Table 6 Cost-effectiveness Outcomes: Cancer

Arving 2014 [34*]

CUA

ICER (cost per QALY gained):

Both INS & IPS dominated usual care. The differences between the INS and SC were estimated as of €-7130 (95% CI €-4286 to €-11,532) and between IPS and SC €-5381 (95% CI €-2732 to €-9524), respectively.

Parameters varied

Subgroup analysis namely: low and high tolerance levels re: QALY gain, tumour size, lymph node metastases, outliers in the number of intervention sessions, outliers in hospital activity. None except no regional lymph node metastases changed the conclusion.

Cost effective because the health care costs were lower and QALYs were higher compared to usual care alone (dominant).

Mean (SD) utility value change over 2 year timeframe

INS: 0.26 (0.20); IPS:0.17 (0.26); Control: 0.20 (0.24)

Intervention costs

€500 (or 3%of the total costs).

Bootstrapping Y/N (iterations):

Y; 1000 replications

Chatterton et al. 2016 [41*]

CUA

Intervention costs (high distress; low distress):

Intervention (psychologist led): $202; Control (nurse led): $60

Intervention: $181; Control: $60 Between group differences were significant at the 0.05 level.

QALYs derived from AQoL-8D (range depending on distress level)

Intervention: 0.614 to 0.760; Control: 0.577 to 0.744; Not significantly different. 0.037 (95% CI: 0.045 to 0.118) high distress and 0.016 (95% CI:0.027 to 0.060) low distress

ICER (cost per QALY gained):

Intervention dominates for high distress patients. ICER: AUD$20,937.50 for low distress patients.

Bootstrapping Y/N (iterations):

Y, 50,000 iterations.

The height of the curve would need to be above 97.5% to be confident that the PI is a good value compared with the NI.

PI is likely to be cost-effective compared with the NI for highly distressed cancer patients...conclusions for low-distress patients/carers support the use of the nurse-led self-management intervention

Total costs (high distress; low distress):

WTP Threshold(s)

AUD$50000 per QALY was taken as the benchmark for cost-effectiveness in Australia

Probability cost effective at WTP threshold:

81% at WTP threshold of AUD$50000 for high distress patients. 73% for low distress patients at the same threshold.

Jansen et al. 2017 [58*]

CUA

ICER (cost per QALY gained):

Intevention dominant as had higher QALY’s and statisically signficant lower cumulative costs.

Bootstrapping Y/N (iterations):

Y (5000 replications)

Stepped care likely to be cost effective

Costs without productivity losses:

Intervention: €6287; Control: €9175; Difference: −2888 (95% CI: − 5630 to − 424 P < 0.05)

QALYs gained (without productivity losses):

Intervention: 0.885; Control: 0.767; Difference: 0.118 (95%CI: 0.009 to 0.227 P < 0.05)

Probability cost effective at WTP threshold:

96% of iterations in south-east quadrant of cost-effectiveness plane.

Johannsen et al. 2017 [60*]

CEA

n/N (%) achieving clinically relevant (2 point reduction in 0-10 scale) change in self-reported pain intensity

Intervention: 19/36 (52.8%); Control: 14/48 (29.2%); OR: 2.71 (higher odds of achieving MCID), p = 0.03

ICER (per additional MCID reduction in self-reported pain scale):

Intervention dominates as has lower costs and higher odds of achieving MCID

Bootstrapping Y/N (iterations):

Y; (1000 iterations).

Cost-effective as 2.71 higher odds of achieving minimal clinically important difference and lower cost

Intervention cost

Intervention: €240 per MBCT participant Control: N/A

WTP Threshold(s)

€0, €1000

Probability cost effective at WTP threshold:

At €0 per additional participant meeting MCID, MBCT was cost-effective with a probability of 85%. At a WTP of €1000 per additional participant with MCID, MBCT was cost-effective with a probability of 90%.

Lengacher et al. 2015 [67*]

CUA (SF-12)

QALY gain at 12 weeks

Intervention: 0.033; Control: 0.021; Incremental gain 0.03 (95%, confidence interval [CI] = 0.02-0.04).

ICER (cost per QALY gained

$22,200 (healthcare perspective); $19,733 (patient/out of pocket perspective)

Parameters varied

ICERs were calculated with the upper and lower bounds of the 95% CI for both costs and MBSR(BC) effects. Assumed effect is sustained over longer time horizon and explored impact on ICERs

Appears to provide for significantly improved HRQOL at a comparatively low cost (fairly reasonable conclusion although could have provided a better sensitivity analysis to confirm).

