Background

Patient-reported outcomes (PROs) are defined as reports of a patient’s health condition that come directly from the patient, without any interpretation of the patient’s response by a clinician or anyone else [1]. Discrepancies in symptom assessment between providers and patients have been reported in cancer care, and the use of PROs has been recommended for patients receiving palliative care and chemotherapy [2, 3]. Moreover, in cancer care, PROs have been shown to improve communication between patients and healthcare providers, identify patient symptoms, improve patient assessment and care, and increase patient satisfaction [4,5,6,7,8,9]. Furthermore, patients undergoing routine PRO evaluations during chemotherapy have an improved quality of life and prolonged survival [7] (Additional file 1).

In a previous randomized controlled trial (RCT) conducted in outpatient cancer patients, 60% of patients were eligible for a PRO evaluation according to study inclusion criteria and reported symptoms when using the administered questionnaire [10]. Further, a systematic review conducted by a British research group reported that many PROs are administered in an outpatient setting [8]. Moreover, in an international palliative care setting, the most frequently used instruments were reported to be the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire-Core 30 (EORTC QLQ-C30), the Edmonton Symptom Assessment System (ESAS), and the Palliative care Outcome Scale (POS) [8]. However, in palliative care clinical practice, it is estimated that about 60% of patients are unable to complete the relevant questionnaires due to the state of their health and related factors [11].

In Japan, the main providers of specialized palliative care services are palliative care units (PCUs), palliative care teams (PCTs), palliative care outpatient services, and home hospices. Specialized palliative care services provide care for patients with complex needs [12]. In a review article evaluating the POS, it was reported that the Support Team Assessment Schedule (STAS) was more frequently used in Japan and China; this scale collects information on healthcare provider assessments rather than PROs [13]. This finding was considered to be due to a preference for a paternalistic medical culture across the healthcare system in Japan. However, this finding may not reflect the actual clinical situation. For example, five years have passed since this survey, and screening of patient distress (whether through PROs or other modalities) has since become obligatory in Japanese designated cancer hospitals [14]. In addition, one of the PROMs CCNSq was developed in Japan. The CCNSq was developed with the aim of standardizing an evaluation tool for the assessment of symptoms in cancer patients eligible for palliative care in Japan and to screen for distress and identify symptomology in patients undergoing chemotherapy. This tool is also used in the consideration of the introduction of palliative care at an early stage. Moreover, the CCNSq has been shown to be reliable and valid in previous studies [15, 16]. The questionnaire form for the CCNSq includes the following fields: the area of most concern, an NRS, a Distress and Impact Thermometer, a request for consultation with a specialist team. Moreover, the momentum for PROs has been building in Japan (as successors to the Japanese-language version of the STAS, the STAS-J), and the Japanese version of the patient-rated Integrated POS (IPOS) has been verified for reliability and validity [17].

In the present study, we aimed to investigate the routine use of PROs in specialized palliative care in Japan. Moreover, in situations in which PROs were not implemented, we evaluated the reasons for not evaluating PROs. We also aimed to clarify the strategies implemented by the institutions that did evaluate PROs in order to devise new ways to use PROs in a wider range of settings. This study is the first report among routine patient assessment and the use of PROs in SPC in Japan and exploratory research.

Methods

This study administered questionnaires and conducted online or telephone interviews. The questionnaires were sent to all institutions providing specialized palliative care in Japan and that were willing to participate in this portion of the study. The questionnaire surveyed respondents (healthcare providers) about their routine use of PROMs and their opinions on PROs. The interviews were conducted with informants who were healthcare providers. The interview asked healthcare providers for specific examples in regard to the usefulness of PROs as well as the barriers to their use; the interview also queried healthcare providers about effective methods used for collecting PROs. Respondents were to be representatives reporting on the institutional reality, not on their personal experience. Although we did not apply for ethical approval for this study, we conducted the study in accordance with Japanese research guidelines, including anonymity and respect for the respondents’ right to free will to participate in the study. For this study, the return of the questionnaire was regarded as the provision of consent for participation in the survey, and consent for participation in the interviews was obtained by e-mail or telephone. Participation in the survey was voluntary. All the survey results were statistically processed and the names of the hospitals were not disclosed to ensure ethical consideration for the respondents.

Definition of terms

We defined “routine assessments” as regular daily or weekly evaluations during the course of inpatient or outpatient palliative care. Conversely, we defined “screening” as a single assessment, such as a distress screening at hospital admission or the patient’s first palliative care visit.

