HI*TRACK: Solving Newborn Hearing Screening Tracking Issues

1. HI*TRACK: Solving Newborn Hearing Screening Tracking Issues Karl R. White, PhD National Center for Hearing Assessment and Management Utah State University www.infanthearing.org

2. Percentage of Newborns Screened Prior to Discharge

3. Rate Per 1000 of Permanent Childhood Hearing Loss in UNHS Programs Sample Prevalence Site Size Per 1000 Rhode Island (3/93 - 6/94) 16,395 1.71 Colorado (1/92 - 12/96) 41,976 2.56 New York (1/96 - 12/96) 27,938 1.65 Utah (7/93 - 12/94) 4,012 2.99 Hawaii (1/96 - 12/96) 9,605 4.15

4. Rate Per 1000 of Permanent Childhood Hearing Loss in UNHS Programs Sample Prevalence % of Refers Site Size Per 1000 with Diagnosis Rhode Island (3/93 - 6/94) 16,395 1.71 42% Colorado (1/92 - 12/96) 41,976 2.56 48% New York (1/96 - 12/96) 27,938 1.65 67% Utah (7/93 - 12/94) 4,012 2.99 73% Hawaii (1/96 - 12/96) 9,605 4.15 98%

5. Tracking "Refers" is a Major Challenge (continued) Initial Rescreen Births Screened Refer Rescreen Refer Rhode Island 53,121 52,659 5,397 4,575 677 (1/93 - 12/96) (99%) (10%) (85%) (1.3%) Hawaii 10,584 9,605 1,204 991 121 (1/96 - 12/96) (91%) (12%) (82%) (1.3%) New York 28,951 27,938 1,953 1,040 245 (1/96-12/96) (96.5%) (7%) (53%) (0.8%)

6. Examples of JCIH Benchmarks and Quality Indicators • % of infants screeened during birth admission • % of infants who do not pass birth admission screen • % of families who refuse hearing screeening • % of infants and families whose care is coordinated between the medical home and related professionals • % of infants with completed audilogic and medical evaluations by 3 months of age • % of infants with confirmed hearing loss : • referred for otologic evaluation • that have a signed IFSP by 6 months of age • % of infants with hearing aids receiving audiologic monitoring at least every 3 months

7. Data Required for MCHB Project Annual Reports • # of infants screened(95%) • # of infants referred for audiologic diagnosis • # and age of infants receiving audiologic diagnosis(before 3 months) • # of infants • in a medical home • connected with family-to-family support • # and age at which identified infants are enrolled in early intervention services (before 6 months)

8. CDC EHDI Reporting System • # of live births • # screened prior to discharge • # screened before 1 month of age • # referred from screening for audiologic evaluation • # with audiological diagnosis by 3 months of age • # with permanent congenital hearing loss (0-7 years) • Hearing loss classified by type, degree and laterality • Average/median age at which hearing loss diagnosised • # of infants receiving intervention by 6 months of age

9. OPERATING SUCCESSFUL EHDI PROGRAMS out Then a miracle occurs Start Good work, but I think we might need just a little more detail right here.

10. Research Program Improvement and Quality Assurance Screening Diagnosis Intervention Medical, Audiological and Educational Purposes of an EHDI Data System

11. Nature and Use of Information is Different For: Hospitals State Departments of Health National Agencies

12. Computerized Patient/Data Management for Hospital-based UNHS Programs Tracking/scheduling related to screening, follow-up, diagnosis, and intervention Communication with stakeholders (e.g., parents, physicians, audiologists) Reporting to funding and administrative agencies Program management, quality control, and risk management

13. Statewide EHDI Data System Monitoring program status to identify in-service and technical support needs. Safety net for babies who "fall through the cracks" Assisting with follow-up / enrollment for diagnostic and intervention programs Access to data for public health policy and administrative decisions. Linking to other Public Health Information databases (e.g., Immunization, WIC, Vital Statistics, Early Intervention, Birth Defects)

14. Examples of Benefits from Linking EHDI Database with Other Public Health Information Systems • An infant referred from the hospital-based UNHS program, but lost to follow-up, could be identified and provided with EHDI services when he or she comes in for the DPT Immunization at eight weeks of age. • By linking the Birth Defects Registry and EHDI data, children with birth defects that make them substantially more likely to develop late onset losses could be monitored and provided with assistance at a much earlier time. • Many of the children who become “lost” for immunizations or birth defects tracking are the same children who are lost for EHDI. By sharing information, fewer resources are needed to more successfully find and provide services to “lost” children. • Linking the EHDI and vital statistics allows a population-based system to be created so that every live birth in the state is included in the EHDI system.

