Journal of Medical Virology 37:4%53
(1992)
Low-Cost Hepatitis B Vaccine Improves Uptake Among Self-payingHealth-Care Students Ross A. Pennie, Annette M. O’Connor,Corinne S. Dulberg, Anna Bottiglia, Pranlal Manga, and C. Yong Kang Department of Pediatrics, Faculty of Medicine (R.A.P.); School of Nursing, Faculty of Health Sciences (A.M.O.); Department of Epidemiology & Community Medicine, Faculty of Medicine (C.S.D.); Children’s Hospital of Eastern Ontario, Ottawa (A.B.);Faculty of Administration (P.M.); and Department of Microbiology & Immunology, Faculty of Medicine (C.Y.K.), University of Ottawa, Ottawa, Ontario, Canada.
Advisory committees recommend hepatitis B (HBV) immunization for professional and student health-care workers. However, the currently Iicensed vaccines are expensive, and previous surveys have shown that few students (14%) have been immunized in Canada. A low-cost immunization program was offered t o health-care students in order t o determine whether the effectiveness of HBV immunization could be improved by substantially reducing the vaccine cost t o recipients. The immunogenicity, side effects, and 3-dose completion rate of a low-cost Korean HBV vaccine were compared with a similar U.S.-made vaccine. A total of 922 postsecondary students enrolled in 6 health-care disciplines in Ottawa, Canada were surveyed for hepatitis-6 immunization status. Nonimmunized students were subsequently offered HBV vaccine at total cost of $15 (Canadian), randomly allocated t o receive 3 intramuscular doses of either Korean or U.S.-made plasma-derived HBV vaccine in a double-blind fashion, surveyed about side effects, and tested for hepatitis B surface antibody seroconversion. Only 12% of the 922 surveyed students had been previously immunized when vaccine was obtainable only at high cost. However, 66% of those not immunized participated in the vaccine trial and paid the $1 5 fee. Hepatitis-B surface antibody seroconversion ( 3 1 0 sample ratio units by radioimmunoassay) occurred in 291/311 (93.6%) and 299/310 (96.5%) of recipients of 3 doses of the Korean and U.S. vaccines, respectively ( P = 0.10). There were no meaningful differences in vaccine adverse effects, and 92.6% of recipients of either vaccine completed 3 doses. These results support the view that HBV immunization would be more effective (vaccine acceptance rates higher, greater proportions of persons protected) if lower cost vaccines were marketed. 8 1992 WiIey-Liss, Inc. 0 1992 WILEY-LISS. INC.
KEY WORDS: HBV, hepatitis B immunization, health-care workers, Korean hepatitis B vaccine, U.S. hepatitis B vaccine
INTRODUCTION Hepatitis B (HBV) is a viral infection of the liver transmitted by sexual activity and by inoculation with infected blood and body fluids. Among those infected as adults, 6 1 0 % become lifetime carriers of the virus [United States Immunization Practices Advisory Committee, 19851. Carriers can transmit infection to others, and themselves have a greater than 25% incidence of chronic hepatitis I United States Immunization Practices Advisory Committee, 19851, which can lead to hepatic cirrhosis and primary cancer of the liver at a rate up to 300 times that of noncarriers [Palmer-Beasley et al., 19811. In North America, HBV prevalence is significantly increased among health-care workers [Pattison et al., 1975; Dienstag et al., 1982; Chaudhary, 1983; United States Immunization Practices Advisory Committee, 1985; Malm e t al., 19861. As a consequence, the United States Immunization Practices Advisory Committee [19851and the Canadian National Advisory Committee on Immunization 119871 recommend that professional and student health-care workers receive prophylaxis with HBV vaccine. Despite these recommendations, most health-care students are not being immunized [Kwon et al., 1984; Pennie and Miller, 19871. A survey of health-care students in Ottawa, Canada confirmed the low vaccine Accepted for publication November 15, 1991. Address reprint requests to Dr. R. A. Pennie, Infectious Diseases and Microbiology, Room 3N27G, McMaster University Medical Centre, 1200 Main Street West, Hamilton, Ontario, Canada L8N 325.
Low-Cost Hepatitis B Vaccine uptake rate and suggested that cost appeared to be the major barrier to student immunization [Pennie e t al., 19911. The present study was conducted in order to determine whether the effectiveness of HBV immunization could be improved by reducing substantially the vaccine cost to recipients through the use of a low-cost Korean vaccine. The immunogenicity, side effects, and 3-dose completion rate of the plasma-derived Korean HBV vaccine were compared with a similar U.S.-made vaccine.
METHODS Student Recruitment In September 1989, students enrolled in six healthcare programs in Ottawa were approached in class by a research nurse for participation in a n experimental immunization program. Students were asked to (1)read a statement about HBV risks and vaccine recommendations; (2) complete a questionnaire eliciting current HBV immunization status, demographic characteristics, and willingness to pay for the vaccine; and (3)read a statement about a proposed HBV vaccine trial. For the price of $15 towards the cost of the vaccine, they would receive 3 doses of either a currently licensed American plasma-derived vaccine or a similarly but not identically prepared vaccine made in Korea that was currently available in Canada only for use in this study. Students were advised of the dates of immunization sessions and were asked to appear if they wished to receive the vaccine. The study was approved by the ethics committee of the University of Ottawa and the Children’s Hospital of Eastern Ontario. Vaccines The vaccines used were Heptavax-B lot M0286 (Merck Sharp & Dohme Canada, Kirkland, Quebec) and Hepavax-B lot H0655 (Korea Green Cross Corporation, Seoul, Korea). Both vaccines were plasma derived and contained 20 pglml of inactivated HBV surface antigen (HBsAg) formulated with alum adjuvant and thimerosal preservative. The retail cost of the Merck vaccine in Ottawa a t the time of the study varied from $135 to $175 (Canadian funds). The estimated retail cost of the Korean vaccine, not yet licensed in Canada, was $15. Immunization and Blood-Testing Schedule Prospective participants who attended the immunization sessions scheduled for their class were screened for the following exclusion criteria: age less than 17 years, poor general health, current immunosuppression or pregnancy, past severe reaction to any vaccine, previous doses of HBV vaccine or HBV immune globulin, and previous known positive tests for HBV. Following informed consent, participants were randomized (in blocks of 24 containing 12 of each vaccine) to receive either the Merck Sharp & Dohme vaccine (USV) or the Korea Green Cross vaccine (KRV). The
49
participants and investigators were blinded with respect to vaccine assignment. The immunization schedule was designed to fit into the &month school year in order to optimize student compliance. Within the first 2 weeks of the 1989-90 school year, each participant (1)had a preimmunization blood sample for measurement of hepatitis-B surface antigen (HBsAg), antibody to hepatitis B core antigen (anti-HBc), and antibody to HBsAg (anti-HBs); (2) was given the first dose of 1.0-ml vaccine intramuscularly in the deltoid muscle on the same day; and ( 3 )paid the $15 fee. One and 6 months following the first dose of vaccine, blood was again obtained for anti-HBs, and the second and third doses of vaccine, respectively, were administered. Six weeks following the third dose of vaccine, a blood sample was obtained for measurement of HBsAg, anti-HBc, and anti-HBs.
