Evaluating Influenza Vaccine Efficacy and Effectiveness Studies
The efficacy (i.e., prevention of illness among vaccinated persons in controlled trials) and effectiveness (i.e., prevention of illness in vaccinated populations) of influenza vaccines depend in part on the age and immunocompetence of the vaccine recipient, the degree of similarity between the viruses in the vaccine and those in circulation (see Effectiveness of Influenza Vaccination when Circulating Influenza Virus Strains Differ from Vaccine Strains), and the outcome being measured. Influenza vaccine efficacy and effectiveness studies have used multiple possible outcome measures, including the prevention of medically attended acute respiratory illness (MAARI), prevention of laboratory-confirmed influenza virus illness, prevention of influenza or pneumonia-associated hospitalizations or deaths, or prevention of seroconversion to circulating influenza virus strains. Efficacy or effectiveness for more specific outcomes such as laboratory-confirmed influenza typically will be higher than for less specific outcomes such as MAARI because the causes of MAARI include infections with other pathogens that influenza vaccination would not be expected to prevent (100). Observational studies that compare less-specific outcomes among vaccinated populations to those among unvaccinated populations are subject to biases that are difficult to control for during analyses. For example, an observational study that determines that influenza vaccination reduces overall mortality might be biased if healthier persons in the study are more likely to be vaccinated (101,102). Randomized controlled trials that measure laboratory-confirmed influenza virus infections as the outcome are the most persuasive evidence of vaccine efficacy, but such trials cannot be conducted ethically among groups recommended to receive vaccine annually.
Influenza Vaccine Composition
Both LAIV and TIV contain strains of influenza viruses that are antigenically equivalent to the annually recommended strains: one influenza A (H3N2) virus, one influenza A (H1N1) virus, and one influenza B virus. Each year, one or more virus strains in the vaccine might be changed on the basis of global surveillance for influenza viruses and the emergence and spread of new strains. All three vaccine virus strains were changed for the recommended vaccine for the 2008—09 influenza season, compared with the 2007—08 season (see Recommendations for Using TIV and LAIV During the 2008—09 Influenza Season). Viruses for both types of currently licensed vaccines are grown in eggs. Both vaccines are administered annually to provide optimal protection against influenza virus infection (Table 1). Both TIV and LAIV are widely available in the United States. Although both types of vaccines are expected to be effective, the vaccines differ in several respects (Table 1).
Major Differences Between TIV and LAIV
During the preparation of TIV, the vaccine viruses are made noninfectious (i.e., inactivated or killed) (103). Only subvirion and purified surface antigen preparations of TIV (often referred to as “split” and subunit vaccines, respectively) are available in the United States. TIV contains killed viruses and thus cannot cause influenza. LAIV contains live, attenuated viruses that have the potential to cause mild signs or symptoms such as runny nose, nasal congestion, fever or sore throat. LAIV is administered intranasally by sprayer, whereas TIV is administered intramuscularly by injection. LAIV is licensed for use among nonpregnant persons aged 2—49 years; safety has not been established in persons with underlying medical conditions that confer a higher risk of influenza complications. TIV is licensed for use among persons aged >6 months, including those who are healthy and those with chronic medical conditions (Table 1).
Correlates of Protection after Vaccination
Immune correlates of protection against influenza infection after vaccination include serum hemagglutination inhibition antibody and neutralizing antibody (14,104). Increased levels of antibody induced by vaccination decrease the risk for illness caused by strains that are antigenically similar to those strains of the same type or subtype included in the vaccine (105—108). The majority of healthy children and adults have high titers of antibody after vaccination (106,109). Although immune correlates such as achievement of certain antibody titers after vaccination correlate well with immunity on a population level, the significance of reaching or failing to reach a certain antibody threshold is not well understood on the individual level. Other immunologic correlates of protection that might best indicate clinical protection after receipt of an intranasal vaccine such as LAIV (e.g., mucosal antibody) are more difficult to measure (103,110).
Immunogenicity, Efficacy, and Effectiveness of TIV
Children aged >6 months typically have protective levels of anti-influenza antibody against specific influenza virus strains after receiving the recommended number of doses of influenza vaccine (104,109,111—116). In most seasons, one or more vaccine antigens are changed compared to the previous season. In consecutive years when vaccine antigens change, children aged <9 years who received only 1 dose of vaccine in their first year of vaccination are less likely to have protective antibody responses when administered only a single dose during their second year of vaccination, compared with children who received 2 doses in their first year of vaccination (117—119).
When the vaccine antigens do not change from one season to the next, priming children aged 6—23 months with a single dose of vaccine in the spring followed by a dose in the fall engenders similar antibody responses compared with a regimen of 2 doses in the fall (120). However, one study conducted during a season when the vaccine antigens did not change compared with the previous season estimated 62% effectiveness against ILI for healthy children who had received only 1 dose in the previous influenza season and only 1 dose in the study season, compared with 82% for those who received 2 doses separated by >4 weeks during the study season (121).
The antibody response among children at higher risk for influenza-related complications (e.g., children with chronic medical conditions) might be lower than those typically reported among healthy children (122,123). However, antibody responses among children with asthma are similar to those of healthy children and are not substantially altered during asthma exacerbations requiring short-term prednisone treatment (124).
Vaccine effectiveness studies also have indicated that 2 doses are needed to provide adequate protection during the first season that young children are vaccinated. Among children aged <5 years who have never received influenza vaccine previously or who received only 1 dose of influenza vaccine in their first year of vaccination, vaccine effectiveness is lower compared with children who receive 2 doses in their first year of being vaccinated. Two recent, large retrospective studies of young children who had received only 1 dose of TIV in their first year of being vaccinated determined that no decrease was observed in ILI-related office visits compared with unvaccinated children (121,125). Similar results were reported in a case-control study of children aged 6—59 months (126). These results, along with the immunogenicity data indicating that antibody responses are significantly higher when young children are given 2 doses, are the basis for the recommendation that all children aged <9 years who are being vaccinated for the first time should receive 2 vaccine doses separated by at least 4 weeks.
Certain studies have demonstrated vaccine efficacy or effectiveness among children aged >6 months, although estimates have varied. In a randomized trial conducted during five influenza seasons (1985—1990) in the United States among children aged 1—15 years, annual vaccination reduced laboratory-confirmed influenza A substantially (77%—91%) (106). A limited 1-year placebo-controlled study reported vaccine efficacy against laboratory-confirmed influenza illness of 56% among healthy children aged 3—9 years and 100% among healthy children and adolescents aged 10—18 years (127). A randomized, double-blind, placebo-controlled trial conducted during two influenza seasons among children aged 6—24 months indicated that efficacy was 66% against culture-confirmed influenza illness during 1999—2000, but did not significantly reduce culture-confirmed influenza illness during 2000—2001 (128). In a nonrandomized controlled trial among children aged 2—6 years and 7—14 years who had asthma, vaccine efficacy was 54% and 78% against laboratory-confirmed influenza type A infection and 22% and 60% against laboratory-confirmed influenza type B infection, respectively. Vaccinated children aged 2—6 years with asthma did not have substantially fewer type B influenza virus infections compared with the control group in this study (129). Vaccination also might provide protection against asthma exacerbations (130); however, other studies of children with asthma have not demonstrated decreased exacerbations (131). Because of the recognized influenza-related disease burden among children with other chronic diseases or immunosuppression and the long-standing recommendation for vaccination of these children, randomized placebo-controlled studies to study efficacy in these children have not been conducted because of ethical considerations.
