Unwanted outcomes in cognitive behavior therapy for pathological health anxiety: a systematic review and a secondary original study of two randomized controlled trials

Figures & data

Figure 1. Flowchart pertaining to the systematic review of randomized controlled trials of cognitive behavior therapy compared to other control conditions for pathological health anxiety. CBT: cognitive behavior therapy; OATD: Open access Theses and Dissertations; RCT: randomized controlled trial.

Table 1. Adverse events and dropout rates reported in randomized controlled trials of cognitive behavior therapy for pathological health anxiety as compared to controls other than cognitive behavior therapy, sorted by year of publication.

						Adverse events					Dropouts			Missing data rate ^a
							At least one, overall		Serious events only			Incidence
First author	Year	Country	Recruitment	CBT format	Control	Method	CBT	Control	CBT	Control	Definition	CBT	Control	CBT	Control
Warwick	1996	UK	Routine care	FTF-CBT	WLC						”dropped out,” ‘treatment refusal’	1/16 = 6%		1/16 = 6%	0/16 = 0%
Clark	1998	UK	Routine care	FTF-CBT	FTF-BSM WLC									1/23 = 4%	0/23 = 0% 0/15 = 0%
Visser	2001	Netherlands	Routine care	FTF-CBT^b	WLC						‘terminating treatment’	10/52 = 19%^f		10/52 = 19%^f	6/26 = 23%
Barsky	2004	USA	Mixed	FTF-CBT	TAU						<4 out of 6 sessions	25%		17/102 = 17%	9/85 = 11%
Buwalda	2007	Netherlands	Unclear	CBGT	PSGT						‘dropped out of the course’	3/25 =  12%	1/23 = 4%	^g	^g
Greeven	2007	Netherlands	Mixed	FTF-CBT	ADM PPL	Closed		^j			‘dropped out of treatment prematurely’	10/40 = 25%	11/37 = 30% 9/35 = 26%
Seivewright	2008	UK	Routine care	FTF-CBT	TAU						‘had no treatment’ or ‘dropped out early’	3/23 = 13%		3/23 = 13%	3/26 = 12%
Hedman	2011	Sweden	Mixed	G-ICBT	AEC						completed <6 out of 12 modules	6/40 = 15%		0/40 = 0%	0/41 = 0%
Sørensen	2011	Denmark	Routine care	Hybrid	FTF-PDT WLC						‘refused to accept treatment’/”dropped out”	3/36 = 8%	7/36 = 19%	4/36 = 11%	9/36 = 25%
Bovell	2012	Canada	Not routine care	G-BIB-CBT	WLC						‘dropped out’	4/10 = 40%		^h
Hedman	2014	Sweden	Mixed	G-ICBT	G-IBSM	Open-ended	12/75 = 16%	8/70 = 11%	0/75 = 0%	0/70 = 0%	completed <6 out of 12 modules	17/79 = 22%	17/79 = 22%	3/79 = 4%	4/79 = 5%
Tyrer	2014	UK	Routine care	FTF-CBT	TAU	Unclear			0/219 = 0%	1/225 = 0%	‘receiving no treatment’	15/219 = 7%		22/219 = 10%	21/225 = 9%
Weck	2015	Germany	Not routine care	FTF-CBT^b	WLC						‘drop-out’	7/75 = 9%		7/75 = 9%	7/42 = 17%
Hedman	2016	Sweden	Mixed	G-ICBT U-ICBT U-BIB-CBT	WLC	Open-ended	6/31 = 19% 3/32 = 9%^c 3/32 = 9%	2/33 = 6%	0/31 = 0% 0/32 = 0% 0/32 = 0%	0/33 = 0%				1/32 = 3% 1/33 = 3% 3/34 = 9%	0/33 = 0%
Fallon	2017	USA	Mixed	FTF-CBT	ADM^d PPL	Closed	w12: 9/53 = 17% w24: 2/53 = 4%	w12: 10/45 = 22% w24: 4/45 = 9% w12: 6/44 = 14% w24: 2/44 = 5%	1/53 = 2%^e	0/45 = 0% 0/44 = 0%	‘dropped out of treatment’	21/53 = 40%	13/45 = 29% 24/44 = 55%	24/53 = 45%^i	17/45 = 38%^i 29/44 = 66%^i
de Meza Skjernov	2017	Denmark	Routine care	CBGT	WLC	Open-ended	8/32 = 25%		0/32 = 0%		<8 out of 13 meetings	6/42 = 14%		2/42 = 5%	0/42 = 0%
Newby	2018	Australia	Unclear	G-ICBT	G-EDU	Open-ended	4/37 = 11%	^k	0/37 = 0%	0/32 = 0%	<4 out of 6 lessons or fact sheets	11/45 = 24%	11/41 = 27%	14/51 = 27%	19/51 = 37%
Morriss	2019	UK	Routine care	R-CBT	WLC	Unclear	0/78 = 0%	0/78 = 0%	0/78 = 0%	0/78 = 0%	<5 out of 6–12 (+ up to 3 booster) sessions	23/78 = 29%		22/78 = 28%	22/78 = 28%
Shabahang	2021	Iran	Not routine care	U-ICBT	WLC						‘dropping out of treatment’	0/76 = 0%		1/76 = 1%	1/76 = 1%

