Aggregate DCEA Tutorial

The hardware and bandwidth for this mirror is donated by dogado GmbH, the Webhosting and Full Service-Cloud Provider. Check out our Wordpress Tutorial.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]dogado.de.

Overview

This tutorial walks through aggregate DCEA step-by-step for a hypothetical NSCLC (lung cancer) treatment, following the Love-Koh et al. (2019) method.

Step 1: Define CEA inputs

icer            <- 28000   # £/QALY
inc_qaly        <- 0.45    # incremental QALYs per patient
inc_cost        <- 12600   # incremental cost per patient (£)
population_size <- 12000   # eligible patients in England
wtp             <- 20000   # NICE standard WTP (£/QALY)
occ_threshold   <- 13000   # opportunity cost threshold (£/QALY)

Step 2: Load baseline health distribution

baseline <- get_baseline_health("england", "imd_quintile")
baseline
#> # A tibble: 5 × 14
#>   imd_quintile group quintile_label      group_label     mean_hale mean_hale_all
#>          <int> <int> <chr>               <chr>               <dbl>         <dbl>
#> 1            1     1 Q1 (most deprived)  Q1 (most depri…      52.1          52.1
#> 2            2     2 Q2                  Q2                   56.3          56.3
#> 3            3     3 Q3                  Q3                   59.8          59.8
#> 4            4     4 Q4                  Q4                   63.2          63.2
#> 5            5     5 Q5 (least deprived) Q5 (least depr…      66.8          66.8
#> # ℹ 8 more variables: mean_hale_male <dbl>, mean_hale_female <dbl>,
#> #   se_hale <dbl>, se_hale_all <dbl>, pop_share <dbl>, cumulative_rank <dbl>,
#> #   year <int>, source <chr>

Step 3: Run aggregate DCEA

result <- run_aggregate_dcea(
  icer                       = icer,
  inc_qaly                   = inc_qaly,
  inc_cost                   = inc_cost,
  population_size            = population_size,
  disease_icd                = "C34",
  wtp                        = wtp,
  opportunity_cost_threshold = occ_threshold
)

Step 4: Interpret outputs

summary(result)
#> == Aggregate DCEA Result ==
#>   ICER:             £28,000 / QALY
#>   Incremental QALY: 0.4500
#>   Incremental cost: £12,600
#>   Population size:  12,000
#>   Net Health Benefit: -6230.77 QALYs
#>   SII change:         0.0649
#>   Decision:           Lose-Lose (efficiency loss + equity loss)
#> 
#> -- Per-group results --
#> # A tibble: 5 × 4
#>   group_label         baseline_hale post_hale    nhb
#>   <chr>                       <dbl>     <dbl>  <dbl>
#> 1 Q1 (most deprived)           52.1      52.0 -1558.
#> 2 Q2                           56.3      56.2 -1402.
#> 3 Q3                           59.8      59.7 -1246.
#> 4 Q4                           63.2      63.1 -1090.
#> 5 Q5 (least deprived)          66.8      66.7  -935.
#> 
#> -- Inequality impact --
#> # A tibble: 4 × 5
#>   index           pre     post    change pct_change
#>   <chr>         <dbl>    <dbl>     <dbl>      <dbl>
#> 1 sii        18.1     18.2     0.0649         0.358
#> 2 rii         0.304    0.306   0.00162        0.533
#> 3 gini        0.0487   0.0490  0.000259       0.533
#> 4 atkinson_1  0.00374  0.00379 0.0000402      1.07

Per-group results

result$by_group
#> # A tibble: 5 × 10
#>   group group_label   baseline_hale post_hale pop_share patient_share n_patients
#>   <int> <chr>                 <dbl>     <dbl>     <dbl>         <dbl>      <dbl>
#> 1     1 Q1 (most dep…          52.1      52.0       0.2         0.25        3000
#> 2     2 Q2                     56.3      56.2       0.2         0.225       2700
#> 3     3 Q3                     59.8      59.7       0.2         0.2         2400
#> 4     4 Q4                     63.2      63.1       0.2         0.175       2100
#> 5     5 Q5 (least de…          66.8      66.7       0.2         0.15        1800
#> # ℹ 3 more variables: health_gain_qaly <dbl>, opp_cost_qaly <dbl>, nhb <dbl>

Inequality impact

result$inequality_impact
#> # A tibble: 4 × 5
#>   index           pre     post    change pct_change
#>   <chr>         <dbl>    <dbl>     <dbl>      <dbl>
#> 1 sii        18.1     18.2     0.0649         0.358
#> 2 rii         0.304    0.306   0.00162        0.533
#> 3 gini        0.0487   0.0490  0.000259       0.533
#> 4 atkinson_1  0.00374  0.00379 0.0000402      1.07

Step 5: Visualise

plot_equity_impact_plane(result)

plot_ede_profile(result, eta_range = seq(0, 10, 0.2))

Step 6: Generate NICE submission table

generate_nice_table(result, format = "tibble")
#> # A tibble: 6 × 6
#>   `Equity subgroup`   `Baseline HALE (years)` `Post-intervention HALE (years)`
#>   <chr>                                 <dbl>                            <dbl>
#> 1 Q1 (most deprived)                     52.1                             52.0
#> 2 Q2                                     56.3                             56.2
#> 3 Q3                                     59.8                             59.7
#> 4 Q4                                     63.2                             63.1
#> 5 Q5 (least deprived)                    66.8                             66.7
#> 6 Total / Summary                        NA                               NA  
#> # ℹ 3 more variables: `Change in HALE (years)` <dbl>,
#> #   `Net Health Benefit (QALYs)` <dbl>, `Population share` <chr>

References

Love-Koh J et al. (2019). Value in Health 22(5): 518-526. https://doi.org/10.1016/j.jval.2018.10.007

These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.
Health stats visible at Monitor.