Pain management for post-treatment survivors of complex cancers: a qualitative study of opioids and cannabis

Abstract

Aim:

We aimed to understand experiences with opioids and cannabis for post-treatment cancer survivors.

Patients & methods:

We conducted seven focus groups among head and neck and lung cancer survivors, using standard qualitative methodology to explore themes around 1) post-treatment pain and 2) utilization, perceived benefits and perceived harms of cannabis and opioids.

Results & conclusion:

Survivors (N = 25) experienced addiction fears, stigma and access challenges for both products. Opioids were often perceived as critical for severe pain. Cannabis reduced pain and anxiety for many survivors, suggesting that anxiety screening, as recommended in guidelines, would improve traditional pain assessment. Opioids and cannabis present complex harms and benefits for post-treatment survivors who must balance pain management and minimizing side effects.

As of January 2022, more than 18 million people in the USA have been diagnosed with cancer (survivors), and survivors are living longer (67% 5 year survival) than ever before [1,2]. However, up to 40% of cancer survivors experience at least moderate pain, which may persist or arise at any time after treatment, diminishing function and quality of life [3–7]. Head and neck (HNC) and lung (LC) cancer survivors experience a particularly high pain burden, as they often are treated with aggressive multimodal therapies that especially affect their aerodigestive tracts; in addition, pain can arise from prevalent psychosocial effects from cancer and its treatment and from comorbidities not directly due to cancer, which are common [8–32]. Chronic tobacco exposure remains the leading risk factor for HNC and LC, and survivors often have tobacco-related comorbid illnesses, such as pulmonary and cardiovascular disease [33–40]. Survivors with a history of smoking face a 5% per year risk of a new primary cancer [33,34,38,39]. In one study of patients receiving radiation therapy, HNC patients had higher pain intensity ratings than prostate and lung cancer patients, and LC patients had the most severe ratings of their worst pain [41]. Indeed, 42% of HNC and 25–30% of long-term LC survivors report ongoing pain [19,30,31,42].

In HNC, pain often arises from treatment. Common and potentially painful late effects of HNC include dry mouth, neck and shoulder dysfunction and lymphedema in the neck and face [27,43–46]. More debilitating late effects include destruction of the jaw, difficulty swallowing and difficulty opening the mouth [47–50]. In LC, late effects can include difficulty breathing, cough, shoulder dysfunction, fatigue and pneumonitis [16–19]. Pain among survivors of both cancers may continue for years; a cross-sectional study of HNC survivors (median 6.6 years from diagnosis) reported that 11.5% experience severe pain, and a review in LC survivors found that 25–30% of long-term (>5 years) survivors experienced pain [19,30,51]. Pain in LC and HNC survivors is strongly correlated with diminished quality of life and function [31,51,52].

The range of sources and presentations of pain makes these cancer survivors complex. Unfortunately, post-treatment care for cancer survivors is often fragmented and suboptimal. As the focus of care shifts from acute treatment to survivorship, there is little consensus regarding which providers should oversee post-treatment care [53]. Care utilization varies widely by the providers that survivors visit, and not visiting a primary care provider (PCP) is associated with lower quality of care of acute and chronic conditions [54–57]. Patients who receive initial pain management (with or without opioids and cannabis) during cancer treatment may no longer visit the same providers for pain management after treatment completion. The complexity of HNC and LC survivors may exacerbate care coordination problems and impede high-quality pain management.

The shifting policy landscape further complicates pain management. Cannabis use in the general population and among cancer patients during treatment is on the rise, with 12–21% of cancer patients reporting cannabis use in the past month [58–61]. Access to medical cannabis is broadening, and medical cannabis is legal in 38 states. However, with few trials of cannabis products in cancer, multiple reviews and guidelines have found insufficient evidence to recommend for or against cannabis use in addressing symptoms in cancer [62–66]. Furthermore, the extent of cannabis use – and experiences with cannabis – remain largely unknown among post-treatment cancer survivors, who continue to experience long-term symptoms of their cancer but may not be overseen in the oncology setting.

