Children s Medical Service Network Partners in Care : Together for Kids

Transcription

1 Program For All-Inclusive Care For Children Children s Medical Service Network Partners in Care : Together for Kids February Evaluation Year 3 Annual Evaluation for 2008 Institute for CHILD HEALTH POLICY UNIVERSITY OF FLORIDA

2 Prepared By Caprice Knapp, PhD Assistant Research Scientist Epidemiology and Health Policy Research Institute for Child Health Policy University of Florida Vanessa Madden, BSc Research Program Coordinator Institute for Child Health Policy University of Florida Hu Wang, MS Research Program Coordinator Institute for Child Health Policy University of Florida Elizabeth Shenkman, PhD Director, Institute for Child Health Policy Professor Epidemiology and Health Policy Research University of Florida

3 Table of Contents Executive Summary Section I. Enrollment Trends Introduction Enrollment Trends Disenrollment Trends Summary and Recommendations Section II. Parent Satisfaction Survey Introduction Survey Results Summary and Recommendations Section III. Administrator Survey Introduction Survey Results Summary and Recommendations Section IV. Nurse Care Coordinator Survey Introduction Survey Results Summary and Recommendations Section V. Claims and Encounter Data Introduction Cost Patterns Summary and Recommendations Endnotes List of Figures Figure 1. Quarterly New Enrollments for Eight Sites, January 2006 to December Figure 2. Active Enrollment Trends for Eight Sites, January 2006 to December Figure 3. Active Caseload by Site, December Figure 4. Active Caseload By Site, June 2007 to December Figure 5. Disenrollment Trends for Eight Sites, January 2006 to December Figure 6. Ratio of New Enrollees to Disenrollees for Five Sites, by Year Figure 7. How Respondents Learned About the Program Figure 8. Time Frame of Receipt of a PIC:TFK Service After Being Invited to Participate Figure 9. Very Satisfied to Satisfied with PIC:TFK Benefits, By Site Figure 10. Parent Report of Overall Quality of Care, By Site Figure 11. Very Satisfied to Satisfied with PIC:TFK Services Figure 12. Satisfaction with the Frequency of PIC:TFK Services Figure 13. Children s Unmet Demand for PIC:TFK Services Figure 14. Very Satisfied to Satisfied with PIC:TFK Services Received by Family Members Figure 15. Family Members Unmet Demand for PIC:TFK Services Figure 16. Respondents Scores on the Five Domains of the Decisional Conflict Scale

4 Table of Contents Figure 17. Number of Nurse Care Coordinators in Florida, by CMSN Area Office FIgure 18. Average Caseload of CMSN Children, by CMSN Area Office Figure 19. Time Spent Coordinating Care after a PIC:TFK Referral Figure 20. NCC Agreement with The PIC:TFK program operates in my area, by PIC:TFK Program Participation Figure 21. NCC Agreement with A child must be in CMSN to be eligible for the PIC:TFK program, by PIC:TFK Program Participation Figure 22. NCC Agreement with To be eligible for the PIC:TFK program, a child s primary care physician must certify that the child has a life limiting illness, by PIC:TFK Program Participation Figure 23. NCC Responses to Parents of children enrolled in PIC:TFK can receive PIC:TFK services, by PIC:TFK Program Participation Figure 24. NCC Responses to Siblings of children with life-limiting illnesses cannot receive PIC:TFK services, by PIC:TFK Program Participation Figure 25. NCC Responses to Children enrolled in the PIC:TFK program can receive services while hospitalized, by PIC:TFK Program Participation Figure 26. NCC Agreement with I think there is a way to reimburse providers in my area for pediatric palliative care prior to the end stage of illness, by PIC:TFK Program Participation Figure 27. NCC Agreement with I am not familiar with hospice services in this community, by PIC:TFK Program Participation Figure 28. NCC Agreement with I am uncertain of the types of service covered under the hospice benefit, by PIC:TFK Program Participation Figure 29. NCC Agreement with I think that it is difficult to know when, in the course of illness, a child should be referred to palliative care, by PIC:TFK Program Participation Figure 30. NCC Agreement with I do not think it is appropriate to approach families whose children have just been diagnosed with a life-limiting illness to talk about palliative care, by PIC:TFK Program Participation Figure 31. NCC Agreement with I do not think it is appropriate to approach families whose children have a life-limiting illness and are in the mid stage of illness to talk about palliative care, by PIC:TFK Program Participation Figure 32. Percentage of Correct Responses to the Knowledge Quiz, by PIC:TFK Program Participation 46 Figure 33. Percentage of Correct Responses to the PCQN, by PIC:TFK Program Participation

5 List of Tables Table 1. Representativeness of Survey Sample Table 2. Percentage of Parents who Report their Children Have Received Services, By Site Table 3. Percentage of Parents who Report that Family Members Received Services, By Site Table 4. Affirmative Responses to the Inclusion of Additional PIC:TFK Services Table 5. Ranking of Usefulness of PIC:TFK Protocols, By CMSN Site Table 6. Ranking of Usefulness of PIC:TFK Protocols, By Hospice Site Table 7. Disenrollment Criteria- Loss of Eligibility Table 8. Disenrollment Criteria- Reference Guide Table 9. Disenrollment Criteria- Reasons Other than Loss of Eligibility Table 10. Demographic Profile of Respondents, by PIC:TFK Site Participation Table 11. Familiarity with Eligibility Criteria for the PIC:TFK Program Table 12. Familiarity with PIC:TFK Service Restrictions Table 13. Attitudes towards Hospice Care Table 14. Referral to Palliative Care Services, by Diagnosis Table 15. Timing of Referral to Palliative Care Services, by Diagnosis and PIC:TFK Program Participation. 47 Table 16. Cost by Service Type, July 2007 to June Table 17. Average Yearly Costs Pre and Post Implementation, by Service Category Table of Contents

6 Executive Summary Background and Purpose In July 2005, Florida s Partners in Care: Together for Kids (PIC:TFK) program for children with lifelimiting illnesses began 1. Prior to the establishment of the PIC:TFK program, children with life-limiting illnesses received hospice care under the Medicare model. Designed for the elderly, the Medicare model of hospice care requires a patient to be in the last six months of life to receive hospice services and the patient must forego any curative treatments. Because the life span of a child with a life-limiting illness is diffi cult to predict and the specifi c factors associated with childhood illnesses may require treatment up to the time of death, the Medicare model of hospice care is inappropriate for a pediatric population. PIC:TFK is the fi rst publicly-fi nanced health program for children in the nation to utilize a pediatric palliative care model which integrates palliative with curative or life-prolonging therapies. PIC:TFK operates under the authority of the Children s Medical Services Network (CMSN). The CMSN is Florida s Title V program for Children with Special Health Care Needs (CSHCN). Children must be clinically and fi nancially eligible for the CMSN to enroll in PIC:TFK. The PIC:TFK program is based on the Children s Hospice International Program for All- Inclusive Care for Children and their Families model of pediatric palliative care, which strives to provide a continuum of care for children and families from the time that a child is diagnosed with a lifethreatening condition, with hope for a cure, through the bereavement process if cure is not attained 2. As such, children enrolled in the program represent children at all stages of their disease trajectories. The PIC:TFK program was authorized to operate in eight sites 3 across the state: Gainesville, Ft. Myers, Jacksonville, Miami, Pensacola, Sarasota, St. Petersburg, and West Palm Beach. As of January 2009, the program operates in six sites: Gainesville, Ft. Myers, Jacksonville, Pensacola, Sarasota, and St. Petersburg 4. The following are included as reimbursable PIC:TFK program services: pain and symptom consultation, specialized nursing, specialized personal care, respite care, support counseling for children and their families, and expressive therapies for children and their siblings. Bereavement counseling is also provided, but is a non reimbursable service. The federal Centers for Medicare and Medicaid Services (CMS) approved the PIC:TFK program as a 1915(b) Medicaid waiver program in As a result, the State was charged under federal guidelines with conducting an Independent Assessment (IA) of the PIC:TFK program. The Institute for Child Health Policy at the University of Florida (the Institute) is conducting the assessment. The purposes of this report are to: Describe the characteristics of children enrolled in the program from January 2006 to December 2008, Describe the results from the parent satisfaction and quality of health care survey, At a Glance PIC:TFK is the fi rst publicly fi nanced pediatric palliative care program in the U.S. to provide supportive care for children and their families from the point of diagnosis onward. Describe the fi ndings from the CMSN and hospice administrative surveys and site visits, Describe the results from a pediatric palliative care survey with CMSN Nurse Care Coordinators (NCCs), and Describe the PIC:TFK cost patterns. Data Sources Five data sources are used in this report. The aims of this report are presented by section as described below. Section I- Enrollment Trends This section of the report describes the characteristics of children enrolled in the program from January 2006 to December Program enrollments and disenrollments were tracked over pg. 6

7 time using monthly enrollment data submitted by each PIC:TFK site to the Institute. Regional differences in enrollment patterns are explored. Section II- Parent Satisfaction Survey This section of the report describes the results from a parent satisfaction and quality of health care survey conducted over the telephone. Survey data were collected for parents whose children are enrolled in PIC:TFK and for parents whose children disenrolled from the program in the three months prior. Parents of recent enrollees are asked about their experiences with the enrollment procedures, and parents of disenrollees are asked for their reasons for leaving the program. Section III- Administrator Survey This section of the report describes the fi ndings from administrative surveys with CMSN and hospice staff. Responses were collected by , and then discussed during scheduled site visits. The main focus of the site visits is to establish the reasons why children are disenrolling from the program, to establish standardized disenrollment categories, and to train the CMSN nurses about submitting the monthly enrollment data in a new electronic database. Section IV- Nurse Care Coordinator Survey This section of the report describes the results from a pediatric palliative care survey with CMSN NCCs from both PIC:TFK sites and non PIC:TFK sites. CMSN Regional Nursing Directors distributed surveys to the NCCs who completed either a hardcopy or online survey. Survey questions test the NCCs knowledge of the PIC:TFK program, and their general attitudes towards, and knowledge of, palliative care services. The survey aims to determine any gaps in the NCCs knowledge base, which will be useful information for developing or refi ning pediatric palliative care training modules in the future. Section V- Claims and Encounter Data This section of the report describes the PIC:TFK cost patterns for Medicaid enrollees, and the issues associated with calculating reliable cost estimates for this population. Claims and encounter data were analyzed for state fi scal year Summary and Recommendations The PIC:TFK program is a complex and novel program in its early phases of operation. Much has been accomplished as evidenced by the establishment of sites with multiple program partners, enrollment of over 600 children, standardization of program policies and procedures, and 89% of parental reports of satisfaction with the program. However, due to the newness and the complexity of the program, there are several areas necessary for improvement and the ongoing success of this waiver program. Some of the areas necessary for improvement are within the control of the PIC:TFK Program Director, however some areas require system level changes. In recognition of this fact, tiered recommendations are in the Executive Summary whereby Tier 1 recommendations refl ect those that are systemic and Tier 2 represents those where there is more direct control to implement changes. At A Glance Five data sources are used in this report: Enrollment data Parental satisfaction survey Administrator survey Nurse Care Coordinator survey Claims and encounter data pg. 7

