4 Evaluating efficacy and areas of clinical application for LI interventions Tereza Ruzickova Psychopharmacology and Emotion Research Lab University of Oxford 5th May 2022 What do you remember from our practical lecture last week? What problems might come up in BA? Problems that can come up in BA • Forgetting to do the activities • Lack of motivation/energy • Lack of time for activities • Exhaustion/overwhelm • Can’t focus or enjoy the activity - stuck in negative thoughts • Reducing bad activities vs increasing good activities • Big practical / necessary problems in life Possible solutions • Be kind, empathetic, validating • What happened/what activity did you do instead? • Di ff erentiate short-term and long-term consequences • What got in the way? •How could we make the activity easier? Or start with a di ff erent, easier activity? • Action First / 5min rule -> it will get easier with time More possible solutions • Praise every small achievement, fi ght against all-or-nothing/ black and white thinking (-> more advanced therapy) • Always try to map out barriers and come up with plan B/C/D ahead of the week • Evaluate schedule to fi nd more time / stick with small goals • Psycho-education on boom and bust • Psycho-education about rumination, problem-solving, mindfulness/distraction techniques Explaining LI interventions in simple terms Evaluating the efficacy of BA and other LI interventions • What are the best ways to evaluate the e ffi cacy of psychological treatment? Evaluating the efficacy of BA and other LI interventions • What are the best ways to evaluate the e ffi cacy of psychological treatment? • Why are randomised controlled trials randomised? • Through randomisation, we are hoping to distribute any potentially confounding variables equally between both groups • However, we should still check whether that succeeded (as much as we can) • By conducting an experiment, we can manipulate the temporal order of events (e.g. does an early increase in activation lead to a later decrease in depression, rather than the other way around?) • Randomisation should prevent any third (confounding) variable from explaining the association Evaluating the efficacy of BA and other LI interventions • What are the best ways to evaluate the e ffi cacy of psychological treatment? • Why are randomised controlled trials randomised? • Why are randomised controlled trials controlled? • A (passive) control group allows us to compare our intervention to rates of spontaneous recovery with time and regression to the mean • Crucially, it allows us to check if we’re not making patients worse than they would be without it! • An active control group additionally allows us to compare our intervention to placebo e ff ects, although this is harder to do with psychological interventions Mayou, Ehlers and Hobbs (2000) Psychological debrie fi ng used to be a popular intervention after trauma until it was found it signi fi cantly slowed down recovery when compared to a passive control group! Efficacy of LI interventions for depression Ekers et al. (2014) • 26 randomised controlled trials (over 1500 participants) • BA more e ff ective than control groups or antidepressant medication • More studies with long term follow-up needed Cuijpers et al. (2007) MA • Large e ff ect size when comparing BA and control conditions • No signi fi cant di ff erence from other established treatments for depression Mazzuchelli et al. (2009) MA • Again, large e ff ect sizes found in comparison to control conditions • No signi fi cant di ff erence from other established treatments “BA may be considered a well-established and advantageous alternative to other treatments of depression.” Orgeta et al. (2017) • SR and MA of 18 randomised controlled trials • BA e ff ective for reducing depression symptoms for older people in the community (55 and older) • Larger studies needed Tindall et al. (2017) • SR and MA of 3 randomised controlled trials and 7 pre-post studies • Examining BA for younger people (18 and below) • BA may be e ff ective, but more studies with better methodology are needed Zabihi et al. (2020) • SR & MA of BA for depression in informal caregivers • 12 randomised controlled trials • BA reduces depression symptoms after intervention and at 1-year follow-up Mir et al. (2015) • Qualitative study on BA adapted for muslim communities • Incorporating religious beliefs into the intervention narrative • Interviews showed the intervention as acceptable and feasible Moradveisi et al. (2013) • Comparing BA vs antidepressant medication in routine clinical practice in Iran • More e ff ective and better retention than antidepressant medication • More e ff ective in patients with more severe depression Bryant et al. (2017) • Randomised controlled trial of brief BA for women with