The present results clarified the availability, status of implementation, and factors related to the implementation of interventional procedures for refractory pain in patients with cancer using a nationwide survey completed by specialists.

In the present study, the proportions of PSs who responded that they were able to perform CPN, phenol saddle block, Epi, and IA were 49.5, 55.2, 75.2, and 40.9%, respectively. In the past three years, almost 50% reported that they had not performed Epi; furthermore, most responded that they had not performed the three other procedures. Previous surveys on specialist pain services examined the availability of interventional procedures. In the UK, procedures were available at 24.5% (CPN), 24.5% (intrathecal neurolysis), and 85.8% (spinal analgesia; 22% for EPI only, 18% for IA only, and 45% for both) of facilities [16]. In Japan, procedures were available at 66% (CPN), 67.4% (intrathecal neurolysis), 88.2% (Epi), and 54.2% (IA) of facilities [23]. Thus, many pain specialist facilities provide interventional analgesia for cancer patients; however, PSs had few opportunities to perform these procedures.

Based on a previous Japanese study [15], we estimated that 3.3% of the 373,584 patients who died due to cancer in 2018 (approximately 12,000 patients) may have required interventional procedures for cancer pain management. Our survey revealed that 1,530 interventional procedures were performed annually by 545 PSs. Assuming that the 1,112 PSs that responded to our survey performed interventional procedures at the same frequency as the 545 PSs, the expected annual number of interventional procedures was 3,122, which is markedly less than the estimated demand. Thus, interventional procedures do not appear to be sufficiently utilized.

Factors related to the implementation of interventional procedures warrant further study. Previous studies reported the following barriers to the implementation of specialist pain management, such as neural blockade and neuraxial infusion: the underutilization of specialists [16, 17]; access issues/geographical issues [18, 19]; inter-facility issues [19]; inability to get appointments [20]; need for repeating procedures [20]; cost issues [17, 18, 21]; the short survival of patients following referral to palliative care services [21]; time on the part of the specialist for evaluation and discussion [16, 21]; complexity [21]; continuity issues, such as the handling of pumps and catheters, creating a pump, procurement of drugs, and management at home [21]; the inexperience of palliative care physicians [18]; perception issues among palliative care physicians (interest or lack of awareness of potential benefits) [18, 21]; and the lack of training for specialists [21]. In the present study, the number of cancer patients with pain seen annually, difficulty in gaining experience, lack of time, and lack of institutional acceptance were associated with the implementation of procedures, with the first three factors being consistent with previously reported associated factors (involvement of specialists in palliative care [16], time on the part of the specialist for evaluation and discussion [16], and the lack of training for specialists [21]). These factors are important because the results of the present study support previous findings.

The following measures may increase the number of interventional procedures being performed. First, in the present survey, the number of cancer patients with pain seen annually (contributing factor) and difficulty in gaining experience and acquiring skills due to the limited number of cases (barrier) were identified as factors related to implementation. Moreover, previous studies reported the lack of training of experts as a barrier to implementation [21]. Thus, PSs need to increase their experience treating such patients. To increase the experience of PSs, several strategies may be effective, including further specialization for the treatment of cancer pain, a region-wide networking system for identifying potential candidates for interventional procedures, and establishing designated teaching facilities. Second, the effective use of time by PSs to practice palliative medicine may increase the implementation of procedures. In the present study, lack of time was associated with the implementation of two procedures: CPN and phenol saddle block. Moreover, increasing the time spent in palliative medicine practice may compensate for lack of experience. In a 2007 survey of lead anesthetists in UK pain clinics [16], joint consulting arrangements were rare, and only 25% of anesthetists’ job plans had time allocated for palliative medicine referrals; however, there was a positive correlation with the number of referrals. Therefore, promoting opportunities for PSs to be involved in palliative medicine may, in turn, increase the number of interventional procedures performed. Third, efforts are needed to educate palliative care physicians who will serve as bridges. The present survey of HHPs and oncologists revealed that they had knowledge of the implementation of procedures, but no experience or may not be able to refer patients to specialists. Palliative care physicians need to act as a bridge to connect patients to specialists who perform these procedures. Previous studies also reported a lack of experience and awareness among palliative care physicians [18, 21]; thus, further education and awareness on indications for and effects of interventional therapies among palliative care physicians are needed. Fourth, the education of IVR specialists may be important for promoting the implementation of CPN because even though many IVR specialists responded that they are willing to perform CPN, actual implementation rates were low.

Since there are few evidence-based interventional procedures, it may be difficult for specialists to provide a rationale for the procedure; furthermore, palliative care physicians who act as bridges may not be able to propose a procedure with confidence and obtain approval from institutions. Further studies to evaluate the efficacy of these interventional procedures are needed.

There are several limitations to the present study. First, although the status of implementation by specialists nationwide was surveyed, we did not obtain information on the implementation status of each facility. A survey of facilities, including designated cancer hospitals, hospitals without designated cancer departments, and home hospices is warranted to obtain more detailed data on interventional procedures for cancer pain management. Second, the valid response rate for each expert, which ranged between 46.8 and 51.5%, may not reflect the overall situation. However, the response rate was sufficient for a survey of individual experts. Third, as Japan has a universal health insurance system, no restrictions on access to medical facilities, and a small geographical area, we considered it unnecessary to ask about geographical distance and cost issues.