Abstract
Elevated body mass index (BMI) has been linked to severe influenza illness and impaired vaccine immunogenicity, but the relationship between BMI and clinical vaccine effectiveness (VE) is less well described. This secondary analysis of data from a test-negative study of outpatients with acute respiratory illness assessed BMI and VE against medically attended, PCR-confirmed influenza over seven seasons (2011–12 through 2017–18). Vaccination status was determined from electronic medical records (EMR) and self-report; BMI was estimated from EMR-documented height and weight categorized for adults as obesity (≥ 30 kg/m2), overweight (25–29 kg/m2), or normal and for children based on standardized z-scales. Current season VE by virus type/subtype was estimated separately for adults and children. Pooled VE for all seasons was calculated as 1—adjusted odds ratios from logistic regression with an interaction term for BMI and vaccination. Among 28,089 adults and 12,380 children, BMI category was not significantly associated with VE against outpatient influenza for any type/subtype. Adjusted VE against A/H3N2, A/H1N1pdm09, and B in adults ranged from 16–31, 46–54, and 44–57%, and in children from 29–34, 57–65, and 50–55%, respectively, across the BMI categories. Elevated BMI was not associated with reduced VE against laboratory confirmed, outpatient influenza illness.
Keywords: Vaccine effectiveness, Body mass index, Influenza, Obesity
Subject terms: Influenza virus, Preventive medicine, Body mass index
Introduction
According to the World Health Organization, about 16% of adults aged 18 years and older worldwide had obesity in 2022. The worldwide prevalence of obesity more than doubled between 1990 and 20221. In the United States, obesity affects over 40% of adults and nearly 20% of children2,3. Obesity is associated with metabolic syndrome and other comorbid conditions, and it may contribute to more severe influenza illness in children and adults4–11.
The effects of obesity on the immune response to influenza vaccination are complex and poorly understood. Post-vaccination hemagglutinin inhibition (HI) titers to influenza antigens are largely similar in individuals with obesity and normal weight12–14. However, antibody titers appear to decline more rapidly in people with higher body mass index (BMI), and aspects of the cell-mediated immune response may be inhibited13,15–17.
The relationship between BMI and clinical vaccine effectiveness (VE) is even less clear as few studies have addressed this question. A prospective observational study reported similar protection against PCR-confirmed influenza illness in vaccinated children with and without obesity18. While there are no published VE estimates comparing adults with and without obesity, a two-season prospective study in vaccinated adults found that influenza or influenza-like illness (ILI) was associated with obesity despite similar levels of seroprotective antibodies19. However, the study was limited by a small sample size, nonspecific endpoint (ILI), likely residual confounding, and lack of an unvaccinated comparison group.
To further elucidate the relationship between obesity and influenza VE in the outpatient setting, we analyzed data from the US influenza (Flu) VE network over seven seasons.
Methods
This analysis included data from the Flu VE Network from 2011–12 through 2017–1820. Details of the study population and procedures have been previously described21,22. In brief, patients seeking medical care for an acute respiratory illness (ARI) with cough and illness of ≤ 7 days duration were recruited from outpatient, urgent care, and emergency department facilities in Michigan, Pennsylvania, Texas, Washington, and Wisconsin. The institutional review boards (IRB) at the Centers for Disease Control and Prevention (CDC) and all participating sites (Marshfield Clinic Health System IRB, Group Health Cooperative Human Subjects Review Committee/Kaiser Permanente-Washington Region IRB, Baylor Scott & White Research Institute IRB, University of Michigan IRBMED, and University of Pittsburgh Human Research Protection Office) approved the study. The study was performed in accordance with Health and Human Services regulations for the protection of human subjects in research (specifically 45 CFR 46, 21 CFR 50, and 21 CFR 56). Participants/legal guardians provided informed consent prior to participating. Consenting participants/legal guardians completed an enrollment interview to ascertain demographics, symptoms, onset date, current season vaccination, and subjective assessments of general and current health status. Anterior or mid-turbinate nasal and throat swabs were collected from participants by research staff and tested for influenza type and subtype by real-time reverse transcription polymerase chain reaction (rRT-PCR) at network sites. Presence of underlying medical conditions was defined as one or more International Classification of Diseases, Ninth/Tenth Revision (ICD-9/10) codes diagnosed for medical encounters in the previous year.