a history of genderbased violence in Kenya • Provided by lay workers from the community after 8 days of training • Moderate reductions in psychological distress maintained at 3-month follow-up Efficacy of LI interventions during crisis periods Ruzickova et al. (2021) • 4-week online BA administered by non-specialists after 15h of training • BA remains e ff ective even when activity options are signi fi cantly limited due to social distancing • Signi fi cant reduction in depression + anhedonia and increase in activation + social support • Bene fi ts remained at one-month follow-up • BA may be particularly appropriate for societal crisis periods with increased mental health burden Efficacy of LI interventions for anxiety Chen et al. (2013) • 8 weeks of group BA vs passive control for excessive worry (transdiagnostic treatment) • daily activity monitoring, identifying avoidant behaviours, goal setting • signi fi cant (but small) e ff ects on excessive worry, intolerance of uncertainty, cognitive avoidance, problem solving • however, no signi fi cant e ff ects on anxiety or stress symptom scores • possible similarity between BA and exposure treatments in anxiety? • other strategies probably needed for a larger e ff ect e.g. active problem-solving training, examining core beliefs, “worry time”, relaxation Hopko et al. (2016) • BA found e ff ective for anxiety symptoms in 70 breast cancer patients • Pre-post design without a control group Essau et al. (2014) • BA found e ff ective for anxiety symptoms in 60 school children • Pre-post design without a control group Hopko et al. (2016) • BA found e ff ective for anxiety symptoms in 70 breast cancer patients • Pre-post design without a control group Essau et al. (2014) • BA found e ff ective for anxiety symptoms in 60 school children • Pre-post design without a control group Ali et al. (2017) • Longitudinal cohort of over 400 patients • 50% participants found to relapse within 1 year, particularly if they had residual symptoms at the end of treatment • Compare to relapse rates from full CBT (30%) or antidepressant medication (76%) as found by Hollon et al. (2005) • Or mindfulness based cognitive therapy (40%) as found by Piet and Hougaard (2011) • Recommend that patients are treated until residual symptoms are gone and follow-up care is provided “We suggest taking the long view, recognising that problems like depression often have to be managed as recurrent long-term conditions.” Remaining questions… • Which people are most likely to bene fi t from LI interventions vs other treatments? • Which people are most likely to relapse? • What is the role of severity, comorbidity, age of onset, demographic factors..? • Can LI interventions be e ff ective when administered repeatedly? • Can they be e ff ectively combined with other treatments to reduce risk of relapse - e.g. medication or other CBT components? A note on activation & grief-related avoidance • Prolonged avoidance of painful thoughts, emotions or physical reminders is associated with worse mental and physical health outcomes in grief (see Baker et al., 2017 for review) •Avoidance has been found to mediate the link between grief-related rumination (repetitive thinking about negative past experiences) and depression (Eisma et al., 2013) • Rumination as Avoidance Hypothesis (Stroebe et al., 2007) repeated analysis of the past impedes acceptance and “exposure” to present reality • Internet-based behavioural activation may facilitate gradual exposure to loss-related stimuli as well as other enriching experiences (Eisma et al., 2015) • But care should perhaps be taken to avoid facilitating further avoidance through purely loss-avoiding activities Efficacy of LI interventions for substance abuse Martinez-Vispo et al. (2018) • SR of BA for depression comorbid with substance use • 6 randomised controlled trials, 2 pre-post designs • smoking, alcohol, opiate addiction • BA improved substance use outcomes in 7/8 and improved depression in 6/8 studies Focus on regularly interacting with other positively reinforcing stimuli Efficacy of LI interventions for physical health problems Upho ff et al. (2020) • Cochrane systematic review of BA for depression comorbid with noncommunicable diseases (diabetes, cancer, cardiovascular, respiratory) • 2 randomised controlled trials • Insu ffi cient evidence to conclude e ff ect (possibly due to variability in physical conditions and requirement of formal depression diagnosis) Hedman-Lagerlof et al. (2021) • 12 weeks of internet-based CBT with therapist guidance for atopic dermatitis • randomised controlled trial • signi fi cant reduction in itch intensity, perceived stress, sleep problems Thank you for your attention! Don’t forget you can give anonymous feedback: https://forms.gle/xDaPaTfGyHb4FDre7