The PCT is a multidisciplinary team consisting of physicians, nurses, pharmacists, and other professionals such as psychologists and social workers that work across different hospital department [18]. The palliative care outpatient clinic provides to outpatients by physicians and nurses with expertise in palliative care. The PCU is a specialized ward for cancer and AIDS patients with the aim of providing them with a comfortable end of life. The home hospice provides in-home palliative care by home physicians and nurses.

Settings and participants

Specialized palliative care service institutions (SPC) providing specialized palliative care in Japan were included in the present evaluation. In Japan, SPCs include PCTs, PCUs, home hospices, and palliative care outpatient clinics. All designated cancer hospitals have a PCT and a palliative care outpatient clinic. Moreover, screening for palliative care is mandatory for all cancer inpatient and outpatient settings of designated cancer hospitals since 2015 [14].

PCUs are required to admit patients with malignant tumors in need of pain management. Although there is no clear definition/regulatory designation for home hospice, in this study, home hospice was defined as an enhanced function home support clinic with a track record of emergency house calls and end-of-life care. Eligible institutions were identified from an open access list of palliative care service by the Ministry of Health, Labour and Welfare. In the implementation of this survey, administrators at designated cancer hospitals asked physicians or nurses on the PCT to respond to the survey. The questionnaire was mailed to 427 designated cancer hospitals, 423 PCUs, and 197 home hospices.

Of the institutions that responded to the study questionnaire, 13 designated cancer hospitals, nine PCUs, and two home hospices responded that they were willing to cooperate in the interview on the questionnaire were interviewed (i.e., a representative healthcare provider was interviewed) after reconfirming their willingness to cooperate in the study.

Survey

Questionnaire survey

We administered the survey between September and December 2019. The questionnaire included the following items: facility background, evaluation tools used at the institution, whether PROMs or non-PROM evaluation tools were used, whether evaluation tools were used routinely, whether evaluation tools were used for screening, patient and provider opinions on PROs, if the facility had discontinued the use of PROMs, and the reasons for discontinuing the use of PROMs if so. We sent one reminder only to the institutions that did not respond to our initial contact (i.e., mailing the study questionnaire and relevant information). The questionnaire was developed through discussion by researchers who were experts in palliative care based on previous studies.

Interview

In addition to the study questionnaire, we conducted semi-structured interviews administered by trained researchers online or by telephone according to an interview guide prepared by our research team. Interviews were conducted by NI, who had interview training and experience in qualitative research. The researcher explained the purpose of the study to the participating institutions. Interviews were conducted between February and April 2020. The interview was administered to gather more insight into specific examples illustrating the usefulness of PROs as well as barriers to the effective collection and use of PROs in palliative care clinical practice; it also helped to gain a better understanding of information on the improvement of processes relevant to the administration of PROMs at each facility.

The interviews included questions relevant to the healthcare providers’ opinions on the use of PROMs for patients receiving palliative care, what the providers felt had been helpful or good about using PROMs for patients receiving palliative care, the usefulness of PROMs in informing care, and improvements in processes for administering PROMs at the healthcare providers’ institution (e.g., staff training and operational methods). The interviews were recorded on an integrated circuit (IC) recorder with the institution’s consent. The coding was compared and two authors (NI, YI) discussed discrepancies to inform the final coding with supervision.

Data analysis

Questionnaire data analysis

We calculated descriptive statistics for each variable evaluated in the questionnaire survey according to the question/category. We performed Fisher’s exact probability tests for the PROMs that were routinely implemented at each institution, the reasons for discontinuing PROMs, and healthcare providers’ thoughts opinions in regard to PROs. We conducted two-tailed tests with a significance level of 5%. Of the returned questionnaires, those that answered at least 80% of the items were included in the analysis. Missing data were excluded. All analyses were performed using JMP 15 data analysis software (SAS Institute, Cary, NC, USA).

Interview data analysis

We also conducted a thematic content analysis of the interview data using verbatim transcripts of the recorded interviews [19]. We coded information on the usefulness of PROMs, barriers to the use of PROMs, and effective methods used for evaluating PROs, grouping similar semantic content according to categories and subcategories. The first two authors iteratively reviewed the analyses until they reached a consensus; another co-author (palliative care researcher M.M.) reviewed their selected categories and subcategories and judged them to be reliable.

Results

Questionnaire survey

Institution characteristics

We sent the study questionnaire to 1,047 institutions providing SPC in Japan and received valid responses from 458 institutions (representing a 44% response rate), with the following distribution: 227 designated cancer hospitals (50%), 212 PCUs (46%), and 19 home hospices (4%).

