- Email: [email protected]
Influenza vaccination and risk of hospitalization among adults with laboratory confirmed influenza illness
Vaccine 32 (2014) 453–457
Contents lists available at ScienceDirect
Vaccine journal homepage: www.elsevier.com/locate/vaccine
Influenza vaccination and risk of hospitalization among adults with laboratory confirmed influenza illness夽 Huong Q. McLean ∗ , Jennifer K. Meece, Edward A. Belongia Marshfield Clinic Research Foundation, Marshfield, WI, USA
a r t i c l e
i n f o
Article history: Received 13 September 2013 Received in revised form 12 November 2013 Accepted 15 November 2013 Available online 26 November 2013 Keywords: Influenza vaccine Severity Hospitalization Effectiveness
a b s t r a c t Background: Influenza vaccine is moderately effective for preventing influenza illness. It is not known if vaccination reduces the risk of subsequent hospital admission among patients with vaccine failure and laboratory confirmed influenza illness. Methods: Patients in a community cohort presenting with acute respiratory illness were prospectively enrolled and tested for influenza during 8 seasons to estimate seasonal vaccine effectiveness. Hospital admissions within 14 days after illness onset were identified for all participants aged ≥20 years with laboratory confirmed influenza. The association between vaccination and hospital admission was examined in a propensity score adjusted logistic regression model. The model was validated by examining the association between vaccination and hospital admission in participants without influenza. Results: Influenza was identified in 1393 (28%) of 4996 participants. Sixty-two (6%) of 1020 with influenza A and 17 (5%) of 369 with influenza B were hospitalized. Vaccination was not associated with a reduced risk of hospital admission among all participants with influenza [adjusted odds ratio (aOR) = 1.08; 95% CI: 0.62, 1.88]; or among those with influenza A (aOR = 1.35; 95% CI: 0.71, 2.57) or influenza B (aOR = 0.67; 95% CI: 0.21, 2.15). Influenza vaccination was not associated with hospitalization after non-influenza respiratory illness (aOR = 1.14; 95% CI: 0.84, 1.54). Conclusions: Influenza vaccination did not reduce the risk of subsequent hospital admission among patients with vaccine failure. These findings do not support the hypothesis that vaccination mitigates influenza illness severity. © 2013 The Authors. Published by Elsevier Ltd. All rights reserved.
1. Introduction Influenza is an important cause of death and serious illness, particularly among adults aged ≥65 years and those with certain underlying chronic conditions. In the United States, approximately 226,000 hospital admissions are attributed to influenza each year [1]. As a result, annual influenza vaccination is recommended for all persons aged ≥6 months to prevent seasonal influenza infection and its complications [2]. However, influenza vaccine failure is common even during seasons with optimal antigenic match between circulating and vaccine viruses. Among adults, vaccine efficacy in preventing laboratory confirmed influenza illness is estimated to be approximately 60% [3]. Similar efficacy has been reported for preventing hospital admission with laboratory
夽 This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. ∗ Corresponding author at: Marshfield Clinic Research Foundation (ML2), 1000 N. Oak Ave, Marshfield, WI 54449, USA. Tel.: +1 715 389 3707; fax: +1 715 389 3880. E-mail address: mclean.huong@marshfieldclinic.org (H.Q. McLean).
confirmed pandemic or seasonal influenza [4–10]. It is not clear if influenza vaccination prevents serious outcomes by primary prevention of influenza infection, by reducing severity of influenza illness, or both. We conducted a population based study of laboratory confirmed influenza among adults aged ≥20 years over multiple seasons to determine if receipt of same-season influenza vaccine was associated with reduced risk of hospital admission within 14 days after onset of influenza illness. 2. Methods 2.1. Study population and design This was a secondary analysis of data from population-based studies of influenza vaccine effectiveness during eight influenza seasons, 2004–05 through 2012–13, in Marshfield, Wisconsin [11–14]. In this community, residents receive nearly all outpatient and inpatient care from the Marshfield Clinic. A single acute care hospital (St. Joseph’s) serves the study population, and both inpatient and outpatient diagnoses are accessible through a combined electronic medical record. The electronic medical record captures
0264-410X/$ – see front matter © 2013 The Authors. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.vaccine.2013.11.060
454
H.Q. McLean et al. / Vaccine 32 (2014) 453–457
90% of outpatient visits, 95% of hospital discharges, and 99% of deaths for the residents in the area [15–18]. During each influenza season, eligible community dwelling residents were recruited by trained research coordinators during or after an inpatient or outpatient medical encounter for acute respiratory illness. Research coordinators used an electronic appointment system to identify and recruit eligible persons in all primary care clinics and in urgent care on weekdays, evenings, and weekends. Eligible persons were also recruited at the hospital that is contiguous with Marshfield Clinic. Most ill persons who were not approached during a clinical encounter were identified on the following day by use of electronic diagnosis codes entered by attending physicians (ICD-9-CM codes 382.0, 382.4, 382.9, 460–466, 480, 483–486, 487, 490, 780.6, and 786.2). These individuals were contacted by telephone, and a swab sample was obtained at home from those who were eligible and consented. Participants completed a short interview to assess illness symptoms and onset date; nasopharyngeal swabs were obtained for influenza testing. Real-time reverse transcription polymerase chain reaction (RT-PCR) and viral cultures were performed at the Marshfield Clinic Research Foundation as previously described [11]. Culture alone was performed on samples collected in 2004–05 and RT-PCR was performed in subsequent years. Subtype results based on RT-PCR were not available for 11% of influenza A positive samples. For those samples, the subtype was assumed to be the predominant subtype identified among study participants during that season. This analysis excluded the 2009–10 season because monovalent vaccine was not available to the local population when the pandemic wave arrived in October–November 2009, and influenza was absent from the study population in the subsequent winter months. 2.2. Influenza vaccination status Influenza vaccination status was determined by a real-time, internet-based vaccination registry used by all public and private vaccination providers serving the population (http://www. recin.org). A validation study of the registry during the 2006–07 and 2007–08 influenza seasons demonstrated that the registry captured 95% of all influenza vaccinations that were received by study participants [19]. A similar high level of capture was demonstrated in a validation study during the 2011–12 season (unpublished data). Adults were classified as vaccinated if they had received influenza vaccine ≥14 days before the onset of illness. 2.3. Ascertainment of hospital admissions and clinical history Dates of hospital admission and discharge diagnoses were identified from the electronic medical record for a 14 day period after onset of influenza illness. To adjust for use of antiviral drugs, we extracted dates of antiviral prescriptions for all participants. 2.4. Outcome and covariates The main outcome was an acute care hospital admission occurring within 14 days of influenza symptom onset. Although most hospital admissions occurred after an outpatient enrollment, some participants were initially enrolled and swabbed after admission to the hospital. Covariates included age, gender, antiviral prescription, specific high risk medical conditions, year, and influenza type/subtype [A/H3N2, A/H1N1, pandemic H1N1 (A/H1N1pdm09), B]. Study participants were classified as having a high risk medical condition if they had at least one visit during a recent 12 month period with an ICD-9 CM diagnosis code of interest. High risk conditions were classified into the following groups: cancer, cardiovascular disease, diabetes, pulmonary, and other.