A retrospective study conducted among approximately 30,000 children aged 6 months—8 years during an influenza season (2003—04) with a suboptimal vaccine match indicated vaccine effectiveness of 51% against medically attended, clinically diagnosed pneumonia or influenza (i.e., no laboratory confirmation of influenza) among fully vaccinated children, and 49% among approximately 5,000 children aged 6—23 months (125). Another retrospective study of similar size conducted during the same influenza season in Denver but limited to healthy children aged 6—21 months estimated clinical effectiveness of 2 TIV doses to be 87% against pneumonia or influenza-related office visits (121). Among children, TIV effectiveness might increase with age (106,132).
TIV has been demonstrated to reduce acute otitis media in some studies. Two studies have reported that TIV decreases the risk for influenza-associated otitis media by approximately 30% among children with mean ages of 20 and 27 months, respectively (133,134). However, a large study conducted among children with a mean age of 14 months indicated that TIV was not effective against acute otitis media (128). Influenza vaccine effectiveness against acute otitis media, which is caused by a variety of pathogens and is not typically diagnosed using influenza virus culture, would be expected to be relatively low when assessing a nonspecific clinical outcome.
Adults Aged <65 Years
One dose of TIV is highly immunogenic in healthy adults aged <65 years. Limited or no increase in antibody response is reported among adults when a second dose is administered during the same season (135—139). When the vaccine and circulating viruses are antigenically similar, TIV prevents laboratory-confirmed influenza illness among approximately 70%—90% of healthy adults aged <65 years in randomized controlled trials (139—142). Vaccination of healthy adults also has resulted in decreased work absenteeism and decreased use of health-care resources, including use of antibiotics, when the vaccine and circulating viruses are well-matched (139—141,143—145). Efficacy or effectiveness against laboratory-confirmed influenza illness was 50%—77% in studies conducted during different influenza seasons when the vaccine strains were antigenically dissimilar to the majority of circulating strains (139,141,145—147). However, effectiveness among healthy adults against influenza-related hospitalization, measured in the most recent of these studies, was 90% (147).
In certain studies, persons with certain chronic diseases have lower serum antibody responses after vaccination compared with healthy young adults and can remain susceptible to influenza virus infection and influenza-related upper respiratory tract illness (148—150). Vaccine effectiveness among adults aged <65 years who are at higher risk for influenza complications is typically lower than that reported for healthy adults. In a case-control study conducted during 2003—2004, when the vaccine was a suboptimal antigenic match to many circulating virus strains, effectiveness for prevention of laboratory-confirmed influenza illness among adults aged 50—64 years with high risk conditions was 48%, compared with 60% for healthy adults (147). Effectiveness against hospitalization among adults aged 50—64 years with high-risk conditions was 36%, compared with 90% effectiveness among healthy adults in that age range (147). A randomized controlled trial among adults in Thailand with chronic obstructive pulmonary disease (median age: 68 years) indicated a vaccine effectiveness of 76% in preventing laboratory-confirmed influenza during a season when viruses were well-matched to vaccine viruses. Effectiveness did not decrease with increasing severity of underlying lung disease (151).
Studies using less specific outcomes, without laboratory confirmation of influenza virus infection, typically have demonstrated substantial reductions in hospitalizations or deaths among adults with risk factors for influenza complications. In a case-control study conducted in Denmark among adults with underlying medical conditions aged <65 years during 1999—2000, vaccination reduced deaths attributable to any cause 78% and reduced hospitalizations attributable to respiratory infections or cardiopulmonary diseases 87% (152). A benefit was reported after the first vaccination and increased with subsequent vaccinations in subsequent years (153). Among patients with diabetes mellitus, vaccination was associated with a 56% reduction in any complication, a 54% reduction in hospitalizations, and a 58% reduction in deaths (154). Certain experts have noted that the substantial effects on morbidity and mortality among those who received influenza vaccination in these observational studies should be interpreted with caution because of the difficulties in ensuring that those who received vaccination had similar baseline health status as those who did not (101,102). One meta-analysis of published studies did not determine sufficient evidence to conclude that persons with asthma benefit from vaccination (155). However, a meta-analysis that examined effectiveness among persons with chronic obstructive pulmonary disease identified evidence of benefit from vaccination (156).
TIV produces adequate antibody concentrations against influenza among vaccinated HIV-infected persons who have minimal AIDS-related symptoms and normal or near-normal CD4+ T-lymphocyte cell counts (157—159). Among persons who have advanced HIV disease and low CD4+ T-lymphocyte cell counts, TIV might not induce protective antibody titers (159,160); a second dose of vaccine does not improve the immune response in these persons (160,161). A randomized, placebo-controlled trial determined that TIV was highly effective in preventing symptomatic, laboratory-confirmed influenza virus infection among HIV-infected persons with a mean of 400 CD4+ T-lymphocyte cells/mm3; however, a limited number of persons with CD4+ T-lymphocyte cell counts of <200 were included in that study (161). A nonrandomized study of HIV-infected persons determined that influenza vaccination was most effective among persons with >100 CD4+ cells and among those with <30,000 viral copies of HIV type-1/mL (77).
On the basis of certain small studies, immunogenicity for persons with solid organ transplants varies according to transplant type. Among persons with kidney or heart transplants, the proportion who developed seroprotective antibody concentrations was similar or slightly reduced compared with healthy persons (162—164). However, a study among persons with liver transplants indicated reduced immunologic responses to influenza vaccination (165—167), especially if vaccination occurred within the 4 months after the transplant procedure (165).
Pregnant Women and Neonates
Pregnant women have protective levels of anti-influenza antibodies after vaccination (168,169). Passive transfer of anti-influenza antibodies that might provide protection from vaccinated women to neonates has been reported (168,170—172). A retrospective, clinic-based study conducted during 1998—2003 documented a nonsignificant trend towards fewer episodes of MAARI during one influenza season among vaccinated pregnant women compared with unvaccinated pregnant women and substantially fewer episodes of MAARI during the peak influenza season (169). However, a retrospective study conducted during 1997—2002 that used clinical records data did not indicate a reduction in ILI among vaccinated pregnant women or their infants (173). In another study conducted during 1995—2001, medical visits for respiratory illness among the infants were not substantially reduced (174). However, studies of influenza vaccine effectiveness among pregnant women have not included specific outcomes such as laboratory-confirmed influenza in women or their infants.
Adults aged >65 years typically have a diminished immune response to influenza vaccination compared with young healthy adults, suggesting that immunity might be of shorter duration (although still extending through one influenza season) (175,176). However, a review of the published literature concluded that no clear evidence existed that immunity declined more rapidly in the elderly (177). Infections among the vaccinated elderly might be associated with an age-related reduction in ability to respond to vaccination rather than reduced duration of immunity (149—150).
The only randomized controlled trial among community-dwelling persons aged >60 years reported a vaccine efficacy of 58% against influenza respiratory illness during a season when the vaccine strains were considered to be well-matched to circulating strains, but indicated that efficacy was lower among those aged >70 years (178). Influenza vaccine effectiveness in preventing MAARI among the elderly in nursing homes has been estimated at 20%—40% (179,180), and reported outbreaks among well-vaccinated nursing home populations have suggested that vaccination might not have any significant effectiveness when circulating strains are drifted from vaccine strains (181,182). In contrast, some studies have indicated that vaccination can be up to 80% effective in preventing influenza-related death (179,183—185). Among elderly persons not living in nursing homes or similar chronic-care facilities, influenza vaccine is 27%—70% effective in preventing hospitalization for pneumonia and influenza (186—188). Influenza vaccination reduces the frequency of secondary complications and reduces the risk for influenza-related hospitalization and death among community-dwelling adults aged >65 years with and without high-risk medical conditions (e.g., heart disease and diabetes) (187—192). However, studies demonstrating large reductions in hospitalizations and deaths among the vaccinated elderly have been conducted using medical record databases and have not measured reductions in laboratory-confirmed influenza illness. These studies have been challenged because of concerns that they have not adequately controlled for differences in the propensity for healthier persons to be more likely than less healthy persons to receive vaccination (101,102,183,193—195).
