Balestra+JGIM+1993

Balestra+JGIM+1993 - ORlGINAL ARTICLES Should Adult Tetanus...

Info iconThis preview shows pages 1–8. Sign up to view the full content.

View Full Document Right Arrow Icon
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 2
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 4
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 6
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 8
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: ORlGINAL ARTICLES Should Adult Tetanus Immunization Be Given as a Single Vaccination at Age 65? A Cost— Effectiveness Analysis DOMIN/C J. BALESTRA, MD, BENJAMIN LITTENBERG, MD Objecn've: To compare three vaccination strategies for the prevention ofadult tetanus. Each strategy includes childhood primary immuni- zation and wound prophylaxis, and one of the following: 1) the currently recommended booster every ten years; 2) a single booster at 65 years of age; or 3) no intervention after age 6 except for wound prophylaxis. illetbods: Cost—effectiveness analysis was used to compare the three different strategies. A Markov model, cycled annually from age 5 through age 85, was applied to each strategy to predict the incidence and costs of tetanus for the U.S. adult population. Results: The three strategies have very similar eifects on life expect, ancy but different costs. Expressed incremental to no intervention after childhood primary immunization, the decennial booster strat- egy is least COStACffCCKiVC. with a discounted incremental cost— ell‘ectiveness ratio of $14 3,138 per year of life saved compared with $4,527 for the single-booster strategy. Sensitivity analysis demon‘ suates that the decennial Strategy is more efi‘ective but more costly over a wide range of model assumptions. Conclusions: The current policy of recommending tetanus booster vaccinations every ten years is effective but much more cosrly than a more easily implemented policy that also provides considerable pro- tection against tetanus. The authors recommend forsaking decennial boostch in favor of a policy of including a single booster at age 65 along with other recommended health maintenance maneuvers re- served for that age. Key words: tetanus; preventive medicine; cost—benefit analysis; im- munization policy; vaccines. J GEN INTERN MED 1993.8:405—412. THE US. PREVENTIVE SERVICES TASK FORCE, the Centers for Disease Control and Prevention (CDC), and other policymakers recommend that all adults, irrespective of their primary tetanus immunization status, receive tetanus booster immunization every ten years.“5 This policy supplements additional recommendations for primary tetanus immunization of children as well as recommendations for tetanus wound prophylaxis.6 Tetanus boosrer immunization is advised because the vaccine is considered inexpensive and highly elli- Reccived from the Department of Medicine, Veterans Affairs Medical and Regional Office Center (DJB). White River junction, Vermont 05009; and the Technology Assessment Program, Department of Medicine, Dartmouth—Hitchcock Medical Center (8L), Lebanon, New Hampshire. Presented in part at the annual meeting ofthc Society ot‘General internal Medicine, Seattle, Washington, May 2. 1991. Dr. Littenberg is an American College ofl’hysicians George Morris Piersol Teaching and Research Scholar. Address correspondence and reprint requests to Dr. llalestra. cacious. The cost—efiectiveness of vaccinating adults against tetanus in the United States has not been pre— viously assessed, nor has the current recommendation been compared with alternative strategies. (Although one study" addressed the cost—effectiveness of using mailed reminders from family physicians to increase primary tetanus vaccination coverage among elderly Canadians, we are nor aware of an analysis of vaccina' tion itself.) We constructed a Markov model to compare the cost- effectiveness of currently recommended rou- tine decennial tetanus booster vaccinations with those of several Other strategies for adults. Although tetanus vaccine is generally combined with diphtheria vaccine for adults, our analysis restricts itself to the quesrion of tetanus boomers. Diphtheria is nor assessed because it is rare, the eficacy of the vac- cine is controversial, and the disease seems locally epi- demic rather than endemith13 METHODS The three strategies analyzed are: the decennial- booster strategy (T), a single tetanus booster immuni- zation at 65 years of age (B), and no intervention after age 6 except for wound prophylaxis (N). The model assumes that the subject population has received pri— mary immunization for tetanus with the last booster administered before age 6. The model also assumes that the Strategies will be applied to the US. population prospectively and predicts the incidence and costs of tetanus for 6-year—olds