Health technology assessment
Cost-effectiveness of HPV-vaccination of boys aged 12 in a Norwegian setting
Health technology assessment
|Updated
This economic evaluation examines the cost-effectiveness of vaccinating both 12-year-old boys and girls against HPV-infection compared to maintaining the current practice of vaccinating only girls.
Key message
Infection with high-risk human papilloma virus (HPV) types 16 and/or 18 is documented to be related to cervical, vulvar, vaginal, penile, anal and oropharyngeal cancer, while infection with low-risk HPV types 6 and/or 11 is documented to be related with genital warts. This economic evaluation examines the cost-effectiveness of vaccinating both 12-year-old boys and girls against HPV-infection compared to maintaining the current practice of vaccinating only 12-year-old girls.
Two vaccines (both delivered in a 3-dose schedule) are available on the Norwegian market with documented effect against HPV-infection: the quadrivalent vaccine is directed at HPV 6, 11, 16 and 18 and the bivalent vaccine is directed at HPV 16 and 18. In this report, we estimated the costeffectiveness of the quadrivalent vaccine for the target population, as this is the vaccine currently offered in the vaccine program for girls. The costeffectiveness of the bivalent vaccine is discussed in a scenario analysis.
• The main finding of the evaluation is the following: From a societal perspective, vaccinating boys in addition to girls aged 12 with the quadrivalent vaccine is probably not cost-effective. The incremental cost-effectiveness ratio (ICER) was NOK 1,626,261 for a quality-adjusted life-year (QALY).
Although there is no official cost-effectiveness threshold value in Norway, such high ICERs are generally associated with the intervention not being accepted for implementation in the Norwegian health sector.
Summary
Background
Infection with human papilloma virus (HPV) is documented to be related to cervical, vulvar, vaginal, penile, anal and oropharyngeal cancer and genital warts. The Norwegian Institute of Public Health is responsible for managing the publicly funded vaccination program against the human papillomavirus (HPV). Currently, this program covers the expenses of vaccinating 12-year-old girls only with the quadrivalent vaccine (directed at HPV 6, 11, 16 and 18), delivered in a 3-dose schedule.
Gender-neutral vaccination has been recommended in a few countries, e.g. United States (1) and Australia (2). In the European Union, only Austria (3) has recommended it. The UK government decided in 2014 to reconsider whether HPV vaccination should be offered to men who have sex with men and/or adolescent males (4).
In this economic evaluation, we evaluate the cost-effectiveness of administering vaccination against HPV-infection to 12-year-old boys in addition to the current practice of vaccinating 12-year-old-girls, compared to maintaining the current practice.
Objective
To evaluate the epidemiological impact, costs, health benefits and cost-effectiveness of administering the quadrivalent HPV-vaccine to 12-year-old-boys in addition to the current practice of vaccinating 12-year-old-girls, compared to maintaining the current practice. We evaluate in addition how alternative scenarios would influence results and conclusions.
Method
We used an already published American economic model consisting of a deterministic, dynamic population-based model that estimated the proportion of people in every contemporary and future cohort having experienced infection with HPV 6/11, 16 and/or 18, from 2016-2115.
The model was adapted to the Norwegian setting with respect to incidence rates of HPV-related outcomes, costs and health related quality-of-life (HRQoL). In addition, the model was modified in order to incorporate the findings on clinical endpoints reported in our systematic reviews. Finally, the model was made probabilistic.
The focus in our base case analysis was on evaluating the consequences of HPV-vaccination of boys (using the quadrivalent vaccine in a 3-dose schedule) may have for 184 birth cohorts during the first 100 years: The 91 cohorts aged 8-99 at the moment of starting the vaccination of boys and the coming 93 cohorts of 8-year-old boys and girls.
We assumed that 82% of all boys and girls aged 12 would get on average 2.54 and 2.78 doses of the HPV-vaccine, respectively. Furthermore, we assumed that the vaccine would only have effect on genital warts, cervical, vulvar, vaginal and anal precancerous lesions and cancer as documented in our own systematic reviews (vaginal and anal outcomes were included only in sensitivity analyses). Finally, the price of the vaccine was set equal to the maximum pharmacy retail price (PRP) of the quadrivalent vaccine in December 2014, NOK 1,113.40/dose.
The economic evaluation was performed from two different costs perspectives: a health-care perspective focusing on costs to the National health system; and a societal perspective in which we excluded the value-added tax (VAT) from prices and included the deadweight loss of taxation, the costs to patients for time used under treatment and the work-related productivity costs due to disease.
For each perspective, an Incremental Cost Effectiveness Ratio (ICER) in terms of NOK per quality-adjusted life year (QALY) gained was calculated. To determine whether vaccinating boys aged 12 was cost-effective, the resulting ICER was compared to a range of potential willingness-to-pay (WTP) values between NOK
250 000 - 2 000 000/ gained QALY. ICERs lower than the chosen WTP value typically supports the hypothesis that vaccinating boys is cost-effective and therefore yields good value for money, while ICERs above the chosen WTP value suggest the opposite.
In addition, we assessed the uncertainty around the results and estimated the Expected Value of Perfect Information (EVPI).
