Case-Control Studies HLTH 4104
Popular in Epidemiology
Popular in Liberal Arts
verified elite notetaker
This 4 page Class Notes was uploaded by Mayah Baker on Wednesday October 28, 2015. The Class Notes belongs to HLTH 4104 at University of North Carolina - Charlotte taught by Dr. Huber in Summer 2015. Since its upload, it has received 27 views. For similar materials see Epidemiology in Liberal Arts at University of North Carolina - Charlotte.
Reviews for Case-Control Studies
Report this Material
What is Karma?
Karma is the currency of StudySoup.
Date Created: 10/28/15
Epidemiology Notes October 19 2015 CaseControl Study Design gt A type of observational study gt Review 0 Study subjects are selected based on disease 0 Researchers obtain exposure history 0 Sources of cases Patient rosters Death certi cate Disease registries 0 Selection of Controls It is always useful to think of a casecontrol study as being nested within some sort of a cohort Le a source population that produced the cases that were identi ed and enrolled In view of this there are two key principles that should be followed in selecting controls 1 The comparison group quotcontrolsquot should be representative of the source population that produced the cases 2 The quotcontrolsquot must be sampled in a way that is independent of the exposure meaning that their selection should not be more or less likely if they have the exposure of interest 0 Strengths Ef cient inexpensive Good for rare diseases multiple exposures 0 Limitations Temporal sequence the order in which events unfold in time Control selection is bias Information is bias gt Rationales 0 There are more exposures in cases than in controls gt Steps in conducting a casecontrol study 0 Identify group of incident cases of disease 0 Draw sample of controls 0 Determine exposure status 0 Compare prevalence of exposure gt Example 0 Association between alcohol use and stillbirths 0 Obtain list of women who delivered stillbirths infants at a hospital in Charlotte 0 Take random sample of live born infants delivered at the same hospital during the same time period 0 Contact the women regarding their alcohol use prior to pregnancy 0 Seems as though this would be very simple but selecting cases and controls can be tricky 0 Possible case de nition Women between ages 1835 years admitted to Carolinas Medical Center 101102 who have con rmed stillbirth on medical records gt Case Selection 0 Need to consider who the case is 0 Develop strict criteria Age Race Sex gt Incident cases vs Prevalence cases 0 Incident case new cases of diseases l easy for them to remember exposure 0 Prevalence cases have had disease for some time Use these with rare diseases Seldom used when looking etiology missing the people who died gt Where to nd cases 0 Populationbased cases All cases in a population in a speci ed timer pedod Advantage Give entire picture of diseases 0 Can compute IR Disadvantages o Expensive 0 Low participation rate 0 Hospitalbased Cases All cases diagnosed in hospitals over a speci ed time period Advantages Easy to identify cases 0 lnexpengve Disadvantages 0 Being a case is dependent on medical services 0 Hospital may be a referral center Physician of ces Health maintenance organization HMOs Places of employment Military service records gt Where to nd controls and Control Selection 0 How you collect data has to match for both cases an controls 0 Population This sample would come from the same place that your cases came from eg random digit dialing birth and death certi cates voting lists Advantages Gives entire picture of disease 0 Direct computation of IR Disadvantages o Expensive 0 Lower participation rates Dif cult to contact healthy people 0 Quality of info may not be comparable 0 Hospitals Controls seeking medical care at the same institutions for a different reasons than what you are asking them of as the cases Advantages o Easily identi ed Suf cient numbers 0 lnexpengve From same population 0 More likely to be comparable with awareness of prior exposure 0 More willing to cooperate than healthy 0000 people Disadvantages Ill and may not want to be bothered by participating in a study 0 Hospital may be a referral center 0 Friends or Neighbors as controls Get a friend neighbor relative of the case as your control 0 Asking people to name friends and neighbors who they know that are involved Advantages Healthy 0 More likely to cooperate Control for confounding Disadvantages Controls will soon be cases due to similar exposures gt Collecting exposure information 0 Exposure information should be the same 0 Potential sources Medical records Interviews Questionnaires Vital records Employment records Environmental testing 0 Looking at electromagnetic elds Thinking about casecontrol studies Example Looking at possible association between woman39s own birth weight and subsequent risk of ovarian cancer gt Case De nitionl Can include age range can include geographic area gt How are you obtaining casesl OBGYN of ce gt Who are the controls how are they obtainedl women just receiving an annual pap smear at the OBGYN of ce gt How are you collecting exposure info medical records