Management of terminal cancer pain in Sweden: a nationwide survey

Pain, 54 (1993) 169-179

169

0 1993 Elsevier Science Publishers B.V. All rights reserved 0304-3959/93/$06.00

PAIN 02320

Management

of terminal cancer pain in Sweden: a nationwide survey Narinder

Rawal a, Jan Hylander

a and Staffan

ArnCr b

’ Department of Anesthesiology and Intensive Care, 6rebro Medical Center Hospital, Grebro (Sweden) and b Department of Anesthesiology and Intensive Care, Karolinska Hospital, Stockholm (Sweden)

(Received 30 September 1992, revision received 7 December 1992, accepted 8 February 1993)

Summary The aim of this nationwide survey was to evaluate the extent of pain problems, the principles of pain management and the use of newer drug delivery techniques in terminal cancer patients. ‘Terminal period’ was defined as an expected life span of about 2 weeks. Another aim was to evaluate the level of knowledge of physicians and nurses and to assess their educational needs regarding management of cancer pain. A total of 456 questionnaires were sent to chairmen and head nurses of 6 major specialities (228 departments) that treat cancer patients in Sweden. Depending on the speciality, each questionnaire consisted of 50-60 questions. The response rate was high, ranging from 79% (surgeons) to 100% (gynecological oncologists). During a l-year period 3767 patients (about 10% of total cancer population) were admitted to hospitals due to severe cancer pain. The results show that almost all physicians follow the analgesic ‘ladder’ principle recommended by WHO, prefer the oral route for morphine administration, prescribe opioid analgesics on a regular ‘by-the-clock’ basis and in unrestricted doses. However, about 30% of physicians believe that all patients have moderate to severe pain at the time of their death. Up to 78% of physicians and nurses believe that periodic severe pain is common in terminal cancer patients. In general physicians do not evaluate different pain types nor do they use any instruments for measurement of pain intensity. Surprisingly intermittent subcutaneous (s.c.) or intramuscular (i.m.> injections of morphine are preferred to continuous s.c., intravenous (i.v.1 infusions or to intermittent S.C. administration through an indwelling butterfly needle. Few physicians routinely prescribe prophylactic laxatives. Antidepressive medication is rarely used. Neuropathic and bone metastasis pain appear to be poorly managed. More than 50% of physicians and nurses admit that they have inadequate knowledge about pain evaluation techniques, newer analgesics and newer drug delivery systems. More aggressive use of opioids administered by more humane routes and better use of adjuvant therapy can be expected to improve pain management for the terminal patient. There is also a need for better physician and nurse education. Key words: Cancer pain; Morphine;

Peripherally

acting analgesics; Adjuvant tnerapy; Educational

Introduction

Pain is one of the most common symptoms in patients with advanced cancer. It has been estimated that one-third to one-half of cancer patients in active therapy, both adults and children, and about two-thirds of those with advanced disease have significant pain. In terminal stages this proportion may be as high as

Correspondence to: Narinder Rawal, M.D., Ph.D., Department of Anesthesiology and Intensive Care, &ebro Medical Center Hospital, S-701 85 drebro, Sweden. Tel.: 46-19-151129; FAX: 46-19-127479.

needs

80-90% (Foley 1985; Levy 1985). Management of cancer pain has been receiving much attention lately as evidenced by increasing acceptance of hospices and palliative care units (Kaiser 1984; Murray Parkes 1985). However, in spite of such advances, it is generally agreed that cancer pain is often inadequately managed especially in the terminal period and remains a major clinical problem (Oster et al. 1978; Cleeland 1981; Levy 1985; Bonica 1986). Although many pharmacological, anesthesiological, surgical and radio-oncological techniques are available several reports indicate severe undertreatment of cancer pain (Marks et al. 1973; Hardy et al. 1977; Charap 1978; Oster et al. 1978;

170 TABLE I PARTICIPATING

SPECIALITIES,

Speciality

Surgery Gynecology Urology Pulmonary medicine Oncology Gynecological oncology Total number of departments Total number of questionnaires

NUMBER OF DEPARTMENTS Total number of departments in Sweden 92 53 30 30 16 7 228 456

Response rate Nurse response number (%I

Physician response (number 1%)

83 (90% J 45 (85%) 27 (90%) 28 (93%) 14 (88%) 6 (86%)

73 (79%) 49 (92%) 26 (87%) 29 (97%) 14 (88%) 7 (100%)

(physician and nurse questionnaire)

Cleeland 1981; McKegney et al. 1981; Bonica 1986). In 1979 a Swedish National Board of Health and Welfare ’ expert group concluded that terminal cancer pain was poorly managed in Sweden. Ten years later another expert group from the Board came to a similar conclusion (Am& et al. 1989). The present nationwide questionnaire survey was undertaken with the following aims: (1) to study the extent of pain problem in terminal cancer patients; (2) to assess the principles of cancer pain management in Sweden and to study the extent to which newer drugs and drug delivery techniques are being used; and (3) to evaluate the knowledge level of physicians and nurses and to assess their educational needs regarding management of cancer pain.