over an 80-year period (to age 85). Other assumptions inherent in the model are that the published life expectancy of “all races” applies to the target population and that current all«cause mortal- ity rates for each age in the population apply into the future.” in general, where assumptions must be made, our model uses values that favor the current decennial strategy. Immune Status Though an antitoxin titer of 0.0} lU/ml. is gener~ ally accepted as defining protective immunity, some evidence suggests this standard is conservative. About 10% of American soldiers and a higher percentage of 405 406 Balesrra. Littenberg. CosnEFFtCtsVENt-Es or TETANUS VACCiNA'i‘ION Canadian soldiers in World War Ii had titers below 0.01 IU/mL one year after primary immunization. The ex- tremely low incidence of tetanus in these two groups suggests that protection was very high.15 The porency of toxoids has since markedly increased. Between World War ii and 1959, only seven cases of tetanus were reported among all US. military personnel who received basic immunization and an emergency boomer at the time of injury. Two researchers with serum antitoxin levels between 0.005 and 0.01 lU/mL suffered no symptom or sequela from self—challenge with a dose of tetanus toxin threefold the minimal lethal dose for humans.15 Conversely, only rarely has cl inical tetanus occurred in adults when antitoxin titers exceeded 0.01 iU/ml..“”‘18 Hence, many persons with antitoxin titers below the accepted standard of 0.01 IU/mL may be protected against tetanus. From 16 to 35% of persons claiming no previous tetanus immunization and up to 69% of persons report- ing a single tetanus immunization have serum antitoxin titers > 0.01 lU/mli.”22 Thus, the use of medical his- tory alone to detect candidates for immunization may unnecessarily commit a large proportion of the popula— tion to vaccination. Our model assumes waning immunity from age 6. Because the relationship between measured antitoxin titer and eifective immunity is somewhat unclear, we elected to model eifecrive immunity rather than titer. Vaccine-induced immunity Although individuals vary greatly in responsive- ness to tetanus toxoid, the decrease in antitoxin over time is also variable, and protective titers may persist well beyond ten years.23 One study showed that 72% of persons had protective antitoxin titers 25 to 30 years after immunization, and one dose of adsorbed tetanus toxoid induced protective titers in the others.“ An- other study showed that nonadsorbed toxoid provided immunity beyond 12 years in children who received four doses.25 The rate of decay of tetanus immunity, based on the work of White et al. ,25 was estimated to be 1% yearly in our model. Each year, 1% of the previous year's immune population is assumed to become non- immune unless revaccination occurs. We tested many annual decay rates, from 0.25% to 5.00%. Single—dose tetanus immunization produces pro- tective antitoxin titers in young adults for at least one year.27 Further, single-dose immunization can be boosted with a single vaccination to > 001 IU/ml. eight to 13 years later.28 Elderly adults mount an ade- quate antibody response to tetanus booster immuniza- tion.” 22 Hence, veterans (who have very high primary immunization rates) are unlikely to require booster im- munization, and single—dose tetanus immunization can be eflicacious for previously immunized or nonim— munized persons. In our model, the efficacy of primary tetanus im~ munization was assumed to be 100% and booster eifi- cacy was estimated to be 90%.“- 2"- 29 In the sensitivity analyses, we varied borh primary eliicacy and booster- eificacy from 50% to 100%. Costs The direct cost of tetanus-diphtheria toxoid to a Veterans Affairs pharmacy in 1986 was $0.43 per dose; , the cost (Red Book) to a commercial druggist was $0.75 per dose. Another source cites approximately, 50.60 per dose. 1 In our base-case analysis, we assumed that toxoid injections would be given as part of other care and would incur no cost beyond the cost of the toxoid itself. This assumption strongly favors more fre- quent vaccination strategies. We also assumed that vac- cination costs will not change in the future. In the sensi- tivity analysis, we allowed the cost ofvaccination to rise as high as S 100 to allow for additional materials, stall, and physician time. TWO factors are important in assessing the medical costs of treating patients who have tetanus: the incuba- tion period and the availability of intensive medical care. Incubation periods longer than ten days are asso— ciated with a 21% mortality compared with 68% for shorter (less than five days) incubation periods.30 In- tensive care unit observation and treatment reduce mortality to 10%, though an average stay of 31 days is required?