Effective buprenorphine use and tapering strategies: Endorsements and insights by people in recovery from opioid use disorder on a Reddit forum

Background of preprint

While buprenorphine is one of the most effective medications used in medication-assisted treatment (MAT) for opioid-use disorder (OUD), its tapering strategies have not been optimised. For most medications, advice to patients on medication use comes from two major sources: their healthcare professionals, and package inserts that contains drug information from drug manufacturers. In the case of buprenorphine tapering strategies, both are relatively limited due to the relative dearth of clinical studies.

Consequently, many patients rely on social media to seek advice and discuss strategies pertaining to buprenorphine tapering. In their preprint, Graves et al. systemically mine and analyse data pertaining to buprenorphine tapering strategies from the anonymous social media platform Reddit. Specifically, Graves et al. used Natural Language Processing (NLP) to mine and analyse buprenorphine-related posts from the subreddit forum /r/suboxone.

Key findings of preprint: Testing the limits and breaking through

Graves et al. first categorised the posts according to the information they provided (Table 1, preprint Table 1).

Table 1. Overall information described in the Reddit posts.

The authors then manually characterised descriptions of buprenorphine use and tapering strategies. Two observations stood out: 1) longer tapering schedules were reported to be more effective for sustained recovery, and 2) tapering from 2.0 mg to 0.0 mg of buprenorphine use was particularly challenging. Specifically, the authors used Reddit’s “upvote” system to qualitatively and quantitatively characterise the success of various tapering schedules and strategies, as well as their accompanying side effects.

By analysing their results with and without scaling the posts to their respective upvotes, the authors find that the most effective quitting doses were 0.063 mg and 0.125 mg; the highest quitting dose was 2 mg, which was generally not popular. From the 23 users who provided enough information for Graves et al. to quantify the number of days involved in the tapering of buprenorphine from 2 mg to 0 mg (the most difficult phase of tapering), the authors found that the median time to taper was 93 days (preprint Figure 2).

Among adverse events reported by patients in terminating buprenorphine, diarrhoea, insomnia, fatigue, and restless leg syndrome were the most common (preprint Figure 4a). The authors also investigated techniques that the Reddit patients used to overcome these adverse events. They found that clonidine and loperamide (Imodium) were the most common pharmacological strategies employed; exercise was among the most effective non-pharmacological strategy (preprint Figure 4b). These strategies are indeed intuitive: loperamide is commonly used to treat diarrhoea, and clonidine has been reported to help in reducing the symptoms of opioid withdrawal, including restless leg syndrome. Physical activity is also one of the most common interventions that healthcare professionals employ in motivational interviewing when encouraging patients to change their lifestyle in overcoming addiction.

What I like about this preprint: Not a kingdom of isolation

I chose this preprint for two reasons. First, the formation of patient support groups on social media show patients that they are not alone in coping with their disease. They also provide a valuable trove of information. As Graves et al. suggest, the complementary advances in NLP and increase in social media use by these patient support groups has enabled healthcare researchers to quickly mine these platforms for information pertaining to the monitoring of epidemiological trends, as well as for studying self-management of buprenorphine. The use of machine learning to rapidly assemble large datasets from other social media platforms (i.e. Twitter) has also accelerated this process [1,2].

Second, this preprint indirectly tackles the opioid epidemic, which is responsible for an alarming rise in overdose-related deaths in the USA in recent years. Much of the problems pertaining to opioid misuse can be attributed to the lack of alternatives in medications for pain relief, which drives prescriptions for opioids. Unfortunately, the paucity of large-scale trials on weaning patients off opioids only serves to compound these problems, as healthcare associations do not have enough evidence to offer high-grade guidance for physicians to taper patients off opioids where indicated. Graves et al. posit that relying on self-reports by patients on social media platforms, especially ones like Reddit, may be an alternative way to gather data and run analyses.

Future directions: Tapering never bothered us anyway

Future directions from this preprint will involve two areas. The first revolves around expanding on the findings from this preprint. Patient advocacy groups are rising, and other tailored websites such as PatientsLikeMe are being developed for patients to offer advice and support one another. Capitalising on these platforms may soon become another way for researchers to analyse and monitor epidemiological trends.

The second area for future directions will involve overcoming the limitations in this preprint. One limitation discussed by Graves et al. points towards the demographic of Reddit users—given that they are likely to be “younger, male, and Hispanic… (they) may not comprise a representative sample of the United States population”. The reliability of the posts may also be compromised by the self-reporting and the public influence on social media; and the anonymity of the Reddit posts complicate the traceability of outcomes employed by users.

Despite these difficulties, data mining from social media platforms and their analyses are here to stay. Improvements in NLP and machine learning strategies in the coming decade will accelerate and refine these strategies, empowering both healthcare professionals and their patients to use and manage opioids appropriately. Hopefully, OUD will one day be a thing of the past.

Questions for authors

• It is interesting to note the use of other psychotropic substances (marijuana and kratom), as well as anxiolytics (benzodiazepines) and even other pain medications (gabapentin) in relieving withdrawal symptoms arising from buprenorphine (preprint Figure 4b). Were these medications grouped by mechanism of action and analysed? A back-of-the-envelope summation of all the benzodiazepines, for example, may result in a rather significant effect on aiding in the management of withdrawal symptoms arising from buprenorphine use.
• How do your observations compare to current guidance in addiction medicine for OUD? How do the tapering strategies compare to strategies used to overcome other addictions, e.g. in smoking cessation?

References

[1] Sarker A, O’Connor K, Ginn R, Scotch M, Smith K, Malone D, Gonzalez G, Social Media Mining for Toxicovigilance: Automatic Monitoring of Prescription Medication Abuse from Twitter, Drug Safety 39(3) (2016) 231-240.

[2] Graves RL, Tufts C, Meisel ZF, Polsky D, Ungar L, Merchant RM, Opioid Discussion in the Twittersphere, Substance Use & Misuse 53(13) (2018) 2132-2139.

Tags: buprenorphine, medication assisted treatment, opioid use disorder, reddit, social media use

doi: https://doi.org/10.1242/prelights.15827

Read preprint (No Ratings Yet)