Skip to main content
. 2018 Sep 12;16:179. doi: 10.1186/s12955-018-0994-8

Table 4.

Disutilities and decrements for adverse event health states in patients with previously treated mNSCLC

Relevance Author, year, country, referencea Instrument and respondent Utility type Health state/disutility Mean HSUV (SD) [SE] {95% CI} HTA suitability
Advanced/mNSCLC Nafees 2008, UK [69] SG
General public
UID Stage IV NSCLC, 2 L, stable disease Does not meet HTA body reference case as vignette-based utility completed by general public respondents. Has been used in multiple HTA submissions, however. Specifically 2 L and UK, good sample size (n=100) and measure of dispersion available.
+ Diarrhoea 0.61
+ Fatigue 0.58
+ Febrile neutropenia 0.56
+ Hair loss 0.61
+ Nausea/vomiting 0.61
+ Neutropenia 0.56
+ Rash 0.62
Stage IV NSCLC, 2 L, responding disease
+ Diarrhoea 0.63
+ Fatigue 0.60
+ Febrile neutropenia 0.58
+ Hair loss 0.63
+ Nausea/vomiting 0.62
+ Neutropenia 0.58
+ Rash 0.64
D Stage IV NSCLC, 2 L,
moving from stable to progressive
−0.18 [0.022]
Neutropenia −0.09 [0.015]
Febrile neutropenia −0.09 [0.016]
Fatigue −0.07 [0.018]
Nausea and vomiting −0.05 [0.016]
Diarrhoea −0.05 [0.016]
Hair loss −0.04 [0.015]
Rash −0.03 [0.012]
Response gain 0.02 [0.007]
Tabberer 2006, UK [52] EQ-5D (tariff NR but likely UK TTO tariff as UK sample)
General public
D Compared with stable disease (advanced NSCLC, line not specified) Not suitable as general public respondents, line of treatment not specified, and no measure of dispersion reported. Good sample size, however (n=154) and Nafees et al. 2008 in 2 L does not provide disutilities for neuropathy or stomatitis so these values from Tabberer et al. may be the best available.
Febrile neutropenia −0.27
Rash −0.06
Neuropathy −0.15
Neutropenia −0.14
Nausea −0.14
Stomatitis −0.14
Diarrhoea −0.13
Doyle 2008, UK [65] SG
General public
UID Metastatic NSCLC, line not specified, SDis no additional symptoms Does not meet reference case as general public respondents. However, these are the only disutilities for severe symptoms for cough, dyspnoea and pain in mNSCLC, so are best option in spite of not meeting HTA derivation method preferences (n=101)
+ Cough 0.58
+ Dyspnoea 0.58
+ Pain 0.56
+ Cough, dyspnoea and pain 0.46
D Cough −0.05 [0.011]
Dyspnoea −0.05 [0.012]
Pain −0.07 [0.012]
Cough, dyspnoea and painc − 0.17 b
Responding disease gain vs SDis 0.09 [0.015] As line not specified, data from Nafees et al. 2008 should be used in preference.
Handorf 2012, USA [70] Expert opinion UID Stage IV NSCLC adenocarcinoma (1 L SDis)
+ neutropenia 0.67 Does not meet reference case as expert opinion-derived. This AE is covered by Nafees et al. 2008, which uses a better derivation method than expert opinion.
+ pneumothorax 0.63 Does not meet reference case as expert opinion-derived, but these are the only estimates for these AE health states. SDis estimate was 0.670 (oral therapy) and 0.653 (i.v. chemotherapy) for disutility calculation.
+ haemorrhage 0.63
+ thrombocytopenia 0.65
+ thrombosis 0.56
Earlier stage NSCLC (curative intent) Grutters 2010, country NR [44] EQ-5D-5 L (tariff NR)
Patients
UID NSCLC, curative intent stage, line not specified, grade III+ dyspnoea 0.52 (median) Patient-derived EQ-5D but tariff and measure of dispersion NR. Only source of grade III+ dyspnea. Utility for NSCLC patients without dyspnoea in this sample was 0.81, i.e. disutility − 0.29
Advanced/mLC (NSCLC+SCLC) Yokoyama 2013, Japan [55] EQ-5D (tariff NR)
Patients
D Stage IIIB/IV NSCLC/SCLC with bone metastases + skeletal related event (pathologic fracture, radiation or surgery to bone lesion, spinal cord compression or hypercalcaemia) − 0.05d Provides NSCLC/SCLC (mixed) patient-derived EQ-5D decrement for (mixed) SREs. Data for these AEs are limited, so although this estimate is not robust (n=9 and response % low at 32%) it does provide an indication. No variability measure reported
Breast cancer and lung cancer Grunberg 2009, USA [58] SG
Patients
UID Base state: continuous nausea and vomiting 0.53f Nafees et al. 2008 provide data for nausea and vomiting but if different levels of nausea and vomiting need to be discerned then these utilities can be considered. Patient-derived SG but mixed lung/breast cancers. Good sample size (n=96) but no measure of dispersion. As the Copyright fee for Grunberg 2009 was high, these data are reported from Shabaruddin 2013 [79] (a previous SR that extracted the graphical data from Grunberg 2009)
Increment Limited nausea and limit vomiting vs continuous nausea and vomiting + 0.53f
Increment Limited nausea vs continuous nausea and vomiting + 0.55f
Increment Limited vomiting vs continuous nausea and vomiting + 0.50f
Advanced Cancer Matza 2014, UK [67] TTO
General public
U Stage IV cancer with bone metastases (no skeletal-related events) 0.47 (0.41) Does not meet reference case as general population respondents. However, as there are no alternative utilities for bone metastases these UK utilities could be considered for NICE or SMC. Good sample size (n=126), SD available.
UID + spinal cord compression without paralysis 0.25 (0.50)