‘I The typical hospital cosr per case of nonfa- tal tetanus managed in 1986 was approximately $16,000 (5512 X 31 days). We used this value as our base—case estimate. This cosr does not include physi‘ cian fees or lost productivity due to tetanus morbidity, but our sensitivity analysis assessed costs of care up to $250,000 to allow for such fees or lost wages. The cost of determining immune status is so great relative to vaccine cost or the risk of adverse reaction to vaccination that objective determination of immune status has little utility. Apreliminary report ofthe feasi- bility of an inexpensive (SS) enzyme-linked immuno- sorbent assay (ELISA) for antitoxin titers is encourag- ing.32 Nonetheless, an assessment of strategies employing antitoxin titer determination is beyond the scope of this work. Burden of Tetanus Clinical tetanus is rare. In the 1 985 ~ 86 biennium there were only 140 reported cases of tetanus in the United States; only nine of the patients had received any prior tetanus immunizationfi’3 Seventy—one percent of the patients were over 50 years of age. The incidence rate for tetanus remained relatively stable from 1974 to 1986, whereas a steep decline had occurred from 1 955 to 1970. Crude annual incidence rates for tetanus in the 1983 ~84 biennium increased progressively by age decile, with particularly sharp increases in JOURNAL or GENERAL lNTERNAL MEDICINE, Volume 8 (August) 1993 407 C; 0.12" o .3 £5 :3 ‘1‘ 0.1“ o Q C O 0.08— FIGURE 1. incidence of tetanus. 0, 131' age (in years). The rate of reported 8 tetanus cases in the United States during r—l 1982—84 increases dramatically after '5 0-05 age 60.29 Q a) o l: 0.04— a) 'U 'H o C: "‘ 0.02 0 the 50 —~ Sdear-old decile and especially above age 60 (Fig. 1). In fact, patients older than 60 accounted for 59% of all tetanus cases.29 Data from serosurveVs parallel these statistics; 11% of persons aged 18—39 years and 49-66% of persons older than 60 years lacked presumably prOtective (0.01 lU/ml.) levels of antitoxin.‘9~ 2‘- 3“ We used data from the prevaccine era to esrimate the incidence of tetanus in populations that have not received boosters. Because the no-booster strategy (N) includes primary pediatric immunization and wound prophylaxis, tetanus incidence will be lower in this strategy than in historical prevaccine populations. On the other hand, we expect incidence under this Strategy to be higher than that observed in recent US. popula- tions, many of whom received decennial boosters. For our base case, we used four times the incidence 0b5€fV€d in 1985—86 in the United States. This is somewhat lower than that observed in 1920 in Massa— chusetts, before vaccine was available (personal com— munication, 1990, Massachusetts Department of Health). This assumption favors the frequent-booster Strategy and corrects for the underreporting of tetanus?‘5 The case—fatality rate for tetanus in 1982 — 84 var» ied from 11% to 52%, with higher rates in older age gro ups.29 The overall case-fatality rate was 26%, and no death occurred among patients under the age of 30. Patients older than 60 had four times the case—fatality rate (52% vs 13%) of the younger cohort. We used a case-fatality rate of 13% up to age 64 and 52% for per- sons 65 Or older. Although there was no fatality re- ported from tetanus for younger age groups in 1982- 5-19 20—29 30.39 4049 50-59 >59 Age 86 CDC data, we assumed a 13% rate. This assumption favors the decennial strategy. Morbidity of Immunization A large study of nearly 10,000 adsorbed tetanus administrations showed that the incidence of severe or moderate local reactions increased from 0.24% for a first injection to 1.45% for boosrer injections; systemic reactions occurred at a rate of only 0.06%.36 High pre- vaccination antitoxin titers and adverse reactions are positively correlated; antitoxin titers > 5 lU/ml. triple the risk of reaction?“11 Anaphylactic reaction to teta- nus toxoid is rare}8 We assumed that there would be no significant adverse reacrion to vaccine. This assumption favors more frequent vaccination. Model Structure A Markov model was constructed to follow a cohort of 6-year-olds of the US population prospectively over 80 years, to age 85.42 The model is cycled annually to account for non— tetanus—related mortality, age-spe- ciiic incidence rates for tetanus, age-specific mortality rates for tetanus, and decay in immunity. The model is repeated for each of the three vaccine strategies stud- ied. Cumulative costs and cumulative benefits (sur- vival) for each of the three strategies are then compared on a per-person basis. We used the base-case assumptions described tits. We varied each above to calculate the base-case resr variable over the ranges described in Chart 1 for the univariate sensitivity analyses. Balestra. Littenoerg. COSTeEFFECTIVENESS OF TEl‘ANUS VACCiNATlON 