TIV Dosage, Administration, and Storage
The composition of TIV varies according to manufacturer, and package inserts should be consulted. TIV formulations in multidose vials contain the vaccine preservative thimerosal; preservative-free single dose preparations also are available. TIV should be stored at 35°F—46°F (2°C—8°C) and should not be frozen. TIV that has been frozen should be discarded. Dosage recommendations and schedules vary according to age group (Table 2). Vaccine prepared for a previous influenza season should not be administered to provide protection for any subsequent season.
The intramuscular route is recommended for TIV. Adults and older children should be vaccinated in the deltoid muscle. A needle length of >1 inch (>25 mm) should be considered for persons in these age groups because needles of <1 inch might be of insufficient length to penetrate muscle tissue in certain adults and older children (196). When injecting into the deltoid muscle among children with adequate deltoid muscle mass, a needle length of 7/8—1.25 inches is recommended (197).
Infants and young children should be vaccinated in the anterolateral aspect of the thigh. A needle length of 7/8—1 inch should be used for children aged <12 months.
Adverse Events after Receipt of TIV
Studies support the safety of annual TIV in children and adolescents. The largest published postlicensure population-based study assessed TIV safety in 215,600 children aged <18 years and 8,476 children aged 6—23 months enrolled in one of five health maintenance organizations (HMOs) during 1993—1999. This study indicated no increase in biologically plausible, medically attended events during the 2 weeks after inactivated influenza vaccination, compared with control periods 3—4 weeks before and after vaccination (198). A retrospective study using medical records data from approximately 45,000 children aged 6—23 months provided additional evidence supporting overall safety of TIV in this age group. Vaccination was not associated with statistically significant increases in any medically attended outcome, and 13 diagnoses, including acute upper respiratory illness, otitis media and asthma, were significantly less common (199).
In a study of 791 healthy children aged 1—15 years, postvaccination fever was noted among 11.5% of those aged 1—5 years, 4.6% among those aged 6—10 years, and 5.1% among those aged 11—15 years (106). Fever, malaise, myalgia, and other systemic symptoms that can occur after vaccination with inactivated vaccine most often affect persons who have had no previous exposure to the influenza virus antigens in the vaccine (e.g., young children) (200,201). These reactions begin 6—12 hours after vaccination and can persist for 1—2 days. Data about potential adverse events among children after influenza vaccination are available from the Vaccine Adverse Event Reporting System (VAERS). A recently published review of VAERS reports submitted after administration of TIV to children aged 6—23 months documented that the most frequently reported adverse events were fever, rash, injection-site reactions, and seizures; the majority of the limited number of reported seizures appeared to be febrile (202). Because of the limitations of passive reporting systems, determining causality for specific types of adverse events, with the exception of injection-site reactions, usually is not possible using VAERS data alone.
In placebo-controlled studies among adults, the most frequent side effect of vaccination was soreness at the vaccination site (affecting 10%—64% of patients) that lasted <2 days (203,204). These local reactions typically were mild and rarely interfered with the recipients’ ability to conduct usual daily activities. Placebo-controlled trials demonstrate that among older persons and healthy young adults, administration of TIV is not associated with higher rates for systemic symptoms (e.g., fever, malaise, myalgia, and headache) when compared with placebo injections (139,155, 203—205).
Pregnant Women and Neonates
FDA has classified TIV as a “Pregnancy Category C” medication, indicating that animal reproduction studies have not been conducted to support a labeling change. Available data indicate that influenza vaccine does not cause fetal harm when administered to a pregnant woman or affect reproductive capacity. One study of approximately 2,000 pregnant women who received TIV during pregnancy demonstrated no adverse fetal effects and no adverse effects during infancy or early childhood (206). A matched case-control study of 252 pregnant women who received TIV within the 6 months before delivery determined no adverse events after vaccination among pregnant women and no difference in pregnancy outcomes compared with 826 pregnant women who were not vaccinated (169). During 2000—2003, an estimated 2 million pregnant women were vaccinated, and only 20 adverse events among women who received TIV were reported to VAERS during this time, including nine injection-site reactions and eight systemic reactions (e.g., fever, headache, and myalgias). In addition, three miscarriages were reported, but these were not known to be causally related to vaccination (207). Similar results have been reported in certain smaller studies (168,170,208), and a recent international review of data on the safety of TIV concluded that no evidence exists to suggest harm to the fetus (209).
Persons with Chronic Medical Conditions
In a randomized cross-over study of children and adults with asthma, no increase in asthma exacerbations was reported for either age group (210), and a second study indicated no increase in wheezing among vaccinated asthmatic children (130). One study (123) reported that 20%—28% of children with asthma aged 9 months—18 years had local pain and swelling at the site of influenza vaccination, and another study (113) reported that 23% of children aged 6 months—4 years with chronic heart or lung disease had local reactions. A blinded, randomized, cross-over study of 1,952 adults and children with asthma demonstrated that only self-reported “body aches” were reported more frequently after TIV (25%) than placebo-injection (21%) (210). However, a placebo-controlled trial of TIV indicated no difference in local reactions among 53 children aged 6 months—6 years with high-risk medical conditions or among 305 healthy children aged 3—12 years (114).
Among children with high-risk medical conditions, one study of 52 children aged 6 months—3 years reported fever among 27% and irritability and insomnia among 25% (113); and a study among 33 children aged 6—18 months reported that one child had irritability and one had a fever and seizure after vaccination (211). No placebo comparison group was used in these studies.
Data demonstrating safety of TIV for HIV-infected persons are limited, but no evidence exists that vaccination has a clinically important impact on HIV infection or immunocompetence. One study demonstrated a transient (i.e., 2—4 week) increase in HIV RNA (ribonucleic acid) levels in one HIV-infected person after influenza virus infection (212). Studies have demonstrated a transient increase in replication of HIV-1 in the plasma or peripheral blood mononuclear cells of HIV-infected persons after vaccine administration (159,213). However, more recent and better-designed studies have not documented a substantial increase in the replication of HIV (214—217). CD4+ T-lymphocyte cell counts or progression of HIV disease have not been demonstrated to change substantially after influenza vaccination among HIV-infected persons compared with unvaccinated HIV-infected persons (159,218). Limited information is available about the effect of antiretroviral therapy on increases in HIV RNA levels after either natural influenza virus infection or influenza vaccination (73,219).
Data are similarly limited for persons with other immunocompromising conditions. In small studies, vaccination did not affect allograft function or cause rejection episodes in recipients of kidney transplants (162,164), heart transplants (163), or liver transplants (165).
Immediate and presumably allergic reactions (e.g., hives, angioedema, allergic asthma, and systemic anaphylaxis) occur rarely after influenza vaccination (220,221). These reactions probably result from hypersensitivity to certain vaccine components; the majority of reactions probably are caused by residual egg protein. Although influenza vaccines contain only a limited quantity of egg protein, this protein can induce immediate hypersensitivity reactions among persons who have severe egg allergy. Manufacturers use a variety of different compounds to inactivate influenza viruses and add antibiotics to prevent bacterial contamination. Package inserts should be consulted for additional information.
Persons who have had hives or swelling of the lips or tongue, or who have experienced acute respiratory distress or who collapse after eating eggs, should consult a physician for appropriate evaluation to help determine if vaccine should be administered. Persons who have documented immunoglobulin E (IgE)-mediated hypersensitivity to eggs, including those who have had occupational asthma related to egg exposure or other allergic responses to egg protein, also might be at increased risk for allergic reactions to influenza vaccine, and consultation with a physician before vaccination should be considered (222—224).