Similarly, opioids are an accepted mainstay of pain management for cancer, in appreciation that opioids can be an appropriate pain management strategy during treatment, for metastatic disease and at the end of life. Accordingly, clinical practice guidelines directly address recommended approaches for pain management during cancer [67–69]. However, cancer-related pain may persist after treatment, and only recently have clinical practice guidelines addressed this survivor population [3]. Recent studies indicate that long-term opioid use is higher among LC and HNC survivors than survivors of other cancers, and unsafe prescribing (e.g. co-prescribing with benzodiazepines or at high average daily doses) is concerningly common years after diagnosis [70–72].

Guidelines for pain management in cancer survivors cautiously recommend opioids and cannabis for subgroups with certain pain phenotypes, despite limited evidence regarding benefits and harms of both opioids and cannabis in survivor populations [3]. Both cannabis and opioids are commonly used during cancer treatment, and the extension of the use of these supportive care approaches into the post-treatment setting is not well understood.

Because the intersection of these topics (pain among vulnerable post-treatment survivors, cannabis use in a rapidly evolving social and legal context, and opioid use after cancer treatment completion) is largely unexplored, we selected a qualitative approach to identify themes for further quantitative research. Specifically, we conducted focus groups with complex cancer survivors to explore experiences and attitudes around pain management, including opioids and cannabis.

Methods

Participants & setting

We enrolled a convenience sample of survivors (identified by study champions or by medical record review and invited by email and phone) from one academic cancer center in New York City (where medical and recreational cannabis use is legal) and two affiliated urban community oncology hospitals – one in Connecticut, where both medical and recreational cannabis use is legal, and another in Florida, where only medical cannabis is legal. Eligible participants had completed treatment for stage I–IVA HNC or stage I or IIA non-small cell LC cancer 3–36 months prior, had no evidence of disease, and, during a telephone screening process, reported any post-treatment pain in the past year (or since treatment completion, if treatment ended more recently) from their cancer or its treatment.

Focus groups

Aligned with standardized focus group methodology, a qualitative methods expert (SC) led seven 90 min online focus groups (range 2 to 7 participants) with a semi-structured guide (Appendix A) asking participants about experiences of pain and pain management, including: the nature and timing of post-cancer pain; clinicians managing pain (oncology, primary care or other providers); and effective treatments (pharmacologic and non-pharmacologic). The moderator probed attitudes around pain management, such as perceived harms and benefits of opioid and cannabis. We defined cannabis as products containing tetrahydrocannabinol (THC), cannabidiol (CBD) or both; we described this range of products to participants and distinguished cannabis products during discussion. Two other authors (SJ and DK or TS) were present at each focus group. The team debriefed after each group, iteratively changing the guide to address emerging themes. We continued adding groups until saturation was achieved.

Medical record data

We abstracted demographic, diagnosis and treatment data from medical records.

Qualitative analysis

Focus groups were video recorded, transcribed, and analyzed using grounded theory and standard content analysis [73–75]. We used QDA Miner software. Four authors (SC, TS, SJ, JB) developed an initial coding scheme based on our primary aims. (Appendix B) We used standard content analysis to identify and analyze themes, deriving additional themes based on findings. The first transcript was independently coded by all coders, with follow-up discussions to refine the scheme and resolve inter-coder discrepancies. The second transcript was independently coded by two authors (SJ and JB), with discrepancies resolved through discussion among all four coders. This iterative process established a high level of intercoder reliability (Kappa = 0.872), at which point the remaining transcripts were coded by one of two authors (SJ and JB).

Privacy

Given the potentially sensitive topics of cannabis and opioid use, participants were encouraged to use pseudonyms, and medical record and focus group data were not linked. Due to this minimal risk, the Memorial Sloan Kettering Cancer Center Institutional Review Board exempted the study from human subjects review, and a formal informed consent process was not required. Our findings are reported according to the COREQ reporting checklist (Appendix C).

