used for monitoring
Sudden changes in occurrence
Follow long-term trend/pattern
Changes in risk factors
Identify whether a health problem exists, is improving, is getting worse
Reportable and Notifiable Disease Statistics: physicians/health care providers
must report cases of certain diseases (reportable/notifiable diseases) to health
authorities in a time frame
o Usually infectious/communicable diseases and may endanger
population
o Penalties for violations
Deficiencies: population coverage might be incomplete
No ifo is aailale aout pesos ho deelop disease ad do’t seek
medical attention
o Asymptomatic
Physicians and healthcare providers may fail to fill out proper forms
o Stigma
o Behavioral Risk Factor Surveillance System (BRFSS): used by U.S. to monitor at state
level behavioral risk factors associated w/ chronic diseases
Examples: health risk behavior, preventative health practice, health care access
o Case Registries:
Registry: centralized database for collection of info about a disease
Register: refers to the document used to collect the information
National program of cancer registries
Surveillance, epi, and end results program (SEER)
Health status measures that can be calculated from data sources:
o Life expectancy
o Birth: crude birth rate, general fertility rate
o Death: infant mortality rate, maternal mortality rate, fetal mortality (fetal death rate,
late fetal death rate), perinatal mortality rate
find more resources at oneclass.com
find more resources at oneclass.com
You're Reading a Preview

Unlock to view full version