M&hods A questionnaire was sent to all chairmen of 6 major specialities that treat cancer patients in Sweden. These included all chairmen of departments of general surgery, pulmonary medicine, urology, gynecology, oncology and gynecological oncology. Depending on the speciality, each questionnaire consisted of 50-60 questions. The questionnaire was sent to each chairman with the request that the physician in the department who was considered most knowledgeable in cancer pain management answer the questions. It was also requested that the physician answering the questions restrict the answers to the general consensus in the department. To study nursing attitudes and practices a modified version of the questionnaire with a similar request was sent to the chief nurse of the same departments. A total of 456 questionnaires, 228 physician questionnaires and 228 nurse questionnaires, were mailed. The survey was supported by the Swedish Society of Anesthesiolo~ and Intensive Care (SFAI). The aims of the survey were explained in a covering letter, it was emphasized that the information requested pertained only to pain management in terminal stages; for practical reasons a terminal period was defined as an expected life span of about 2 weeks. A reminder was sent to the non-responders. Possible answers were provided for every question and the physician or nurse only needed to select and tick the appropriate box. Space for additional

’ National Board of Health and Welfare, Pharmacological

AND RESPONSE RATE

pain management in terminal care, SOSFS(M) 21 f1979), 1-9 (Swedishf.

comments was provided for each question. Most of the answers consisted of either ‘yes’ or ‘no’ alternatives, or ‘frequently’, ‘sometimes’ and ‘never’ alternatives. If not otherwise specified all ranges are percentages, reflecting range in responses according to speciality. Statistical analysis of data was difficult because of too many variables such as (a) at least 2-3 possible answers for each question, (b) 6 different responding specialities, (c) nurse and physician responses and (d) most data are based on opinions rather than exact figures. For information about questions, please see Appendix at the end of the manuscript.

Results The questionnaire was mailed in mid 1990 but the data are from 1989. The response rate was quite high, ranging from 79 to 100% for physicians and 85 to 93% for nurses. The number of departments in the participating specialities and the response rates are shown in Table 1. In the opinion of physicians a total of 3767 cancer patients were admitted during the period from 1 January to 31 December 1989 with severe pain as their predominant symptom. In general most of the physician and nurse answers were similar. Data from Cancer Incidence Register of Swedish Health Board shows that 39,780 persons (20,132 males and 19,648 females) had some form of cancer during 1989. Pain evaluation The use of different pain evaluation instruments is shown in Table II. In 97% of the departments pain was evaluated by patient history only. The use of visual analog scales (VAS) in different specialities ranged from 7% (gynecology) to 29% (gynecological oncology). Questionnaires or body image pain charts for pain evaluation were rarely used. The WHO ‘ladder’ principle of pain management was practiced in 86100% of departments. However, one of the commonest comments by the nurses was that many physicians stopped peripherally acting analgesics at the start of opioid therapy. For patients treated with opioids, 49% of departments had special patient info~ation about

171 TABLE II EVALUATION

OF PAIN AND ITS INTENSITY

Table showing use of clinical history and questionnaires for evaluation of pain and the use of VAS for assessment of pain intensity at the time of starting analgesic therapy. These data are based on physician answers; however, the nurse answers were similar. Speciality

Clinical pain evaluation No. of departments (%I

Use of questionnaires No. of departments (%I

Use of VAS

Surgery Pulmonary medicine Urology Gynecology Gynecological oncology Oncology

27 (38%) 19 (68%) 13 (52%) 23 (50%) 3 (50%) 13 (92%)

3 (4%) 2 (7%) 1 (4%) - (-) 104%) - (-1

13 (20%) 4 (14%) 2 (8%) 3 (7%) 2 (29%) 4 (25%)

these drugs and 27% provided special information for family members. Ah information was oral; onhy 2-4% of departments provided written information, e.g., brochures. Pain in spite of treatment

According to 21-78% of physicians and nurses, periodic pain was common in terminal cancer patients. Fig. 1 shows frequency of occurrence of severe pain and its reIationship to various types of cancer. The physicians and nurses concurred that skeletal metastases pain was particularly difficult to manage. Of note, about 50% of physicians believe that in spite of treatment this pain is especially common in patients with uterine and pancreatic cancer. Fig. 1 also shows that about 30% of physicians believe that severe pain is quite common in all terminal cancer patients in spite of treatment.

Route limits

of

Fig. 1. Frequency of constant pain in spite of treatment in terminal patients suffering from different kinds of cancer. Figure shows the percentage of physicians who believe that constant pain is ‘common’ or ‘very frequent’. Physicians were asked according to their speciaiity. Numbers on bars represent the number of respondents for every question.

dose

The most common routes of morphine administration in Sweden were oral (tablets, mixtures, slow-release tablets) and intermittent se. injections. Surprisingly continuous S.C. morphine was used in very few departments (Fig. 2). Epidural and intrathecal (i.t.1 opioids were used frequently in 42% of gynecological oncology, 22% of surgical and 16% of urological departments. In the remaining specialities the technique was used in less than 10% of departments. Patient-controlled analgesia (PCA) was regularly used in only 3%

*fvY

-

a~l~es~~ admi~~t~ation and m~imum

Pulm.rMd.

h’wY

Qyr*co~

Qynoncology

onwlooy

Fig. 2. Figure showing the most common routes of morphine administration for management of terminal cancer pain. Figure shows only ‘frequently used’ physician answers (answers ‘sometimes used’ or ‘rarely used’ are not inchtded). Numbers on bars represent the number of respondents for every question. * Represents use of regular oral morphine tablets only. Many physicians also prescribe oral morphine as mixture or slow-release tablets.

