DRAFT EVALUATION PLAN

• Plan also available in Portable Document Format (PDF)

New York Department of Health

New York Medicaid Redesign Team Section 1115 Demonstration

Start Date of Demonstration Period:        November 30, 2016
End Date of Demonstration Period:           March 31, 2021

DEMONSTRATION EVALUATION

On or before January 31, 2017, the state must submit to CMS for approval a draft design for the demonstration evaluation. At a minimum, the draft design must include a discussion of the goals, objectives, and hypotheses, with consideration of the beneficiaries, providers, plans, market areas, and/or expenditures specific to each of the programs. A separate design should be developed for each program, with sufficient methodological detail to determine scientific rigor, including its ability to demonstrate program effects, a plan for statistical analysis, and a description of the data sources to be used for each program.

To obtain public comment for inclusion into the evaluation design, a draft will be posted to the NYSDOH Web site by December 20, 2016, with comments requested from the public by January 10, 2017 to allow for incorporation of comments received prior to submission of the draft design to CMS.

The demonstration evaluation covers the overall demonstration, and should include following domains of focus:

• MLTC
• MMMC
• Individuals Moved from Institutional Settings to Community Settings for Long Term Services and Supports
• Temporary Assistance to Needy Families
• Twelve-Month Continuous Eligibility Period
• Express Lanes Eligibility

The evaluation of DSRIP, HARP and the Self-Direction Pilot are to be conducted separately and are described elsewhere in the document. The state may revise research questions with approval from CMS.

TECHNICAL APPROACH

As noted above, the primary goals of the Demonstration are to increase access, improve quality, and expand coverage to low income New Yorkers. To accomplish these goals, the Demonstration includes several key activities including enrollment of new populations, quality improvement and coverage expansions. This evaluation plan will assess the degree to which the key goals of the Demonstration have been achieved and/or the key activities of the Demonstration have been implemented.

Evaluation Plan Approach

The process of designing the evaluation plan first involved identifying and documenting the Demonstration´s key activities and associated goals, which were included in the State´s Demonstration extension proposal and the Special Terms and Conditions.

With key activities and goals identified, the process of designing the evaluation plan involved selecting evaluation questions that correspond to each of the major Demonstration goals and activities, building on the previous evaluation plan. The evaluation itself will seek to answer the evaluation questions and test related hypotheses, which in turn will assess the degree to which the Demonstration has been effective in implementing the key activities identified, directly achieving the goals of the Demonstration, or both.

The specific evaluation questions to be addressed by the evaluation were based on the following criteria:

1. Potential for improvement, consistent with the key goals of the Demonstration;
2. Potential for measurement, including (where possible and relevant) baseline measures that can help to isolate the effects of Demonstration initiatives and activities over time; and
3. Potential to coordinate with the DOH´s ongoing performance evaluation and monitoring efforts.

Once research questions were selected to address the Demonstration´s major program goals and activities, specific variables and measures were then identified to correspond to each research question. Finally, a process was developed for identifying data sources that are most appropriate and efficient in answering each of the evaluation questions.

The evaluation team will use all available data sources. The timing of data collection periods will vary depending on the data source, and on the specific Demonstration activity.

Analysis Plan

While the Demonstration seeks to generate cost savings and promote quality care, observed changes may be attributed to the Demonstration itself and/or external factors, including other State- or national-level policy or market changes or trends. For each Demonstration activity, a conceptual framework will be developed depicting how specific Demonstration goals, tasks, activities, and outcomes are causally connected to serve as the basis for the evaluation methodology. Methods chosen will attempt to account for any known or possible external influences (such as policy changes or market shifts) and their potential interactions with the Demonstration´s goals and activities. The evaluation will seek to isolate the effects of the Demonstration on the observed outcomes in several ways:

1. To the extent possible, credible contextual information will be gathered that attempts to isolate the Demonstration´s contribution to any observed effects as well as describe the relative contributions of other factors that may influence the observed effects. This will include documenting any relevant legal, regulatory, or policy changes or other trends – including the sequence, scope, and duration of such changes – at both a State and national level that are likely to influence the observed outcomes.
2. Where possible and relevant, the evaluation will incorporate baseline measures, and account for secular trends, for each of the selected variables included in the evaluation. Data for each of the targeted variables and measures will be collected regularly so that changes in outcome measures and variables can be observed on a longitudinal basis.
3. The evaluation will compare rates of performance and measures with State and national benchmarks, where relevant and feasible. Incorporating benchmark measures will allow for external comparisons of Demonstration measures to State and national trends, further isolating the impacts of the Demonstration by controlling for external factors influencing the observed effects.

The evaluation features described above (analysis of qualitative contextual information, the use of baseline measures, ongoing data collection, and benchmarking) represent quasi-experimental means by which the evaluation team will determine the effects of the Demonstration. Evaluation conclusions will include key findings associated with individual research questions addressed as well as integrated information combining the results of individual evaluation questions to make broad conclusions about the effects of the Demonstration as a whole. In addition, the evaluation will include specific recommendations of best practices and lessons learned that can be useful for DOH, other States, and CMS.

A competitive bidding process will be used to contract with an independent entity to conduct the evaluation, in which a Request for Proposals (RFP) will be developed and issued by NYSDOH. This RFP will describe the scope of work, the major tasks, and contract deliverables, with a period during which potential bidders can submit questions. Proposals received will undergo review by a panel of NYSDOH staff, using a scoring system developed for this RFP. Eligible bidders must not be employees or entities of the New York Department of Health, and not have any business relationship with any administrative or provider entities involved in Demonstration activities. Applicants will be evaluated on the basis of related work experience, staffing level and expertise, environment and resources, data analytic capacity, and ability to act as an independent, unbiased third party in conducting the evaluation.

Evaluation of Demonstration Activities

Individuals Moved from Institutional Settings to Community Settings for Long Term Services and Supports

The broad goals of New York´s Home and Community based services expansion (HCBS) program are to assess the impact of the demonstration on: 1) Improve care coordination; and 2) Improve patient safety and quality of care for consumers. Toward these goals, the following evaluation questions will be addressed using the data set from Money Follows the Person:

Goal 1: Improve care coordination

• Question 1: For the HCBS Expansion population that transitioned from an institutional setting, what is the average time in nursing facility prior to transition?
• Question 2: For the HCBS Expansion population that entered a Managed Long-Term Care plan (MLTC) after transitioning from an institutional setting, what are their demographic characteristics?

Goal 2: Improve patient safety and quality of care for consumers

• Question 1: For the HCBS Expansion population that entered MLTC after transitioning from an institutional setting, what percent had an emergency room visits in the last 90 days? What are the rates for falls requiring medical intervention and how have they changing since 2012?
• Question 2: For the HCBS Expansion population that entered MLTC after transitioning from an institutional setting, how long on average are they staying in the community before re-entering a nursing facility?

Evaluation

Annually, New York will perform the calculation of the proposed evaluation questions. The Department of Health has extensive experience with the computation and evaluation of quality performance measurement with a variety of service delivery entities, such as hospitals, managed care organizations, managed long-term care organizations, and nursing home.

Data Sources

Uniform Assessment System-NY (UAS-NY) Community Health data
The MLTC plans are required to collect and report to the NYSDOH information on enrollees´ levels of functional and cognitive impairment, behaviors, and clinical diagnoses. This information is collected at enrollment and then semi-annually thereafter. The UAS-NY is an electronic system based on a uniform data set, which standardizes and automates needs assessments for home and community based programs in New York. The UAS-NY is based on the interRAI suite of assessment instruments. interRAI is a collaborative network of researchers in over 30 countries committed to improving health care for persons who are elderly, frail, or disabled. Their goal is to promote evidence-based clinical practice and policy decisions through the collection and interpretation of high quality data. The interRAI organization and its assessment tools are used in many states, as well as Canada and other countries. Using the UAS- NY tool facilitates access to programs and services, eliminates duplicative assessment data, and improves consistency in the assessment process. With the UAS-NY, functional status data demonstrates its importance to inform eligibility for the MLTC program, provide the basis for the MLTC plans´ care management planning processes, and facilitate a plan´s identification of areas where the patient´s status differs from optimal health or functional status.