8 A summary of the key fi ndings and recommendations for this study are presented below by section. A complete list of fi ndings and recommendations can be found in the corresponding sections of the report. Section I- Enrollment Trends This section of the report examines trends in program enrollments and disenrollments to identify areas where recruitment and retention efforts need strengthening. Regional differences in enrollment trends are explored, with the goal of identifying those sites with the lowest and the highest enrollment numbers. The key fi ndings are as follows: The composition of the PIC:TFK program changed markedly during 2008, with two hospices exiting the program and one hospice rejoining the program. Overall, there are one-third fewer new enrollments in 2008 as compared with Adjusting for those sites who no longer participate in the program, the ratio of new enrollees to disenrollees in 2008 is 1.2:1. Active enrollees fill approximately one-quarter of the 940 program slots in December The largest active caseload is in Ft. Myers, followed by Jacksonville and St. Petersburg. Across sites, the active caseload remains relatively constant over a one-year period. Between December 2007 and December 2008, four sites report an pg. 8 increase in their active caseload of between two to nine children and one site reports that their active caseload has decreased by nine children. The Institute makes the following Tier 2 recommendations (no Tier 1 recommendations are made for this section): CMSN should investigate differences in capacity across sites. For example, one site serves 88 children while the active caseload at the other sites is between 20 and 57 children which could be due to capacity constraints. As the program expands to include other sites in Florida, CMSN headquarters should consider developing an online enrollment database which interfaces with the Child Assessment and Plan System (CAP) and other existing CMSN databases. By developing and maintaining a standardized database, CMSN staff will have access to the latest enrollment information and IT support from CMSN headquarters, and the PIC:TFK Program Director will have direct access to the enrollment information. Section II- Parent Satisfaction Survey This section of the report describes parental satisfaction with the frequency and quality of PIC:TFK services received by themselves, their children, and by other family members. The survey asked parents about their satisfaction with the scope of the PIC:TFK services, or if they have unmet needs. Parents of children recently joining the program are asked about their satisfaction with enrollment procedures, and parents of children recently leaving the program are asked the reason for leaving the program. The key fi ndings are as follows: Approximately two-thirds of parents of new enrollees indicate that they fi rst learned about the program from their children s NCC and 35% do not recall receiving a list or being told about PIC:TFK services. Of the parents who did not receive information, one-half are located at one PIC:TFK site. The majority (89%) of parents are very satisfi ed to satisfied with the program benefi ts, and the overall satisfaction rating has increased 6% from Seventy percent of parents report that the quality of care that their children receive in PIC:TFK is excellent or very good. The quality of care rating has increased 5% compared with the previous reporting period. Ninety-three percent to100% of parents are satisfi ed with the different services received by their children (support counseling, personal care, nursing care, music therapy, art therapy, and play therapy). Eighty-four percent to 94% of parents are satisfi ed with the frequency of each service. Ninety-four percent to 100% of parents are satisfi ed with the services received by themselves or their other family members (support counseling, respite,

9 music therapy, art therapy, play therapy, and volunteer services). Thirty-one percent of respondents report that no one other than their ill child received any PIC:TFK services. Not all families who would like to receive certain services are receiving them. For example, 36% of parents would like for their children to start receiving support counseling and 29% of parents would like for themselves or another family member to start receiving support counseling. The greatest unmet need for children is for art, music and play therapies. The greatest unmet need for family members is art therapy for siblings, followed by in-home respite services. One-quarter of parents report that they would like to see additional services included in the PIC:TFK benefi ts package. The greatest reported demand is for the continuation of PIC:TFK services to the parent or family members while the ill child is hospitalized. The Institute makes the following Tier 1 recommendations: The PIC:TFK Steering Committee should investigate options for increasing the response rate among parents of Black children. The PIC:TFK Steering Committee should consider investigating the reasons why some families who would like to receive specifi c PIC:TFK services are currently not receiving those services. For example, the greatest unmet need is for expressive therapies, however this is likely due to the lack of registered expressive therapists in Florida. The Institute makes the following Tier 2 recommendations: The Institute recommends that CMSN headquarters consider developing an online enrollment database which interfaces with CAP and other existing CMSN databases. The Institute also recommends that hospice staff have restricted access to this online database and the ability to update contact information in real time. Giving hospice staff access to the enrollment database will also aid joint care planning efforts. Parents of all disenrollees should be sent a disenrollment letter as a standard practice and this letter should be printed in English and Spanish. This will ensure that all parents are aware of the reason for their children s disenrollment and will give parents an opportunity to contact the hospice for further information. The Institute recommends that survey questions should be added to determine why parents have unmet needs for services that are offered through the PIC:TFK program. These reasons could be that the service was not a part of their care plan, they were not told about the service, they used the maximum number of allowable service units, etc. It is important to determine why unmet needs exist. Section III- Administrator Survey Administrative surveys and site visits are important components of the program evaluation, as they give valuable insights into whether the program is operating as intended. Findings from previous administrative surveys have described the sites operational practices, standards of care, and procedures, as well as issues arising during the implementation phase of the PIC:TFK program 5. The main focus of the site visits, and this section of the report, is to investigate the reasons why children disenroll from the program. The key fi ndings from the administrative surveys and site visits are as follows: The sites report operational issues with regards to reimbursement and staffi ng. Many of the sites report that the reimbursement rates and the funding allocated to administrative costs are too low. The CMSN sites report that the standardized forms are the most useful protocol. The statewide meeting ranks highly for both CMSN and hospice staff, and the technical assistance calls are viewed universally as the least useful. One-half of children leave the program because they no longer meet the eligibility criteria. The most common reason for disenrollment is relocation (17%), followed by the loss of Medicaid coverage or CMSN coverage (16%), death (12%), relocation to full hospice or a long term care facility (4%) and non recertifi cation by the PCP (1%). pg. 9

10 One-half of children disenroll from the program for reasons other than loss of eligibility. The most reported reason for program disenrollment is the families lack of interest, or non responsiveness (21%), followed by families not using services (10%) or the hospice determining that services are no longer required (10%). It is important to note that CMSN staff report that only one family is dissatisfi ed with PIC:TFK services. The Institute makes the following Tier 1 recommendation: The Institute recommends that the PIC:TFK Steering Committee internally review reimbursement fees and hospice staffi ng issues. However, setting reimbursement rates is beyond the authority of the PIC:TFK Steering Committee. The Institute makes the following Tier 2 recommendations: A special study to investigate the indirect costs associated with providing services to PIC:TFK participants should be considered as many hospices continue to question the adequacy of reimbursement. The study could enumerate the administrative costs of providing services (travel costs, training, care planning, care coordination, staffi ng) which would also be helpful for establishing staffi ng patterns and capacity limits for hospices. The Institute recommends that the PIC:TFK Program Director pg. 10 shares the disenrollment fi ndings with all new sites entering the PIC:TFK program. It is important for hospices to be aware that one-quarter of families entering the program are either not contactable or are unresponsive. Hospices can then factor this indirect cost information into their break-even cost estimates and adjust staffi ng rosters accordingly. Section IV- Nurse Care Coordinator Survey The primary persons responsible for initiating referrals to the PIC:TFK program are the CMSN NCCs. As such, it is important to determine the NCCs knowledge of, and attitudes towards, pediatric palliative care. This section describes the fi rst palliative care survey undertaken with CMSN NCCs in Florida. The survey compares responses from the PIC:TFK sites and non PIC:TFK sites. The key fi ndings are as follows: Fifteen percent of respondents were previously employed by hospice and 14% report ever completing a formal palliative care training course, such as the End-of-Life Nursing Education Consortium (ELNEC) course. The average caseload for a NCC is 181 children, however the average caseload varies signifi cantly across the state. PIC:TFK respondents are signifi cantly more likely than their counterparts at non PIC:TFK sites to know whether the PIC:TFK program operates in their area, to correctly identify that CMSN enrollment and PCP certifi cation are required for admission to the program, and that both siblings and parents may also receive PIC:TFK services. One in twenty NCCs can correctly identify that enrolled children are not eligible to continue receiving PIC:TFK services while hospitalized, as this would constitute a duplication of Medicaid reimbursable services. NCCs correctly answer 63% of questions on a knowledge quiz about hospice policies and the mean percentage of correct answers on the Palliative Care Quiz for Nurses is 54%. Respondents from the PIC:TFK sites are more likely than their counterparts at non PIC:TFK sites to refer children to palliative care services before the end of life for 12 diagnoses eligible for the program. The Institute makes the following Tier 2 recommendations (no Tier 1 recommendations are made for this section): Although the PIC:TFK NCCs scored higher than the non PIC:TFK NCCs, there were several gaps in knowledge that could be addressed in future training such as the fact that there is no allowable reimbursement when children are hospitalized. Very few NCCs have completed a formal palliative care training program and the PIC:TFK Steering Committee should consider if this requirement, as

