 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
PEDIATRICS Vol. 108 No. 2 August 2001, pp. 317-325
,
From * The Lewin Group, San Francisco, California; Objective. To investigate the
economic implications of a 2-dose hepatitis B virus vaccination regimen
compared with the current 3-dose vaccination regimen for adolescents in
3 settings: public schools, public health clinics, and private sector
settings in the United States.
Methods. To measure resource utilization and costs
associated with the administration of the 3-dose regimen and to assess
vaccination compliance rates with this regimen, primary data were
collected with the use of questionnaires tailored for each setting.
Conservative modeling assumptions were used to derive 2-dose compliance
rates from 3-dose compliance data. The results were incorporated into a
decision analytic model, which was used to examine short-term and
lifetime scenarios for an adolescent cohort receiving the 2-dose versus
the 3-dose regimen. In the short-term analysis, the vaccination program
costs were compared for the 2 regimens. In the lifetime analysis, the
model also incorporated long-term disease costs for those individuals
who contract hepatitis B.
Results. Predicted increases in compliance with a 2-dose
vaccination regimen contributed to a higher probability of
seroprotection in each setting. In the lifetime analysis, this positive
impact of improved compliance resulted in a lower infection rate and greater cost-effectiveness for the 2-dose regimen in all settings, including private sector settings, where it cost an average of only
$964 per year of life gained, and in public schools, costing an average
of $1246 per year of life gained. In public health clinics, the 2-dose
regimen had both lower expected lifetime costs and better clinical
outcomes than the 3-dose regimen. In the short-term analysis, costs
were higher for the 2-dose regimen, reflecting higher total vaccine
acquisition costs without the long-term offset of cost savings from
reduced infection. Sensitivity analyses identified cost per dose of
vaccine and the probability of completing the regimens as the most
sensitive model variables.
Conclusions. Improved compliance with a 2-dose regimen
would contribute to a higher probability of adolescents' achieving
seroprotection. When the long-term consequences of hepatitis B virus
infection are included, the 2-dose regimen would be cost-effective
compared with the 3-dose regimen in all settings and cost saving in
public health clinic settings.
Clinical
and Health Economic Statistics, Merck Research Laboratories, West
Point, Pennsylvania; § Louisiana State University, Health Science
Center, Baton Rouge, Louisiana; and Quintiles Canada Inc, Quebec,
Canada.
ABSTRACT
Top
Abstract
Methods
Results
Discussion
References
Estimates of the annual number of hepatitis B virus (HBV)
infections in the United States suggest that there are 200 000 to 300 000 infections per year,1 20 000 of which lead to chronic disease with approximately 4000 deaths each year as a result of
acute fulminant hepatitis or the sequelae associated with chronic
disease.1-3 Population-based seroprevalence studies
estimate that the race-adjusted average lifetime risk of HBV infection
in the United States is 4.8%.4 HBV infection is
especially problematic for infants and adolescents because the younger
the age at infection, the higher the likelihood of becoming a long-term
carrier of the disease. In addition, because the sequelae of chronic
disease can take decades to develop, infection at younger ages means
that these people are more likely to experience the long-term
consequences of chronic disease during their lifetimes.
Existing evidence in the literature indicates that the current 3-dose
HBV vaccine is a cost-effective method of preventing HBV infection in
infants, adolescents, and high-risk groups.4 Clinical
practice is consistent with the evidence in the literature; it is
widely accepted that immunization against HBV is appropriate and
desirable in these populations. Many studies have evaluated the
trade-off between the costs associated with instituting a vaccination
program and the resulting decrease in HBV prevalence compared with the
clinical and economic outcomes found in the absence of a vaccination
program.4-9 Margolis et al,4 using decision
analytic modeling, estimated that preventing perinatal HBV infection or
instituting routine infant or adolescent vaccination (based on a 3-dose
vaccination schedule) were cost saving from a societal perspective,
compared with no vaccination, when both direct costs and productivity
costs were included. These results are widely accepted, and many
leading organizations, including the World Health
Organization,10 the American College of
Physicians,11 the Centers for Disease Control and
Prevention, and the Advisory Committee on Immunization Practices, have
recommended adolescent vaccination, in addition to universal infant
immunization, as an effective means of controlling HBV
infection.3
A significant problem with HBV vaccination among adolescents is that
compliance to the existing 3-dose regimen is highly variable. Two
studies in US health clinics found that only 11% and 14% of adolescents who initiated HBV vaccination regimens completed the 3-dose
series,12,13 whereas other studies14,15 showed compliance rates of 71% and 72% in primary care and
school-based adolescent clinics. In some cases, >80% of adolescents
have completed the series in school-based programs.16
Development of alternative cost-effective HBV vaccination regimens that
require fewer doses may have a significant impact on overall
vaccination coverage and vaccination compliance rates among
adolescents.