Vaccine Immunogenicity The HBV serology was examined using the following commercially available radioimmunoassay kits: Ausria-11, Corab, and Ausab (Abbott Laboratories). AntiHBs levels were expressed semiquantitatively as the ratio of test sample value to mean negative control value in sample ratio units (SRU). A vaccine-induced seroconversion was defined as the development of protective levels of anti-HBs, i.e., a10 SRU [Francis et al., 1982; Hadler et al., 19861in the absence of seroconversion of anti-HBc in a participant whose first blood sample was negative for HBsAg and anti-HBs. Vaccine immunogenicity was defined as the proportion of vaccineinduced seroconversions in participants who received 3 doses of vaccine and the follow-up anti-HBs test. A clinically meaningful difference in the immunogenicity of the KRV compared to the licensed USV was defined as a difference in proportions of more than 0.10. Participants were considered previously infected and not given more than 1dose of vaccine if their preimmunization blood sample tested (1)positive for HBsAg, (2) positive for anti-HBs plus anti-HBc, or ( 3 ) anti-HBs 3 1 0 SRU and anti-HBc negative. Participants whose preimmunization sample showed very low levels (1.5-9 SRU) of anti-HBs alone, or anti-HBc alone, were defined as not interpretable and were offered 3 doses of vaccine but were not included in the analysis of vaccine immunogenicity . Survey of Side Effects At the time of each dose of vaccine, participants were given thermometers and asked to record their oral temperature once daily for 5 days. They were also asked to complete a check list of the duration and severity of side effects on a standardized form that was to be returned to a class representative within 10 days. A difference of 0.25 in the reported incidence of the various symptoms following the 2 vaccines was considered clinically meaningful.
Pennie et al.
50
Participant Compliance The completion rate was defined as the proportion of participating students who completed all 3 doses of vac-
TABLE I. Characteristics of Students Receiving HeDatitis B Vaccine Vaccine administered KRVa USVb
cine.
Effectiveness of Immunization Programs The effectiveness of the immunization program measured the proportion of nonimmunized students attending the HBV information session in class who (1)decided to accept immunization, and (2) received all 3 doses of vaccine, and (3) developed protective levels of anti-HBs. The calculations of effectiveness, outlined below, did not take into account students who received fewer than 3 doses of vaccine.
Effectiveness of the Low-Cost Program Using KRV Vaccine Worst-case (observed) scenario: [proportion of students who participated in the immunization trial after being approached in class] x [KRV immunogenicity] x [KRV completion rate]. Best-case (theoretical) scenario: [proportion of students who said they would pay $15 for licensed vaccine] x [KRV immunogenicity] x [KRV completion rate].
Effectiveness of the High-Cost Program Using USV Vaccine (Theoretical) [Proportion of students who said they would pay the full price for vaccine] x [USV immunogenicityl x [USV completion rate]. The chi square test was applied to the immunogenicity and side-effect data to test for differences between the vaccine groups.
RESULTS Student Recruitment The questionnaire was administered to 922 students in the 6 health-care programs listed in Table I. Class attendance was 75% or greater on the test day, and over 95% of the students present in class completed the questionnaire. Of these, 12% of first year and 13% of upperclass students reported that they had already received HBV immunization, 33% stated that they would pay $175 for the vaccine, and 96% said they would accept the vaccine if it cost $15. When the vaccine trial began, 5351811 (66%) of the nonimmunized, surveyed students participated, a s did a further 1581419 (37.7%) of the other health-care students on campus, including those whose classes had not been surveyed but had heard of the study by word of mouth. Table I illustrates the characteristics of the 693 participating students. The groups were similar with regard to demographic, educational, and clinical characteristics.
Vaccine Immunogenicity Table I1 illustrates the immunogenicity of the 2 vaccines. In the 621 participants who started with negative
Total number students enrolled Sex: Males Age: Range (years) Median (years) Program of study Ambulance & Emergency Dental Assistant Medicine Medical Technology Nursing Rehabilitation Therapy Race White Black Oriental West Asian HBV Tests a t Enrollment Carriers of HBsAg Anti-HBs 210 SRU Dropped Out before 3rd doseC Completed 3 doses of vaccine
349 23.5%
344 25.6%
17-47 21
17-48 21
3.7% 5.4% 18.3% 8.7% 55.3% 8.6%
17.4% 9.5% 57.0% 8.2%
90.8% 3.2% 2.3% 3.7%
85.2% 6.1% 3.2% 5.5%
1.1%
0.6% 5.4% 92.8%
3.5% 4.4%
0.9% 2.3% 4.4%
92.4%
aKorea Green Cross. bMerck Sharp & Dohme. CExcluding students HBsAg positive or anti-HBs 210 SRU a t enrollment.
