Keywords: cancer pain, analgesic prescribing, audit
The aim of this audit was to ascertain if analgesics were being prescribed according to the World Health Organization guidelines and to identify common errors in cancer pain management on oncology wards.
A retrospective chart audit was conducted. The charts of patients who had been admitted in the radiation oncology ward of the Christian Medical College for at least a week in 2001 (group A), and 2002 (group B), were screened. The charts of the first 25 patients with pain in each group were selected for the audit. Shorter admissions were excluded because most would have been for chemotherapy administration.
31 of the patients were men and 19 were women. The patient's age ranged from 19 to 68 years. Pain due to skeletal metastases was the commonest pain syndrome documented in 20 patients (40%).
Pain had been assessed and recorded in 88% of patients in 2001 and in all patients in 2002. There was an increase in pain recording in 2002 after the introduction of a pain-monitoring chart maintained by the nursing staff.
92% of patients in Group A and 96% in Group B received analgesics round the clock. The response to analgesics was recorded in 32% of charts in group A and 84% in Group B. 36% in Group A and 60% in Group B had extra prn analgesics prescribed if needed for breakthrough pain.
Non-opioids were not prescribed for all patients receiving opioids. Only 36% of patients in Group A and 60% in Group B were prescribed non-opioids along with Step 2 and Step 3 opioids. Kangarooing between non-opioids was not a common problem in either group.
The non-opioids used were ibuprofen, rofecoxib, ketrolac, diclofenac and paracetamol. The commonest non-opioid used was diclofenac: 28% in Group A and 36% in Group B, followed by paracetamol in 16% and 28% in Groups A and B respectively. There was a trend to increased use of rofecoxib in Group B (16%) as compared to Group A (4%).
All 50 patients had been prescribed opioids. Dextropropoxyphene was the commonest weak opioid and had been used in 76% of group A and 52% of group B. Tramadol was used in 7 patients in group B.
Buprenorphine was the commonest strong opioid in group A (10 patients) and oral morphine (6) in group B. The two patients who were given oral morphine in group A were given inadequate doses. One patient in group A and 2 in group B were prescribed transdermal fentanyl.
Adjuvant analgesics were commonly used. Corticosteroids were used in both groups. They were prescribed for pain, chemotherapy emesis, cerebral oedema and spinal cord compression. The other adjuvants used were tricyclic antidepressants and antiepileptics for neuropathic pain.
It was encouraging to note that pain had been recorded and regular analgesics prescribed for nearly all patients. There was an improvement in monitoring pain and in prescribing analgesics for breakthrough pain in group B. Nevertheless some fundamental errors were identified in both groups. [Table - 1]
16 patients in group A and 10 in group B were not given non opioids. Non-steroidal anti-inflammatory drugs (NSAIDs) play an important role in cancer pain. A prospective randomized controlled study showed that patients who received ketorolac in addition to morphine showed a better analgesia after a week in comparison to the group treated with morphine only. Morphine escalation was slower and the maximum morphine dose was lower in the group treated with ketorolac (Mercadante et al 2002).
NSAIDs are found to be effective in cancer related neuropathic pain. In a randomized double-blind trial done on 20 patients with malignant nerve pain, comparing naproxen 1500mg and slow-release morphine 60mg, daily for 1 week, followed by cross-over medication during the second week, it was found on the 7th day, that there was significant pain relief in 32% of patients in the naproxen group, and 21% in the morphine group. Patients on the morphine arm needed approximately twice as much paracetamol rescue than patients in the naproxen arm. This study favors the combination of both an anti-inflammatory drug and an opioid for symptomatic pain relief (Dellemjin et al 1994).
Oral morphine was very rarely used in 2001. The two patients who did receive oral morphine had inadequate doses. A patient with pain uncontrolled on 0.8 micrograms of buprenorphine (equivalent to 48 mg of oral morphine) was switched to 2.5 mg of morphine three times a day.
In a survey from France, et l reported that 50% of medical oncologists and 76% of primary physicians were reluctant to use morphine. In this survey older oncologists and women physicians were less likely to prescribe. Fear of addiction was the strongest predictor to prescribing morphine at any stage of the illness. Fear of tolerance, adverse effects, the constraints of prescribing forms and a poor 'public image' of morphine were cited as other reasons for not using morphine (Larue et al 1995)
Common side effects of analgesics should be anticipated and treated prophylactically. We found that laxatives had been used in 30 patients. But in half the cases in group A, bulk laxatives were used after patients developed constipation. Stimulant laxatives are generally recommended for opioid induced constipation. These were used in 8 patients in group A and 10 in group B. Antiemetics and gastroprotectors were also commonly used but it was difficult to assess if this was for coexistent symptoms, chemotherapy or for the adverse effects of analgesics.
This small audit has several limitations. Outpatients and short stay patients were not studied. It is possible that we may have missed patients in whom mild or moderate pain was not documented or addressed.The outcome of pain management was difficult to ascertain. We have not analysed the cost effectiveness of various combinations nor was it clear if all patients needed the various adjuvants that were prescribed.
Although we found that education resulted in increased use of morphine, appropriate laxatives and non-opioids, this was not true in all cases. Proformas and practical guidelines may help translate theoretical teaching into clinical practice. We have found that the introduction of a pain monitoring sheet has resulted in consistent improvement in nurse led pain monitoring. Sample prescriptions incorporating simple, effective and economical drugs may lead to more uniform improvements in analgesic prescribing by doctors in the ward.
[Table - 1]