A recent clinical trial17 demonstrated that a new 2-dose
HBV vaccination regimen (Recombivax HB; 10 µg/1.0 mL given at 0 and
4-6 months) was as effective as the standard 3-dose regimen (Recombivax HB, Merck & Co, Inc, Whitehouse, NJ; 5 µg/0.5 mL given at
0, 1, and 6 months) in providing HBV seroprotection (defined as
serologic levels of antibody to hepatitis B surface antigen [anti-HBs] Study Design
This study was designed to assess how a 2-dose vaccination
regimen would affect costs of HBV vaccination for adolescents in the
United States. Because the 2-dose regimen was not in use at the time of
the study, the necessary data about cost and compliance were obtained
by measuring these components for a 3-dose regimen and then adjusting
the values obtained to reflect likely changes associated with a 2-dose
regimen. Initial comparisons between regimens and overall results were
generated by a decision analytic model with the use of a set of
base-case values as input. Sensitivity analyses were conducted to
determine how changes in these values affected the results.
Short-term costs were defined to include only those associated with the
vaccination program itself. As comprehensive record reviews were too
labor intensive to be feasible, these costs were estimated with the use
of information obtained from a set of questionnaires, tailored to and
administered in 3 different vaccination settings: public school-based
programs, public health clinics, and private sector physician offices.
The same set of tailored questionnaires also was used to estimate the
degree of compliance with each dose of a 3-dose regimen among
adolescents targeted for vaccination. Separate short-term cost and
compliance estimates were generated for each setting.
A more complete lifetime economic perspective was obtained by adding to
short-term costs the long-term costs associated with HBV infection and
the savings that would result from its prevention. These outcomes were
estimated by constructing and using a lifetime model of HBV infection
and sequelae similar to one originally described by Margolis et
al.4 By incorporating both short-term vaccination costs
and long-term costs, the complete model allowed the calculation of
per-patient costs and cost-effectiveness ratios comparing the 2-dose
and 3-dose regimens from different perspectives.
Primary Data Collection
The questionnaire that was used to collect information about the
3-dose regimen was specifically designed for use in a telephone interview. The sites selected for the study reflected the diversity of
typical HBV immunization programs across the United States. The
selected sites and the estimated number of adolescents who were
eligible for vaccination at each site are shown in Table
1. The individuals interviewed were
actively involved in the organization and administration of HBV
vaccination at each site. All interviews were administered by telephone
by personnel from The Lewin Group.
TABLE 1
10 mIU/mL measured at month 7) in adolescents (11-19 years of age). In addition, the 2-dose regimen provided much better interim seroprotection between doses; approximately 73% of recipients achieved seroprotection with 1 dose of the 2-dose regimen, when measured at month 6, compared with only 44% after 1 dose of the 3-dose
regimen. In this study, we used the studied 2-dose regimen and a
decision analytic model incorporating all components of HBV infection
to assess the clinical and economic impact of this 2-dose regimen
versus the standard 3-dose regimen in adolescents.
METHODS
Top
Abstract
Methods
Results
Discussion
References
Sites Participating in the Assessment of Resource Use and Costs in
Administration of the 3-Dose Vaccination Regimen and the Number of
Eligible Adolescents per Site*
Although the specific questions varied slightly by type of setting, the questionnaires all included 3 main components. Specifically, each questionnaire evaluated resources and costs associated with the following: 1) preparatory work for each vaccine dose administration (including vaccine acquisition), 2) administration and disposal of vaccine doses, and 3) follow-up of patients who may have missed a scheduled dose.
