Deficiencies with CAGE-AID questionnaire in identifying nonmedical opioid use—a report of two cases

Introduction

Opioids are frequently used in palliative care for cancer-related pain. While opioids are necessary for symptom management, opioid misuse is a growing concern (1). Recent studies have found that a substantial minority of patients with cancer-pain receiving opioids are at risk for nonmedical opioid use (NMOU) (2,3). NMOU is defined by behaviors such as increased use of opioids without any clinical evidence for disease progression, frequent unscheduled outpatient or emergency room visits requesting early refills of opioids, seeking opioids from multiple outside providers commonly referred to as “doctor shopping”, reporting stolen or missing opioids without filing a police report, using opioids for recreational purposes and family concerns over opioid use (2,4-6). Some of the known risk factors for NMOU behavior include a history of alcohol use or substance use disorder (SUD), younger age (<45 years), male sex, history of mental health disorder, and low socioeconomic status (2,5). SUD may be considered as a sever form of NMOU.

Major pain guidelines encourage providers to screen patients for NMOU and SUD before initiating opioids by utilizing screening questionnaires (7). The CAGE questionnaire is widely used to screen adults for excessive alcohol use (5,8). It consists of the following questions: Have you ever felt you should cut down on your drinking? Have people annoyed you by criticizing your drinking? Have you ever felt bad or guilty about your drinking? Have you ever had a drink first thing in the morning to steady your nerves or get rid of a hangover (eye-opener)? Later, the CAGE questionnaire was modified to include illicit drug use. The CAGE Adapted to Include Drugs (CAGE-AID) questionnaire substitutes’ ‘drink’ with ‘drink or drugs’. The patients answer the questions with a yes or no response, and each question with a yes response is assigned one point, a score of ≥2 out of 4 is considered positive for being at risk of SUD or NMOU (8-11).

Despite its wide use, the CAGE-AID questionnaire may not accurately identify patients at risk for NMOU (5). We present two cases wherein the results of the CAGE-AID misidentified the risk for NMOU. We present this article in accordance with the CARE reporting checklist (available at https://apm.amegroups.com/article/view/10.21037/apm-23-192/rc).

Case presentation

Patient 1

A man in his 60s with metastatic prostate cancer was admitted to the hospital for low back pain. Palliative care (PC) was consulted to manage his uncontrolled cancer-associated pain. At his initial consultation, the patient was using gabapentin 300 mg three times a day, methocarbamol 500 mg twice a day, hydromorphone 2 mg every 4 hours as needed, and tramadol 50 mg every 6 hours as needed without any relief. His home morphine equivalent daily dose (MEDD) was 40 mg. On examination, he appeared to be in acute distress with a reported pain intensity of 10/10 and he exhibited lumbar spinal tenderness. A recent magnetic resonance imaging (MRI) showed increased spinal metastasis. Previously, his CAGE-AID questionnaire (3 of 4 questions) was positive for alcohol consumption, indicating that he was at risk for NMOU. He was initiated on a continuous morphine infusion, which was titrated up over the next 48 hours to a MEDD of 75 mg.

During the hospital course, he continued to express uncontrolled pain. He underwent a rotation to a continuous hydromorphone infusion with a MEDD of 100 mg. The PC providers were cautious when titrating the opioids due to his positive CAGE-AID status and concerns for NMOU. A repeat spine MRI showed further disease progression. He was subsequently initiated on oral methadone 5 mg every 8 hours and intravenous dexamethasone with gradual improvement of the pain. Our PC providers discussed the CAGE-AID results with his wife, who expressed that the patient never had any alcohol or illicit drug-related problems. The patient consumed alcohol only during social events in limited amounts. In addition, the patient perceived his rare use of alcohol negatively and felt guilty that it might have contributed to his cancer diagnosis. After two weeks in the hospital, he was discharged home on methadone 10 mg twice daily and 5 mg in the afternoon, and oral hydromorphone 4 mg every 4 hours as needed for breakthrough pain. His MEDD at the time of discharge was 165 mg (a conversion factor of 5 was used to calculate MEDD for methadone and hydromorphone). In the following two months, his pain remained controlled with a stable MEDD.

Patient 2

A male patient in his 40s with metastatic cholangiocarcinoma and biliary obstruction with two external biliary drains was admitted with intractable nausea, vomiting, and uncontrolled pain. He had been hospitalized at another facility for two weeks and was discharged 24 hours prior to his admission to our hospital. His medical history was significant for multiple hospitalizations over the past two months for similar complaints. He was formerly a construction worker who was currently unemployed. He depended on his wife for financial support for the family. PC was consulted for cancer associated pain. His CAGE-AID screening questionnaire was negative (score of 0/4). His physical examination was positive for mild right upper quadrant tenderness, and the patient reported extremely high scores (10/10) of both pain and nausea. On review of the prescription drug monitoring program (PDMP), he was previously prescribed fentanyl and methadone. However, the patient expressed side effects such as having hallucinations and sever nausea with both medications. At the time of PC visit, patient was not allowed to take anything by mouth except for small sips of water with medications, he was initiated on as-needed doses of oral hydromorphone 4 mg every 4 hours for moderate pain, intravenous (IV) hydromorphone 1 mg every 3 hours for severe pain, IV metoclopramide 10 mg, and IV ondansetron 8 mg to help with his nausea.