11 described in the implementation guidelines, is needed given the differences in scores between the PIC:TFK and non PIC:TFK sites. Once new sites are approved and begin enrolling children, it is recommended that the NCCs in those sites complete the same survey so that a pre, post analysis can be performed. Section V- Claims and Encounter Data The PIC:TFK program is the fi rst publicly-funded program to provide supportive care for children and families from the point of diagnosis onwards. Costs analyses are important to establish fi scal sustainability and impact. The key fi ndings are as follows: Support counseling is the most used service, which corroborates results from the parent survey, and inpatient respite is used the least. The results show that average yearly inpatient costs fell after enrollment in PIC:TFK and emergency department, pharmacy, and outpatient costs increased. However, the cumulative effect was a decrease in total costs of about $10, The Institute makes the following Tier 2 recommendations (no Tier 1 recommendations are made for this section): The Institute recommends that the reimbursement regulations associated with inpatient respite be clarifi ed by AHCA. Utilization of this service may increase if the sites were given more defi nitive information. Costs should continue to be monitored on a quarterly basis by the PIC:TFK Program Director, AHCA, and CMSN. Costs for children eligible for Title XXI should be collected and analyzed. At a Glance Two hospices exited and one hospice rejoined the PIC:TFK program in The six sites that provide PIC:TFK services in January 2009 are: Gainesville Ft. Myers Jacksonville Pensacola Sarasota St. Petersburg pg. 11

12 Section 1. Enrollment Trends pg. 12

13 Introduction An important aspect of program evaluation is the ability to track program enrollment over time. Furthermore, examining regional differences in enrollment trends allows program planners to offer additional support to sites with low enrollment numbers and to look for best practices at the sites with the highest enrollment numbers. It is equally important to determine the reasons that children disenroll from the program. Establishing the specifi c reasons for disenrollment gives program planners clues about gaps in care, whether these may be related to local or statewide issues, and to develop plans to address disenrollments if necessary. Background Children must meet fi nancial and medical screening criteria to be eligible for enrollment in CMSN and PIC:TFK. A further clinical eligibility requirement for PIC:TFK mandates that a child be diagnosed with a potentially life-limiting condition. Children must also meet the CMSN fi nancial eligibility requirements associated with Title XIX Medicaid (for children under 21) or Title XXI Florida KidCare (for children under 19). The enrollment of Safety-Net eligible children in PIC:TFK (whose families have higher incomes than Title XIX or Title XXI families) is optional and based on funding availability. As of January 2009, eligible children receive PIC:TFK services at six sites across the state: Gainesville, Ft. Myers, Jacksonville, Pensacola, Sarasota, and St. Petersburg. At each site, CMSN staff collaborates with a local hospice organization with demonstrated expertise in caring for children. Participating local hospices are: Hospice of the Nature Coast, Hope Hospice and Community Care Service, Community Hospice of Northeast Florida, Covenant Hospice, Tidewell Hospice, and Hospice of the Florida Suncoast. The PIC:TFK program is authorized, but is not currently operating, in two additional sites 7 : Miami and West Palm Beach. CMSN NCCs are charged with identifying children that might be potentially eligible for PIC:TFK. The primary care physician for a potentially eligible child is contacted for a referral to the PIC:TFK program, and if given, the family is then approached by the CMSN NCC about enrollment. Once a family agrees to enroll, hospice staff performs an initial intake assessment with the family and a plan of care is developed. Parents and the primary care physician also receive a copy of the care plan. Hospice and CMSN staff work together to ensure that the care plan is appropriate and is being followed. After a plan of care is agreed upon, the hospice provides the palliative care services as agreed upon in the care plan. Data Source Since November 2005, the sites have collected the following key data elements on enrollees: child s name, age, race, gender, funding source (Title XIX, Title XXI, or Safety-Net), contact information, diagnoses, stage of illness (newly diagnosed, mid stage, end stage), admission date, disenrollment date, and reason for disenrollment. Initially this information was submitted on a monthly basis to the PIC:TFK Program Director, however the Institute subsumed the role of processing the monthly enrollment fi le during Data was initially collected using an EXCEL template, which included data on both enrollees and those children who were potentially eligible for PIC:TFK but not enrolled 8. The EXCEL template was problematic for two reasons. First, the sites reported that the template was not user-friendly. Second, the Institute observed that, despite training, there was a substantial amount of missing data, incomplete data, and incorrectly coded data 9. In June 2008, the Technical Advisory Panel advised that a new system be developed to capture and report the enrollment data. In response, the Institute developed an ACCESS database in consultation with all the PIC:TFK sites 10. All sites received face-toface training before the database became operational in September pg. 13

14 The advantages of the ACCESS database are: 1. Software availability. The ACCESS program is part of the standard Microsoft Offi ce package, which is available to all CMSN NCCs. 2. User-friendly interface. Each child s details are displayed on a single page and data is organized into colorcoded sections. Only essential information is recorded. 3. Quick results. At the touch of a button, users can search records or print lists of children in each stage of the enrollment process. 4. Quality control. The database is designed to prevent simple data entry errors. For example, users receive an error message if they type an incomplete telephone number. 5. Simpler coding. Users may select codes from drop-down lists that standardize the collected information and prevent coding errors. pg. 14

15 Enrollment Trends Data collected in the ACCESS database are used in this section to describe enrollment trends. Specifi cally, three measures are used to assess the program s growth and sustainability: Figure 1. Quarterly New Enrollments for Eight Sites, January 2006 to December 2008 Number of new children entering the program each month (new enrollees), New Enrollments Number of children receiving PIC:TFK services each month (active enrollees), and Number of children leaving the program each month (disenrollees). To be considered PIC:TFK enrollees, children must fi rst receive an initial assessment from hospice. Children are not counted as PIC:TFK enrollees if an initial appointment was scheduled but not completed. All enrollment fi gures presented below represent children who received an initial assessment from hospice Q Q Q Q Q approximately one-third fewer new enrollees in 2008 as compared with Active Enrollments Cumulative enrollment in the PIC:TFK program is 615 children. Of those 615 children who ever received PIC:TFK services, 42%, or Q Quarter Q Q Q Q Q Q out of 615, remain enrolled as of December Figure 2 shows both the total caseload of children served by the sites each month (cumulative active enrollments denoted by a line) and the new enrollments each month (monthly enrollments denoted by blue bars). The total active caseload increased by 155 children during calendar year 2007, but decreased by 41 children Figure 2. Active Enrollment Trends for Eight Sites, January 2006 to December 2008 New Enrollments Figure 1 shows a steady increase in the number of new enrollees during calendar years 2006 and 2007, reaching a peak of 86 new enrollments during the third quarter of The second largest peaks in enrollment are seen in the second quarter of 2007 and the third quarter of During calendar years 2006, 2007, and 2008, the program saw 124, 256, and 161 new enrollments, respectively. There are Active Enrollments Monthly Enrollments Cumulative Enrollments Potential List Implemented Gainesville left program Sarasota joined program Miami left program Gainesville rejoined program Potential List Stopped West Palm Beach left program J F M A M J J A S O N D J F M A M J J A S O N D J F M A M J J A S O N D Month pg. 15

16 during calendar year During 2008 both the Miami program and the West Palm Beach program stopped providing PIC:TFK services. In December 2008, the majority of active enrollees are in midstage (84%), while 12% are newly diagnosed 12 and 4% are end stage. Ninety-three percent of active enrollees are Title XIX eligible, 3% Title XXI, and 4% are eligible through Safety-Net. Fifty-four percent of enrollees are male. Seventy percent of enrollees are White, 25% are Black, and 5% are Hispanic 13. Children are categorized by their primary diagnosis and the most common diagnoses are: congenital anomaly, brain injury, cystic fi brosis, respiratory disorder, cerebral palsy, cardiovascular disorder, convulsions, developmental delay, muscular dystrophies, immunodefi ciency, and neoplasm. More than 100 different primary diagnoses are represented in the PIC:TFK program, with many children having multiple diagnoses. Of the active enrollees, 3% had disenrolled from the program at some point previously, and then reenrolled. Most of the re-enrollees were subsequently disenrolled from the program because they were not using services, or were not contactable. The average time that these children were disenrolled is 243 days. Active Enrollees Figure 3. Active Caseload by Site, December 2008 Figure 4. Active Caseload By Site, June 2007 to December Ft. Myers Jacksonville Pensacola Sarasota St. Petersburg West Palm Beach June 2007 December 2007 June 2008 December Figure 3 shows a snapshot of the active caseload for each site in December The Ft. Myers site has the largest active caseload, followed by Jacksonville and St. Petersburg. Active enrollees fi ll approximately one-quarter of the 940 funded program slots as of December Looking at enrollment trends over time, Figure 4 shows that Ft. Myers has almost doubled their caseload between June 2007 and December 2007, and has sustained the largest active caseload between December 2007 and December pg. 16

17 Disenrollment Trends Data collected in the ACCESS database are used in this section to describe enrollment trends. Since the PIC:TFK program began in 2005, 356 children have been disenrolled from the program. Three sites left the program, directly accounting for 24% of the disenrollments. The reasons for the other three-quarters of disenrollments will be discussed in further detail in Section Three. Figure 5 shows the number of children who left the program each month (denoted by blue bars) and the cumulative number of disenrollments (denoted by a line) for the eight sites. Cumulative disenrollments have increased steadily, increasing from 67 disenrollees by January 2007 to 356 disenrollees by December In 2008, there are 244 new disenrollments. Two sites left the program during 2008, directly accounting for 27% of the new disenrollments during Disenrollments Figure 5. Disenrollment Trends for Eight Sites, January 2006 to December 2008 Monthly Disenrollments Cumulative Disenrollments Gainesville left program J F M A M J J A S O N D J F M A M J J A S O N D J F M A M J J A S O N D Month Miami left program West Palm Beach left program Figure 6 shows the ratio of new enrollees to disenrollees for fi ve sites: Ft. Myers, Jacksonville, Pensacola, Sarasota, and St. Petersburg. The fi gure does not include three sites that ceased to provide PIC:TFK services during 2006 to In calendar year 2006, equal numbers of children joined and left the program. During calendar years 2007 and 2008 there are more new enrollees joining the program than there are children disenrolling from the program; however, the ratio of new enrollees to disenrollees decreases markedly from calendar year 2007 to calendar year Figure 6. Ratio of New Enrollees to Disenrollees for Five Sites, by Year Ratio of New Enrolleess to Disenrollees pg. 17

