Group therapy decreases chronic pain

For low-income/low-literacy patients, modified cognitive behavioral therapy (CBT) and group pain education (EDU were associated with decreases in chronic pain (see chronic pain behav therapy lowincome aim2018 in dropbox, or doi:10.7326/M17-0972).

Details:
--290 patients with chronic pain in 4 low-income community health centers in Alabama, from 2013-15. Chronic pain was defined as pain most days of the month for at least 3 months
--71% women, 33% white/67% African-American, 66% single, 72% at or below poverty level, 36% below 5th grade reading level, 13% employed, 43% no insurance, 83% disabled, 50% chronic LBP/13% knee pain/8% neck pain, 89% musculoskeletal pain/76% arthritis/47% headache, mean PHQ-9 of 12.1, mean Brief Pain Inventory (BPI)-interference (a measure of pain-related function)=6.6, mean BPI-intensity (a measure of pain intensity)=6.5 [both with scores 0-10, higher score more pain or worse function]
--on average, the sample reported >6 pain sites, >4 causes of pain, and pain lasting >15yrs
--therapies: 68% on opioids (tramadol 53%, acetaminophen/hydrocodone 57%), 57% NSAIDs, 50% muscle relaxants, 26% "nerve pain meds"/30% psychotropic meds, 34% surgery, 34% nerve blocks
--exclusions: cancer pain, significant cognitive impairment, self-reported substance abuse, uncontrolled mental illness, reading level below 1st grade
--comparing 3 groups:
    --usual care (UC)
    --literacy-adapted and simplified group CBT: 10 weekly 90-minute group sessions, with simplified CBT behavioral techniques, including motivational reinforcement, pain education, and pain management skills
    --EDU: 10 weekly 90-minute group sessions pain self-management and group discussion, but no specific skill-building execises
--both intervention groups also received usual care: nonpsych pain management (drugs, chiropractic/physical therapy)
--participants were paid $20/session for travel to group sessions. each received a workbook and weekly audio CD with summary of that week's material
--outcomes: BPI-intensity scores (primary outcome), BPI-invervention score, and PHQ-9 for depression. A 30% change was considered clinically meaningful [though in using BPI’s, a 0.7 change is actually considered clinically meaningful: see http://gmodestmedblogs.blogspot.com/2018/03/opioids-not-better-for-chronic.html​ ]

Results:
--see graphic representation below (much easier to see the trends and differences)
--baseline to post-treatment
     --BPI-intensity score:
        --UC: -0.26; CBT: -1.06, change vs UC: -0.80; EDU: -0.83, change vs UC: -0.57
    --BPI-interference score:
        --UC: -0.30; CBT: -1.66, change vs UC: -1.37; EDU: -2.00, change vs UC: -0.70
    --PHQ-9:
        --UC: -1.10; CBT: -2.43, change vs UC: -1.33; EDU: -2.24, change vs UC: -1.15
--post-treatment to 6-month follow-up
     --BPI-intensity score:
        --UC: -0.24; CBT: 0.35, change vs UC: 0.60; ​EDU: 0.14, change vs UC: 0.38
    --BPI-interference score:
        --UC: 0.18; ​CBT: 0.54, change vs UC 0.36; EDU: 0.25, change from UC 0.07
    --PHQ-9:
        --UC: -0.24; CBT: 0.61,change vs UC 0.85; EDU: 0.43, change vs UC 0.67
    --the proportion of people with >30% improvement in pain score (for CBT, EDU, and UC respectively) was 30.5%, 20.0% and 11.5% at posttreatment, and 21.7%, 16.4% and 8.5% at 6-month followup. Using the minimally important change criterion: CBT and EDU groups exceeded this criterion at post-treatment and 6-months later.
    --a 30% improvement in physical function(for CBT, EDU, and UC respectively): 40.2%, 28.8%, and 18.4% at post-treatment, and 34.8%, 20.6%, and 15.7% at 6-month followup . the CBT and EDU exceeded their minimally important change criterion of 14% at posttreatment, though only the CBT did so at 6 months.
    ​--depression:  the proportion of patients with clinically meaningful depression improvement were: 39.5%, 42.5% and 28.2% at posttreatment and 42.6%, 33.8%, and 31.0% at 6 months; a significant improvement in both intervention groups post-treatment and at 6 months:
--adverse events: 9.4% of CBT, 17.5% of EDU and 18.4% of UC patients went to the ED, mostly for temporary pain exacerbations

Commentary:
--statistics: 20% of physician visits and 10% of drug sales are for chronic pain; 116 million Americans have chronic pain ($600 billion/yr), pain disproportionately affects economically disadvantaged/ethnic minorities/women/older adults
--CBT has been shown to be effective for chronic pain management (eg, see http://gmodestmedblogs.blogspot.com/2016/03/low-back-pain-improves-with-stress_29.html  , or http://gmodestmedblogs.blogspot.com/2017/04/home-based-cbt-for-low-back-pain.html  ), but had not been tested in low-income communities
--the patient workbooks and therapist supplements are available for free at http://pmt.ua.edu/publications.html 
--overall, the most improvement in the intervention groups was from baseline to post-treatment, with some decrement at the 6-month follow-up mark. As in the graph below, at that point the CBT and EDU interventions were pretty equivalent, with both BPI-intensity around 6 (vs UC around 6.4, not much difference), but BPI-interference (reflecting how much the pain interfered with daily life) being around 5.5 with UC at 6.5 (0.7 difference is clinically meaningful).  PHQ-9 was around 9 in both intervention groups, but around 10.3 in UC (the cutpoint of 10 = “probable depression”).
--there was a self-selection bias (they included only patients who wanted to be included), which interferes with the generalizability of the results.
--it would be interesting to know the response to these interventions by type of pain, and whether a positive response led to decreasing pain meds (esp opioid use)



cid:8bc6c0c8-6efe-4d06-9ff6-024b37a5759e


so,
--given that CBT was not more effective than EDU at 6-months, one wonders if the main actual intervention was just having group sessions (patients get a lot of information, often learn more from their peers than us guys, get support and feel better just by being in groups; this has has been found in other studies of group interventions, such as with diabetics). And, if group sessions were to continue for the long-term, instead of being limited to 10 sessions, the pain and level of function might not just continue to be better than usual care but actually continue to improve.​
--and, an oft-repeated issue, it is pretty striking that 43% of those in their study had no insurance......  makes it a bit hard to develop long-term, coherent, effective, and inclusive programs to help people... [we really  do need universal health care!!]

Comments

Popular posts from this blog

cystatin c: better predictor of bad outcomes than creatinine

diabetes DPP-4 inhibitors and the risk of heart failure

UPDATE: ASCVD risk factor critique