HIPAA and its Implications on Epidemiological Research Using Large Databases .

Uploaded on:
Category: Art / Culture
Brief diagram of this presentation. Utilizing vast connected computerized information for general wellbeing researchData advancement procedures to guarantee HIPAA-complianceExamples Some musings. Two sorts of information for general wellbeing exploration. Essential dataProspectively collectedWell-planned information accumulation toolInformed consentSecondary dataData initially gathered for other purposesMay be proprietaryPrivacy and confiden
Slide 1

K. Arnold Chan, MD, ScD Harvard School of Public Health Channing Laboratory, Birgham & Women\'s Hospital and Harvard Medical School HIPAA and its Implications on Epidemiological Research Using Large Databases 1

Slide 2

Brief diagram of this presentation Using expansive connected robotized information for general wellbeing research Data improvement procedures to guarantee HIPAA-consistence Examples Some musings

Slide 3

Two sorts of information for general wellbeing research Primary information Prospectively gathered Well-outlined information accumulation apparatus Informed assent Secondary information Data initially gathered for different purposes May be exclusive Privacy and secrecy (especially essential if no earlier approval ) Different information frameworks

Slide 4

Large connected human services databases Health protection claims information Medicaid Medicare Managed Care Organizations (MCO) Automated therapeutic records Hospital/Clinic IT frameworks Availability of composed records Need to contact patients/people ?

Slide 5

Public wellbeing research inside MCOs Harvard Community Health Plan (along these lines got to be Harvard Pilgrim HealthCare) Kaiser Permanente (a few states) Group Health Cooperative (Seattle range) Others HMO Research Network 10+ MCOs over the U.S.

Slide 6

Public wellbeing research inside MCOs Different sorts of MCOs Group display Staff demonstrate Different association with healing centers Implications on information get to MCOs with research programs Separate research divisions Full-time agents and bolster staff

Slide 7

Data components in the MCO information Demographic data Membership Start date, end date, advantage arrange, ... Office visits Type of visit, diagnosis(es), extraordinary techniques Special examinations Radiology, Laboratory examinations Hospitalizations Drug dispensings Linkable by a one of a kind ID

Slide 8

HIPAA and Research with Databases Authorization from individual research subjects not doable Individual approval might be postponed by Institutional Review Board or Privacy Board Minimal Risk Data reported in total design No single-case report "Least essential" standard De-recognizable proof

Slide 9

HIPAA and Research with Databases Single MCO studies Investigators and research staff are MCO workers Multiple-MCO studies May include transferral of information crosswise over MCOs or to a Data Center Other sorts of studies not shrouded in this presentation e.g. Create a de-distinguished dataset for open or business utilize

Slide 10

HIPAA and information advancement Do not move singular level information unless totally vital Generate rundown tables at every review site Combine the tables for definite report Smalley et al. Contraindicated utilization of cisapride: the effect of a FDA administrative activity. JAMA 2000; 284: 3036-9.

Slide 12

HIPAA and information improvement Randomly created Study ID to supplant True ID Crosswalk between the two put away at secured area Destroy the crosswalk after effective linkage of information and quality check Implications for capacity and go down

Slide 13

HIPAA and information advancement Roll-up/change factors Age - > Age bunches National Drug Code - > Drug or Group of medications ICD-9 conclusion code - > Disease e.g. A man conceived on Dec 10, 1934 with determination code xxx.yy got durg 55555-333-22 65-70 y/o m with Heart Failure got Digoxin

Slide 14

HIPAA and information improvement Preserve transient grouping of occasions yet disguise the genuine dates e.g. Sedate use amid pregnancy ponder 29 year-old got 55555-333-22 on Nov 25, 1999 and conveyed a child on Dec 10, 1999 - > 26-30 year-old mother conveyed in 1999, infant presented to amoxicillin at - 16 days

Slide 15

HIPAA and information advancement Only concentrate data pertinent to the review e.g. An investigation of osteoporosis does not require data on subjects\' psychological well-being status Co-dreary conditions might be significant Use intermediary measures to depict level of comorbidity Charlson\'s Index (in light of accompanying conclusions) Chronic Disease Score (in light of co-pharmaceuticals)

Slide 16

HIPAA and information improvement Geocoding Describe social-financial status of study subjects in view of registration tract information Send out (Study ID, address) to a geocoding firm (Study ID, X1, X2, X3) returned X1 : instruction level X2 : pay level X3 : race/ethnicity data

Slide 17

An illustration Finkelstein et al. Diminishing Antibiotic Use Among US Children: The Impact of Changing Diagnosis Patterns. Pediatrics 2003; 112: 620-7. Information components included Date of birth, sexual orientation Membership Drug dispensings Diagnoses in closeness to anti-microbials dispensings Data from nine MCOs

Slide 18

Finkelstein et al. Pediatric anti-infection agents utilize examine Data advancement at each MCO Extract anti-microbials utilize data Extract finding of intrigue (diseases) Use date of birth, sexual orientation, and enrollment information to compute individual time of intrigue Refined, total information sent to the Data Center Rate of anti-microbials utilize = # of anti-microbials utilize/1,000 man years for every age-sex gather

Slide 19

HIPAA and information improvement Individual distinguishing proof is required for specific sorts of research Obtain restorative records Contact patient to direct meeting and additionally ask for example Linkage with outside information Cancer registry National Death Index

Slide 20

HIPAA and information improvement The procedure Data extraction, change, diminishment, and de-ID completed at each MCO Governed by State laws and neighborhood HIPAA-consistent Standard Operating Procedures Principle of Limited Dataset/Minimum important The objective Highly prepared and de-recognized information accessible for connection crosswise over review locales and complex examinations

Slide 21

k-obscurity and substantial datasets The objective A de-distinguished dataset at a specific level of individual namelessness A 43 year-old man with hypertension, diabetes, and nervousness, taking atenolol, rosiglitazone, and lorazepam versus A man 40-45 taking a beta-blocker and a thiazolidenedione

Slide 22

HIPAA, Data Storage and Access Implications on Data Backup Plans Data should be devastated after the report is distributed Data just used to bolster pre-characterized examinations Ancillary investigation are conceivable after IRB survey and endorsement

Slide 23

Epidemiology ponders utilizing vast databases In the past times ... Give me every one of the information, do what I say ... Imagine a scenario where the examiner/analyst need to do THIS investigation ?. Utilize existing datasets to test new theory Good research rehearse Define important information components as indicated by research convention Pre-characterized explanatory arrangement

Slide 24

Epidemiology ponders utilizing vast databases Keys to assurance of human subjects Competent, dependable agents and staff IRB survey and oversight Data advancement rules e.g. Great Epidemiology Practice Information innovation Some sensible standards/rules are superior to no rule

View more...