18 Summary and Recommendations This section of the report examines trends in program enrollments and disenrollments to identify areas where recruitment and retention efforts need strengthening. Regional differences in enrollment trends are explored, with the goal of identifying those sites with the lowest and the highest enrollment numbers. The key fi ndings are as follows: The composition of the PIC:TFK program changed markedly during 2008, with two hospices (Miami and West Palm Beach) exiting the program and one hospice (Gainesville) rejoining the program. Overall, there are one-third fewer new enrollments in 2008 as compared with Adjusting for the sites that no longer serve children, the ratio of new enrollees to disenrollees is 1.2:1 in Active enrollees fi ll approximately one-quarter of the 940 program slots as of December The largest active caseload is in Ft. Myers, followed by Jacksonville and St. Petersburg. Across sites, the active caseload remains relatively constant over a one-year period. Between December 2007 and December 2008, four sites report an increase in their active caseload of between two to nine children and one site reports that their active caseload has decreased by nine children. The ability of the sites to recruit and retain participants is crucial. The active caseload of children has remained constant at many of the sites, indicating that the hospices may not have the capacity to serve more than a certain number of children under the current model of reimbursement. In light of the fi ndings, the Institute recommends the following: CMSN should continue to address site specifi c barriers. Also, CMSN should determine what each hospice s capacity is to serve children in the program. As the program expands to include other sites in Florida, CMSN headquarters should consider developing an online enrollment database which interfaces with the Child Assessment and Plan System (CAP) and other existing CMSN databases. By developing and maintaining a standardized database, CMSN staff will have access to the latest enrollment information and IT support from CMSN headquarters, and the PIC:TFK Program Director will have direct access to the enrollment information. At a Glance PIC:TFK has provided supportive care services to 615 children and their families since the program s inception. 42%, or 259 out of 615, remain enrolled as of December One-third of the 259 enrollees are located at the Ft. Myers site. pg. 18

19 Section II. Parent Satisfaction Survey pg. 19

20 Introduction The PIC:TFK program aims to deliver all inclusive care to children and their families, by offering supportive care services to the entire family unit. As such, parental satisfaction with the frequency and quality of PIC:TFK services is an important indicator of program quality. An equally important goal is to establish whether families are satisfi ed with the scope of services, or if they have unmet needs. The National Survey of Children with Special Health Care Needs (CSHCN) shows that 22% of CSHCN in Florida have an unmet need for a specifi c health care service 14. Furthermore, only half of the parents of CSHCN in Florida report that they partner in decision making at all levels and are satisfi ed with the services that they receive. As the primary caregiver of a child with a life-limiting illness, parents may experience decisional confl ict, making partnership with their health care providers an important aspect of care. Due to the importance of these issues the Institute added modules to the parental survey to assess: 1) Enrollment Procedures, 2) Overall Satisfaction, 3) Satisfaction with PIC:TFK Services, 4) Unmet Needs, 5) Decisional Confl ict, 6) Disenrollment Reasons, and 7) Demographics. Data Sources Parental Satisfaction Survey Parental satisfaction data were collected by a telephone survey. Survey data were collected for three groups of parents: 1) parents whose children had been enrolled in PIC:TFK less than three months, 2) parents whose children had been enrolled in PIC:TFK more than three months, and 3) parents whose pg. 20 children disenrolled from PIC:TFK during the previous three months. The Institute developed and tested survey instruments to capture family satisfaction within the program in During the contract year, the Institute conducted the fi rst evaluation of the PIC:TFK program. Survey instruments were further refi ned, and several new evaluation components were added for year two. Each year the Institute receives advice from its Technical Advisory Panel (TAP) about how to best refi ne the parental satisfaction survey. In June 2008 the consensus of the TAP was to shorten the parental satisfaction survey and ask more targeted questions about unmet needs and satisfaction with individual PIC:TFK services. This information is valuable for planning purposes, both at the state level and the individual site level. Previous validation efforts have shown the PedsQL to be an inadequate survey instrument for children with life-limiting illnesses. In the absence of an appropriate quality of life instrument for this patient population, quality of life questions were not asked in this survey. The Institute continues to work with national experts to try and identify an appropriate tool to measure quality of life. The parent survey contains the following modules. Enrollment Procedures: This module is administered to parents of children enrolled in the program for less than three months. Parents are asked how they fi rst learned about the program, how they were informed of the available services, and how quickly their family started receiving services. At a Glance 140 parents completed the survey. The parent survey includes the following modules: Enrollment procedures Satisfaction with care Satisfaction with PIC:TFK services Unmet needs Decisional confl ict Disenrollment reasons Overall Satisfaction: Parents are asked whether they would recommend the PIC:TFK program, whether they had ever fi led a complaint, and their general satisfaction with the program and their children s overall quality of care. PIC:TFK Services: Parents are asked to rate their satisfaction with each PIC:TFK service received

21 by themselves, their children, or by another family member. For each service, parents are asked whether they were satisfi ed with the frequency of visits, and if not, whether they would prefer more or less frequent visits. If no one in the family received a particular service, parents are asked whether they would like to start receiving that service. Unmet Needs: Parents are asked whether they would like to see additional services included in the PIC:TFK program. Parents are fi rst asked if they would like genetic counseling, as this was an identifi ed unmet need in the National Survey of CSHCN 15. Two further questions center on continuity of care during, and after, their children s hospitalization, including: 1) the continuation of PIC:TFK services to family members during hospitalization, and 2) services to help parents care for their child at home after discharge from the hospital. Parents are also asked if they would like short-term daycare for their children in a location other than their home and whether they would like to participate in community groups for children with similar diagnoses. Decisional Conflict Scale (DCS): The DCS measures parents perceptions of uncertainty in choosing treatment options for their children, and is the preferred instrument for measuring decisional confl ict in this population 16. The DCS consists of 16 items which measure fi ve domains relevant to decision making: informed feeling (3 items), values clarity (3 items), support (3 items), uncertainty (3 items), and effective decision making (4 items). The response category for each item includes: strongly disagree, disagree, neither agree nor disagree, agree, and strongly agree. The domain scores are calculated by averaging scores of items in a specifi c domain and then transforming scores to a 100 point scale. A score of 0 indicates low decisional confl ict and a score of 100 indicates high decisional confl ict. A score below 25 is associated with implementing decisions; scores exceeding 37.5 are associated with decision delay or feeling unsure about making a decision 17. A previous study conducted by the Institute confi rmed that the DCS is a valid and reliable tool to use with PIC:TFK parents 18. Disenrollments: This module is administered to parents of children who were disenrolled from the PIC:TFK program in the three months prior to the survey. Parents are asked to choose from one of 17 possible reasons for program disenrollment. Administrative Data Administrative data is used to compare the demographic characteristics of children whose parents responded to the survey (responders) and those who did not respond (non responders). The following children s characteristics were compared to determine the representativeness of the sample: age, gender, race, insurance type, stage of illness, and enrollment status. Claims and encounter data were used to categorize children into one of six Clinical Risk Groups (healthy, signifi cant acute, minor chronic condition, moderate chronic condition, major chronic condition, unassigned). Analysis Multivariate analyses were conducted using STATA version Multivariate regressions were conducted to estimate the relationships between quality of care and satisfaction on factors known to infl uence parent-reported health care experiences, such as race/ethnicity, parental education, etc. Each regression uses a logistic functional form where the dependent variable takes on a zero or one value. Satisfaction scores are transformed into a binary variable by collapsing into two categories (1, 0). If a parent s satisfaction score is very satisfi ed or satisfi ed, it is assigned a value of one, and zero otherwise. Quality of care scores are transformed into a binary variable by collapsing into two categories (1, 0). If a parent s reported quality of care is Excellent to Very Good, it is assigned a value of one, and zero otherwise. The regressions control for child s race/ ethnicity (White, Black, Hispanic, and Other), parental educational level (less than high school, high school, some college, and college graduate), child s gender, child s age, child s stage of illness (early, mid, late), parent s marital status (married or other), household type (single versus two parent), and dummy variables for the PIC:TFK sites. A summary of the logistic regression results is contained in this section of the report. Complete regression results are contained in the Technical Appendix. pg. 21

22 Survey Results Three hundred and fi fty-four parents were eligible to participate in the telephone survey. In total, 140 parents completed the survey. Compared with the reporting period, there is a 65% increase in the number of completed surveys. The overall response rate is 63% of those parents with valid contact information. More than one-third of parents did not have valid contact information. Profile of Responders and Non Responders Table 1 compares the demographic characteristics of respondents children with the characteristics of non respondents children. Bivariate analyses were conducted to determine if there were signifi cant differences between the responders and non responders. Corresponding p-values are presented in Table 1 and p-values of 0.05 or less denote statistically signifi cant differences. Three factors were found to be signifi cantly different between the responders and non responders: children s race/ethnicity, age, and stage of illness. Parents of children who are Black are less likely to respond to the survey and White parents are more likely to respond. Respondents children are older than the children of non respondents. Parents of newly diagnosed children and children at the end of life are more likely to participate in the survey than parents of children in the mid stage of their illness trajectory. Demographics of Parent Respondents Of the 140 parent respondents whose children are enrolled in PIC:TFK, 91% of parents are female. The average parental age is 41.7 years with a standard deviation of 11.0 years. Fifty-fi ve percent of parents are White, 13% Black, and 24% Hispanic, and 8% other races. One-quarter of PIC:TFK parents have a high school diploma and 14% have not graduated from high school. Forty-seven percent are married, 26% are single or common law married, 22% are divorced or separated, and 5% are widowed. Fifty-six percent of respondents describe their household as a twoparent household. The majority of PIC:TFK parents (79%) speak English at home. Table 1 shows that the majority of respondents report that their child has participated in the program for more than three months (76%). Eleven percent of surveyed parents report that their child disenrolled from the program in the past three months while 13% note that their child has been enrolled for less than three months. Enrollment Procedures Questions in this section are restricted to the parents of new enrollees, meaning those whose children have been enrolled for three months or less, to reduce recall bias. As a result, 18 parents Table 1. Representativeness of Survey Sample Responders Non Responders P Value Child s Enrollment Status New Enrollees 18 13% 24 11% Active Enrollees % % P=0.90 Disenrollees 16 11% 25 12% Child s Gender Male 72 51% % P=0.72 Female 68 49% % Child s Clinical Risk Group Healthy 3 2% 2 1% Significant Acute 1 1% 3 1% Minor Chronic Condition 1 1% 2 1% P=0.55 Moderate Chronic Condition 29 22% 46 23% Major Chronic Condition 93 69% % Unassigned 7 5% 4 2% Child s Age* Mean ( ± S.D.) 9.9 years (± 5.8) 8.7 years (± 5.4) P=0.048 Child s Race/Ethnicity* Black % White 87 77% 98 56% P=0.003 Hispanic 3 3% 13 7% Child s Insurance* Medicaid % % Title XXI 7 5% 5 2% P=0.23 Safety Net 6 4% 5 2% Child s Stage of Illness* Less than six months after diagnosis 18 13% 15 7% Mid Stage % % P=0.01 End of Life 10 7% 5 2% PIC:TFK Site Site % 31 15% Site 2 6 4% 11 5% P=0.69 Site % 63 29% Site % 29 14% Site % 37 17% Site % 33 15% Site 7 9 6% 10 5% *Demographic information is incomplete. S.D.= Standard Deviation pg. 22