172 TABLE III PERCENTAGE

OF PHYSICIANS WHO USE HYPNOTICS

AND ANTIDEPRESSANTS

FOR MANAGEMENT

OF TERMINAL CAN-

CER PAIN Speciality

Surgeons Pulmonologists Urologists Gynecologists Gynecological oncologists Oncologists

In about 50% of patients

In occasional patients

~tidepressants (Physicians)

Hypnotics (Physicians)

Antidepressants (Physicians)

Hypnotics (Physicians)

~tidepressants (Physicians)

Hypnotics (Physicians)

4% _

23% 35% 16% 41% 57% 38%

9% 3% 4% 11% 29% 15%

49% 48% 56% 33% 43% 62%

87% 97% 96% 87% 71% 85%

28% 17% 28% 26%

In almost all patients

2% -

of surgical and 34% of ~necological departments. Regarding the route of opioid administration, the order of preference in Sweden was oral > intermittent S.C.injection > intermittent intramuscular (i.m.) injection > continuous intravenous (i.v.1 infusion (Fig. 2). For oral administration the most common drugs in Sweden are morphine, ketobemidone (a synthetic opioid) and slow-release morphine (Dolcontin) tablets in that order. The use of slow-release morphine in gynecological departments was not as extensive as it was in other departments. ’ Oral mixtures and rectal administration are unusual for terminal cancer patients; however, if these routes were selected, the most common drug as mixture was morphine and as suppository ketobemidone. The most common opioids for i.m. administration were morphine, meperidine and ketobemidone; all 3 drugs were used equally frequently. Opioids were administered ‘by-the-clock’ in 86% of departments; only 4% of departments gave opioids on demand. Radiation therapy for pain management was practiced at 52-100% of departments, the exception was gynecological departments where this modality was rarely used. Nerve blocks were used in only 5% of departments. Neurosurgery for cancer pain management was used infrequently in Sweden. When patients were discharged from hospital and home care therapy was instituted oral morphine was preferred by 95% of physicians. The remaining 5% of physicians preferred epidural/i.t. opioids or S.C.morphine infusion. There were no maximum limits for opioids when treating terminal cancer pain. Only 1% of Swedish departments had such limits. The maximum daily dose of mo~hine that was administered by any nurse (during 1989) to a single patient was 360 mg (oral). Another patient received a maximum daily dose of 400 mg morphine epidurally. The maximum reported daily dose for oral morphine by a physician was 500 mg.

The most common supplementary drugs prescribed were hypnotics. Sedatives such as diazepam were prescribed for about 50% of patients (Table III). Antidepressives were rarely used in Sweden. About 50% of physicians did not answer questions about the use of antidepressants in their practice. Of the respondents only 2% of gynecologists and 4% of surgeons routinely prescribed antidepressants to terminal cancer patients. This is notewo~hy because about 50% of nurses believed that anxiety and restlessness were frequent problems, and 24% of nurses believed that depression was a frequent problem in terminal cancer patients (Table IV>. Neuropathic pain

As can be seen in Fig. 3 there was a wide variation in the choice of treatment modalities for the management of predominantly neuropathic pain. With the

’ Oral slow-release morphine (Dolcontin) was introduced in Sweden in late 1988. Recent Swedish pharmacy statistics show an impressive increase in the use of this drug. It has repiaced regular morphine tabiets and is currently the oral opioid of choice.

Fig. 3. Figure shows general consensus in departments for management of predominantly neuropathic pain in Sweden. Numbers on bars represent the number of respondents for every question.

173

skeletal metastasis. Gynecologists are generally more restrictive than other speciaiists regarding the use of NSAIDs (48%), steroids (17%) or radiotherapy (21%). Also, in general, steroids are used rather sparingly in Sweden (Fig. 4).

WflPpY

puhlumd.

“rO,Dpy

W-UWY

onco,*y

Gyna~lqly

Fig. 4. Figure showing how pain due to skeletal metastasis is managed by different specialities. Numbers on bars represent the number of respondents for every question.

exception of oncologists ail physicians used morphine as the most common method to treat this type of pain. Antiepileptics or antidepressives were rarely used to treat this type of pain. Most oncologists (83%) used steroids when neuropathic pain was diagnosed, however, other physicians used steroids considerably less frequently. Radiation therapy, nerve block and TENS were used occasionally. Non-pha~acoiogical methods (acupuncture, hypnosis) were almost never used. Pain due to skeletal metastasis Fig. 4 shows that there is a considerable variation in the choice of adjuvant therapy for pain associated with

Adverse effects of opioids Not surprisingly nausea, constipation and sedation were the most common reported adverse effects of opioid therapy. Table IV shows the frequency of occurrence of various adverse effects. Over 80% of physicians reported constipation as the most frequent adverse effect of morphine therapy. However, routine prophylactic laxative use varied considerably, i.e., 32% (gynecology), 54% (surgery), 65% (urology), 71% (gynecological oncology), 83% (pulmonary medicine) and 100% (oncology). Routine use of prophylactic antiemetics ranged from 6% t~ecolo~) to 23% (urology). Benzodiazepines were administered to manage anxiety and restlessness at 63-72% of departments. Addiction was believed to be a frequent problem of opioid therapy by 8% of physicians, 18% believed that addiction may be seen sometimes, the remaining physicians considered this to be a rare or a non-existent problem. About 9% of respondents believed that respiratory depression may sometimes be associated with opioid therapy in terminal cancer patients. Tolerance development About 1 in 4 (23%) physicians believed that tolerance to opioids will occur irrespective of type of opioid or the route of its administration. Almost all physicians increased morphine doses by 25-50% when the originally prescribed doses did not provide adequate analgesia. However, 62% of physicians responded that they did not know if the type of opioid or the route of opioid administration affected development of tolerance.

