Weaving New Quality Components into the Fabric of Coverage ......Enrique Martinez-Vidal ... Child and Adolescent Major Depressive Disorder: Suicide Risk Assessment (SRA) Oral Health

Weaving New Quality Components into the Fabric of Coverage for Pregnant Women and Children

National Association of Medicaid Directors 2015 Fall Conference

November 3, 2015

Enrique Martinez-Vidal

Vice President for State Policy and Technical Assistance

2

Today’s Goals Brief overview of child health quality measures

and reporting

National evaluation of the CHIPRA Quality Demonstration

Measuring early elective deliveries and complicated births

Data linkages project

Maternal and Infant Health Initiative Looking to the future

3

2015 Child Core SetPreventive CareChlamydia Screening in Women (CHL)

Childhood Immunization Status (CIS)

Well-Child Visits in the First 15 Months of Life (W15)

Immunizations for Adolescents (IMA)

Developmental Screening in the First Three Years of Life (DEV)

Well-Child Visits in the Third, Fourth, Fifth and Sixth Years of Life (W34)

Human Papillomavirus Vaccine for Female Adolescents (HPV)

Adolescent Well-Care Visit (AWC)

Care of Acute and Chronic Conditions Weight Assessment and Counseling for Nutrition and Physical Activity for Children/Adolescents – Body Mass Index Assessment for Children/Adolescents (WCC)

Medication Management for People with Asthma (MMA)

Ambulatory Care – Emergency Department (ED) Visits (AMB)

Experience of Care Consumer Assessment of Healthcare Providers and Systems (CAHPS®) 5.0H (Child Version Including Medicaid and Children with Chronic Conditions Supplemental Items) (CPC)

4

2015 Child Core SetMaternal and Perinatal Health Pediatric Central Line-Associated Bloodstream Infections – Neonatal Intensive Care Unit and Pediatric Intensive Care Unit (CLABSI)

PC-02: Cesarean Section (PC02)

Live Births Weighing Less Than 2,500 Grams (LBW)

Frequency of Ongoing Prenatal Care (FPC)

Prenatal & Postpartum Care: Timeliness of Prenatal Care (PPC)

Behavioral Health Risk Assessment (for Pregnant Women) (BHRA)

Behavioral HealthFollow-Up Care for Children Prescribed Attention-Deficit/Hyperactivity Disorder (ADHD) Medication (ADD)

Follow-Up After Hospitalization for Mental Illness (FUH)

Child and Adolescent Major Depressive Disorder: Suicide Risk Assessment (SRA)

Oral Health Prevention: Dental Sealants for 6–9 Year-Old Children at Elevated Caries Risk (SEAL)

Percentage of Eligibles Who Received Preventive Dental Services (PDENT)

Access to CareChild and Adolescents’ Access to Primary Care Practitioners (CAP)

5

Measure Specifications

Measures that states report to CMS should include data on entire population of children in Medicaid/CHIP in the state

Two-thirds are based on HEDIS health plan measures Data sources:

– Primarily Medicaid/CHIP administrative data (enrollment and claims or managed care encounters)

– Some measures can use HEDIS hybrid methods (administrative data plus medical chart review)

– Some perinatal measures require vital records data – States can link to other administrative data sources, including

immunization registries – One survey-based measure (CAHPS) – Two EHR measures added in 2013 and 2015

6

States’ Quality Measurement and Reporting Strategies

Calculate measures

Use measures to drive QI

Improve quality of care

• Report results to stakeholders• Align QI priorities• Support provider-level improvement

7

Child Core Set: Reporting on Child Health Quality (FFY 2013)

All states reported 2 or more measures 33 states reported at least 13 7 states reported 24 measures 2 states (NC & SC) reported 25 measures Median # of measures reported by states: 16 (up

from 12 in 2011) Completeness is improving: 41 states now include

both Medicaid and CHIP populations in at least 1 measure (up from 34 in 2011)

Most frequently reported: access to primary care, well-child visits, use of dental services

Source: 2014 Annual Report on the Quality of Care for Children in Medicaid and CHIP

8

Child Core Set Fills a Gap by Providing Uniform State-level Measures for Children Measure set for voluntary annual reporting by

Medicaid and CHIP agencies (24 measures in 2015) Updated annually Includes: access to care, preventive care, maternal &

perinatal care, care of acute & chronic conditions, oral health, experience of care

Data sources: Medicaid/CHIP admin data, HEDIS hybrid methods, vital records data, immunization registries, CAHPS, EHRs

9

AHRQ-CMS Pediatric Quality Measures Program (PQMP)

Created under CHIPRA 2009 Following identification of the initial Child Core Set in 2010:

– AHRQ-CMS partnership to:• Develop future enhanced and improved Child Core Sets• Provide for development of new measures for use by others public and

private programs, plans, providers, and patients– Process: Cooperative agreement (grant) awards to:

• 7 AHRQ-CMS PQMP Centers of Excellence• Duration: February 2011-February 2016 (for most)

– Two CHIPRA quality demo grantees (IL, MA) are undertaking new quality measure development as part of their grants

10

PQMP Accomplishments: Measures, Science of Measurement, and Dissemination

Multiple measures on different topics– Prenatal care– Children with special health care needs– Mental Health

Science– Publications: http://www.ahrq.gov/policymakers/CHIPRA/chipra-

publications.html– Many conference presentations

Dissemination– Care Coordination Measures for Children with Medical Complexity:

CHIPRA Webinar– Adoption of the updated Child Core Set

11

CHIPRA Quality Demonstration Grants

12

CHIPRA Quality Demonstration Grants Children’s Health Insurance Program Reauthorization

Act of 2009 – $100 million program– One of the largest federally funded efforts focused on health

care for children 5-year grants awarded by CMS

– 10 grants that included 18 states total– Feb 2010-Feb 2015, with some extensions– Approximately $10 million per grantee

National Evaluation – CMS-funded; Overseen by AHRQ– Mathematica, AcademyHealth, Urban Institute– Aug 2010-Sept 2015, with continuation through mid-2017

13

Projects Pursued by Demonstration States (#1)

Quality Measures Reporting (10 states) Service Delivery

– Testing and improving provider-based models (14 states)

– Developing, improving, and sustaining care management entities for children with serious emotional disorders (3 states)

Health Information Technology (14 states)– Developing and enhancing current health IT applications, or

providing incentives for their adoption and use

– Analyzing health IT data, and using analyses to develop QI activities

14

Projects Pursued by Demonstration States (#2)

Electronic Health Records (2 states)– Evaluating the model EHR format for children

Other (11 states)– Improve services for youth with complex behavioral needs

– Develop stakeholder collaborations/partnerships

– Focus on School-based Health Centers

15

Ten Demo States Implemented Quality Measures Projects

16

Measure-focused Demo States Saw Greater Increase in Median # of Reported Measures

0

5

10

15

20

25

FFY 2010 FFY 2011 FFY 2012 FFY 2013

Med

ian

num

ber o

f mea

sure

s

Reporting period

Measure-focused demonstration states (n = 10)Other demonstration states (n = 8)Non-demonstration states and DC (n = 33)

17

Demo States Used a Variety of Strategies to Expand Measure Reporting

Hiring dedicated computer programmers for technical aspects of measure calculation

Contracting with Medicaid managed care plans and EQROs to support reporting

Supporting pay-for-reporting programs

Fielding CAHPS more systematically

Developing standard procedures to assemble multiple data files and check their accuracy

Comparing performance with national benchmarks

Establishing statewide group to provide oversight

Identifying variations and monitor changes in performance

18

State Experiences: Maine Increased reported measures from

14 in 2010 to 18 in 2014– Used HIE data to calculate measures– Added billing code modifier to

distinguish between global developmental and autism screenings to report rates separately

Formed a stakeholder workgroup, Maine Child Health Improvement Partnership

Disseminated annual reports on 21 measures (18 from Child Core Set)

Implemented strategies to improve measure performance

19

State Experiences: Massachusetts Reported on 22 out of 26 measures in

2014– Linked data from demo practices,

health plans, and the state’s database on Medicaid and commercial data

Fielded a survey on caregiver perceptions of care in Medicaid and CHIP (patient experience measure)

Produced quality measure reports for practices, families, Medicaid/CHIP policymakers, and commercially insured patients– Solicited feedback on usability via

interviews with practices and focus groups with families and used info to improve reports

20

State Experiences: North Carolina Increased reported measures from 2

in 2010 to 25 in 2014– Leveraged data from various state

agencies Improved existing practice-level

quality reports Using pediatric QI specialists to

analyze network- and practice-level data and work with practices to set QI goals, NC reported:– Improved care quality in over 200

practices– Improved performance on quality

measures statewide in 15 months– Demo staff believe that collaboration

between the QI specialists and practices contributed to the changes

21

Demo States Encountered Similar Challenges

Combining data from different programs/sources– Medicaid FFS, Medicaid MCOs, CHIP (if separate

agency)

Linking state data sources– Vital records, state immunization registry

Reporting measures from EHRs Adapting state-level measures to practice-level for

QI activities

22

Key Takeaways Calculating the measures took more time and

resources than states anticipated Some measures were more challenging than others

But… States can overcome many of the challenges to

reporting the Child Core Set measures if they invest in data quality and reporting systems, identify staff or contractors who have expertise in quality measurement, and make use of TA and financial support