Routine use of PROMs and non-PROMs

The routine use of PROMs and non-PROM evaluations is shown in Table 1. Overall, only 108 (24%) of the surveyed institutions used PROMs, including 35 PCTs (15%), 66 outpatient palliative care services (29%), 24 PCUs (11%), and one home hospice (5%). Additionally, 93 PCUs (44%) routinely used non-PROM evaluation instruments such as STAS-J. In regard to screening, 76 PCUs (36%) used non-PROM instruments more often than PROMs for both routine assessments and when screening for patient distress.

Table 1 Routine use of PROMs and non-PROM PRO assessment methodologies
Full size table

Table 2 shows the PROMs in use at institutions that reported routine use of PROMs in palliative care clinical practice. The most commonly used PROM instrument was the CCNSq, which was used by the PCT at 20 institutions (57%) and at outpatient palliative care services at 54 institutions (82%). For PCUs, the CCNSq was routinely used at 10 institutions (42%), whereas the Distress and Impact Thermometer was used at 12 institutions (50%). The CCNSq includes the Distress and Impact Thermometer, and some institutions had multiple responses. The QOL scale was used by one home hospice only (5%).

Table 2 PROMs routinely used in palliative care clinical practice
Full size table

Attitude towards PROs

Table 3 shows the opinions about PROs of healthcare providers at institutions that routinely using PROMs. Of the 108 institutions routinely using PROMs, 99 respondents (92%) stated that these instruments were useful in symptom management (representing a higher prevalence than the perceptions at the institutions that did not routinely use PROMs; p = 0.002). Further, 37 of the respondents (34%) that routinely used PROMs and 143 of the respondents (41%) that did not routinely use PROMs had negative opinions in regard to the burden placed on patients through the use of these instruments. There was also a significant difference in opinions regarding burdens on healthcare providers between respondents that did or did not routinely use PROMs (p = 0.007). Moreover, > 50% of the respondents that routinely used PROMs stated (via the surveyed healthcare providers employed therein) that patients cannot be assessed when their disease progresses and that results are influenced by the patient’s level of cognitive function. Fewer than 10% of healthcare providers responded that the healthcare provider can evaluate outcomes even if the patient does not self-evaluate (i.e., reflecting a paternalistic view of medical care).

Table 3 Opinions about PROs of healthcare providers in institutions that routinely use PROMs
Full size table

Interview

Institution characteristics

A total of 24 institutions that responded to the questionnaire agreed to be interviewed: 13 designated cancer hospitals (54%), nine PCUs (38%), and two home hospices (8%). The respondents held a range of positions: 11 were physicians (42%), 13 were nurses (50%), one was a pharmacist (4%), and one was a psychologist (4%). Of the respondents 15 (63%) routinely used PROMs, whereas three had discontinued their use.

Assessment of the usefulness, barriers, and attitudes of healthcare provider towards PROMs

Table 4 shows respondents’ views on the usefulness of PROMs, the barriers to their use, and the practical effective methods used for PROMs. Responses were recorded for usefulness and barriers for patients and healthcare providers regarding the perspective of healthcare providers. Patients reported usefulness in regard to facilitating communication with healthcare providers (n = 8) and improving self-care skills (n = 13), while healthcare providers’ determination of usefulness was in regard to gaining a better understanding of the patient's condition and using the findings of PROMs in the provision of more optimal care (n = 25), building good relationships with patients (n = 6), and educational benefits to healthcare providers (n = 7).

Table 4 Views on the usefulness of and barriers to using PROMs and relevant effective methods
Full size table

In regard to barriers to the use of PROMs, the most common patient-reported barriers included difficulty in using PROMs due to incomprehension or cognitive decline (n = 34), difficulty using PROMs due to worsening symptomology or a decreased state of consciousness (n = 8), the burden placed on patients (n = 5), and difficulty in filling out the form by themselves due to progression of their disease (n = 4). The barriers reported by healthcare providers included the burden placed on healthcare providers (n = 20), difficulty in using PROMs and linking the findings of these reports to care decisions (n = 15), the need for more comprehensive healthcare provider education on the use of PROMs (n = 12), and the need for training in regard to the evaluation and interpretation of PROMs (n = 4).

We also asked about practical effective methods employed in the use of PROMs. Respondents reported that practical effective methods had been implemented in efforts to reduce the burden placed on patients (n = 45), improving healthcare providers’ education on the use of PROMs (n = 21), implementing standard procedures for the use of PROMs (n = 28), and increasing awareness and consideration towards healthcare providers involved in the evaluation of PROs (n = 16).