TABLE IV INCIDENCE OF VARIOUS SYM~OMS

IN THE LAST 2 WEEKS OF LIFE IN CANCER PATIENTS

Data based on nurse responses from 203 departments which treat terminal cancer pain. Symptom

Anxiety, restlessness

Depression Tiredness Breathing difficulties Bedridden due to pain Sleep disturbance due to breakthrough pain Sleeping difficulties Nasogastric or iv. nutrition Poor appetite Swallowing difficulties Dryness of mouth

Percentage of nurses who believe the symptom is seen frequently

the symptom is seen sometimes

the symptom is seen rarely or never

49% 24% 83% 15% 30% 14% 15% 50% 92% 32% 73%

46% 63% 17% 45% 32% 66% 58% 33% 6% 48% 26%

5% 13% 40% 38% 20% 27% 17% 2% 20% 1%

174

therapy. Our study shows that 12% of nurses believe that under-reporting of pain and stoic acceptance of pain by patients as an inevitable consequence of cancer are a frequent occurrence while 52% believe that this is not an unusual situation. Even if the cancer patient was experiencing severe pain or breakthrough pain 35% of nurses were not allowed to give additional doses of opioids without consulting a physician. About 45% of nurses had no maximum limits and could administer the amount of opioids they considered necessary to treat terminal cancer pain. The frequency of various symptoms in the last 2 weeks of life is shown in Table IV. The most common problems were poor appetite, tiredness and dryness of mouth.

%

Discussion

0 Pal” w*hnon

~IY

dtug

Mb-Y tdnllqlm”

Psycho

Gr,t

FgF

WHO “WdW

*rulgrlc druga’

Fig. 5. Percentage of physicians and nurses who believe they have inadequate knowledge (see text) regarding various aspects of cancer pain management. Numbers on bars represent the number of respondents for every question. * Only nurse responses. ** Includes epidural and i.t. opioids and PCA.

Educational needs It should be realized that it is difficult to assess the

knowledge level of individuals by their responses to a few questions. In this study the questions about knowledge level were formulated as ‘educational need’; low educational need was interpreted as high level of knowledge and vice versa. In general the physicians and nurses considered their knowledge levels about pain management as relatively low; this was particularly valid for educational needs regarding newer analgesic drugs, pain evaluation techniques and about drug delivery systems such as PCA and spinal opioids (Fig. 51. In general this educational need was lower among physicians as compared to nurses. A ‘pain expert’ was available at 83% of departments, and 70% of departments frequently cooperated with this expert. Nurse questions

According to 47% of nurses loneliness and according to 66% of nurses excessive anxiety were a frequent cause of inadequate pain relief by analgesics. Only 50% of nurses responded to the question about the amount of their working time spent on pain management and on how much additional time they believed would be necessary to improve pain treatment. On an average the nurses spent 28% of their working time on duties related to pain treatment. The nurses estimated that they would need to increase this time by an average of 24% to provide optimal pain treatment for the terminal patient. Almost all nurses believed that family members were rarely concerned about addiction risks with opioid

It is emphasized that this survey assesses the physicians’ and nurses’ opinions, it does not provide accurate data about the patterns of practice. Our study has limitations such as (a) only one physician and one nurse surveyed per department, (b) no independent validation that these persons truly are ‘experts’ or truly answered questions in terms of departmental policy rather than personal knowledge and (c) no prior validation of our questionnaire. The questionnaire response rate from physicians and nurses was quite high. In the opinion of the physicians, about 10% of patients (3767 patients) who were being treated at home for cancer during a l-year period found it necessary to seek admission because of severe pain. Our survey shows that almost all Swedish physicians (86-100%) follow the WHO guidelines for use of analgesics in the management of cancer pain. However, it appears that many Swedish physicians are not aware of the analgesiapotentiating effects of combining peripherally and centrally acting analgesics. Similar results have been reported by other workers (Bruera et al. 1989). Thus, the most common comment by nurses was that many physicians terminated non-opioids when they started administration of strong opioids for the management of terminal cancer pain. Pain in spite of treatment

In the opinion of about 30% of Swedish physicians most cancer patients in active therapy have significant pain at the time of their deaths. Thus the situation is similar to that reported by others (Hardy et al. 1977; Charap 1978; Oster et al. 1978; Cleeland 1981; Bonica 1986). Up to 78% of physicians and nurses believe that periodic severe pain is common in terminal cancer patients, particularly in patients suffering from uterine and pancreatic cancer. Many nurses would like to have more time to be able to provide adequate pain relief. In spite of departmental consensus allowing unrestricted amounts of opioids, Swedish physicians and

175

nurses do not appear to be particularly liberal with these drugs. The maximal oral morphine daily dose administered by a nurse was 360 mg and by a physician 500 mg. This may have contributed to inadequacy of pain relief; however, our data does not permit us to draw such conclusions. Pain evaluation

Most patients with advanced cancer have more than one type of pain (Twycross 1984). The incidence and severity of cancer pain is related to the location and pathophysiology of the primary tumor or its metastases (Foley 1979; Bonica 1980). The site, character, intensity, quality and variation of each pain should therefore be evaluated because effective treatment regimen is best developed when the etiology and intensity of pain are assessed. The use of antidepressants, anticonvulsants or steroids for neuropathic pain, interventional radiotherapy and palliative surgery for visceral pain and NSAIDs with radiotherapy for bone metastasis pain is gaining increasing acceptance (Levy 1985). Thus the importance of differentiating approach to pain treatment has been emphasized so that different types of pain are managed appropriately in a specific manner rather than by treatments based only on pain intensity (Walsh 1983; Feinman 1985; Swerdlow 1986; Jorgensen et al. 1990; ArnCr 1991). Although there are some centers in Sweden where such pain evaluation is undertaken the majority of departments treat cancer pain on the basis of history and clinical assessment only. In general physicians in Sweden do not use VAS or other instruments for accurate measurements of pain intensity nor do they use pain questionnaire or body image pain charts which may be valuable as an aid for evaluation of underlying mechanism of pain and as a baseline for future reference. Routes of opioid administration