Influenza vaccination status
Participants with electronic medical record (EMR) documented vaccine receipt of ≥ 1 dose of current season influenza vaccine or undocumented but plausible self- or parent-reported vaccination 14 or more days before symptom onset were classified as vaccinated. Plausible self-report was determined by sites as receipt of vaccination during times influenza vaccination is typically offered and from providers (e.g. Veterans Affairs, Indian Health Service) not routinely reported to immunization information systems. Participants vaccinated < 14 days before onset were excluded from analysis.
Body mass index (BMI)
Height and weight measurements closest in time to enrollment were extracted from EMR. BMI in adults was grouped into categories according to National Heart, Lung and Blood Institute (NHLBI) definitions: < 18.5 kg/m2 (underweight), 18.5 to < 25 kg/m2 (normal, reference category), 25 to < 30 kg/m2 (overweight), and ≥ 30 kg/m2 (obesity)23,24. For children and adolescents, the publicly available SAS Program for the 2000 CDC Growth Charts was used to calculate BMI category for children 24 months (2 years) to 17 years of age based on percentile ranges for BMI by age and sex, with < 5th percentile, underweight, 5 to < 85th percentile, normal (reference category), 85 to < 95th percentile, overweight and ≥ 95th percentile, obesity25,26.
Height and weight data were only included if within plausible ranges. For adults ≥ 18 years of age, height ≥ 44 inches and ≤ 90 inches, weight ≥ 55 pounds and ≤ 1000 pounds, and BMI ≥ 12 and ≤ 70 were considered plausible27. For children, plausible values for height, weight, and BMI were defined by z scores calculated from the percentile of weight-, height- and BMI-for-age (in months) scores of the child25. We excluded pregnant women, underweight individuals, adults with weight measurements obtained more than 12 months from enrollment date, and children with height or weight measurements more than 6 months from enrollment date. Analyses were restricted to children aged ≥ 24 months.
Analytic approach
Influenza cases were persons with rRT-PCR positive tests for influenza A/H3N2, A/H1N1pdm09, or B viruses. Participants with unsubtypable influenza A or inconclusive results or with co-infections with multiple influenza viruses were excluded. Controls were persons testing rRT-PCR negative for all influenza subtypes/types. The analysis period was defined at each site, for each season, and for each type/subtype. The period was determined as the calendar week of enrollment from the first rRT-PCR positive influenza case of the specific type/subtype at that site through the last confirmed case of that type/subtype at that site. Seasons with 50 or more virus type/subtype specific cases were pooled; sites with 10 or more virus type/subtype specific cases were included in pooled analyses for the respective seasons and type/subtype.
We generated multivariable logistic regression models with influenza case status as the outcome. VE was estimated as (1-aOR) × 100, where aOR is the adjusted odds ratio for influenza among vaccinated compared with unvaccinated persons28,29. Vaccination status, age, network site, and enrollment season, were included in all adjusted models a priori. Age was modeled as a linear tail-restricted natural cubic spline for adults and as categories (2–8 years and 9–17 years) for children and adolescents. The pediatric age cut off was selected for consistency with influenza vaccine recommendations30. Records with missing data for influenza result, BMI, or model covariates were not included in the final analyses. Additional covariates were evaluated and retained in the model if the coefficient for the vaccination status differed by ≥ 10%. Covariates examined included: calendar time relative to the season peak at each site and type/subtype; time from illness onset to enrollment; presence of ≥ 1 underlying medical condition other than obesity; sex; self-reported health status; race/ethnicity; and smoking status (in adults only). We evaluated modification of VE by participant BMI by including an interaction term in models. Evidence of interaction was defined a priori at p < 0.10. Separate models were generated for children and adults, and A/H3N2, A/H1N1pdm09, and type B associated illness.
We used models restricted to vaccinated participants, with odds of influenza as the outcome and BMI as the exposure, to assess differential odds of breakthrough illness among vaccinated participants with obesity versus those with normal BMI as the reference group. Time from vaccination to illness onset (< 90 days vs ≥ 90 days) was included as a covariate in addition to the other a priori covariates to account for waning vaccine effects.
Sensitivity analyses
For A/H3N2, the 2014–15 season was excluded in sensitivity analyses because circulating A/H3N2 viruses were not antigenically matched to the vaccine virus. Among adults, a BMI of ≥ 40 (class 3 obesity) is currently used to define individuals at increased risk for severe outcomes from influenza so we performed an additional sensitivity analysis creating a four level BMI term by dividing the obesity category into obesity (30 to < 40) and class 3 obesity (BMI ≥ 40). To assess potential differences among older adults, sensitivity analyses were restricted to adults ≥ 65 years of age.