The resources assessed included 1) advertising and promotional material used to encourage vaccination, 2) equipment used in program set-up, 3) vaccine and vaccine-related materials (eg, syringes, swabs), and 4) staff time required for administrative and clinical duties.
In addition to resources expended and other identified costs, the questionnaires asked respondents to estimate compliance with each dose of the 3-dose regimen. Target populations evaluated were adolescents 11 to 15 years of age with no previous HBV infection or vaccination and no personal reasons to refuse vaccination. This age group was selected because it is the only group for which the 2-dose regimen is indicated and because efficacy data were available. For partial vaccination series, separate estimates were provided for the percentage of adolescents who received no dose, 1 dose, or 2 doses of the 3-dose regimen. The mean estimated compliance rates for each health care setting are shown in Table 2 and were used as the base-case values for the model. Probable 2-dose compliance data were derived from the 3-dose data, as shown in Table 2, by assuming that compliance with the full 2-dose series would equal the total compliance associated with at least 2 doses of the 3-dose regimen.
|
Model of HBV Vaccination and Infection
A condensed version of the decision analytic model constructed for this study is shown in Fig 1. The model incorporates 3 components: 1) the proportion of adolescents who received a partial or complete vaccine series for each vaccine regimen; 2) the subsequent impact of the level of compliance on the proportion of adolescents who achieved seroprotection against HBV; and 3) the lifetime clinical and economic consequences associated with HBV infection (updated from the Margolis model), which are not shown in the figure.
|
Adolescents enter the model, as designated in Fig 1 by the square at
the left of the figure, and receive either the 2-dose or the 3-dose
regimen. For each regimen, adolescents have some chance of completing
the full series, receiving only a partial series, or receiving no dose,
as indicated by the percentages in Table 2. Depending on the number of
doses received, a certain proportion of adolescents will develop
vaccine-induced immunity (ie, anti-HBs 10 mIU/mL),18
whereas some may have no protective response. The degree of
seroprotection resulting from partial or complete 2- or 3-dose regimens
was obtained from the clinical trial previously discussed (Table
3),17 in which adolescents
received either a 2-dose or a 3-dose vaccination regimen. It was
assumed that an adolescent with vaccine-induced immunity would not
develop HBV infection, even if exposed to HBV, whereas an adolescent
with no protective response (ie, an anti-HB level <10 mIU/mL) was
assumed to have a 5% chance of lifetime HBV infection.
|
Data that defined the lifetime clinical and economic consequences of HBV infection were obtained directly from the Margolis et al4 study and are shown in Table 4. By attaching unit costs to each HBV infection, designated in Fig 1 by the triangles labeled HBV at the right of the figure, we were able to calculate lifetime costs associated with HBV infection. The methods used to estimate these costs are described in the next section.
|
Short-Term Cost Estimation
For the public sector, Centers for Disease Control and Prevention contract prices19 were used for vaccine costs and the national average Medicaid reimbursement of $8.84/dose for administration fees was used for administration costs. Private sector drug costs represent national average reimbursement rates as of March 2001 obtained from samples of n = 454 health plans (2-dose) and n = 958 health plans (3-dose). Private sector administration costs of $8.97 reflect the average per dose reimbursement for administration found in the subset of n = 911 health plans from samples reporting these costs.
Other costs of resources consumed by the vaccination program in each health care setting were estimated either by applying unit costs to each item or by using the total costs indicated on the questionnaires. Because the reported costs from each of the 3 distinct models of private physician offices investigated were consistent with each other, the costs from all private physician offices were pooled to provide costs for the private sector setting. Vaccination program costs used for the base-case values in the model are shown in Table 5.