The order of preference for opioid administration in Sweden is oral > intermittent S.C. injection > intermittent i.m. injection > continuous i.v. infusion. It is generally accepted that morphine should be administered as a continuous S.C. or i.v. infusion to avoid the pain and discomfort of repeated injections. If intermittent S.C.injections are used, the technique practiced in most institutions in the USA should be introduced in Sweden. With this technique morphine is administered through a butterfly needle inserted under the skin. The needle can be left in place for many days. In appropriate cases, the PCA technique will provide excellent analgesia. The S.C.infusion technique was used by only 4-16% of physicians. Interestingly pain management by morphine was better for home care as compared to hospitalized patients. The order of preference for home care patients was oral > spinal opioids > S.C. morphine infusions.

Adjuvant therapy

It is generally agreed that the most important and consistently encountered adverse effects of morphine therapy are constipation and nausea, which must be aggressively treated and prevented. Prevention is especially important in the coincidental pain of constipation. According to WHO recommendations all patients treated with opioids should receive prophylactic laxatives (WHO 1986). Prophylactic antinauseants are recommended by some for patients being treated with opioids (Levy 1985; WHO 1986). It is interesting to note that treatment strategies for specific symptoms such as constipation and vomiting vary considerably between different specialities. The most common adjuvant drugs in Sweden are benzodiazepines; about 50% of terminal cancer patients received diazepam. However, physicians in Sweden are very restrictive with the use of antidepressive agents and neuroleptic agents such as haloperidol. This is noteworthy in view of our data showing that 24% of nurses believe that depression and 49% of nurses believe that anxiety are frequent symptoms; 63% of nurses believe that depression is not unusual. Management of neuropathic and bone metastasis pain

The role of opioids in the management of neuropathic pain is somewhat controversial. It has been shown by some that certain forms of chronic neuropathic non-malignant pains are resistent to opioid therapy (Am& 1991). Others believe that a neuropathic mechanism may reduce opioid responsiveness but does not result in an inherent resistance to these drugs (Portenoy et al. 1990). However, it is generally agreed that cancer patients should not be deprived of an open trial of the analgesic effectiveness of opioids (Am& et al. 1989; Portenoy et al. 1990; Wall 1990). It is also well accepted that adjuvant drugs such as corticosteroids (Hanks 1988; Payne 1989; Epstein et al. 1991), anticonvulsants (Swerdlow 1986; Epstein et al. 19911, antidepressives (Walsh 1983; Feinman 1985; Hanks 1988; Payne 1989; Epstein et al. 1991) and alternative techniques such as regional anesthetic blocks (Hanks 1988; Payne 1989; ArnCr et al. 1990) have a role in the management of neuropathic pain which is perhaps the most difficult pain to treat in cancer patients (Hanks 1988; Payne 1989). About 60% of Swedish physicians treat neuropathic pain with morphine alone. It must be emphasized that there are some exceptions; thus about 20-30% of oncologists and gynecological oncologists use antidepressants and about 30% of pulmonologists and urologists and 80% of oncologists use corticosteroids as adjuncts to morphine for the management of neuropathic pain. Bone metastases pain is also managed differently by different specialists. Thus about 50% of gynecologists use NSAIDs while 100% of oncologists use NSAIDs

176

for this type of pain. Similarly use of steroids varied from 10% in gynecological oncology departments to 60% in pulmonary medicine departments. (In Sweden gynecological oncology is a separate speciality.) Patients with gynecological cancer may be admitted arbitrarily to either gynecology or gynecological oncology departments). It is also interesting to note that patients with gynecological cancer are four times more likely to receive radiotherapy for their bone metastasis pain if they happen to be admitted to gynecological oncology as compared to regular gynecology departments (Fig. 5). Knowledge level of physicians and nurses

As mentioned earlier it is difficult to evaluate knowledge of individuals on the basis of such surveys. Two points are emphasized: (1) low ‘educational need’ was interpreted as high knowledge level and vice versa and (2) the physicians and nurses answering the questions were considered most knowledgeable by their chairman and head nurses. It has been suggested that the reason for the current inadequacy in the management of cancer pain is more due to improper application of current knowledge rather than lack of knowledge regarding pain mechanisms and treatment modalities (Levy 1985). However, our study shows that a significant number of Swedish physicians and nurses admit that they have inadequate knowledge of pain evaluation techniques, newer analgesic drugs and newer drug delivery techniques. Although our survey methodology was different from that of Jorgensen et al. (1990) in Denmark and Vainio (1988) in Finland, our conclusions regarding inadequate knowledge among physicians and nurses are similar. Our conclusions are also consistent with findings of the expert group of Swedish National Board of Health and Welfare that there is a scope for improvement in many areas of pain management in the terminally ill cancer patient (Am& et al. 1989). Although national guidelines are available their clinical application seems to be unsatisfactory.