15. Utah EHDI Data System Hospital 1 Hospital 2 Hospital 3 State Department of Health . . . . Hospital 21

16. Iowa EHDI System Hospital 1 Hospital 2 . Area Education . Agency #1 Hospital 9 Hospital 10 Hospital 11 Area Education . Agency #2 . State Department of Health Hospital 16 . . Hospital 17 . Hospital 25 Hospital 26 Area Education . Agency #9 . Hospital 35

17. Hawaii EHDI System Hospital 1 Hospital 2 Zero-to-Three State Department Project of Health Hospital 3 . . . . Early Intervention Hospital Programs

18. Hospitals Most Likely to Participate in a State EHDI Database If: it provides locally useful data gathering data is quick transfer to the state is trouble-free it reduces other reporting requirements It reduces risk

19. Who Needs the Data? • Screeners and program coordinators • Hospital administrators • Health care providers • Public Health officials

20. What Type of Data is Needed? Collected continuously by CORE VARIABLES: everyone. Everyone agrees they would be OPTIONAL VARIABLES: nice, but some may not have resources to collect (may not be collected continuously). Some people think they are RESEARCH VARIABLES: important; others should be aware that some are collecting them.

21. Examples of Possible: CORE VARIABLES OPTIONAL VARIABLES RESEARCH VARIABLES Gestational Age Infant's last name Time of Birth Specific Results of Medical ID# Sex Diagnostic Tests Date of Birth Nursery Type Date and Time of Screening Test Mother's Maiden Name Birthweight Type of Delivery Birth Hospital Amplification Mother's Occupational Screening Hospital Age at Amplification Noise Exposure Inpatient Screen Result Days in NICU Outpatient Screen Result JCIH Risk Indicators Diagnostic Result Age at Diagnosis

22. Options for Developing an EHDI Patient/Data Management System • Develop your own • Modify an existing system, for example • “Heelstick” data management system • Electronic Birth Certificate (EBC) • Purchase an existing system • Whatever system you choose, should it be web-based?

23. Combining EHDI Data Management with Existing Systems is Logical Because : • Combining EHDI with Heelstick is attractive because: • Both do initial screening of babies in the nursery prior to hospital discharge • Both do 2nd stage or outpatient screening for a significant number of babies • Poor follow-up is currently the biggest challenge for EHDI programs • Heelstick programs have been extremely successful with follow-up • The infrastructure for Heelstick follow-up system is already in place • Combining with Electronic Birth Certificate is an attractive option because the EBC is: • Legally required for every birth • Contains wealth of demographic and medical data

24. North Carolina Heelstick Form

25. Heelstick Screening Procedures • Small sample of blood collected and put on Heelstick form (filter paper) prior to discharge, but after 24 hours of age • Form sent to laboratory within hours or days for analysis • A significant number of initial screenings need to be redone because of poor technique • Results reported to State Follow-up Coordinator who contacts physicians and parents about “abnormals” (urgency depends on disease) • Depending on state, about 1% to 2 % are abnormal. Additional blood is collected for these babies to confirm the screening result (diagnosis).

26. Diagnosis n=10 Fail=80 Outpatient Screening n=80 Fail=10 Hearing Loss=3 Normal Hearing=7 Inpatient Screening 2 Stage OAE Pass=920 Pass=90 Diagnosis n=40 Hearing Loss=3 Normal Hearing=37 Fail=40 1 Stage AABR Inpatient Screening Pass=960 Inpatient Screening Fail=20 Diagnosis n=20 Hearing Loss=3 Normal Hearing=17 1 Stage OAE / AABR Pass=980 Typical UNHS Screening Protocols (example for 1,000 newborns)

27. Inpatient Hearing Screening • Multiple attempts are very common • Different screeners often attempt the same baby • Screening can be done any time from shortly after birth to minutes before discharge • Use of both OAE and AABR becoming more common • Successful management requires more than knowing whether baby passed or referred

28. Outpatient Screening • Depending on protocol, outpatient screening required for 2-10% of all births • Usually done between 2-14 days following discharge • Sometimes done at a different location from inpatient screening • Requires coordination with baby’s doctor

29. Audiological Diagnosis • Often done at location other than screening hospital • Requires coordination with baby’s doctor and ENT • One visit often not sufficient • Advantages in coordinating with Part C, IDEA Child Find activities

30. Enrollment in Early Intervention • Continued need for data management and tracking because: • Early Intervention requires ongoing, multidisciplinary services • Coordination is needed with the baby’s medical home • Important to link late-identified children with original screening results

31. Issues to Consider Before Combining EHDI and Heelstick • Heelstick Screening has added many new tests over the years. But, Newborn Hearing Screening (NBHS) is not just another analysis of the bloodspot • NBHS screening personnel often involved in collection and analysis of screening data, follow-up, and diagnostic procedures • When, where, how, and by whom NBHS screening is done is quite different than Heelstick

32. Issues to Consider Before Combining EHDI with Heelstick or EBC • Heelstick Screening has added many new tests over the years. But, Newborn Hearing Screening (NBHS) is not just another analysis of the bloodspot • Screening personnel often involved in collection and analysis of screening data, follow-up, and diagnostic procedures • When, where, how, and by whom NBHS screening is done is quite different than Heelstick • Timing of data collection and entry • Ideal if Heelstick or EBC is always followed by NBHS, but it doesn’t happen that way • When are you finished with NBHS? • How are outpatient NBHS screenings updated?