23

Primary Care Providers & Child Health Care Quality Reports

→Survey of pediatricians and family physicians providing care for children in Medicaid & CHIP

→Two demonstration states (NC, PA) and one comparison state (OH)

→Questions about physicians’ experiences with and attitudes toward pediatric quality reports

→727 physicians responded (response rate: 45%)

24

Primary Care Providers & Child Health Care Quality Reports

25

Primary Care Providers & Child Health Care Quality Reports

26

How States Are Using the Child Core Set Measures to Improve Quality

Reporting Results to Stakeholders – Used existing data sources– Created targeted reports for different stakeholders

Aligning Measures and QI Priorities – Formed multi-stakeholder QI workgroups– Encouraged consistent quality reporting standards across programs – Required managed care organizations to meet quality benchmarks

Supporting Provider-Level Engagement– Paid providers for reporting measures and demonstrating

improvement

– Changed reimbursement to support improvements– Hosted learning collaboratives– Provided individualized TA

27

Learn More About the CHIPRA Quality Demo

http://www.ahrq.gov/policymakers/chipra/demoeval/index.html

Evaluation Highlights:– How are CHIPRA quality Demonstration States using

quality reports to drive health care improvements for children?

– How are the CHIPRA Quality Demonstration States encouraging health care providers to put quality measures to work?

– How are CHIPRA Quality Demonstration States improving perinatal care?

28

Early Elective Deliveries and Complicated Births

29

Measuring Early Elective Deliveries in Medicaid

CMS made EEDs a priority measure in the Child Core Set

Medicaid Medical Directors Learning Network Study– 12 states provided Medicaid admin data linked

with vital statistics data– 10 states provided only vital statistics data

EEDs accounted for almost 9% of births paid for by Medicaid

Source: Tara Trudnak Fowler, Jeff Schiff, Mary S. Applegate, Katherine Griffith, and Gerry L. Fairbrother “Early Elective Deliveries Accounted for Nearly 9 Percent of Births Paid For Medicaid” Health Affairs, 33, no.12 (2104)

30

Early Elective Deliveries in Medicaid: Policy Implications

Downward trend in EEDs but more reduction needed Policy interventions:

– partnering with hospitals on “hard stop” policies to prohibit EEDs – attached to reporting/ reimbursement

– learning collaboratives – prior authorization requirements – educational and feedback efforts targeting physicians and patients– participating in national initiatives:

• Strong Start for Mothers and Newborns Initiative• Hospital Engagement Networks/CMS Partnership for Patients• Collaborative Improvement and Innovation Network (CoIIN) to Reduce Infant Mortality

(HRSA)

Participating states have common analytic code and protocols to continue monitoring EEDs for quality improvement

31

Costs of Complicated Births Birth complications are an especially important area for

Medicaid– 2/3 of adult women Medicaid beneficiaries are of childbearing age– About half of all Medicaid hospital stays are for pregnancy, childbirth,

and newborns – In 2011, Medicaid paid for approximately 45 percent of all births in the

United States and, under ACA expansion, it is increasing…

Avg. cost for complicated newborn increased from 2002-2009 Proportion of complicated births billed to Medicaid increased

from 2006-2009– Increase in proportion in women 15-44 covered by Medicaid– By 2009, Medicaid covered more complicated births than private

insurance (47.5% vs 44.4%)

From 2002-2009, avg cost/admission for complicated newborn stays paid by Medicaid consistently higher than private insured

Source: Fowler TT, et al. Trends in complicated newborn hospital stays & costs, 2002-2009: Implications for the future. Medicare and Medicaid Research Review 2014;4(4):E1-E17

32

Leading Diagnoses for Complicated Newborn Stays in 2009

Preterm birth/low birth weight– 23% of diagnoses and 33% of aggregate costs– More likely to be billed to Medicaid

Respiratory distress – 18% of diagnoses and 28% of aggregate costs– More likely to be billed to Medicaid

Jaundice– 10% of diagnoses and 3% of aggregate costs– Less likely to be billed to Medicaid

33

Improved Data and Data Linkages Would Improve Study

Maternal-child linkages not possible with HCUP-NIS– Missing key maternal factors that may influence

complicated births (e.g., maternal age, education, smoking, birth history, type of delivery)

Clinical data to determine severity of conditions are lacking in administrative data

Only expected payer source, not actual, given billing or hospital discharge abstract data

Discharges billed to CHIP may not be consistently classified as a specific payer type

34

Data Linkages

35

Addressing Data Linkage Difficulties: LBW and C-Section Measures

Initiative provided technical assistance to states for measuring & reporting two CMS Child Core Set measures• Rate of low birth weight• Rate of C-section delivery