Results

Participants

26 survivors participated in the study: 24 HNC, 1 LC and 1 esophageal cancer, erroneously enrolled. Because esophageal cancer survivorship is similarly complex to LC and HNC, involves similar anatomy (aerodigestive tract), and commonly involves ongoing pain, this participant remained in the discussion but was not included in analysis. Most participants were male (80%) and white (92%). Survivors ranged from 7 to 35 months from treatment completion (Table 1). Below, we describe five major themes: impact of pain, experiences with providers and pain treatments, experiences with opioids, experiences with cannabis and changes in ellxpectations.

Table 1. . Patient characteristics (N = 25).

Characteristic	Mean (SD)
Age at time of study	55.9 (10)
Months since treatment completion	18.3 (9.2)

Characteristic	N (%)
Cancer
Head and neck	24 (96)
Lung	1 (4)
Chemotherapy
Yes	15 (60)
No	10 (40)
Radiation therapy
Yes	23 (92)
No	2 (8)
Surgery
Yes	20 (80)
No	5 (20)
Sex
Male	20 (80)
Female	5 (20)
Race/ethnicity
White non-Hispanic	23 (92)
White Hispanic	2 (8)
Highest level of education completed
Less than high school	1 (4)
Graduated high school or attended college or trade school	9 (36)
Graduated college	6 (24)
Completed graduate school	9 (36)
Oncology setting
Academic hospital	14 (56)
Community hospital	11 (44)
History of tobacco use
Current	0 (0)
Former	9 (36)
Never	15 (60)
Missing	1 (4)
History of alcohol use
Yes	2 (8)
No	23 (92)
History of recreational drug use
Yes	4 (16)
No	21 (84)

SD: Standard deviation.

Impact of pain

Survivors enumerated many ways in which pain in the prior 6 months had interfered with daily life. For many, pain complicated essential physical activities, like eating, work or exercise:

“It hurts to swallow still, and they said that would be chronic.”

“I still have burns on my tongue…I cannot eat a lot of foods.”

For several, pain interfered with work or exercise:

“My right arm [has] constant pain… I'm a composer; I play the piano and I can't play for more than half-hour. I can't do the things that I'm supposed to be doing regularly”

“I groom dogs and cats. I own a business, so I couldn't work for the month of December. I just couldn't. It's a very physical job.”

“I was a marathoner. If I go out for a run now…[I] get so dry throated and [I] can't spit. It's just very uncomfortable.”

For many, pain diminished psychological health and impacted relationships:

“The pain is always a reminder of the cancer, and is it coming back? … Anxiety.”

“[The pain] does really put stress on the marriage and the family.”

Experiences with providers & pain treatments

All but one participant (24/25, 96%) rated their pain care as good or excellent, and 24 (96%) reported no problems getting pain care. Participants identified many provider types as helpful in post-treatment pain management. Oncology doctors and nurses had most often discussed (96%) or treated (84%) participants' pain. Most participants had tried multiple pain management approaches since treatment completion (Table 2). Opioids were a common strategy, with 69% having previously used them and 8% currently using. 54% had used any cannabis product since treatment completion, and 35% currently use cannabis.

Table 2. . Self-reported pain management strategies used since completing cancer treatment.

Medications	Currently use N (%)	Used in past but stopped N (%)	Never used N (%)
Traditional systemic therapies
Opioids	2 (8)	18 (72)	5 (20)
Cannabis	8 (32)	5 (20)	12 (48)
Non-steroidal anti-inflammatory drugs	7 (28)	10 (40)	8 (32)
Acetaminophen	7 (28)	10 (40)	8 (32)
Muscle Relaxants	2 (8)	3 (12)	20 (80)
Anxiolytics	8 (32)	4 (16)	13 (52)
Antidepressants	3 (12)	5 (20)	17 (68)
Anticonvulsants	4 (16)	9 (36)	12 (48)
Intravenous medications	0 (0)	7 (28)	18 (72)
Traditional locally acting therapies
Pain medicine pumps	0 (0)	3 (12)	22 (88)
Topical medicines	7 (28)	12 (48)	6 (24)
Therapeutic injections	1 (4)	3 (12)	21 (84)
Surgeries	1 (8)	5 (20)	19 (76)
Neurostimulation	0 (0)	1 (4)	24 (96)
Physical therapy	9 (36)	6 (24)	10 (40)
Exercise/muscle strengthening	18 (72)	4 (16)	3 (12)
Integrative, complementary or behavioral approaches
Acupuncture	2 (8)	3 (12)	20 (80)
Massage	9 (36)	4 (16)	12 (48)
Mind-based therapy	14 (56)	0 (0)	11 (44)
Herbal therapies	2 (8)	0 (0)	23 (92)
Mushrooms	0 (0)	1 (4)	24 (96)
Other	0 (0)	0 (0)	25 (100)