TABLE 11. Anti-HBs Response to Two Hepatitis B Vaccines in Evaluable ReciDientsa Vaccine administered KRVb USVc No. of evaluable recipientsa 311 310 Vaccine immunogenicitya 93.6% 96.5% P = 0.10 (Proportion who developed anti-HBs 210 SRUd) Distribution of Anti-HBs valuesa (SRUd) 500 80.4% 84.9% aExcluding students positive for HBsAg, anti-HBs, or anti-HBc a t enro!lment and including only those who received 3 doses of vaccine. bKorea Green Cross. CMerck Sharp & Dohme. dSRU: sample ratio units (sample value/mean negative control) measured 6 weeks after 3rd dose of vaccine. P values were determined by chi square.
blood tests for HBV serology and completed the 3-dose program and follow-up HBV serology, the rates of seroprotection (anti-HBs a10 SRU) were 93.6% for KRV and 96.5% for USV (P = 0.10). There was no significant difference in the distribution of peak anti-HBs values between the vaccine groups (x2 = 3.31, D.F. = 3, P = 0.35). The sample size was sufficiently large to ensure
Low-Cost Hepatitis B Vaccine
51
TABLE 111. Adverse Effects Recorded Within 5 Days of Receiving Hepatitis B Vaccine by Students With Negative Hepatitis B Serology at Time of Enrollmenta Vaccine doses administered KRVb usvc n = 957 n = 961 Recipients reported Sore arm after injection 70.6% 64.9% P c 0.01 Median duration sore arm 1 day 1 day Temperature 238°C 1.3% 2.6% P = 0.03 Malaise 11.1% 10.4% Median-duration malaise 1 day 1 day Excessive tiredness 13.2% 13.3% Median duration tiredness 2 days 2 days 1.3% 2.0% Missed school or work Other minor symptoms 7.6% 8.6% a1918 doses of vaccine given to 659 students. bKorea Green Cross. CMerck Sharp & Dohme. P values were determined by chi square.
with 80%power and 95%confidence that the proportion of participants seroconverting after KRV was within 6%of the proportion seroconverting after USV. Vaccine Side Effects There were no serious adverse effects of immunization, and none of the 621 blood samples taken after the third dose of vaccine was consistent with acute infection with live HBV. The incidence of reported side effects did not differ for the first, second, or third doses of either vaccine; therefore, all three doses were analyzed together, as illustrated in Table 111. The 17 students whose first blood tests indicated previous HBV infection were analyzed as a separate group: 41.2% had a sore arm, 11.8%had malaise, none had fever, and none missed school. These results do not differ from those of students who tested negative for HBV markers when given the first dose of either vaccine: 58.4%sore arm, 11.5%malaise, 2.1% fever, and 1.2%missed school.
effectiveness using the KRV in the theoretical bestcase scenario was 83.4%:[96%said they would pay the low price] x [93.6%immunogenicityl x [92.8%completion]. The theoretical effectiveness of a n immunization program using the high-cost USV was only 29.4%:[33% said they would pay the high pricel X [96.5%immunogenicityl x 192.4%completion]. Cost Effectiveness of the Vaccines Given the statistically equivalent results with respect to immunogenicity, side effects, and completion rate, the 2 vaccines were virtually identical when used within the blinded immunization trial. During this immunization program, the unit cost of capital, administration, and labor was identical for the 2 vaccines, but the ingredient cost of the KRV was significantly lower than the USV. Therefore, the KRV would be more cost effective.
DISCUSSION The two vaccines studied in this trial were shown to be equally safe and immunogenic, with hepatitis-B surface antibody seroprotection rates comparable to those reported by others [Szmuness et al., 1980, 1982; Krugman et al., 1981; Francis et al., 1982; Dienstag et al., 19841.The low-cost vaccine program resulted in greatly increased student uptake of HBV vaccine. Therefore, a vaccine such a s the KRV, marketed a t low cost, would be more effective than the currently licensed more costly vaccines. Even when HBV vaccine was offered in the context of a trial involving the use of a foreign, plasma-derived Effectiveness of Immunization P r o g r a m s vaccine, students accepted it in large numbers. This The observed effectiveness of the immunization pro- suggests that the current perception by the medical gram (proportion of students offered vaccine who sero- community th a t the public will not accept plasmaconverted) using the low-cost KRV in the context of the derived vaccines, which has led to the recent withvaccine trial was 57.3%:[66%participation] x [93.6% drawal from the market by Merck of its plasma-derived immunogenicity] x [92.8%completion]. The program HBV vaccine, may not be entirely accurate. Theoretical
Participant Compliance The proportion of participants completing 3 doses of vaccine was 92.8%for KRV and 92.4%for USV. Fiftyone participants did not complete 3 doses: 7 were found to be positive for HBsAg; 10 were positive for anti-HBc and anti-HBs, indicating previously resolved HBV infection; 16 left school; 6 became pregnant; 5 were fearful of further injections; 4 lost interest; 1 was in hospital for unrelated reasons; and 2 had suspected but unproven allergic reaction to the vaccine (1 KRV, 1 USV).
Pennie et al.
52
factors, such as worries about getting AIDS from plasma-derived vaccine, fear of needles, and the inconvenience of a 3-dose regimen become much less important impediments to HBV immunization when the cost is made affordable. A limitation of this study is t h a t it did not directly compare a low-cost HBV immunization program with a high-cost program. The increased HBV immunization uptake that occurred as a result of this study may not be due only to the low cost. Other factors, such as convenience and peer enthusiasm, may have encouraged enrollment. However, the current study and the survey carried out the previous year [Pennie et al., 19911demonstrated that, despite their awareness of the recommendations regarding HBV immunization, students a t all levels had low immunization uptake when only the costly vaccines were available. This was likely a n accurate reflection of the true immunization rate. HBV immunization is not easily forgotten because the vaccine is expensive and not given jointly with routine immunizations. The 3-dose completion rate of 92.6% reflected the high motivation of health-care students and justified the implementation of a HBV immunization program for them. Although the proportion of sore arms after injection was high, and up to 11% of students said they felt unwell for a day or 2, the number of missed school days was small. The aim of a n efficient immunization program is to induce immunity against HBV in as many students as possible without subjecting many students to unnecessary tests or giving vaccine to many students who cannot benefit from i t because they are HBV carriers or already immune. As a result of this study design, in which HBV test results were not available before the administration of the first dose of vaccine, 2.4% of recipients did not benefit from the vaccine because they had been infected previously. In populations such as the one studied here, with a low prevalence of previous HBV infection and a high seroconversion rate, a program that immunizes all participants without pre- or postimmunization blood work would appear to be warranted because i t minimizes cost and inconvenience while protecting over 90% of those who receive 3 doses. This is especially the case if a low-cost vaccine is used t h a t minimizes the cost of the small number (2.4%) of unnecessary doses. Furthermore, this study confirmed the observation [Dienstag et al., 19821 that giving vaccine to people who have already had HBV infection was not associated with a n increase in reported side effects. Seven of the students who enrolled in the study were found to be HBsAg seropositive, yielding a n HBsAgseroprevalence rate in this population of 1%. This figure was higher than the previously reported rates of 0.1-0.3% in U.S. blood donors and Canadian armed forces personnel [Mosely et al., 1975; Smith et al., 1976; Chaudhary e t al., 19831. Notably, 71% of the seropositive students had recently immigrated to Canada from areas including Subsaharan Africa and Southeast Asia, where hepatitis B is highly endemic. The influence of
HBsAg-seropositive immigrants on the incidence of new cases of hepatitis B in a low-endemicity host country such as Canada warrants further study. Although this study only examined the immunization behavior of Canadian health-care students in a context of self-pay, our results should be applicable to individuals employed in informal settings (doctors’ offices, foster care, private nursing homes, and overseas voluntary service) where employees must self-pay for HBV vaccines currently marketed in industrialized countries at high cost. The observed increased effectiveness of the $15 vaccine program suggests that if a lowcost HBV vaccine were readily available, the vaccine uptake rate might increase substantially. Therefore, it would be prudent to market low-cost HBV vaccines in order that HBV immunization be within the reach of all those who would benefit from it. Inexpensive vaccines may one day allow HBV immunization to be universal.
