Palliative care (PC) is the active total care of patients whose disease does not respond to curative treatment. PC aims to achieve the best possible quality of life for patients and their families [1]. World Health Organization (WHO) emphasizes that PC is a component of universal health coverage, integrated into the Sustainable Development Goals. [2,3,4].

PC is not only designed for patients with end-stage cancer but also for people suffering from non-oncological diseases from stages before the so-called end-of-life stage [5]. PC is focused on those patients who experience severe health-related suffering (SHS), that is when the suffering produced by an injury or illness cannot be alleviated without the intervention of a professional and when it compromises physical, social, spiritual, and/or emotional functioning [6]. Thus, people suffering from a disease with SHS can benefit from PC, which not only contemplate therapies focused on managing the pathology, but also those actions and research-oriented to understand and manage ─ more efficiently ─ the distressing clinical complications suffered by these patients [7].

Access to PC is an emerging challenge for global public health. Patients with diseases that require end-of-life PC experience a significant economic burden [8]. Several studies have reported that the costs related to end-of-life care represent around 25–30% of insurer medical expenditures [9,10,11]. In addition, there is a relevant cost component related to patient care that is incurred by society. [12,13,14]. Thus, access to PC in the coverage schemes should be prioritized because the evidence supports that this health service alleviates the suffering of patients, families, and saves money for healthcare systems and society [15,16,17]. Moreover, annually more than 61 million people worldwide experience about 6 billion days of SHS that can be potentially alleviated with access to a PC [18]. Regarding the need for PC, it is estimated that by 2060, 47% of worldwide deaths will experience a high SHS burden, potentially requiring PC [19].

Chile has made progress in providing access to PC since the implementation of the Explicit Health Guarantees (GES) regime, which included access to a set of PC services for cancer patients [20]. Although this is an important achievement, there is still a need to promote interdisciplinary management, provide continuous care with good quality home assistance and extend its coverage to non-oncologic diseases. [21].

According to the Global Atlas of Palliative Care report, around 30% of the population requiring PC are patients with malignant neoplasms [9]. In the Chilean settings, the latter reveals a major gap in terms of access to PC, with a significant fraction of the target population (around 70% of patients who experience SHS) excluded from the social security system. Regarding the distributional consequences of the lack of access, this is more detrimental to patients with low socioeconomic status, since they tend to have greater barriers to access to health care, and therefore, to PC [22]. In addition, the demand for non-oncological PC may be increasing by demographic and epidemiological changes that result in a high incidence of complications of chronic diseases. [9, 23, 24].

However, in the context of a new palliative care law which was recently launched in Chile [25], PC services will be expanded to patients with non-oncologic diseases. For this purpose, the local authority required a precise estimate of the resources needed to provide them to the healthcare sector adequately. Although some efforts have been made to develop estimates of the population in need of PC [6, 7, 19, 26,27,28,29,30,31,32]; none of them has reached the technical robustness and legitimacy to support the national health policy. This report is the result of the collaboration between the Ministry of Health of Chile (MoH) and the technical group of Health Technology Assessment at Pontificia Universidad Católica de Chile, to generate a validated methodology and a reliable estimate of the demand for PC in Chile.

Methodology

To estimate the need for PC, we considered the aim of the law project proposed by the Chilean Ministry of health, which is to provide PC coverage in the last year of life through domiciliary care. Thus, a retrospective approach was used based on annual death records. This approach assumes that the number of people needing palliative care during the year represents those who finally died during the same period. This approach is considered adequate in those cases where palliative care services are focused on providing end-of-life care [6, 28, 31], which is precisely the purpose of the present study.

The estimation of the need for PC was carried out according to the following steps: (1) Identification of the diseases susceptible to receiving non-oncological PC; (2) Estimation of the annual deaths whose cause corresponds to the group of diseases identified; (3) Estimation of the proportion of deceased patients for each disease that will require palliative care; (4) Estimation of the number of patient-years based on the expected survival adjustment.

Identification of diseases susceptible to receiving non-oncologic palliative care

There are several sources related to the identification of diseases susceptible to receiving PC. Three different approaches were considered to explore those variations for the identification of diseases susceptible to non-oncologic PC. For the first approach, a group of diseases defined according to their ICD-10 category (International Classification of Diseases, 10th edition) were identified based on the proposal of Murtagh et al.[28] and the Lancet Commission report [6], which included a larger number of diseases. In addition, we validated this list with local clinical experts (a physician specializing in PC with more than ten years of clinical expertise, academic training in this field and advisor experience to the Health Ministry in PC policies), which led to the incorporation some additional codes. Oncologic pathologies were discarded, but the code for specific neoplasms associated with HIV.

For the second approach, we restricted the set of diseases identified in the first approach excluding those, according to expert judgment, who have a low probability of utilization of PC for end-of-life care. They included HIV-related infectious diseases, acute life-threatening diseases such as acute myocardial infarction, pulmonary embolism, acute pericarditis, acute myocarditis, acute renal failure, or infectious diseases of the respiratory system. The main assumption is that the patient suffering any of these conditions as the cause of death and diagnosed close to the date of death, would not have been a candidate for PC before that date. Finally, for the third approach, we used the group of pathologies proposed by Murtagh et al. [28]. The groups of pathologies to be considered for each approach, together with the ICD-10 codes, are presented in Table 1.

Table 1 Health conditions are included in the estimate of the population requiring palliative care Estimate of annual deaths

Data on the number and causes of death were obtained from the Department of Health Statistics and Information of the MoH which is publicly available [33]. This provides cause of death data for each fatality in Chile and other demographic details. These data were filtered according to the cause of death for 2018–2020 and by month.

Estimate the proportion of deceased patients for each disease receiving palliative care

To estimate the proportion of deceased patients receiving PC for each disease, we used the weights developed by the Lancet Commission [6]. Briefly, the expert’s commission developed a multiplier that informs the proportion of people from each health condition experiencing SHS [6] and hence who required PC. For those pathologies not considered by the Lancet Commission, similarities were discussed with clinical experts to match each code in order to provide an equal proportion of palliative care needs [6]. Supplementary Table 1 reports the proportion of patients who will require PC for each pathology. We applied this approach to all scenarios, including the one that considers the set of pathologies reported by Murtagh et al.

Estimation of the number of patient-years based on the expected survival adjustment

The purpose of the present study was to estimate the demand expressed as the number of patient-years. This estimate is equivalent to the expected number of patients who will require PC for a whole year. Standardization is needed because not all patients who die in a year require the same amount of PC time. While some will require a few days, others will require full-year support.

To estimate the average non-oncologic PC length of stay (LOS), similar derivation between oncologic and non-oncologic average PC LOS were assumed. This derivation was obtained from Jordan et al. [34]. This study reported an average oncologic PC LOS of 28 days and 24.27 days for non-oncologic PC LOS. Consequently, 13.3% of reduction was calculated and applied to Chilean data regarding the average oncologic PC. In addition, there is an average of 180 days of survival for cancer patients who receive PC in Chile [35]. Thus, the expected average non-oncologic average PC LOS were 156 days. This approach assumes that the hazard of death between day one and the day end of the month is constant. This value provides information for forecasting deaths and can be applied to allocating resources to palliative care.