Antiviral prescription was defined as a prescription of oseltamivir, zanamivir, amantadine, or rimantadine within 14 days of symptom onset for persons not hospitalized and between symptom onset and hospital admission for persons who were hospitalized. 2.5. Analyses We restricted the analysis of hospital admissions to enrolled adults aged ≥20 years because influenza-related hospitalization was less common in children, and potential confounding factors are likely to be different for adults and children. Studies of influenza vaccination and hospital admission are particularly susceptible to confounding, since persons who are vaccinated may be more likely to have pre-existing chronic medical conditions or other risk factors for hospital admission. To minimize confounding by indication for vaccination, we used a propensity score regression adjustment [20,21]. Propensity scores allow adjustment for multiple potential confounders when the data are too sparse to include a separate effect for each potential confounder in the regression model. The propensity scores were generated from a multivariable logistic regression model that assessed the probability of influenza vaccination as a function of the potential confounders. In the propensity model, the dependent variable was influenza vaccination status and the independent variables were potential confounders identified a priori. The propensity score covariates included age, gender, cancer, cardiovascular disease, diabetes, pulmonary disorders, other high risk conditions, and year. The propensity scores from the model were then included as a continuous variable in the final logistic regression model that assessed the association between influenza vaccination and hospital admission. To determine the effect of influenza vaccination among persons with laboratory confirmed influenza, the final logistic regression model predicting hospital admission included the following covariates: propensity score, influenza vaccination, age group, influenza type/subtype, receipt of antiviral drug prescription. The primary analysis included all study participants with laboratory confirmed influenza. Secondary analyses included subgroups based on influenza type (A or B). We excluded the small number of participants with both A and B infection because the risk of hospitalization may be different for those co infected with both types and persons with unknown vaccination status. Since the primary outcome included all hospital admissions during a 14 day period, we performed a secondary analysis restricted to hospital admissions that were directly related to influenza infection. These included individuals who received any discharge diagnosis (among the top three diagnosis codes) for influenza, pneumonia, bronchitis, exacerbation of chronic pulmonary disease, or acute respiratory infection. In addition, one individual with a discharge diagnosis of fever was included in this group because symptoms of influenza like illness were present at the time of admission. We also performed an analysis restricted to persons who were enrolled in the outpatient setting and subsequently admitted to the hospital. Finally, we evaluated residual confounding by examining the association between influenza vaccination and hospital admission among study participants with a negative influenza test in a logistic regression model. The propensity scores for study participants with a negative influenza test (i.e., non-influenza respiratory illness) were generated using the same method as described above. If the propensity scores adequately adjusted for confounding, there should be no association between influenza vaccine receipt and hospital admission in that group. We assumed that confounders would be the same for influenza negative and influenza positive study participants.
H.Q. McLean et al. / Vaccine 32 (2014) 453–457 Table 1 Clinical and demographic characteristics and influenza vaccination status for adult study participants with laboratory confirmed influenza. Patient characteristics
Age 20–49 years 50–59 years 60–69 years 70–79 years ≥80 years Male Received antiviral prescription High risk conditions Cancer Cardiovascular Diabetes Pulmonary disordersa Other high riskb Season 2004–05 2005–06 2006–07 2007–08 2008–09 2010–11 2011–12 2012–13
Vaccinated n = 552 No. (%)
Unvaccinated n = 837 No. (%)
208 (38) 108 (20) 102 (18) 88 (16) 46 (8) 195 (35) 117 (21)
584 (70) 161 (19) 57 (7) 24 (3) 11 (1) 388 (46) 199 (24)
64 (12) 130 (23) 98 (18) 122 (22) 109 (20)
20 (2) 79 (9) 42 (5) 84 (10) 69 (8)
80 (14) 16 (3) 12 (2) 161 (29) 43 (8) 41 (7) 36 (7) 163 (30)
34 (4) 14 (2) 22 (3) 339 (41) 101 (12) 73 (9) 56 (7) 198 (24)
Table 2 Characteristics associated with influenza vaccination among adults hospitalized with confirmed influenza.
p
<0.0001
<0.001 0.3 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001
a Pulmonary disorders included pulmonary tuberculosis, other respiratory tuberculosis, other diseases due to mycobacteria, sarcoidosis, cystic fibrosis, alpha-1 anti-trypsin deficiency, chronic bronchitis, emphysema, asthma, bronchiectasis, extrinsic allergic alveolitis, chronic airway obstruction, and coal workers. b Kidney disease, liver disease, blood disorders, immunosuppressive disorders, metabolic disorders, and neurological/musculoskeletal disorders.
Unadjusted risk ratios were used to compare the risk of influenza vaccination among adults hospitalized with influenza. All analyses were performed using SAS 9.3 (SAS Institute Inc., Cary, NC); p-values <0.05 were considered significant. 3. Results During the 8 influenza seasons, 4996 adults with acute respiratory illness seeking medical care were enrolled. Influenza infection was laboratory confirmed for 1393 persons; 1020 (73%) had type A infection, 369 (26%) had type B infection, and 4 (<1%) were positive for both type A and B. Most (84%) influenza A infections were H3N2 subtype, followed by H1N1 (10%) and H1N1pdm09 (6%). The number of influenza A positive study participants ranged from 18 in the 2005–06 season to 356 in the 2007–08 season. The number of influenza B positive study participants ranged from 5 in the 2006–07 season to 144 in the 2007–08 season. Among persons with laboratory confirmed influenza and known vaccination status, 583 (42%) were males, 540 (39%) had at least one high risk condition, 316 (23%) were prescribed antiviral medications, and 31 (2%) were enrolled after admission to the hospital. The proportion vaccinated differed with respect to age, gender, and presence of high risk conditions (Table 1). In particular, influenza vaccination was more common in older adults and women. The median age was 55 years [interquartile range (IQR): 41, 69] among adults who were vaccinated and 41 years (IQR: 30, 52) among adults who were not vaccinated (p < 0.001). Vaccination was also more common among persons with cancer, cardiovascular disease, diabetes, pulmonary disorders, and other high risk conditions compared to those without these high risk conditions. Similar patterns were observed when examined by influenza type. Seventy-nine patients with laboratory confirmed influenza were admitted to the hospital within 14 days of symptom onset: 62 (6%) of 1020 with influenza A and 17 (5%) of 369 with influenza B.
455
Age 20–49 years 50–59 years 60–69 years 70–79 years ≥80 years Gender Male Female High risk conditions Cancer Yes No Cardiovascular Yes No Diabetes Yes No Pulmonary disorders Yes No Other high riska Yes No
Total persons No.