Results
From 2011–12 through 2017–18, 31 647 adults were enrolled in the Flu VE Network, of whom 28 089 were included in this analysis (Supplemental Fig. 1). Among eligible adults, 6 259 (22%) had a normal weight BMI, 8 338 (30%) had an overweight BMI, and 13 492 (48%) had an obese BMI (Table 1). Among those with obese BMI, 25% (12% of adults overall) would be considered class 3 obesity with a BMI ≥ 40. Most participants were non-Hispanic White (80%), 26% had an underlying medical condition other than obesity, and 54% were vaccinated with 37% of those receiving vaccine less than 90 days prior to illness onset. 9% of vaccination dates were determined from self-report (data not shown). Compared to individuals of normal weight, a greater percentage of participants with obesity were non-Hispanic Black, or Hispanic, had an underlying medical condition, and had received the current season influenza vaccine. A higher percentage of adult participants with BMI in normal ranges reported excellent/very good general health compared to individuals with BMI in the obesity range.
Table 1.
Characteristics of adults categorized by body mass index, 2011–12 through 2017–18.
| BMI category* | ||||
|---|---|---|---|---|
| Total (n = 28089) | Normal weight (n = 6259, 22%) | Overweight (n = 8338, 30%) | Obesity (n = 13492, 48%) | |
| Age, mean (SD) | 48.2 (17.6) | 43.8 (19.5) | 50.1 (18.2) | 49.1 (16.0) |
| Site | ||||
| Michigan | 3941 (14) | 808 (13) | 1159 (14) | 1974 (15) |
| Pennsylvania | 5511 (20) | 1442 (23) | 1675 (20) | 2394 (18) |
| Texas | 5155 (18) | 954 (15) | 1459 (18) | 2742 (20) |
| Washington | 7829 (28) | 2040 (33) | 2487 (30) | 3302 (24) |
| Wisconsin | 5653 (20) | 1015 (16) | 1558 (19) | 3080 (23) |
| Male | 10365 (37) | 1840 (29) | 3649 (44) | 4876 (36) |
| Race | ||||
| Non-hispanic white | 22490 (80) | 5063 (81) | 6755 (81) | 10672 (79) |
| Non-hispanic black | 1754 (6) | 227 (4) | 431 (5) | 1096 (8) |
| Non-hispanic, other race | 2050 (7) | 638 (10) | 629 (8) | 783 (6) |
| Hispanic, any race | 1716 (6) | 308 (5) | 496 (6) | 912 (7) |
| Missing | 23 (0) | 27 (0) | 29 (0) | 79 (0) |
| Underlying medical condition other than obesity | 7245 (26) | 1069 (17) | 1981 (24) | 4195 (31) |
| Self-reported general health status | ||||
| Excellent/very good | 17147 (61) | 4546 (73) | 5632 (68) | 6969 (52) |
| Good | 8310 (30) | 1308 (21) | 2145 (26) | 4857 (36) |
| Fair/poor | 2601 (9) | 395 (6) | 553 (7) | 1653 (12) |
| Refused | 31 (0) | 10 (0) | 8 (0) | 13 (0) |
| Vaccinated | 15285 (54) | 3170 (51) | 4640 (56) | 7475 (55) |
| < 90 days before onset | 5596 (37) | 1167 (37) | 1749 (38) | 2680 (36) |
| Influenza test result | ||||
| Negative | 20747 (74) | 4555 (73) | 6044 (72) | 10148 (75) |
| A/H3N2 | 4008 (14) | 925 (15) | 1247 (15) | 1836 (14) |
| A/H1N1pdm09 | 1461 (5) | 363 (6) | 454 (5) | 644 (5) |
| B | 1873 (7) | 416 (7) | 593 (7) | 864 (6) |
| Onset to enrollment interval | ||||
| 0–2 days | 8170 (29) | 1823 (29) | 2340 (28) | 4007 (30) |
| 3–4 days | 11103 (40) | 2431 (39) | 3338 (40) | 5334 (40) |
| 5–7 days | 8816 (31) | 2005 (32) | 2660 (32) | 4151 (31) |
| Season | ||||
| 2011–12 | 1958 (7) | 429 (7) | 575 (7) | 954 (7) |
| 2012–13 | 3356 (12) | 754 (12) | 1053 (13) | 1549 (11) |
| 2013–14 | 3616 (13) | 814 (13) | 1123 (13) | 1679 (12) |
| 2014–15 | 5296 (19) | 1181 (19) | 1575 (19) | 2540 (19) |
| 2015–16 | 4075 (15) | 994 (16) | 1160 (14) | 1921 (14) |
| 2016–17 | 4449 (16) | 920 (15) | 1334 (16) | 2195 (16) |
| 2017–18 | 5339 (19) | 1167 (19) | 1518 (18) | 2654 (20) |
Data presented as number (n), column percent (%) unless otherwise noted.
