Advancing Pediatric Palliative Care In A Low-Middle Income Country: A Descriptive Analysis of the Implementation, A challenging But Not Impossible Task

Background: The abyss of access to palliative care and pain management in Latin America remains to be an unaddressed global health issue. Efforts to improve the development level of Palliative Care (PC) provision have traditionally targeted services for adults, leaving the pediatric population unaddressed. Examples of such services are scarce and should be portrayed in scientic literature to inform decision-makers and service providers on models of care available to tackle the burden of Pediatric Palliative Cares (PPC) in Low-and middle-income countries (LMIC). The purpose of this study is to describe the implementation of a pediatric palliative care program “Taking Care of You” (TCY), in a tertiary care university hospital in Cali, Colombia. Methods: A program’s database was built with children between 0 to 18 years old and their families, from year 2017 to 2019. Descriptive analysis to evaluate the impact of the program and service delivery was conducted. A theory-based method was directed to describe the PPC program, according to the implementation of self-designed taxonomy, mapping theoretical levels and domains. Clinical outcomes in patients were included in the analysis. Results: Since 2017 the program has provided PPC services to 1.965 children. Most of them had an oncologic diagnosis and were referred from hospitalization services (53%). The number of outpatients increased by 80% every trimester between 2017-2018. A 50% increase was reported in hospitalization, emergency, and intensive care units during the same time period. Conclusions: The program addressed a gap in the provision of PPC to children in Cali. It shows effective strategies used to implement a program, and how the referral times, coordination of care, communication with other hospital services were improved. Providing compassionate/holistic care to children with Life-limiting and threatening diseases, and in end-of-life. The implementation of this program has required the onset of specic strategies and arrangements to promote awareness and education proving it a hard task, yet not impossible.

services (7). The statistics about PC need are expected to be larger in Latin America due to the health inequalities that the region faces. The Lancet Commission on Palliative Care reported an unaddressed lack of access to PC and pain relief globally, which affects low-and middle-income countries (LMICs) starkly (8) where the provision for PPC and pain relief are estimated to be neglected (8). A systematic review reported that 66.7% of the countries in South America did not have any PPC activity until 2011, when the rst integrated PPC service was documented (9). The eld of PC has continued to develop and become an active area of research and advocacy. Even though a new de nition of the concept is available (10), the generalized provision of healthcare in Latin America has focused on improving access to specialized adult services. This results in an untrained health workforce in the eld of pediatrics (11) and a lack of PPC services to address the current need.
The 2012 Atlas of Palliative Care in Latin America reported Colombia as having four hospice-type residencies, only one second-level service, and 13 tertiary care units that serve both adults and children (12). There are still no available data regarding the national need for PC; however, according to the Population-based Cancer Registry (PCR), the incidence of pediatric cancer from 1986 to 2005 suggests that the cases increase over the years and by age group (13).
To date, national PC services are estimated to have grown almost 500% in the past 5 years. However, developing PPC in Colombia requires overcoming even greater barriers than for adults, mainly due to the absence of PC academic training programs for pediatricians and the lack of related educational objectives for other healthcare professionals (9,14).
In light of the pressing need to promote PPC in Latin America and considering the characteristics of PPC in Colombia based on the recommendations from the American Academy of Pediatrics (AAP), the Institute of Medicine (IoM), and the International Meeting for Palliative Care in Children's Trento (IMPaCCT) to introduce PPC teams (12,15,16), the objective of this paper is to report the process of implementing the PPC program "Taking Care of You" (TCY) in the city of Cali in Colombia. As with other LMIC health inequalities and a lack of scienti c literature on the subject, Cali represents a context where promoting PPC might seem to be an impossible task. This article re ects on the strategies attempted for developing a multidisciplinary program that could provide coordinated care and symptom management to children with life-threatening and limiting conditions, geared toward reducing suffering and improving patients and families' quality of life.