Persons vaccinated No. (%)
Risk ratio (95% CI)
21 10 15 14 19
6 (29) 6 (60) 9 (60) 11 (79) 15 (79)
1.0 2.10 (0.90, 4.89) 2.10 (0.95, 4.64) 2.75 (1.33, 5.70) 2.76 (1.35, 5.65)
34 45
21 (62) 26 (58)
1.07 (0.74, 1.54) 1.0
14 65
11 (79) 36 (55)
1.42 (1.00, 2.01) 1.0
29 50
22 (76) 25 (50)
1.52 (1.07, 2.14) 1.0
14 65
10 (71) 37 (57)
1.25 (0.85, 1.86) 1.0
29 50
19 (66) 28 (56)
1.17 (0.82, 1.68) 1.0
20 59
15 (75) 32 (54)
1.38 (0.98, 1.95) 1.0
a Kidney disease, liver disease, blood disorders, immunosuppressive disorders, metabolic disorders, and neurological/musculoskeletal disorders.
The median time from symptom onset to hospital admission was 3 days (IQR: 2–5 days). Seventy (89%) had discharge diagnoses codes that were consistent with an acute respiratory illness or exacerbation of chronic pulmonary disease. Among hospitalized patients, those who were older were more likely to be vaccinated compared to those aged 20–49 years and those with a cardiovascular high risk condition were more likely to be vaccinated compared to those without a cardiovascular high risk condition (Table 2). Vaccination status among hospitalized patients was not associated with gender or the other high risk conditions examined. Among patients with laboratory confirmed influenza, influenza vaccination was not associated with a decreased risk of hospitalization following onset overall or by influenza type (Table 3). The propensity score adjusted odd ratio of hospitalization for vaccinated compared to unvaccinated patients was 1.08 (95% CI: 0.62, 1.88), 1.35 (95% CI: 0.71, 2.57), and 0.67 (95% CI: 0.21, 2.15) overall, for type A infection, and for type B infection, respectively. The results were similar in secondary analyses where the outcome was hospital admission for acute respiratory illness rather than all hospital admissions and for analyses restricted to patients enrolled in the outpatient setting and subsequently hospitalized (Table 3). In the 3603 adults with non-influenza respiratory illness, there was no association between influenza vaccination and hospital admission within 14 days after illness onset (propensity score adjusted OR = 1.14; 95% CI: 0.84, 1.54; p = 0.4). 4. Discussion In this multi-season study, we examined the hypothesis that vaccination may mitigate influenza illness severity and reduce the risk of hospital admission. We found that vaccinated and unvaccinated individuals with influenza had a similar risk of hospital admission after adjustment for propensity to be vaccinated, regardless of influenza type. This suggests that influenza vaccination prevents serious outcomes by primary prevention of influenza infection.
456
H.Q. McLean et al. / Vaccine 32 (2014) 453–457
Table 3 Propensity score adjusted association between influenza vaccination status and hospital admission within 14 days after illness onset for study participants with laboratory confirmed influenza. Propensity scores incorporated factors associated with receipt of influenza vaccine, including age, gender, cancer, cardiovascular disease, diabetes, chronic pulmonary disease, other high risk conditions, and year. All hospital admissions within 14 days of illness onset
Any influenza virus Influenza A virus Influenza B virus
Admissions for acute respiratory illnessa
Admission after outpatient enrollmentb
No. cases (% vaccinated)
Vaccinated No. (%) hospitalized
Unvaccinated No. (%) hospitalized
Crude OR (95% CI)
Adjusted OR (95% CI)
Adjusted OR (95% CI)
Adjusted OR (95% CI)
1389 (40)
47 (9)
32 (4)
1020 (43)
40 (9)
22 (4)
369 (31)
7 (6)
10 (4)
2.34 (1.47, 3.72) 2.57 (1.50, 4.39) 1.58 (0.59, 4.27)
1.08c (0.62, 1.88) 1.35d (0.71, 2.57) 0.67 (0.21, 2.15)
0.95c (0.52, 1.70) 1.23d (0.62, 2.44) 0.54 (0.16, 1.81)
1.11c (0.56, 2.21) 1.21d (0.54, 2.71) 1.02 (0.28, 3.82)
a
Nine patients with hospital admission that did not appear to be associated with influenza infection were excluded. Thirty one patients were enrolled after hospital admission and were excluded. c Adjusted for propensity score, age, influenza type/subtype (A/H3N2, A/H1N1, A/H1N1pdm09, B), and receipt of antiviral drug prescription. d Adjusted for propensity score, age, influenza subtype (A/H3N2, A/H1N1, A/H1N1pdm09), and receipt of antiviral drug prescription. Adjusted for propensity score, age, and receipt of antiviral drug prescription. b
In the past decade, multiple observational studies of vaccine effectiveness have been performed using medically attended influenza (confirmed by RT-PCR) as the primary endpoint. Most of these studies have assessed vaccine effectiveness for preventing outpatient influenza illness, but few have focused on vaccine effectiveness for preventing hospitalization with laboratory confirmed influenza [4–10,22–25]. In these studies where the comparison groups were those without influenza, vaccine effectiveness estimates ranged from 25% to 74%. An important finding from these studies is that vaccination provides moderate benefit against influenza hospitalization, presumably due to primary prevention of influenza illness. To our knowledge, one other study has examined the association between vaccination and hospital admission among persons with influenza. Despite a different study population and most cases being caused by A/H1N1pdm09, they had similar findings to our study: vaccination did not reduce the risk of hospitalization [9]. Additionally, they found that hospitalized patients who were vaccinated were less likely to have had severe disease. However, because the study was observational, it is not possible to know whether this association was due to vaccination, residual confounding, or confounding from unmeasured factors. Due to the limited number of hospitalized cases in our study, we were unable to assess the impact of vaccination on severity of cases among those hospitalized. We attempted to minimize confounding with a propensity score that adjusted for the likelihood of influenza vaccination based on multiple covariates. The propensity score model was tested in study participants with non-influenza respiratory illness, since an association between vaccination and hospital admission is not biologically plausible in the absence of influenza. The model with propensity score adjustment showed no evidence of confounding in this group: the odds ratio for hospital admission in vaccinated versus unvaccinated adults with non-influenza illness was 1.1 (p = 0.4). This supported the validity of the propensity score model to examine the association between vaccination and hospitalization among patients with influenza. Strengths of this study included systematic recruitment and sample collection from a community cohort with medically attended acute respiratory illness, use of a highly sensitive and specific RT-PCR assay, access to a validated immunization registry, and complete capture of hospital admissions from the electronic medical record. However, several limitations should be acknowledged. First, hospitalization due to influenza is rare in healthy adult populations. Despite eight seasons, there were few hospitalizations in
our study, all of which were from a single hospital in central Wisconsin. Second, antigenic characterization was not performed for many positive samples, and minor antigenic drift can be difficult to detect and interpret. As a result, we were not able to assess the potential impact of antigenic variability. The 2007–08 season accounted for the majority of A (H3N2) infections, and during that year there was circulation of A/Brisbane/10/2007-like viruses that were minor antigenic variants from the vaccine strain [26]. Third, our classification of high risk medical conditions was based on ICD-9 diagnosis codes without medical record validation. However, all diagnoses were entered by physicians and automatically mapped to ICD-9 codes in the electronic medical record, which reduced the potential for coding error. Finally, our study population included primarily outpatient influenza cases and there may have been differential health care seeking behavior between vaccinated and unvaccinated individuals. We cannot exclude the possibility that vaccinated individuals had milder influenza illness and did not seek medical attention. In that scenario, vaccination would have reduced illness severity, leading to fewer outpatient visits and hospitalizations, but this would not be evident when comparing the risk of hospitalization in vaccinated and unvaccinated outpatients. However, we note that estimates of vaccine effectiveness in the outpatient setting are generally similar to estimates of efficacy based on randomized clinical trials, and the primary endpoint for clinical trials is influenza illness rather than severity. Because of these limitations, results should be interpreted with caution. Hospitalization is an important complication of influenza infection from a public health and an economic perspective. Available evidence suggests that influenza vaccine provides moderate protection against influenza-related hospitalization. Further research is warranted to assess the impact of vaccination in preventing severe outcomes among vaccine failures, including differences by type, subtype, and lineage.