Discussion

This study is the first report of the routine use of PROs in SPC in Japan. This study showed that (1) only 108 institutions (24%) routinely used PROMs in SPC in Japan, (2) the most frequently used PROM instrument was the CCNSq, and (3) most institutions using PROMs responded that they were useful (i.e., via representative healthcare providers).

In SPC, PROMs were used somewhat more frequently in palliative care outpatient clinics (29%) than in other settings. However, PROMs were used less routinely in PCUs and were rarely used in home hospice care (5%). In palliative care clinical practice, it is estimated that approximately 60% of patients are unable to complete PROMs [11]. This is because patients receiving palliative care have deteriorating medical conditions and experience cognitive decline; this challenge was also mentioned as a barrier to use of PROMs in the current study. Previous study have reported a decrease in the percentage of patients using PROMs as their conditions became more severe [20]. Moreover, patients who are able to go to outpatient clinics are more likely to use PROMs than those who are hospitalized simply because they by definition have a greater state of health and ability (i.e., they are able to attend outpatient clinics), and PROMs can effectively be used to assess and alleviate their symptomology [20].

As evidenced in the current evaluation, the most frequently used PROM was the CCNSq. The CCNSq is a tool developed in Japan specifically in order to screen for distress in cancer patients for the purpose of deciding on the early introduction of palliative care; this instrument has been shown to be both reliable and valid [15, 16]. Although the CCNSq does not include questions on spiritual aspects, it is considered to be a widely used and popular option because it allows for the assessment of a patient's condition from multiple perspectives. The CCNSq can be used as a screening instrument for patient distress independent from decisions regarding early introduction of palliative care.

Many PROMs have been developed worldwide. One instrument, the ESAS, has been translated into more than 20 languages; however, the items mainly concern physical symptomology, and it is therefore difficult to conduct a comprehensive assessment using this tool [21]. However, the choice of PROMs varies from institution to institution and reasons for selection vary by institution. It is necessary and critically important to use a tool that is not burdensome to the patient and can comprehensively assess distress.

In the present study, 92% of institutions that routinely used PROMs reported that these instruments are useful in symptom management, a significant difference from the positive response rate among institutions that did not routinely use PROMs. Moreover, according to the study interview, the reason given for the usefulness of PROMs was that this methodology facilitated communication between patients and healthcare providers. We believed that the use of PROMs allows patients to talk to healthcare providers about their symptoms and concerns, and facilitates communication [22]. As a result, it likewise helped healthcare providers to better understand patients’ symptoms and conditions. We consider that gaining a better understanding of patients’ conditions may lead to a better assessment of care goals (as reported in previous studies) [23]. From these results, we can infer that the use of PROMs allows providers to appropriately assess patients' conditions as well as to provide proper care management for their symptomology. However, institutions that routinely use PROMs did not only respond affirmatively in regard to their usefulness, but also reported in regard to the burden placed on patients and healthcare providers as a result of their implementation. In this study, we did not ask patients directly, so the actual burden of using PROMs for patients is not clear. Patients are the experts on their experience with their diseases, and we believe that patient self-reporting can provide unbiased information and help identify symptom and needs and improve care. In addition, a study of healthcare providers at a cancer center reported that the strengths outweigh the weaknesses, such as the possibility for patients to express their health condition and the facilitation of communication between patients and healthcare providers through the use of PROMs [24]. We consider that the use of PROMs and/or Patient-Reported Experience measures (PREMs) may be useful to patients and that we need to examine ways to reduce the burden use of these tools [25].

More specifically, the results from the interview administered in the present study revealed barriers in the use of PROMs not only for patients but also for healthcare providers. We consider that this needs to be overcome, as evidence by the barriers to PRO implementation, such as concerns about the increase work in using PROMs and lack of training on the use of the tool.

As reported in previous studies, we conclude that the use of PROMs for patients admitted to PCUs requires that providers understand that only some patients should be targeted for provision of this self-report methodology [26]. In patients who are eligible in palliative care setting, the situation may make it difficult to response PROMs due to their condition. In this survey, more than 50% of the respondents answered that PROs were difficult to use due to the progression of the patient’s disease or cognitive decline. On the other hand, Seipp et al. reported the use of PROs may be feasible if the patient’s condition is carefully considered with the use of PROMs [27]. The careful selection of eligible patients may reduce barriers to the use of PROMs that occur as a result of patient conditions (i.e., patients may be screened as to their state of health and the likely burden that the use of PROMs would entail). We strongly recommend that these modifications and nuances be communicated in the education and training of healthcare providers. It is important that healthcare providers are trained in the use of PROs and how they can be used in their care, rather than thinking that it is impossible for all patients to use PROs because they are vulnerable, such as patient in PCUs to reduce barriers. This survey did not directly ask patients, so the actual burden on patients from use of PRMs is not clear. However, we consider that PROMs can be used even if the burden is caused, through patients and healthcare providers recognition that PROs are provided to patients for their care.

