Staffing in California Public Hospital Palliative Care Clinics: A Report from the California Health Care Foundation Palliative Care in Public Hospitals Learning Community

Introduction

Although outpatient specialty palliative care (PC) services have spread in the United States, detailed information about their current state is limited.^1–5 Little is known about how these clinics function, including how they are staffed, which patients are being served, and which clinical services are being offered. ⁶ Although some of this information will likely emerge from the Center to Advance PC's National Registry, and a few studies have been done in this area, current descriptive data are limited, and no available sources have explored this issue in safety net populations or health systems.^1,3–5,7 “Safety net” populations generally consist of patients insured by Medicaid and those with limited or no medical insurance, and health systems that serve these populations have a mission or government mandate, or both, to care for patients regardless of their ability to pay.

Based on a survey of California safety net medical centers, we report on the staffing, patients, services, and current capacity in clinic-based PC. While such descriptive information is not available for comparator safety net institutions in other states, this California data may be of use nationally as public institutions seek to match PC needs with service capacity. California Senate Bill 1004 (SB 1004) is a 2018 California law requiring Medicaid-managed care plans to make PC available to eligible Medicaid patients. Our survey sought to assess the possible impacts of SB 1004 on PC clinics housed in systems that serve large numbers of Medicaid patients.

Methods

A survey was developed by two study authors (M.W.R., K.K.) to assess specialty PC clinics among California safety net medical centers. Survey topics included clinic availability, clinic staffing, scope of services offered, clinic financing, patient characteristics, and respondents' impression of the extent to which SB 1004 had influenced clinic development or expansion. Survey questions asked for quantitative responses primarily, with some narrative responses regarding staffing.

Participation in the survey was solicited from the 15 public hospital safety net medical centers participating in the two-year “California Public Hospital Community-based Palliative Care Learning Community Project” (2018–2019), funded by the California Health Care Foundation. Learning Community activities included webinars, technical assistance, and two in-person conferences to support the expansion of community-based PC among California's public medical centers. Programs were excluded from participating in the survey if their clinic had been in operation for less than one year or was held less than one-half day weekly.

The survey was sent to clinic directors through e-mail in May 2018, with the request that the clinic director complete the survey, assisted as needed by other members of the PC program or medical center. Clinic directors were assured that responses would be presented in aggregate and that data would not be attributed to individual programs. Two e-mail reminders were sent to nonresponders, and as needed the study team followed up by e-mail regarding missing or ambiguous responses. Responses were entered into an Excel spreadsheet, which was used to generate descriptive statistics for the quantitative data. Narrative data was collated and summarized by one author (K.K.) and subsequently reviewed and confirmed by two other authors (M.W.R., M.P.). IRB approval was waived given that no patient-level data were collected.

Results

Discussion

The 10 California public hospital PC clinic programs we studied are providing PC services to over 1000 patients annually, offering a combined 27.5 half-day sessions each week. On average, PC clinics in California safety net institutions offer 2.75 sessions per week, which are staffed with 1.37 nurse FTE, 0.79 social worker FTE, 0.69 physician FTE, 0.52 chaplain FTE, and 0.51 nurse practitioner FTE. Compared with an earlier study by Smith of 20 nonsafety net PC clinics nationally, ⁵ nonsafety net clinics generally offer more clinics/week (average 4.4 half-day clinics vs. 2.75), but are staffed with fewer providers. Looking just at physician, nurse practitioner, nurse, and social work FTE, nonsafety net clinics in the Smith study averaged 2.22 FTE, whereas safety net clinics averaged 3.36 FTE for these disciplines. The need for more FTE per patient in safety net clinics may be explained by the complexity of the patient popualtion, where a high incidence of psychosocial issues, poverty, health disparities, and language and cultural diversity, can make delivering pallative care more complex and time consuming than is the case in other settings.

Our results indicate that service delivery is tied to the particulars and governing structure of each hospital's bureaucracy. There is significant variation however in how critical programmatic decisions are made, including how the outpatient service is staffed (discipline and FTE), where staff practice (inpatient and/or outpatient), and which funds and hiring processes can be used for the program. Half the programs indicated that staff recruitment and engagement was a challenge in the face of workforce shortages or an inability to financially compete for trained staff.

All of the safety net clinics' IDTs included a physician, 80 percent included a social worker, and more than half the teams included a nurse practitioner and registered nurse. Compared with data from the Smith study, safety net PC clinics appear more likely to have a social worker on the team. Even so, safety net teams reported that they wanted to increase the social work FTE, a unsurprising finding given the complexity of the patient population.