Acknowledgements We thank the following individuals for their contribution to this work: Burney Kieke, Sarah Kopitzke, Pam Squires, Jim Donahue, Stephanie Irving, David Shay, and Alicia Fry. Conflicts of interest: HQM, JKM, and EAB receive research funding from MedImmune, LLC. Funding: Centers for Disease Control and Prevention(cooperative agreement U18 IP000183).
H.Q. McLean et al. / Vaccine 32 (2014) 453–457
References [1] Thompson WW, Shay DK, Weintraub E, et al. Influenza-associated hospitalizations in the United States. JAMA 2004;292(11):1333–40. [2] Centers for Disease Control Prevention. Prevention and control of influenza with vaccines: recommendations of the Advisory Committee on Immunization Practices (ACIP)–United States, 2012-13 influenza season. MMWR 2012;61(32):613–8. [3] Osterholm MT, Kelley NS, Sommer A, Belongia EA. Efficacy and effectiveness of influenza vaccines: a systematic review and meta-analysis. Lancet Infect Dis 2012;12(1):36–44. [4] Puig-Barbera J, Diez-Domingo J, Arnedo-Pena A, et al. Effectiveness of the 2010-2011 seasonal influenza vaccine in preventing confirmed influenza hospitalizations in adults: a case–case comparison, case-control study. Vaccine 2012;30(39):5714–20. [5] Dominguez A, Castilla J, Godoy P, et al. Effectiveness of pandemic and seasonal influenza vaccines in preventing pandemic influenza-associated hospitalization. Vaccine 2012;30(38):5644–50. [6] Talbot HK, Griffin MR, Chen Q, Zhu Y, Williams JV, Edwards KM. Effectiveness of seasonal vaccine in preventing confirmed influenza-associated hospitalizations in community dwelling older adults. J Infect Dis 2011;203(4): 500–8. [7] Castilla J, Martinez-Artola V, Salcedo E, et al. Vaccine effectiveness in preventing influenza hospitalizations in Navarre Spain, 2010-2011: cohort and case-control study. Vaccine 2012;30(2):195–200. [8] Cheng AC, Kotsimbos T, Kelly HA, et al. Effectiveness of H1N1/09 monovalent and trivalent influenza vaccines against hospitalization with laboratoryconfirmed H1N1/09 influenza in Australia: a test-negative case control study. Vaccine 2011;29(43):7320–5. [9] Castilla J, Godoy P, Dominguez A, et al. Influenza vaccine effectiveness in preventing outpatient, inpatient, and severe cases of laboratory-confirmed influenza. Clin Infect Dis 2013;57(2):167–75. [10] Talbot HK, Zhu Y, Chen Q, Williams JV, Thompson MG, Griffin MR. Effectiveness of influenza vaccine for preventing laboratory-confirmed influenza hospitalizations in adults, 2011-2012 influenza season. Clin Infect Dis 2013;56(12):1774–7. [11] Belongia EA, Kieke BA, Donahue JG, et al. Effectiveness of inactivated influenza vaccines varied substantially with antigenic match from the 2004-2005 season to the 2006-2007 season. J Infect Dis 2009;199(2):159–67. [12] Belongia EA, Kieke BA, Donahue JG, et al. Influenza vaccine effectiveness in Wisconsin during the 2007-08 season: comparison of interim and final results. Vaccine 2011;29(38):6558–63.
457
[13] Treanor JJ, Talbot HK, Ohmit SE, et al. Effectiveness of seasonal influenza vaccines in the United States during a season with circulation of all three vaccine strains. Clin Infect Dis 2012;55(7):951–9. [14] Centers for Disease Control Prevention. Interim adjusted estimates of seasonal influenza vaccine effectiveness–United States February 2013. MMWR 2013;62:119–23. [15] DeStefano F, Eaker ED, Broste SK, et al. Epidemiologic research in an integrated regional medical care system: the Marshfield epidemiologic study area. J Clin Epidemiol 1996;49(6):643–52. [16] Greenlee RT. Measuring disease frequency in the Marshfield Epidemiologic Study Area (MESA). Clin Med Res 2003;1(4):273–80. [17] Greenlee RT, Naleway AL, Vidaillet H. Incidence of myocardial infarction in a general population: the Marshfield Epidemiologic Study Area. WMJ 2002;101(7):46–52. [18] Belongia EA, Chyou PH, Greenlee RT, Perez-Perez G, Bibb WF, DeVries EO. Diarrhea incidence and farm-related risk factors for Escherichia coli O157:H7 and Campylobacter jejuni antibodies among rural children. J Infect Dis 2003;187(9):1460–8. [19] Irving SA, Donahue JG, Shay DK, Ellis-Coyle TL, Belongia EA. Evaluation of selfreported and registry-based influenza vaccination status in a Wisconsin cohort. Vaccine 2009;27(47):6546–9. [20] D’Agostino Jr RB. Propensity score methods for bias reduction in the comparison of a treatment to a non-randomized control group. Stat Med 1998;17(19):2265–81. [21] D’Agostino Jr RB, D’Agostino Sr RB. Estimating treatment effects using observational data. JAMA 2007;297(3):314–6. [22] Rondy M, Puig-Barbera J, Launay O, et al. 2011-12 seasonal influenza vaccines effectiveness against confirmed A(H3N2) influenza hospitalisation: pooled analysis from a European network of hospitals. A pilot study. PloS One 2013;8(4):e59681. [23] Cheng AC, Holmes M, Irving LB, et al. Influenza vaccine effectiveness against hospitalisation with confirmed influenza in the 2010-11 seasons: a testnegative observational study. PloS One 2013;8(7):e68760. [24] Kwong JC, Campitelli MA, Gubbay JB, et al. Vaccine effectiveness against laboratory-confirmed influenza hospitalizations among elderly adults during the 2010-2011 season. Clin Infect Dis 2013;57(6):820–7. [25] Thompson M, Sokolow LZ, Almendares O, et al. Effectiveness of non-adjuvanted monovalent A(H1N1)pdm09 vaccines for preventing RT-PCR-confirmed pandemic influenza hospitalizations: case-control study of children and adults at ten U.S influenza surveillance network sites. Clin Infect Dis 2013;57:1587–92. [26] Centers for Disease Control Prevention. Influenza activity–United States and worldwide, 2007-08 season. MMWR 2008;57(25):692–7.