Assessment data are submitted by plans to the UAS-NY electronically as assessments are conducted, and are added to the database upon submission. Each year, MLTC UAS-NY submissions are created into two static assessment files. One containing the most recent assessment for enrollees in each plan from January through June. The second containing the most recent assessment for enrollees in each plan from July through December. These two files will be used to describe and evaluate the MLTC plan performance.

Money Follows the Person (MFP) data
The cohort for this evaluation will be defined by participation in the MFP program and utilize their tracking system. In January 2007, the federal Centers of Medicare and Medicaid Services (CMS) approved New York´s application to participate in the Money Follows the Person Rebalancing Demonstration Program. The MFP Demonstration, authorized under the Deficit Reduction Act and extended through the Affordable Care Act, involves transitioning eligible individuals from long-term institutions like nursing facilities and intermediate care facilities into qualified community-based settings. The initiative assists people who want to leave institutional care and receive services in their community of choice. The MFP Rebalancing Demonstration Grant helps states rebalance their Medicaid long-term care systems.

Minimum Data Set (MDS 3.0)
MDS 3.0 is a federally required standardized assessment and the basis of the comprehensive assessment for all residents of long-term care facilities. NY will uses this data to calculate the member´s time in a nursing facility prior to discharge to the community.

Managed Long Term Care Program

The broad goals of the New York Managed Long-Term Care (MLTC) program evaluation are to assess the impact of the demonstration on:

1. Improve care coordination for Medicaid´s highest risk/highest cost population;
2. Improve patient safety and quality of care for consumers;
3. Reduce preventable acute hospital admissions;
4. Improve satisfaction for consumers.

Toward these goals, the following evaluation questions will be addressed:

Goal 1: Improve care coordination

• Question 1: How has enrollment in MLTC plans increased since 2012?
• Question 2: What are the demographic characteristics of the MLTC population? Have they changed since 2012?
• Question 3: What are the functional and cognitive deficits of the MLTC population? Have they changed since 2012?
• Question 4: Are the statewide and plan-specific overall functional indices decreasing or staying the same since 2012?
• Question 5: Are the average cognitive and plan specific attributes decreasing or staying the same since 2012?

Goal 2: Improve patient safety and quality of care for consumers

• Question 1: What percent of members did not have an emergency room visit in the last 90 days? What are the rates for falls requiring medical intervention and how have they changed since 2012?
• Question 2: To what extent are enrollees able to receive timely access to personal, home care and other services such as dental care, optometry and audiology?
• Question 3: Are enrollees accessing necessary services such as flu shots and dental care?

Goal 3: Reduce preventable acute hospital admissions

• Question 1: What is the rate of potentially avoidable hospitalizations? Is the rate stable or decreasing?
• Question 2: What are the per member per month (PMPM) costs of the population?

Goal 4: Improve satisfaction for consumers

• Question 1: What is the percent of members who rated their managed long-term care plan within the last six months as good or excellent? Percent of members who rated the quality of care manager/case manager services within the last six months as good or excellent? Percent of members who in the last six months rated their home health aide/personal care aide/personal assistant, care manager/case manager, regular visiting nurse or covering/on call nurse services were usually or always on time? Percent of members who rated the quality of home health aide/personal care aide/personal assistant services within the last six months as good or excellent?

Evaluation

Annually, New York will perform the calculation of the proposed evaluation questions. The Department of Health has extensive experience with the computation and evaluation of quality performance measurement with a variety of service delivery entities, such as hospitals, managed care organizations, managed long-term care organizations, and nursing home.