408 CHART i I The Basetine Cost—Effectiveness of Difierent Vaccination Strategies Cost in Life Lifetime Cost in Life Cost per Lifetime Lilelime Undiscounted Expectancy Probability Discounted Expectancy Year of Number Cost in Dollars Undiscounted oi Tetanus Dollars Discounted Lite Saved, of Cases Discounted Years Years incremental Undiscounted Dollars to Strategy N Strategy Per person Per cohort of six-year-old children No booster (N) $0.832 68.315319 0.038% 80.060 19.464526 -— 1,391 $219,600 Booster at age 65 (8) 50.996 68.315417 0.030% 80.073 19.464529 $4,527 1,098 $267,180 Booster every ten years (T) $4.135 68.315464 0.016% 80.919 19.464532 $143,138 586 $3,363,540 RESULTS cosr -effectiveness values, for all the variables over the Base-case Analysis The absolute risk of tetanus is small. Even with no adult booster (except wound prophylaxis), the life‘ long probability of tetanus is less than i in 2,700. Full compliance with the decennial booster policy lowers the risk to 1 in 6,000. The Strategy of providing a rou- tine booster only at age 65 reduces the risk to about 1 in 5,500. The dilference in average life expectancies among these three Strategies is small. After discounting, there is only about a 2-minute survival advantage for the decennial-booster strategy (T) compared with the sin- gle-boosrer Strategy (B). Applying the model to all 3.6 million 6—year-olds in a typical year, Strategies B and T reduce the lifetime number of tetanus cases by 295 and 805, respectively, compared with no booster (N ). Both strategies B and T reduce predicted mortality from tetanus, with the strat- egy T conserving the most life. The marginal benefit between the most-effective and least-effective strate- gies is about 3 minutes per person. However, this small difference in effect increases the discounted costs of tetanus (vaccination costs plus costs of care) from 50.06 per person for strategy N, to 8.074 for the strat- egyB, to 50.92 for srrategyT (Chart 1).To change from strategy N to strategy B would cost 54,527 per year of life saved. To go from N to T would cost 8 143,138 per year of life saved. To go from B to T would cost $281,748 per year of additional life saved. Compared with a policy of no booster (N), the single-booster (B) and decennial-booster (T) policies cost 8 1 62 and 53,906, respectively, per case of tetanus averted. One—way Sensitivity Analyses For most variables, over the ranges we tested, the cost—effectiveness changed for all three strategies. However, in most cases the changes were similar for all three strategies and there was no reordering among the strategies. In Other words, the most cost-effective strat- egy at low values for the variable remained most cost: efl’ective as the variable increased. The values for the expected life spans and costs, as well as the marginai ranges we tested, are shown in Chart 2. The rate of decay in tetanus immunity applies across all strategies. Slower annual decay rates make policies with less intervention increasingly attractive (Fig. 2). Even at a decay rate of 5%, the decennial- booster strategy (T) remains less cosr~effective than the single-booster (B) Strategy. Because they depend mainly on boosters, strate- gies B and T are affected little by the efficacy of primary vaccination. Therefore, the marginal cost of T com- pared with B changes little with this variable. However, as efiicacy declines, N becomes less effecrive and more costly. As booster emcacy decreases, strategies 8 and T become less cost-effective. However, the relative difi {erences are small. Within realistic estimates of booster efficacy, there is no change in the relative cosr— effectiveness among strategies. At a boosrer efficacy of 50%, the three strategies have identical expected life span effects with substantially more costs associated with strategy T. Cost of vaccination has little effect on the relative cost among strategies. Emergency room or primary care office visit charges are obviated if vaccine administra- tion is ancillary to other services, such as a blood pres sure check or health maintenance in the office, or wound suturing in the emergency room. Over the range we examined (80— 5100 per vaccination), the cost of vaccination influences all strategies but plays little role in the relative choice among them. Because strategy '1‘ involves many more vaccine administrations, its cost increases much more rapidly with increasing cost of vaccination than do the costs of the other strategies. We were able to detect threshold effects for costs of care. For a cost of care higher than $34,770 per tetanus case, strategy B actually results in both cost savings and improved survival compared with strategy N. Even at a cost of $250,000 per case, strategnyails to dominate strategy N. Strategy T does not dominate strat- egy N until the cost of a case of tetanus exceeds $416,699. It dominates strategy B oniy at costs higher than 561 1.333 per case. Given reasonable estimates of the cost of care, strategy B is a reasonable choice in- JOURNAL OF GENERAL INTERNAL MEDICINE. Volume 8 (August). 