+ spinal cord compression with paralysis 0.13 (0.49)
+ fracture of the leg 0.42 (0.41)
+ fracture of the rib 0.44 (0.42)
+ fracture of the arm 0.44 (0.41)
+ radiation treatment (2 weeks, 5 appointments/week) 0.42 (0.42)
+ radiation treatment (2 appointments) 0.45 (0.41)
+ surgery to stabilize bone 0.40 (0.44)
Matza 2014, Canada [67] TTO
General public
U Stage IV cancer with bone metastases (no skeletal-related events) 0.47 (0.45) Does not meet reference case as general population respondents. However, as there are no alternative utilities for bone metastases these Canadian utilities could be considered for CADTH. Reasonable sample size (n=61), SD available.
UID + spinal cord compression without paralysis 0.25 (0.54)
+ spinal cord compression with paralysis 0.19 (0.53)
+ fracture of the leg 0.40 (0.48)
+ fracture of the rib 0.43 (0.47)
+ fracture of the arm 0.43 (0.48)
+ radiation treatment (2 weeks, 5 appointments/week) 0.41 (0.50)
+ radiation treatment (2 appointments) 0.45 (0.45)
+ surgery to stabilize bone 0.39 (0.50)
Matza 2014, UK and Canada [67] TTO
General public
U Stage IV cancer with bone metastases (no skeletal-related events) 0.47 (0.42) Does not meet reference case as general population respondents. However, as there are no alternative utilities for bone metastases these UK+Canadian utilities could be considered for NICE, SMC or CADTH. Good sample size (n=187), SD available.
UID + spinal cord compression without paralysis 0.25 (0.21)
+ spinal cord compression with paralysis 0.15 (0.50)
+ fracture of the leg 0.41 (0.43)
+ fracture of the rib 0.44 (0.43)
+ fracture of the arm 0.43 (0.43)
+ radiation treatment (2 weeks, 5 appointments/week) 0.41 (0.45)
+ radiation treatment (2 appointments) 0.45 (0.42)
+ surgery to stabilize bone 0.40 (0.46)
Matza 2014, UK [67] TTO
General public
D Stage IV cancer with bone metastases As above
+ spinal cord compression without paralysis −0.22 (0.31)
+ spinal cord compression with paralysis −0.34 (0.36)
+ fracture of the leg −0.05 (0.09)
+ fracture of the rib −0.03 (0.08)
+ fracture of the arm −0.03 (0.07)
+ radiation treatment (2 weeks, 5 appointments/week) −0.05 (0.12)
+ radiation treatment (2 appointments) −0.02 (0.07)
+ surgery to stabilize bone −0.07 (0.17)
Matza 2014, Canada [67] TTO
General public
D Stage IV cancer with bone metastases As above
+ spinal cord compression without paralysis −0.22 (0.32)
+ spinal cord compression with paralysis −0.28 (0.30)
+ fracture of the leg −0.07 (0.19)
+ fracture of the rib −0.04 (0.17)
+ fracture of the arm −0.04 (0.07)
+ radiation treatment (2 weeks, 5 appointments/week) −0.06 (0.21)
+ radiation treatment (2 appointments) −0.02 (0.11)
+ surgery to stabilize bone −0.08 (0.21)
Matza 2014, UK and Canada [67] TTO
General public
D Stage IV cancer with bone metastases As above
+ spinal cord compression without paralysis −0.22 (0.31)
+ spinal cord compression with paralysis −0.32 (0.34)
+ fracture of the leg −0.06 (0.13)
+ fracture of the rib −0.03 (0.12)
+ fracture of the arm −0.04 (0.11)
+ radiation treatment (2 weeks, 5 appointments/week) −0.06 (0.15)
+ radiation treatment (2 appointments) −0.02 (0.08)
+ surgery to stabilize bone −0.07 (0.18)
Cancer, unclear stage Lloyd 2008, UK [59] SG
General public
UID Anaemia associated with cancer treatment Does not meet reference case as general population sample respondent for SG exercise.
Haemoglobin level (g/dL) 7.0–8.0 0.58 {0.067}
8.0–9.0 0.61 {0.064}
9.0–10.0 0.64 {0.060}
10.0–10.5 0.64 {0.062}
10.5–11.0 0.66 {0.061}
11.0–12.0 0.70 {0.056}
>12.0 0.71 {0.057}
VAS
General public
UID Haemoglobin level (g/dL) 7.0–8.0 16.9 {2.6}
8.0–9.0 22.3 {3.0}
9.0–10.0 27.6 {2.9}
10.0–10.5 32.9 {3.4}
10.5–11.0 38.8 {3.6}
11.0–12.0 45.9 {4.2}
>12.0 51.2 {4.3}
TTO
Cancer patients with recent experience of chemotherapy-related anaemia or fatigue
UID Haemoglobin level (g/dL) 7.0–8.0 0.30 {0.127}
8.0–9.0 0.36 {0.126}
9.0–10.0 0.41 {0.125}
10.0–10.5 0.45 {0.122}
10.5–11.0 0.45 {0.111}
11.0–12.0 0.55 {0.105}
>12.0 0.61 {0.112}
U Own current health 0.85 {0.034}
EQ-5D current health 0.87 {0.076}
VAS
Cancer patients with recent experience of chemotherapy-related anaemia or fatigue
UID Haemoglobin level (g/dL) 7.0–8.0 21.7 {5.7}
8.0–9.0 32.4 {6.6}
9.0–10.0 34.2 {6.7}
10.0–10.5 41.9 {6.6}
10.5–11.0 44.7 {6.6}
11.0–12.0 52.2 {6.8}
>12.0 62.4 {7.9}
U Own current health 87.6 {4.9}
EQ-5D current health 84.2 {4.57}
NR Westwood 2014, NR [71] NR D Anaemia 0.073 [0.018] Disutilities for anaemia and treatment mode have been used in previous NICE submissions. However, there is no information concerning their derivation.
2 L NSCLC Oral therapy (ERL) 0.014 [0.012]
2 L NSCLC i.v. therapy 0.043 [0.020]
Patients without NSCLC
1 L setting
Nafees 2016, Multinational and UK [68]g TTO
Patients (but not NSCLC patients) from the general public in UK, Australia, France, China, S. Korea, Taiwan
D Bleeding vs BL (stable no side effects) −0.25 No
Hypertension vs BL (stable no side effects) −0.03
UID Responding + bleeding vs BL 0.534
Responding + hypertension vs BL 0.749
Stable + bleeding vs BL 0.508
Stable + hypertension vs BL 0.729