None of these randomized controlled trials reported deterioration rates. ADM: antidepressant medication; AEC: attention/engagement control; CBGT: cognitive behavioral group therapy; CBT: cognitive behavior therapy; FTF-BSM: face-to-face behavioral stress management; FTF-CBT: face-to-face cognitive behavior therapy; FTF-PDT: face-to-face psychodynamic therapy; G-BIB-CBT: guided cognitive behavioral bibliotherapy; G-EDU: guided education about pathological health anxiety; G-IBSM: guided Internet-delivered behavioral stress management; G-ICBT: guided Internet-delivered cognitive behavior therapy; R-CBT: remote cognitive behavior therapy other than pure bibliotherapy or only via the Internet; PPL: pill placebo; PSGT: problem-solving group therapy; TAU: treatment-as-usual; U-BIB-CBT: unguided cognitive behavioral bibliotherapy; U-ICBT: unguided Internet-delivered cognitive behavior therapy; WLC: waitlist control.

^aMissing data rates are tabulated here because these have sometimes been used as a proxy measure of dropout status. We tabulated rates at the post-treatment assessment pertaining to the primary self-report health anxiety outcome, or the self-report health anxiety outcome that was reported first in the primary publication. For Warwick et al. (1996) and Clark et al. (1998), we tabulated missing data rates pertaining to the ‘Time spent worrying about health’ and ‘Time seriously worried about health’ visual-analogue scales.

^bThese trials compared variations on FTF-CBT, focused either on cognitive strategies or exposure, to each other.

^cIncorrectly listed as 4 in the original publication.

^dThis trial also administered a FTF-CBT plus ADM combination which was not tabulated in this overview of CBT versus non-CBT controls.

^e‘gastrointestinal problems.’

^fIn the original publication, a drop-out rate of 22/78 = 28% was presented based on the pooling of participants randomized to CBT and those crossed over from the WLC.

^gIn the original publication, missing data rates were listed for treatment completers only, and pertaining to the combination of all assessment points only.

^hAt least 4/10 in CBT.

ⁱIn this trial, participants with insufficient response were withdrawn from the protocol which accounted for 5/53 in FTF-CBT, 4/45 in ADM, and 5/44 in PPL.

^jThis trial reported the frequency of each type of adverse event, but not for CBT, and not the proportion of participants reporting at least one adverse event.

^kIn the original publication, 29 are reported as not having experienced an adverse event.

Figure 2. Flowchart pertaining to the two randomized controlled trials of cognitive behavior therapy for pathological health anxiety on which the secondary original study was based.