Contents lists available at ScienceDirect
Vaccine journal homepage: www.elsevier.com/locate/vaccine
Influenza vaccination and risk of hospitalization among adults with laboratory confirmed influenza illness夽 Huong Q. McLean ∗ , Jennifer K. Meece, Edward A. Belongia Marshfield Clinic Research Foundation, Marshfield, WI, USA
a r t i c l e
i n f o
Article history: Received 13 September 2013 Received in revised form 12 November 2013 Accepted 15 November 2013 Available online 26 November 2013 Keywords: Influenza vaccine Severity Hospitalization Effectiveness
a b s t r a c t Background: Influenza vaccine is moderately effective for preventing influenza illness. It is not known if vaccination reduces the risk of subsequent hospital admission among patients with vaccine failure and laboratory confirmed influenza illness. Methods: Patients in a community cohort presenting with acute respiratory illness were prospectively enrolled and tested for influenza during 8 seasons to estimate seasonal vaccine effectiveness. Hospital admissions within 14 days after illness onset were identified for all participants aged ≥20 years with laboratory confirmed influenza. The association between vaccination and hospital admission was examined in a propensity score adjusted logistic regression model. The model was validated by examining the association between vaccination and hospital admission in participants without influenza. Results: Influenza was identified in 1393 (28%) of 4996 participants. Sixty-two (6%) of 1020 with influenza A and 17 (5%) of 369 with influenza B were hospitalized. Vaccination was not associated with a reduced risk of hospital admission among all participants with influenza [adjusted odds ratio (aOR) = 1.08; 95% CI: 0.62, 1.88]; or among those with influenza A (aOR = 1.35; 95% CI: 0.71, 2.57) or influenza B (aOR = 0.67; 95% CI: 0.21, 2.15). Influenza vaccination was not associated with hospitalization after non-influenza respiratory illness (aOR = 1.14; 95% CI: 0.84, 1.54). Conclusions: Influenza vaccination did not reduce the risk of subsequent hospital admission among patients with vaccine failure. These findings do not support the hypothesis that vaccination mitigates influenza illness severity. © 2013 The Authors. Published by Elsevier Ltd. All rights reserved.
1. Introduction Influenza is an important cause of death and serious illness, particularly among adults aged ≥65 years and those with certain underlying chronic conditions. In the United States, approximately 226,000 hospital admissions are attributed to influenza each year [1]. As a result, annual influenza vaccination is recommended for all persons aged ≥6 months to prevent seasonal influenza infection and its complications [2]. However, influenza vaccine failure is common even during seasons with optimal antigenic match between circulating and vaccine viruses. Among adults, vaccine efficacy in preventing laboratory confirmed influenza illness is estimated to be approximately 60% [3]. Similar efficacy has been reported for preventing hospital admission with laboratory
夽 This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. ∗ Corresponding author at: Marshfield Clinic Research Foundation (ML2), 1000 N. Oak Ave, Marshfield, WI 54449, USA. Tel.: +1 715 389 3707; fax: +1 715 389 3880. E-mail address: mclean.huong@marshfieldclinic.org (H.Q. McLean).
confirmed pandemic or seasonal influenza [4–10]. It is not clear if influenza vaccination prevents serious outcomes by primary prevention of influenza infection, by reducing severity of influenza illness, or both. We conducted a population based study of laboratory confirmed influenza among adults aged ≥20 years over multiple seasons to determine if receipt of same-season influenza vaccine was associated with reduced risk of hospital admission within 14 days after onset of influenza illness. 2. Methods 2.1. Study population and design This was a secondary analysis of data from population-based studies of influenza vaccine effectiveness during eight influenza seasons, 2004–05 through 2012–13, in Marshfield, Wisconsin [11–14]. In this community, residents receive nearly all outpatient and inpatient care from the Marshfield Clinic. A single acute care hospital (St. Joseph’s) serves the study population, and both inpatient and outpatient diagnoses are accessible through a combined electronic medical record. The electronic medical record captures
0264-410X/$ – see front matter © 2013 The Authors. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.vaccine.2013.11.060
454
H.Q. McLean et al. / Vaccine 32 (2014) 453–457
90% of outpatient visits, 95% of hospital discharges, and 99% of deaths for the residents in the area [15–18]. During each influenza season, eligible community dwelling residents were recruited by trained research coordinators during or after an inpatient or outpatient medical encounter for acute respiratory illness. Research coordinators used an electronic appointment system to identify and recruit eligible persons in all primary care clinics and in urgent care on weekdays, evenings, and weekends. Eligible persons were also recruited at the hospital that is contiguous with Marshfield Clinic. Most ill persons who were not approached during a clinical encounter were identified on the following day by use of electronic diagnosis codes entered by attending physicians (ICD-9-CM codes 382.0, 382.4, 382.9, 460–466, 480, 483–486, 487, 490, 780.6, and 786.2). These individuals were contacted by telephone, and a swab sample was obtained at home from those who were eligible and consented. Participants completed a short interview to assess illness symptoms and onset date; nasopharyngeal swabs were obtained for influenza testing. Real-time reverse transcription polymerase chain reaction (RT-PCR) and viral cultures were performed at the Marshfield Clinic Research Foundation as previously described [11]. Culture alone was performed on samples collected in 2004–05 and RT-PCR was performed in subsequent years. Subtype results based on RT-PCR were not available for 11% of influenza A positive samples. For those samples, the subtype was assumed to be the predominant subtype identified among study participants during that season. This analysis excluded the 2009–10 season because monovalent vaccine was not available to the local population when the pandemic wave arrived in October–November 2009, and influenza was absent from the study population in the subsequent winter months. 2.2. Influenza vaccination status Influenza vaccination status was determined by a real-time, internet-based vaccination registry used by all public and private vaccination providers serving the population (http://www. recin.org). A validation study of the registry during the 2006–07 and 2007–08 influenza seasons demonstrated that the registry captured 95% of all influenza vaccinations that were received by study participants [19]. A similar high level of capture was demonstrated in a validation study during the 2011–12 season (unpublished data). Adults were classified as vaccinated if they had received influenza vaccine ≥14 days before the onset of illness. 2.3. Ascertainment of hospital admissions and clinical history Dates of hospital admission and discharge diagnoses were identified from the electronic medical record for a 14 day period after onset of influenza illness. To adjust for use of antiviral drugs, we extracted dates of antiviral prescriptions for all participants. 2.4. Outcome and covariates The main outcome was an acute care hospital admission occurring within 14 days of influenza symptom onset. Although most hospital admissions occurred after an outpatient enrollment, some participants were initially enrolled and swabbed after admission to the hospital. Covariates included age, gender, antiviral prescription, specific high risk medical conditions, year, and influenza type/subtype [A/H3N2, A/H1N1, pandemic H1N1 (A/H1N1pdm09), B]. Study participants were classified as having a high risk medical condition if they had at least one visit during a recent 12 month period with an ICD-9 CM diagnosis code of interest. High risk conditions were classified into the following groups: cancer, cardiovascular disease, diabetes, pulmonary, and other.
