Data Sources
Clinician Reporting
Clinician reporting directly to public health authorities would be used for Tier 1 case reports and would include reports outside of those made using hospitalization or emergency department data. Information from patient records would likely include chief complaint, diagnosis, and physician or triage notes. Reporting of case reports could include electronic case reporting (eCR), as NAS is scheduled to be released as a condition within the Reportable Conditions Knowledge Management System (RCKMS) in early 2024. Patient records could include both birthing parent and newborn records and could be used for case ascertainment and case classification.
Strengths
Reporting from patient records has the most comprehensive information, including demographic information, clinical
impression, physician or triage notes, laboratory test results, medical treatments and procedures, medical history, and
comorbidity. Ideally, both birthing parent and newborn medical records would be used to gather information for case
classification. When compared with utilizing discharge data sets that require analysis and cleaning, clinician reporting can
provide a timelier representation of NAS.
Limitations
Birthing parent and infant visit records may not be equally accessible, and identification of birthing parent/infant pairs may
be difficult and could result in duplicate cases. Outpatient records will normally only be inclusive of birthing parent data.
Clinical reporting directly to public health authorities can be burdensome in both reporting and abstraction and may lead to
under-ascertainment.
Laboratory Reporting Systems
Laboratory data are useful for Tier 1 case classification as described in Appendix 3 of 23-MCH-01. Required clinical laboratory reporting is mandated by states for selected reportable conditions. Data elements include laboratory test results (and the lab reference range), the patient contact information, and the health care provider who ordered the lab test. There is variability in reporting of personal identifiers. These reports are used to determine morbidity, exposure and take public health actions.
.
Strengths
Birthing parent and neonate clinical specimen testing may be performed at hospitals, commercial, state and federal labs.
Provides valuable information on in utero exposures when sensitive and specific tests are conducted close to the time of
delivery. Some information is provided to national databases through the Electronic Laboratory Reporting System (ELR).
Limitations
Ons Laboratory testing of both pregnant women and neonates can be highly variable in who receives testing and what substances
are tested for, especially when testing may result in punitive action. Relying solely on laboratory testing for case
ascertainment can lead to underreporting. Patient identifiers are not standardized to align with identifiers within medical
records. Diverse systems of data aggregation.
Hospital Inpatient Discharge Data (State and National)
Most states have access to their hospitals’ inpatient discharge data, which contain case-specific discharge data including utilization data, clinical data, and demographic data for patients admitted to acute care hospitals. Some states obtain patient identifiers and others do not. These databases also contain utilization, revenue, expense, and payer data. Hospital discharge data may be used for Tier 2 ascertainment and classification of cases.
.
Strengths
Hospitals provide their records to a data intermediary for processing. Formatting varies from state to state. Data elements
include: dates of admission and discharge, nature of admission (e.g., emergency), residence, sex, age, race, date of birth,
diagnosis and procedure codes (ICD-10-CM), expected principal source of payment (including workers compensation), and
charges
Limitations
State data sets are records of all hospitalizations and duplication may exist.
Most states do not include federal and non-acute care facilities. There may be a lag in data availability.
Nationwide Emergency Department Data (State and National)
Approximately 42 states have emergency department data on all ED visits, including demographic, diagnosis and procedure information. Data elements can include: birth date, hospital and physician information, sex, race (includes Hispanic), type of
visit, source of payment, diagnosis and procedure codes (ICD-10-CM), reason for visit, mode of transport, medical record number.
Hospitals are required to report certain data for each outpatient ED visit, by law or regulation, in selected states. The purposes of the data are to accurately quantify and track the number and type of ED visits and to provide case-mix information to
hospitals and communities. ED databases available through the Healthcare Cost and Utilization Project (HCUP) are limited to only patients seen in the ED who are not subsequently admitted as an inpatient as a part of the encounter.
.
.
Strengths
Important for estimating burden of conditions in the population that does not get hospitalized. State and local jurisdictions
may have access to a more comprehensive version of these data than HCUP and it may include encounters for patients who
are subsequently admitted. A patient’s emergency department discharge disposition may be useful to determine birthing
parent and possibly newborn status if available.
Limitations
Each state has its own procedures. Data with identifiers are not available to all states. Care should be taken to ensure
duplicate records do not occur, as infants may also be present in other data sources if previously/subsequently hospitalized
or admitted to a pediatric recovery center. Not available in all states, but complete visit counts in those states that have
mandatory systems and can be used for patient follow-back and longitudinal data for special studies when patient identifiers
are included.
Syndromic Surveillance Data
As part of the public health electronic data reporting requirements within the Medicare Promoting Interoperability Program emergency departments are incentivized to electronically report data to public health in near real time. Data elements can
include: age, hospital, sex, race, ethnicity, chief complaint, diagnosis, medical record number, and discharge disposition. The purposes of the data are to track disease trends in near real time to provide situational awareness. Depending on local resources, ED data may be used as a supplement for ascertainment and classification of Tier 1 and/or Tier 2 cases
.
.
Strengths
Case ascertainment can occur through querying the chief complaint or diagnosis fields to identify cases that may have been
missed through provider reporting. A patient’s emergency department discharge disposition may be useful to determine
birthing parent and possibly newborn status.
Limitations
Each state has its own procedures. Data with identifiers are not available to all states. Care should be taken to ensure
duplicate records do not occur. Coverage of emergency departments participating in syndromic surveillance may vary
between jurisdictions. Syndromic surveillance is potentially useful only as a supplement to additional case ascertainment
activities as it contains information only on neonates presenting to an emergency department.
Birth Defects Registries
Birth defects registries aligned with the National Birth Defects Prevention Network (NBDPN) are statewide, population based surveillance systems that have identified birth defects in children born in each state. Birth defects registries may be used for ascertainment and classification of Tier 1 cases.