Hypersensitivity reactions to other vaccine components can occur but are rare. Although exposure to vaccines containing thimerosal can lead to hypersensitivity, the majority of patients do not have reactions to thimerosal when it is administered as a component of vaccines, even when patch or intradermal tests for thimerosal indicate hypersensitivity (225,226). When reported, hypersensitivity to thimerosal typically has consisted of local delayed hypersensitivity reactions (225).
Guillain-Barré Syndrome and TIV
The annual incidence of Guillain-Barré Syndrome (GBS) is 10—20 cases per 1 million adults (227). Substantial evidence exists that multiple infectious illnesses, most notably Campylobacter jejuni gastrointestinal infections and upper respiratory tract infections, are associated with GBS (228—230). The 1976 swine influenza vaccine was associated with an increased frequency of GBS (231,232), estimated at one additional case of GBS per 100,000 persons vaccinated. The risk for influenza vaccine-associated GBS was higher among persons aged >25 years than among persons aged <25 years (233). However, obtaining strong epidemiologic evidence for a possible small increase in risk for a rare condition with multiple causes is difficult, and no evidence exists for a consistent causal relation between subsequent vaccines prepared from other influenza viruses and GBS.
None of the studies conducted using influenza vaccines other than the 1976 swine influenza vaccine have demonstrated a substantial increase in GBS associated with influenza vaccines. During three of four influenza seasons studied during 1977—1991, the overall relative risk estimates for GBS after influenza vaccination were not statistically significant in any of these studies (234—236). However, in a study of the 1992—93 and 1993—94 seasons, the overall relative risk for GBS was 1.7 (CI = 1.0—2.8; p = 0.04) during the 6 weeks after vaccination, representing approximately one additional case of GBS per 1 million persons vaccinated; the combined number of GBS cases peaked 2 weeks after vaccination (231). Results of a study that examined health-care data from Ontario, Canada, during 1992—2004 demonstrated a small but statistically significant temporal association between receiving influenza vaccination and subsequent hospital admission for GBS. However, no increase in cases of GBS at the population level was reported after introduction of a mass public influenza vaccination program in Ontario beginning in 2000 (237). Data from VAERS have documented decreased reporting of GBS occurring after vaccination across age groups over time, despite overall increased reporting of other, non-GBS conditions occurring after administration of influenza vaccine (203). Cases of GBS after influenza virus infection have been reported, but no other epidemiologic studies have documented such an association (238,239). Recently published data from the United Kingdom’s General Practice Research Database (GPRD) found influenza vaccine to be protective against GBS, although it is unclear if this was associated with protection against influenza or confounding because of a “healthy vaccinee” (e.g., healthier persons might be more likely to be vaccinated and are lower risk for GBS) (240). A separate GPRD analysis found no association between vaccination and GBS over a 9 year period; only three cases of GBS occurred within 6 weeks after influenza vaccine (241).
If GBS is a side effect of influenza vaccines other than 1976 swine influenza vaccine, the estimated risk for GBS (on the basis of the few studies that have demonstrated an association between vaccination and GBS) is low (i.e., approximately one additional case per 1 million persons vaccinated). The potential benefits of influenza vaccination in preventing serious illness, hospitalization, and death substantially outweigh these estimates of risk for vaccine-associated GBS. No evidence indicates that the case fatality ratio for GBS differs among vaccinated persons and those not vaccinated.
Use of TIV among Patients with a History of GBS
The incidence of GBS among the general population is low, but persons with a history of GBS have a substantially greater likelihood of subsequently experiencing GBS than persons without such a history (227). Thus, the likelihood of coincidentally experiencing GBS after influenza vaccination is expected to be greater among persons with a history of GBS than among persons with no history of this syndrome. Whether influenza vaccination specifically might increase the risk for recurrence of GBS is unknown. However, avoiding vaccinating persons who are not at high risk for severe influenza complications and who are known to have experienced GBS within 6 weeks after a previous influenza vaccination might be prudent as a precaution. As an alternative, physicians might consider using influenza antiviral chemoprophylaxis for these persons. Although data are limited, the established benefits of influenza vaccination might outweigh the risks for many persons who have a history of GBS and who are also at high risk for severe complications from influenza.
Vaccine Preservative (Thimerosal) in Multidose Vials of TIV
Thimerosal, a mercury-containing anti-bacterial compound, has been used as a preservative in vaccines since the 1930s (242) and is used in multidose vial preparations of TIV to reduce the likelihood of bacterial contamination. No scientific evidence indicates that thimerosal in vaccines, including influenza vaccines, is a cause of adverse events other than occasion local hypersensitivity reactions in vaccine recipients. In addition, no scientific evidence exists that thimerosal-containing vaccines are a cause of adverse events among children born to women who received vaccine during pregnancy. Evidence is accumulating that supports the absence of substantial risk for neurodevelopment disorders or other harm resulting from exposure to thimerosal-containing vaccines (243—250). However, continuing public concern about exposure to mercury in vaccines was viewed as a potential barrier to achieving higher vaccine coverage levels and reducing the burden of vaccine-preventable diseases. Therefore, the U.S. Public Health Service and other organizations recommended that efforts be made to eliminate or reduce the thimerosal content in vaccines as part of a strategy to reduce mercury exposures from all sources (243,245,247). Since mid-2001, vaccines routinely recommended for infants aged <6 months in the United States have been manufactured either without or with greatly reduced (trace) amounts of thimerosal. As a result, a substantial reduction in the total mercury exposure from vaccines for infants and children already has been achieved (197). ACIP and other federal agencies and professional medical organizations continue to support efforts to provide thimerosal preservative—free vaccine options.
The benefits of influenza vaccination for all recommended groups, including pregnant women and young children, outweigh concerns on the basis of a theoretical risk from thimerosal exposure through vaccination. The risks for severe illness from influenza virus infection are elevated among both young children and pregnant women, and vaccination has been demonstrated to reduce the risk for severe influenza illness and subsequent medical complications. In contrast, no scientifically conclusive evidence has demonstrated harm from exposure to vaccine containing thimerosal preservative. For these reasons, persons recommended to receive TIV may receive any age- and risk factor—appropriate vaccine preparation, depending on availability. An analysis of VAERS reports found no difference in the safety profile of preservative-containing compared with preservative-free TIV vaccines in infants aged 6—23 months (202).
Nonetheless, certain states have enacted legislation banning the administration of vaccines containing mercury; the provisions defining mercury content vary (251). LAIV and many of the single dose vial or syringe preparations of TIV are thimerosal-free, and the number of influenza vaccine doses that do not contain thimerosal as a preservative is expected to increase (Table 2). However, these laws might present a barrier to vaccination unless influenza vaccines that do not contain thimerosal as a preservative are easily available in those states.
The U.S. vaccine supply for infants and pregnant women is in a period of transition during which the availability of thimerosal-reduced or thimerosal-free vaccine intended for these groups is being expanded by manufacturers as a feasible means of further reducing an infant’s cumulative exposure to mercury. Other environmental sources of mercury exposure are more difficult or impossible to avoid or eliminate (243).
LAIV Dosage, Administration, and Storage
Each dose of LAIV contains the same three vaccine antigens used in TIV. However, the antigens are constituted as live, attenuated, cold-adapted, temperature-sensitive vaccine viruses. Additional components of LAIV include egg allantoic fluid, monosodium glutamate, sucrose, phosphate, and glutamate buffer; and hydrolyzed porcine gelatin. LAIV does not contain thimerosal. LAIV is made from attenuated viruses that are only able to replicate efficiently at temperatures present in the nasal mucosa. LAIV does not cause systemic symptoms of influenza in vaccine recipients, although a minority of recipients experience nasal congestion, which is probably a result of either effects of intranasal vaccine administration or local viral replication or fever (252).
