October 18, 2013

Cost-effectivness and meningococcal vaccine

There’s a story in the Herald arguing for mass vaccination in NZ against group C meningococcal disease.  There’s a good case for the vaccination — it’s a really nasty disease and there aren’t any other good treatment or prevention approaches — but I don’t see how the cost-effectiveness claims in the story can be right.

Dr Gravatt is quoted as saying there would be zero net medical cost for the vaccination program

Dr Gravatt estimated the net cost of vaccinating children at 12 months and again at 18 years would be zero for each “quality-adjusted” year of life saved at a vaccine price of $25 to $40 per dose. The listed prices were around $43 to $87, but discounts would be likely in a bulk contract.

He refers to a British vaccine study — which one isn’t specified.  However, there are cost-effectiveness analyses of the whole British group C meningococcal vaccination program (they vaccinate against group C but not group B, the opposite of NZ).  One, from 2002, estimates  a cost per life year saved of £6259. The second, from 2006, considers a wider range of possible vaccination programs: for the one that’s most similar to what Dr Gravatt proposed it estimates net medical costs of £9082 per quality-adjusted life year, and for the cheapest one, £3653 £2760 (update)per quality-adjusted life year. And that’s assuming the vaccine costs only £12 per dose, and in a country where group C meningococcal disease was slightly more common than it is here.

Another way to look at it: there are about 60000 births per year in NZ, so after the initial catch-up phase we’d be paying for 120000 doses per year. Even at $25 per dose, that comes to $3 million per year. According to the Herald’s story there are about 25 cases per year, so for the vaccination to be cost-neutral even at this lowest suggested price, the average disease case would have to cost the healthcare system $120000. That’s more than ten times what the UK cost-effectiveness study estimated (roughly £4000) including both the immediate cost of treating the disease and the cost of treating the long-term effects.

The vaccine looks to be a lot more cost-effective than, say, the prostate cancer drug that Campbell Live was pushing last week, and because herd immunity is important for this disease, there’s a relatively stronger case for vaccinating everyone, but that would be because the cost is worth it, not because the cost is zero.

Updates, including a lot more technical detail: 

[Update: Dr Gravatt’s letter with his calculations is up at the New Zealand Medical Journal,  unfortunately for subscribers/institutions only. The difference between the UK conclusions and his numbers is in a big difference in estimated total cost for long-term care of complications in survivors. I don’t yet understand why.]

Update again: Ok. Comparing Dr Gravatt’s calculations with the 2002 UK analysis I now understand the differences in long-term care costs. The minor difference is in the assumed proportion of people needing long-term institutional care: the UK paper assumes 10% of 7% of cases, ie, 0.7%, and Dr Gravatt is using 2.1%.

The second difference is more important: the UK paper is discounting costs and benefits in the future, at a rate of 3% per year and Dr Gravatt isn’t discounting. It’s pretty uncontroversial that monetary costs and savings should be discounted: even after inflation-adjustment, it’s better to have money now than later. That means the current value of a lifetime’s care for a disabled person is less than the total of the yearly payments. Since the costs of vaccination are incurred now and the cost savings are realised long into the future, the discount rate matters.  Using Dr Gravatt’s figures for rate of sequelae and discounting just the future monetary costs at 3% (the UK rate) gives a present value of costs from not vaccinating that’s about half the undiscounted costs, suggesting a cost-effective but not cost-neutral treatment. Pharmac’s discount rate is 3.5%, which would give a slight further reduction in the estimated cost savings.

The UK paper also discounted the benefits (life years) in the future.  Discounting non-monetary benefits (life years saved) is fairly standard but is controversial; fortunately I’m only addressing the monetary costs so I don’t need to take a position on discounting the QALYs.

The second UK paper gives a very much lower present value of costs for long-term sequelae.   That’s the paper I was initially relying on, and I don’t understand why it’s so different. The difference in long-term costs doesn’t have a huge impact on the conclusion, because the acute costs are high and the long-term costs are discounted, but it’s still confusing. I’ve emailed the author to ask.

