Family Planning Waiver Evaluation. Final Report. Presented to Florida Agency for Health Care Administration and Florida Department of Health

Final Report Presented to Florida Agency for Health Care Administration and Florida Department of Health June 6, 2003

Summary Preface The Family Planning Waiver Evaluation report is organized into two components: the Evaluation Summary and the Supporting Data. The Evaluation Summary contains an overview of the evaluation, including the original Family Planning Waiver hypotheses, significant findings, conclusions and recommendations. The Supporting Data includes the data upon which the summary is based. This report is the independent evaluation of the Maternal and Child Health and Education Research Data Center (the Research Center) at the University of Florida, College of Medicine, a Branch of the Lawton and Rhea Chiles Center for Healthy Mothers and Babies at the University of South Florida. The Department of Health and the Medicaid Division of the Agency for Health Care Administration provided the data on birth vital statistics, Medicaid eligibility, enrollment in the family planning waiver, family planning services and Medicaid claims. Their cooperation was always professional, timely, collegial, and thought provoking. Prepared by the Maternal Child Health and Education Research and Data Center at the University of Florida, College of Medicine, a Branch of the Lawton and Rhea Chiles Center for Healthy Mothers and Babies at the University of South Florida ii

Summary Family Planning Waiver Evaluation Table of Contents Overview......1 Scope of Work...1 Methodology...1 Subjects...1 Family Planning Eligibility and Services...1 Analysis...2 Summary of Key Findings...3 Summary of Additional Findings...5 Hypotheses and Outcomes from the Waiver Submitted Sept. 1997...6 Conclusion...10 Limitations...10 Section I- Operational Definitions of Key Terms and Abbreviations...12 Operational Definition of Key Terms...13 Abbreviations...17 Section II- pre-waiver Descriptive Statistics...19 Pre-Waiver 1995 Cohort-Summary of Subsequent Statistics.20 Pre Waiver 1995 Cohort Sociodemographic Descriptive Statistics...22 Pre Waiver 1995 Mothers Who Delivered and Had Subsequent Deliveries Within Two Years...24 Section III-1st-Year Waiver Implementation Descriptive Statistics...26 Post-Waiver 1995 Deliveries Cohort-Summary of Subsequent Statistics...27 Post Waiver 1999 Cohort Sociodemographic Descriptive Statistics...29 Post-Waiver 1999 Mothers who Delivered and had a Subsequent within Two Years...31 Section IV- pre-waiver to 1st-Year Waiver Implementation Comparisons...33 Pre-Waiver to post-waiver 1995 Deliveries Cohort-Summary of Subsequent Statistics...34 Section V- Medicaid Eligibility Distribution Pre-Family Planning Waiver (1995 Cohort...38 1995 pre-waiver Medicaid Eligibility Status SOBRA, non-sobra, None, and Medicaid Eligibility Status SOBRA, non-sobra, None at 240 Days Past...39 Medicaid Eligibility Status is a Key for the Family Planning Waiver Evaluation...40 1995 pre-waiver Cohort of Mothers Medicaid Eligibility Status by 30-Day Increments After Based on Mothers SOBRA Medicaid Eligibility at...41 Prepared by the Maternal Child Health and Education Research and Data Center at the University of Florida, College of Medicine, a Branch of the Lawton and Rhea Chiles Center for Healthy Mothers and Babies at the University of South Florida iii

Summary 1995 pre-waiver Cohort of Mothers Medicaid Eligibility Status by 30-Day Increments After Based on Mothers with non-sobra Medicaid Eligibility at...42 1995 pre-waiver Cohort of Mothers Medicaid Eligibility Status by 30-Day Increments After Based on Mothers with any Medicaid Eligibility at SOBRA and non-sobra...43 Family Planning Waiver (1995)- Medicaid SOBRA pre-waiver Birth Cohort Groups... 44 Family Planning Waiver (1995)- Medicaid non-sobra Groups...45 Pre-Family Planning Waiver (1995) Implementation...46 Pre-Waiver Period Medicaid Eligibility Status 1995...47 Pre-Waiver Period Medicaid Groups for Mother Who Delivered in 1995...49 Medicaid Eligibility Status is a Key for Maternal Child Health Outcomes...50 Mother s Medicaid Eligibility Status Prior to and After 1995-1998 Cohorts...51 Section VI-Medicaid Eligibility 1st-Year Waiver Implementation...55 1st-Year Waiver Implementation 1999 Post-Birth Distribution of Mother s Medicaid Eligibility Status by Number of Days After Based on Mother s with SOBRA Medicaid Eligibility at...56 1st-Year Waiver Implementation 1999 Post-Birth Distribution of Mother s Medicaid Eligibility Status by Number of Days After Birth Based on Mother s with non-sobra Medicaid Eligibility at...57 1st-Year Waiver Implementation 1999 Post-Birth Distribution of Mother s Medicaid Eligibility Status by Number of Days After Based on Mother s with Medicaid Eligibility at SOBRA and non-sobra...58 Family Planning Waiver 1st-Year Waiver Implementation Period 1999 Deliveries...59 Family Planning Waiver 1st-Year Waiver Implementation 1999 Lost Eligibility, Had Family Planning Eligibility, and Did or Did Not Receive Family Planning Services...60 1st-Year Family Planning Waiver Implementation 1999 Medicaid Eligibility Distributions...61 Post-Waiver Period Medicaid Eligibility Status 1999 Deliveries...62 Mother s Medicaid Eligibility Status Prior to and After 1999 Cohort...63 Section VII- Descriptive Statistics Subsequent Deliveries for Family Planning Waiver Evaluation Cohort ( Cohorts 1995-1999)...64 Summary of Descriptive Statistics...65 5 1-year Trends of Repeat Deliveries for 1995 Cohort by Mother s Medicaid Status...66 Prepared by the Maternal Child Health and Education Research and Data Center at the University of Florida, College of Medicine, a Branch of the Lawton and Rhea Chiles Center for Healthy Mothers and Babies at the University of South Florida iv