LAIV is intended for intranasal administration only and should not be administered by the intramuscular, intradermal, or intravenous route. LAIV is not licensed for vaccination of children aged <2 years or adults aged >49 years. LAIV is supplied in a prefilled, single-use sprayer containing 0.2 mL of vaccine. Approximately 0.1 mL (i.e., half of the total sprayer contents) is sprayed into the first nostril while the recipient is in the upright position. An attached dose-divider clip is removed from the sprayer to administer the second half of the dose into the other nostril. LAIV is shipped to end users at 35°F—46°F (2°C—8°C). LAIV should be stored at 35°F—46°F (2°C—8°C) on receipt and can remain at that temperature until the expiration date is reached (252). Vaccine prepared for a previous influenza season should not be administered to provide protection for any subsequent season.
Shedding, Transmission, and Stability of Vaccine Viruses
Available data indicate that both children and adults vaccinated with LAIV can shed vaccine viruses after vaccination, although in lower amounts than occur typically with shedding of wild-type influenza viruses. In rare instances, shed vaccine viruses can be transmitted from vaccine recipients to unvaccinated persons. However, serious illnesses have not been reported among unvaccinated persons who have been infected inadvertently with vaccine viruses.
One study of children aged 8—36 months in a child care center assessed transmissibility of vaccine viruses from 98 vaccinated to 99 unvaccinated subjects; 80% of vaccine recipients shed one or more virus strains (mean duration: 7.6 days). One influenza type B vaccine strain isolate was recovered from a placebo recipient and was confirmed to be vaccine-type virus. The type B isolate retained the cold-adapted, temperature-sensitive, attenuated phenotype, and it possessed the same genetic sequence as a virus shed from a vaccine recipient who was in the same play group. The placebo recipient from whom the influenza type B vaccine strain was isolated had symptoms of a mild upper respiratory illness but did not experience any serious clinical events. The estimated probability of acquiring vaccine virus after close contact with a single LAIV recipient in this child care population was 0.6%—2.4% (253).
Studies assessing whether vaccine viruses are shed have been based on viral cultures or PCR detection of vaccine viruses in nasal aspirates from persons who have received LAIV. One study of 20 healthy vaccinated adults aged 18—49 years demonstrated that the majority of shedding occurred within the first 3 days after vaccination, although the vaccine virus was detected in one subject on day 7 after vaccine receipt. Duration or type of symptoms associated with receipt of LAIV did not correlate with detection of vaccine viruses in nasal aspirates (254). Another study in 14 healthy adults aged 18—49 years indicated that 50% of these adults had viral antigen detected by direct immunofluorescence or rapid antigen tests within 7 days of vaccination. The majority of samples with detectable virus were collected on day 2 or 3 (255). Vaccine strain virus was detected from nasal secretions in one (2%) of 57 HIV-infected adults who received LAIV, none of 54 HIV-negative participants (256), and three (13%) of 23 HIV-infected children compared with seven (28%) of 25 children who were not HIV-infected (257). No participants in these studies had detectable virus beyond 10 days after receipt of LAIV. The possibility of person-to-person transmission of vaccine viruses was not assessed in these studies (254—257).
In clinical trials, viruses isolated from vaccine recipients have been phenotypically stable. In one study, nasal and throat swab specimens were collected from 17 study participants for 2 weeks after vaccine receipt (258). Virus isolates were analyzed by multiple genetic techniques. All isolates retained the LAIV genotype after replication in the human host, and all retained the cold-adapted and temperature-sensitive phenotypes. A study conducted in a child-care setting demonstrated that limited genetic change occurred in the LAIV strains following replication in the vaccine recipients (259).
Immunogenicity, Efficacy, and Effectiveness of LAIV
LAIV virus strains replicate primarily in nasopharyngeal epithelial cells. The protective mechanisms induced by vaccination with LAIV are not understood completely but appear to involve both serum and nasal secretory antibodies. The immunogenicity of the approved LAIV has been assessed in multiple studies conducted among children and adults (106,260—266). No single laboratory measurement closely correlates with protective immunity induced by LAIV (261).
A randomized, double-blind, placebo-controlled trial among 1,602 healthy children aged 15—71 months assessed the efficacy of LAIV against culture-confirmed influenza during two seasons (267,268). This trial included a subset of children aged 60—71 months who received 2 doses in the first season. In season one (1996—97), when vaccine and circulating virus strains were well-matched, efficacy against culture-confirmed influenza was 94% for participants who received 2 doses of LAIV separated by >6 weeks, and 89% for those who received 1 dose. In season two, when the A (H3N2) component in the vaccine was not well-matched with circulating virus strains, efficacy (1 dose) was 86%, for an overall efficacy over two influenza seasons of 92%. Receipt of LAIV also resulted in 21% fewer febrile illnesses and a significant decrease in acute otitis media requiring antibiotics (267,269). Other randomized, placebo-controlled trials demonstrating the efficacy of LAIV in young children against culture-confirmed influenza include a study conducted among children aged 6—35 months attending child care centers during consecutive influenza seasons (270), in which 85%—89% efficacy was observed, and a study conducted among children aged 12—36 months living in Asia during consecutive influenza seasons, in which 64%-70% efficacy was documented (271). In one community-based, nonrandomized open-label study, reductions in MAARI were observed among children who received 1 dose of LAIV during the 1990—00 and 2000—01 influenza seasons even though antigenically drifted influenza A/H1N1 and B viruses were circulating during that season (272). LAIV efficacy in preventing laboratory confirmed influenza has also been demonstrated in studies comparing the efficacy of LAIV with TIV rather than with a placebo (see Comparisons of LAIV and TIV Efficacy or Effectiveness).
A randomized, double-blind, placebo-controlled trial of LAIV effectiveness among 4,561 healthy working adults aged 18—64 years assessed multiple endpoints, including reductions in self-reported respiratory tract illness without laboratory confirmation, work loss, health-care visits, and medication use during influenza outbreak periods (273). The study was conducted during the 1997—98 influenza season, when the vaccine and circulating A (H3N2) strains were not well-matched. The frequency of febrile illnesses was not significantly decreased among LAIV recipients compared with those who received placebo. However, vaccine recipients had significantly fewer severe febrile illnesses (19% reduction) and febrile upper respiratory tract illnesses (24% reduction), and significant reductions in days of illness, days of work lost, days with health-care—provider visits, and use of prescription antibiotics and over-the-counter medications (273). Efficacy against culture-confirmed influenza in a randomized, placebo-controlled study was 57%, although efficacy in this study was not demonstrated to be significantly greater than placebo (155).
Adverse Events after Receipt of LAIV
Healthy Children Aged 2—18 Years
In a subset of healthy children aged 60—71 months from one clinical trial (233), certain signs and symptoms were reported more often after the first dose among LAIV recipients (n = 214) than among placebo recipients (n = 95), including runny nose (48% and 44%, respectively); headache (18% and 12%, respectively); vomiting (5% and 3%, respectively); and myalgias (6% and 4%, respectively). However, these differences were not statistically significant. In other trials, signs and symptoms reported after LAIV administration have included runny nose or nasal congestion (20%—75%), headache (2%—46%), fever (0—26%), vomiting (3%—13%), abdominal pain (2%), and myalgias (0—21%) (106,260,263, 265,270,273—276). These symptoms were associated more often with the first dose and were self-limited.