A variety of clinic models were reported across the studied programs. Some were embedded in specialty practices (most in oncology), some were operated as stand-alone clinics, and some were a mix of the two. The data did not indicate which arrangement programs were found more advantageous. Each is likely to carry some benefits. The embedded model (where the patients typically have the disease seen in the host clinic) offers the potential for economies of scale by sharing staff (e.g., nurses, clinic coordinator, scheduler), administrative and scheduling responsibilities, space, and the possibility for warm hand-offs between the host clinic and PC. By contrast, stand-alone clinics can see patients with a range of diseases and have more independence in directing resources (staff, space, appointments) to maximize efficiencies. Because eight of the 10 programs all reported a mix of follow-up and new patient appointments, with an average of 1.88 new appointments and four follow-up appointments in their schedule templates, all the clinic arrangements seem to work. New programs choosing a clinic model should select one that works best for their environment and needs.

Despite the variation and multiplicity of funding streams across the 10 programs, use of professional billing as a source of support for program staffing was rare. There may be regulatory explanations for this, including limitations in how Federally Qualified Health Centers (FQHCs)—a designation common among public hospitals in California—can generate revenues for outpatient services. In addition, most safety net clinics have relatively few patients insured by Medicare Advantage plans or commercial health plans that offer pallative care benefits, two revenue sources that are increasingly common for nonsafety net PC programs. Eight safety net clinics reported that approximately three-quarters of their financing came from their health systems. Comparing safety net program funding to that of outpatient clinics in the survey literature, ⁴ it appears that safety net programs are more likely to enjoy institutional support, but less likely to be able to take advantage of both traditional (professional fee billing revenue) and emerging (Medicare Advantage) funding streams.

Most of our respondents reported that the recent (January 2018) implementation of California's SB 1004 law, which mandates access to PC for eligible patients enrolled in a managed Medicaid program, had little influence on the development or expansion of their PC clinics. The reasons for this are not clear from the data we collected, but it is possible to hypothesize. The eligibility criteria for SB 1004 are such that most qualifying patients have very advanced disease, and likely need home-based services. As a result, most of California's Medicaid-managed care plans initially contracted with organizations that offer other types of home-based care (such as hospice and home health) when assembling an SB 1004 provider network. It is possible that, in time, safety net PC clinics and Medicaid-managed care plans will collaborate more fully as a result of SB 1004, in particular in the area of identifying seriously ill patients whose disease has progressed to a point where home-based services are more appropriate than clinic-based services. The public hospital pallative care clinics are uniquely well postioned to serve as a referral hub for SB 1004, and as data indicating the value of providing PC to seriously ill Medicaid enrollees becomes available, managed care plans may see the safety net clinics as an important partner in increasing appropriate use of the SB 1004 benefit.

Staff burnout is an often overlooked area in the analysis of health care delivery. Unchecked, staff burnout can lead to reduced service quality, emotional exhaustion, feelings of ineffectiveness on the job, and staff turnover.^8,9 In light of the myriad challenges public hospitals face, including financial burdens, strained resources, and staff recruitment and retention challenges, staff burnout is a critical concern. ¹⁰ In fact, nearly all of the safety net PC programs (90%) reported addressing staff wellbeing through various activities, highlighting a shared value of supporting both team and individual resiliency. Our study did not assess staff longevity, satisfaction, or turnover, although future research into these topics would be valuable.

Finally, the 10 programs reported providing services and processes that are consistent with the recommendations of the National Consensus Project's Clinical Practice Guidelines for Quality Palliative (NCP Guidelines). ¹¹ All the teams routinely provide a comprehensive assessment and pain and symptom management; 9 of 10 routinely provide advance care planning and care coordination; 8 of 10 routinely provide emotional support and a PC-focused plan of care; and, 7 of 10 and 6 of 10, respectively routinely provide family caregiver support and social service referrals. Spiritual support, although not routinely provided, is provided on an as-needed basis by eight of the 10 programs.

This survey study has limitations. The data were collected through self-report and not confirmed by independent review. Our survey included the majority of California safety net PC programs, but not all, and key data may be missing. Among the 10 programs responding, only eight programs submitted evaluable financial data and only nine programs submitted evaluable patient demographic data. Our findings may only generalize to the state of California. Finally, outpatient PC programs are not static; they grow and shrink depending on multiple factors, such as available funds, hospital priorities, patient needs, etc. Accordingly, the presented analysis must be interpreted as a snapshot in time. The analysis did not investigate differences between newer (one to three years) versus more mature programs (more than three years). A future study analyzing these differences with respect to staffing stability, services provided, and number of patients seen may offer additional important information.

Conclusion

As the field of PC continues to develop, attention should be paid to increasing access to specialty care for safety net populations. This descriptive analysis demonstrates how California's public hospital outpatient PC programs are working to meet this need. Despite significant and ongoing challenges of providing health care in safety net systems, our respondents use an IDT approach to deliver guideline-concordent specialty PC to a significant number of complex patients. More research is needed to understand the most effective and efficient staffing models for meeting the needs of safety net population, with the goal of better supporting safety net health systems in developing rational approaches to staffing and funding their pallative care services.