The Generalized Anxiety Disorder Workbook
Melisa Robichaud & Michel Dugas (New Harbinger, 2015)
Built on the intolerance-of-uncertainty model these researchers pioneered. The best single match if your worry is about money and the future.
This site is a scaffold, not a substitute for the actual programmes. The workbooks below were written by the researchers who ran the trials, and working through one properly will do more for you than any website — including this one.
Structured self-help workbooks tied directly to the trial literature.
Melisa Robichaud & Michel Dugas (New Harbinger, 2015)
Built on the intolerance-of-uncertainty model these researchers pioneered. The best single match if your worry is about money and the future.
Pia Callesen (2017/2020)
The most accessible metacognitive therapy self-help book. Callesen trained at Wells’ MCT Institute.
Caveat: Written primarily around depression and overthinking, and its main efficacy claim is Callesen’s own 2020 Scientific Reports study — good, but not a large trial base.
Robert Leahy (2005)
Comprehensive, readable, structured programme for worry from a leading cognitive therapist.
David A. Clark & Aaron T. Beck (Guilford, 2011)
Classic CBT skills from the founder of cognitive therapy and a leading anxiety researcher.
Russ Harris
The most usable ACT self-help and practitioner texts. The Phase 3 reading.
Judson Brewer (2021)
Habit-loop and mindfulness approach. The companion app has genuine RCT support (Roy et al., JMIR, 2021, n=65: 67% vs. 14% GAD-7 reduction vs. usual care at two months).
Caveat: The trials come from Brewer’s own group and the app is commercial.
For Phase 4 consolidation.
Mark Williams & Danny Penman (2011)
Williams co-developed MBCT; the closest consumer version of an evidence-based programme, with audio.
Jon Kabat-Zinn
The foundational MBSR text — the programme tested in the JAMA Psychiatry non-inferiority trial.
Segal, Williams & Teasdale
The MBCT structure in workbook form.
Practitioner texts, if you want the machinery rather than the programme.
Adrian Wells (Guilford, 2009)
The primary MCT text.
Robichaud, Koerner & Dugas (2nd ed., 2019)
The definitive IU-based clinician manual.
Edward Watkins (Guilford, 2016)
The source for the concreteness work in Phase 2.
David H. Barlow
The transdiagnostic frame.
Steven Hayes
The ACT founder’s own self-help programme.
Explanatory rather than therapeutic — but they change how the worry feels.
Sendhil Mullainathan & Eldar Shafir (2013)
Why financial worry hijacks the mind, based on their Science research. Explanatory, not a workbook — but understanding bandwidth as a documented load rather than a personal failing changes a lot.
Our driver profile is a routing aid, not a measurement. If you want real numbers on these constructs — for yourself or to take to a clinician — these are the validated instruments. We do not reproduce them because several are copyrighted or distributed by their authors.
PSWQ — Penn State Worry Questionnaire
Worry — the trait tendency, not the last two weeks.
Meyer et al., 1990. 16 items, 16–80; around 62 is a common probable-GAD threshold.
IUS-12 — Intolerance of Uncertainty Scale, short form
The construct behind the uncertainty track. The proper version of our driver profile’s IU items.
Carleton et al., 2007. 12 items, 12–60; Prospective Anxiety (7) + Inhibitory Anxiety (5) subscales.
MCQ-30 — Metacognitions Questionnaire 30
Metacognitive beliefs, including uncontrollability/danger and positive beliefs about worry.
Wells & Cartwright-Hatton, 2004. 30 items, 30–120; five subscales.
RRS — Ruminative Responses Scale (brooding subscale)
Brooding specifically — the harmful form of rumination.
Treynor, Gonzalez & Nolen-Hoeksema, 2003. Brooding subscale is 5 items, 5–20.
AAQ-II — Acceptance and Action Questionnaire II
Psychological inflexibility. Note its known overlap with negative affect and neuroticism.
Bond et al., 2011. 7 items, 7–49; roughly 24–25 and above suggests high psychological inflexibility. Free from contextualscience.org.
Every effect size quoted anywhere on this site, in one place, so you can check it.
CBT for GAD, pooled effect vs. control
Cuijpers et al., Clinical Psychology Review, 2014;34(2):130–140. 41 studies, 2,132 patients; g = 0.84 (95% CI 0.71–0.97) across 38 comparisons.
CBT for anxiety vs. placebo
Carpenter et al., Depression & Anxiety, 2018. 41 placebo-controlled RCTs; Hedges’ g ≈ 0.56 on target symptoms.
MCT vs. CBT at 9 years
Solem et al., Brain and Behavior, 2021. 39 of 60 patients; recovery 57% (MCT) vs. 38% (CBT) on completer analysis; GAD re-diagnosis 9.5% vs. 23.1%.
MBSR vs. escitalopram
Hoge et al., JAMA Psychiatry, 2023 (TAME, NCT03522844), n=276. CGI-S improvement 1.35 vs. 1.43; difference −0.07 (95% CI −0.38 to 0.23), within the −0.495 non-inferiority margin. Study-related adverse events in 15.4% vs. 78.6%.
MBCT and depression relapse
Kuyken et al., JAMA Psychiatry, 2016. Individual-patient-data meta-analysis; HR 0.69 (95% CI 0.58–0.82).
Exercise and anxiety
Wipfli et al., 2008. 49 RCTs; effect size −0.48 vs. no-treatment controls. Recent three-level meta-analysis in clinical samples: 27 RCTs, 1,447 participants, d ≈ −0.42.
Cyclic sighing
Balban et al., Cell Reports Medicine, 2023;4(1):100895 (NCT05304000), n=111. Exhale-focused cyclic sighing beat mindfulness meditation on mood and respiratory rate over one month of 5 min/day.
Job insecurity and mental health
Llosa et al., Anales de Psicología, 2018. 56 samples, 53,405 participants; subjective insecurity correlates with depression, anxiety and emotional exhaustion independently of actual job loss.
Financial scarcity and cognition
Mani, Mullainathan, Shafir & Zhao, Science, 2013;341(6149):976–980. Inducing financial thoughts reduced cognitive performance in poor but not well-off participants; the same farmers performed worse pre-harvest than post-harvest.
Guided vs. unguided self-help
Cuijpers et al., 2010. Guided self-help is roughly comparable to face-to-face therapy (post-test difference d ≈ −0.02); unguided is reliably less effective (d ≈ 0.28 vs. control for self-guided depression treatment).
Rumination-focused CBT
Watkins et al., British Journal of Psychiatry. RFCBT added to treatment-as-usual reduced residual depression and rumination, and comorbid GAD fell from 11 patients at baseline to 1 post-intervention.
Stepped care
NICE CG113. Step 1 psychoeducation and monitoring; Step 2 low-intensity self-help; Step 3 high-intensity CBT or an SSRI; Step 4 specialist services.
The NICE CG113 model: least intrusive effective intervention first, stepping up if it does not work.