Conclusion

Our study shows that about 10% of cancer patients seek hospital admission because of severe pain. Our study also shows that the Swedish physicians follow many but not all of the principles of accepted pain management. They follow the analgesic ‘ladder’ principle recommended by WHO, prefer the oral route of morphine administration, prescribe opioid analgesics on regular ‘by-the-clock’ basis and in unrestricted doses (although the maximum daily doses reported were not particularly impressive). However, there are many deficiencies. (a) The achievable goal of effective pain relief

for all hospitalized terminal patients is far from reality in Sweden. Up to 78% of physicians and nurses believe that periodic severe pain is common in terminal cancer patients. About 30% of physicians believe that all patients have moderate to severe pain at the time of their death. In spite of severe pain 35% of nurses were not allowed to give additional doses of opioids without consulting a physician. (b) In general, physicians do not evaluate different pain types nor do they use any instruments for measurement of pain intensity. (c) It is noteworthy that physicians choose intermittent S.C. or i.m. injections of morphine rather than continuous S.C. or i.v. infusion or intermittent S.C.injection through an indwelling butterfly needle when oral morphine is ineffective or when patients are unable to take it due to nausea or vomiting. (d) A large number of physicians do not prescribe routine prophylactic laxatives and an even larger number do not prescribe prophylactic antiemetics. (e) Depending on the speciality of the treating physician the treatment modalities for specific painful conditions such as neuropathic pain and bone metastasis pain differ greatly in Sweden. (f) Swedish physicians are very restrictive in prescribing antidepressive and anxiolytic medication. There is a lack of uniform pattern of prescription of these drugs. Furthermore most physicians seem to be unaware of the analgesia potentiating properties of antidepressants. (g> Although morphine is a good choice for a majority of cancer patients, Swedish physicians and nurses who treat terminally ill cancer patients have to be made aware that morphine is not a panacea. They should be made to realize that there are some types of pain especially neuropathic pain and generalized bone metastasis pain which may respond poorly to opioids alone but such pain may respond much better when adjuvant drugs such as NSAIDs, corticosteroids, antidepressants, anticonvulsants or other techniques such as radiotherapy and regional blocks are used. (h) Many Swedish physicians and nurses admit that they have inadequate knowledge about newer analgesic drugs, pain evaluation techniques and newer drug delivery systems. There is a need for active education of physicians and nurses. Greater efforts should be made to persuade physicians to better follow Health Board guidelines. There is also a need for better patient information. Stoic acceptance of pain as an inevitable consequence of cancer should be discouraged.

Acknowledgements

This project was financially supported by Grebro County Council, Lions Cancer Research Fund, Kabi Pharmacia, Syntex, Karo Bio Nordic, Meda AB and Lundbeck. We thank all the chairmen and head nurses of participating departments for their cooperation and

177

support. We also thank Gunnar Westman, M.D., Ph.D., Chairman, Department of Oncology, &ebro, Bengt Wadman, M.D., Ph.D., Member of WHO Advisory Panel on Cancer, &ebro, Bertil Widman, M.D., Ph.D., Swedish Health Board, Stockholm, Henny Olsson, CRN, Interim Professor, Center for Caring Sciences, Grebro, Elisabeth Killander, CRN, University Pain Center, Linkoping, Lars Thaning, M.D., Senior Consultant, Pulmonary Medicine, Grebro and Ann-Marie

Akerling, CRN, Oncology Department, for their help in designing the questionnaire. We express our appreciation of the following chairmen in 6rebro: Jan-Erik Johansson, M.D., Ph.D. (urology), Lars Forslin, M.D., Ph.D. (gynecology), Lennart Domellof, M.D., Ph.D. (surgery) and Bo Frankendal, M.D., Ph.D. (gynecological oncology). Finally we thank Ing-Marie Dimgren for secretarial help.

APPENDIX

THE FOLLOWING

QUESTIONS

WERE ASKED

Evaluation of pain

(1)

Do you routinely use a visual analog scale (VAS) or any other instrument to measure pain intensity?

(2a)

Do you use any special pain evaluation form? (A body image pain chart showing the characteristic features of nociceptive, visceral and neuropathic pain was provided in the questionnaire. Such a pain chart showing body image is used in many pain centers; it allows the patient to localize and mark the type of pain sensation on this figure.)

(2b)

Do you rely on history and physical examination alone for cancer pain evaluation or do you use some kind of pain questionnaire?

Pharmacological management (3)

Do you follow the WHO ‘ladder’ principle for cancer pain management? problems with the ‘ladder’ treatment?

(4)

How do you inform the patients and families before starting opioid therapy? written information (brochure), (c) both?

(5)

In your opinion, how frequently do terminal cancer patients experience pain during the last 2 weeks of their lives in spite of ongoing treatment? Is this pain constant or periodic? What type of cancer pain is most difficult to treat in your practice?

(6)

Which of the following routes of morphine administration are used at your department? (The physicians and nurses were asked to select from the following alternatives.) (a) oral (tablets, mixture slow-release tablets), (b) rectal, (cl subcutaneous, (d) intramuscular, (e) intravenous, (f) patient-controlled analgesia (PCA), (g) epidural or intrathecal, (h) various combinations of these routes

(7)

Do you administer opioids at fixed intervals or ‘on demand’?

(8)

How frequently do patients in your department receive supplementary medication such as: (a) sedatives, (b) antidepressives, cc> hypnotics? (To avoid confusion, trade names of the most common drugs in each category were provided.) What are the indications for these drugs at your department?