Antiviral prescription was defined as a prescription of oseltamivir, zanamivir, amantadine, or rimantadine within 14 days of symptom onset for persons not hospitalized and between symptom onset and hospital admission for persons who were hospitalized. 2.5. Analyses We restricted the analysis of hospital admissions to enrolled adults aged ≥20 years because influenza-related hospitalization was less common in children, and potential confounding factors are likely to be different for adults and children. Studies of influenza vaccination and hospital admission are particularly susceptible to confounding, since persons who are vaccinated may be more likely to have pre-existing chronic medical conditions or other risk factors for hospital admission. To minimize confounding by indication for vaccination, we used a propensity score regression adjustment [20,21]. Propensity scores allow adjustment for multiple potential confounders when the data are too sparse to include a separate effect for each potential confounder in the regression model. The propensity scores were generated from a multivariable logistic regression model that assessed the probability of influenza vaccination as a function of the potential confounders. In the propensity model, the dependent variable was influenza vaccination status and the independent variables were potential confounders identified a priori. The propensity score covariates included age, gender, cancer, cardiovascular disease, diabetes, pulmonary disorders, other high risk conditions, and year. The propensity scores from the model were then included as a continuous variable in the final logistic regression model that assessed the association between influenza vaccination and hospital admission. To determine the effect of influenza vaccination among persons with laboratory confirmed influenza, the final logistic regression model predicting hospital admission included the following covariates: propensity score, influenza vaccination, age group, influenza type/subtype, receipt of antiviral drug prescription. The primary analysis included all study participants with laboratory confirmed influenza. Secondary analyses included subgroups based on influenza type (A or B). We excluded the small number of participants with both A and B infection because the risk of hospitalization may be different for those co infected with both types and persons with unknown vaccination status. Since the primary outcome included all hospital admissions during a 14 day period, we performed a secondary analysis restricted to hospital admissions that were directly related to influenza infection. These included individuals who received any discharge diagnosis (among the top three diagnosis codes) for influenza, pneumonia, bronchitis, exacerbation of chronic pulmonary disease, or acute respiratory infection. In addition, one individual with a discharge diagnosis of fever was included in this group because symptoms of influenza like illness were present at the time of admission. We also performed an analysis restricted to persons who were enrolled in the outpatient setting and subsequently admitted to the hospital. Finally, we evaluated residual confounding by examining the association between influenza vaccination and hospital admission among study participants with a negative influenza test in a logistic regression model. The propensity scores for study participants with a negative influenza test (i.e., non-influenza respiratory illness) were generated using the same method as described above. If the propensity scores adequately adjusted for confounding, there should be no association between influenza vaccine receipt and hospital admission in that group. We assumed that confounders would be the same for influenza negative and influenza positive study participants.
H.Q. McLean et al. / Vaccine 32 (2014) 453–457 Table 1 Clinical and demographic characteristics and influenza vaccination status for adult study participants with laboratory confirmed influenza. Patient characteristics
Age 20–49 years 50–59 years 60–69 years 70–79 years ≥80 years Male Received antiviral prescription High risk conditions Cancer Cardiovascular Diabetes Pulmonary disordersa Other high riskb Season 2004–05 2005–06 2006–07 2007–08 2008–09 2010–11 2011–12 2012–13
Vaccinated n = 552 No. (%)
Unvaccinated n = 837 No. (%)
208 (38) 108 (20) 102 (18) 88 (16) 46 (8) 195 (35) 117 (21)
584 (70) 161 (19) 57 (7) 24 (3) 11 (1) 388 (46) 199 (24)
64 (12) 130 (23) 98 (18) 122 (22) 109 (20)
20 (2) 79 (9) 42 (5) 84 (10) 69 (8)
80 (14) 16 (3) 12 (2) 161 (29) 43 (8) 41 (7) 36 (7) 163 (30)
34 (4) 14 (2) 22 (3) 339 (41) 101 (12) 73 (9) 56 (7) 198 (24)
Table 2 Characteristics associated with influenza vaccination among adults hospitalized with confirmed influenza.
p
<0.0001
<0.001 0.3 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001
a Pulmonary disorders included pulmonary tuberculosis, other respiratory tuberculosis, other diseases due to mycobacteria, sarcoidosis, cystic fibrosis, alpha-1 anti-trypsin deficiency, chronic bronchitis, emphysema, asthma, bronchiectasis, extrinsic allergic alveolitis, chronic airway obstruction, and coal workers. b Kidney disease, liver disease, blood disorders, immunosuppressive disorders, metabolic disorders, and neurological/musculoskeletal disorders.
Unadjusted risk ratios were used to compare the risk of influenza vaccination among adults hospitalized with influenza. All analyses were performed using SAS 9.3 (SAS Institute Inc., Cary, NC); p-values <0.05 were considered significant. 3. Results During the 8 influenza seasons, 4996 adults with acute respiratory illness seeking medical care were enrolled. Influenza infection was laboratory confirmed for 1393 persons; 1020 (73%) had type A infection, 369 (26%) had type B infection, and 4 (<1%) were positive for both type A and B. Most (84%) influenza A infections were H3N2 subtype, followed by H1N1 (10%) and H1N1pdm09 (6%). The number of influenza A positive study participants ranged from 18 in the 2005–06 season to 356 in the 2007–08 season. The number of influenza B positive study participants ranged from 5 in the 2006–07 season to 144 in the 2007–08 season. Among persons with laboratory confirmed influenza and known vaccination status, 583 (42%) were males, 540 (39%) had at least one high risk condition, 316 (23%) were prescribed antiviral medications, and 31 (2%) were enrolled after admission to the hospital. The proportion vaccinated differed with respect to age, gender, and presence of high risk conditions (Table 1). In particular, influenza vaccination was more common in older adults and women. The median age was 55 years [interquartile range (IQR): 41, 69] among adults who were vaccinated and 41 years (IQR: 30, 52) among adults who were not vaccinated (p < 0.001). Vaccination was also more common among persons with cancer, cardiovascular disease, diabetes, pulmonary disorders, and other high risk conditions compared to those without these high risk conditions. Similar patterns were observed when examined by influenza type. Seventy-nine patients with laboratory confirmed influenza were admitted to the hospital within 14 days of symptom onset: 62 (6%) of 1020 with influenza A and 17 (5%) of 369 with influenza B.
455
Age 20–49 years 50–59 years 60–69 years 70–79 years ≥80 years Gender Male Female High risk conditions Cancer Yes No Cardiovascular Yes No Diabetes Yes No Pulmonary disorders Yes No Other high riska Yes No
Total persons No.
