Cost-effectiveness explorer

Move a new treatment around the cost-effectiveness plane, and see what happens when its carbon footprint is priced in.

A new treatment costs more (or less) than current care and produces more (or fewer) QALYs. Is it worth funding? Change the inputs below.

viewof dCost = Inputs.range([-20000, 60000], {value: 15000, step: 500, label: "Extra cost per patient (£)"})
viewof dQaly = Inputs.range([-1, 2], {value: 0.6, step: 0.05, label: "Extra QALYs per patient"})
viewof threshold = Inputs.range([0, 50000], {value: 20000, step: 1000, label: "Threshold (£ per QALY)"})
viewof dCO2 = Inputs.range([-2000, 2000], {value: 0, step: 50, label: "Extra emissions (kg CO₂e)"})
viewof carbonPrice = Inputs.range([0, 300], {value: 0, step: 10, label: "Carbon value (£ per tonne)"})

carbonCost = dCO2 / 1000 * carbonPrice
totalCost = dCost + carbonCost
icer = dQaly === 0 ? NaN : totalCost / dQaly
nmb = threshold * dQaly - totalCost   // net monetary benefit
verdict = nmb >= 0 ? "✅ Cost-effective at this threshold" : "❌ Not cost-effective at this threshold"

md`**ICER:** ${Number.isFinite(icer) ? "£" + Math.round(icer).toLocaleString("en-GB") + " per QALY" : "undefined (no QALY difference)"}
· **Net monetary benefit:** £${Math.round(nmb).toLocaleString("en-GB")}
${carbonPrice > 0 ? ` · **Carbon cost included:** £${Math.round(carbonCost).toLocaleString("en-GB")}` : ""}

### ${verdict}`

Plot.plot({
  width: 640, height: 420, grid: true, marginLeft: 60,
  x: {domain: [-1, 2], label: "Δ QALYs →"},
  y: {domain: [-25000, 65000], label: "↑ Δ Cost (£)"},
  marks: [
    Plot.ruleX([0]), Plot.ruleY([0]),
    Plot.line([[-1, -threshold], [2, 2 * threshold]], {stroke: "#eddcc6", strokeDasharray: "6 4"}),
    Plot.text([[1.6, 1.6 * threshold]], {text: ["threshold"], dy: -10, fill: "#eddcc6"}),
    carbonPrice > 0 ? Plot.dot([[dQaly, dCost]], {r: 6, stroke: "gray"}) : null,
    Plot.dot([[dQaly, totalCost]], {r: 8, fill: nmb >= 0 ? "#7fd1a0" : "#ff8a80"})
  ]
})

Things to try

Set extra QALYs to a negative value with negative cost. Which quadrant are you in, and what does the ICER mean there?
Keep the treatment just cost-effective, then raise the carbon value. When does the decision flip?
Why do economists often prefer the net monetary benefit to the ICER?