Summary Section VIII-Descriptive Statistics Subsequent Deliveries Within 2-5 Years, After Previous, Family Planning Implementation Cohort 1999...71 Summary of Descriptive Statistics For Subsequent Deliveries...72 Trends of Subsequent Deliveries within 2 Years by Mother Medicaid Status for 5 Cohorts...73 Trends of Subsequent Deliveries within 3 Years by Mother s Medicaid Status for 4 Cohorts...74 Trends of Subsequent Deliveries within 4 Years by Mother s Medicaid Status for 3 Cohorts...75 Trends of Subsequent Deliveries within 5 Years by Mother s Medicaid Status for 2 Cohorts...76 Subsequent Deliveries within 6 Years by Mother s Medicaid Status for 1995 Cohorts...77 Section IX-Descriptive Statistics Subsequent Deliveries Within 2-5 Years, by Cohorts (1995-1999)...78 Summary of Distribution of Subsequent Deliveries...79 Distribution of Number and Percent of Subsequent Deliveries within 2 Years for 5 Cohorts by Mother s Medicaid Status...80 Distribution of Number and Percent of Subsequent Deliveries within 3 Years for 4 Cohorts by Mother s Medicaid Status...82 Distribution of Number and Percent of Subsequent Deliveries within 4 Years for 3 Cohorts by Mother s Medicaid Status...85 Summary of Family Planning Waiver Evaluation Descriptive Statistics...86 Family Planning Waiver Evaluation Descriptive Statistics...87 Section X- Inter-birth Interval 1st-Year Implementation Cohort...89 1st-Year Waiver Implementation Inter-birth Intervals, 1999 Cohort By Medicaid Eligibility Groupings...90 Inter-birth Interval 1st-Year Implementation of Family Planning Waiver and Time Period When Received by Medicaid Eligibility Groups...92 1999 1st-Year Implementation Group Inter-birth Interval for Subsequent Birth Within 24 Months of Previous Birth...93 1st-Year Implementation Group Birth Interval by Time Period of When Family Planning Services Rendered (Less than 60 Days After ) and by Medicaid Eligibility Status Within 24 Months Based on Subsequent Births...94 1st-Year Implementation Group Birth Interval by Time Period of When Family Planning Services Rendered (Less Than or Greater Than 60 Days) and by Medicaid Eligibility Status Within 24 Months Based on Subsequent Birth...95 1st-Year Implementation Group Birth Interval by Time Period of When Family Planning Services Rendered (Greater than 60 Days After ) and by Medicaid Eligibility Status Within 24 Months Based on Subsequent Birth...96 Prepared by the Maternal Child Health and Education Research and Data Center at the University of Florida, College of Medicine, a Branch of the Lawton and Rhea Chiles Center for Healthy Mothers and Babies at the University of South Florida v

Summary Pre Waiver Inter-birth Intervals 1995 Cohort by Medicaid Eligibility Groupings...97 Post Waiver Interpregnancy Birth by Time Period of When Family Planning Services Rendered (Any Claims) and By Medicaid Eligibility Status Within 24 Months Based on Subsequent Birth Deliveries...98 Inter-birth Interval for First Family Planning Waiver Group for Women Who Received Family Planning Services, By Time Period and By Family Planning Services Groupings...99 Section XI-Descriptive Statistics Family Planning Services Pre Waiver...100 Family Planning Medicaid...101 Description of Family Planning Services By 7 Categories...102 Family Planning Services Received, 1st-Year Waiver By Group, Mothers Medicaid Eligibility SOBRA at...104 Family Planning Services Received, 1st-Year Waiver By Group, Mothers Medicaid Eligibility non-sobra at...105 Distribution of Family Planning Services By Type of Service During 1999 and 2000...106 1st-Year Family Planning Waiver-Number and Percent of Women Who Delivered in 1999 Who Had Received a Service in 1999-2000...107 1st-Year Family Planning Waiver-Distribution By Time Period When Waiver Eligible Women Receive Family Planning Services...108 Section XII-1st-Year Waiver Implementation Access to Family Planning Services Evaluation...109 Classification and Comparison of Women by Medicaid Status and Days After...110 Results...113 Conclusion...114 Section XIII-1st-Year Waiver Evaluation Utilization of Family Planning Services in Number and Rate of Subsequent Deliveries...115 1995 Pre Waiver Medicaid Eligibility Status-Number and Rate of Subsequent Deliveries...116 1st-Year Waiver Subsequent Evaluation Comparison Based on Family Planning Services Utilization by Medicaid Eligibility at Birth and at 240 Post...117 Number and Rate of Subsequent Deliveries by Time Period, Types of Family Planning Services and Medicaid Eligibility 1st-Year Implementation 1999...119 1st-Year Waiver Subsequent Evaluation Comparison Based on Family Planning Services Utilization Comparisons By Medicaid Eligibility Groups Based on Evaluation Design Groupings...121 Subsequent Deliveries and Associated Rates...123 Prepared by the Maternal Child Health and Education Research and Data Center at the University of Florida, College of Medicine, a Branch of the Lawton and Rhea Chiles Center for Healthy Mothers and Babies at the University of South Florida vi