23 Percentage Percentage Percentage 100% 80% 60% 0% 100% 80% 60% Figure 7. How Respondents Learned About the Program 0% 100% 80% 60% 0% 68% Nurse Care Coordinator 39% are asked these questions. Sixtyeight percent of these parents indicate that they fi rst learned about the program from their children s NCC (Figure 7). Respondents often learned about the program from more than one source. Next, respondents are asked if they were given a list or were told about the PIC:TFK services available for their children. Thirty-fi ve percent 17% 17% 17% Provider Brochure Friend or Family Member Social Service Agency 44% Hospital Figure 8. Time Frame of Receipt of a PIC:TFK Service After Being Invited to Participate 44% 50% 11% Internet Less than 2 weeks 2-4 weeks More than 1 month 89% Figure 9. Very Satisfied to Satisfied with PIC:TFK Benefits, By Site 84% 88% 83% of parents do not recall receiving a list or being told about PIC:TFK services. Of the parents who did not receive information, one-half are located at one PIC:TFK site. Figure 8 shows the amount of time from when the child was invited to participate in PIC:TFK to when the child received services. Fortyfour percent of the parents report having received a service within two weeks. 6% 93% 96% Overall Site 1 Site 3 Site 4 Site 5 Site 6 Satisfaction with Benefits It is important to note that sometimes the timing of service provision may be at the request of the parents. Parents may have scheduling issues that prevent them from having services provided in a short time period. The percentage of parents who report that this time period was due to personal choice was 38%. Overall Satisfaction This survey module asks specifi c questions about the respondents overall satisfaction with the PIC:TFK program. All parents answered these questions. The goal of this series of questions is to identify strengths and weaknesses in the program so that improvements can be made and best practices acknowledged. However, to do so, programmatic and site level feedback is needed. In this report, results are presented by program and site. A site key has been provided to CMSN and AHCA headquarters, but is not provided in this report due to privacy concerns. Table 1 shows the number of survey respondents by site. It should be noted that the stratifi ed results should be interpreted with caution since the sample sizes across the sites are small. Furthermore, stratifi ed responses for Sites 2 and 7 will not be presented in this report due to the low number of survey respondents. However, any overall percentages reported include sites 2 and 7. Parents are asked about their satisfaction with the PIC:TFK program benefi ts. Figure 9 shows that 89% of parents are very satisfi ed to satisfi ed with the program benefi ts. The overall satisfaction rating has increased 6% from pg. 23

24 Ninety-four percent of respondents overall would recommend the program to a family member or friend. At three of the six sites, 100% of parents indicate that they would recommend the program to a family member or friend. Eight PIC:TFK parents made a complaint with CMSN, although there were no more than two complaints at any one site. When asked about the quality of care their child receives in PIC:TFK, 70% overall indicate that the quality Percentage 100% 80% 60% 0% 70% 68% 68% 23% 7% 16% 16% is excellent or very good (Figure 10). This quality of care rating has increased 5% from Multivariate Regression Results As described previously, two logistic regressions were performed to estimate the relationships between quality of care and satisfaction and a variety of individual predictor variables. 12% Excellent or Very Good quality of care. Children in the mid stage of illness were 94% less likely than children in the early stage to report an Excellent of Very Good quality of care. Based on the results of the multivariate analyses, the signifi cant factors for satisfaction were other race, college graduate, not being married, mid stage, and Site 2. Children of other races were about 95% less likely than White children to report high levels of satisfaction. Figure 10. Parent Report of Overall Quality of Care, By Site 64% 67% 32% 33% Overall Site 1 Site 3 Site 4 Site 5 Site 6 5% Excellent to Very Good Good Fair to Poor 0% 75% 21% 4% Children in the mid stage of illness were 95% less likely than children in the early stage to report high levels of satisfaction. Not being married and college graduation was associated with higher levels of satisfaction as compared to their referent groups (married and less than high school). Parents in Site 2 report lower level of satisfaction than the referent group (which is Site 3). In summary, results from the multivariate analyses suggest that there are few differences in quality of care even after controlling for individual factors and site enrollment. There are however several signifi cant differences in satisfaction. Satisfaction with PIC:TFK Services Children s Services Parents are next asked about what PIC:TFK services their children received. Overall, 94% of parents report that their children received one or more PIC:TFK service. Table 2 shows parental reports of children s service usage across all sites 19. Between 33% and 57% of children receive support counseling 20. Parents also report use of music, art and play therapy sessions; however, it is unclear from the parent report data whether these modalities are delivered as part of the support counseling session or separately by a licensed therapist. Parents are also asked whether their children s PIC:TFK nurses or therapists changed during the last four months. Overall, of Parameter estimates are considered to be statistically signifi cant if their p values are less than, or equal to, Based on the results of the multivariate analyses, the signifi cant factors for quality of care were other race and mid stage of illness. Children of other races were about 85% less likely than White children to report Table 2. Percentage of Parents who Report their Children Have Received Services, By Site Overall Site 1 Site 3 Site 4 Site 5 Site 6 Support Counseling 43% 53% 51% 35% 57% 33% Personal Care 35% 50% 33% 14% 57% 23% Nursing Care 60% 74% 61% 35% 80% 50% Music Therapy 44% 5% 73% 84% 13% 4% Art Therapy 25% 21% 52% 0% 13% 8% Play Therapy 18% 26% 18% 0% 27% 8% pg. 24

25 parents report a change in their children s therapists although responses are not uniform across sites. No parents at Site 5 report a change in their therapists, whereas 38% of parents at Site 1 report seeing different therapists within four months. Whether or not the children s PIC:TFK nurses or therapists changed in the last four months did not signifi cantly affect parental satisfaction levels for any of the six PIC:TFK services. Percentage 100% 80% 60% 0% Figure 11. Very Satisfied to Satisfied with PIC:TFK Services 95% 93% 95% 98% 98% 100% Support Counseling Personal Care Nursing Care Music Therapy Art Therapy Play Therapy Very Satisfied to Satisfied w ith Service Figure 12. Satisfaction with the Frequency of PIC:TFK Services The next series of questions ask parents to report their satisfaction with each PIC:TFK service, if their child received that service. Responses cannot be reported at the individual site level due to low samples sizes. Percentage 100% 80% 60% 0% 92% 94% 94% 90% 83% 84% Support Counseling Personal Care Nursing Care Music Therapy Art Therapy Play Therapy Satisfied with Frequency of Services Figure 11 shows that parents are overwhelmingly satisfi ed with all types of PIC:TFK services received by their children. Figure 12 shows that most parents are also satisfi ed with the frequency of services, and that parents are least satisfi ed with the frequency of art and play therapies. Although not shown in the fi gure, eight out of nine parents who were not satisfi ed with the frequency of art and play therapies would prefer more frequent services. Parents are asked what other services their children might have received. Sixteen families from two sites report that their children received massage therapy and all of those families are very satisfi ed to satisfi ed with the service. Two parents would have preferred more frequent massage services. For each PIC:TFK service, parents whose children did not receive the service are asked whether they would like for their child to receive that service. Responses are Percentage 100% 80% 60% 0% Figure 13. Children s Unmet Demand for PIC:TFK Services 36% 38% Support Counseling shown in Figure 13. The greatest unmet need is for music, art, and play therapies, perhaps refl ecting the lack of registered expressive therapists in Florida. One-half of parents report that they would like their children to receive nursing care. Family Services For each PIC:TFK service, respondents are asked whether they or another family member received that service. Parents are only asked if another family member has received music, art, or play therapies if the enrolled 51% 59% 51% 55% Personal Care Nursing Care Music Therapy Art Therapy Play Therapy child has have a sibling living in the household. Sixty-fi ve percent of parents report that there is more than one child under the age of 21 years in their household. Table 3 shows the number of parents who report that a service was used by another family member, by type of service 21. The results show that family members use counseling more than any other service. Respite is the second most commonly received service; however, receipt of services varies by location. Few families receive volunteer services or expressive therapies (art, music pg. 25

26 Percentage 100% 80% 60% 0% Table 3. Percentage of Parents who Report that Family Members Received Services, By Site Overall Site 1 Site 3 Site 4 Site 5 Site 6 Support Counseling 37% 32% 49% 39% 29% Music Therapy 17% 0% 43% 0% 0% Art Therapy 16% 0% 39% 0% 22% 0% Play Therapy 11% 25% 11% 22% 7% Respite 27% 39% 12% 3% 29% 44% Volunteer Services 11% 11% 16% 0% 27% 4% No Services 31% 42% 27% 30% 29% Figure 14. Very Satisfied to Satisfied with PIC:TFK Services Received by Family Members 94% 100% 100% 100% 97% 100% Support Counseling Music Therapy Art Therapy Play Therapy In Home Respite Very Satisfied to Satisfied w ith Service Volunteer Services Unmet Needs One-quarter of parents report that they would like to see six additional services covered by PIC:TFK. Table 4 shows that the greatest unmet need is for provision of services to the parent or other family members while the child is in the hospital. Currently, there is no billing mechanism to enable family members to continue receiving PIC:TFK services while a child is hospitalized. The second greatest need is the need for interaction with families of children with similar diagnoses. It is important to note that some of these services are offered by Medicaid and it is unclear if the parents know they are offered Medicaid services or if they have used the maximum allowable number of units and cannot get more services. Percentage Figure 15. Family Members Unmet Demand for PIC:TFK Services 100% 80% 60% 0% 29% 32% Support Counseling and play therapies), and receipt of these services varies by site. As shown in Figure 14, parents are overwhelmingly satisfi ed with all the PIC:TFK services received by family members. 34% 35% 32% Music Therapy Art Therapy Play Therapy In Home Respite Volunteer Services asked whether they would like to receive that service. Responses are shown in Figure 15. Between 29% and of parents report that their family members would like to start receiving a PIC:TFK service. Decisional Conflict Scale (DCS) The DCS aims to capture the amount of decisional confl ict experienced by parents when making a treatment decision for their children. The DCS measures whether respondents feel like they made an informed values-based decision, are likely to stick with their choice, and are satisfi ed with their decision. The DCS was administered to all parents, even though only 59% of respondents Thirty-one percent of respondents report family members do not receive PIC:TFK services. These families are enrolled in Site 1 (8 families), Site 3 (12 families), Site 4 (7 families), Site 5 (3 families), and Site 6 (7 families). Parents who report that family members did not receive a PIC:TFK service are pg. 26 Table 4. Affirmative Responses to the Inclusion of Additional PIC:TFK Services Service Agree with Inclusion % Services provided by my child s PIC:TFK therapist/nurse to me or my family members while my child is in the hospital 79% Community groups for children with similar diagnoses 77% Nutritional counseling 62% Services that teach me how to care for my child at home once 59% they are discharged from the hospital Short-term day care for my child in a location other than my 53% home Genetic counseling 44%