aSome studies identified in this systematic review were not included in this table for the following reasons: Grunberg 2009 [58] and Grutters 2010 [44] did not report values

bThe italics indicate a calculated utility. This is calculated from the values reported in Doyle 2008 [65] Table 3 (difference between stable disease no other symptoms and stable disease with cough, dyspnoea and pain, to obtain the disutility). It is not calculated by adding the disutilities, as this would not be valid

cValues were calculated from ‘SDis + cough, dyspnea and pain’ utility minus ‘SDis no additional symptoms’ utility

dThe study did not indicate if mean or median

eSome studies identified in this systematic review were not included in this table for the following reasons: Grunberg 2009 [58] and Grutters 2010 [44] did not report values, and Nafees 2016 [68] reported variation in two decrements (bleeding, hypertension) based on the different populations in which valuation was undertaken

fAs reported in Shabarruddin 2013 [79], base state and utility increments were presented on different scales: base state was based on standard gamble scale between perfect health (arbitrary score of 100) or immediate death (arbitrary score of 0) while the utility increments were based on a scale between perfect health (arbitrary score of 100) and the surrogate negative anchor of continuous nausea/vomiting (re-set to an arbitrary score of 0)

gValues are presented for global population (United Kingdom, Australia, France, China, Taiwan, Korea). Note that country-specific data are also available

Abbreviations: 1 L first line, BL baseline, CI confidence interval, D decrement, ERL erlotinib, HSUV health state utility value, HTA health technology assessment, i.v. intravenous, LC lung cancer, mLC metastatic lung cancer, mNSCLC metastatic non-small cell lung cancer, NICE National Institute for Health and Care Excellence, NR not reported, NSCLC non-small cell lung cancer, SCLC small cell lung cancer, SD standard deviation, SE standard error, SG standard gamble, TTO time trade-off, U utility, UID utility incorporating decrement for adverse events, VAS visual analogue scale