Contextual Factors
Geographic and demographic context Colombia has a population of 48,258,494 inhabitants, 51.2% of the population is female, and 22.6% are between 0 and 14 years of age(17). Cali is a city with 2.5 million inhabitants (18). There is limited data available regarding the number of pediatric patients with a need for PC or PPC services available in the country; however, between 1986 and 2005, the incidence of pediatric cancer in the city was 109.9 cases per 100,000 person-years in children below 1 year of age, 146.9 cases per 100,000 person-years in children between 1 and 14 years of age and 157.4 cases per 100,000 person-years between 15 and 19 years of age (13). The past two years reported a boom in the enablement and creation of PPC teams in the main cities of Colombia.

Colombian legal framework for PC regulation
Since 2014, the country has passed relevant measures related to PPC regulation. Important milestones include the regulation of services in the country, including those aimed at the pediatric population (19), guidelines for inclusive care in PC, differentiating attention for adults and children (20), speci c standards for Children Cancer Care Units (CCCU) where PC and pain treatment must be guaranteed from the beginning of treatment(21), a cancer control plan with a mention of PC (22), and regulating the process of dignifying death for children and adolescents, allowing euthanasia (23). (Colombian regulatory framework is summarized in Appendix 1 under supplementary data) The program "Taking Care of You" The program operates in the Fundacion Valle del Lili, a nonpro t university hospital that functions as a referral center for the southwestern region of Colombia. It has 177 pediatric beds and cares for approximately 50,000 children annually.
Our institution has a general PC program that has operated since 2007 and is led by family medicine physicians, nursing staff, social workers, and psychologists. The general PC program has attended to the entire population, including children until late 2017, when the specialized PC program for pediatric patients called Taking Care of You (TCY) was launched, led by a pediatrician with a team that included a family medicine physician, a nurse, a social worker, and a psychologist.
The main objective of the program is to provide coordinated, multidisciplinary and humanized care to promote quality of life for patients under the age of 18 with life-threatening and limiting conditions in any hospital care service (emergency room, hospital room, pediatric intensive care unit (PICU), neonatal intensive care unit (NICU), and outpatient setting). Patients are treated through an inter-consultation or rst-time assessment by the multidisciplinary PPC team.
This multidisciplinary team designed exclusive approaches for the pediatric population that provided coordination and support to the attending healthcare professionals during hospitalization, clinical decision-making, accompaniment throughout the patients' disease trajectory, around-the-clock availability, activities that seek to promote quality of life, workshops for bereaved parents(24), among others that will be described later.

Program strategies
The program has been implemented following national and international standards (25), which include clinical excellence and safety (including 24 hours, 7 days a week of attention), compliance with statutes and regulations, ethical behavior and consumer rights, inclusion and access, organizational excellence, patient and family-centered care, performance measurement, stewardship and accountability, and workforce excellence.
To ful ll the baseline program goals and objectives, an eight-step strategy was implemented. This included education and awareness in PPC (a question-based strategy, summarized in Appendix 2 under supplementary data), institutional support, the participation of the PPC team in academic and healthcare activities, advocacy with other actors in the healthcare system, capacity building in PPC, the formation of a multidisciplinary team led by a pediatrician with training in PPC, and research that is described in Table  1.

Methods
Two approaches were followed to describe the outcomes of the program. First, a categorization of the strategies was conducted to assist with its implementation, and second, descriptive statistics were used to describe the program outcomes.

Mapping implementation strategies
A theory-based method was applied to describe the implementation process of our program. The approach is explained in detail elsewhere (26). The matrix combines two frameworks: rst, the levels of organization in uenced by implementation (27), and second, the domains of implementation (28). These combined frameworks provide a comprehensive matrix where de ned project activities can be positioned according to their intended target domain and level, to facilitate a structured description of the project (26,27). Different sources of information were cross-referenced in a database to improve the quality of the information.
To analyze and display the strategies implemented by the project, a matrix exercise using the Cochrane Effective Practice and Organization of Care (EPOC) Review Group Taxonomy 2015(29,30) was conducted. Strategies were grouped into subcategories (Table 2), namely, implementation strategies, nancial arrangements and delivery arrangements. Strategies were placed in a matrix according to the domain of implementation and level of organization.