ACKNOWLEDGMENTS This study was supported by research grant #02503 of the Ministry of Health of the Province of Ontario, Canada. The authors would like to thank Dr. Peter Rowe and Dr. David Scheifele for their careful and constructive review of this manuscript. REFERENCES Chaudhary RK (1983): Prevalence of hepatitis B markers in selected health care personnel from the Canadian Forces. Canada Diseases Weekly Report 9-50:197-198. Dienstag JI, Ryan DM (1982): Occupational exposure of hepatitis B virus in hospital personnel. Infection or immunization? American Journal of Epidemiology 115:2639. Dienstag J L , Stevens CE, Bhan AK, Szmuness W (1982): Hepatitis B vaccine administered to chronic carriers of hepatitis B surface antigen. Annals of Internal Medicine 96575-579. Dienstag JL, Werner BG, Polk BF, Snydman DR, Cravan DE, Platt R, Crumpacker CS, Ouellet-Hellstrom R, Grady GF (1984): Hepatitis B vaccine in health care personnel. Safety, immunogenicity, and indicators of efficacy. Annals of Internal Medicine 101:3440. Francis DP, Hadler SC, Thompson SE, Maynard JE, Ostrow DG, Altman N, Braff EH, OMalley P, Hawkins D, Judson FN, Penley K, Nylund T, Christie G, Myers F, Moore J N , Gardner A, Doto IL, Miller JH, Reynolds GH, Murphy BL, Schable CA, Clark BT, Curran J (1982): The prevention of hepatitis B with vaccine. Report of the Centers for Disease Control multi-center efficacy trial among homosexual men. Annals of Internal Medicine 97:362-366. Hadler SC, Francis DP, Maynard J E , Thompson SE, Judson FN, Echenberg DF, Ostrow DG, O’Malley P, Penley KA, Altman NL, Braff E, Shipman GF, Coleman PJ, Mandel E J (1986):Long-term immunogenicity and efficacy of hepatitis B vaccine in homosexual men. New England Journal of Medicine 315:209-214. Krugman S, Holley HP, Davidson M, Simberkoff MS, Matsaniotis N (1981): Immunogenic effect of inactivated hepatitis B vaccine. Comparison of 20 pg and 40 pg doses. Journal of Medical Virology 8:119-121. Kwon H-J, Keenan KM, Colman H, Sundeen KM, Waite DE (1984): Compliance with a hepatitis B vaccination program. Journal of Dental Education 48:563-565. Malm DN, Mathias RG, Turnbull KW, Kettyls GD (1986): Prevalence of hepatitis B in anaesthesia personnel. Canadian Anaesthesiology Society Journal 33:167-172. Mosely JW, Edwards VM, Casey G, Redeker AG, White E (19751: Hepatitis B virus infection in dentists. New England Journal of Medicine 293:729-734. National Advisory Committee on Immunization (1987):Statement on immunizing agents for the prevention of viral hepatitis. Canada Disease Weekly Report 13-22:99-106. Palmer-Beasley R, Lin C-C, Hwang L-Y, Chien C-S (1981):Hepatocel-
Low-Cost Hepatitis B Vaccine lular carcinoma and hepatitis B virus: A prospective study of 22,707 men in Taiwan. Lancet 2:1129-1132. Pattison CP, Maynard JE, Berquist KR, Webster HM (1975):Epidemiology of hepatitis B in hospital personnel. American Journal of Epidemiology 101:59-64. Pennie RA, Miller A (1987): Evaluation of hepatitis B vaccine given intradermally to paediatric house staff. Clinical and Investigative Medicine 10:91. Pennie RA, O’Connor AM, Garvock MJ, Drake ER (1991): A study of the factors influencing the acceptance of hepatitis B vaccine by students in health disciplines in Ottawa. Canadian Journal of Public Health 82:12-15. Smith JL, Maynard J E , Berquist KR, Doto IL, Webster HM, Sheller
53 MJ (1976):Comparative risk of hepatitis B among physicians and dentists. Journal of Infectious Diseases 133:705-706. Szmuness W, Stevens CE, Harley EJ, Zang EA, Oleszko WR, William DC, Sadovsky R, Morrison JM, Kellner A (1980):Hepatitis B vaccine: Demonstration of efficacy in a controlled clinical trial in a high-risk population in the United States. New England Journal of Medicine 303:833-841. Szmuness W, Stevens CE, Harley E J , Zang EA, Alter HJ, Taylor PE, DeVera A, Chen GTS, Kellner A (1982): Hepatitis B vaccine in medical staff of hemodialysis units. Efficacy and subtype crossprotection. New England Journal of Medicine 307:1481-1486. United States Immunization Practices Advisory Committee (1985): Recommendations of the ACIP. Morbidity and Mortality Weekly Report 34313435.