33. Issues to Consider Before Combining EHDI with Heelstick or EBC(continued) • Will hospital’s staff have timely access to the data for program improvement and follow-up? • Screener performance • Scheduling outpatient screening, referring for Diagnostic Assessments, confirmed hearing loss • Can hospitals update data • Who decides which data is most accurate?

34. Issues to Consider Before Combining EHDI with Heelstick or EBC(continued) • Will the Heelstick or EBC form include all the “fields” you need? • Heelstickor EBC forms with NBHS fields usually only include type of test, left ear result, right ear result. Do you need….? • Screener ID • Mother’s language • Type of insurance • Who decides if and when you can add or modify “fields” • Hearing loss risk factors • Results for multiple tests or attempts • Outpatient screening results

35. Issues to Consider Before Combining EHDI with Heelstick or EBC(continued) • Will the Heelstick or EBC form include all the “fields” you need? • Heelstick and EBC forms with NBHS fields usually only include type of test, left ear result, right ear result. Do you need….? • Screener ID • Mother’s language • Type of insurance • Who decides if and when you can add or modify “fields” • Can you transfer data from screening machines directly to the Heelstick or EBC? • Duplicate data entry • Transmission errors • Hearing loss risk factors • Results for multiple tests or attempts • Outpatient screening results

36. Issues to Consider Before Combining EHDI with Heelstick or EBC(continued) • Combining EHDI with Heelstick or EBCisn’t free • Costs of modifying and reprinting forms is very small • Cost of adding fields to Heelstick follow-up software and generating new letters / reports can be substantial ($50K+) • Cost of developing software to process EBC data for EHDI data management system can be even more expensive • Costs and risks of duplicate data entry are significant (screener records info, transfers to Heelstick form, lab personnel keypunch)

37. Issues to Consider Before Combining EHDI with Heelstick or EBC(continued) • Follow-up of babies requires substantial personnel resources whether or not NBHS is combined with Heelstick or EBC • Although it varies widely, Heelstick follow-up typically requires about 1 FTE per 30,000 births - - - expect similar resources for NBHS • 2% to 10% of babies will require some type of follow-up for NBHS • Do Heelstick follow-up staff understand EHDI issues well enough to do follow-up?

38. Issues to Consider Before Combining EHDI with Heelstick or EBC(continued) • Sources of information are quite different for diagnostic confirmation of screening results • For Heelstick: New blood specimen is submitted to lab by doctor or hospital, lab does analysis and sends to Heelstick Coordinator • For NBHS: Information is reported in various forms to • Physician, hospital, and / or state EHDI coordinator • from hospitals, community-based audiologists, physicians

39. Issues to Consider Before Combining EHDI with Heelstick or EBC(continued) • Sources of information are quite different for diagnostic confirmation of screening results • For Heelstick: New blood specimen is submitted to lab by doctor or hospital, lab does analysis and sends to Heelstick Coordinator • For NBHS: Information is reported in various forms to hospital or state EHDI coordinator from hospitals, community-based audiologists, physicians

40. Is a Web-based System the Answer? • Access? • Speed? • Linkages with existing data? • Flexibility? • Security?

41. Demonstrations of: Stand Alone system Web-based system • (Demos of HI*TRACK are also available at www.hitrack.org)

42. Thin-Client Architecture Benefits Issues • Reduced user interface functionality. • Slower response times for user interactions. • If network stops, work stops. • Difficult to integrate with third party screening software. • Installation on the client machine is not required. • Software updates do not require any maintenance on the client machines. • Cheaper to deploy. Database Server Thin Client Presentation & Business Rule Layers Software = UI (user interface) is Web Browser

43. Medium-Client Architecture Benefits Issues • Better responsiveness than thin-client. • More feature rich user interface. • “Business rule” changes require no change on clients. • Better integration with third party screening software. • Client requires software to be installed. • If network stops, work stops. • User interface changes require the clients to be updated. Database Server Business Rule Layer Medium Client Software = UI & Presentation

44. Fat-Client Architecture Benefits Issues • Full feature user interface. • Even better user responsiveness. • Good integration with third party screening software. • Software changes require the clients to be updated. • If network stops, some features not available Database Server Fat Client Software = UI & Presentation & Business Rules

45. Stand-alone Architecture Benefits Issues • Full feature user interface. • Best user responsiveness. • Work is not dependent on the network. • Best integration with third party screening software. • Software changes require updates to be installed. • Can only be accessed from the user’s machine Stand Alone Software = UI & Presentation & Business Rules and Data Base

46. OPERATING SUCCESSFUL EHDI PROGRAMS out Then a miracle occurs Start Good work, but I think we might need just a little more detail right here.