Experiences with opioids

Most participants who used opioids viewed them as highly effective:

“It was very, very helpful…noticeably useful and relieving.”

“There were a couple of times that [the pain] was just absolutely unbearable. [With] Fentanyl, it would just go [away]…It was amazing.”

However, most participants were also highly cautious about opioids, enumerating side effects and other perceived harms. Some described physical side effects, like constipation, nausea, and dizziness as well as being mentally incapacitated:

“The problem with opioids is constipation. For me, I took them for as little as possible because of that problem. It just was unbearable.”

“It made me very, very dizzy. I couldn't really do anything else when I was taking that.”

“Some of the oral opioids really made me very nauseous.”

“Knocked me out…It was awful.”

Many also expressed fear of addiction:

“I told [my doctors]…it didn't matter how much pain I was in, I just wasn't taking [opioids]…cause you hear how people get hooked.”

“I was really afraid of taking too much Oxy… just to make sure I didn't become dependent.”

Stigma was another theme, and accordingly, some participants reported access issues:

“I get the feeling [some providers] were judging me, that I was just trying to get [high].”

“The pharmacy put the script in and I said to [the pharmacist], why are you giving me, I think he gave me 10 [tablets]. And he said the regular BS, the government, this and that, and he said, ‘When you finish, come back and get another script.’ Okay, did it and he gave me grief again and I told him to take the script and shove it.”

Given these issues, many participants would only consider taking opioids again for severe pain or if no other options were available:

“It would have to be pretty severe pain.”

“It would be a last resort.”

Experiences with cannabis for pain

About half of participants (13/25, 52%) had tried cannabis for pain after treatment completion, including edibles, medical cannabis pills, smoking, vaping, CBD and THC tinctures, CBD creams and salves and combinations. About a third (8, 32%) were still using these products.

The vast majority who had tried cannabis said it helped them. Many emphasized relief from physical pain:

“[CBD] helps dull the pain down, so it really does not get in the way.”

“Scar tissue makes it impossible to move the neck too much…A few drops of [CBD oil] loosens up the neck muscle and helps me out.”

“Everything I've tried definitely helped with my pain. It was both THC edibles and tinctures, as well as CBD salve and oil.”

Some compared cannabis favorably to opioids, with better relief and fewer side effects. Others liked that cannabis was ‘natural’:

“They worked better than the opioids… And no side effects whatsoever.”

“I think they're a better alternative than opioids, and I felt they worked better.”

“It felt a lot more natural for me.”

Several noted that cannabis helped them with nausea and appetite:

“Marijuana did increase my appetite, which was…beneficial.”

“Cannabis…would relax me enough where I could eat and not feel nauseated.”

Others found cannabis helpful for sleep:

“I used CBD oil under the tongue, and I used it for, as I was saying, anxiety, and also to help me go to sleep which was a major, major problem at one point. I had really bad insomnia that I think was anxiety-based. So, I used the CBD in conjunction with other stuff, like Trazodone”

“The cannabis … just took away the pain [so] I slept great.”

“The pain right now, I smoke a little bit sometimes at night to go to sleep.”

Participants also emphasized how cannabis had relieved anxiety and psychological pain:

“[THC] really helps with my anxiety.”

“I have a lot of anxiety around eating because I have a fear of choking, so sometimes it helps that.”

A few, however, said the products they had tried were ineffective. Others said that cannabis exacerbated anxiety:

“I ordered some gummies…I really didn't see any change.”

“I don't take the [THC] pills anymore because they really weren't strong enough.”