(1992)
Low-Cost Hepatitis B Vaccine Improves Uptake Among Self-payingHealth-Care Students Ross A. Pennie, Annette M. O’Connor,Corinne S. Dulberg, Anna Bottiglia, Pranlal Manga, and C. Yong Kang Department of Pediatrics, Faculty of Medicine (R.A.P.); School of Nursing, Faculty of Health Sciences (A.M.O.); Department of Epidemiology & Community Medicine, Faculty of Medicine (C.S.D.); Children’s Hospital of Eastern Ontario, Ottawa (A.B.);Faculty of Administration (P.M.); and Department of Microbiology & Immunology, Faculty of Medicine (C.Y.K.), University of Ottawa, Ottawa, Ontario, Canada.
Advisory committees recommend hepatitis B (HBV) immunization for professional and student health-care workers. However, the currently Iicensed vaccines are expensive, and previous surveys have shown that few students (14%) have been immunized in Canada. A low-cost immunization program was offered t o health-care students in order t o determine whether the effectiveness of HBV immunization could be improved by substantially reducing the vaccine cost t o recipients. The immunogenicity, side effects, and 3-dose completion rate of a low-cost Korean HBV vaccine were compared with a similar U.S.-made vaccine. A total of 922 postsecondary students enrolled in 6 health-care disciplines in Ottawa, Canada were surveyed for hepatitis-6 immunization status. Nonimmunized students were subsequently offered HBV vaccine at total cost of $15 (Canadian), randomly allocated t o receive 3 intramuscular doses of either Korean or U.S.-made plasma-derived HBV vaccine in a double-blind fashion, surveyed about side effects, and tested for hepatitis B surface antibody seroconversion. Only 12% of the 922 surveyed students had been previously immunized when vaccine was obtainable only at high cost. However, 66% of those not immunized participated in the vaccine trial and paid the $1 5 fee. Hepatitis-B surface antibody seroconversion ( 3 1 0 sample ratio units by radioimmunoassay) occurred in 291/311 (93.6%) and 299/310 (96.5%) of recipients of 3 doses of the Korean and U.S. vaccines, respectively ( P = 0.10). There were no meaningful differences in vaccine adverse effects, and 92.6% of recipients of either vaccine completed 3 doses. These results support the view that HBV immunization would be more effective (vaccine acceptance rates higher, greater proportions of persons protected) if lower cost vaccines were marketed. 8 1992 WiIey-Liss, Inc. 0 1992 WILEY-LISS. INC.
KEY WORDS: HBV, hepatitis B immunization, health-care workers, Korean hepatitis B vaccine, U.S. hepatitis B vaccine
INTRODUCTION Hepatitis B (HBV) is a viral infection of the liver transmitted by sexual activity and by inoculation with infected blood and body fluids. Among those infected as adults, 6 1 0 % become lifetime carriers of the virus [United States Immunization Practices Advisory Committee, 19851. Carriers can transmit infection to others, and themselves have a greater than 25% incidence of chronic hepatitis I United States Immunization Practices Advisory Committee, 19851, which can lead to hepatic cirrhosis and primary cancer of the liver at a rate up to 300 times that of noncarriers [Palmer-Beasley et al., 19811. In North America, HBV prevalence is significantly increased among health-care workers [Pattison et al., 1975; Dienstag et al., 1982; Chaudhary, 1983; United States Immunization Practices Advisory Committee, 1985; Malm e t al., 19861. As a consequence, the United States Immunization Practices Advisory Committee [19851and the Canadian National Advisory Committee on Immunization 119871 recommend that professional and student health-care workers receive prophylaxis with HBV vaccine. Despite these recommendations, most health-care students are not being immunized [Kwon et al., 1984; Pennie and Miller, 19871. A survey of health-care students in Ottawa, Canada confirmed the low vaccine Accepted for publication November 15, 1991. Address reprint requests to Dr. R. A. Pennie, Infectious Diseases and Microbiology, Room 3N27G, McMaster University Medical Centre, 1200 Main Street West, Hamilton, Ontario, Canada L8N 325.
Low-Cost Hepatitis B Vaccine uptake rate and suggested that cost appeared to be the major barrier to student immunization [Pennie e t al., 19911. The present study was conducted in order to determine whether the effectiveness of HBV immunization could be improved by reducing substantially the vaccine cost to recipients through the use of a low-cost Korean vaccine. The immunogenicity, side effects, and 3-dose completion rate of the plasma-derived Korean HBV vaccine were compared with a similar U.S.-made vaccine.
METHODS Student Recruitment In September 1989, students enrolled in six healthcare programs in Ottawa were approached in class by a research nurse for participation in a n experimental immunization program. Students were asked to (1)read a statement about HBV risks and vaccine recommendations; (2) complete a questionnaire eliciting current HBV immunization status, demographic characteristics, and willingness to pay for the vaccine; and (3)read a statement about a proposed HBV vaccine trial. For the price of $15 towards the cost of the vaccine, they would receive 3 doses of either a currently licensed American plasma-derived vaccine or a similarly but not identically prepared vaccine made in Korea that was currently available in Canada only for use in this study. Students were advised of the dates of immunization sessions and were asked to appear if they wished to receive the vaccine. The study was approved by the ethics committee of the University of Ottawa and the Children’s Hospital of Eastern Ontario. Vaccines The vaccines used were Heptavax-B lot M0286 (Merck Sharp & Dohme Canada, Kirkland, Quebec) and Hepavax-B lot H0655 (Korea Green Cross Corporation, Seoul, Korea). Both vaccines were plasma derived and contained 20 pglml of inactivated HBV surface antigen (HBsAg) formulated with alum adjuvant and thimerosal preservative. The retail cost of the Merck vaccine in Ottawa a t the time of the study varied from $135 to $175 (Canadian funds). The estimated retail cost of the Korean vaccine, not yet licensed in Canada, was $15. Immunization and Blood-Testing Schedule Prospective participants who attended the immunization sessions scheduled for their class were screened for the following exclusion criteria: age less than 17 years, poor general health, current immunosuppression or pregnancy, past severe reaction to any vaccine, previous doses of HBV vaccine or HBV immune globulin, and previous known positive tests for HBV. Following informed consent, participants were randomized (in blocks of 24 containing 12 of each vaccine) to receive either the Merck Sharp & Dohme vaccine (USV) or the Korea Green Cross vaccine (KRV). The
49
participants and investigators were blinded with respect to vaccine assignment. The immunization schedule was designed to fit into the &month school year in order to optimize student compliance. Within the first 2 weeks of the 1989-90 school year, each participant (1)had a preimmunization blood sample for measurement of hepatitis-B surface antigen (HBsAg), antibody to hepatitis B core antigen (anti-HBc), and antibody to HBsAg (anti-HBs); (2) was given the first dose of 1.0-ml vaccine intramuscularly in the deltoid muscle on the same day; and ( 3 )paid the $15 fee. One and 6 months following the first dose of vaccine, blood was again obtained for anti-HBs, and the second and third doses of vaccine, respectively, were administered. Six weeks following the third dose of vaccine, a blood sample was obtained for measurement of HBsAg, anti-HBc, and anti-HBs.