In a randomized trial published in 2007, LAIV and TIV were compared among children aged 6—59 months (277). Children with medically diagnosed or treated wheezing within 42 days before enrollment, or a history of severe asthma, were excluded from this study. Among children aged 24—59 months who received LAIV, the rate of medically significant wheezing, using a pre-specified definition, was not greater compared with those who received TIV (277); wheezing was observed more frequently among younger LAIV recipients in this study (see Persons at Higher Risk from Influenza-Related Complications). In a previous randomized placebo-controlled safety trial among children aged 12 months—17 years without a history of asthma by parental report, an elevated risk for asthma events (RR = 4.06, CI = 1.29—17.86) was documented among 728 children aged 18—35 months who received LAIV. Of the 16 children with asthma-related events in this study, seven had a history of asthma on the basis of subsequent medical record review. None required hospitalization, and elevated risks for asthma were not observed in other age groups (276).
Another study was conducted among >11,000 children aged 18 months—18 years in which 18,780 doses of vaccine were administered for 4 years. For children aged 18 months—4 years, no increase was reported in asthma visits 0—15 days after vaccination compared with the prevaccination period. A significant increase in asthma events was reported 15—42 days after vaccination, but only in vaccine year 1 (278).
Initial data from VAERS during 2007—2008, following ACIP recommendation for LAIV use in children aged 2—4 years, do not suggest a concern for wheezing after LAIV in young children. However data also suggest uptake of LAIV is limited and continued safety monitoring for wheezing events after LAIV is indicated (CDC, unpublished data, 2008).
Adults Aged 19—49 Years
Among adults, runny nose or nasal congestion (28%—78%), headache (16%—44%), and sore throat (15%—27%) have been reported more often among vaccine recipients than placebo recipients (252,279). In one clinical trial among a subset of healthy adults aged 18—49 years, signs and symptoms reported more frequently among LAIV recipients (n = 2,548) than placebo recipients (n = 1,290) within 7 days after each dose included cough (14% and 11%, respectively); runny nose (45% and 27%, respectively); sore throat (28% and 17%, respectively); chills (9% and 6%, respectively); and tiredness/weakness (26% and 22%, respectively) (279).
Persons at Higher Risk for Influenza-Related Complications
Limited data assessing the safety of LAIV use for certain groups at higher risk for influenza-related complications are available. In one study of 54 HIV-infected persons aged 18—58 years and with CD4 counts >200 cells/mm3 who received LAIV, no serious adverse events were reported during a 1-month follow-up period (256). Similarly, one study demonstrated no significant difference in the frequency of adverse events or viral shedding among HIV-infected children aged 1—8 years on effective antiretroviral therapy who were administered LAIV, compared with HIV-uninfected children receiving LAIV (257). LAIV was well-tolerated among adults aged >65 years with chronic medical conditions (280). These findings suggest that persons at risk for influenza complications who have inadvertent exposure to LAIV would not have significant adverse events or prolonged viral shedding and that persons who have contact with persons at higher risk for influenza-related complications may receive LAIV.
Serious Adverse Events
Serious adverse events after administration of LAIV requiring medical attention among healthy children aged 5—17 years or healthy adults aged 18—49 years occurred at a rate of <1% (252). Surveillance will continue for adverse events, including those that might not have been detected in previous studies. Reviews of reports to VAERS after vaccination of approximately 2.5 million persons during the 2003—04 and 2004—05 influenza seasons did not indicate any new safety concerns (281). Health-care professionals should report all clinically significant adverse events occurring after LAIV administration promptly to VAERS after LAIV administration.
Comparisons of LAIV and TIV Efficacy or Effectiveness
Both TIV and LAIV have been demonstrated to be effective in children and adults, but data directly comparing the efficacy or effectiveness of these two types of influenza vaccines are limited. Studies comparing the efficacy of TIV to that of LAIV have been conducted in a variety of settings and populations using several different outcomes. One randomized, double-blind, placebo-controlled challenge study among 92 healthy adults aged 18—41 years assessed the efficacy of both LAIV and TIV in preventing influenza infection when challenged with wild-type strains that were antigenically similar to vaccine strains (282). The overall efficacy in preventing laboratory-documented influenza from all three influenza strains combined was 85% and 71%, respectively, when challenged 28 days after vaccination by viruses to which study participants were susceptible before vaccination. The difference in efficacy between the two vaccines was not statistically significant in this limited study. No additional challenges to assess efficacy at time points later than 28 days were conducted. In a randomized, double-blind, placebo-controlled trial, conducted among young adults during an influenza season when the majority of circulating H3N2 viruses were antigenically drifted from that season’s vaccine viruses, the efficacy of LAIV and TIV against culture-confirmed influenza was 57% and 77%, respectively. The difference in efficacy was not statistically significant and was based largely on a difference in efficacy against influenza B (155).
A randomized controlled clinical trial conducted among children aged 6—71 months during the 2004—05 influenza season demonstrated a 55% reduction in cases of culture-confirmed influenza among children who received LAIV compared with those who received TIV (277). In this study, LAIV efficacy was higher compared with TIV against antigenically drifted viruses as well as well-matched viruses (277). An open-label, nonrandomized, community-based influenza vaccine trial conducted during an influenza season when circulating H3N2 strains were poorly matched with strains contained in the vaccine also indicated that LAIV, but not TIV, was effective against antigenically drifted H3N2 strains during that influenza season. In this study, children aged 5—18 years who received LAIV had significant protection against laboratory-confirmed influenza (37%) and pneumonia and influenza events (50%) (278).
Although LAIV is not licensed for use in persons with risk factors for influenza complications, certain studies have compared the efficacy of LAIV to TIV in these groups. LAIV provided 32% increased protection in preventing culture-confirmed influenza compared with TIV in one study conducted among children aged >6 years and adolescents with asthma (283) and 52% increased protection compared with TIV among children aged 6—71 months with recurrent respiratory tract infections (284).
Effectiveness of Vaccination for Decreasing Transmission to Contacts
Decreasing transmission of influenza from caregivers and household contacts to persons at high risk might reduce ILI and complications among persons at high risk. Influenza virus infection and ILI are common among HCP (285—287). Influenza outbreaks have been attributed to low vaccination rates among HCP in hospitals and long-term—care facilities (288—290). One serosurvey demonstrated that 23% of HCP had serologic evidence of influenza virus infection during a single influenza season; the majority had mild illness or subclinical infection (285). Observational studies have demonstrated that vaccination of HCP is associated with decreased deaths among nursing home patients (291,292). In one cluster-randomized controlled trial that included 2,604 residents of 44 nursing homes, significant decreases in mortality, ILI, and medical visits for ILI care were demonstrated among residents in nursing homes in which staff were offered influenza vaccination (coverage rate: 48%), compared with nursing homes in which staff were not provided with vaccination (coverage rate: 6%) (293). A review concluded that vaccination of HCP in settings in which patients were also vaccinated provided significant reductions in deaths among elderly patients from all causes and deaths from pneumonia (294).
Epidemiologic studies of community outbreaks of influenza demonstrate that school-age children typically have the highest influenza illness attack rates, suggesting routine universal vaccination of children might reduce transmission to their household contacts and possibly others in the community. Results from certain studies have indicated that the benefits of vaccinating children might extend to protection of their adult contacts and to persons at risk for influenza complications in the community. However, these data are limited and studies have not used laboratory-confirmed influenza as an outcome measure. A single-blinded, randomized controlled study conducted during as part of a 1996—1997 vaccine effectiveness study demonstrated that vaccinating preschool-aged children with TIV reduced influenza-related morbidity among some household contacts (295). A randomized, placebo-controlled trial among children with recurrent respiratory tract infections demonstrated that members of families with children who had received LAIV were significantly less likely to have respiratory tract infections and reported significantly fewer workdays lost, compared with families with children who received placebo (296). In nonrandomized community-based studies, administration of LAIV has been demonstrated to reduce MAARI (297,298) and ILI-related economic and medical consequences (e.g., workdays lost and number of health-care provider visits) among contacts of vaccine recipients (298). Households with children attending schools in which school-based LAIV vaccination programs had been established reported less ILI and fewer physician visits during peak influenza season, compared with households with children in schools in which no LAIV vaccination had been offered. However a decrease in the overall rate of school absenteeism was not reported in communities in which LAIV vaccination was offered (298). These community-based studies have not used laboratory-confirmed influenza as an outcome.