27 report making a decision about their children s treatment in the past six months. Overall, the respondents average score is 20.3 with a standard deviation of 15.8, indicating a low level of decisional confl ict. Parents responses to the fi ve individual DCS domains are shown in Figure 16 and the bars indicate the standard deviation from the mean score. The mean score is below 25 for all domains. Individual responses to the sixteen statements can be found in the technical appendix. Disenrollments In the three months prior to the telephone survey, the administrative database shows that 41 children disenrolled from the program. Of this group, 16 parents completed the survey. Additionally, three parents indicated that their children disenrolled from the program, for a total of 19 disenrollees. The main reasons given for disenrollment are listed in descending order: Parents do not know the reason why their children stopped receiving services (6), Parents do not know what the PIC:TFK program is (3), Children are no longer in CMSN because they lost Medicaid or insurance coverage (3), Parents think that their children are still enrolled in PIC:TFK (2), Children move to full hospice (1), Children receive private Score Figure 16. Respondents Scores on the Five Domains of the Decisional Conflict Scale insurance (1), Informed Values Clarity Support Uncertainty Effective Decision *Standard deviation bars are shown Parents stop services because they are too busy (1), Parents stop services because they can t get the services that they want (1), and Parental refusal (1). Of the eight parents who believe that their child is still enrolled in PIC:TFK, or do not know why their children stopped receiving services, six are located in Site 2. Site 2 stopped providing services in mid 2008, and it appears that parents in Site 2 are not aware that services have ceased or would like further information. Two parents indicate that they are either too busy to continue receiving services, or that they cannot get the services that they want. However, it should be noted that no parents report stopping services because they are unhappy with the quality, frequency, or type of PIC:TFK services. Although parents may choose to stop services, they are not dissatisfi ed with care. Section III, which is presented next, discusses the process, and importance of, creating new disenrollment categories. In addition, the section cross-walks the disenrollment reasons from the parent survey to the disenrollment reasons noted by the CMSN staff in the PIC:TFK administrative database. pg. 27

28 Summary and Recommendations This section of the report describes parental satisfaction with the frequency and quality of PIC:TFK services received by themselves, their children, and other family members. The survey asked parents about their satisfaction with the scope of the PIC:TFK services and if they have unmet needs. Parents of children recently joining the program are asked about their satisfaction with enrollment procedures and parents of children recently leaving the program are asked the reason for leaving the program. The key fi ndings are as follows: Three signifi cant factors were found to infl uence parents participation in the PIC:TFK telephone survey: children s race, children s age, and the children s stage of illness. Approximately two-thirds of parents of new enrollees indicate that they fi rst learned about the program from their children s NCC. Thirty-fi ve percent of parents of new enrollees do not recall receiving a list or being told about PIC:TFK services and one-half are located at one PIC:TFK site. The majority (89%) of parents are very satisfi ed to satisfi ed with the program benefi ts, and the overall satisfaction rating has increased 6% from Regression results show that being married and graduating from college signifi cantly increase satisfaction. Seventy percent of parents report that the quality of care that their children receive in PIC:TFK is excellent or very good. Regression results show that children of other races and those in the mid stage of illness have signifi cantly lower levels of satisfaction. Ninety-three percent to 100% of parents are satisfi ed with the different services received by their children (support counseling, personal care, nursing care, music therapy, art therapy, and play therapy). Eighty-four percent to 94% of parents are satisfi ed with the frequency of each service. Ninety-four percent to 100% of parents are satisfi ed with the services received by themselves or their other family members (support counseling, respite, music therapy, art therapy, play therapy, and volunteer services). Thirty-one percent of respondents report that no one other than their ill child received any PIC:TFK services. Some families report that they are not receiving some PIC:TFK services but would like to. For example, 36% of parents would like for their children and 29% of parents would like for themselves or another family member to start receiving support counseling. The greatest unmet need for children is for art, music and play therapies. The greatest unmet need for family members is art therapy for siblings, followed by in-home respite services. One-quarter of parents report that they would like to see additional services included in the PIC:TFK benefi ts package and the greatest reported demand is for the continuation of PIC:TFK services to the parent or family members while the ill child is hospitalized. Of the 19 parents of recent disenrollees, six parents are unsure why their children had stopped receiving services, two thought that their children are At a Glance Parents of black children are less likely to participate in the survey. 93% to 100% of parents are satisfi ed with the PIC:TFK services (support counseling, personal care, specialized nursing care, in-home respite, music, art and play therapies). pg. 28

29 still enrolled in PIC:TFK, and three parents did not know what the PIC:TFK program is. The Institute makes the following recommendations. The PIC:TFK Steering Committee, as well as the Technical Advisory Panel, should investigate options for increasing the response rate among parents of Black children. CMSN staff currently enter enrollment information into the PIC:TFK enrollment database, however key demographic information is missing for some children (especially race, gender, and diagnosis) and some contact information is out of date. The Institute recommends that CMSN headquarters consider developing an online enrollment database which interfaces with CAPS and other existing CMSN databases. The Institute also recommends that hospice staff have restricted access to this online database, and the ability to update contact information in real time. Giving hospice staff access to the enrollment database will also aid in joint care planning efforts. Given that 35% of new enrollees were not given a list or told about the available PIC:TFK services, further communication efforts on the part of CMSN and hospice may be necessary. Based on the results from the unmet need portion of the parent survey, the PIC:TFK Steering Committee should investigate if the service package needs to be changed or if further research into the parents needs should be conducted. Parents of all disenrollees should be sent a disenrollment letter as a standard practice and the letter should be printed in both English and Spanish. Survey questions should be added to determine why the parents have unmet needs. For example, if a parent indicates that she would like for her child to receive support counseling follow up questions should be asked to determine if this service was or was not part of her child s care plan, if she asked her nurse for the service and was turned down, etc. At a Glance Some families report that they are not receiving some PIC:TFK services but would like to. Future research should address why some parents have unmet needs and if the PIC:TFK service package needs to change. pg. 29

30 Section III. Administrator Survey pg. 30

31 Introduction At a Glance 15 standard categories are established to track disenrollments. More than 99% of families leave the program for reasons other than being unhappy with PIC:TFK services. Administrative surveys are an important component of the program evaluation, as they give valuable insights into whether the program is operating as intended. Findings from previous administrative surveys have described the sites operational practices, standards of care, and procedures, as well as issues arising during the implementation phase of the PIC:TFK program 22. It is important to monitor practices over time, and thus the current administrative survey covers similar ground to previous surveys. However, the focus of the site visits this year, and this section of the report, is to investigate the reasons why children disenroll from the program. As previously noted in Section I, disenrollments have steadily risen over time, but the reasons for disenrollment are unclear. In part, disenrollments are due to children losing program eligibility, mainly through loss of insurance, death, or moving out of the service area. In past PIC:TFK program reports, other reasons for disenrollment were poorly defi ned and did not clearly distinguish whether parents were dissatisfi ed with an aspect of care, or simply that the program no longer met the children s or families needs. In this report, information from the site visits were used to better describe the reasons why children are disenrolling from the program, and to categorize and enumerate those disenrollment reasons. Data Sources Two data sources are used in this section of the report. In August 2008, administrative surveys were distributed by to CMSN and hospice staff at the six participating sites. Qualitative survey data was then collected from key CMSN and hospice personnel during scheduled site visits in August. During each site visit, the reason for disenrollment was reviewed for each child by the Institute. The fi rst review was with CMSN staff, followed by a more detailed review with hospice staff. As a result of the site visits, the Institute established 15 disenrollment categories and, together with the CMSN and hospice staff, recoded a disenrollment reason for each child. The site visits also provided a timely opportunity for the Institute to undertake face-to-face training with CMSN staff on the new enrollment database. As mentioned previously, the database contains a drop-down list of disenrollment categories. This feature ensures that the collected information is standardized and helps to reduce coding errors associated with typing in text. The Institute trained CMSN staff on which disenrollment category to choose in which situation, and provided staff with a reference table. The Institute also distributed a Database User Manual, explaining how to add data, sort data, print lists, and other functions. pg. 31

32 Survey Results Administrative Surveys Responses from the administrative surveys cover similar ground to the evaluation fi ndings. As reported previously, the sites are complying with the PIC:TFK operational guidelines in terms of referrals, timely follow-up, interagency meetings, care planning, and reporting enrollments. However, the sites still report operational issues with regards to reimbursement and staffi ng. Many of the sites report that the reimbursement rates and the funding allocated to administrative costs are too low. Both of these issues are beyond the auspices of the PIC:TFK Program Director and/or CMSN headquarters and could be addressed by the PIC:TFK Steering Committee. The sites are asked to report the most useful PIC:TFK protocol, and the responses are given in Tables 5 and As of August 2008, the CMSN sites report that the standardized forms are the most useful protocol. The statewide meeting ranks highly for both CMSN and hospice staff, and the technical assistance calls are viewed as the least useful. Site Visits Based on their experiences, CMSN sites are asked to explain why program disenrollments had increased in the last year. Their answers fall into the following four broad categories: Children or family members reach their therapeutic goals and will not benefi t from continued services, Table 5. Ranking of Usefulness of PIC:TFK Protocols, By CMSN Site Families cannot commit to the program because they feel stressed or overwhelmed and don t want another agency coming into their homes, Children lose Medicaid or CMSN coverage, and Families ask for more than the program can offer. The following quotations illustrate some of their sentiments. Some families do ask for more than the program can offer such as increased respite hours, a Home Health Aid every day, using the PIC:TFK physician visits to replace primary care visits. CMSN Office Site Protocol One Three Four Five Six Seven Total Ranking Standardized forms PIC:TFK implementation guidelines Statewide meeting Training by Program Director Site visits by Program Director Chart reviews by Program Director Technical assistance calls Table 6. Ranking of Usefulness of PIC:TFK Protocols, By Hospice Site Hospice Site One Three Four Five Six Seven Total Ranking PIC:TFK implementation guidelines Statewide meeting Site visits by Program Director Chart reviews by Program Director Training by Program Director Standardized forms Technical assistance calls *Site 7 submitted incomplete data and their contributions are not counted towards the total As children spend more time in the program they begin to meet their goals, others have become stable and no longer need the PIC:TFK program or palliative care or upon review there are no further benefi ts for the child from the PIC:TFK program children that are disenrolled can be re-enrolled at a later date if there is a decline in their health status or other services are needed. [Families may feel overwhelmed because of the situation at home] and do not want another person in the home or one more appointment to keep. Analysis of the December 2008 disenrollment data reveal that 49% of children left the program because they no longer meet the eligibility criteria. Table 7 shows the seven ways in which children lose their pg. 32