|
Long-Term Cost Estimation
The costs associated with HBV infection and the subsequent sequelae were obtained from the Margolis et al study,4 using the medical component of the consumer price index to adjust the costs to 2001 dollars (Table 4).20 As originally calculated by Margolis et al,4 these costs included the loss of earnings as a result of illness, as well as work time lost in medical care visits and the consequences of premature mortality measured in terms of lost productivity. These productivity losses were built into the cost estimates using the methods described by Margolis et al4 and assumed that individuals were employed productively between the ages of 20 and 65. Average lifetime earnings during this period were estimated from sources21 that were current at the time of the Margolis study and were adjusted for unemployment and a 1% annual increase in productivity.4
Economic Analyses
Two types of economic analyses were undertaken: 1) analysis of short-term costs (ie, those related to the vaccination program alone), and 2) analyses of total lifetime HBV-related costs, involving both short-term costs and long-term costs associated with HBV infection and sequelae. For each health care setting, per-patient short-term costs were expressed as the cost per person in the target population, cost per person achieving seroprotection, and the cost per HBV infection prevented. The short-term cost per person achieving seroprotection represented the vaccination program costs invested to yield 1 adolescent with vaccine-induced immunity. The short-term cost per HBV infection prevented represented the vaccination program costs invested to prevent 1 HBV infection.
Per-patient costs also were expressed in terms of total lifetime HBV-related costs. In these lifetime analyses, the per-patient costs for each health care setting and each vaccination regimen included the per-patient vaccination program costs, as well as the long-term direct medical costs and productivity losses associated with HBV infection and related sequelae. For these lifetime analyses, future costs and life years were discounted at 5% per year. This discount rate is consistent with rates traditionally used for cost-effectiveness analyses and is within the range of currently recommended values.22
Incremental cost-effectiveness ratios also were determined for short-term and total lifetime HBV-related outcomes, expressed in terms of the 2-dose vaccination regimen versus the 3-dose. For example, the incremental cost-effectiveness ratio for the cost per additional HBV infection prevented with the 2-dose regimen was the difference in costs between the 2 regimens divided by the difference in the number of HBV infections prevented. Incremental cost-effectiveness ratios also were expressed in terms of life years gained, using total lifetime HBV-related cost data. The incremental ratio was expressed as the difference in lifetime HBV-related costs between the 2 regimens divided by the difference in the number of years of life associated with each regimen.
Sensitivity Analyses
Sensitivity analyses were conducted to test the robustness of the model to changes in the values of all input variables. Both 1-way, in which the value assigned to a single variable was changed, and multivariate analyses were performed.
For the sensitivity analyses, most variables were varied from 50% (lower limit) to 150% (upper limit) of the base-case values, and outcomes were determined over this range. However, a different approach was used for the probability of completing all 3 doses of the 3-dose vaccination regimen. In these analyses, the lower limit was set at 50% of the base-case value and the upper limit was set at the base-case probability of completing all 3 doses of the 3-dose regimen. The reasoning behind this was based on the premise that because compliance data from the questionnaire might already overstate compliance rates (as suggested by the rate observed in a multichart audit),12 there was no reason for an upper range value of 150% of the base case for compliance. Such a value likely would be unrealistic and overestimate compliance values observed in real clinical practice.
To account for the possibility that patients might be equally compliant with both regimens, we conducted a sensitivity analysis setting the compliance rates for the full series equal for the 2 regimens. We also investigated how the regimens would compare if compliance with the complete 2-dose regimen were considerably higher than with the complete 3-dose regimen. Examining these 2 situations required the elimination of the base-case assumption that equated compliance with the full series of the 2-dose regimen to compliance with at least 2 doses of the 3-dose regimen. Instead, for the first situation, compliance with the complete 3-dose regimen was set equal to the base-case compliance with the complete 2-dose regimen. For the second situation, compliance with the complete 3-dose regimen was fixed at 50% of the base-case value, which is approximately the midpoint of the range of 3-dose compliance reported in the literature.12-16 Compliance with the complete 2-dose regimen was varied from that value up to its base-case value.
Multivariate sensitivity analyses were performed with the use of Monte Carlo simulations.23 The aim of these analyses was to assess the distribution of model results when the variables identified as particularly sensitive in 1-way sensitivity analyses were varied simultaneously while keeping the value of all other variables constant.