Vaccine Immunogenicity The HBV serology was examined using the following commercially available radioimmunoassay kits: Ausria-11, Corab, and Ausab (Abbott Laboratories). AntiHBs levels were expressed semiquantitatively as the ratio of test sample value to mean negative control value in sample ratio units (SRU). A vaccine-induced seroconversion was defined as the development of protective levels of anti-HBs, i.e., a10 SRU [Francis et al., 1982; Hadler et al., 19861in the absence of seroconversion of anti-HBc in a participant whose first blood sample was negative for HBsAg and anti-HBs. Vaccine immunogenicity was defined as the proportion of vaccineinduced seroconversions in participants who received 3 doses of vaccine and the follow-up anti-HBs test. A clinically meaningful difference in the immunogenicity of the KRV compared to the licensed USV was defined as a difference in proportions of more than 0.10. Participants were considered previously infected and not given more than 1dose of vaccine if their preimmunization blood sample tested (1)positive for HBsAg, (2) positive for anti-HBs plus anti-HBc, or ( 3 ) anti-HBs 3 1 0 SRU and anti-HBc negative. Participants whose preimmunization sample showed very low levels (1.5-9 SRU) of anti-HBs alone, or anti-HBc alone, were defined as not interpretable and were offered 3 doses of vaccine but were not included in the analysis of vaccine immunogenicity . Survey of Side Effects At the time of each dose of vaccine, participants were given thermometers and asked to record their oral temperature once daily for 5 days. They were also asked to complete a check list of the duration and severity of side effects on a standardized form that was to be returned to a class representative within 10 days. A difference of 0.25 in the reported incidence of the various symptoms following the 2 vaccines was considered clinically meaningful.
Pennie et al.
50
Participant Compliance The completion rate was defined as the proportion of participating students who completed all 3 doses of vac-
TABLE I. Characteristics of Students Receiving HeDatitis B Vaccine Vaccine administered KRVa USVb
cine.
Effectiveness of Immunization Programs The effectiveness of the immunization program measured the proportion of nonimmunized students attending the HBV information session in class who (1)decided to accept immunization, and (2) received all 3 doses of vaccine, and (3) developed protective levels of anti-HBs. The calculations of effectiveness, outlined below, did not take into account students who received fewer than 3 doses of vaccine.
Effectiveness of the Low-Cost Program Using KRV Vaccine Worst-case (observed) scenario: [proportion of students who participated in the immunization trial after being approached in class] x [KRV immunogenicity] x [KRV completion rate]. Best-case (theoretical) scenario: [proportion of students who said they would pay $15 for licensed vaccine] x [KRV immunogenicity] x [KRV completion rate].
Effectiveness of the High-Cost Program Using USV Vaccine (Theoretical) [Proportion of students who said they would pay the full price for vaccine] x [USV immunogenicityl x [USV completion rate]. The chi square test was applied to the immunogenicity and side-effect data to test for differences between the vaccine groups.
RESULTS Student Recruitment The questionnaire was administered to 922 students in the 6 health-care programs listed in Table I. Class attendance was 75% or greater on the test day, and over 95% of the students present in class completed the questionnaire. Of these, 12% of first year and 13% of upperclass students reported that they had already received HBV immunization, 33% stated that they would pay $175 for the vaccine, and 96% said they would accept the vaccine if it cost $15. When the vaccine trial began, 5351811 (66%) of the nonimmunized, surveyed students participated, a s did a further 1581419 (37.7%) of the other health-care students on campus, including those whose classes had not been surveyed but had heard of the study by word of mouth. Table I illustrates the characteristics of the 693 participating students. The groups were similar with regard to demographic, educational, and clinical characteristics.
Vaccine Immunogenicity Table I1 illustrates the immunogenicity of the 2 vaccines. In the 621 participants who started with negative
Total number students enrolled Sex: Males Age: Range (years) Median (years) Program of study Ambulance & Emergency Dental Assistant Medicine Medical Technology Nursing Rehabilitation Therapy Race White Black Oriental West Asian HBV Tests a t Enrollment Carriers of HBsAg Anti-HBs 210 SRU Dropped Out before 3rd doseC Completed 3 doses of vaccine
349 23.5%
344 25.6%
17-47 21
17-48 21
3.7% 5.4% 18.3% 8.7% 55.3% 8.6%
17.4% 9.5% 57.0% 8.2%
90.8% 3.2% 2.3% 3.7%
85.2% 6.1% 3.2% 5.5%
1.1%
0.6% 5.4% 92.8%
3.5% 4.4%
0.9% 2.3% 4.4%
92.4%
aKorea Green Cross. bMerck Sharp & Dohme. CExcluding students HBsAg positive or anti-HBs 210 SRU a t enrollment.