We ascertained as well the impact on the base case results by changing the value of potentially important inputs and modelling assumptions. We did this by conducting the following sensitivity analyses (from a health care perspective):
- Lower vaccine purchase prices per dose.
- The exclusion of the effect of the vaccine on genital warts in order to examine the cost-effectiveness of using a bivalent vaccine.
- Increasing the coverage among girls instead of vaccinating boys.
- The exclusion of the HRQoL-impact of vaccination.
- The inclusion of the vaccine effect on anal cancer (in both genders) and vaginal cancer and vaginal intraepithelial neoplasia grade 2 and 3 (VaIN 2+).
- The exclusion of the effect of the vaccine in reducing the number of conization-related preterm births.
- A 2-doses schedule providing the full vaccination effect reported in published studies.
- The reduction in the incidence of several relevant outcomes under different vaccine coverage assumptions.
In addition, we conducted a series of one-way sensitivity analysis where we examined how the base case results changed when increasing or reducing certain groups of variables (epidemiologic, effect, costs, HRQoL) by 25%. We present the results in the form of a Tornado diagram.
Results
From a health-care perspective, the base case results showed that including boys in the current vaccination program would lead to a discounted, incremental cost of NOK 1.851 billion (1 billion = 1 thousand million) and a discounted, incremental health gain of 1,034.59 QALYs over the 100 years horizon of the model (2016-2115). This resulted in an ICER of NOK 1,789,463/QALY. The scatter-plot of the ICER showed that both the incremental costs and the health gain were positive for all iterations.
The ICER of vaccinating boys is lower from a societal perspective than from a public health perspective. The incremental costs were approximately 9% lower, NOK 1.677 billion vs. NOK 1.851 billion in the public health-care perspective, while the incremental effect was the same, leading to an ICER of NOK 1,626,261 /QALY.
The results of the scenario analyses were the following:
- Using prices of NOK 250, 500 and 750/dose resulted in lower incremental costs and therefore lower ICERs of NOK 351,975/QALY, NOK 765,909/QALY and NOK 1,186,606/QALY, respectively.
- 2.Excluding the vaccine effect on genital warts resulted in both somewhat higher incremental costs and a considerably lower incremental health effect than in the base case. The resulting ICER was NOK 3,754,854/QALY. Assuming these results apply to the bivalent vaccine, and that the price of the quadrivalent vaccine is equal to the public price of NOK 1,113.40/dose, we estimated that the price of the bivalent vaccine had to be approximately 550 NOK/dose or lower in order to be as cost-effective as the quadrivalent vaccine.
- 3.Increasing the first dose coverage among girls aged 12 from 82% to 92% (for at least one dose) instead of vaccinating boys of same age resulted in a 42% lower incremental effect and an 89% lower incremental costs than vaccinating boys. The ICER was NOK 336,755/QALY, considerably below the ICER of the vaccination program for boys.
- Ignoring all HRQoL-gains (reduction of morbidity) and focusing exclusively on lifetime gains (reduction in mortality) in terms of life-years gained (LYG) resulted in a 71% decrease in the nominal incremental effect of the program, compared to the base case (where reductions in both morbidity and mortality are assessed). This is partly due to the large HRQoL-gain associated with preventing genital warts, an important outcome for males in our model. The lower incremental effect leads to a considerably higher ICER, NOK 6,188,344 /LYG.
- 5.When incorporating the potential effect of the vaccine in reducing the number of cases of VaIN 2+ and vaginal and anal cancer, the ICER decreased to NOK 1,538,578/QALY.
- When excluding the vaccine effect in reducing the number of conization-related preterm births, the ICER increased to NOK 1,848,515/QALY.
- Assuming all children get two vaccination doses led to an ICER of NOK 1,389,853 million/QALY, approximately 22% lower than in the base case.
- Depending on the outcome, including boys in the vaccination program led to incidence reductions after 100 years that were 1-9 percentage points higher than increasing coverage among girls and 3-13 percentage points higher than vaccinating only girls at the current coverage rate. The greatest incidence reduction accomplished by vaccinating boys was registered for genital warts among males.
Finally, our one-way sensitivity analyses showed that changes in the incidence of HPV-related outcomes and the HRQoL-losses associated with these outcomes had a considerable impact on the cost-effectiveness results, although less than changes in the vaccine effect estimates. Changes in the HPV-acquisition rates and treatment costs had very limited impact on the results.
Conclusion
From both perspectives and given the current public price of NOK 1,113.4/dose of the quadrivalent vaccine, vaccinating boys in addition to girls aged 12 years is probably not cost-effective. The incremental ICER was NOK 1,789,463/QALY from a health care perspective and NOK 1,626,261/QALY from a societal perspective.
Although there is no official cost-effectiveness threshold value in Norway, such high ICERs are generally associated with the intervention not being accepted for implementation in the Norwegian health sector.
The price scenario analysis shows that a lower price per vaccine dose has a major, positive effect on the cost-effectiveness of the program.
Increasing coverage among girls aged 12 from 82% to 92% seems to be more cost-effective than vaccinating boys and girls at a coverage rate of 82%.
The price of the bivalent vaccine should not be higher than approximately NOK 550/dose for it to be deemed as cost-effective as the quadrivalent vaccine.