Mean (SD) patient opportunity costs

$592 ($494)

Mewes et al. 2015 [77*]

CEA and CUA (SF-36) using Markov model with hypothetical cohort of 1000 and time horizon of 5 years.

Cost (per clinically relevant change in FACT-ES):

CBT: €1051, PE: €1315.

Parameters varied

PSA were propagated through the model using 5000 Monte Carlo simulations

In relative terms, CBT is likely the most cost-effective strategy compared to PE and control but results sensitive to uncertainties so overall cost-effectiveness uncertain.

Intervention costs (including labour, training, admin, materials):

CBT: €190, PE:€197

Cost (per clinically relevant change in HFRS):

CBT: €1067, respectively PE: No clinically relevant difference seen between PE and the control.

WTP Threshold(s)

€20 k to €80 k, with €30 k per QALY commonly accepted as the prevailing ceiling ratio

Total QALY gain

CBT: 4.400; PE: 4.399; Control: 4.392

ICER (cost per QALY gained):

CBT: Incremental cost/QALY €22,502; PE: Incremental cost /QALY €28,087;

Probability cost effective at WTP threshold:

PE has the highest probability of being cost effective up to WTP values of €26,000/QALY above which CBT has the highest probability of being cost-effective, with a probability of 49% at a ceiling ratio of €30,000/QALY, up to 56% at €80,000/QALY

Prioli et al. 2017 [87*]

CUA (SF-6D)

Intervention (MBAT): $992.49; Control (BCSG): $562.71

Difference between groups $429.79

ICER (cost per QALY gained):

MBAT: $196,236 compared with baseline.

BCSG: $128, 404 compared with baseline;

Parameters varied

Included cost components were varied. Yielded MBAT costs ranging from $241 to $792 (varying session leaders and art supply costs Other sensitivity analyses suggested that if if the session leader cost is less than $550, MBAT can be less costly than a BCSG.

An MBAT intervention is more costly than usual support group care and has a similar effect on utility as a BCSG (i.e. not likely to be cost effective).

Intervention cost included screening, labour, materials, staff travel costs & those of participants (varied in sensitivity analysis):

QALY gain

Intervention: 0.00433;

Control: 0.00433

van der Spek et al. 2018 [97*]

CUA (EQ-5D)

Mean (SE) change in utility score:

MCGP-CS: 0.540 (0.016). SGP: 0.511 (0.014); Control: 0.507 (0.014); Difference between MCGP-CS vs control: 0.033 (95%CI:− 0.007 to 0.074); Difference between SGP vs control: 0.004 (95%CI:− 0.036 to 0.044).

ICER (cost per QALY gained):

MCGP was dominant (lower costs and more QALYs gained).

Parameters varied

Complete case analysis and costs/effects at different time point.

MCGP-CS is highly likely a cost-effective intervention (likely but there is considerable uncertainty and the sensitivity analysis could have explored this in more detail).

Intervention costs per patient:

MCGP-CS: €288; SGP: €286; Control: N/A

WTP Threshold(s)

€0 to €30,000.

Probability cost effective at WTP threshold:

At €0, MCGP-CS has a 78% probability of being cost-effective compared to CAU, increasing to 85% at €10,000 and to 92% at €30,000. At €0 SGP has an 80% probability of being cost-effective compared to CAU, this does not increase if society is willing to pay more.

Compared to SGP, MCGP-CS has a 52% probability of being cost-effective at €0, increasing to 63% at €10,000 and to 77% at €30,000.

Zhang & Fu 2016 [106*]

CUA (EQ-5D)

Incremental change in EQ-5D score

BF + group vs INP = 0.054 p < 0.05; BF + group vs control (usual care) 0.008 (95% CI:0.041, 0.058) p = 0.74 BF + phone = 0.057. p < 0.05 BF + phone versus control (usual care) 0.016 (95% CI: 0.033, 0.065) p = 0.53; Results are significant compared to non-participating group but not usual care.

ICER (cost per QALY gained):

Provider and patient ICERs were $16,759 and $12,561/QALY for support and telephone groups respectively. (Societal) ICERs compared with non-participating group was $17,276 for BF + group and $11,612 for BF + phone. No further analysis against usual care as results were not significantly different for this group.

WTP Threshold(s)

$50,000/QALY, the consensus threshold to determine cost-effectiveness for society.

The interventions of BF+ group or BF + phone were cost-effective compared with those of patients who were eligible but declined (INP group) participation (really depends on INP group motivations).