Some studies have also documented reductions in influenza illness among persons living in communities where focused programs for vaccinating children have been conducted. A community-based observational study conducted during the 1968 pandemic using a univalent inactivated vaccine reported that a vaccination program targeting school-aged children (coverage rate: 86%) in one community reduced influenza rates within the community among all age groups compared with another community in which aggressive vaccination was not conducted among school-aged children (299). An observational study conducted in Russia demonstrated reductions in ILI among the community-dwelling elderly after implementation of a vaccination program using TIV for children aged 3—6 years (57% coverage achieved) and children and adolescents aged 7—17 years (72% coverage achieved) (300). In a nonrandomized community-based study conducted over three influenza seasons, 8%—18% reductions in the incidence of MAARI during the influenza season among adults aged >35 years were observed in communities in which LAIV was offered to all children aged >18 months (estimated coverage rate: 20%—25%) compared with communities with such vaccination programs (297). In a subsequent influenza season, the same investigators documented a 9% reduction in MAARI rates during the influenza season among persons aged 35—44 years in intervention communities, where coverage was estimated at 31% among school children, compared with control communities. However, MAARI rates among persons aged >45 years were lower in the intervention communities regardless of the presence of influenza in the community, suggesting that lower rates could not be attributed to vaccination of school children against influenza (301).
Effectiveness of Influenza Vaccination when Circulating Influenza Virus Strains Differ from Vaccine Strains
Manufacturing trivalent influenza virus vaccines is a challenging process that takes 6—8 months to complete. This manufacturing timeframe requires that influenza vaccine strains for influenza vaccines used in the United States must be selected in February of each year by the FDA to allow time for manufacturers to prepare vaccines for the next influenza season. Vaccine strain selections are based on global viral surveillance data that is used to identify trends in antigenic changes among circulating influenza viruses and the availability of suitable vaccine virus candidates.
Vaccination can provide reduced but substantial cross-protection against drifted strains in some seasons, including reductions in severe outcomes such as hospitalization. Usually one or more circulating viruses with antigenic changes compared with the vaccine strains are identified in each influenza season. However, assessment of the clinical effectiveness of influenza vaccines cannot be determined solely by laboratory evaluation of the degree of antigenic match between vaccine and circulating strains. In some influenza seasons, circulating influenza viruses with significant antigenic differences predominate and, compared with seasons when vaccine and circulating strains are well-matched, reductions in vaccine effectiveness are sometimes observed (126,139,145, 147,191). However, even during years when vaccine strains were not antigenically well matched to circulating strains, substantial protection has been observed against severe outcomes, presumably because of vaccine-induced cross-reacting antibodies (139,145,147,273). For example, in one study conducted during an influenza season (2003—04) when the predominant circulating strain was an influenza A (H3N2) virus that was antigenically different from that season’s vaccine strain, effectiveness among persons aged 50—64 years against laboratory-confirmed influenza illness was 60% among healthy persons and 48% among persons with medical conditions that increase risk for influenza complications (147). An interim, within-season analysis during the 2007—08 influenza season indicated that vaccine effectiveness was 44% overall, 54% among healthy persons aged 5—49 years, and 58% against influenza A, despite the finding that viruses circulating in the study area were predominately a drifted influenza A H3N2 and a influenza B strain from a different lineage compared with vaccine strains (302). Among children, both TIV and LAIV provide protection against infection even in seasons when vaccines and circulating strains are not well matched. Vaccine effectiveness against ILI was 49%—69% in two observational studies, and 49% against medically attended, laboratory-confirmed influenza in a case-control study conducted among young children during the 2003—04 influenza season, when a drifted influenza A H3N2 strain predominated, based on viral surveillance data (121,125). However, continued improvements in collecting representative circulating viruses and use surveillance data to forecast antigenic drift are needed. Shortening manufacturing time to increase the time to identify good vaccine candidate strains from among the most recent circulating strains also is important. Data from multiple seasons and collected in a consistent manner are needed to better understand vaccine effectiveness during seasons when circulating and vaccine virus strains are not well-matched.
Cost-Effectiveness of Influenza Vaccination
Economic studies of influenza vaccination are difficult to compare because they have used different measures of both costs and benefits (e.g., cost-only, cost-effectiveness, cost-benefit, or cost-utility). However, most studies find that vaccination reduces or minimizes health care, societal, and individual costs, or the productivity losses and absenteeism associated with influenza illness. One national study estimated the annual economic burden of seasonal influenza in the United States (using 2003 population and dollars) to be $87.1 billion, including $10.4 billion in direct medical costs (303).
Studies of influenza vaccination in the United States among persons aged >65 years have documented substantial reductions in hospitalizations and deaths and overall societal cost savings (186,187). Studies comparing adults in different age groups also find that vaccination is economically beneficial. One study that compared the economic impact of vaccination among persons aged >65 years with those aged 15—64 years indicated that vaccination resulted in a net savings per quality-adjusted life year (QALY) and that the Medicare program saved costs of treating illness by paying for vaccination (304). A study of a larger population comparing persons aged 50—64 years with those aged >65 years estimated the cost-effectiveness of influenza vaccination to be $28,000 per QALY saved (in 2000 dollars) in persons aged 50—64 years compared with $980 per QALY saved among persons aged >65 years (305).
Economic analyses among adults aged <65 years have reported mixed results regarding influenza vaccination. Two studies in the United States found that vaccination can reduce both direct medical costs and indirect costs from work absenteeism and reduced productivity (306,307). However, another United States study indicated no productivity and absentee savings in a strategy to vaccinate healthy working adults, although vaccination was still estimated to be cost-effective (139).
Cost analyses have documented the considerable cost burden of illness among children. In a study of 727 children at a medical center during 2000—2004, the mean total cost of hospitalization for influenza-related illness was $13,159 ($39,792 for patients admitted to an intensive care unit and $7,030 for patients cared for exclusively on the wards) (308). Strategies that focus on vaccinating children with medical conditions that confer a higher risk for influenza complications are more cost-effective than a strategy of vaccinating all children (309). An analysis that compared the costs of vaccinating children of varying ages with TIV and LAIV indicated that costs per QALY saved increased with age for both vaccines. In 2003 dollars per QALY saved, costs for routine vaccination using TIV were $12,000 for healthy children aged 6—23 months and $119,000 for healthy adolescents aged 12—17 years, compared with $9,000 and $109,000 using LAIV, respectively (310). Economic evaluations of vaccinating children have demonstrated a wide range of cost estimates, but have generally found this strategy to be either cost-saving or cost-beneficial (311—314).
Economic analyses are sensitive to the vaccination venue, with vaccination in medical care settings incurring higher projected costs. In a published model, the mean cost (year 2004 values) of vaccination was lower in mass vaccination ($17.04) and pharmacy ($11.57) settings than in scheduled doctor’s office visits ($28.67) (315). Vaccination in nonmedical settings was projected to be cost saving for healthy adults aged >50 years and for high-risk adults of all ages. For healthy adults aged 18—49 years, preventing an episode of influenza would cost $90 if vaccination were delivered in a pharmacy setting, $210 in a mass vaccination setting, and $870 during a scheduled doctor’s office visit (315). Medicare payment rates in recent years have been less than the costs associated with providing vaccination in a medical practice (316).