To assess short-term costs, the model was run allowing the compliance for the 3-dose vaccine and the cost of vaccine administration to vary over the chosen ranges. For the lifetime version of the model, in addition to these aforementioned variables, the probability of HBV infection given no protective response and the cost of HBV infection also were allowed to vary in the simulation. Although identified as sensitive in the 1-way analyses, the per-dose cost of the 2-dose regimen was not varied in these simulations because this cost is not one that would be subject, in reality, to random variation.
| |
RESULTS |
|---|
|
Top
Abstract
Methods
Results
Discussion
References
|
|---|
Base-Case Results
The base-case probabilities of both achieving immunity and preventing infection were higher for the 2-dose regimen. However, these improved clinical outcomes were associated with an increased short-term cost (Table 6). This was primarily attributable to higher total vaccine acquisition costs associated with improved compliance with the 2-dose regimen. However, when long-term costs associated with HBV infection were included in the model to provide a lifetime societal perspective, the 2-dose regimen was cost-effective when compared with the 3-dose regimen in all settings. For the public health clinic settings, the 2-dose regimen is a dominant strategy, meaning that there were both better clinical outcomes (ie, higher probability of seroprotection) and lower expected costs compared with the 3-dose regimen. The expected cost per person with immunity and the cost per infection prevented were lower with the 2-dose regimen (Table 6). The 2-dose regimen also was dominant for public health clinics in terms of cost per year of life gained.
|
In the base-case analyses for the private sector third-party payer and the school-based clinics, a higher probability of immunity was achieved with the 2-dose vaccination regimen, which was associated with a higher lifetime cost than the 3-dose regimen. This was reflected in the incremental cost-effectiveness ratios of the 2-dose regimen versus the 3-dose, which for the private sector were: $1517 per infection prevented and $964 per year of life gained. For the public school setting, the incremental cost-effectiveness ratios were $1960 per infection prevented and $1246 per year of life gained.
Sensitivity Analyses
One-way sensitivity analyses revealed that the most sensitive variables that affected cost outcomes were the cost per dose of vaccine for the 2-dose regimen and the probability of completing the vaccine series. The sensitivity to these variables was apparent in all settings for both the short-term and lifetime cost scenarios.
When the per-dose cost of the vaccine itself was varied for the 2-dose regimen, while remaining constant for the 3-dose regimen, there was a large change in the relative cost per adolescent. Figure 2 shows the difference between the 2 regimens in total expected lifetime costs per adolescent as the per-dose vaccine cost changes for the 2-dose regimen. The results for each setting are shown by a separate line. In each setting, if the per-dose cost is low enough, the difference in expected cost per adolescent is negative (ie, the cost per adolescent is less for the 2-dose regimen than for the 3-dose regimen). The sensitivity analysis shows that the difference in expected costs increases to a positive value (ie, the 2-dose regimen becomes more expensive) as the per-dose cost increases.
|
For each setting, there is a threshold value, defined as the per-dose cost at which the expected lifetime costs for the 2 regimens are equal. The threshold values correspond to the points in Fig 2 where the 3 sloping lines cross the horizontal $0-difference line. The threshold value is $21.60 for school-based programs, $24.90 for public health clinics, and $52.00 for third-party payers. Thus, the cost per dose for the 2-dose regimen can be higher for third-party payers than in the other 2 settings and still result in lifetime costs per adolescent that are lower than the 3-dose regimen for third-party payers.
The sensitivity analyses show how the lifetime costs per adolescent change when per-dose vaccine costs vary around their base-case values. These values are shown in Fig 2 at the base of the vertical dotted lines. For example, the base-case cost per dose of vaccine for the 2-dose regimen among private payers is $54.65, for which the 2-dose regimen would be almost $5 more expensive than the 3-dose regimen (in terms of lifetime costs per adolescent). In the school-based setting, the 2-dose regimen would be approximately $2 more expensive than the 3-dose regimen, assuming a base-case cost per dose of $23.75 for the 2-dose vaccine. For the public health clinic setting, the base-case value ($23.75) is slightly below the threshold value, meaning that the 2-dose regimen is cost saving at the base-case value.