33 eligibility, and the corresponding percentage of children who disenrolled from the PIC:TFK program for those reasons. The most common reason is relocation, followed by the loss of Medicaid coverage. It should be noted that children can be reenrolled in the program after their insurance is reinstated, or upon transfer to Safety-Net status. Reenrolled children are not included in the disenrollment statistics below. Similarly, if a child disenrolls multiple times, only the most recent disenrollment reason is captured in the disenrollment statistics in Table 7. While 49% of children left the program because they no longer met the eligibility criteria, 51% disenrolled for other reasons. To assist the CMSN sites with categorization challenges, the Institute produced a reference guide that details scenarios for each disenrollment category. The reference guide is reproduced in Table 8. Both the PIC:TFK Program Director and the CMSN sites had the opportunity to provide feedback on these disenrollment categories. Category Family Not Contactable Family Not Interested Family Not Using Services Services are No Longer Required Family is Receiving Services Elsewhere Family Wants Services Not On Offer Unhappy with Services Exogenous Factors Table 7. Disenrollment Criteria- Loss of Eligibility Reason for Loss of Eligibility % Moved out of the county 17% Died 12% Lost Medicaid 12% Closed to CMSN 4% Moved to a long term care facility 3% Moved to full hospice 1% PCP did not recertify child 1% Table 8. Disenrollment Criteria- Reference Guide Scenario Select this reason if you genuinely can not contact the family. For example: Phone number is disconnected. Family has moved, or is away for an extended period of time. Select this reason if you have tried contacting the family and they are nonresponsive. For example: Family member is not home at scheduled appointment times. You repeatedly leave voic messages, with no response. You send a letter informing the family that they will be disenrolled unless they call back, with no response. Select this reason if the family is receptive to the PIC:TFK program, however the program is not a priority. For example: Family says that they are too busy to schedule any appointments. Family says that they feel overwhelmed with another provider in the house. Family says that PIC:TFK isn t a top priority for them at the moment. Select this reason if the child s or family s goals have been met. For example: Family may be in crisis after diagnosis and require support counseling. Over time, the situation stabilizes and family no longer wants counseling. PIC:TFK therapist teaches the family new techniques and the family follows the therapist s recommendations. Several months later, the family and therapist agree that the child s therapeutic goals have been met and that he/she will not benefit from further sessions. Select this reason if services are being duplicated. For example: Child is receiving counseling at school, from a behavioral health provider, or community counseling service. Family is receiving 24/7 nursing care. Select this reason if the family does not understand program goals. For example: Family is more interested in tangible supplies than supportive services. Family only wants services not offered at your site (i.e. registered art therapist). Select this reason if the family has voiced dissatisfaction with the quality or frequency of care. For example: Family reports an incident or expresses dissatisfaction. Family doesn t want to see a particular PIC:TFK therapist. Family expresses their frustration with less frequent services. Select this reason if events occur beyond your control. It is not safe for the case worker to enter the house. Child is incarcerated or there is suspicion of child abuse/domestic violence. pg. 33

34 Table 9 shows the percentage of children who disenrolled from the program for reasons other than loss of eligibility. The most frequent reason for program disenrollment is the families lack of interest, or non responsiveness followed by family not using services. It is important to note that only one family was reported as being dissatisfi ed with PIC:TFK services. Table 9 shows that one-fi fth of parents are not interested, even though they initially agreed to participate in the program. It is unclear why these parents do not return phone calls or miss scheduled appointments. Since hospice resources can be tied up making multiple follow-up attempts, several sites have developed a disenrollment letter to send to non responsive families. The disenrollment letter informs families that services will be discontinued unless they contact the PIC:TFK program. These sites anecdotally report some success with this approach. Currently the disenrollment letters are sent in English, but the sites plan to translate the letters into Spanish. In Section II, parents of children who left the program in the past three months are asked the reason for their children s disenrollment. The Institute cross-checked these parent reported reasons with the PIC:TFK enrollment database reported reasons. Discounting those parents whose children stopped receiving services when Site 2 left the program, there is 50% agreement between what the parents indicated was the disenrollment reason and what the nurses indicated was the disenrollment reason. In the parent satisfaction survey, two parents report that they do not know why services stopped. For those same two parents the database notes that one family did not use services and the other family was not interested in receiving services. Another parent reports that they cannot get the services they want; however, the site reports that the family is not interested in receiving services. These disagreements illustrate the diffi culties inherent in trying to understand the complex reasons behind disenrollment. In some instances it is diffi cult for the PIC:TFK sites to know whether a non responsive family wants to continue receiving services or not. Table 9. Disenrollment Criteria- Reasons Other than Loss of Eligibility Reason for Disenrollment % Family Not Interested 21% Family Not Using Services 10% Services are No Longer Required 10% Family Wants Services Not On Offer 4% Family Not Contactable 2% Exogenous Factors 2% Family is Receiving Services Elsewhere 1% Unhappy with Services <1% pg. 34

35 Summary and Recommendations Administrative surveys and site visits are important components of the program evaluation, as they give valuable insights into whether the program is operating as intended. Findings from previous administrative surveys have described the sites operational practices, standards of care, and procedures, as well as issues arising during the implementation phase of the PIC:TFK program 24. The main focus of the site visits, and this section of the report, is to investigate the reasons why children disenroll from the program. The key fi ndings from the administrative surveys and site visits are as follows: The sites report operational issues with regards to reimbursement and staffi ng. Many of the sites report that the reimbursement rates and the funding allocated to administrative costs are too low. Both of these issues are causes for concern for the long term sustainability of the program, however they are beyond the auspices of the PIC:TFK Program Director and/or CMSN headquarters. The CMSN sites report that the standardized forms are the most useful protocol. The statewide meeting ranks highly for both CMSN and hospice staff, and the technical assistance calls are viewed universally as the least useful. One-half of children leave the program because they no longer meet the eligibility criteria. The most common reason for disenrollment is relocation, followed by the loss of Medicaid coverage or CMSN coverage, death, relocation to full hospice or a long term care facility, and non re-certifi cation by the PCP. One-half of children disenroll from the program for reasons other than loss of eligibility. The most reported reason for program disenrollment is the families lack of interest, or non responsiveness (21%), followed by families not using services (10%) or the hospice determining that services are no longer required (10%). Only one family who disenrolled was dissatisfi ed with PIC:TFK services. The Institute makes the following recommendations. A special study to investigate the indirect costs associated with providing services to PIC:TFK participants should be conducted as many hospices continue to question the adequacy of reimbursement. The study could enumerate the administrative costs of providing services (travel costs, training, care planning, care coordination, staffi ng) which would also be helpful for establishing staffi ng patterns and capacity limits for hospices. The Institute recommends that the PIC:TFK Program Director shares the disenrollment fi ndings with all new sites entering the PIC:TFK program. It is important for hospices to be aware that one-quarter of families entering the program are either not At a Glance CMSN staff report that one-half of children leave the program because they no longer meet the eligibility criteria (e.g., relocation, loss of Medicaid, death). One-third of children disenroll because the parents are non-responsive (21%), or are not using services (10%). contactable or are unresponsive. Hospices can then factor this indirect cost information into their break-even cost estimates and adjust staffi ng rosters accordingly. pg. 35

36 Section IV. Nurse Care Coordinator Survey pg. 36

37 Introduction At a Glance The NCC survey includes the following modules: PIC:TFK program knowledge Attitudes and knowledge towards hospice and palliative care Referral patterns Nationally, experts have called for the integration of pediatric palliative care into ongoing medical management during the early stages of illness, from the point of diagnosis onward 25,26,27. Evidence from studies with the adult population repeatedly show that physicians delay referring patients to hospice because of a lack of familiarity with palliative care, a reluctance to make a terminal diagnosis, discomfort communicating poor prognoses, complicated hospice admission criteria, diffi culties deciding when to refer, and perceived family reluctance 28. Conversely, hospice efforts to educate and engage with physicians in the community can be an enabling factor to physician referral 29. Less is known about providers willingness to refer children to pediatric palliative care services. One study shows that 91% of physicians are likely to refer children with life-threatening and terminal illnesses to a home health agency that provides pediatric palliative care and hospice 30,31. However, physicians are not asked about the timing of referrals, which can be an additional challenge in the pediatric setting as many pediatric illnesses are characterized by prognostic uncertainty and variable lengths of survival 32,33. Although physicians are the primary gatekeepers for hospice referrals, studies show that nurses and physicians face similar barriers to hospice referrals,34,35. The most common obstacles reported by the hospital nurses are the unwillingness of a patient or the patient s family to accept the prognosis or the hospice referral, nurses discomfort, belief of physician s hesitance, sudden death of the patient, and nurses desire to maintain hope among patients and patients families 36. The same survey showed that 65% of the hospital nurses who cared for terminally ill patients in the last three months did not discuss hospice with the patient s family, and their willingness to discuss hospice was predicated on their own comfort levels and their belief in the benefi ts of hospice 37. The PIC:TFK program is the fi rst publicly-funded program to provide supportive care for children and families from the point of diagnosis onwards. In Florida, the primary persons responsible for initiating referrals to the PIC:TFK program are the CMSN Nurse Care Coordinators (NCCs). As such, it is important to determine the NCCs knowledge of, and attitudes towards, pediatric palliative care. This section describes a survey undertaken with all CMSN NCCs in Florida. The survey compares responses from the PIC:TFK sites and the non PIC:TFK sites. All NCCs in the PIC:TFK sites annually receive on-site training from the PIC:TFK Program Manager. The survey aims to determine any gaps in the NCCs knowledge base, which can be useful for developing or refi ning pediatric palliative care training modules in the future. Data Source There are 21 CMSN Area Offi ces located in Florida, yet a centralized list of NCCs is not readily available. Instead, a group-administered survey approach was adopted. Each Regional Nursing Director was asked to provide by the number of NCCs in their Area Offi ce. The Institute then mailed the requisite number of survey packets to each Area Offi ce. Each packet contained a survey, replypaid envelope, and instructions for completing the survey online. A tracking ID links respondents to their respective Area Offi ces; however, surveys are completely pg. 37