(9)*

How frequently do you employ the following drugs or techniques for the management of predominantly neuropathic pain? (a) antiepileptics, (b) steroids, (c> opioids, (d) antidepressives, (e) radiation therapy, (f) nerve blocks, (g> TENS, (h) acupuncture.

(lo)*

In patients with skeletal metastases how frequently do you enploy the following methods for pain management at your department? (a) radiation therapy, (b) NSAIDs, (cl steroids, (d) opioids, (e) other drugs or techniques.

(11)

Do you have a maximum opioid dose limit at your department when treating patients with terminal cancer pain? What is the maximum amount of opioid you have administered to such a patient?

If yes, have you encountered (a) verbal information,

any (b)

178

Adverse effects

(12)

In your opinion how and other opioids? (respiratory rate < (dose escalation not

frequently do terminal cancer patients experience the following side effects of morphine (a) nausea, (b) vomiting, (c) constipation, (d) sedation, (e) respiratory depression lO/min, cyanosis, breathing problems), (f) hallucinations, (g) euphoria, (h) tolerance related to increased pain associated with disease progression), (i) dependence?

Tolerance development

(13)

What is the routine at your department when you increase the dose of morphine due to inadequate analgesic response? Increase dosage by about (a) 25%, (b) 26-50%, (c) 51-75%, (d) 76-lOO%, (e) other routine.

(14)

Do you believe that tolerance administration of opioid?

(15)

If providing analgesia to a terminal cancer patient is becoming difficult, would you increase opioid doses even if the risks of (a) respiratory depression, (b) disorientation are very high?

(16)

Do you routinely prescribe prophylactic morphine and other opioids?

(17a)

If your patients require medication because of anxiety, which of the following drugs would you select? (a)benzodiazepines, (b) neuroleptics, (c) antidepressives, (d) others.

(17b)

What is your prescribing routine? (a) on demand, (b) regular administration,

development

is related

(a) antiemetics,

to (a) type of opioid

used, (b) the route of

(b) laxatives when treating cancer patients with

e.g., 2-3 times/day.

(18) * In your practice how frequent is the use of the following techniques for the management of terminal cancer pain? (a) radiation therapy, (b) cytostatic drugs, (c) nerve blocks, (d) epidural and intrathecal opioids, (e) neurosurgery, (f) hormone therapy, (g) acupuncture, (h) hypnosis, (i) other non-pharmacological methods. What are the indications for using these techniques? (19) * Of the following routes of morphine administration for ‘home care’ patients, which is your order of preference (l-5)? (a) oral, (b) continuous subcutaneous infusion, (c) intermittent injection of epidural or intrathecal opioids, (d) continuous infusion of epidural or intrathecal opioids, (e) others. (20)

During the period January 1 to December 31, how many patients were admitted to your department treatment of terminal cancer pain only?

for the

Educational needs regarding pain management (21)

What is the level of knowledge in your department regarding the following? (The physicians and nurses were asked to select one of the following three alternatives: ‘high level’, ‘average level’ or ‘low level’.) (a) WHO ‘ladder’, (b) pharmacokinetics and pharmacodynamics of the most common analgesics, (c) adverse effect profiles of the most common analgesics, (d) difference between tolerance, dependence and addiction, (e) types of pain (somatic, visceral, neuropathic), (f) ‘newer’ drugs (slow-release morphine, sublingual buprenorphine, transdermal fentanyl patch, (g) ‘newer’ drug delivery techniques (PCA, epidural and intrathecal opioids alone or in combination with local anesthetics), (h) non-pharmacological techniques (acupuncture, hypnosis).

(22)

Do you have a ‘pain center’ or a ‘pain specialist’ at your hospital? If so, how frequently consulted?

(23)

Which of the following specialities do you consult on a regular basis when pain management in the terminal cancer patient is becoming difficult? (a) anesthesiology, (b) neurosurgery, (c) neurology, (d) clinical pharmacology, (e) rehabilitation, (f) others

is this specialist

(All of the above questions, except those marked with asterisks, were common to physicians and nurses. Questions with asterisks, i.e., questions 9, 10, 18, 19, were physician questions only. Additionally the following questions were asked of the nurses only.) (24)

In spite of apparent adequate analgesic therapy, patients may not be ‘pain free’ due to anxiety and loneliness. In your opinion, how frequent are (a) anxiety, (b) loneliness the cause of inadequate analgesia?

(25)

What percentage of your total time is used for providing pain relief to terminal cancer patients? Do you think that you need more time to provide satisfactory analgesia? If so, how much more?

179

(26)

How often do (a) cancer patients themselves or (b) family members refuse morphine treatment addiction risk?

(27)

In your opinion, what percentage of patients accept their pain stoically (‘pain has to be accepted as part of cancer’) and do not report pain?

(28)

If a terminal cancer patient is in severe pain in spite of ongoing treatment (break-through pain), how much morphine are you allowed to give before contacting your physician? (a) as much as is necessary, (b) increase usual dose by 25%, Cc) up to 50%, Cd) up to lOO%, (e) not allowed to give additional morphine without physician’s orders.

(29)

In your opinion, how common are the following problems during the last 2 weeks of a cancer patient’s life? (a) bedridden due to pain, (b) sleeping problems, Cc) disturbed sleep due to break-through pain, (d) patient requires i.v. or nasogastric feeding, (e) lack of appetite, (f) dry mouth, (g) restlessness, anxiety, (h) depression, (i) nausea, vomitting, (j) constipation, (k) diarrhoea, (1) breathing difficulties?