To gain insight into the sensitivity of the results to changes in compliance with the complete 2-dose regimen when compliance with the complete 3-dose regimen was lowered, we fixed 3-dose compliance at 50% of base-case values and varied the rate of 2-dose regimen compliance. The sensitivity of the total expected lifetime HBV-related costs to changes in 2-dose compliance under these conditions is depicted by setting in Fig 3. We calculated threshold 2-dose regimen compliance values, which reflect the level of 2-dose regimen compliance at which the 2-dose regimen and 3-dose regimen are equivalent in cost. The threshold 2-dose regimen compliance rates are 62%, 51%, and 65% for private sector, school-based, and public health settings, respectively.
|
Higher probabilities of completing the 2-dose regimen lead to lower expected lifetime costs for the 2-dose regimen compared with the 3-dose regimen in all settings. This result is observed because greater compliance leads to a larger proportion of adolescents with seroprotection and, hence, fewer HBV infections, long-term sequelae, and long-term costs. The higher the probability of completing the 2-dose regimen, the lower its expected lifetime costs are when compared with a 3-dose regimen for third-party payers.
A 1-way sensitivity analysis setting compliance with the complete series, equal for both regimens, was conducted. Results are virtually the same as for the base-case values reported in Table 6. For example, the lifetime costs per person for the 3-dose regiment in public schools would be $89.85, compared with the base-case value of $90.04.
Multivariate sensitivity analyses with the use of Monte Carlo simulations of 1000 iterations support the robustness of the results. These analyses showed short-term costs for the 2-dose regimen greater than for the 3-dose regimen in 100% of iterations. However, when long-term costs also were considered, the 2-dose regimen was less expensive than the 3-dose regimen in both the public school and the clinic settings in 86.3% and 88.8%, respectively, of 1000 iterations of the lifetime model. The 2-dose regimen was always more effective in these settings. These results are consistent with the estimates from the base-case analysis showing the dominance of the 2-dose regimen in the public clinic setting. The base-case analysis for the public school setting, however, showed a slight increase in the lifetime cost per adolescent associated with the 2-dose regimen. In 95% of iterations, this increase resulted in a cost-effectiveness ratio of less than $320 per year of life gained (compared with the base case of $1246). The maximum value of the cost-effectiveness ratio was $3804 per year of life gained. For private third-party payers, the 2-dose regimen always remained more effective, but it was more expensive in 40.8% of the observed simulations. Again, the additional cost was low, with 95% of the 1000 iterations resulting in a cost-effectiveness ratio of less than $1575 per year of life gained (compared with the base case of $964). The maximum value of the ratio was $5940 per life year gained, supporting the base-case finding of cost-effectiveness.
| |
DISCUSSION |
|---|
|
Top
Abstract
Methods
Results
Discussion
References
|
|---|
The results of this study are comparable to previously published HBV-related costs for the 3-dose regimens. In 2 previous studies,9,24 the cost per adolescent successfully vaccinated ranged from $44 (in 1994 US dollars, adjusted to $57.84 in 2001 US dollars using the Medical Care Consumer Price Index) to $83.21 in a school setting. This range is consistent with the cost determined in the present study from questionnaire responses for the public school-based program ($77.00 per person achieving immunity). The lifetime HBV model predicted costs per HBV infection prevented of between approximately $1900 and $3100, depending on setting (based on the 3-dose vaccination regimen). Approximate calculations from the Margolis et al study4 showed a cost per HBV case avoided by adolescent vaccination of $3352 under the same discounting conditions as the present study. The Margolis et al analysis used a vaccination program cost of $85, higher than the short-term program costs used in the present study, which may account for a significant portion of the difference in the costs per infection prevented between the 2 studies. Similarly, Bloom et al25 estimated that in an adolescent cohort that received a 3-dose vaccine schedule, the cost per case prevented was $7241 with both costs and outcomes discounted at 5% per year. However, the vaccination cost used in that study was $225, several times greater than the cost used in the present study. Furthermore, Bloom's analysis incorporated a booster vaccine shot after 10 years, which would increase overall direct medical costs. In the present study, no booster shots were incorporated in the analysis, because vaccine efficacy was assumed to be constant for the lifetime of a vaccinated adolescent. Overall, the results of the present study primarily are consistent with costs reported in the literature, and the differences are explainable given the methodological differences between the studies.4,25
As in any study incorporating decision analytic modeling, there may be limitations. Vaccine administration costs were determined by surveying selected sites in each health care setting. These costs may reflect some regional variation. However, the aim of the present study was to compare the different vaccination regimens; variation in absolute costs between sites does not affect the comparison of the 2-dose with the 3-dose vaccine within each setting.