Over the next 3 days, the patient refused to take oral hydromorphone and required multiple doses of IV hydromorphone, accounting for a MEDD of approximately 80 mg. During the PC team visit he had hypoactive bowel sounds, was tolerating a liquid diet, was able to ambulate independently in the room, and appeared to be in no acute distress. He denied any nausea or vomiting and had 3 bowel movements in the past 24 hours. He was encouraged to take oral hydromorphone as needed and using IV hydromorphone only for severe pain not well controlled with oral hydromorphone. Shortly afterward, the patient refused all oral medications, due to complaints of nausea. This included opioids and antiemetics. Subsequently, he requested and received IV hydromorphone multiple times in the next 24 hours.

Moreover, he demanded the nurses to rapidly push the IV hydromorphone into his central line rather than the standard slow 2-minute IV administration. He also requested that IV hydromorphone not to be diluted with normal saline prior to administration, raising concerns for NMOU. He began watching the clock waiting for his next available dose of IV hydromorphone and was irritated with the staff when the IV hydromorphone was not administered as per his demands.

New imaging was consistent with stable disease without any changes to support the high severity of pain and nausea. When concerns were raised about his current use of IV opioids, the patient became defensive and argumentative and began negotiating the amount and frequency of IV hydromorphone he could receive. Our team addressed the importance of taking oral opioids for pain and discontinued IV hydromorphone. We offered a nerve block to help with the pain, along with a referral to integrative medicine for consideration for acupuncture and oncologic massage. However, he left the hospital against medical advice once it was clear that he would not receive any further IV hydromorphone. All procedures performed in this study were in accordance with the ethical standards of the institutional and national research committee(s) and with the Helsinki Declaration (as revised in 2013). Publication of this case report was waived from patient consent according to the university of Texas MD Anderson cancer center institutional review board.

Discussion

NMOU is a common entity among patients with cancer pain that is frequently underdiagnosed (2,12) and carries a high rate of morbidity and mortality (13). Identifying NMOU behavior promptly is essential to prevent serious complications (5). Many screening tools, such as the CAGE-AID questionnaire, the Screener and Opioid Assessment for Patients with Pain questionnaire (SOAPP), and the Opioid Risk Tool (ORT), are used to identify patients at risk for NMOU or SUD, and are recommended by multiple pain guidelines to be used before initiating opioids. Although the administration of these tools is quick and effective, the results must be interpreted with caution as these are meant for screening rather than diagnosing NMOU (5).

In the first case, the patient had a positive CAGE-AID score with clear evidence of cancer progression in the painful area, and no evidence of NMOU. The providers were concerned about the positive CAGE-AID score as a risk-factor and that the reported uncontrolled pain and demand for frequent opioids was related to NMOU, delaying the escalation of opioids which resulted in poor pain control. Despite only an occasional use of alcohol in social events, the patient answered “yes” on the CAGE-AID questionnaire due to his concern that even the rare consumption of small amounts of alcohol may have contributed to his cancer diagnosis. Once the opioid doses were appropriately increased, his pain was well controlled, and he improved his functional capacity.

The CAGE-AID questionnaire is highly sensitive but not specific (5). The results of screening tools must be interpreted with caution and should not supersede the clinician’s judgment. Clinicians should not rely solely on the CAGE-AID to identify patients at risk for NMOU. A positive CAGE-AID implies that the patient is at risk for NMOU but does not establish the diagnosis.

In the second case, the patient's demands for a specific opioid dose, route of administration, refusal to any oral medication, persistent pain despite opioid dose increases, and frequent hospitalizations for persistent pain without a clear etiology, were consistent with NMOU. In addition, he also refused interventions that could help pain control and left the hospital against medical advice. Our team did not unearth any previous history of alcohol or substance abuse. It is unclear if the exposure to opioids in the setting of cancer pain might have triggered his new NMOU behavior without any prior history of substance abuse. In such instances, the patient may have scored as negative on CAGE-AID as he might not have had any previous desire to cut down on alcohol or drugs, or feelings of guilt, nor need to use any substance as an eye-opener. The patient might also have been aware of such screening questionnaires from his previous hospitalizations and deliberately denied using alcohol and other recreational drugs.

Additionally, the patient’s male gender, young age, and poor socioeconomic status with his advanced malignancy are known risk factors for NMOU behaviors (2,5). His negative CAGE-AID did not exclude him from close monitoring by the PC team. He received extensive education regarding the safe use of opioids by the medical providers.