Table 2. Baseline characteristics in two randomized controlled trials of cognitive behavior therapy for pathological health anxiety.

	FTF-CBT	G-ICBT	U-ICBT	BIB-CBT	WLC
	(n = 102)	(n = 134)	(n = 33)	(n = 34)	(n = 33)
Clinical trial	II	I + II	I	I	I
Sociodemographics
Age, M (SD)	39 (13)	39 (12)	37 (12)	35 (12)	42 (14)
Female	71 (70%)	96 (72%)	24 (73%)	24 (71%)	26 (79%)
Post-secondary education	78 (76%)	100 (75%)	27 (82%)	28 (82%)	27 (82%)
Employed full or part time	65 (64%)	94 (70%)	27 (82%)	21 (62%)	22 (67%)
Clinical variables
Health anxiety (HAI-18), M (SD)	34.2 (6.4)	33.7 (6.5)	34.4 (8.0)	36.3 (6.7)	34.4 (6.8)
Health anxiety (HAI-18), range	20–51	15–51	16–46	20–53	18–44
Severity of pathological health anxiety^a
Mild	43 (42%)	63 (47%)	13 (39%)	11 (32%)	10 (30%)
Moderate	40 (39%)	52 (39%)	11 (33%)	15 (44%)	18 (55%)
Substantial	19 (19%)	19 (14%)	9 (27%)	8 (24%)	5 (15%)
Depression (MADRS-S), M (SD)	14.7 (7.0)	13.4 (7.0)	13.8 (7.9)	14.4 (7.6)	12.7 (6.8)
Disability (SDS), M (SD)	11.6 (6.8)	11.1 (7.3)	9.2 (7.3)	10.8 (6.9)	11.9 (7.4)

BIB-CBT: cognitive behavioral bibliotherapy; FTF-CBT: face-to-face cognitive behavior therapy; G-ICBT: therapist-guided Internet-delivered cognitive behavior therapy; HAI-18: 18-item Health Anxiety Inventory; MADRS-S: Montgomery-Åsberg Depression Rating Scale – Self-report version; SDS: Sheehan Disability Scale; U-ICBT: unguided Internet-delivered cognitive behavior therapy; WLC: waiting-list control.

^aBased on the first 14 items of the HAI-18, in accordance with Österman et al. (2022).

Table 3. Adverse events in two randomized controlled trials of cognitive behavior therapy for pathological health anxiety.