38 anonymous. Reminders were sent to the Nursing Directors two, four, and six weeks after the fi rst mailing. Figure 17 shows the number of NCCs in each CMSN Area Offi ce, as reported by the Nursing Directors. There are 454 NCCs in Florida, with the highest number of NCCs clustered in Tampa and the lowest number located in Marathon- Monroe. There are 206 NCCs employed at PIC:TFK sites and 248 NCCs employed at the non PIC:TFK sites. The overall survey response rate is 61% (279/454) 38. The response rate for NCCs from PIC:TFK sites is 68% (141/206) compared with a 56% response rate from non PIC:TFK sites (138/248). Survey Instrument The survey instrument is a modifi ed version of a survey developed by the Institute to capture referral practices of Floridian and Californian primary care practitioners 39. The original survey was validated by fi ve pediatricians and three pediatric nurse practitioners. After consultation with the Technical Advisory Panel, additional questions were added about the NCCs knowledge of PIC:TFK. Two series of questions were also included to assess NCCs general attitudes towards, and knowledge of, palliative care services. The NCC survey is comprised of the following six modules. Demographics: Demographic questions capture respondent s age, gender, race/ethnicity, qualifi cations, prior experience with palliative care, and participation in palliative care training. Respondents are also asked about the number of CMSN children assigned to their current caseload, whether they work full-time or parttime, and whether they conduct their case management solely by telephone. PIC:TFK Program: The NCCs are asked eight questions to test their knowledge of the PIC:TFK program. NCCs are asked whether the PIC:TFK program operates in their area and if so, whether they have referred a child, along with Figure 17. Number of Nurse Care Coordinators in Florida, by CMSN Area Office pg. 38

39 general knowledge questions about eligibility criteria and the scope of PIC:TFK services. Attitudes: An existing survey instrument is used to test the NCCs attitudes towards palliative care. The instrument developers previously tested the survey questions with 264 area physicians 40. In addition to the six original questions, NCCs are asked six supplemental questions developed by the Institute. NCCs are asked to report their agreement with each statement on a 5-point scale from strongly agree to strongly disagree ; responses are then collapsed into agree, disagree, and neither agree or disagree. Knowledge Quiz: An existing survey instrument is used to test the NCCs knowledge of hospice policies and services 41. NCCs are asked seven knowledge questions with possible responses of agree, disagree, and not sure. Responses are then translated into correct, incorrect, and not sure based on criteria specifi ed by the authors. For the statistical analyses, the not sure responses are grouped with the incorrect responses. The overall percentage of correct responses is calculated for the Knowledge Quiz. Palliative Care Quiz for Nurses: The Palliative Care Quiz for Nurses (PCQN) is a 20-item scale used to assess nurses general knowledge of palliative care 42. The PCQN tests the nurses knowledge of the philosophy and principles of palliative care, the management and control of pain and other symptoms, and the provision of psychological and spiritual care to individuals and their families. Possible response categories are true, false, and not sure. Responses are then translated into correct, incorrect, and not sure based on criteria specifi ed by the authors. For the statistical analyses, the not sure responses are grouped with the incorrect responses. The overall percentage of correct responses is calculated for the PCQN. Referrals: These survey questions aim to assess whether or not NCCs will refer children with specifi c diagnoses to palliative care services, and if so, to determine the timing of those referrals. The diagnoses include cancer, cystic fi brosis, HIV, extreme prematurity, genetic diseases, metabolic diseases, sickle cell disease, cerebral palsy, brain injury, muscular dystrophies, asthma, diabetes, and Attention Defi cit Hyperactivity Disorder. For each of the diagnoses, respondents are asked to choose from the following answers: does not need referral, needs referral at diagnosis, needs referral once the child and family had been living with the disease for a while, but the condition is stable, needs referral once the child s condition becomes unstable, and needs referral at the end of life, once curative treatment is no longer an option. Analysis Respondents are split into two groups depending on whether or not their CMSN Area Offi ce participates in the PIC:TFK program. Area Offi ces classifi ed as participants are Ft. Myers, Gainesville, Jacksonville, Miami North, Miami South, Pensacola, Sarasota, St. Petersburg, and West Palm Beach 43. Area maps in this report are created with EpiInfo 44, and all data analyses are conducted with STATA version Chi-square tests assess whether or not there are signifi cant differences in the attitudes toward palliative care for NCCs who are employed in Area Offi ces that participate in the PIC:TFK program versus those who are in a non-pic:tfk Area Offi ce. T-tests assess if there are signifi cant differences in the overall scores on the Knowledge Quiz and the PCQN between the participating and non participating Area Offi ces. Logistic regression models are used to determine whether participation in the PIC:TFK program affects referral patterns. Specifi cally, NCCs are asked if they would refer children who have one of 12 diagnoses. If they respond positively, NCCs are then asked when the most appropriate point in the child s disease trajectory that referral should be made. In the logistic regression, only respondents who indicate that a referral should be made are included. A value of one is assigned to those respondents who indicate that the child should be referred at diagnosis or when the child had been living with the disease for a while and a value of zero is assigned to those who responded that palliative care is only appropriate once curative therapy has been forgone or if they responded that palliative care is not needed at all. Predictor variables that control for demographics and experience were included in the models. pg. 39

40 Survey Results At a Glance Of the 454 CMSN Nurse Care Coordinators in Florida, 61% participated in a pediatric palliative care survey. Table 10. Demographic Profile of Respondents, by PIC:TFK Site Participation Total Non PIC:TFK Site PIC:TFK Site Age Category years 4% 6% 3% years 12% 15% 10% years 24% 21% 26% 50+ years 59% 58% 61% Gender Female 99% 99% 99% Male 1% 1% 1% Race/ethnicity White non-hispanic 81% 83% 79% Black non-hispanic 11% 10% 12% Hispanic 4% 4% 4% Other 4% 2% 5% Years as a Practicing Nurse 0-9 years 14% 17% 11% years 22% 24% 20+ years 64% 63% 65% Highest Degree ARNP 1% 0% 2% MS 5% 2% 8% PhD 0% 0% 0% RN 94% 98% 90% Demographics Table 10 shows that the majority of respondents are older than 50, White, female, and have more than 20 years experience as a practicing nurse. The demographic profi le of respondents at the PIC:TFK sites and non PIC:TFK sites is similar, with the exception that more PIC:TFK respondents hold a Masters degree. Respondents are asked about their prior employment history with either hospice or within a palliative care setting. Fifteen percent of respondents were previously employed by hospice. Seven percent of NCCs had ever worked as a member of a palliative care team in a hospital, and 27% had ever provided palliative care in a nursing facility or a home health agency. PIC:TFK and non PIC:TFK respondents do not signifi cantly differ in their palliative care experience. Of those respondents with prior palliative care experience, the average number of years employed with hospice was three years, compared with six years with a hospital palliative care team, and fi ve years with a nursing facility/home health agency. Fourteen percent of NCCs report ever completing a formal palliative training course. Completion rates do not differ signifi cantly based on PIC:TFK participation. Of those respondents who completed a training course, the most popular course is the End-of-Life Nursing Education Consortium (ELNEC) curriculum followed by the Initiative for Pediatric Palliative Care (IPPC) curriculum. Respondents are asked several questions pertaining to their employment as NCCs. Most NCCs are employed full-time (90%) and, on average, they have worked as an NCC for 7.5 years. There are more part-time employees at PIC:TFK sites (16%) than at non PIC:TFK sites (5%). Thirtyseven percent of respondents report that they conduct all of their case management by telephone, and this does not vary signifi cantly by PIC:TFK site participation. The average self-reported caseload for a NCC is 181 children, although Figure 18 shows that the average caseload varies signifi cantly across pg. 40

41 Figure 18. Average Caseload of CMSN Children, by CMSN Area Office the state. Orlando NCCs report the highest caseloads, followed by Ocala, Rockledge, Ft. Lauderdale, and Jacksonville. NCCs report the lowest caseloads in Marathon- Monroe, followed by Ft. Myers and Pensacola. On average, PIC:TFK respondents report signifi cantly smaller caseloads than non PIC:TFK respondents. The average self-reported caseload at a PIC:TFK site is 166 children compared with 196 children at sites where the PIC:TFK program does not operate. Percentage 100% 80% 60% 0% Statement Figure 19. Time Spent Coordinating Care after a PIC:TFK Referral 49% 27% 24% No Difference More Time Less Time Time Spent Coordinating Care Table 11. Familiarity with Eligibility Criteria for the PIC:TFK Program Correct (%) Incorrect (%) Not Sure (%) The PIC:TFK program operates in my area 56% 11% 33% A child must be in the Children s Medical Services Network 34% 25% 41% to be eligible for the PIC:TFK program To be eligible for the PIC:TFK program, a child s primary care physician must certify that the child has a life limiting illness 70% 4% 26% PIC:TFK Questions NCCs at the PIC:TFK sites are asked whether they have referred a child to the PIC:TFK program. Sixty-three percent of NCCs have made a referral, and of those, 51% report that the referral impacts the amount of time spent coordinating the child s care. Figure 19 shows that 24% of NCCs think that the referral results in less time spent coordinating care, while 27% report the opposite. All NCCs are asked three questions about their familiarity with the PIC:TFK program s eligibility criteria (Table 11). For all three questions, more than one-quarter of NCCs respond that they are unsure of the correct answer. Eleven percent of respondents can not correctly identify whether or not the PIC:TFK program operates in their area. pg. 41