Quality Measures & Domain Modules¶
Beyond general-purpose algorithms, medicaid-utils includes domain-specific modules that package covariates and outcome definitions from published Medicaid research.
Topics Modules¶
The topics package contains condition-specific variable construction developed for
peer-reviewed publications.
Opioid Use Disorder (OUD)¶
The topics.oud module provides:
Medication detection – buprenorphine treatment identification via procedure codes and NDC
OUD medication flags – medication-assisted treatment indicators
Behavioral health services – mental health and substance use service identification
Care setting classification – FQHC, outpatient hospital, physician office
from medicaid_utils.topics.oud import medication_and_behavioral_health as mbh
# Flag buprenorphine claims in pharmacy data
df_rx = mbh.flag_rx_buprenorphine(rx.dct_files["base"])
# Flag behavioral health treatment in outpatient claims
df_ot = mbh.flag_proc_behavioral_health_trtmt(ot.dct_files["base"])
OB/GYN¶
The topics.obgyn module provides:
Delivery outcome identification – ICD-9 and ICD-10 based delivery detection
Preterm birth flags – gestational age classification
Multiple birth detection
Provider classification – religious vs. secular hospital identification
Chronic condition comorbidities – diabetes, hypertension, CKD, depression, COPD, tobacco use
from medicaid_utils.topics.obgyn import hospitalization
# Flag delivery-related hospitalizations
df_ip = hospitalization.flag_delivery(ip.dct_files["base"], cms_format="TAF")
External Data Integration¶
The other_datasets module integrates external provider and geographic data:
NPI Registry – National Provider Identifier lookups
HCRIS – Healthcare Cost Report Information System (hospital cost reports)
UDS – Uniform Data System (health center data)
FQHC – Federally Qualified Health Center provider identification
Geographic crosswalks – RUCA (Rural-Urban Commuting Area), RUCC (Rural-Urban Continuum), PCSA (Primary Care Service Area) codes via ZIP code
from medicaid_utils.other_datasets import fqhc
# Get FQHC NPI crosswalk for a given start year
pdf_fqhc_crosswalk = fqhc.get_fqhc_crosswalk(start_year=2012)