 

avatar

Thomas Lumley (@tslumley) is Professor of Biostatistics at the University of Auckland. His research interests include semiparametric models, survey sampling, statistical computing, foundations of statistics, and whatever methodological problems his medical collaborators come up with. He also blogs at Biased and Inefficient See all posts by Thomas Lumley »

Comments

  • avatar
    Lance Gravatt

    Dear Professor Lumley,

    Each individual whose life is saved by the men C vaccination is assumed to go on and live a normal life span (84 years). Thus, the number of life-years saved per person saved is multiplied significantly since the average age of men C deaths is low. Secondly, the British study has calculated the cost per life-year saved for those that are not killed by meningococcal disease but suffer either “mild” or “serious” life-long disability at a cost to the overall health budget (deafness, multiple amputations and severe CNS disability). I would be happy to forward you my spreadsheet for your critique. The more expertise brought to bear to rid us of this scourge the better.

    10 years ago

    • avatar
      Thomas Lumley

      The young age issue is why the vaccine is even worth considering at that price, and goes in the denominator of cost-effectiveness — the number of years of life saved per case prevented is large because of the high case fatality rate and the age distribution, and the number of QUALYs saved is even higher because of long-term sequelae. But young age affects the net costs only through the long-term cost of treating the disabilities, and the UK cost-effectiveness studies (by Caroline Trotter) I referred to estimate the costs of sequelae as only about 10% of the acute costs. I was surprised their estimate was so low, and it will be sensitive to what care you assume for people with CNS disability, but even increasing the cost for sequelae by a factor of ten wouldn’t have been enough to make the UK calculations reverse.

      There’s no question that the cost, financial and personal, of each case is extremely large, but the disease is also relatively rare. I’d be interested to see your spreadsheet. As I said, I don’t see how the vaccination can be cost-neutral for the national health system, but I can easily see it being cost-effective.

      10 years ago

      • avatar
        Lance Gravatt

        The 2006 update analysis from UK offers 6 vaccinations strategies. The most cost effective one at 2,760 pounds per QALY being strategy 5 which we have chosen (vaccination at 12 months and 18 years catch up). The strategies that yield a cost per QALY of around 7,000 pounds vaccinate at 4 months and 18 year catch up which roughly doubles the number of doses needed.

        Second the 2006 analysis includes a 3.5 million advertising campaign by Government which we have not included as an incremental cost because annual vaccination campaigns are already in operation eg. Australia has a single vaccine that now combines men C plus Hib so there is no extra jab or extra visit.

        Finally, the 2006 analysis assumes that herd immunity is achieved.

        10 years ago

        • avatar
          Thomas Lumley

          You’re right about the cheapest strategy being #5 at GBP2760/qualy, slightly better than the 3600 I quoted. I’ve updated the post to show that.

          Herd immunity is assumed in the UK calculation, but that improves the cost-effectiveness — herd immunity seems to be particularly important for this vaccine. The paper says the indirect benefits were more than three times the direct benefits. That’s why the strategies assumed an advertising campaign, since getting high coverage is necessary to get the benefits.

          I’d willingly stipulate that a campaign wouldn’t be needed here, but that 3.5 million pounds was a small fraction of the costs involved in the UK, so it won’t change the calculations much. It still looks to me as if it’s cost-effective but not even close to cost-neutral.

          I’ve got the NZMJ site ready to read your letter when it turns up on line.

          10 years ago

        • avatar
          Lance Gravatt

          Thanks for these useful insights.

          I have also taken a scenario where I have applied the pro rata rate of confirmed men C cases over all confirmed strain cases to the cases reported in NZ as “untyped” and “probable”. This increases the annual incidence of men C cases from 0.60 to 1.05 per 100,000

          10 years ago

  • avatar
    Lance Gravatt

    The studies referred to in the UK and Netherlands’ analyses will be in the Letter to the Editor of NZMJ online this afternoon.