“[CBD] oil sent me into an anxiety spin…I ended up in the hospital.”

“It made me so anxious. I felt like I couldn't breathe. It was not a good experience.”

Others said cannabis interfered with sleep or left them groggy:

“[With] the tincture, I had really hard time sleeping.”

“[Cannabis made me] tired and a little bit slow in the morning.”

For many HNC patients, cannabis exacerbated dry mouth:

“[Edibles] helped with the pain, but it caused severe dry mouth, which made it almost impossible to swallow.”

“Edibles definitely increased my dry mouth…At night it was severe.”

Some feared cannabis use would interfere with employment or be otherwise stigmatizing:

“If I'm going for a new job and they do a drug screen, I don't want that stuff in me.”

“Shame on anyone that used that stuff, that's how I grew up.”

Others, however, were unconcerned about stigma:

“If it's something making me feel good, I don't care what people think.”

“Most…cancer patients don't really give a damn what other people think.”

The majority perceived declining stigma around cannabis:

“Shame on anyone that used that stuff, that's how I grew up. And that's how I was my whole life until last year…I think it was just the shock and the horror of having to go through what I went through that gave me the attitude of what do I have to lose. And I'm glad I tried it. I think completely different now. I think society is changing, too, that's why you've been seeing it legalized state by state, not just for medical use, that's been going for a while, but now for recreational use.”

“Everybody's doing it now. It's become semi-legal in many locations. It's an acceptable form of entertainment and pain relief, so I don't think there's nearly the stigma with it that there was years ago.”

Nonetheless, several detailed access problems, including cost:

“When I talked [to providers] about marijuana they didn't want to talk about it. They couldn't recommend it because there's no clinical data.”

“You have to know someone to get it.”

“[The CBD] was really expensive. A little bottle is $150.”

Most would consider taking cannabis for pain in the future, particularly if provider-approved:

“[If my provider said] it's the right way to go, it's definitely something I would at least try if I was in a lot of pain.”

Others, however, remained uninterested:

“I have happily avoided it, this far.”

Changes in pain management expectations

The balance between pain management and minimizing side effects changed after treatment completion.

“Now, I see [side effects] very differently…I have to think about them before I take something so I am able to function.”

“When I was doing treatment…[I] did whatever I needed medicine-wise, just to manage the pain and if I couldn't function…it didn't really matter. But since treatments ended and I started titrating off the Gabapentin…I'm in more pain than I was, but it's just something that it's manageable.”

Survivors considered how to manage pain while maintaining function.

“The only challenge…[is] taking enough medication to be able to go to sleep but then still be able to get up in the morning. And finding a balance there was really hard.”

“Where [the doctor has] got me now…if I go down any lower, it's not going to work. If I go up any higher, I'm going to be not able to function.”

Survivors reported adjusting expectations to live with some pain.

“You just have to work around it. You can't complain or stay in bed all day. You just got to do what you got to do.”

“There's going to be some pain, but if you stay positive…that helps.”

Discussion

For survivors of complex cancers, post-treatment pain significantly impairs everyday functioning. Despite general satisfaction with pain care, relief remained elusive for many in this study. Our novel focus on post-treatment use of cannabis and opioids revealed complex tradeoffs of perceived harms and benefits when managing pain. Neither cannabis nor opioids emerged as straightforward approaches for pain. Perceived harms varied from side effects (e.g., constipation from opioids, anxiety from cannabis) to practical issues (e.g. cost of cannabis, stigma regarding both products). Perceived benefits similarly ranged widely (e.g. relief from acute pain from opioids, pain and anxiety relief from cannabis).

Given the unclear effectiveness of cannabis and opioids for pain management among survivors of complex cancers, provider involvement is critical. In our study, post-treatment pain management involved mostly oncology providers rather than primary care providers. It is unclear whether oncology is the optimal setting for managing long-term, post-treatment pain. Providers of ongoing care for survivors vary, with oncology, primary care and other providers (e.g. pain specialists, physical therapists, massage therapists, integrative medicine practitioners) having different degrees of oversight depending on the survivor and time since diagnosis [76–81]. For survivors with long-term pain, it is critical that providers, regardless of practice type, are coordinated in their pain management efforts and knowledgeable about pain-management approaches – including opioids and cannabis, which likely requires better clinical education about cannabis and the communication of cannabis harms and benefits to patients [82].