Data Sources

Uniform Assessment System-NY (UAS-NY) Community Health data
The MLTC plans are required to collect and report to the NYSDOH information on enrollees´ levels of functional and cognitive impairment, behaviors, and clinical diagnoses. This information is collected at enrollment and then semi-annually thereafter. The UAS-NY is an electronic system based on a uniform data set, which standardizes and automates needs assessments for home and community based programs in New York. The UAS-NY is based on the interRAI suite of assessment instruments. interRAI is a collaborative network of researchers in over 30 countries committed to improving health care for persons who are elderly, frail, or disabled. Their goal is to promote evidence-based clinical practice and policy decisions through the collection and interpretation of high quality data. The interRAI organization and its assessment tools are used in many states, as well as Canada and other countries. Using the UAS-NY tool facilitates access to programs and services, eliminates duplicative assessment data, and improves consistency in the assessment process. With the UAS-NY, functional status data demonstrates its importance to inform eligibility for the MLTC program, provide the basis for the MLTC plans´ care management planning processes, and facilitate a plan´s identification of areas where the patient´s status differs from optimal health or functional status.

Assessment data are submitted by plans to the UAS-NY electronically as assessments are conducted, and are added to the database upon submission. Each year, MLTC UAS-NY submissions are created into two static assessment files. One containing the most recent assessment for enrollees in each plan from January through June. The second containing the most recent assessment for enrollees in each plan from July through December. These two files will be used to describe and evaluate the MLTC plan performance.

Satisfaction data
In 2007, the NYSDOH, in consultation with the MLTC plans, developed a satisfaction survey of MLTC enrollees. The survey was field tested and is now administered by the NYSDOH´s external quality review organization, IPRO. New York State sponsors the biennial MTLC satisfaction survey. The survey contained three sections: health plan satisfaction; satisfaction with select providers and services, including timeliness of care and access; and self-reported demographic information.

Express Lane-like Eligibility

Express Lane-like Eligibility refers to a Medicaid procedure in which individuals applying for Temporary Assistance (TA) are automatically considered for Medicaid enrollment without having to file a separate application. The underlying rationale is that Medicaid eligibility determination and enrollment can be facilitated given that, in most cases, applicants for TA are also eligible for Medicaid given the lower income threshold for the former. While Express Lane Eligibility does not represent a newly implemented Medicaid enrollment procedure, it´s authority under the 1115 Waiver, applied to adults, is a recent change.

Given the program objective of increasing access to health insurance through Medicaid by streamlining the application and enrollment process, the following questions would be addressed in the evaluation:

Question 1: How many and what percentage of Medicaid recipients are enrolled through Express Lane-like Eligibility?

Question 2: What are the demographic and clinical characteristics of Medicaid beneficiaries enrolled through Express Lane-like Eligibility as compared to those not enrolled through this mechanism?

Methods

Evaluation Objectives: Evaluation of the Express Lane-like eligibility initiative will provide feedback to program staff regarding the number and characteristics of Medicaid recipients enrolled through this mechanism, providing insights into how effectively the program reaches potential recipients in terms of both number and characteristics. Information gained could potentially be used to enroll potential recipient groups who may be underrepresented in this enrollment mechanism.

Question 1. How many and what percentage of Medicaid recipients are enrolled through Express Lane-like Eligibility?

While Express Lane-like eligibility is not a new Medicaid enrollment procedure, tracking of the number of recipients enrolled into Medicaid under this mechanism will be begin as soon as possible after November 30, 2016, the start date of Medicaid Redesign Team section 1115 demonstration. The number and percentage of recipients enrolled through the Express Lane-like eligibility mechanism will be determined monthly and annually over the duration of the demonstration.

Question 2. What are the demographic and clinical characteristics of Medicaid beneficiaries enrolled through Express Lane-like Eligibility as compared to those not enrolled through this mechanism?