In keeping with the "prospective" component of the study, compliance rates for the 3-dose vaccination regimens were based on questionnaire data. These values were used to derive compliance rates for the 2-dose regimen and are a major component in the subsequent development of the HBV model. Our reliance on compliance data from this single source for each health care setting may be viewed as a potential limitation, as compliance rates for HBV vaccination do show considerable variation in the literature.4,9,25 There also is evidence that physicians often overestimate vaccine compliance rates,12 which would affect the private sector analysis. This suggests the possibility that the compliance values elicited from the questionnaires for the 3-dose vaccine may overestimate real-world compliance, in which case the compliance used for the 2-dose regimen also would be overestimated and may be more in the range of published studies. It also is possible, however, that compliance for the 2-dose vaccine actually would be higher in the real world than the estimate used in this analysis. However, comprehensive sensitivity analyses of compliance showed little substantive impact on the economic outcomes. In general, the 2-dose regimen was always either a cost-saving or a cost-effective strategy in the long-term compared with the 3-dose regimen.
Our results suggest that the introduction of a 2-dose vaccination regimen could have a significant impact on the clinical and economic outcomes of HBV vaccination. Although we found the 2-dose regimen to be marginally more expensive than the 3-dose regimen in the private sector and the public school settings, it was still a cost-effective strategy in those settings. The median cost per year of life saved for medical interventions in the literature has been shown to be $19 000 per life-year.26 The incremental cost per year of life gained for the 2-dose regimen compared with the 3-dose regimen was only $964 in the private sector setting and $1246 in the public school setting. These costs per life year gained are extremely low. Furthermore, analyses performed under potentially more realistic 3-dose compliance conditions indicate that the 2-dose regimen would be equivalent in expected lifetime costs at a threshold compliance value of only 62% and 53% for the private sector and school-based settings, respectively.
For the public health clinic setting, we found that the 2-dose vaccine was the dominant strategy in terms of cost-effectiveness, regardless of the effectiveness measure used. This means that over the lifetime of an adolescent, a 2-dose HBV vaccine was found to have better clinical outcomes and be cost saving compared with a 3-dose vaccine. Because adolescent vaccination with a 3-dose regimen has already been shown to have a favorable cost-effectiveness ratio,4,24 the present study suggests that the 2-dose vaccine would be even more cost-effective in the public health clinic setting.
| |
ACKNOWLEDGMENT |
|---|
This study was made possible by a grant from Merck.
| |
FOOTNOTES |
|---|
Received for publication May 9, 2000; accepted Nov 27, 2000.
Reprint requests to (H.P.L.) The Lewin Group, 490 Second St, Suite 201, San Francisco, CA 94107. E-mail: hugh.levaux{at}lewin.com
| |
ABBREVIATIONS |
|---|
HBV, hepatitis B virus; HBs, hepatitis B surface antigen.
| |
REFERENCES |
|---|
|
Top
Abstract
Methods
Results
Discussion
References
|
|---|
California, Louisiana, and
Oregon, 1992-1994. MMWR Morb Mortal Wkly Rep.
1994:43:605-609all consumers:
medical care component. Available at:
http://146.142.4.24/cgi-bin/srgate (CPI multiplier: 1.244). Accessed
March 2001This article has been cited by other articles:
 |
|
 |
|
  L. Tilson, L. Thornton, D. O'Flanagan, H. Johnson, and M. Barry Cost effectiveness of hepatitis B vaccination strategies in Ireland: an economic evaluation Eur J Public Health, June 1, 2008; 18(3): 275 - 282. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2001 American Academy of Pediatrics. All rights reserved. |
||
|
||