Finally, in rare instances, certain behaviors that can occur with undertreated chronic cancer-related pain, such as opioid pseudo-addiction, may lead to a presumption of NMOU in patients with a positive CAGE-AID score. Pseudo-addiction occurs when patients have improperly treated pain resulting in analgesic demands and mistrust that mimic addictive behaviors (14,15). It is important to correlate all the clinical findings and make appropriate adjustments to the opioids with a close follow-up clinical evaluation (14,15). One limitation of the CAGE-AID is that there is not much evidence to support its use to identify NMOU or SUD among patients with cancer diagnosis. More research is needed in validating screening tools for NMOU in this patient population.

Conclusions

Our cases highlight that the CAGE-AID questionnaire should be used only as a screening tool and not to diagnose NMOU. Clinicians should be aware that the implementation of such screening tools is mainly to identify the risk factors for NMOU. It is essential to interpret its results and other screening tools with caution and in conjunction with clinical findings, regular monitoring of PDMP, and conducting random urine drug screens. Management changes solely based on the CAGE-AID questionnaire might result in undertreating pain or worsening NMOU. More research is needed to determine the best practices of using screening tools for NMOU among patients with cancer pain.

Acknowledgments

Funding: This work was supported in part by the National Institutes of Health (Award Number 1UL1TR003167).

Footnote

Reporting Checklist: The authors have completed the CARE reporting checklist. Available at https://apm.amegroups.com/article/view/10.21037/apm-23-192/rc

Peer Review File: Available at https://apm.amegroups.com/article/view/10.21037/apm-23-192/prf

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://apm.amegroups.com/article/view/10.21037/apm-23-192/coif). Akhila Reddy serves as an unpaid editorial board member of Annals of Palliative Medicine from February 2022 to January 2024. The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. All procedures performed in this study were in accordance with the ethical standards of the institutional and/or national research committee(s) and with the Helsinki Declaration (as revised in 2013). Publication of this case report was waived from patient consent according to the university of Texas MD Anderson cancer center institutional review board.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Yennurajalingam S, Arthur J, Reddy S, et al. Frequency of and Factors Associated With Nonmedical Opioid Use Behavior Among Patients With Cancer Receiving Opioids for Cancer Pain. JAMA Oncol 2021;7:404-11. [Crossref] [PubMed]
2. Yennurajalingam S, Edwards T, Arthur J, et al. The development of a nomogram to determine the frequency of elevated risk for non-medical opioid use in cancer patients. Palliat Support Care 2021;19:3-10. [Crossref] [PubMed]
3. Zylla D, Steele G, Gupta P. A systematic review of the impact of pain on overall survival in patients with cancer. Support Care Cancer 2017;25:1687-98. [Crossref] [PubMed]
4. Yennurajalingam S, Edwards T, Arthur JA, et al. Predicting the risk for aberrant opioid use behavior in patients receiving outpatient supportive care consultation at a comprehensive cancer center. Cancer 2018;124:3942-9. [Crossref] [PubMed]
5. Arthur J, Bruera E. Balancing opioid analgesia with the risk of nonmedical opioid use in patients with cancer. Nat Rev Clin Oncol 2019;16:213-26. [Crossref] [PubMed]
6. Kim YJ, Dev R, Reddy A, et al. Association Between Tobacco Use, Symptom Expression, and Alcohol and Illicit Drug Use in Advanced Cancer Patients. J Pain Symptom Manage 2016;51:762-8. [Crossref] [PubMed]
7. Dowell D, Ragan KR, Jones CM, et al. CDC Clinical Practice Guideline for Prescribing Opioids for Pain - United States, 2022. MMWR Recomm Rep 2022;71:1-95. [Crossref] [PubMed]
8. O'Brien CP. The CAGE questionnaire for detection of alcoholism: a remarkably useful but simple tool. JAMA 2008;300:2054-6. [Crossref] [PubMed]
9. Ewing JA. CAGE questionnaire allows doctors to avoid focusing on specifics of drinking. BMJ 1998;316:1827. [Crossref] [PubMed]
10. Ewing JA. Detecting alcoholism. The CAGE questionnaire. JAMA 1984;252:1905-7. [Crossref] [PubMed]
11. SAMHSA 2001;Pages. Accessed at U.S. Department of Health & Human Services. Available online: https://store.samhsa.gov/sites/default/files/d7/priv/sma12-3581.pdf2001.
12. Amaram-Davila JS, Arthur J, Reddy A, et al. Managing Nonmedical Opioid Use Among Patients With Cancer Pain During the COVID-19 Pandemic Using the CHAT Model and Telehealth. J Pain Symptom Manage 2021;62:192-6. [Crossref] [PubMed]
13. Hser YI, Mooney LJ, Saxon AJ, et al. High Mortality Among Patients With Opioid Use Disorder in a Large Healthcare System. J Addict Med 2017;11:315-9. [Crossref] [PubMed]
14. Weissman DE, Haddox DJ. Opioid pseudoaddiction--an iatrogenic syndrome. Pain 1989;36:363-6. [Crossref] [PubMed]
15. Elander J, Lusher J, Bevan D, et al. Understanding the causes of problematic pain management in sickle cell disease: evidence that pseudoaddiction plays a more important role than genuine analgesic dependence. J Pain Symptom Manage 2004;27:156-69. [Crossref] [PubMed]