1993 409 CHART Z I Sensitivity Analyses Variable N N B B T T B compared to N T 00m ared to N T com 008101 Expec‘ed 005101 1.50390in 005101 Expected Marginal Cost pepr Year of Life Sarvegamd to B Program L119 Years Program L110 Years Program Life Years (douafs) Base Case Analysis $0.060 19.454526 $0.074 19.464529 $0.919 19.464532 $4.527 $143,138 $281,748 Annual Decay in Immunity 0.00% $0.000 19.464536 $0.025 19.464536 $0.893 19.454536 No life saved N0 life saved N0 life saved 0.25% $0.018 19.454533 $0.042 19.464534 $0.909 19.464534 $23,798 $890,978 No life saved 0.50% 50.034 19.464531 $0.053 19.464532 $0.914 19.464533 619.743 $440,154 $860,564 1.00% $0.060 19.464526 $0.074 19.464529 $0.919 19.464532 $4.527 $143,138 $281,748 2.00% $0.098 19.464520 $0.105 19.464525 $0.926 19.464531 $1,321 $75,293 $135,937 5.00% $0.151 19.454512 $0.154 19.464518 $0.947 19.454523 $524 $49,707 $79,217 Efficacy of Primary Vaccinafion 50% $0.125 19.464516 $0.085 19.464528 $0.920 19.464532 ($3,373) $49,662 $208,766 60% $0.112 19.464518 $0.085 19.464528 $0.920 19.464532 ($2.741) $57,689 $208,766 70% $0099 19.464520 $0.085 19.464528 $0.920 19.454532 ($1,793) $68,393 $208,766 80% $0.086 19.464522 $0.085 19.464528 $0.920 19.464532 ($213) $83,378 $208,766 90% $0.073 19.464524 $0.085 19.464528 $0.920 19.464532 $2.946 $105,856 $208,766 100% $0.060 19.454526 $0.074 19.464529 $0.919 19.464532 $4.527 $143,138 $281,748 Booster Efficacy 50% 80.060 19.464526 $0.085 19.464526 $0.953 19.454526 N0 life saved No lile saved N0 life saved 75% $0.060 19.464526 $0.079 19.464528 $0.939 19.454529 $9,251 $293,043 $860,625 100% 80.050 19.464526 $0.070 19.464530 $0.902 19.464535 52.575 $93,523 $166,262 0051 per vaccination $0 $0.060 19.464526 $0.049 19.464529 $0.026 19.464532 ($3.859) ($5.716) ($7.572) $25 $0.060 19.464526 $0.677 19.464529 $22,354 19.464532 $205,787 $3,715,513 S7.Z25.439 $50 $0.060 19.464526 $1.306 19.464529 344.682 19.464532 $415,433 $7,436,942 514.458.4330 375 30.060 19.464526 $1.935 19.464529 367.010 19.464532 $625,078 $11,158,270 $21,691.461 3100 50.060 19.464526 $2.564 19.464529 589.338 19.464532 8834.724 $14,879,598 $28,924,473 Cost per case 01 tetanus $0 $0.000 19.464526 $0.025 19.454529 $0.893 19464532 38,386 $148,853 $289,321 $50,000 $0.188 19.464526 $0.177 19.464529 $0.974 19.464532 ($3,673) $130,992 8265.657 5100.000 $0.376 19.464526 $0.329 19.464529 $1.055 19.464532 . ($15,732) $113,131 $241,994 $150,000 $0.564 19.464526 $0.480 19.464529 $1.135 19.464532 ($27,791) $95,270 $218,331 $200,000 $0.751 19.464526 $0.632 19.464529 $1.216 19.454532 ($39,850) $77,409 3194.668 5250.000 $0.939 19.464526 $0.784 19.464529 $1.297 19.464532 ($51,909) $59,548 $171,005 Fatality Hate Multiplier 0 $0.060 19.464536 $0.074 19.464536 $0.919 19.464536 No life saved N0 life saved N0 Eife saved 1 $0.060 19.464526 $0.074 19.464529 $0.919 19.464532 $4.527 $143,138 3281.748 2 $0.060 19.464517 $0.074 19.464523 $0.919 19.464528 $2,264 $78,075 $169,049 3 $0.060 19.454512 $0.074 19.464518 $0.919 19.464526 $2,263 $61,345 $105,656 4 $0.060 19.464506 $0.074 19.464512 $0.919 19.464523 82.264 $50,519 $76,840 5 $0.060 19.454501 $0.074 19.454507 $0.919 19.464520 62263 $45,201 865.019 Incidence Rate Multiplier 0 $0.000 19.454536 $0.025 19.464536 $0.893 19.464536 N0 life saved No life saved No life saved 0.5 $0.008 19.464535 $0.031 19.464535 $0.896 19.464536 N0 life saved $888,832 $865,122 1 $0.015 19.464534 $0.037 19.454534 $0.900 19.464535 N0 life saved $884,546 $862,282 1.5 $0.023 19.464532 $0.043 19.464534 $0.903 19.464535 $10,408 $293,420 $859,443 2 $0.030 19.464531 $0.049 19.464533 $0.905 19.464534 59,685 $291,991 $856,604 3 $0.045 19.464529 $0.062 19.464531 $0.912 19.464533 $8.237 $216,850 $425,462 4 $0.060 19.464526 $0.074 19.464529 $0.919 19.454532 $4.527 $143,138 $281,748 6 $0.090 19.454522 $0.098 19.464526 $0.932 19.464530 $1,948 $105,210 $208,471 8 $0.120 19454517 $0.12 19.464523 $0.945 19.464528 $334 $74,958 $164,506 10 $0.150 19.464512 $0.147 19.464519 $0.958 19.454526 ($541) 3157.670 5115.881 Discount rate 0.00% $0.832 68.315319 $0.996 68.315417 $4.135 66.315464 $1.678 $22,783 866.790 2.50% 30.205 32.435779 $0.254 32.435796 $1.742 32.435806 $2,884 $56,948 5145.856 5.00% $0.060 19.464526 $0.074 19.464529 $0.919 19.454532 $4,527 $143,138 $281,748 7.50% $0.021 13.590168 $0.025 13.590159 $0.562 13.590169 $3.665 $540,577 No 1119 saved 10.00% $0.009 10.411912 $0.010 10.411912 $0.377 10.411912 N0 1119 saVed N0 lile saved N0 1119 saved Values in parentheses indcate that the program saves lives and money. 