	FTF-CBT	G-ICBT	U-ICBT	BIB-CBT	Pooled CBT (any)
Participants reporting ≥ 1 adverse event	17/97 (18%)	25/127 (20%)	3/32 (9%)^a	3/32 (9%)	48/288 (17%)
Number of adverse events	18	27	3	3	51
Number of serious adverse events^b	0	0	0	0	0
Classification of adverse events
Distress/deterioration	17/18 (94%)	27/27 (100%)	2/3 (67%)	3/3 (100%)	49/51 (96%)
Relatedness to treatment (1–5)	3.9 (0.8)	3.9 (0.9)	2.5 (0.7)	4.3 (0.6)	3.9 (0.9)
Distress at adverse event (0–3)	2.5 (0.6)	1.9 (0.8)	2.0 (0.0)	3.0 (0.0)	2.2 (0.8)
Distress at end of treatment (0–3)	1.3 (0.9)	0.9 (0.9)	1.0 (1.4)	1.3 (1.2)	1.0 (0.9)
Failure	0/18 (0%)	2/27 (7%)	0/3 (0%)	1/3 (33%)	3/51 (6%)
Relatedness to treatment (1–5)	N/A	4.8 (0.4)	N/A	5.0 (N/A)	4.8 (0.3)
Distress at adverse event (0–3)	N/A	2.0 (0.0)	N/A	3.0 (N/A)	2.3 (0.6)
Distress at end of treatment (0–3)	N/A	1.5 (0.7)	N/A	2.0 (N/A)	1.7 (0.6)
Hopelessness	0/18 (0%)	0/27 (0%)	0/3 (0%)	1/3 (33%)	1/51 (2%)
Relatedness to treatment (1–5)	N/A	N/A	N/A	5.0 (N/A)	5.0 (N/A)
Distress at adverse event (0–3)	N/A	N/A	N/A	3.0 (N/A)	3.0 (N/A)
Distress at end of treatment (0–3)	N/A	N/A	N/A	2.0 (N/A)	2.0 (N/A)
Insufficient quality of treatment	0/18 (0%)	3/27 (11%)	0/3 (0%)	0/3 (0%)	3/51 (6%)
Relatedness to treatment (1–5)	N/A	5.0 (0.0)	N/A	N/A	5.0 (0.0)
Distress at adverse event (0–3)	N/A	2.3 (0.6)	N/A	N/A	2.3 (0.6)
Distress at end of treatment (0–3)	N/A	1.0 (1.0)	N/A	N/A	1.0 (1.0)
Stigma	0/18 (0%)	0/27 (0%)	1/3 (33%)	0/3 (0%)	1/51 (2%)
Relatedness to treatment (1–5)	N/A	N/A	3.0 (N/A)	N/A	3.0 (N/A)
Distress at adverse event (0–3)	N/A	N/A	1.0 (N/A)	N/A	1.0 (N/A)
Distress at end of treatment (0–3)	N/A	N/A	0.0 (N/A)	N/A	0.0 (N/A)
Unclear	1/18 (6%)	0/27 (0%)	0/3 (0%)	0/3 (0%)	1/51 (2%)

Participants completed free-text fields to report adverse events after treatment. The events were then classified by two independent raters. More than one category could be used for each adverse event (free-text description). BIB-CBT: cognitive behavioral bibliotherapy; CBT: cognitive behavior therapy; FTF-CBT: face-to-face cognitive behavior therapy; G-ICBT: therapist-guided Internet-delivered cognitive behavior therapy; N/A: not applicable; U-ICBT: unguided Internet-delivered cognitive behavior therapy; WLC: waiting-list control.

^aIncorrectly reported as 4 in the primary publication.

^bLeading to death or requiring immediate care or hospitalization.

Table 4. Deterioration and dropouts in two randomized controlled trials of cognitive behavior therapy for pathological health anxiety.

	FTF-CBT	G-ICBT	U-ICBT	BIB-CBT	WLC
Deterioration status
Primary: reliable, r = 0.87, SD × 1, α = 0.05	0/97 (0%)	1/128 (1%)	0/32 (0%)	0/31 (0%)	0/33 (0%)
Sensitivity 1: reliable, but r = 0.70	0/97 (0%)	0/128 (0%)	0/32 (0%)	0/31 (0%)	0/33 (0%)
Sensitivity 2: reliable, but SD × 0.84, α = 0.20	2/97 (2%)	2/128 (2%)	1/32 (3%)	3/31 (10%)	2/33 (6%)
Sensitivity 3: 30% deterioration	0/97 (0%)	1/128 (1%)	0/32 (0%)	0/31 (0%)	0/33 (0%)
Sensitivity 4: ≥ 3 point sum score increase	2/97 (2%)	3/128 (2%)	1/32 (3%)	3/31 (10%)	3/33 (9%)
Dropout status
Primary: ≥ 3 weeks inactivity and did not return^a	No data	19/102 (19%)	No data	No data	N/A
… of which began week 1–6	No data	12/19 (63%)	No data	No data	N/A
… of which began week 7–12	No data	7/19 (37%)	No data	No data	N/A
Sensitivity 1: ≤ 5 sessions or modules	10/102 (10%)	24/134 (18%)	9/33 (27%)	5/34 (15%)	N/A
Sensitivity 2: missing post-treatment	5/102 (5%)	6/134 (4%)	1/33 (3%)	3/34 (9%)	0/33 (0%)^b

BIB-CBT: cognitive behavioral bibliotherapy; CBT: cognitive behavior therapy; FTF-CBT: face-to-face cognitive behavior therapy; G-ICBT: therapist-guided Internet-delivered cognitive behavior therapy; N/A: not applicable; U-ICBT: unguided Internet-delivered cognitive behavior therapy; WLC: waitlist control.