References Art&, S., Differentiation of Pain and Treatment Efficacy (thesis), Stockholm, 1991, ISBN 91-628-0366-2. Art&, S. and Meyerson, B.A., Opioid sensitivity of neuropathic pain. A controversial issue. In: G. Guilbaud and J.M. Besson (Eds.), Lesions of Primary Afferent Fibers as a Tool for the Study of Clinical Pain, Elsevier, Amsterdam, 1991, in press. Arntr, S., Bertler, A., Bolund, C. and Meyerson, B.A., The management of pain in the terminal stage of life, The Swedish National Board of Health and Welfare, Allmlnna Fiirlaget, 1 (1989) 47 (Swedish). A&r, S., Lindblom, U., Meyerson, B.A. and Molander, C., Prolonged relief of neuralgia after regional anesthetic blocks. A call for further experimental and systematic clinical studies, Pain, 43 (1990) 287-297. Bonica, J.J., Past and current status of pain research and therapy, Semin. Anesth., V (1986) 82-99. Bonica J.J., Cancer pain. In: J.J. Bonica (Ed.), Pain, Raven Press, New York, 1980, 325, pp. Bruera, E., Brenneis, C., Michaud, M. and MacDonald, R.N., Influence of the pain and symptom control team (PSCI) on the patterns of treatment of pain and other symptoms in a cancer center, J. Pain Sympt. Manag., 4 (1989) 112-116. Charap, A.D., The knowledge, attitudes and experience of medical personnel treating pain in the terminally ill, Mt. Sinai J. Med., 45 (1978) 561-580. Cleeland, C., The impact of pain on patients with cancer, Cancer, 54 (1981) 2635-2641. Epstein, J.B., Schubert, M.M. and Scully, C., Evaluation and treatment of pain in patients with orofacial cancer: a review, Pain Clin., 4 (1991) 3-20. Feinman, C., Pain relief by antidepressants, possible modes of action, Pain, 23 (1985) l-8. Ferrer-Brechner, T. and Ganz, P., Combination therapy with ibuprofen and methadone for chronic pain, Am. J. Med., 77 (1984) 78-83. Foley, K.M., Pain syndromes in patients with cancer. In: J.J. Bonica and V. Ventafridda (Eds.), Advances in Pain Research and Therapy, Vol. 2, Raven Press, New York, 1979, p. 59. Foley, K.M., The treatment of cancer pain, N. Engl. J. Med., 313 (1985) 84-95. Greenwald, H.P., Bonica, J.J. and Bergner, M., The prevalence of pain in four cancers, Cancer, 60 (1987) 2563-2569. Hanks, G.W. (Ed.), Pain and Cancer, Cancer Surveys, 7 (1988) l-222.

because of

Hardy, T.K. and Pritchard, R.I., Physical distress suffered by terrninal cancer patients in hospital, Anaesthesia, 32 (1977) 647-649. Jorgensen, L., Mortensen, M.B., Jensen, N.H. and Eriksen, J., Treatment of cancer pain patients in a multidisciplinary pain clinic, Pain Clin., 3 (1990) 83-89. Kaiser, C.B., Hospice inpatient care: characteristics of an institution and its patients, Conn. Med., 48 (1984) 146-151. Levy, M.H., Pain management in advanced cancer, Sem. Oncol., 12 (1985) 394-410. Marks, M.R. and Sachar, E.J., Undertreatment of medical inpatients with narcotic analgesics, Ann. Intern. Med., 78 (1973) 173-181. McKegney, F.P., Bailey, L.R. and Yates, J.W., Prediction and management of pain in patients with advanced cancer, Gen. Hosp. Psychiat., 3 (1981) 95-101. Murray Parkes, C., The dying patient. Terminal care: home, hospital or hospice?, Lancet, i (1985) 155-157. Norton, W.S. and Lack, S.A., Control of symptoms other than pain. In: R.G. Twycross and V. Ventafridda (Eds.), The Continuing Care of Terminal Cancer Patients, Pergamon, Oxford, 1980, 167 PP. Oster, M.W., Vizel, M. and Turgeon, L.R., Pain of terminal cancer patients, Arch. Intern. Med., 138 (1978) 1801-1802. Payne, R., Cancer pain: anatomy, physiology and pharmacology, Cancer, 63 (1989) 2266-2274. Portenoy, R.K., Foley, K.M. and Inturrisi, C.E., The nature of opioid responsiveness and its implications for neuropathic pain: new hypotheses derived from studies of opioid infusions, Pain, 43 (1990) 273-286. Swerdlow, M., Anticonvulsants in the therapy of neuralgic pain, Pain Clin., 1 (1986) 9-19. Twycross, R.G., Incidence of pain, Clin. Oncol., 3 (1984) 5-15. Vainio, A., Treatment of terminal cancer pain in Finland. A questionnaire survey, Acta Anaesthesiol. Stand., 32 (1988) 260-265. Vainio, A., Practising physicians’ experiences of treating patients with cancer pain, Acta Oncol., 28 (1989) 177-182. Wall, P.D., Neuropathic pain (Editorial), Pain, 43 (1990) 267-268. Walsh, T.D., Antidepressants in chronic pain, Clin. Neuropharmacol., 6 (1983) 271-295. Walsh, T.D. and Saunders, C.M., Hospice care: the treatment of pain in advanced cancer, Recent Results Cancer Res., 89 (1984) 201-211. Wilke, E., Some problems in cancer management, Proc. R. Sot. Med., 67 (1974) 100-1005. World Health Organization, Cancer pain relief, Geneva, WHO, 1986.