Persons vaccinated No. (%)
Risk ratio (95% CI)
21 10 15 14 19
6 (29) 6 (60) 9 (60) 11 (79) 15 (79)
1.0 2.10 (0.90, 4.89) 2.10 (0.95, 4.64) 2.75 (1.33, 5.70) 2.76 (1.35, 5.65)
34 45
21 (62) 26 (58)
1.07 (0.74, 1.54) 1.0
14 65
11 (79) 36 (55)
1.42 (1.00, 2.01) 1.0
29 50
22 (76) 25 (50)
1.52 (1.07, 2.14) 1.0
14 65
10 (71) 37 (57)
1.25 (0.85, 1.86) 1.0
29 50
19 (66) 28 (56)
1.17 (0.82, 1.68) 1.0
20 59
15 (75) 32 (54)
1.38 (0.98, 1.95) 1.0
a Kidney disease, liver disease, blood disorders, immunosuppressive disorders, metabolic disorders, and neurological/musculoskeletal disorders.
The median time from symptom onset to hospital admission was 3 days (IQR: 2–5 days). Seventy (89%) had discharge diagnoses codes that were consistent with an acute respiratory illness or exacerbation of chronic pulmonary disease. Among hospitalized patients, those who were older were more likely to be vaccinated compared to those aged 20–49 years and those with a cardiovascular high risk condition were more likely to be vaccinated compared to those without a cardiovascular high risk condition (Table 2). Vaccination status among hospitalized patients was not associated with gender or the other high risk conditions examined. Among patients with laboratory confirmed influenza, influenza vaccination was not associated with a decreased risk of hospitalization following onset overall or by influenza type (Table 3). The propensity score adjusted odd ratio of hospitalization for vaccinated compared to unvaccinated patients was 1.08 (95% CI: 0.62, 1.88), 1.35 (95% CI: 0.71, 2.57), and 0.67 (95% CI: 0.21, 2.15) overall, for type A infection, and for type B infection, respectively. The results were similar in secondary analyses where the outcome was hospital admission for acute respiratory illness rather than all hospital admissions and for analyses restricted to patients enrolled in the outpatient setting and subsequently hospitalized (Table 3). In the 3603 adults with non-influenza respiratory illness, there was no association between influenza vaccination and hospital admission within 14 days after illness onset (propensity score adjusted OR = 1.14; 95% CI: 0.84, 1.54; p = 0.4). 4. Discussion In this multi-season study, we examined the hypothesis that vaccination may mitigate influenza illness severity and reduce the risk of hospital admission. We found that vaccinated and unvaccinated individuals with influenza had a similar risk of hospital admission after adjustment for propensity to be vaccinated, regardless of influenza type. This suggests that influenza vaccination prevents serious outcomes by primary prevention of influenza infection.
456
H.Q. McLean et al. / Vaccine 32 (2014) 453–457
Table 3 Propensity score adjusted association between influenza vaccination status and hospital admission within 14 days after illness onset for study participants with laboratory confirmed influenza. Propensity scores incorporated factors associated with receipt of influenza vaccine, including age, gender, cancer, cardiovascular disease, diabetes, chronic pulmonary disease, other high risk conditions, and year. All hospital admissions within 14 days of illness onset
Any influenza virus Influenza A virus Influenza B virus
Admissions for acute respiratory illnessa
Admission after outpatient enrollmentb
No. cases (% vaccinated)
Vaccinated No. (%) hospitalized
Unvaccinated No. (%) hospitalized
Crude OR (95% CI)
Adjusted OR (95% CI)
Adjusted OR (95% CI)
Adjusted OR (95% CI)
1389 (40)
47 (9)
32 (4)
1020 (43)
40 (9)
22 (4)
369 (31)
7 (6)
10 (4)
2.34 (1.47, 3.72) 2.57 (1.50, 4.39) 1.58 (0.59, 4.27)
1.08c (0.62, 1.88) 1.35d (0.71, 2.57) 0.67 (0.21, 2.15)
0.95c (0.52, 1.70) 1.23d (0.62, 2.44) 0.54 (0.16, 1.81)
1.11c (0.56, 2.21) 1.21d (0.54, 2.71) 1.02 (0.28, 3.82)
a
Nine patients with hospital admission that did not appear to be associated with influenza infection were excluded. Thirty one patients were enrolled after hospital admission and were excluded. c Adjusted for propensity score, age, influenza type/subtype (A/H3N2, A/H1N1, A/H1N1pdm09, B), and receipt of antiviral drug prescription. d Adjusted for propensity score, age, influenza subtype (A/H3N2, A/H1N1, A/H1N1pdm09), and receipt of antiviral drug prescription. Adjusted for propensity score, age, and receipt of antiviral drug prescription. b
In the past decade, multiple observational studies of vaccine effectiveness have been performed using medically attended influenza (confirmed by RT-PCR) as the primary endpoint. Most of these studies have assessed vaccine effectiveness for preventing outpatient influenza illness, but few have focused on vaccine effectiveness for preventing hospitalization with laboratory confirmed influenza [4–10,22–25]. In these studies where the comparison groups were those without influenza, vaccine effectiveness estimates ranged from 25% to 74%. An important finding from these studies is that vaccination provides moderate benefit against influenza hospitalization, presumably due to primary prevention of influenza illness. To our knowledge, one other study has examined the association between vaccination and hospital admission among persons with influenza. Despite a different study population and most cases being caused by A/H1N1pdm09, they had similar findings to our study: vaccination did not reduce the risk of hospitalization [9]. Additionally, they found that hospitalized patients who were vaccinated were less likely to have had severe disease. However, because the study was observational, it is not possible to know whether this association was due to vaccination, residual confounding, or confounding from unmeasured factors. Due to the limited number of hospitalized cases in our study, we were unable to assess the impact of vaccination on severity of cases among those hospitalized. We attempted to minimize confounding with a propensity score that adjusted for the likelihood of influenza vaccination based on multiple covariates. The propensity score model was tested in study participants with non-influenza respiratory illness, since an association between vaccination and hospital admission is not biologically plausible in the absence of influenza. The model with propensity score adjustment showed no evidence of confounding in this group: the odds ratio for hospital admission in vaccinated versus unvaccinated adults with non-influenza illness was 1.1 (p = 0.4). This supported the validity of the propensity score model to examine the association between vaccination and hospitalization among patients with influenza. Strengths of this study included systematic recruitment and sample collection from a community cohort with medically attended acute respiratory illness, use of a highly sensitive and specific RT-PCR assay, access to a validated immunization registry, and complete capture of hospital admissions from the electronic medical record. However, several limitations should be acknowledged. First, hospitalization due to influenza is rare in healthy adult populations. Despite eight seasons, there were few hospitalizations in
our study, all of which were from a single hospital in central Wisconsin. Second, antigenic characterization was not performed for many positive samples, and minor antigenic drift can be difficult to detect and interpret. As a result, we were not able to assess the potential impact of antigenic variability. The 2007–08 season accounted for the majority of A (H3N2) infections, and during that year there was circulation of A/Brisbane/10/2007-like viruses that were minor antigenic variants from the vaccine strain [26]. Third, our classification of high risk medical conditions was based on ICD-9 diagnosis codes without medical record validation. However, all diagnoses were entered by physicians and automatically mapped to ICD-9 codes in the electronic medical record, which reduced the potential for coding error. Finally, our study population included primarily outpatient influenza cases and there may have been differential health care seeking behavior between vaccinated and unvaccinated individuals. We cannot exclude the possibility that vaccinated individuals had milder influenza illness and did not seek medical attention. In that scenario, vaccination would have reduced illness severity, leading to fewer outpatient visits and hospitalizations, but this would not be evident when comparing the risk of hospitalization in vaccinated and unvaccinated outpatients. However, we note that estimates of vaccine effectiveness in the outpatient setting are generally similar to estimates of efficacy based on randomized clinical trials, and the primary endpoint for clinical trials is influenza illness rather than severity. Because of these limitations, results should be interpreted with caution. Hospitalization is an important complication of influenza infection from a public health and an economic perspective. Available evidence suggests that influenza vaccine provides moderate protection against influenza-related hospitalization. Further research is warranted to assess the impact of vaccination in preventing severe outcomes among vaccine failures, including differences by type, subtype, and lineage.