TABLE 11. Anti-HBs Response to Two Hepatitis B Vaccines in Evaluable ReciDientsa Vaccine administered KRVb USVc No. of evaluable recipientsa 311 310 Vaccine immunogenicitya 93.6% 96.5% P = 0.10 (Proportion who developed anti-HBs 210 SRUd) Distribution of Anti-HBs valuesa (SRUd) 500 80.4% 84.9% aExcluding students positive for HBsAg, anti-HBs, or anti-HBc a t enro!lment and including only those who received 3 doses of vaccine. bKorea Green Cross. CMerck Sharp & Dohme. dSRU: sample ratio units (sample value/mean negative control) measured 6 weeks after 3rd dose of vaccine. P values were determined by chi square.
blood tests for HBV serology and completed the 3-dose program and follow-up HBV serology, the rates of seroprotection (anti-HBs a10 SRU) were 93.6% for KRV and 96.5% for USV (P = 0.10). There was no significant difference in the distribution of peak anti-HBs values between the vaccine groups (x2 = 3.31, D.F. = 3, P = 0.35). The sample size was sufficiently large to ensure
Low-Cost Hepatitis B Vaccine
51
TABLE 111. Adverse Effects Recorded Within 5 Days of Receiving Hepatitis B Vaccine by Students With Negative Hepatitis B Serology at Time of Enrollmenta Vaccine doses administered KRVb usvc n = 957 n = 961 Recipients reported Sore arm after injection 70.6% 64.9% P c 0.01 Median duration sore arm 1 day 1 day Temperature 238°C 1.3% 2.6% P = 0.03 Malaise 11.1% 10.4% Median-duration malaise 1 day 1 day Excessive tiredness 13.2% 13.3% Median duration tiredness 2 days 2 days 1.3% 2.0% Missed school or work Other minor symptoms 7.6% 8.6% a1918 doses of vaccine given to 659 students. bKorea Green Cross. CMerck Sharp & Dohme. P values were determined by chi square.
with 80%power and 95%confidence that the proportion of participants seroconverting after KRV was within 6%of the proportion seroconverting after USV. Vaccine Side Effects There were no serious adverse effects of immunization, and none of the 621 blood samples taken after the third dose of vaccine was consistent with acute infection with live HBV. The incidence of reported side effects did not differ for the first, second, or third doses of either vaccine; therefore, all three doses were analyzed together, as illustrated in Table 111. The 17 students whose first blood tests indicated previous HBV infection were analyzed as a separate group: 41.2% had a sore arm, 11.8%had malaise, none had fever, and none missed school. These results do not differ from those of students who tested negative for HBV markers when given the first dose of either vaccine: 58.4%sore arm, 11.5%malaise, 2.1% fever, and 1.2%missed school.
effectiveness using the KRV in the theoretical bestcase scenario was 83.4%:[96%said they would pay the low price] x [93.6%immunogenicityl x [92.8%completion]. The theoretical effectiveness of a n immunization program using the high-cost USV was only 29.4%:[33% said they would pay the high pricel X [96.5%immunogenicityl x 192.4%completion]. Cost Effectiveness of the Vaccines Given the statistically equivalent results with respect to immunogenicity, side effects, and completion rate, the 2 vaccines were virtually identical when used within the blinded immunization trial. During this immunization program, the unit cost of capital, administration, and labor was identical for the 2 vaccines, but the ingredient cost of the KRV was significantly lower than the USV. Therefore, the KRV would be more cost effective.
DISCUSSION The two vaccines studied in this trial were shown to be equally safe and immunogenic, with hepatitis-B surface antibody seroprotection rates comparable to those reported by others [Szmuness et al., 1980, 1982; Krugman et al., 1981; Francis et al., 1982; Dienstag et al., 19841.The low-cost vaccine program resulted in greatly increased student uptake of HBV vaccine. Therefore, a vaccine such a s the KRV, marketed a t low cost, would be more effective than the currently licensed more costly vaccines. Even when HBV vaccine was offered in the context of a trial involving the use of a foreign, plasma-derived Effectiveness of Immunization P r o g r a m s vaccine, students accepted it in large numbers. This The observed effectiveness of the immunization pro- suggests that the current perception by the medical gram (proportion of students offered vaccine who sero- community th a t the public will not accept plasmaconverted) using the low-cost KRV in the context of the derived vaccines, which has led to the recent withvaccine trial was 57.3%:[66%participation] x [93.6% drawal from the market by Merck of its plasma-derived immunogenicity] x [92.8%completion]. The program HBV vaccine, may not be entirely accurate. Theoretical
Participant Compliance The proportion of participants completing 3 doses of vaccine was 92.8%for KRV and 92.4%for USV. Fiftyone participants did not complete 3 doses: 7 were found to be positive for HBsAg; 10 were positive for anti-HBc and anti-HBs, indicating previously resolved HBV infection; 16 left school; 6 became pregnant; 5 were fearful of further injections; 4 lost interest; 1 was in hospital for unrelated reasons; and 2 had suspected but unproven allergic reaction to the vaccine (1 KRV, 1 USV).
Pennie et al.
52
factors, such as worries about getting AIDS from plasma-derived vaccine, fear of needles, and the inconvenience of a 3-dose regimen become much less important impediments to HBV immunization when the cost is made affordable. A limitation of this study is t h a t it did not directly compare a low-cost HBV immunization program with a high-cost program. The increased HBV immunization uptake that occurred as a result of this study may not be due only to the low cost. Other factors, such as convenience and peer enthusiasm, may have encouraged enrollment. However, the current study and the survey carried out the previous year [Pennie et al., 19911demonstrated that, despite their awareness of the recommendations regarding HBV immunization, students a t all levels had low immunization uptake when only the costly vaccines were available. This was likely a n accurate reflection of the true immunization rate. HBV immunization is not easily forgotten because the vaccine is expensive and not given jointly with routine immunizations. The 3-dose completion rate of 92.6% reflected the high motivation of health-care students and justified the implementation of a HBV immunization program for them. Although the proportion of sore arms after injection was high, and up to 11% of students said they felt unwell for a day or 2, the number of missed school days was small. The aim of a n efficient immunization program is to induce immunity against HBV in as many students as possible without subjecting many students to unnecessary tests or giving vaccine to many students who cannot benefit from i t because they are HBV carriers or already immune. As a result of this study design, in which HBV test results were not available before the administration of the first dose of vaccine, 2.4% of recipients did not benefit from the vaccine because they had been infected previously. In populations such as the one studied here, with a low prevalence of previous HBV infection and a high seroconversion rate, a program that immunizes all participants without pre- or postimmunization blood work would appear to be warranted because i t minimizes cost and inconvenience while protecting over 90% of those who receive 3 doses. This is especially the case if a low-cost vaccine is used t h a t minimizes the cost of the small number (2.4%) of unnecessary doses. Furthermore, this study confirmed the observation [Dienstag et al., 19821 that giving vaccine to people who have already had HBV infection was not associated with a n increase in reported side effects. Seven of the students who enrolled in the study were found to be HBsAg seropositive, yielding a n HBsAgseroprevalence rate in this population of 1%. This figure was higher than the previously reported rates of 0.1-0.3% in U.S. blood donors and Canadian armed forces personnel [Mosely et al., 1975; Smith et al., 1976; Chaudhary e t al., 19831. Notably, 71% of the seropositive students had recently immigrated to Canada from areas including Subsaharan Africa and Southeast Asia, where hepatitis B is highly endemic. The influence of
HBsAg-seropositive immigrants on the incidence of new cases of hepatitis B in a low-endemicity host country such as Canada warrants further study. Although this study only examined the immunization behavior of Canadian health-care students in a context of self-pay, our results should be applicable to individuals employed in informal settings (doctors’ offices, foster care, private nursing homes, and overseas voluntary service) where employees must self-pay for HBV vaccines currently marketed in industrialized countries at high cost. The observed increased effectiveness of the $15 vaccine program suggests that if a lowcost HBV vaccine were readily available, the vaccine uptake rate might increase substantially. Therefore, it would be prudent to market low-cost HBV vaccines in order that HBV immunization be within the reach of all those who would benefit from it. Inexpensive vaccines may one day allow HBV immunization to be universal.