410 Balestra. Littenberg. COST—EFFECTIVENESS OF Tetanus VACClNATION 5 4:. "Q C (1 U3 3 O .5: 3: ca IO Z O a.» o' .E “U Q) > rd 0) 239 C 0 L4 53 Q) >3 Li Q.) Q 4.; U) 0 C.) 0 0.005 0.01 0.0I5 0.02 0.025 0.03 0.035 FIGURE 2. Sensitivity anaiysis of decay rate in tetanus antitoxin titer. The horizontal axis represents different decay rates in immunity. The vertical axis repre- sents the discounted cost—effectiveness of each strategy in thousands of dollars per year of life saved. expressed incrementai to no intervention (N) after pediatric immuni- zation. Each line represents a strategy at different annual decay rates for immunity. As the decay rate increases. both strategies become more attractive. especially the do cennial~booster strategy. B = booster at age 65 only; T = decenniai boosters from age 6 onward. 0.04 0.04 5 0.05 Annual Rate of Decay in Immunity volving a tradeofi’ between resources spent and lives saved. Varying the fatality rates up to five times the base- case esrimates had little impacr on the cosr— efiectiveness of the three choices. A threshold applies to incidence rates for tetanus (Fig. 3). The age-specific incidence rates were varied from one-half the 1982—84 rates to tenfold the base rates. Strategy B resulted in cost savings and life savings relative to strategy N at incidence rates for tetanus about ninefold the 1982—84 rates. At more realistic incidence rates for tetanus, strategies N and B provide reasonable alternatives in which resources are con— served at the expense of some survival. As the discount rate increases, all strategies be- come less attractive, with strategy T affected most ad— versely. However, at any given discount rate, the rela tive cost—effectiveness among the three strategies is the same. Generally, the relative choices among the three strategies are remarkably insensitive to wide fluctua- tions in the variables we analyzed. Very high incidence rates for tetanus, extreme costs for care, and more rapid decay in immunity improve the reiative merit of strat- egy T. Multi—way Sensitivity Analyses We varied two or more variables simultaneously to estimate what conditions would be necessary for strat- egy T to become a clear-cut winner in terms of both cost and efficacy. In other words, when does T dominate both E and N? it takes very extreme combinations of key variables to produce such resuits, For instance, if the cost of caring for a case of tetanus exceeds $180,000 and the rate of decay in immunity exceeds 5% per year. T is clearly preferred. Similarly, ifthe incidence oftent- nus is tenfold higher than that reported in 1982—855 and the cost of care is $100,000 and the decay in immunity exceeds 3% per year, '1‘ is preferred. What if the booster were much better than we esti mated? T becomes the preferred strategy only if the booster is 100% elfective in restoring immunity and it costs less than $0.06 per dose to deliver. What if we have underestimated both the inci‘ dence and the lethality of tetanus? We found no combi- nation of these two variables so high that T dominates B. Even under conditions such that T is much more elfec- tive, B is less costly. DISCUSSION We found that the decennial-booster strategy is most attractive when efiectiveness alone is considered, but suffers from relatively high cost. Further, the cur- rently recommended strategy is hindered by poor compliance, as judged by serosurveys and chart audits.” 21- 34' 43 A singie booster at age 65 is a conve- nient component of a health maintenance program in— cluding screening tests and other vaccinations. in part, the compromised cost—eifectivcness of the current standard results from its failure to target those at great est risk from tetanus, attempting instead to maintain immunity in the entire population at all times. This standard has not been achievable.” 45 Because the de- cennial-booster strategy is difficult to administer, be- cause incidence data suggest that much eifort is di. rected at those unlikely to get the disease, and because JOURNAL or GENERAL lNTERNAL MEDICINE. Volume 8 (August). 7993 41 1 FIGURE 3. Sensitivity analysis of various incidence rates for tetanus. The hori- zontal axis represents multiples of the 1982—84 incidence rates for tetanus. The vertical axis represents the discounted cost— effectiveness of each strategy in thou— sands of dollars per year of life saved. ex- pressed incremental to no intervention (N) after pediatric immunization. The base case uses an incidence multiplier of4. Strategy B (booster at age 65 only) remains more cost- effective than strategy T (decennial boosters from age 6 onward) over a wide range of incidence rates for tetanus. At rates greater than nine times the 1982—84 incidence. strategy B becomes tost~saving and dips below the bottom of the graph. (Thousands) Cost per Year of Life Saved (Inc. to N) health insurance is nearly universal