##
This **preview** shows page *1*
out of 4 **pages**.

*View Full Document*

End of preview. Want to read all 4 pages?

Upload your study docs or become a GradeBuddy member to access this document.

View Full Document**Unformatted text preview:**

Range of values within which the true value (average or mean) probably fallsThe narrower the confidence interval’s range, the more certain you can be that your results include the population valueDepends upon the size of your sample and the level of certainty you set (95% or 99%)+/- an estimate – think of political pollsStatistical PowerStudies with Low Power:False negative results (find no effect when there actually is one)Studies with High Power:False positive results (find an effect when there isn’t one)Factors that are related to statistical power:Sample sizeMagnitude of the effectAlpha level or level of certainty (p < .05)Error variabilityMeta-analysis: Combine results of comparable studies to increase powerMammography for breast cancerHIV testsSensitivity:Measure the proportion of actual positives which are correctly identified as such; a sensitive test will generate few false negativesSpecificity: Measures the proportion of negatives which are correctly identified as such; a specific test generates few false positivesHigh Risk Strategy: Individual-Level InterventionPros: Appropriate to individual, motivation is higher, cost-effectiveCons: Costs of screening, palliative/temporary care, limited potentialPopulation Strategy: Population-Level InterventionPros: Radical, large potential of population, behaviorally appropriateCons: Small benefit to individual, low motivation, risk/benefit ratio worrisomePBHL 2001 1st Edition Lecture 4Data and AnalysisCollection of DataBirth CertificatesDeath CertificatesNotifiable DiseaseOther vital statisticsTransmitted from local governments to statesTransmitted from National Center for Health Statistics (part of CDC)Uses of Health DataHealth needs identification Analysis of problems and trendsEpidemiologic researchProgram evaluationProgram planningBudget preparationAdminsitrative decisionHealth educationThe CensusServes as denominator for most public health dataDefining the denominator for the total populationDefining population groups by age, sex, race, ethnicityMeasured every 10 yearsAmerican community survey done in between:EducationHousing Health insuranceAccuracy and Availability of DataData collection is imperfectCensus is most accurate; still there are over counts and undercountsInformation technology increase accuracy and availability Statistical sampling is faster and cheaper Confidentiality of DataGovernment uses safeguards to protect individualRequires data protection committeesHIVTesting for HIV is anonymousIndividuals are tracked by unique identifiersStatistics Statistics refers to both the numbers that describe the health of populations and the science used to prevent those numbersIt answers questions by giving probability RatesRates are used extensively in epidemiologyRates put the raw numbers into perspective by relating them to the size of population being consideredAdjust scale of denominator in proportion to how common the disease or condition isAllows to make accurate comparisons across populations of different sizesCrude Rates: rate calculated from raw data (Ex: ACTUAL number of deaths, number of births)Adjusted Rate: An adjustment to the data to make populations being examined equivalent to one another Life Expectancy: Average number of years of life remaining to people at a particular age; most commonly used is the life expectancy at birthYPLL: Years of potential life lost; give greater weight to death of young peopleRelative Risk or Risk RatioRatio of the incidence rate for persons exposed to the factor to the incidence rate for persons in the unexposed groupRR= 1 no associationRR > 1 increased riskRR < 1 decreased riskOdds Ratio ORUsed to measure the strength of association between exposure and disease; odds of one group compared to another groupStrength of Findingsp-value: and confidence intervals are useful in deciding how seriously to take the experimental resultsPower: Probability of how likely you are to find an effect if one existsP-ValueP-Value: The probability of obtaining the observed result by chance alone (significance level)P ≤ .05 means a result is statistically significant (unlikely to have occurred by chance). When p=.05, there is still a 5% chance that the result is wrong.Confidence IntervalsConfidence Intervals: Interval estimate of a population parameter used to indicate the reliability of an estimate◦ Range of values within which the true value (average or mean) probably falls◦ The narrower the confidence interval’s range, the more certain you can be that your results include the population value◦ Depends upon the size of your sample and the level of certainty you set (95% or 99%)◦ +/- an estimate – think of political pollsStatistical PowerStudies with Low Power:False negative results (find no effect when there actually is one)Studies with High Power:False positive results (find an effect when there isn’t one)Factors that are related to statistical power:◦ Sample size◦ Magnitude of the effect◦ Alpha level or level of certainty (p < .05)◦ Error variability◦ Meta-analysis: Combine results of comparable studies to increase powerStatistics of Screening TestsScreening tests are used to detect disease early◦ Mammography for breast cancer◦ HIV tests Two statistical measures related to the performance of the diagnostics tests Sensitivity:Measure the proportion of actual positives which are correctly identified as such; a sensitive test will generate few false negatives Specificity: Measures the proportion of negatives which are correctly identified as such; a specific test generates few false positivesTake AwayCauses within a community are likely to be individual-level differences of susceptibilityCauses between communities are likely to be environmental (social, physical) differences in exposure to risk factorsPrevention Paradox: a preventive measure that brings a large benefit to the community offers little to each participating individual, e.g., immunization, wearing seat beltsHigh Risk Strategy: Individual-Level InterventionPros: Appropriate to individual, motivation is higher, cost-effectiveCons: Costs of screening, palliative/temporary care, limited potential Population Strategy: Population-Level InterventionPros: Radical, large potential of population, behaviorally appropriateCons: Small benefit to individual, low motivation, risk/benefit ratio

View Full Document