Summary Section XIV-Impact of Utilization of Family Planning Services...124 Received Family Planning Services in 1st-Year Waiver Implemented Period (1999)...125 Number of Subsequent Repeat Deliveries Based on Access Medicaid Grouping for Those Who Did and Did Not Receive Family Planning Services...126 Pre-waiver and 1 st Year Implementation...127 Hierarchical Order of Medicaid/non-Medicaid Eligibility Status and Number and Rate of Subsequent Repeat Deliveries within 24 Months of Their Previous...127 Design Modification Grouping...132 Findings from Reduction and Subsequent Deliveries Based on Receiving Family Planning Services...133 Group 1-Continuously Enrolled Based on Access Grouping...134 Group 2-Not Continuously Enrolled (Lost Eligibility) Waiver Eligible Based on Access Grouping...135 Group 3-Non-Medicaid Based on Modified Design Grouping...136 Group A-Medicaid-Continuously Enrolled in Medicaid Based on Access Grouping...137 Group B-Not Continuously Enrolled (Lost Eligibility) Based on Modified Design Groupings...138 Group C-Non-Medicaid Eligible...139 Group 4- Based on Original Evaluations Design in Proposal...140 Group 5-Original Design in Proposal...141 Group 6-Original Design in Proposal...142 Section XV-Medicaid Cost Associated With Implementation of Family Planning Waiver...143 Findings...144 Medicaid Continuously Enrolled-Type of Family Planning Service Category(s) by Period of Family Planning Service, Family Planning Services, Associated Changes, and Subsequent (s)...145 Waiver Eligible-Type of Family Planning Service Category(s) By Period of Family Planning Service and Family Planning Services, Associated Costs, and Subsequent Deliveries...146 Section XVI-Prototype and an Example of Fiscal Impact Analysis for 1st-Year Family Planning Evaluation...147 Costs Associated With Women Who Only Received Non-Sterilization Services...148 Percentage and Cost of Sterilization Services...149 Costs of Family Planning Services...150 Prototype and Example of Fiscal Methodology Used to Determine Fiscal Cost/ Benefits Pertaining to Family Planning Wavier...151 Prototype and Example of Pregnancy Cost Savings Based on Medicaid Reimbursement...152 Prepared by the Maternal Child Health and Education Research and Data Center at the University of Florida, College of Medicine, a Branch of the Lawton and Rhea Chiles Center for Healthy Mothers and Babies at the University of South Florida vii

Summary Prototype and Example of Distribution of Prevented Subsequent Pregnancies and Subsequent Children by High Risk Pregnancy/ Birth/Infants Requiring Neonatal Intensive Care...153 Prototype and Example of Medical Newborn Cost Savings Based on Medicaid Reimbursement...154 Prototype and Calculations...155 Section XVII Teenage 1999 Deliveries...157 Findings...158 Post-Waiver Period Medicaid Eligibility Status...159 Teenage 1 st -Year Waiver Implementation Inter-birth Intervals 1999 Cohort By Medicaid Eligibility Groupings...160 Teenage 1999 1 st -Year Waiver Implementation Medicaid Eligibility Status Number and Rate of Subsequent Deliveries...161 1 Year Report Deliveries for 1999 Teenage Cohort by Mother s Medicaid Status...162 Teenage 1 st -Year Waiver Subsequent Evaluation Comparison Based on Family Planning Services Utilization by Medicaid Eligibility at Birth...163 Teenage Group 1 Continuously Enrolled Based on Access Grouping...164 Teenage Group 2 Continuously Enrolled Based on Access Grouping...165 Teenage Group 3 Non Medicaid Eligible...166 Teenage Costs of Family Planning Services...167 Section XVIII-Appendices...168 Maternal Child Health and Education Research and Data Center (The Research Center)-Contract Timeline...169 Section XIX-Acknowledgements...170 Prepared by the Maternal Child Health and Education Research and Data Center at the University of Florida, College of Medicine, a Branch of the Lawton and Rhea Chiles Center for Healthy Mothers and Babies at the University of South Florida viii

Summary Overview Scope of Work The Medicaid Family Planning Waiver Evaluation is a quantitative study assessing the effect of providing access to family planning services on the following outcomes: 1) Frequency of subsequent pregnancies 2) Related costs The evaluation measured the impact on these two outcomes for women who were eligible for and actually used at least one family planning service compared to those women who were eligible for but did not use extended family planning services. The evaluation further sub-divided the participants in the waiver into SOBRA and non-sobra categories. Methodology Subjects: All Medicaid pregnant women who gave birth in Florida between 1/1/1995 and 12/31/1995 (pre-waiver population) All Medicaid pregnant women who gave birth in Florida between 1/1/1999 and12/31/1999 (first year Waiver implementation population) All non-medicaid pregnant women who gave birth in the above two time periods. Medicaid eligibility status was determined at three time periods: delivery, 60 days after delivery, and 240 days after delivery. Family Planning Eligibility and Services: Under the Family Planning Waiver, family planning services were available for two years after delivery to pregnant women who delivered under Medicaid but lost eligibility for all other Medicaid services 60 days after delivery. Family planning services were counted in determining participation in the Waiver and the costs of the Waiver if they occurred between 60 days and two years after birth. Sterilizations were not included in the determination of the costs of the Waiver because it was not possible in many instances to determine when the sterilization actually occurred and there was no established method for calculating avoided subsequent births over time related to women who were sterilized. The data on sterilizations and the costs associated with these procedures are included in the Supporting Data component of this report in the event the Agency decides to include these factors in making its decision on continuing the Waiver. (See pp. 147-155.) Prepared by the Maternal Child Health and Education Research and Data Center at the University of Florida, College of Medicine, a Branch of the Lawton and Rhea Chiles Center for Healthy Mothers and Babies at the University of South Florida 1