Program outcomes
A descriptive analysis of the variables was performed using the databases of the program, and no administrative permission was required to access the data. The results were reported using measures of central tendency and dispersion, according to data distribution. Categorical variables were summarized as percentages. The variables analyzed included referrals, precedence, and type of pathologies (i.e., oncologic vs nononcologic), place of death per year. All analyses were performed using Microsoft Excel 2016. Information for referrals was gathered through an online institutional questionnaire applied to multidisciplinary pediatric departments (provided in supplementary data).

Coverage
Since the beginning of the program in 2017, a total of 1,965 children have been referred. In 2017, 146 pediatric patients were assessed by the general PC team in the pediatric services of the institution (emergency room, hospital room, PICU, NICU, and outpatient setting), whereas in 2019, 771 patients were assessed by the speci c PPC program in hospital rooms and the pediatric emergency room, and 324 new patients were assessed in the outpatient setting. It represents an 81% and 93% increase in PPC demand in the inpatient ward and outpatient setting, respectively (see Figures 1 and 2).
A relevant aspect in the program development was making a strategic alliance with the PICU and the pediatric oncology unit, which favored early integration of the PPC team in the unit activities, allowing support in decision-making to promote patient quality of life. The advantages of the strategic alliance with the PICU and the pediatric oncology unit were re ected in the transversal and permanent accompaniment during the patients' end-of-life care, optimizing interdisciplinary communication, and therapeutic adherence.
Disease prevalence at the moment of referral to the program Patients referred were classi ed as oncological and non-oncological. We found a similar proportion in both categories, including hospitalization and outpatient care. In our institution, the most frequent oncological clinical conditions were central nervous system tumors (53%) (mainly medulloblastoma and low-and high-grade gliomas), followed by bone tumors (26%) (Osteosarcoma and Ewing's sarcoma). On the other hand, non-oncological conditions were mainly related to severe neurological compromise in 38% of the cases (including congenital malformations and chromosomal abnormalities, such as trisomy, microcephaly, cerebral palsy), followed by neurodegenerative diseases, such as spinal muscular atrophy and orphan diseases (immunode ciencies or metabolic diseases).

Cause of referral
The most frequent reasons for referral to the PPC program in both hospitalized patients and the outpatient setting are summarized in Table 3. The main cause for referral was support during therapeutic effort limitation (89%), followed by end-of-life care (86%). We estimated the number of patients rst referred to our program and their time of death in 2018 and 2019 and found that 35% (n=296) and 26% (n=379) of patients, respectively, died that same year. Additionally, the median time from referral to death was 12 days, with an interquartile measure of 2 to 61 days.
Place of death and participation in bereavement workshops Table 4 shows where most of the PPC patient deaths took place. Additionally, it displays the families' attendance of bereavement workshops carried out within the program framework. Eleven percent of the relatives attended the bereavement workshops in the 2017-2018 period and 25% in the 2018-2019 period.
The program has had around-the-clock availability, during which it received 81 phone calls in 2018 and 244 phone calls in 2019, which were related to symptom management (58%) and care coordination (42%).
Since the TCY program was created, parents or relatives of the children received follow-up calls and a personalized condolence letter that invited them to the bereavement workshop. The objective is to encourage families to participate in an exclusive bereavement workshop, where psychological and educational interventions take place. The process ends with a symbolic butter y release.