Despite cannabis' benefits for many in our study, access was often challenging, due to providers not offering cannabis, difficulties purchasing conveniently or legally, or patient lack of awareness. These challenges are understandable given regulatory issues, limited evidence regarding cannabis for symptom management, and poor clinician knowledge, which can inhibit patient-provider communication about cannabis. Providers often lack knowledge to counsel about cannabis, with 84% of cancer providers and 87% of oncology fellows reporting knowledge deficits and only 24% of fellows receiving relevant training [82,83]. Furthermore, cannabis costs, a barrier in our study, can be prohibitively high [84]. Some survivors felt stigma around cannabis use, perceiving it as illicit or knowing that others, including some providers, view it that way. Stigma may impede patient-clinician discussions about cannabis.

Few survivors in our study used opioids for ongoing post-treatment pain. Most indicated that benefits were offset by unpleasant side effects or fear of addiction. This fear is a notable source of distress and may cause survivors to use opioids inappropriately (e.g., avoiding when beneficial) [85–87]. Stigma from doctors, pharmacists, family and friends deterred opioid use for some survivors.

Co-occurring psychological pain was another notable finding. The emotional consequences of cancer are well established, particularly fear of recurrence [13,88]. As observed elsewhere, our participants linked physical and emotional pain, with physical pain causing distress and anxiety, which then affected functioning [89,90]. Psychological symptoms and pain have a bidirectional relationship; patients with pre-existing mental-health disorders experience more pain, which can exacerbate psychological symptoms [91,92].

Our study illustrates the interplay between pain treatments and psychological symptoms, with participants for whom cannabis improved anxiety often finding it also reduced pain. The relationship between pain and anxiety is well established, though the mechanisms of mutual influence are not fully described [93]. The American Society of Clinical Oncology cites strong evidence that psychological interventions to relieve anxiety (e.g., relaxation techniques, mindfulness-based stress reduction) are consistently effective at reducing pain and improving quality of life in survivors [3]. If cannabis is shown to reduce anxiety, it may have a similar effect to psychological interventions in reducing pain.

These findings suggest that a sole focus on physical pain is likely insufficient for many survivors. Yet common methods to assess survivors' pain, such as the two-question screen in ASCO's pain management guidelines and the Brief Pain Inventory, typically omit psychological pain [3,94]. Evaluating multiple dimensions of pain may improve how clinicians conceptualize and treat it. Using standardized instruments to screen for anxiety and depression, as recommended in practice guidelines, would likely be a valuable supplement to traditional pain assessment [95–97]. These psychosocial needs underscore the importance of mental healthcare as an integral part of comprehensive survivorship care, as is widely recommended [3,54,98,99].

Our study was limited to a convenience sample of mostly white, male, head and neck cancer survivors with mild pain burden, reflecting the demographics of these patients across the three hospitals. Recruitment challenges limited enrollment of lung cancer survivors which reduces generalizability to this population; however, the burden of the aerodigestive symptoms in lung cancer suggests likely similarities in pain and symptom management that should be explored in future research. A larger, more diverse sample may reveal different challenges, especially because people of color face greater legal risks around cannabis use (i.e., disparate rate of legal enforcement of cannabis laws) and more challenges with pain management [100–105]. Lastly, we did not ask participants explicitly how they obtained opioids and cannabis. Although participants often revealed this information, future research should explore the extent to which cannabis and opioids are prescribed or recommended to survivors, and whether these products were obtained from medical, commercial or illicit sources.

Conclusion

Our study elucidated several important considerations for managing pain in disease-free cancer survivors. Survivors in our study experienced daily challenges. Future studies must address psychological distress as integral to pain assessment and management. Critically, all clinicians who manage survivors' pain must understand myriad treatment options, including cannabis, to educate survivors about their choices and optimize care.