Summary Analysis: The Research Center linked and analyzed birth vital statistics data to Medicaid eligibility files, delivery and birth hospital discharge data files, and Medicaid claims data files. These data files were used to ascertain the impact of Florida s Medicaid Family Planning Waiver on subsequent deliveries, inter-birth intervals, the costs of the family planning services provided under the Waiver, and cost savings, if any, associated with a reduction in the rate of subsequent births to women who were eligible for the Waiver. The initial evaluation design proposed to compare the birth history of the following two groups: Pre-Waiver Cohort or women who delivered in 1995 under the Medicaid program prior to the Waiver, lost their eligibility for Medicaid services 60 days after delivery and were not eligible for family planning services for at least 240 days after the birth of the index child. Waiver Cohort or women who delivered in 1999 under the Medicaid program, lost their eligibility for Medicaid services 60 days after delivery, but retained their eligibility for family planning services for two years postpartum because of the Waiver. Subsequent deliveries within the 24 months of the index birth were ascertained for both groups by linking to birth vital statistics records from previous years. Though subsequent delivery rates could be calculated for both groups (see summary of key data below), the Research Center realized that the pre-waiver cohort could not serve as an adequate comparison group for the following reasons: 1. The introduction of Welfare Reform in 1996 dramatically changed the economic characteristics of women in the post Welfare Reform Waiver cohort by increasing the number of lower income women in this cohort compared to the pre-waiver cohort. Lower income women have significantly higher rates of subsequent births than do other women as illustrated by the fact that women who were continuously enrolled in Medicaid had higher subsequent birth rates both before and after the Waiver (.189 and.208 respectively). Women continuously enrolled in Medicaid are generally lower income women. (See p. 123.) 2. The family planning experience of the pre-waiver cohort is not available in the Medicaid claims data. The pre Waiver cohort includes women who may or may not have received family planning services post-partum independent of Medicaid funding. Therefore, it is not possible to relate the subsequent birth rate of this cohort to their utilization of family planning Prepared by the Maternal Child Health and Education Research and Data Center at the University of Florida, College of Medicine, a Branch of the Lawton and Rhea Chiles Center for Healthy Mothers and Babies at the University of South Florida 2

Summary Summary of Key Findings: Given the lack of comparability between the pre-waiver cohort and the Waiver cohort, the Research Center concluded that a more valid method of evaluating the effectiveness of the Waiver would be to compare women who were eligible for the Waiver and used family planning services funded by the Waiver to women who were eligible, but did not use Waiver services. This approach enabled the Research Center to compare similar groups, determine their respective subsequent birth rates and calculate the cost benefit of Waiver services. The user and nonuser groups were further divided into teenagers and adults, at the request of the Department of Health, to determine if there was a differential impact related to age. For more detailed analysis, the Research Center subdivided all analysis groups into SOBRA and non-sobra categories and all groups were compared to women who never gave birth funded by the Medicaid program. This level of analysis provided interesting information with regard to the birth patterns of these subgroups. This information could be useful in designing strategies to address unique characteristics of subpopulations, but it does not significantly alter the overall conclusions with regard to the impact on subsequent births and related costs experienced in the first year implementation of the Waiver. The following table (on p. 4) summarizes the critical information with regard to effect of the Waiver on subsequent births and costs. Key findings derived from this table are outlined below the table. More detailed data that supports this table and the associated findings is included in the Supporting Data component of this report. The relevant pages in the Supporting Data for each finding are included in parenthesis at the end of the finding. Prepared by the Maternal Child Health and Education Research and Data Center at the University of Florida, College of Medicine, a Branch of the Lawton and Rhea Chiles Center for Healthy Mothers and Babies at the University of South Florida 3

Summary Pre and Post Waiver Implementation Overview Comparisons Between Women Who Did and Did Not Receive Family Planning Services (FPS) By Subsequent Births, Rate of Subsequent Births, Prevented Subsequent Births, and Cost Savings # of Women # Subsequent Deliveries Rate of Subsequent Deliveries Pre-Waiver Cohort 1 45,678 5,618.123 Waiver Cohort 2,4 51,487 6,786.132 # of Prevented Births 3 Cost Avoided $ Cost of Family Planning Services (FPS) 4,5,6 Cost Savings 7 $ Teenagers Who Used FPS >60 Days Post Only Teenagers Who Did Not Use FPS Teenagers Who Used FPS at Both <60 and >60 Days Post Teenagers Who Used FPS <60 Days Post Only Adults Who Used FPS >60 Days Post Only Adults Who Did Not Use FPS Adults Who Used FPS at Both <60 and >60 Days Post Adults Who Used FPS <60 Days Post Only 2,569 444.172 121 1,645,116 576,223 1,068,893 7,569 1,660.219 469 58.124 37 503,052 163,676 339,376 477 96.201 5,493 600.109 38 516,648 892,679 (376,031) 31,847 3,685.116 1,364 106.078 4 34,384 317,823 (263,439) 1,699 137.081 TOTAL 200 2,719,200 1,950,401 768,799 1 The pre-waiver cohort equals women who gave birth in 1995 funded by Medicaid and subsequently lost eligibility for Medicaid, including family planning services, 60 days after delivery and did not regain Medicaid eligibility for at least 240 days after the index birth. 2 The waiver cohort equals women who gave birth in 1999 funded by Medicaid and subsequently lost eligibility for Medicaid 60 days after the index birth, but retained eligibility for family planning services funded by the Medicaid Family Planning Waiver. 3 Prevented subsequent births equals the number of births that did not occur in the first 2 years after the index birth for the waiver cohort in comparison to the pre-waiver cohort. 4 Does not include sterilization costs. 5 The cost of FPS was derived from Medicaid claims data. 6 $13,596 equals the average savings for a prevented delivery and infant birth. 7 See page 151 for cost saving calculations. Prepared by the Maternal Child Health and Education Research and Data Center at the University of Florida, College of Medicine, a Branch of the Lawton and Rhea Chiles Center for Healthy Mothers and Babies at the University of South Florida 4