for Americans aged 65 years and older, an alternative Strategy aimed at 65- year-olds warrants consideration. We provide the cost—effectiveness for each scrat- egy and nore that the relative ordering among the three strategies does not vary over mosr of the ranges tested in our sensitivity analysis. For the decision maker who musr choose among the three strategies without refer» ence to competing health needs, the optimum choice does not change over the ranges we tested. More realis- tically, the decision maker must compare a tetanus—pre- vention strategy with other competing demands. In that case, the decision must be made by comparing the cost~eifecriveness of the three strategies with those of other health programs. For instance, it may be more reasonable to provide screening for hypertension to older men ($8,574 per year of life saved)46 than to provide decennial tetanus boosters ($143,138 in the base case). If the current vaccination standard is forsa- ken and single~boosrer immunization for 65-year-old persons is adopted, 22 life-years will be lost from a cohort of 3.6 million children, but 53.14 million will be saved. If these savings were invested in hypertension screening of 60—year-old men, 375 life-years could be saved. Our model has several important limitations. Se- vere underreporting of tetanus and gross underestima- tion of costs for tetanus care could bias results in favor of Strategies with less intervention. However, it is un- likely that underreporting would greatly exceed the fourfold adjustment used in our base-case analysis. Also, limited information is available about the natural history ofimmunity, initially induced by vaccination in childhood. Natural immunity to tetanus is not consid 3 4 5 6 7 8 9 10 incidence Multiplier ered explicitly in our model and might confound our results.47‘49 Such immunity would lessen the need for frequent boosters. Further research is necessary in several areasAvail- able data suggest that lethal tetanus is a geriatric dis— ease.” 33 As the above—85 age group is the fastest-grow— ing segment of the U.S. population, further Studies of the rate at which immunity wanes, particularly in the geriatric population, will be needed. Should veterans be subjected to the current standard of tetanus booster immunization when available evidence shows strong correlations between long—lasting immunity and prior military service?" 12-34 The issue is significant as vet— erans constituted 26.6% of 50—year—olds in 1986, and an increasing number among the aging US. population will be veterans, Elderly women may be more vulnera- ble to tetanus than elderly men.“ Should special elforts be directed toward elderly women, or women generally? Strategies for tetanus prophylaxis should be aimed at the population with the highest incidence, least pro- tection, and highest case-fatalityrate. Therefore, in lieu of decennial boosters and in addition'to pediatric pri- mary immunization and wound prophylaxis, we recom- mend providing a single tetanus booster to all adults at about age 65. REFERENCES 1. Williams \\"\\", Hickson MA, Harte MA. et :11. Immunization poli~ cies and vaccine coverage among adults. Ann Intern Med. 1988;103:6165. Frame PS. Critical review of adult health maintenance, Part 2. Prevention ofinfectious diseases] Fam Pmct. 1986;22:417—21 Immunization Practices Advisory Committee. General recom— b.) OJ 412 \II 10. 11. 17. 18. 19. . Edsall Baiestra. Lirtenberg. COST—EFFECTIVENESS OF TETANUS VACCINATION mendations on immunization. Ann Intern Med. 1989; 111:133-42. . LaForce FM. Immunizations, immuIto—prophylaxis, and chemo- prophylaxis to prevent selected 1987-, 257:2464~70. infections. gAMA. . US. Preventive Service Task Force. Guide to Clinical Preventive Service. Baltimore: Williams and Wilkins, 1989. . immunization Practices Advisory Committee. Diphtheria, teta~ nus and pertussis: guidelines for vaccine prophylaxis and other preventive measures. MM\VR. 193554240525. . Hutchison BG, Stoddart GL. Cost—effectiveness of primary teta- nus \nccination among elderly Canadians. Can Med Assoc J. 1988;139tl 14551. Rappuoli R, Petugini M, Falsen E. Molecular epidemiology oftlte 1984—1986 outbreak of diphtheria in Sweden. N fingl 3 Med. 1988;318:124. Karzon DT, Eduards KM. Diphtheria outbreaks in immunized populations. N Engl] Med. 1988;318:416. Bader M. Diphtheria in immunized populations. N Engl} Med. 19885 19:446-7. Tarangerj, Rabo E. Diphtheria in immunized populations. N Engl] Med 1988;319:447. . HamischJP, Tronca E, Nolan CM, Turclt M. Holmes iCK. Diphthe- ria among alcoholic urban adults. A decade of experience in Seattle. Ann Intern Med. 1989;111:7132. . Hoblin BA, Townsend TR. Immunity to diphtheria and tetanus in inner-city women of childbearing age. Am J 1>ublic Heaith. 