We note that, although the present survey did not directly ask patients about patient burden as a barrier in the use of PROs, healthcare providers responded affirmatively in this regard. Hence, the current lack of validation regarding the effectiveness of PROs in SPC may lead healthcare providers to erroneously conclude that PROs are a burden to patients. Therefore, studies providing a high level of evidence in support of the effectiveness of PROMs (or the lack thereof), including gold-standard RCTs, are needed in the future.

In previous studies, healthcare providers were concerned about the burden that PROs represented to themselves in regard to the need for increased explanation and assistance in filling out PROMs when administering these tools in clinical practice, and about how to respond to the identification of symptoms and needs derived from these PROMs; these previously reported concerns [28,29,30] are consistent with the results of the present study. Therefore, it is necessary to create a system that enables healthcare providers to provide treatment and care for patients' identified symptoms and needs in order to reduce the burden placed on healthcare providers.

In this study, the development of standard procedures for the use of PROMs and increased education and training for healthcare providers were suggested as ways to utilize PROs more effectively in palliative care practice. It is necessary to not only reduce the burden on patients in the utilization of PROMs, but to also address burden reduction (including in terms of relevant education and training) for healthcare providers [26, 31, 32]. Currently, an implementation method for PROs that considers the burden placed on patients and on healthcare providers has not yet been established. Therefore, future research is needed in order to establish an effective implementation method. Moreover, in the future, it will be important to accumulate and share know-how and expertise, such as in regard to the results of surveys on how PROMs are currently being implemented in palliative care clinics and the effective methods that institutions have successfully implemented in administering PROMs.

We note that, in the present study, there were relatively few opinions reported among the survey results that reflected the more traditional paternalistic view that healthcare providers can evaluate symptoms and outcomes even if the patient cannot self-evaluate. In some circumstances, a proxy evaluation by a healthcare provider may be useful, but symptoms and psychosocial aspects that are invisible to others may cause discrepancies in evaluation between the patient and the healthcare provider. In fact, it is clear that PROMs are not widely used in palliative care clinical practice, and therefore such responses may have been underreported due to social desirability bias. More specifically, it is likely that non-PROM instruments have been routinely used instead of PROMs in palliative care practice due to Japanese cultural characteristics that favor a paternalistic approach [11, 33]. However, since there has been a shift from a paternalistic approach to patient-participatory medicine within Japanese healthcare during these past decades [34], it is possible that social desirability bias may have influenced the study results to a small extent [35]. In Japan, it was found that PROs are not routinely used in clinical palliative care compared to other countries, such as Australia, where PROs are focused and implemented in clinical palliative care [36]. We believed it is necessary to implement PROs in the future to improve the quality of palliative care. This study revealed the use of PROMs and barriers in SPC, but we believe that further RCTs on symptom management, facilitation of patient-healthcare provider communication, and overall survival need to be conducted in order to clarify the effectiveness of PROMs.

Study limitations

One limitation of this study is that the collection rate for the administered questionnaire was less than 50%. Therefore, our results may not reflect the actual situation in regard to the use of PROMs in palliative Japan as a whole, and the overall generalizability of our findings is unclear. Moreover, we evaluated only 19 home hospices in the questionnaire survey, which might not be sufficient to reflect the situation in those settings or in the professional palliative care field in general. The second limitation is that we did not research palliative care for people with other diseases. Therefore, the results do not show the overall results of palliative care.

Conclusions

This study is the first to report the routine use of PROs in SPC in Japan. In sum, the survey administered in the present study quantified the status of the routine use of PROMs in SPC in Japan, revealed barriers to the wider use of PROMs in this country, and identified needed innovations in the implementation of this assessment methodology. Based on the results of the present study, it is necessary to consider the usefulness of PROs in clinical palliative care, as well as the careful selection of PROMs according to the patient's condition and how specifically to introduce and operate PROMs in order to minimize the resulting burden on patients and healthcare providers. In addition to cross-sectional questionnaires and interviews evaluating the perceptions of both patients and healthcare providers, studies objectively evaluating the usefulness of PROMs (including in RCTs) are needed in the future.