Methodologies used in cost-effectiveness models for evaluating treatments in major depressive disorder: a systematic review

Zimovetz, Evelina A; Wolowacz, Sorrel E; Classi, Peter M; Birt, Julie

doi:10.1186/1478-7547-10-1

Cost Effectiveness and Resource Allocation

Table 2 Health-state utility values applied in published models in MDD

From: Methodologies used in cost-effectiveness models for evaluating treatments in major depressive disorder: a systematic review

Author, year	Utility value by health state^a	Primary source and method of utility estimation
Armstrong et al., 2008 [9]; Armstrong et al., 2007 [8]	Treated depression: 0.848 Untreated depression: 0.58	Sullivan et al. [39]: EQ-5D^b
Aziz et al., 2005 [10]	Pharmacotherapy: Depression: 0.43 Partial depression: 0.55 Well: 0.75 ECT: Depression: 0.52 Partial depression: 0.66 Well: 0.90	Multiple sources: Hatziandreu et al. [57]; Mazumdar et al. [58]; Judd et al. [59]; McDonald et al. [60], Sackett and Torrence [61]
Benedict et al., 2010 [11]	Remitters: 0.79 Responders: 0.68 Non-responders: 0.55 Staying in remission: 0.86	Multiple sources: Eli Lilly, HMBU trial^c (data on file): EQ-5D; Revicki and Wood [62]: standard gamble^d
Dardennes et al., 2000 [16]	Remission with follow-up Preventive strategy: 0.875 Episodic strategy: 0.895 Remission without follow-up: 0.895^e Recurrence first 2 months: 0.306^e Recurrence months 3 and 4: 0.725^e Recurrence months 5 and 6:0.795^e	Anton and Revicki [63]: standard gamble^f
Nuijten, 2001 [32]	Depression, on treatment SSRI: 0.70, TCA: 0.64 In remission, treatment prolongation SSRI: 0.80, TCA: 0.72 In remission, off treatment: 0.86 Severe depression: 0.30	Revicki and Wood [62]: standard gamble^d
Perlis et al., 2009 [33]	Recovered Not on treatment: 0.88 Disutility of treatment: 0.04 Depressed Not on treatment: 0.63 Disutility on treatment: 0.04	Multiple sources: Bennett et al. [64]; Revicki et al. [65]; Revicki and Wood [62]: standard gamble^d; Schaffer et al. [66]
Sado et al., 2009 [35]	Severe depression: 0.30^g Moderate depression: 0.63^g Response on treatment: 0.80^g Response, no treatment: 0.86^g	Revicki and Wood [62]: standard gamble^d
Simon et al., 2006 [36]	Severe depression: 0.30 Moderate depression: 0.63 Remission, treatment: 0.80 Remission, no treatment: 0.86	Revicki and Wood [62]: standard gamble^d
Sobocki et al., 2008 [37]	Well: 0.86 Episode: 0.57 Remission: 0.81	Sobocki et al. [67]: The study administered EQ-5D questionnaire to 447 patients treated with antidepressant in primary care.
Sullivan et al., 2004 [39]	Treated depression: 0.848 Untreated depression decrement: -0.268 Decrements for ADRs were applied^h	Sullivan et al. [39]: EQ-5D^b

ADR = adverse drug reaction; DFD = disease-free day; ECT = electroconvulsive therapy; EQ-5D = the EuroQol Five Dimension instrument; GI = gastrointestinal; MDD = major depressive disorder; MEPS = Medical Expenditure Panel Survey; QAD = quality-adjusted day; SG = standard gamble, SSRI = selective serotonin reuptake inhibitor; TCA = tricyclic antidepressant; UK = United Kingdom.
^a Note that some of the studies applied utility values but did not report these explicitly. Francois et al. [19] did not report the utility weights, but stated the source for these as Quality of Life Perspective Study, Lundbeck (data on file). Lenox-Smith et al. [28] transformed DFDs into utility weights using the methodology from Lave et al. [68] using the following assumptions: a non-depressed subject was assumed to have utility score of 1.0 (perfect health); a subject with major depression was assumed to have a utility score of 0.59 (estimated from literature); subjects were assumed to gain 0.41 of QAD for each depression-free day. Trivedi et al. [40] transformed DFDs into utility weights using the methodology from Lave et al. [68], assuming a gain of 0.2 to 0.41 of QAD for each DFD.
^b Utility values for each health state were derived from a direct analysis of the data in the 2000 MEPS, which provided individual and variance adjustment weights. The EQ-5D was administered via self-administered questionnaire in MEPS.
^c Utility study based on trials (Eli Lilly, HMBU trial, data on file) derived utility values for remitters, responders, and non-responders from EQ-5D scores of European patients using the UK tariffs.
^d Utility of remitters staying in remission was obtained from Revicki and Wood [62]. This study used SG method to generate utilities for 11 hypothetical health states. Health states were varied by severity and medication; 70 patients with MDD or dysthymia who completed ≥ 8 weeks of antidepressant treatment were recruited.
^e The same values for both episodic strategy and preventative strategy.
^f The SG method was used to generate utilities for hypothetical depression-related health states; 70 MDD patients were interviewed cross-sectionally to provide utilities for these health states.
^g Both values for combination therapy and antidepressant therapy were assumed to be the same.
^h GI: -0.065; Diarrhoea: -0.044; Dyspepsia: -0.086; Nausea: -0.065; Constipation: -0.065; Sexual: -0.049; Excitation: -0.129; Insomnia: -0.129; Anxiety: -0.129; Drowsiness: -0.085; Headache: -0.115; Other (average of all ADRs): -0.085.

Back to article page

ISSN: 1478-7547

Contact us

General enquiries: journalsubmissions@springernature.com