Summary Summary of Key Findings (continued) Two hundred births were averted during the two-year measurement period for women who used family planning services funded by the Waiver. The net savings related to these averted births was $768,799. The cost per women who actually used Waiver services was $274. The cost saved per averted birth was $13,596. (See p. 155 for labor/delivery and infant hospitalization costs.) Only 9,895 women who were eligible for family planning services under the Waiver actually used these services. This represents 19% of the total eligible women. (See pp. 134-142.) Both teenage and adult women in the Waiver cohort who used family planning services had lower subsequent birth rates than their counterparts who did not use family planning services. (See pp. 159-166.) Teenagers who used family planning services had a greater reduction in subsequent birth rates than did adults who used services. (See pp. 159-166.) Of the 200 births averted, 158 related to teenagers and only 42 to adults. Consequently, all of the cost savings in the program are attributed to reductions in teenage subsequent birth rates. (See table p. 4.) Summary of Additional Findings: In addition to the above key findings, the following findings shed light on the experience of sub-populations within the Waiver cohort. Data related to these findings is included in the Supporting Data component of this report. The relevant pages for each finding are included in parenthesis at the end of the finding. Non-SOBRA Medicaid women who continued Medicaid eligibility post-partum, including access to family planning services, had a lower utilization rate of family planning services than women who were eligible for family planning services because of the Waiver (primarily SOBRA women). (See pp. 104-105.) Non-SOBRA Medicaid women have higher subsequent delivery rates than SOBRA Medicaid women. (See pp. 125-131.) There was essentially no difference in the inter-birth interval between women who did and did not receive family planning services. (See p. 90.) There was no major difference in the mean or median inter-birth intervals between SOBRA and non-sobra Medicaid women. Both of these groups had inter-birth intervals that were slightly lower than the 1995 and 1999 non-medicaid delivery cohorts. (See p. 90.) Prepared by the Maternal Child Health and Education Research and Data Center at the University of Florida, College of Medicine, a Branch of the Lawton and Rhea Chiles Center for Healthy Mothers and Babies at the University of South Florida 5

Hypotheses and Outcomes from the Agency for Health Care Administration Waiver Application Submitted September 1997 The Medicaid Family Planning Waiver included seven hypotheses that the Agency and the Department of Health considered to be appropriate indicators of the success of the Waiver in contributing to improving maternal and child health in Florida. All of these hypotheses were examined in the course of this evaluation. Some of the hypotheses could not be measured in the time period of the evaluation or by the data available to the Research Center. Each hypothesis and its outcome are provided below to the extent the hypothesis could be measured within the limitations of time and data. Hypothesis 1: Measure: Data Required: Data Source: Outcome: The demonstration waiver will result in an increase in the annual proportion of clients receiving Medicaid paid family planning services in Florida. The number and percent of women who receive a Medicaid funded family planning service in the first year of the Waiver compared to the number and percent of women who received Medicaid funded family planning services in the pre-waiver comparison year (1995). 1. The total number of women who were eligible to receive a Medicaid funded family planning service in 1995. 2. The number of women who actually received a Medicaid funded family planning service in 1995. 3. The number of women who were eligible to receive a Medicaid funded family planning service in the first year of the Waiver. 4. The number of women who actually received a Medicaid funded family planning service in the first year of the Waiver. Florida Medicaid Program This hypothesis could not be measured for the following reasons: A comparison between pre- and post-waiver Medicaid recipients would not be valid because prior to the Waiver women who were eligible for Medicaid at delivery, but lost their eligibility 60 days postpartum, also lost their ability to access Medicaid funded family planning services. These women represent higher income Medicaid recipients so that the Medicaid eligible group pre-waiver does not have the same characteristics as the Waiver group, and the number of Medicaid eligible women pre-waiver is significantly less than the Waiver population because of different eligibility requirements. In other words, this comparison would be apples to oranges. Another type of comparison could also be applied to partially address this hypothesis, i.e. a comparison between utilization in the first and second year of the Waiver. This comparison would simply indicate if utilization increased or decreased over the life of the Waiver. This Prepared by the Maternal Child Health and Education Research and Data Center at the University of Florida, College of Medicine, a Branch of the Lawton and Rhea Chiles Center for Healthy Mothers and Babies at the University of South Florida 6

comparison could not be made during this evaluation because Medicaid claims data for 2002 is not currently available. Hypothesis 2: Measure: Data Required: Data Source: Outcome: The proportion of women in the target population who experience repeat Medicaid deliveries within two years will decline. The proportion of women with a Medicaid funded birth during a fiscal year that has a subsequent Medicaid birth funded by the Waiver within two years compared to a second cohort of women in the next year of Waiver implementation. 1. The number of Medicaid paid births for women covered by the Waiver during fiscal year 1999. 2. The number of women covered by the Waiver in 1999 that have another Medicaid birth within two years. Florida Medicaid Program This hypothesis was measured as follows: The proportion of women in the target population who experienced repeat Medicaid deliveries pre and post waiver implementation was essentially the same. A portion of the slight increase in subsequent birth rates for Waiver eligible women compared to pre-waiver women can be largely attributed to the implementation of Welfare Reform, which shifted more low-income women into the Waiver eligible group. In addition, there was a slight increase in subsequent birth rates overall. The magnitude of the increase, however, was less in the waiver-eligible group than among women who were otherwise eligible. Hypothesis 3: Corollary 3a: Measures: Data required: The demonstration waiver will result in a decrease in the annual rate of Medicaid paid deliveries in Florida. A decrease in Medicaid paid deliveries in Florida will result in a decrease in annual expenditures for prenatal, birth, newborn, and infant care. The numbers of fiscal year Medicaid paid deliveries along with associated prenatal, delivery, newborn, and infant care expenditures. The aggregate number of fiscal year Medicaid paid deliveries and associated total expenditures. Outcome: The Waiver resulted in reduction of 200 deliveries in the first two years of implementation with a total cost savings of $768,799. The Waiver did not result in an overall reduction in the annual rate of Medicaid paid deliveries. Prepared by the Maternal Child Health and Education Research and Data Center at the University of Florida, College of Medicine, a Branch of the Lawton and Rhea Chiles Center for Healthy Mothers and Babies at the University of South Florida 7