^aDue to insufficient documentation, this could only be calculated for G-ICBT in Clinical trial II (n = 102).

^bIn the waitlist control condition, completion of the post-treatment assessment was a requirement for being crossed over to treatment.

Table 5. Potential predictors of adverse events and dropouts in two randomized controlled trials of cognitive behavior therapy for pathological health anxiety.

			Dropout status ^a
	Adverse events		Primary analysis	Sensitivity analysis
	At least one reported, open-ended question		≥3 weeks inactivity and did not return	Initiated ≤5 modules
	FTF-CBT	G-ICBT	G-ICBT	G-ICBT
Frequency of outcome	17/97 (18%)	25/127 (20%)	19/102 (19%)	24/134 (18%)
	OR (95% CI)	OR (95% CI)	OR (95% CI)	OR (95% CI)
Sociodemographic predictors
Age (years)	1.01 (0.97–1.05)^c	1.00 (0.96–1.03)^c	0.96 (0.91–1.00)^c	0.95 (0.91–1.00)^c
Female	1.64 (0.49–5.49)	2.33 (0.75–7.31)	0.50 (0.18–1.39)	0.38 (0.15–0.96)*
Post-secondary education	2.94 (0.62–13.97)	0.87 (0.33–2.30)	0.29 (0.10–0.81)*	0.31 (0.12–0.79)*
Clinical predictors
Health anxiety (HAI-18)	1.00 (0.92–1.08)^c	1.01 (0.94–1.08)^c	1.05 (0.97–1.14)^c	1.01 (0.94–1.08)^c
Depression (MADRS-S)	0.93 (0.85–1.01)^c	1.04 (0.97–1.10)^c	1.07 (1.00–1.15)^c	1.02 (0.96–1.09)^c
Therapy-naive (psychological)^b	2.86 (0.36–22.48)	0.76 (0.22–2.64)	4.30 (0.53–34.62)	4.92 (0.61–39.47)
Credibility/expectancy (C/E scale)	1.04 (0.95–1.14)^c	0.99 (0.05–2.58)^c	0.89 (0.83–0.95)* c	0.92 (0.86–0.97)* c

All tests were conducted on multiply imputed data. Prediction of deterioration status could not be modeled due to the low incidence. C/E scale: Credibility/Expectancy scale; FTF-CBT: face-to-face cognitive behavior therapy; G-ICBT: therapist-guided Internet-delivered cognitive behavior therapy; HAI-18: 18-item Health Anxiety Inventory; MADRS-S: Montgomery-Åsberg Depression Rating Scale – Self-report version; OR: odds ratio.

^aAgreement between the two definitions was 92% (κ = 0.67) in FTF-CBT and 94% (κ = 0.81) in G-ICBT, i.e. substantial but not perfect. Prediction of dropout status in terms of ≥3 weeks inactivity and non-return could not be modeled for FTF-CBT due to insufficient documentation. Prediction of dropout status in terms of non-completion, i.e. ≤5 sessions, could not be modeled for FTF-CBT due to the low incidence (10/102, 10%).

^bThis information was available with regard to participants in Clinical trial II only.

^cORs derived from analyses of continuous predictors stand for the proportion of odds as a function of a one-unit increase in the predictor.

*P < 0.05.

Data availability statement

Data derived from the systematic review are reported in Table 1 and will be made freely available via the Open Science Framework upon acceptance (https://osf.io/pa89t/). Individual participant data from the secondary original study fall under Swedish and European Union data protection and privacy legislation and can therefore not be share in their entirety. Reasonable requests may be directed to the corresponding author, and will be considered on a case-by-case basis.