Medicaid claims and enrollment data will be used to compare recipients enrolled through the Express Lane- like mechanism to those enrollees who did not, on demographic and clinical factors. A list of enrollees through this mechanism over a selected two-year period during the demonstration will be used to identify those individuals in the database. It is anticipated that a two year period will be a sufficient time frame in order to identify a sufficient number of enrollees to allow comparisons to be made. From the claims and enrollment data, demographic (age, sex, race/ethnicity, New York State region) and clinical information (presence or absence of chronic diseases, such as mental illness and diabetes, maternal/delivery, etc.) will be extracted, with comparisons to be made between Express Lane-like enrollment vs. non-Express Lane-like using analytic procedures such as chi-square analysis.

Twelve-Month Continuous Eligibility for Modified Adjusted Gross Income (MAGI) Individuals

The Twelve-Month Continuous Eligibility initiative, initiated in 2014 with the Affordable Care Act Marketplace, is to prevent lapses in Medicaid coverage due to fluctuations in recipient income, and applies to Medicaid recipients eligible under Modified Adjusted Gross Income (MAGI) guidelines. MAGI eligibility groups include the following:

• Pregnant women
• Infants and children under the age of 19
• Childless adults who are: not pregnant, age 19-64, not on Medicare, or could be certified as disabled but not on Medicare
• Parents/Caretaker relatives
• Family Planning Benefit Program
• Children in foster care

MAGI recipients remain eligible for Medicaid until renewal after a 12-month period, during which time recipients are not required to report changes in income, and such changes are not considered even if they are reported by the recipient. Changes in eligibility would be made only in the cases of death, moving out of state, or voluntary disenrollment in Medicaid. Given the program objective of limiting gaps in Medicaid coverage, the evaluation will address the following questions:

• Question 1: How many Medicaid beneficiaries have been continuously enrolled over 12- month intervals since its inception, and what are their demographic and clinical characteristics?
• Question 2: Did Medicaid enrollment months per member increase following the implementation of continuous eligibility as compared to pre-implementation?
• Question 3: Was there an increase in the percentage of Medicaid beneficiaries continuously enrolled for 12 months following implementation of continuous eligibility as compared to pre-implementation?
• Question 4: Given the intended continuity in Medicaid coverage, was there greater use of primary care utilization and other preventive services, and an associated reduction in overall health care costs under Medicaid, following the implementation of this initiative as compared to pre-implementation?
• Question 5: How many beneficiaries covered under continuous eligibility would have experienced lapses in Medicaid coverage if not for the Waiver?

Evaluation Objectives: Evaluation of the Twelve-Month Continuous Eligibility for MAGI Individuals program is to provide information to program managers on how effectively continuous enrollment is being implemented, the potential health care benefits associated with 12-month continuous eligibility, as well as possible effects on health care costs. Such information could potentially be used to make program modifications toward increasing effectiveness in preventing lapses in coverage, and/or to ensure greater inclusion of subgroups that may be underserved with this initiative, and to encourage use of preventive services resulting from increased Medicaid coverage to prevent more severe disease and, in turn, prevent potentially higher costs.

Methods

Question 1. How many Medicaid beneficiaries have been continuously enrolled over 12-month intervals since its inception, and what are their demographic and clinical characteristics as compared to Medicaid enrollees not continuously enrolled?

MAGI Medicaid enrollees will be identified, based on aid category codes, in the enrollment data from January 1, 2014 through December 1, 2018. Medicaid enrollment history for these recipients will be used to determine the number and proportion of recipients who had at least one 12-month period of continuous enrollment during this time period.