    10 years ago

  • avatar
    Lance Gravatt

    Dear Professor Lumley,

    I can easily explain where I have taken the incidence and quality of life scores for long-term disabilities. It is, as referenced, from the 2008 US Study by Ortega-Sanchez et al in Clinical Infectious Diseases. I used this reference for the long-term disability sequelae because unlike the Trotter papers it gave data that was broken down to an individual disability level instead of a simply classification of “mild” and “severe”.

    10 years ago

    • avatar
      Thomas Lumley

      Thanks. I wasn’t trying to cast doubt on those numbers, just noting that they were a minor but noticeable part of the difference.

      10 years ago

  • avatar
    Lance Gravatt

    Dear Professor Lumley,

    Discounting in cost-utility analyses is very much a Treasury approach that in layman terms is something of an artificial construct. However, I do accept that it is routinely used by PHARMAC. If in my Letter we use the higher incidence of meningococcal C scenario (which is most likely closer to the real incidence since it does not simply ignore those cases that were untyped), use your discount rate of 3% and take the lower costing of $25 then you will again achieve a near cost neutral result.

    10 years ago

  • avatar
    Lance Gravatt

    Finally I would like to say on the issue of discounting. There is no way the HPV vaccine would be cost effective given the large up front cost of vaccination in teenagers but the need to wait several decades before seeing the benefits of cervical cancers avoided. It was the emotion value that society placed on preventing cancer that got this vaccine over the line. I would argue that the emotional value of saving largely teenage lives is at least equal to that of avoiding cancer in late middle age.

    10 years ago

    • avatar
      Thomas Lumley

      I’m still not convinced the vaccine is cost-neutral, but I do think it’s sufficiently cost-effective to make sense as a public health intervention.

      If I hear anything enlightening from Dr Trotter I will update or post again, but we agree on a lot of things and I think we have got to the point where our remaining disagreements are at least well-sourced and understood, and could be followed up by anyone else who is interested.

      10 years ago

      • avatar
        Lance Gravatt

        Dear Professor Lumley,

        Thank you sincerely for the independent challenge and testing of my analyses. I agree with you that the majority of issues we now concur on and those where we differ and well understood and described.

        It is something of an esoteric debate as to whether vaccination is actually cost neutral or simply very cost effective when it comes to the central question of is it worthy of public funding or not.

        I believe that pricing can be achieved in NZ through a competitive process that would settle the issue in favour of neutrality in any case.

        Let’s keep in touch if there are any developments.

        10 years ago

  • avatar
    Chris Hughes

    The treatment of long term costs of very small numbers of patients bedevils qualy calculations. One study relied on by HM Treasury in considering policy options around compensation for birth injury claims resolved the problem n a straightforward way by ignoring any long-term costs

    10 years ago

  • avatar
    Lance Gravatt

    Been thinking about this issue of NPV or discounting of future savings in the vaccinated group that avoid life long disability such as multiple amputations or severe neurological damage. If we accept that this is a valid consideration in determining the cost effectiveness, do we not also then have to take into account inflation for the scenario of not vaccinating wherein a proportion of individuals have annual lifelong costs? The model as it stands assumes that a minor disability costs $1,000 per year for an average of 68 years.

    Kind Regards,

    Lance

    10 years ago

    • avatar
      Thomas Lumley

      Yes, all the costs (at least all the monetary ones) need to be discounted. That’s what the UK analysis did.

      The discounting will always make vaccination less favorable, since the vaccination costs are incurred before the benefits: often long before.

      For purely monetary costs incurred by the government, I think the case for discounting based on the Treasury’s real cost of funds is pretty strong. If the government spends a million dollars now and recoups a million dollars in (inflation-adjusted) savings over the next fifty years or so, that doesn’t even come close to balancing the books.

      10 years ago