If the absolute number of Medicaid paid deliveries does not decline and the associated expenditures increase, then hypothesis 4 will be tested. Hypothesis 4: Corollary 4a: Measure: Data Required: Data Source: Outcome: Hypothesis 5: Measure: Data Required: The demonstration waiver will result in a slower annual rate of growth in Medicaid paid deliveries in Florida. A slower annual rate of growth in Medicaid paid deliveries in Florida will result in a slower annual rate of growth in expenditures for prenatal, delivery, newborn, and infant care. The percentage change in the number of Medicaid paid deliveries each fiscal year along with the change in associated prenatal, delivery, newborn, and infant care expenditures will be compared to what would be expected given population growth by age cohort, and changes in the economy, as measured by the percent of women of child bearing age in households below 185 percent of the federal poverty level. 1. The aggregate number of fiscal year Medicaid paid deliveries and associated total expenditures. 2. The number of women of childbearing age and percent of women in households below 185 percent of the federal poverty level. Florida Medicaid Program The percentage of Medicaid deliveries has increased since the waiver was implemented. In the 1999 fiscal year, 43.2% of all deliveries were Medicaid; in 2000, 44.1% of all deliveries were Medicaid; and in 2001, 45.7% of all deliveries were Medicaid, a 2.1% increase between 1999 2000 and a 3.6% increase between 2000 2001. The demonstration waiver will produce a net annual savings in State and Federal Medicaid expenditures for birth-related services. The estimated fiscal year Medicaid savings from births prevented by the waiver less fiscal year Medicaid family planning expenditures for the waiver target population. 1. The estimated number of Medicaid paid births, in a given fiscal year, to the waiver target population based on previous years experience. 2. The actual number of Medicaid paid births to this population. 3. The average associated prenatal, birth, newborn, and infant care costs for a Medicaid birth. 4. The total fiscal year Medicaid family planning expenditures for the waiver target population. Prepared by the Maternal Child Health and Education Research and Data Center at the University of Florida, College of Medicine, a Branch of the Lawton and Rhea Chiles Center for Healthy Mothers and Babies at the University of South Florida 8

Outcome: Hypothesis 6: Measure: Data Required: Data Source: Outcome: Hypothesis 7: Measures: Data Required: Data Source: Outcome: The Waiver resulted in a net savings of $768,799 in the first two-year measurement period. These savings were all attributable to a reduction in teenage subsequent birth rates. The demonstration waiver will support a continued decrease in the Florida resident infant death rate. The number of Florida resident infant deaths divided by the number of Florida resident births multiplied by 1000 to establish a rate. 1. The Florida resident infant death rate for the years of the waiver, 1999, 2000, 2001, 2002. 2. The number of averted subsequent births due to the family planning waiver. Florida Department of Health Vital Statistics The number of averted births resulting from the waiver was 200 in the first two-year measurement period of the Waiver. Assuming an infant mortality rate of 7 per 1000 live births, a reduction of 200 births would result in 1.4 infant deaths. This number is too small to have a statistically significant impact on infant mortality. The demonstration waiver will result in a decrease in the annual teen birth rate and associated repeat teen pregnancies. The number of births to teens (age 19 and under) divided by the total number of teens (age 19 and under) multiplied by 1000 to establish a rate. The number of births to teens (age 19 and under) and the total number of teens (age 19 and under). Florida Department of Health Vital Statistics The Family Planning Waiver only serves women after they have given birth funded by Medicaid, and by definition, therefore, could not have any impact on a first pregnancy. Therefore, the waiver could only have an impact on subsequent pregnancies to teenagers. The number of averted pregnancies to teenagers as a result of the waiver was 158 in the first twoyear measurement period. This was the most positive result of the initial implementation phase of the Waiver. Prepared by the Maternal Child Health and Education Research and Data Center at the University of Florida, College of Medicine, a Branch of the Lawton and Rhea Chiles Center for Healthy Mothers and Babies at the University of South Florida 9

Conclusion The Family Planning Waiver had a positive impact on subsequent birth rates and costs to the Medicaid Program for women who chose to utilize family planning services, particularly for teenagers. Because data was not available to examine the impact of the waiver beyond two years after its implementation, the impact of the Waiver has not been satisfactorily assessed. In addition, the level of utilization of Waiver services during the evaluation period suggests that many women were not aware of the services or did not understand that Medicaid would pay for such services even though they were not eligible for Medicaid. This observation is supported by the qualitative study that was conducted by The Chiles Center in the second year of this evaluation. The qualitative study, directly comparable to the time frame for this analysis, found that many eligible women did not understand that they were eligible and consequently did not use the Waiver services. Teenage mothers who utilized Family Planning Services benefited most from the Waiver. This is very significant since the avoidance of a second birth by a teenager is highly correlated with a reduction in poverty, increased high school graduation rates, and reduction in child maltreatment. The cost savings included in the study related only to the cost of family planning services and did not include cost savings resulting from a reduction in negative outcomes associated with second births to teenagers. The Research Center recommends that the Waiver be extended for at least four more years and that this extension be accompanied by an information campaign conducted by the Agency and the Department of Health to inform women of the availability and benefits of participating in the Program. Limitations Associated with Evaluation of Family Planning Waiver 1. No date of services for sterilization is included in the outpatient database. This is needed to determination the proper assignment of women who received sterilizations post-delivery to appropriate groups for Waiver evaluation purposes. 2. An endpoint of 24 months only captures 1/3 of subsequent deliveries within a 48- month period. The utilization of family planning services would have a greater impact over a longer period of analysis than reflected in this report. 3. The Research Center could not establish a methodology for calculating averted birth rates for women who were sterilized and, therefore, could not determine costs or cost savings associated with this service. The Research Center recommends that the costs of sterilization should be spread over time based on some formula that included average birth rates for the affected population over time. This approach would prevent attributing the total costs of sterilization of a single fiscal year. The Agency may wish to construct a procedure for this purpose and use the data available in this report. 4. In some instances, it was difficult to determine if a women should be assigned to the Waiver cohort because some Medicaid non-sobra women at birth had a Family Planning eligibility flag between 60 and 240 days past delivery, and some SOBRA Prepared by the Maternal Child Health and Education Research and Data Center at the University of Florida, College of Medicine, a Branch of the Lawton and Rhea Chiles Center for Healthy Mothers and Babies at the University of South Florida 10