BMI body mass index, SD standard deviation.
*BMI categories: Normal (18.5 to < 25 kg/m2), Overweight (25 to < 30 kg/m2), Obesity (≥ 30 kg/m2).
Compared to non-influenza controls, among adult participants lower percentages of influenza case participants had BMI in the obese range, were male, less likely to rank their general health as poor to fair and to have received current season influenza vaccine (Table 2). Across all types/subtypes, controls were more likely to be enrolled later in illness (5–7 days from onset) compared to cases.
Table 2.
Characteristics of cases and controls by influenza type in adults, 2011–12 through 2017–18*.
| A/H3N2 | A/H1N1pdm09 | B | ||||
|---|---|---|---|---|---|---|
| Controls (n = 13800) | Cases (n = 3919) | Controls (n = 9143) | Cases (n = 1361) | Controls (n = 14632) | Cases (n = 1804) | |
| BMI† | ||||||
| Normal weight | 2963 (21) | 903 (23) | 2052 (22) | 341 (25) | 3201 (22) | 399 (22) |
| Overweight | 4025 (29) | 1217 (31) | 2623 (29) | 417 (31) | 4246 (29) | 569 (32) |
| Obesity | 6812 (49) | 1799 (46) | 4468 (49) | 603 (44) | 7185 (49) | 836 (46) |
| Age, mean (SD) | 48.4 (17.6) | 50.2 (18.2) | 47.9 (17.9) | 45.8 (15.2) | 47.9 (17.8) | 50.0 (16.0) |
| Site | ||||||
| Michigan | 2310 (17) | 661 (17) | 995 (11) | 164 (12) | 1660 (11) | 199 (11) |
| Pennsylvania | 2244 (16) | 852 (22) | 1958 (21) | 433 (32) | 2584 (18) | 331 (18) |
| Texas | 2434 (18) | 600 (15) | 1644 (18) | 156 (11) | 3231 (22) | 354 (20) |
| Washington | 3961 (29) | 848 (22) | 2844 (31) | 246 (18) | 4404 (30) | 451 (25) |
| Wisconsin | 2851 (21) | 958 (24) | 1702 (19) | 362 (27) | 2753 (19) | 469 (26) |
| Male | 4999 (36) | 1531 (39) | 3128 (34) | 580 (43) | 5124 (35) | 763 (42) |
| Race | ||||||
| Non-hispanic white | 11008 (80) | 3175 (81) | 7296 (80) | 1099 (81) | 11694 (80) | 1472 (81) |
| Non-hispanic black | 853 (6) | 234 (6) | 546 (6) | 99 (7) | 813 (6) | 103 (6) |
| Non-hispanic, other race | 1068 (8) | 270 (7) | 707 (8) | 89 (7) | 1133 (8) | 117 (7) |
| Hispanic, any race | 842 (6) | 223 (6) | 566 (6) | 70 (5) | 961 (7) | 107 (6) |
| Missing | 29 (0) | 17 (0) | 28 (0) | 4 (0) | 31 (0) | 5 (0) |
| Underlying medical condition other than obesity | 3827 (28) | 1104 (28) | 2515 (28) | 217 (16) | 4102 (28) | 474 (26) |
| Self-reported general health status | ||||||
| Excellent/very good | 8232 (60) | 2483 (63) | 5529 (60) | 914 (67) | 8814 (60) | 1198 (66) |
| Good | 4212 (31) | 1122 (29) | 2774 (30) | 340 (25) | 4444 (30) | 466 (26) |
| Fair/poor | 1344 (10) | 305 (8) | 833 (9) | 106 (8) | 1362 (9) | 139 (8) |
| Refused | 12 (0) | 9 (0) | 7 (0) | 1 (0) | 12 (0) | 1 (0) |
| Vaccinated | 7968 (58) | 2093 (53) | 5148 (56) | 498 (37) | 8358 (57) | 761 (42) |
| < 90 days before onset | 2961 (37) | 895 (43) | 1871 (36) | 176 (35) | 2816 (34) | 175 (23) |
| Calendar time‡ | ||||||
| Pre-peak | 3384 (25) | 760 (19) | 2602 (28) | 275 (20) | 6196 (42) | 324 (18) |
| Peak | 4636 (34) | 2345 (60) | 3379 (37) | 827 (61) | 6188 (42) | 1112 (62) |
| Post-peak | 5780 (42) | 814 (21) | 3162 (34) | 259 (19) | 2248 (15) | 368 (20) |
| Onset to enrollment interval | ||||||
| 0–2 days | 3512 (25) | 1746 (45) | 2324 (25) | 633 (47) | 3759 (26) | 525 (29) |
| 3–4 days | 5510 (40) | 1506 (38) | 3533 (39) | 496 (36) | 5747 (39) | 803 (45) |
