Skip to main content

Table 2 Estimates of 5-year RS and LLE for setting 1 and scenarios A-D. Point estimates (PE) of 5-year relative survival (RS) and loss in life expectancy (LLE), with lower (LCI) and upper (UCI) confidence intervals, standard errors (SE) and relative % precision (RP) from setting 1, different methods and scenarios for including uncertainty in general population mortality when estimating SEs. Results are presented for men and women, aged 55, 65, 75 and 85 years at diagnosis. General population mortality rates are stratified by age, sex and calendar year. RP illustrates comparison of the conventional method to a bootstrap-based method

From: Assessing the impact of including variation in general population mortality on standard errors of relative survival and loss in life expectancy

Method Setting / scenariob 5-year Relative Survival (RS) Loss in life expectancy (LLE)a
   PE SE LCI UCI RPc PE SE LCI UCI RP
men, 55 years at diagnosis
Conventionald 1 0.638 0.026 0.585 0.687 0.00 12.57 0.86 10.90 14.25 0.00
Bootstrap A 0.638 0.026 0.585 0.687 0.00 12.57 0.86 10.90 14.25 0.03
-basede B 0.638 0.026 0.585 0.687 0.01 12.57 0.86 10.89 14.26 0.30
  C 0.638 0.026 0.585 0.687 0.14 12.58 0.88 10.85 14.31 6.22
  D 0.639 0.026 0.585 0.687 1.39 12.57 1.07 10.46 14.68 57.57
men, 65 years at diagnosis
Conventional 1 0.636 0.021 0.593 0.675 0.00 7.69 0.44 6.83 8.54 0.00
Bootstrap A 0.636 0.021 0.593 0.675 0.00 7.69 0.44 6.83 8.54 0.06
-based B 0.636 0.021 0.593 0.675 0.03 7.69 0.44 6.83 8.54 0.60
  C 0.636 0.021 0.593 0.675 0.48 7.69 0.46 6.79 8.60 12.69
  D 0.636 0.022 0.592 0.677 4.83 7.69 0.65 6.42 8.95 120.05
men, 75 years at diagnosis
Conventional 1 0.629 0.021 0.586 0.669 0.00 4.19 0.23 3.74 4.65 0.00
Bootstrap A 0.629 0.021 0.586 0.669 0.01 4.19 0.23 3.74 4.65 0.11
-based B 0.629 0.021 0.586 0.669 0.07 4.19 0.23 3.73 4.65 1.06
  C 0.629 0.021 0.585 0.669 1.40 4.20 0.26 3.70 4.70 22.25
  D 0.629 0.023 0.583 0.672 14.76 4.21 0.42 3.39 5.02 220.92
men, 85 years at diagnosis
Conventional 1 0.585 0.030 0.523 0.641 0.00 1.90 0.13 1.64 2.16 0.00
Bootstrap A 0.585 0.030 0.523 0.641 0.02 1.90 0.13 1.64 2.16 0.12
-based B 0.585 0.030 0.523 0.641 0.21 1.90 0.13 1.64 2.16 1.27
  C 0.584 0.031 0.521 0.642 4.12 1.90 0.15 1.61 2.19 24.31
  D 0.584 0.036 0.511 0.651 40.37 1.92 0.25 1.43 2.40 244.55
women, 55 years at diagnosis
Conventional 1 0.663 0.025 0.611 0.709 0.00 12.06 0.90 10.30 13.83 0.00
Bootstrap A 0.663 0.025 0.611 0.709 0.00 12.06 0.90 10.30 13.83 0.02
-based B 0.663 0.025 0.611 0.709 0.01 12.06 0.90 10.29 13.83 0.19
  C 0.663 0.025 0.611 0.709 0.34 12.06 0.92 10.26 13.86 4.23
  D 0.663 0.025 0.611 0.710 3.74 12.04 1.08 9.92 14.16 44.26
women, 65 years at diagnosis
Conventional 1 0.660 0.020 0.619 0.697 0.00 7.76 0.47 6.84 8.68 0.00
Bootstrap A 0.660 0.020 0.619 0.697 0.00 7.76 0.47 6.84 8.68 0.04
-based B 0.660 0.020 0.619 0.697 0.02 7.76 0.47 6.84 8.69 0.36
  C 0.660 0.020 0.619 0.697 0.44 7.76 0.49 6.81 8.72 7.87
  D 0.660 0.020 0.619 0.698 4.70 7.75 0.63 6.51 8.99 80.65
women, 75 years at diagnosis
Conventional 1 0.653 0.020 0.612 0.692 0.00 4.56 0.27 4.03 5.09 0.00
Bootstrap A 0.653 0.020 0.612 0.692 0.00 4.56 0.27 4.03 5.09 0.07
-based B 0.653 0.020 0.612 0.692 0.03 4.56 0.27 4.03 5.09 0.64
  C 0.653 0.020 0.612 0.692 0.68 4.56 0.29 4.00 5.13 13.01
  D 0.654 0.021 0.611 0.693 6.59 4.56 0.41 3.76 5.37 129.26
women, 85 years at diagnosis
Conventional 1 0.611 0.027 0.556 0.662 0.00 2.20 0.15 1.90 2.50 0.00
Bootstrap A 0.611 0.027 0.556 0.662 0.01 2.20 0.15 1.90 2.50 0.09
-based B 0.611 0.027 0.556 0.662 0.18 2.20 0.15 1.90 2.50 0.90
  C 0.611 0.028 0.555 0.662 3.42 2.20 0.17 1.88 2.52 17.56
  D 0.611 0.031 0.547 0.669 33.92 2.22 0.25 1.72 2.71 175.91
  1. aLLE is presented in years
  2. bSee Table 1 for information on the different settings and scenarios
  3. cRP is presented in %
  4. dConventional refers to the standard method for estimating SEs, where the general population mortality is assumed to be measured without uncertainty
  5. eBootstrap-based refers to the parametric bootstrap approach used for including uncertainty in population mortality rates in the estimation of SEs
Back to article page

BMC Medical Research Methodology

ISSN: 1471-2288

Contact us