.
Strengths
May include the counts, incidence and prevalence of infants diagnosed with NAS as well as comorbidities. They can
include similar demographic information and some birthing parent/infant health characteristics that might be of use to
the NAS surveillance program.
Limitations
Case definitions vary and include ascertainment throughout the first year of life or beyond with an inherent data availability
delay.
Publicly Funded Health Care: Medicaid Claims Data
Medicaid is one of several publicly funded healthcare organizations serving state citizens. Medicaid is the state and federal partnership that provides health coverage for selected populations of low income and/or medical need. Medicaid claims data may be used for ascertainment and classification of Tier 2 cases.
.
Strengths
Includes demographic characteristics, diagnosis and procedure codes (ICD- 10-CM, CPT, HCPCS), therapies, all covered
services, prescriptions, costs, multiple sources of payment. Healthcare claims data, including pharmacy data can provide
crucial information relating to medications, prescribed treatments, charges, and physician visits not available from other
sources.
Limitations
Program coverage and funding of programs vary state by state. There may be a lag in data availability. Analysis can be
restricted to infants with a code indicating newborn hospitalization to prevent duplicate cases, but this will eliminate cases
identified after the delivery hospitalization.
All Payer Claims Data
All Payer Claims Data (APCD) is available in several states and includes both publicly and privately funded healthcare information on medically- related claims from emergency departments and in-patient, out-patient and skilled nursing facilities. All Payer Claims Data could be used for Tier 2 ascertainment and classification..
.
.
Strengths
Includes demographic characteristics, diagnosis, and procedure codes (ICD-10-CM, CPT, HCPCS), therapies, all covered
services, prescriptions, costs, sources of payment.
Limitations
Covered populations may vary by state depending on participation by federal and self-funded health care plans. There may
be a lag in data availability. Analysis can be restricted to infants with a code indicating newborn hospitalization to prevent
duplicate cases, but this will eliminate cases identified after the delivery hospitalization.
Private Insurers: Health Management Plan Data
Private insurers / managed care plans keep records of all of the claims they process. Data sharing agreements can potentially be established with these entities to gather specific information on treatment, medication, etc. Private insurer data could be used for Tier 2 ascertainment and classification.
.
Strengths
Medical record level data with more specific information on treatment, medication, and follow-up.
Limitations
Difficult / time consuming to obtain agreements with data owners. Population varies depending on the size of the
population served by each care organization. Some insurers may redact records containing information on substance use
disorders. There may be a lag in data availability. Analysis can be restricted to infants with a code indicating newborn
hospitalization to prevent duplicate cases, but this will eliminate cases identified after the delivery hospitalization.
Residential Pediatric Recovery Centers
Health care services provided to infants in residential pediatric recovery centers. Includes multiple diagnosis codes (ICD-10-CM), chief complaint and physician or triage notes for the neonate. May include administrative/discharge records. Residential pediatric recovery center data could be used to ascertain and classify Tier 1 cases.
.
Strengths
Residential treatment centers treat multiple conditions from drug and alcohol addictions and in some states can provide
concurrent care for the birthing parent.
Limitations
Program coverage, data acquisition and funding of programs vary state by state.
Prescription Drug Monitoring Programs (PDMP)
A PDMP is an electronic database that tracks controlled substance prescriptions within a state. Information may be available on substances prescribed during pregnancy if able to link PDMP data with case data.
.
Strengths
The PDMP data source is comprehensive for the state, and if it is possible to access and use it may provide information on
substance exposure during pregnancy as a part of case surveillance or for special studies. For more in-depth analyses, it is
possible to determine the time of the exposure by the date the prescription was filled as well as the amount/volume.
Limitations
There are strict regulations regarding access and use of data from PDMP. Linkage of data between cases and the PDMP
may not yield high results. These data contain information about prescriptions filled, but not necessarily consumed.
Vital Records
A state’s office of vital records is responsible for issuing birth and death certificates. Birth certificates will contain demographic information on the infant and parent(s) and limited health information about the infant at time of birth. An
office of vital records also provides aggregate data on births or deaths within a jurisdiction.
.
Strengths
Aggregate data from birth certificates can be used in the calculation of rates as part of surveillance. Neonatal abstinence
syndrome may be a diagnosis tracked on a birth certificate within some jurisdictions.
Limitations
All infants diagnosed with NAS may not receive an indication of that diagnosis on a birth certificate (if applicable in the
jurisdiction) due to timing differences between the issuance of a birth certificate and diagnosis.
more Resources
Electronic Medical Records
Electronic medical records (EMRs) are digital versions of the paper charts in clinician offices, clinics, and hospitals. EMRs contain notes and information collected by and for the clinicians in that office, clinic, or hospital and are mostly used by
providers for diagnosis and treatment. Data may include diagnoses, medications, immunizations, family medical histories, and diagnostic screening results. Some jurisdictions receive reports generated by a facility’s electronic health record (EHR)
system based on ICD10CM or SNOMED diagnosis codes for Tier 1 case ascertainment to identify neonates diagnosed with NAS (e.g. P96.1) or exposed in utero to opiates, benzodiazepines, or barbiturates(e.g. P04.14). Further review of EMRs can
be used for case classification.
.
Strengths
Records are inclusive of a broader view of a patient’s care including specialists and laboratory specimens. Visits from
providers over time may be included including preventive treatments and tests, screenings, patient monitoring and other
data not normally collected in administrative datasets. Depending on the amount of the EMR available, this data source may be the most helpful in differentiating implicated drugs related to the exposure.
Limitations
Database structure is based on visits and duplication may exist. It may be difficult for jurisdictions to receive access to a
facility’s electronic medical record system; attempts to gain remote access may be time consuming and not always
successful.