women at birth remained in the Medicaid program beyond the allowed 60 day postpartum period. An additional small number of women shifted Medicaid eligibility status between 60 and 240 days. The impact of these categorization issues was not considered large enough to affect the overall conclusions in the report. Prepared by the Maternal Child Health and Education Research and Data Center at the University of Florida, College of Medicine, a Branch of the Lawton and Rhea Chiles Center for Healthy Mothers and Babies at the University of South Florida 11

Section I Operational Definitions of Key Terms And Abbreviations Prepared by the Maternal Child Health and Education Research and Data Center at the University of Florida, College of Medicine, a Branch of the Lawton and Rhea Chiles Center for Healthy Mothers and Babies at the University of South Florida 12

Operational Definition of Key Terms Key Terms Family Planning pre-waiver Period Definitions 1995 Cohort 1/1/95-12/31/95 1st-Year Family Planning Waiver Period 1999 Cohort 1/1/99 12/31/99 Medicaid Eligibility Determination Based on Medicaid eligibility at time of delivery from Medicaid eligibility files Medicaid SOBRA 1995 & 1999 Eligibility at delivery based on SOBRA flag in Medicaid eligibility files non-sobra 1995 &1999 Eligibility at birth based on non- SOBRA flag in Medicaid eligibility files Total Medicaid Cohort 19995 & 1999 SOBRA and non-sobra Medicaid eligibility at delivery non-medicaid No Medicaid eligibility flag in Medicaid eligibility files Prepared by the Maternal Child Health and Education Research and Data Center at the University of Florida, College of Medicine, a Branch of the Lawton and Rhea Chiles Center for Healthy Mothers and Babies at the University of South Florida 13

Operational Definition of Key Terms (continued) Key Terms Group 1- Access Continuously Enrolled Group 2- Access Not Continuously Enrolled (Lost Eligibility) Potentially Waiver Eligible Group 3- Access non-medicaid Eligible Definitions Medicaid SOBRA @ Birth & Medicaid SOBRA @ 240 days Medicaid SOBRA @ Birth & Medicaid non-sobra @ 240 days Medicaid non-sobra @ Birth & Medicaid SOBRA @ 240 days Medicaid non-sobra @ Birth & Medicaid non- SOBRA @ 240 days Family Planning Services @ Birth & Medicaid SOBRA @ 240 days Family Planning Services @ Birth & Medicaid non- SOBRA @ 240 days Medicaid SOBRA @ Birth & non-medicaid @ 240 days Medicaid non-sobra @ Birth & non-medicaid @ 240 days Medicaid SOBRA @ Birth & Family Planning Services @ 240 days Medicaid non-sobra @ Birth & Family Planning Services @ 240 days Family Planning Services @ Birth & non-medicaid @ 240 Days Family Planning Services @ Birth & Family Planning Services @ 240 Days Non-Medicaid @ Birth & Medicaid SOBRA @ 240 days Non-Medicaid @ Birth & Medicaid non-sobra @ 240 days Non-Medicaid @ Birth & non-medicaid @ 240 days Non-Medicaid @ Birth & Family Planning Services @ 240 days Prepared by the Maternal Child Health and Education Research and Data Center at the University of Florida, College of Medicine, a Branch of the Lawton and Rhea Chiles Center for Healthy Mothers and Babies at the University of South Florida 14

Operational Definition of Key Terms (continued) Key Terms Group A- Medicaid-Continuously Enrolled in Medicaid Group B- Modified Design- Not Continuously Enrolled-Lost Eligibility Definitions Medicaid SOBRA @ Birth & Medicaid Non- SOBRA @ 240 days Medicaid non-sobra @ Birth & Medicaid non- SOBRA @ 240 days non-medicaid @ Birth & Medicaid non-sobra @ 240 days Family Planning Services @ Birth & Medicaid non- SOBRA @ 240 days Medicaid SOBRA @ Birth & Medicaid SOBRA @ 240 days Medicaid SOBRA @ Birth & non-medicaid @ 240 days Medicaid SOBRA @ Birth & Family Planning Services @ 240 days Medicaid non-sobra @ Birth & Medicaid SOBRA @ 240 days Medicaid non-sobra @ Birth & non-medicaid @ 240 days Medicaid non-sobra @ Birth & Family Planning Services @ 240 days non-medicaid @ Birth & Family Planning Services@ 240 days Family Planning Services @ Birth & Medicaid SOBRA @ 240 days Family Planning Services @ Birth & non-medicaid @ 240 days Family Planning Services@ Birth & Family Planning Services @ 240 days Group C- Modified Design- non- Medicaid Non-Medicaid @ Birth & Medicaid SOBRA @ 240 days Non-Medicaid @ Birth & non-medicaid @ 240 days Prepared by the Maternal Child Health and Education Research and Data Center at the University of Florida, College of Medicine, a Branch of the Lawton and Rhea Chiles Center for Healthy Mothers and Babies at the University of South Florida 15