Education
There were continuous campaigns in the pediatric units that emphasized the importance and health bene ts from early referral of children with limiting and life-threatening conditions to PPC. This led to the progressive growth of the program and, consequently, a growing number of patients with these conditions in the program. Additionally, since 2019, classroom workshops with multi-professional pro les divided into working groups have been trained in PPC with educational strategies and materials that include roleplay, videos, and clinical case discussions, among others. The core is based on ve essential learning messages: PPC principles, communication, pain management, end-of-life care, and PC as a human right. We managed to train a total of 125 healthcare professionals (11 physicians, 97 nurses, 14 psychologists, and 3 social workers) through 4 educational workshops.
Strategies Figure 3 shows the integration of the EPOC categories on a single matrix displaying the categorization for the domains: implementation strategies, nancial arrangements, and delivery arrangements. The program has focused on advancing individual health professional factors, such as educational and professional interaction, for example, through promoting communication between hospital teams. Second, the program has focused on developing organizational change by generating task shifting, promoting incentives and resources and pursuing change to social, political and legal frameworks. The program has not stressed change at a policy level, though small changes have been reported at generating speci c roles for the program, determining the site of services and providing disease management speci cally from a PPC perspective.

Discussion
This study describes the implementation of a PPC program in a middle-income country of Latin America and the Caribbean, where the existence of training programs in this eld is absent and the assessment by PPC teams is scarce and limited to the main cities of the territory. The program has focused on an unaddressed gap in the provision of PPC for children in Cali. It has improved referral times, coordination of care, communication with other hospital services, provided compassionate/holistic care to children with limiting and life-threatening diseases, and end-of-life care. The implementation of this program has required the onset of speci c strategies and arrangements to promote awareness and education, which has been a di cult task. With this implementation experience, we intend to contribute to the PC philosophy in pediatric patients' care and hope it serves as a model for other regions and countries.
One of the biggest challenges of creating our TCY program was establishing a PC team with skills, training, and exclusive activity assistance for the pediatric population. Nonetheless, a descriptive study of the development of a PPC program at a Children's Hospital of Eastern Ontario, Ottawa, showed a signi cant increase in patient referrals associated with the inclusion of social work resources and experienced PC physicians (31). This was a crucial element to expand PPC demand for children with complex chronic diseases and to increase the program demand three-fold compared with the previous year.
The implementation of the PPC program faced barriers, even with a detailed and comprehensive project plan, adequate economic resources, and a well-trained team. Some studies in this matter identi ed barriers such as resistance from healthcare professionals (32). To decrease this gap, we established solid and continuous educational campaigns throughout the pediatric units highlighting the important role and health bene ts associated with the early referral of children with limiting and life-threatening conditions to PPC, which led to the progressive growth of the program and consequently, a growing number of patients with these conditions in the program (33).
When using the theory-based method description applied to our TCY program implementation, we recognized that the program has focused on advancing professionals and groups of professional in their capacities and competencies and on generating organizational change to nd a niche within the hospital but has not in uenced policy. Small changes at the policy level can be accounted for by the de nition of speci c PPC roles and settings for patient care.
We found the proportion of oncological and non-oncological conditions in our program to be very similar. The most frequent diagnoses in both groups were neurological, similar to other implementation studies carried out worldwide. For example, a descriptive study about the development of a PPC program in Canada with 341 patients reported that 24% of the referred children had a neurological diagnosis (34). We corroborate that children with neurological conditions, even if not progressive, experience high distress due to symptom control di culties (33,35), the high burden associated with care (36,37), and potential complications that could cause premature death (33,34), so the PC team plays a key role in the multidisciplinary management of these patients.
The referral motives regarding our PPC program were associated with supporting therapeutic effort limitation, end-of-life care, comunication support and pain and symptom management, similar to a descriptive study conducted in Argentina, where the reasons for PPC program referrals were pain control (19%) followed by symptom control not associated with pain (16%) (38). This suggests that pain and symptom control are still unattended (39), and treating these symptoms should be a priority and an opportunity for PPC teams to improve the patients' hospitalization experience. We found the average time from referral to our program until death within a year to be 12 days, which shows that there was a late integration of PPC in the wide and varied conditions susceptible to PC in the pediatric population. In the largest retrospective medical record review for a PPC population in the US by Thrane et al., they evidenced that the time from referral to death was 4.4 months (40). This highlights that children are still not being referred to PC until mere days before death, especially in middle-income countries.
Regarding end-of-life care, the global trend is to promote death at home with patient-family comfort and support measures (41); however, in pediatrics, multiple di culties arise when trying to promote this measure, and the hospital scenario is the most frequent place of death (42). Our team recognized the same di culty, and most of our patients (74%) died in the hospital. Some of the barriers identi ed in favoring end-of-life care at home were related to fear from parents or caregivers to face the death of their child alone, the high attention needs of a patient at the end of life, the lack of home care programs with PC training, and the absence of pediatric hospices in the country, similar to what is described in the literature (41)(42)(43). However, some families were able to care for their children at home during their nal days.
Finally, family support during bereavement is an essential biopsychosocial need for the family's wellbeing after the loss of a child (44). Our program does follow-up to support parents in grief. However, bereavement workshop attendance has been low (25%). Therefore, we consider it an important aspect to improve by increasing coverage and offering more family services. Future goals, achievable through My Child Matters grant, include improving inclusion and access by bene tting more than 500 children in the region, accompanying 1000 families and relatives, and educating more than 1500 healthcare professionals. We also seek to make the enlarged program sustainable.