To understand the characteristics of MAGI recipients that receive 12-month enrollment, those with 12-month enrollment over the 4-year period will be compared to MAGI recipients not showing 12-month enrollment in their enrollment histories. Demographic variables on which comparison will be made include sex, race, and age. Additionally, the presence or absence of chronic diseases will be compared between these two groups as of recipients´ first month of enrollment Medicaid occurring on or after January 1, 2014. Comparisons will be made, using chi-square analysis, on the presence or absence of conditions such as HIV/AIDS, diabetes, serious mental illness, asthma, cardiovascular disease and kidney disease. Clinical Risk Group (CRG) categories and/or diagnosis codes on claims will be used to determine the presence of these conditions.

Question 2. Did Medicaid enrollment months per member increase following the implementation of continuous eligibility as compared to pre-implementation?

The hypothesis to be tested under this question is that, given the mechanism of 12-month continuous eligibility to prevent lapses in Medicaid coverage, months of enrollment per member will show an increase over the four years following the implementation of 12-month continuous eligibility as compared to the four years preceding its implementation.

Medicaid enrollment data will be used to determine months of enrollment per recipient. This will be determined for each of the five years prior to implementation of 12-month continuous eligibility (January 1, 2011 – December 1, 2013) and each of the five years following implementation (January 1, 2014 – December 1, 2018). An interrupted time series design1 is proposed to test this hypothesis in assessing the effect of the 12- month continuous eligibility initiative on Medicaid enrollment. This is a quasi-experimental design in which summary measures of the outcome variable (annual months of enrollment per member, in this case) are taken at equal time intervals over a period prior to program implementation, followed by a series of measurements at the same intervals over a period following program implementation. This design was chosen in consideration of the fact that a control group is unlikely to be available, limiting the ability to separate the effects of this initiative from other statewide health care reform initiatives that are ongoing (e.g., DSRIP, the Affordable Care Act). Given the limitation resulting from the likely absence of a comparison group, this design is advantageous in that potential confounders (i.e., other health care reform initiatives) are minimized in that they would have to occur contemporaneously with the introduction of 12-month continuous eligibility in order to exert a confounding effect, which is unlikely, but is recognized as possible nonetheless. This design also has the advantage of accounting for secular trends in the enrollment months per member, to which other health care reform initiatives may contribute.

Segmented regression2 will be used as the primary analytic strategy in the analysis of data under the interrupted time series design in testing this hypothesis. This analysis enables the evaluation of changes in the level and trend in the outcome variable, while controlling, as necessary, for such biases as secular trend, serial autocorrelation, and seasonal fluctuation in the outcome variable. A potential issue to address over the study period is change in characteristics of the Medicaid population over time. This could occur through increased enrollment of younger and healthier people into Medicaid, and/or increased movement of older and sicker people from Medicaid fee-for-service to managed care, either of which could confound the effects of the 12- month continuous eligibility initiative on member months of Medicaid enrollment. This will be addressed through adjustment of the outcome variable by standardizing on factors such as age, sex, and health status (e.g., Clinical Risk Grouping3, Charlson Comorbidity Index4), or inclusion of population-level measures of these variables as covariates in the model. Additionally, stratification will be used to assess differential program effects on months of Medicaid enrollment by recipient subgroups (e.g., sex, race, age, NYS region, mental health status).

Question 3. Was there an increase in the percentage of Medicaid beneficiaries continuously enrolled for 12 months following implementation of continuous eligibility as compared to pre-implementation?

Given the primary objective of the 12-month continuous eligibility initiative of preventing lapses in Medicaid coverage due to fluctuations in income, this question will be addressed through testing the hypothesis that that the percentage of recipients continuously enrolled for 12 months will increase in the years following the implementation of this initiative.

In the same manner and rationale as for Question 2, above, interrupted time series with segmented regression will be used to test this hypothesis. The dependent variable will be the proportion of enrollees continuously enrolled over a 12-month period, in each of the five years prior to implementation of 12-month continuous eligibility, and the five years after. Again, potential confounding due to changes in the Medicaid population will be controlled through standardizing the outcome variable on factors such as age, sex, and health status, or inclusion of such variables in the model, with stratification on various recipient subgroups to assess differential program effects.