Operational Definition of Key Terms (continued) Key Terms Standard Family Planning Services Enhanced Family Planning Service Standard and Enhanced Family Planning Services Definitions SOBRA or non-sobra Medicaid eligible women @ delivery who received Family Planning Services prior to 60 days post delivery Women who would have lost eligibility prior to waiver and received Family Planning Services after 60 days post delivery Family Planning Services received prior to and after 60 days post delivery Prepared by the Maternal Child Health and Education Research and Data Center at the University of Florida, College of Medicine, a Branch of the Lawton and Rhea Chiles Center for Healthy Mothers and Babies at the University of South Florida 16

Abbreviations Abbreviation Description < 60 days Received Family Planning Services prior to 60 days post delivery < 60 days > Received Family Planning Services prior to and after 60 days post delivery 60 > days Received Family Planning Services after 60 days post delivery 1Y 1 st -Year Family Planning Waiver Implementation calendar year 1999 deliveries BVS DoD FP FRS M MFPWE MM P MM T MU MRMP MRMT N NM NM @ 240 Birth Vital Statistics Date of Family Planning Medicaid eligible Family Planning Services Medicaid Eligible Medicaid Family Planning Waiver Evaluation Eligibility for Pregnant Women: Family income cannot exceed 185% of Florida Poverty Level Eligibility for Pregnant Women (Protected Eligibility) If a women becomes financially ineligible, her coverage is protected for the pregnancy 2-months postpartum. Presumptive Eligibility for Pregnant Women Eligibility for Pregnant Women Eligibility for Pregnant Women Continued Number non-medicaid Eligible non-medicaid @ 240 days post delivery Prepared by the Maternal Child Health and Education Research and Data Center at the University of Florida, College of Medicine, a Branch of the Lawton and Rhea Chiles Center for Healthy Mothers and Babies at the University of South Florida 17

NM @ D NS @ 240 NS @ D Post-W Pre-W PW RFPS RFPS < 60 RFPS < 60 > RFPS 60 > RRD S @ 240 S @ D SOBRA non-medicaid @ deliver non-sobra @ 240 days post delivery non-sobra @ delivery 1 st -Year Implementation of Family Planning Waiver Pre-Family Planning Waiver pre-waiver Period calendar year 1999 deliveries Received Family Planning Services Received Family Planning Services Prior to 60 days Received Family Planning Services Prior to and After 60 days Received Family Planning Services After to 60 days Rate of Repeat Deliveries SOBRA Medicaid @ 240 days post delivery SOBRA Medicaid @ Sixth Omnibus Budget Reconciliation Act # Number # RD Number of Repeat Deliveries Prepared by the Maternal Child Health and Education Research and Data Center at the University of Florida, College of Medicine, a Branch of the Lawton and Rhea Chiles Center for Healthy Mothers and Babies at the University of South Florida 18

Section II Pre-Waiver Descriptive Statistics This section presents social demographic statistics for the four contrast groups Medicaid SOBRA, Medicaid non-sobra, Total Medicaid, and non-medicaid Populations in 1995 Prepared by the Maternal Child Health and Education Research and Data Center at the University of Florida, College of Medicine, a Branch of the Lawton and Rhea Chiles Center for Healthy Mothers and Babies at the University of South Florida 19

Pre Waiver 1995 Deliveries Cohort-Summary of Subsequent Statistics SOBRA Versus non-sobra Medicaid Non-SOBRA Medicaid women of varying racial/ethnic backgrounds who delivered in 1995 had greater percentages of subsequent deliveries within two years when compared with SOBRA Medicaid women. The greatest difference in percentage of subsequent deliveries between non-sobra and SOBRA Medicaid women occurred among black mothers (91% non- SOBRA versus 9% SOBRA). Similar percentages of SOBRA and non-sobra Medicaid women who were married (50.4% versus 50%) and delivered in 1995 had a subsequent delivery within two years. However, when comparing single women who delivered in 1995, 14% of those receiving SOBRA Medicaid and 86% receiving non-sobra Medicaid had a subsequent delivery within two years. Non-SOBRA Medicaid women of all ages who delivered in 1995 had greater percentages of subsequent deliveries within two years when compared with SOBRA Medicaid women. Non-SOBRA Medicaid women of varying educational levels who delivered in 1995 had greater percentages of subsequent deliveries within two years when compared with SOBRA Medicaid women. There were greater differences in the percentages of subsequent deliveries of SOBRA and non-sobra Medicaid women at the lower educational levels (81% non-sobra/no High School Education versus 19% SOBRA/High School Education). Regardless of when mothers entered prenatal care, non-sobra Medicaid women who delivered in 1995 had a greater percentage of subsequent deliveries when compared with SOBRA Medicaid women. With regard to Healthy Start screening score, the smallest difference in the percentage of non-sobra and SOBRA Medicaid women who delivered in 1995 and had a subsequent delivery within two years was among those with low Healthy Start screening score (58% SOBRA versus 42% non-sobra). Of those women who experienced complications with labor and delivery, non-sobra Medicaid women who delivered in 1995 had a greater percentage of subsequent deliveries within two years when compared with SOBRA Medicaid women (72% versus 28%). Of women who delivered in 1995 and had a child with a congenital anomaly, a greater percentage of non-sobra Medicaid women had a subsequent delivery when compared to SOBRA Medicaid women (72% versus 29%). Prepared by the Maternal Child Health and Education Research and Data Center, University of Florida, College of Medicine, a Branch of the Lawton and Rhea Chiles Center for Healthy Mothers and Babies at the University of South Florida 20