LIMITATIONS
This study shows the implementation of a PPC program in a university hospital in Cali, Colombia. This is a descriptive study, and the data were retrospectively collected, which implies that selection and information bias could be introduced during its development. To mitigate this situation, our databases were crossed with different sources of institutional information to improve the quality of information. On the other hand, our institution is a specialized reference center for patients with highly complex diseases; therefore, the reported number of patients who bene t from the program is high when compared with other institutions. Additionally, the Colombian healthcare system is based on universal coverage provided by several private insurance companies, and as such, patients are often transferred to other institutions due to individual insurance preferences or administrative limitations, making it hard to keep track of all PPC patients. Likewise, there was no guarantee that all the patients with cancer and life-threatening conditions received outpatient care and completed follow-up through the TCY program because patients often need di cult-to-obtain authorization from their health entity to receive specialized care, such as PPC.
Despite these limitations, our implementation model could be an example to mitigate gaps in the PPC service for Latin America and the Caribbean, where access and opportunity to a PPC team are scarce.

Conclusions
The creation of a specialized PPC service increases patient referrals and favors a comprehensive approach to patients with life-threatening and limiting conditions in a general hospital. Institutional support, philanthropic support, awareness, and education are essential for the viability and impact of PPC programs. There are still many improvement opportunities, such as promoting the family home as the preferred location for PPC patients to die surrounded by their loved ones, the creation of a pediatric hospice and improving the accompaniment of bereaved families, among others. With this program implementation description, we hope other PPC programs can be strengthened or new ones can be created in Latin America and the Caribbean.

Declarations Ethics approval and consent to paticipate
This article was previously approved by the ethics committee, Comite de Etica en Investigacion Biomedica of Fundacion Valle del Lili. Since the study is retrospective, and does not contain any individual persons data, no concent to participate was needed, no administrative permissions were requiered to access to raw data.

Not aplicable
Availability of data and material The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Completing interests
The authors declared no potential con icts of interest concerning the research, authorship and/or publication of this article Funding The program is funded in part by My Child Matters grant, given by Sano Espoir Foundation in 2019, to reduce health inequality for children with cancer in low and middle-income-countries, and includes support for PPC programs as an essential element of healthcare for children with oncological conditions, non the less the present paper received no funding from the grant.
Authors´ contributions XGQ (PI) was the person that conceived and designed the program TCY in collaboration with MICS(Co-PI), who helped conceive the program and create the database. LGPL provided knowledge with his writing expertise and was the leader of the investigation group, he additionally helped through his biostatistics expertise. ACH provided the in-hospital setting research expertise and lled the database. WGG provided perspectives from PC settings in the Caribbean. FD provided expertise with disadvantaged populations, NAC provided broad expertise in palliative care. We certify that all individuals listed as authors of this manuscript, 1) have made substantial contributions to conception and design, or acquisition of data, or analysis and interpretation of data; 2) have been involved in drafting the manuscript or revising it critically for important intellectual content; 3) have given nal approval of the version to be published, and 4) agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.