Question 4. Given the intended continuity in Medicaid coverage, was there greater use of primary care utilization and other preventive services, and an associated reduction in overall health care costs under Medicaid, following the implementation of this initiative as compared to pre-implementation?

Three hypotheses will be tested in addressing this question:

1. The use of primary care and other preventive services will increase following the implementation of 12- monthy continuous eligibility. This is expected due to the anticipated continuity of coverage resulting from the initiative.
2. Health care costs for primary care and selected preventive care services will increase following the implementation of 12-month continuous eligibility, given the expected increase in utilization of these services.
3. Total cost of care per recipient will decrease following the implementation of 12-month continuous eligibility. This result is expected because fewer lapses in coverage should occur in the NYS Medicaid population, making preventive care more accessible and thus preventing more severe illness that is more costly to treat.

The Interrupted time series design with segmented regression analysis will be used to test these three hypotheses, comparing cost and utilization of primary and preventive care before and after program implementation. To control for the effect of year to year fluctuation in Medicaid enrollment on service utilization and cost, per member per year rates will be computed as the dependent variable in each analysis, for each of the four years prior to, and four years after, the start of the 12-month continuous eligibility initiative. For all analyses, potential confounding due to changes in the Medicaid population will be controlled through inclusion in the model of variables such as age, sex, presence of chronic diseases, or by standardizing the outcome variable on these factors. Results will be stratified by demographic and clinical recipient subgroups to assess differential program effects.

To test these three hypotheses, Medicaid claims data will be used to identify primary care and selected preventive services, including vaccines, and screening for type 2 diabetes, HIV, and elevated cholesterol for the outcome variables to test the first hypothesis. Costs associated with these services, as well as total care costs, will also be determined from Medicaid claims, to be used in computing the outcome variables for the second and third hypotheses, respectively. To compute per member per year rates for each of these services, the total number of services of each type paid by Medicaid each year will be determined, and divided by the total number of months of enrollment over all recipients for that year and the resulting quotient multiplied by Cost per member per year associated with primary care and preventive services, and for total health care costs, will be computed in the same manner.

Question 5. How many beneficiaries covered under continuous eligibility would have experienced lapses in Medicaid coverage if not for the Waiver?

Prior to implementation of the 12-Month Continuous Eligibility Initiative, Medicaid enrollees were subject to loss of coverage in the event that their incomes rose above the eligibility threshold. In order to quantify the number of MAGI enrollees who would have lost coverage using the previous eligibility criteria, Medicaid enrollment staff will maintain a record of reported changes in income received from enrollees. Such records will be used from the inception of the program, if available, or retention of these records will begin as soon as is logistically feasible to do so, and will be maintained on an ongoing basis. Given that Medicaid enrollees are not required to provide information on changes in income until time of eligibility renewal after 12 months, individuals who would otherwise have lost coverage will likely be undercounted.

Mainstream Medicaid Managed Care / Temporary Assistance to Needy Families (MMC/TANF)

The overarching goals of the New York MMMC/TANF program´s evaluation are to assess the impact of the demonstration on:

1. Expand MMMC/TANF enrollment;
2. Improve health care access for MMMC/TANF beneficiaries;
3. Continue to improve the quality of care;
4. Reduce the number of uninsured New Yorkers.

Toward these goals, the following evaluation questions will be addressed:

Goal 1: To expand MMC enrollment

• Question 1: How many beneficiaries were enrolled in MMC as a result of the demonstration? What percentage of Medicaid recipients are enrolled in an MMC program?