Vaccination Coverage Levels
Continued annual monitoring is needed to determine the effects on vaccination coverage of vaccine supply delays and shortages, changes in influenza vaccination recommendations and target groups for vaccination, reimbursement rates for vaccine and vaccine administration, and other factors related to vaccination coverage among adults and children. One of the national health objectives for 2010 includes achieving an influenza vaccination coverage level of 90% for persons aged >65 years and among nursing home residents (317,318); new strategies to improve coverage are needed to achieve these objectives (319,320). Increasing vaccination coverage among persons who have high-risk conditions and are aged <65 years, including children at high risk, is the highest priority for expanding influenza vaccine use.
On the basis of the 2006 final data set and the 2007 early release data from the National Health Interview Survey (NHIS), estimated national influenza vaccine coverage during the 2005—06 and 2006—07 influenza seasons among persons aged >65 years and 50—64 years increased slightly from 32% and 65%, respectively to 36% and 66% (Table 3) and appear to be approaching coverage levels observed before the 2004—05 vaccine shortage year. In 2005—06 and 2006—07, estimated vaccination coverage levels among adults with high-risk conditions aged 18—49 years were 23% and 26%, respectively, substantially lower than the Healthy People 2000 and Healthy People 2010 objectives of 60% (Table 3) (317,318).
Opportunities to vaccinate persons at risk for influenza complications (e.g., during hospitalizations for other causes) often are missed. In a study of hospitalized Medicare patients, only 31.6% were vaccinated before admission, 1.9% during admission, and 10.6% after admission (321). A study in New York City during 2001—2005 among 7,063 children aged 6—23 months indicated that 2-dose vaccine coverage increased from 1.6% to 23.7%. Although the average number of medical visits during which an opportunity to be vaccinated decreased during the course of the study from 2.9 to 2.0 per child, 55% of all visits during the final year of the study still represented a missed vaccination opportunity (322). Using standing orders in hospitals increases vaccination rates among hospitalized persons (323). In one survey, the strongest predictor of receiving vaccination was the survey respondent’s belief that he or she was in a high-risk group. However, many persons in high-risk groups did not know that they were in a group recommended for vaccination (324).
Reducing racial and ethnic health disparities, including disparities in influenza vaccination coverage, is an overarching national goal that is not being met (317). Estimated vaccination coverage levels in 2007 among persons aged >65 years were 70% for non-Hispanic whites, 58% for non-Hispanic blacks, and 54% for Hispanics (325). Among Medicare beneficiaries, other key factors that contribute to disparities in coverage include variations in the propensity of patients to actively seek vaccination and variations in the likelihood that providers recommend vaccination (326,327). One study estimated that eliminating these disparities in vaccination coverage would have an impact on mortality similar to the impact of eliminating deaths attributable to kidney disease among blacks or liver disease among Hispanics (328).
Reported vaccination levels are low among children at increased risk for influenza complications. Coverage among children aged 2—17 years with asthma for the 2004—05 influenza season was estimated to be 29% (329). One study reported 79% vaccination coverage among children attending a cystic fibrosis treatment center (330). During the first season for which ACIP recommended that all children aged 6 months—23 months receive vaccination, 33% received one or more dose of influenza vaccination, and 18% received 2 doses if they were unvaccinated previously (331). Among children enrolled in HMOs who had received a first dose during 2001—2004, second dose coverage varied from 29% to 44% among children aged 6—23 months and from 12% to 24% among children aged 2—8 years (332). A rapid analysis of influenza vaccination coverage levels among members of an HMO in Northern California demonstrated that during 2004—2005, the first year of the recommendation for vaccination of children aged 6—23 months, 1-dose coverage was 57% (333). During the 2005—06 influenza season, the second season for which ACIP recommended that all children aged 6 months—23 months receive vaccination, coverage remained low and did not increase substantially from the 2004—05 season. Data collected in 2006 by the National Immunization Survey indicated that for the 2005—06 season, 32% of children aged 6—23 months received at least 1 dose of influenza vaccine and 21% were fully vaccinated (i.e., received 1 or 2 doses depending on previous vaccination history); however, results varied substantially among states (334). As has been reported for older adults, a physician recommendation for vaccination and the perception that having a child be vaccinated “is a smart idea” were associated positively with likelihood of vaccination of children aged 6—23 months (335). Similarly, children with asthma were more likely to be vaccinated if their parents recalled a physician recommendation to be vaccinated or believed that the vaccine worked well (336). Implementation of a reminder/recall system in a pediatric clinic increased the percentage of children with asthma or reactive airways disease receiving vaccination from 5% to 32% (337).
Although annual vaccination is recommended for HCP and is a high priority for reducing morbidity associated with influenza in health-care settings and for expanding influenza vaccine use (338—340), national survey data demonstrated a vaccination coverage level of only 42% among HCP during the 2005—06 season (Table 3). Vaccination of HCP has been associated with reduced work absenteeism (286) and with fewer deaths among nursing home patients (292,293) and elderly hospitalized patients (294). Factors associated with a higher rate of influenza vaccination among HCP include older age, being a hospital employee, having employer provided health-care insurance, having had pneumococcal or hepatitis B vaccination in the past, or having visited a health-care professional during the preceding year. Non-Hispanic black HCP were less likely than non-Hispanic white HCP to be vaccinated (341). Beliefs that are frequently cited by HCP who decline vaccination include doubts about the risk for influenza and the need for vaccination, concerns about vaccine effectiveness and side effects, and dislike of injections (342).
Vaccine coverage among pregnant women has not increased significantly during the preceding decade. (343). Only 12% and 13% of pregnant women participating in the 2006 and 2007 NHIS reported vaccination during the 2005—06 and 2006—07 seasons, respectively, excluding pregnant women who reported diabetes, heart disease, lung disease, and other selected high-risk conditions (Table 3). In a study of influenza vaccine acceptance by pregnant women, 71% of those who were offered the vaccine chose to be vaccinated (344). However, a 1999 survey of obstetricians and gynecologists determined that only 39% administered influenza vaccine to obstetric patients in their practices, although 86% agreed that pregnant women’s risk for influenza-related morbidity and mortality increases during the last two trimesters (345).
Influenza vaccination coverage in all groups recommended for vaccination remains suboptimal. Despite the timing of the peak of influenza disease, administration of vaccine decreases substantially after November. According to results from the NHIS regarding the two most recent influenza seasons for which these data are available, approximately 84% of all influenza vaccination were administered during September—November. Among persons aged >65 years, the percentage of September—November vaccinations was 92% (346). Because many persons recommended for vaccination remain unvaccinated at the end of November, CDC encourages public health partners and health-care providers to conduct vaccination clinics and other activities that promote influenza vaccination annually during National Influenza Vaccination Week and throughout the remainder of the influenza season.
Self-report of influenza vaccination among adults, compared with determining vaccination status from the medical record, is a sensitive and specific source of information (347). Patient self-reports should be accepted as evidence of influenza vaccination in clinical practice (347). However, information on the validity of parents’ reports of pediatric influenza vaccination is not yet available.
CLINIC FOR CHILDREN
JL Taman Bendungan Asahan 5 Jakarta Indonesia 102010
phone : 62(021) 70081995 – 5703646
Clinical and Editor in Chief :
DR WIDODO JUDARWANTO
email : email@example.com
Copyright © 2009, Clinic For Children Information Education Network. All rights reserved.