Technical specs for these measures require linking birth certificate data to Medicaid claims data• Technical assistance to states seeking to expand/improve

quality reporting

• Training to improve technical capacity to build & sustain data linkage infrastructure

Supported by CDC’s Division of Reproductive Health

36

Linkage and Reporting Status 13 Phase I States

– AZ, GA, IN, KY, ME, MA, MI, MS, NM, NV, OK, WV, WY– Executed DUA/MOU: 10– Linked data set: 10 (1 previously linked)

6 Phase II States– CT, DE, DC, NE, NJ, VA– Executed DUA/MOU: 2– Linked data set: 1– Expected to submit measures in Fall 2015:

• Yes=1; Maybe=2; No=3

Helped jumpstart other efforts

37

Linkages Lessons Learned DUAs, DSAs, and MOUs most important, sensitive, &

time-consuming part (sharing identified data/legal processes)

Think carefully about the needed records & data elements Needed more guidance on Medicaid data Document every step in the process to facilitate future

replication Provide feedback to all partners Consider expanding linkage longitudinally across Medicaid

records for the child (not just mother’s record) Don’t stop! Time and Persistence!http://www.academyhealth.org/datalinkageproject/

38

Maternal and Infant Health Initiative

39

CMCS’ Maternal and Infant Health Initiative (MIHI)

Launched in July 2014, MIHI aims to:– Increase the rate and content of postpartum visits– Increase the use of contraception among women in

Medicaid and CHIP 14 states/territories are participating:

– AL, CA, CO, DE, IA, KY, MA, MI, MS, MO, NY, Northern Mariana Islands, OR, and WA

Each state/territory receives up to $400,000 total over 4 years– Project period: September 2015 – September 2019

40

CMCS’ Maternal and Infant Health Initiative (MIHI)

Uses of funds:– Improve data validation methods/initiatives– Evaluate data sources and measures– Train and educate providers in measure

collection and reporting– Train staff in the use of tools for data collection

and analysis (claims, surveys, EHRs)– Develop a plan to sustain data collection

41

Looking to the Future

42

Lessons Learned: Gaps (1)

Data Gaps– COEs often can not get State Medicaid data (privacy, state

resources)– National rates not available on a timely basis (challenges of

combining data from different States/programs; CMS resource issues)

Measure Level Gaps (state level vs practice-level)– Child Core Set and COE-developed measures focused

mostly on state-level measures (i.e., to provide a state-level picture of quality, state-to-state comparisons)

– But there is need in the QI Demos (and other QI implementation) for practice-level measures.

43

Lessons Learned: Gaps (2)

Capacity Gaps– Variation in capacity to adopt measures across state

Medicaid programs– Led to a situation in which all states, regardless of capacity,

were held to the lowest common denominator (i.e., measures acceptable to all or most States).

Communications Gaps– Between legislative reporting requirements and

programmatic needs/realities– Between silos in HHS (CMCS, CMMI, AHRQ)

44

Lessons Learned: Resources Key challenge for CHIPRA 2009 activities was time and

resources– Same people were charged with implementation of multiple new

Obama-era programs: CHIPRA 2009 (multiple programs for enrollment, quality measurement, quality improvement); the early 2009 Investment Act; and ACA in early 2010

– Given pre-existing resources in the Medicaid program, thoughtful, coordinated implementation was a challenge.

Many existing external resources involved in CHIPRA implementation were ready to go on Day 1 with:– Measures (e.g., COEs)– Quality Improvement (e.g., NICHQ, NIPN components, scientific

literature on QI/DI)– National leadership: AHRQ, CMCS, CMMI– State Medicaid leadership

45

Recommendations for the Future (1) Capture the lessons learned across all the investments in

children's health care, especially for Medicaid/CHIP Information from quality measure development lessons

and from improvement evaluations could help drive research and program implementation agendas

Designate a single entity at OS level to coordinate children's healthcare quality activities across investments and do continuous quality improvement across HHS, state and private sector programs

Develop a new strategy for federal Medicaid/CHIP and state Medicaid/CHIP programs to work together and with private sector on quality measurement and improvement and equity

46

Recommendations for the Future (2) Provide more resources to state Medicaid programs, possibly

contingent on coordinated approach with feds and external experts

Pay for state reporting on quality measures, contingent on accuracy of reporting as a result of additional resources provided

Invest in implementing and using children's healthcare quality measures developed by COEs and through the state quality demonstrations

Don’t stop working on children's healthcare quality even though most of the CHIPRA quality money and activities are ending!

47

Thank you!

Enrique [email protected]

202-292-6729

AcademyHealthwww.academyhealth.org/