Acknowledgements We thank the following individuals for their contribution to this work: Burney Kieke, Sarah Kopitzke, Pam Squires, Jim Donahue, Stephanie Irving, David Shay, and Alicia Fry. Conflicts of interest: HQM, JKM, and EAB receive research funding from MedImmune, LLC. Funding: Centers for Disease Control and Prevention(cooperative agreement U18 IP000183).
H.Q. McLean et al. / Vaccine 32 (2014) 453–457
References [1] Thompson WW, Shay DK, Weintraub E, et al. Influenza-associated hospitalizations in the United States. JAMA 2004;292(11):1333–40. [2] Centers for Disease Control Prevention. Prevention and control of influenza with vaccines: recommendations of the Advisory Committee on Immunization Practices (ACIP)–United States, 2012-13 influenza season. MMWR 2012;61(32):613–8. [3] Osterholm MT, Kelley NS, Sommer A, Belongia EA. Efficacy and effectiveness of influenza vaccines: a systematic review and meta-analysis. Lancet Infect Dis 2012;12(1):36–44. [4] Puig-Barbera J, Diez-Domingo J, Arnedo-Pena A, et al. Effectiveness of the 2010-2011 seasonal influenza vaccine in preventing confirmed influenza hospitalizations in adults: a case–case comparison, case-control study. Vaccine 2012;30(39):5714–20. [5] Dominguez A, Castilla J, Godoy P, et al. Effectiveness of pandemic and seasonal influenza vaccines in preventing pandemic influenza-associated hospitalization. Vaccine 2012;30(38):5644–50. [6] Talbot HK, Griffin MR, Chen Q, Zhu Y, Williams JV, Edwards KM. Effectiveness of seasonal vaccine in preventing confirmed influenza-associated hospitalizations in community dwelling older adults. J Infect Dis 2011;203(4): 500–8. [7] Castilla J, Martinez-Artola V, Salcedo E, et al. Vaccine effectiveness in preventing influenza hospitalizations in Navarre Spain, 2010-2011: cohort and case-control study. Vaccine 2012;30(2):195–200. [8] Cheng AC, Kotsimbos T, Kelly HA, et al. Effectiveness of H1N1/09 monovalent and trivalent influenza vaccines against hospitalization with laboratoryconfirmed H1N1/09 influenza in Australia: a test-negative case control study. Vaccine 2011;29(43):7320–5. [9] Castilla J, Godoy P, Dominguez A, et al. Influenza vaccine effectiveness in preventing outpatient, inpatient, and severe cases of laboratory-confirmed influenza. Clin Infect Dis 2013;57(2):167–75. [10] Talbot HK, Zhu Y, Chen Q, Williams JV, Thompson MG, Griffin MR. Effectiveness of influenza vaccine for preventing laboratory-confirmed influenza hospitalizations in adults, 2011-2012 influenza season. Clin Infect Dis 2013;56(12):1774–7. [11] Belongia EA, Kieke BA, Donahue JG, et al. Effectiveness of inactivated influenza vaccines varied substantially with antigenic match from the 2004-2005 season to the 2006-2007 season. J Infect Dis 2009;199(2):159–67. [12] Belongia EA, Kieke BA, Donahue JG, et al. Influenza vaccine effectiveness in Wisconsin during the 2007-08 season: comparison of interim and final results. Vaccine 2011;29(38):6558–63.
457
[13] Treanor JJ, Talbot HK, Ohmit SE, et al. Effectiveness of seasonal influenza vaccines in the United States during a season with circulation of all three vaccine strains. Clin Infect Dis 2012;55(7):951–9. [14] Centers for Disease Control Prevention. Interim adjusted estimates of seasonal influenza vaccine effectiveness–United States February 2013. MMWR 2013;62:119–23. [15] DeStefano F, Eaker ED, Broste SK, et al. Epidemiologic research in an integrated regional medical care system: the Marshfield epidemiologic study area. J Clin Epidemiol 1996;49(6):643–52. [16] Greenlee RT. Measuring disease frequency in the Marshfield Epidemiologic Study Area (MESA). Clin Med Res 2003;1(4):273–80. [17] Greenlee RT, Naleway AL, Vidaillet H. Incidence of myocardial infarction in a general population: the Marshfield Epidemiologic Study Area. WMJ 2002;101(7):46–52. [18] Belongia EA, Chyou PH, Greenlee RT, Perez-Perez G, Bibb WF, DeVries EO. Diarrhea incidence and farm-related risk factors for Escherichia coli O157:H7 and Campylobacter jejuni antibodies among rural children. J Infect Dis 2003;187(9):1460–8. [19] Irving SA, Donahue JG, Shay DK, Ellis-Coyle TL, Belongia EA. Evaluation of selfreported and registry-based influenza vaccination status in a Wisconsin cohort. Vaccine 2009;27(47):6546–9. [20] D’Agostino Jr RB. Propensity score methods for bias reduction in the comparison of a treatment to a non-randomized control group. Stat Med 1998;17(19):2265–81. [21] D’Agostino Jr RB, D’Agostino Sr RB. Estimating treatment effects using observational data. JAMA 2007;297(3):314–6. [22] Rondy M, Puig-Barbera J, Launay O, et al. 2011-12 seasonal influenza vaccines effectiveness against confirmed A(H3N2) influenza hospitalisation: pooled analysis from a European network of hospitals. A pilot study. PloS One 2013;8(4):e59681. [23] Cheng AC, Holmes M, Irving LB, et al. Influenza vaccine effectiveness against hospitalisation with confirmed influenza in the 2010-11 seasons: a testnegative observational study. PloS One 2013;8(7):e68760. [24] Kwong JC, Campitelli MA, Gubbay JB, et al. Vaccine effectiveness against laboratory-confirmed influenza hospitalizations among elderly adults during the 2010-2011 season. Clin Infect Dis 2013;57(6):820–7. [25] Thompson M, Sokolow LZ, Almendares O, et al. Effectiveness of non-adjuvanted monovalent A(H1N1)pdm09 vaccines for preventing RT-PCR-confirmed pandemic influenza hospitalizations: case-control study of children and adults at ten U.S influenza surveillance network sites. Clin Infect Dis 2013;57:1587–92. [26] Centers for Disease Control Prevention. Influenza activity–United States and worldwide, 2007-08 season. MMWR 2008;57(25):692–7.