Goal 2: To improve health care access for Medicaid beneficiaries in New York

• Question 1: How much has access to primary care changed over the course of the waiver? Has the number of primary care visits total and per member per year changed?
• Question 2: How much has access to specialty care changed over the course of the waiver? Has the number of specialty care visits total and per member per year changed?
• Question 3: Has increased adoption of PCMH among Medicaid providers increased access to primary care? Has the proportion of Medicaid providers with PCMH recognition changed?
• Question 4: Are there any barriers to access to care or changes in beneficiary to provider ratios?
• Question 5: What are the differences in access to care for Medicaid subpopulations? Does utilization of services vary by population subgroups (race/ethnicity, rural/urban, aid category, age, gender and special needs)

Goal 3: To continue to improve the quality of care

• Question 1: How has quality of care for MMC changed over the course of the waiver? Measures such as those for preventive care, chronic condition treatment, potentially preventable use of ER and inpatient admissions, tobacco cessation discussions and awareness of services will be monitored for changes.
• Question 2: How does quality of care for NY MMC compare with national benchmarks?
• Question 3: Has the gap in measures of quality and satisfaction narrowed between NY MMC and commercial plans?
• Question 4: How has the expansion of the demonstration into new populations been implemented into measurement of these groups? How do HARP plans compare to MMC for behavioral measures? Has quality changed for members whose BH benefits were moved into MMMC?
• Question 5: Are there any disparities in quality of care for Medicaid enrollees? Does quality of care vary by population subgroups (race/ethnicity, rural/urban, aid category, age, gender and special needs)
• Question 6: How have Medicaid prenatal care standards improved prenatal care for enrollees? Has use of 17-P increase for eligible prenatal patients? Is the use of long-acting, reversible contraception methods increasing?

Goal 4: Reduce the number of Uninsured New Yorkers

• Question 1: How has expanded Medicaid eligibility (through the ACA) affected health coverage for low-income uninsured adults? Has enrollment in the expansion groups changed the proportion of other "traditional" MMMC categories? What are the demographic characteristics of the expansion group and how do they compare to "traditional" MMMC?

Evaluation

Annually, New York will perform the calculation of the proposed evaluation questions. The Department of Health has extensive experience with the computation and evaluation of quality performance measurement with a variety of service delivery entities, such as hospitals and managed care organizations. Enrollment by aid category, restricted/exemption codes and other identifiers will be monitored for changes during each reporting period.

Data Sources

Medicaid data warehouse - will be used for beneficiaries´ eligibility information, claims and encounter data reflecting the transactional data for diagnoses, use of services and procedures.

Provider Network Data System – Managed care plans submitted information for the providers in the network

PCMH File – Provider file from NCQA containing the PCMH certification level for all recognized providers in New York

CAHPS – A nationally recognized satisfaction survey asking members about access to care and their experiences with their health care providers and health

OHIP/EQRO Access & Availability Surveys – Access and availability surveys conducted on a regular basis to ensure timely access to types of providers is available to MMMC members

QARR – NYS´ quality reporting system for health plans contains annually submitted data for quality measures for HEDIS as well as NYS-specific measures

NCQA reports – National Committee for Quality Assurance publishes reports containing Medicaid and Commercial performance benchmarks. CMCS publishes annual Medicaid child and adult reports using their required core measure sets

Medicaid Prenatal Care Provider Reporting – Medicaid prenatal care providers submitted data for prenatal care screening and services

NYSoH Enrollment – QHP and Essential Plan enrollment data

References

1. Cook, TD, Campbell, DT. (1979). Quasi-experimentation: Design and analysis for field Settings. Boston, MA: Houghton Mifflin Company.
2. Gillings, D, Makuc, D, Siegel, E. Analysis of interrupted time series mortality trends: An example to evaluate regionalized perinatal care. American Journal of Public Health 1981;71;38-46.
3. Averill RF, Goldfield NI, Eisenhandler J, et al. Development and Evaluation of Clinical Risk Groups (CRGs). Wallingford, Conn: 3M Health Information Systems. Final Report to the National Institutes of Standards and Technology, US Department of Commerce; 1999.
4. Charlson, M.E., Pompei, P., Ales, K.L., & Mackenzie, C.R. (1987). A new method of classifying prognostic comorbidity in longitudinal studies: Development and validation Journal of chronic diseases, 40(5):373–383.