1989;79:1297-8. . U.S. Bureau of the Census. Statistical Abstracts of the United States; 1990 (1 10th edition). Washington, DC, 1990. G. Specific prophylaxis of tetanus. LAMA. I71 z—‘i 17-27. 1959; . Berger SA, Cherubin LE, Nelson 5, et al. Tetanus despite pte-ex- isting antitetanus antibody. JANA. 1978;240:769-70. Passer: EL, Anderson BR. Clinical tetanus despite a “protective” levei of toxin-neutralizing antibody. JAMA. 1986;255:1 171-5. Maselle S‘i”, Matte R. Moise R, Hofstad T. Neonatal tetanus despite protective serum antitoxin concentration. FEMS Microbiol Im- munol.1991;3(3):171«5. Crossley K. Irvine P, Warren B, et al. Tetanus and diphtheria immunity in urban Minnesota adults. JANA. 1979;242: 2298300. . Ruben FL, Nagei J, Fireman P. Antitoxin responses in the elderly to tetanus—diphtheria (Td) immunization. Am J Epidemiol. 1978;108:1459. . Weiss 8?, Strassburg MA, FeeleyJC. Tetanus and diphtheria im- munity in an elderly population in Los Angeles County. Am J Public Health. 1983;73:802»4. . Simonsen 0, Block AV, Klaerl-x A, et a1. Immunity against tetanus and response to revaccination in surgical patients more than 50 years of age. Surg Gynecol Obstet. 1987;164:329-54. . Trinca JC. Antibody response to successive booster doses of raw nus toxoid in adults. Infect Immun. 1974;10:1-5. . Simonsen O, Kieldsen K, Heron I. Immunity against tetanus and elfcct of revaccination 25—30 years after primary vaccination. Lancet. 1984;11240-2. . Peebles TC, Levine L, Eldrcd MC, Edsall G. Tetanus toxoid emer- gency boosters. A reappraisal. N Engl} Med. 1969;280:57S»81. 26 48. 49. . White WG, Barnes GM, Grillilh All, Gall D, Barker Ii, SmithJW. Duration ol'immunity after active immunization against tetanus, L'IllCCI. 1969:2295-6. Matvcev KI, Sergecva 'l‘l, Zurubina IEK, Minacva 2L. Single-dose immunization of human subjects against tetanus. 1 mm Epidc. miol Microbial Immunol. 1976;20:1121-30, Ulll3erg-Olsson K, Eriksson E. Active immunization against [ct-.1. nus in man. 1. Duration ofanamnesric tcacrion after one dosc of vaccine. Iiur Surg Res. 1975;7249-58. Immunization Practices Advisory Committee. 'I‘ctanus—Unticd States, 1982—1984. MMWR. 1985;34:602-11. Singh GP, Sikka PK, Gupta MM. 'l‘etanus— a method ofscoring to determine prognosis. Indian] Med Sci. 1986;40:124-3 . Edmondson RS, Flowers MW. Intensive care in tetanus: manage- ment, complications, and mortality in 100 cases. BM}. l979;1:1401-4. . ank jW, Schabas R, Arshinofl‘ R, Brant R. Diphtheria—tetanus over~immunization in children with no records: can it be pre- vented? Can Med Assoc). 1989;141:1241-6. Immunization Practices Advisory Committee. Tetanus—«United States, 1935—1986. MMWR. l987;36;477.81_ ’. Mullooly JP. Tetanus immunization of adult members of an HMO. Am] Public Health. 1984;74:841-2. . Sutter RW, Cochi SL, Brink EW, Sirotkin 131. Assessment of vital Statistics and surveillance data for monitoring tetanus mortality, United States, 1979— 1984'. Am} Epidemioi. 1990;131:132-42. White WG. Reactions after plain and adsorbed tetanus vaccines. Lancet. 19801.42. Levine L, Edsall G. Tetanus toxoid: what determines reaction proneness?j Infect Dis. 1981;144:376. Mansfield IE, Ting S, Rzuvls DO, Frederick R. Systemic reactions during cutaneous testing for tetanus toxoid hypersensitivity. Ann allergy. 1986;57zl35-7. . Zimmerman [1, Gold R, Lavi S. Adverse effects of immunization. Postgmd Med. 1987;82:225-32. . Edsall G, Elliott MW, Pecblcs TC, ct al. Excessive use oftctanus toxoid bOOSICFS.}A1\lA. 1967;202:1113. . jacobs RL, Lowe RS. Innicr BL. Adverse reactions to tetanus tox— Oid.in‘le. 1982;247:409. . Beck fR. Pauker SC. The Markov process in medical prognosis. Med Decis Making. 1933;3ri I968. . Hilton E, Singer C, Kozarsky P, Smith MA, L—irdis MP, Borenstein MT. Status ofimmunity to tetanus, measles, mumps, rubella, and polio among US. travelers. Ann Intern Med. 1991;115:326. . InForce FM. Adult immunizations: are they worth the trouble?) Gen Intern Med. 19905 (5/0 suppl):SS7-861. . Belcher DW. Implementing preventive services: success and fai l— ure in an outpatient trial. Arch Intern Med. 1990;150:2533-41. . Littenberg B, Garbet AM, Sox HC. Screening for hypertension. Ann Intern Med. 1990;112:192~202. . Saha K, Sharma VK, Seltgal VN, Agarwol SK. Natural resistance against tetanus in patients with lepromatous leprosy. Trans RSoc Trop Med Hyg. 1981;75:832-4. Dostur FD, Awatramani VP, Dixit SK, et al. Response to single dose of tetanus vaccine in subjects with naturally acquired teta— nus antitoxin. Lancet. 1981;2:219.22. Singh M, Kumar B, Ayagiri A, Kant 5. Natural tetanus immunity in lepromatous leprosy patients. Indian] chr. 1986;58:915- ...
View Full Document

Page1 / 8

Balestra+JGIM+1993 - ORlGINAL ARTICLES Should Adult Tetanus...

This preview shows document pages 1 - 8. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online