ACKNOWLEDGMENTS This study was supported by research grant #02503 of the Ministry of Health of the Province of Ontario, Canada. The authors would like to thank Dr. Peter Rowe and Dr. David Scheifele for their careful and constructive review of this manuscript. REFERENCES Chaudhary RK (1983): Prevalence of hepatitis B markers in selected health care personnel from the Canadian Forces. Canada Diseases Weekly Report 9-50:197-198. Dienstag JI, Ryan DM (1982): Occupational exposure of hepatitis B virus in hospital personnel. Infection or immunization? American Journal of Epidemiology 115:2639. Dienstag J L , Stevens CE, Bhan AK, Szmuness W (1982): Hepatitis B vaccine administered to chronic carriers of hepatitis B surface antigen. Annals of Internal Medicine 96575-579. Dienstag JL, Werner BG, Polk BF, Snydman DR, Cravan DE, Platt R, Crumpacker CS, Ouellet-Hellstrom R, Grady GF (1984): Hepatitis B vaccine in health care personnel. Safety, immunogenicity, and indicators of efficacy. Annals of Internal Medicine 101:3440. Francis DP, Hadler SC, Thompson SE, Maynard JE, Ostrow DG, Altman N, Braff EH, OMalley P, Hawkins D, Judson FN, Penley K, Nylund T, Christie G, Myers F, Moore J N , Gardner A, Doto IL, Miller JH, Reynolds GH, Murphy BL, Schable CA, Clark BT, Curran J (1982): The prevention of hepatitis B with vaccine. Report of the Centers for Disease Control multi-center efficacy trial among homosexual men. Annals of Internal Medicine 97:362-366. Hadler SC, Francis DP, Maynard J E , Thompson SE, Judson FN, Echenberg DF, Ostrow DG, O’Malley P, Penley KA, Altman NL, Braff E, Shipman GF, Coleman PJ, Mandel E J (1986):Long-term immunogenicity and efficacy of hepatitis B vaccine in homosexual men. New England Journal of Medicine 315:209-214. Krugman S, Holley HP, Davidson M, Simberkoff MS, Matsaniotis N (1981): Immunogenic effect of inactivated hepatitis B vaccine. Comparison of 20 pg and 40 pg doses. Journal of Medical Virology 8:119-121. Kwon H-J, Keenan KM, Colman H, Sundeen KM, Waite DE (1984): Compliance with a hepatitis B vaccination program. Journal of Dental Education 48:563-565. Malm DN, Mathias RG, Turnbull KW, Kettyls GD (1986): Prevalence of hepatitis B in anaesthesia personnel. Canadian Anaesthesiology Society Journal 33:167-172. Mosely JW, Edwards VM, Casey G, Redeker AG, White E (19751: Hepatitis B virus infection in dentists. New England Journal of Medicine 293:729-734. National Advisory Committee on Immunization (1987):Statement on immunizing agents for the prevention of viral hepatitis. Canada Disease Weekly Report 13-22:99-106. Palmer-Beasley R, Lin C-C, Hwang L-Y, Chien C-S (1981):Hepatocel-
Low-Cost Hepatitis B Vaccine lular carcinoma and hepatitis B virus: A prospective study of 22,707 men in Taiwan. Lancet 2:1129-1132. Pattison CP, Maynard JE, Berquist KR, Webster HM (1975):Epidemiology of hepatitis B in hospital personnel. American Journal of Epidemiology 101:59-64. Pennie RA, Miller A (1987): Evaluation of hepatitis B vaccine given intradermally to paediatric house staff. Clinical and Investigative Medicine 10:91. Pennie RA, O’Connor AM, Garvock MJ, Drake ER (1991): A study of the factors influencing the acceptance of hepatitis B vaccine by students in health disciplines in Ottawa. Canadian Journal of Public Health 82:12-15. Smith JL, Maynard J E , Berquist KR, Doto IL, Webster HM, Sheller
53 MJ (1976):Comparative risk of hepatitis B among physicians and dentists. Journal of Infectious Diseases 133:705-706. Szmuness W, Stevens CE, Harley EJ, Zang EA, Oleszko WR, William DC, Sadovsky R, Morrison JM, Kellner A (1980):Hepatitis B vaccine: Demonstration of efficacy in a controlled clinical trial in a high-risk population in the United States. New England Journal of Medicine 303:833-841. Szmuness W, Stevens CE, Harley E J , Zang EA, Alter HJ, Taylor PE, DeVera A, Chen GTS, Kellner A (1982): Hepatitis B vaccine in medical staff of hemodialysis units. Efficacy and subtype crossprotection. New England Journal of Medicine 307:1481-1486. United States Immunization Practices Advisory Committee (1985): Recommendations of the ACIP. Morbidity and Mortality Weekly Report 34313435.
