Skip to main content

Table 5 Scenario analyses: incremental costs, QALYsDW gained and ICERs for the CCC intervention compared to business-as-usual (percentage difference to expected value analysis in parentheses)

From: Cancer care coordinators in stage III colon cancer: a cost-utility analysis

Scenario Incremental costs per patient (NZ$) QALYsDW gained per patient ICER (NZ$ per QALYDW)
Expected value analysisa $2,250 0.111 $20,200
Varying phases of CCC intervention    
a. CCC from diagnosis to surgery only $80 (−96 %) 0.009 (−92 %) $9,100 (−55 %)
b. CCC from surgery to chemotherapy only $2,170 (−4 %) 0.104 (−6 %) $20,900 (3 %)
Variations to discount rate    
c. 0 % per annum discount rate $2,520 (12 %) 0.148 (33 %) $17,100 (−15 %)
d. 6 % per annum discount rate $2,080 (−8 %) 0.088 (−21 %) $23,600 (17 %)
Variation to epidemiological parameters    
e. Set all DWs (incl pYLDs) to zero (= ‘life years’ gained) $2,250 (0 %) 0.150 (35 %) $15,000 (−26 %)
f. Exclude improved quality of life impact of CCC $2,250 (0 %) 0.100 (−10 %) $22,400 (11 %)
g. Exclude improved survival due to quicker to surgery $2,240 (0 %) 0.107 (−4 %) $21,000 (4 %)
h. Exclude improved survival due to quicker to chemotherapy $2,130 (−5 %) 0.084 (−24 %) $25,200 (25 %)
i. Exclude increasing % of patients getting chemotherapy $800 (−64 %) 0.061 (−45 %) $13,000 (−36 %)
j. Exclude oxaliplatin $2,020 (−10 %) 0.095 (−14 %) $21,300 (5 %)
Variation to cost parameters    
k. Scale all health system costs up 20 % $2330 (4 %) 0.111 (0 %) $20900 (3 %)
l. Scale all health system costs down 20 % $2170 (−4 %) 0.111 (0 %) $19500 (−3 %)
m. Exclude dietician and social worker intervention costs $1730 (−23 %) 0.111 (0 %) $15600 (−23 %)
n. Exclude unrelated health system costs (i.e. include costs up to cure time only) $1780 (−21 %) 0.111 (0 %) $16000 (−21 %)
  1. All models are expected value only; there is no parameter uncertainty
  2. Dollars are NZ$, for the year 2011. Unless stated otherwise, all costs and benefits discounted at 3 % per annum. All values rounded to three meaningful decimal places
  3. pYLDs prevalent years of life lived with disability, which is used as the ‘expected’ amount of morbidity by sex, age and ethnicity
  4. aNote that these results differ slightly from those in Table 3 due to not including uncertainty about parameters (due to long run time of models). Results in Table 3 are the preferred results, but the results in this table should be compared to this expected value analysis which used the same modelling strategy