| 5–7 days | 4778 (35) | 667 (17) | 3286 (36) | 232 (17) | 5126 (35) | 476 (26) |
| Season | ||||||
| 2011–12 | 1281 (9) | 176 (4) | 0 (0) | 0 (0) | 892 (6) | 51 (3) |
| 2012–13 | 2237 (16) | 738 (19) | 0 (0) | 0 (0) | 1551 (11) | 341 (19) |
| 2013–14 | 0 (0) | 0 (0) | 2667 (29) | 734 (54) | 0 (0) | 0 (0) |
| 2014–15 | 3782 (27) | 1071 (27) | 0 (0) | 0 (0) | 3762 (26) | 226 (13) |
| 2015–16 | 0 (0) | 0 (0) | 3147 (34) | 468 (34) | 2092 (14) | 218 (12) |
| 2016–17 | 3172 (23) | 868 (22) | 0 (0) | 0 (0) | 2961 (20) | 360 (20) |
| 2017–18 | 3328 (24) | 1066 (27) | 3329 (36) | 159 (12) | 3374 (23) | 608 (34) |
Data presented as number (n), column percent (%) unless otherwise noted.
BMI body mass index, SD standard deviation.
*Cases and controls included in this table are from seasons with 50 or more cases of the specific virus and sites that had 10 or more cases of the specific virus.
†BMI categories: Normal (18.5 to < 25 kg/m2), Overweight (25 to < 30 kg/m2), Obesity (≥ 30 kg/m2).
‡Peak defined as the weeks during which the middle 50% of all cases of the type/subtype occurred during the season at each site.
From 2011–12 through 2017–18, 16 636 children were enrolled in the Flu VE Network, of whom 12 380 were included in this analysis (Supplemental Fig. 1). Among eligible children 7 860 (63%) had a normal range BMI, 2 018 (16%) had an overweight range BMI, and 2 502 (20%) had an obese range BMI. Most were non-hispanic white (64%), 4% had an underlying medical condition other than obesity in the previous year, 45% were vaccinated, 18% of vaccinated children received live attenuated influenza vaccine (LAIV) and 41% of vaccinated children received vaccine less than 90 days prior to illness onset (Table 3). Less than 1% of vaccination dates were determined from self-report (data not shown). Children with obese range BMI were more frequently enrolled from the Texas site, non-Hispanic Black or Hispanic, female, and 9–17 years of age compared with children of normal weight and overweight range. Children with obesity were more likely to have an underlying medical condition and less likely to report excellent or very good health status or be vaccinated compared to children with normal weight and children with overweight BMI. There were multiple demographic and clinical differences between pediatric influenza cases and controls (Table 4). Across all influenza types, controls were more likely to be vaccinated.
Table 3.
Characteristics of children and adolescents categorized by body mass index, 2011–12 through 2017–18.
| BMI Category* | ||||
|---|---|---|---|---|
| Total (n = 12380) | Normal weight (n = 7860, 63%) | Overweight (n = 2018, 16%) | Obesity (n = 2502, 20%) | |
| Age category | ||||
| 2–8 Years | 7182 (58) | 4837 (62) | 1139 (56) | 1206 (48) |
| 9–17 years | 5198 (42) | 3023 (38) | 879 (44) | 1296 (52) |
| Site | ||||
| Michigan | 2448 (20) | 1596 (20) | 398 (20) | 454 (18) |
| Pennsylvania | 1539 (12) | 957 (12) | 267 (13) | 315 (13) |
| Texas | 3199 (26) | 1861 (24) | 501 (25) | 837 (33) |
| Washington | 1886 (15) | 1289 (16) | 317 (16) | 280 (11) |
| Wisconsin | 3308 (27) | 2157 (27) | 535 (27) | 616 (25) |
| Male | 6335 (51) | 3965 (50) | 1009 (50) | 1361 (54) |
| Race | ||||
| Non-hispanic white | 7938 (64) | 5252 (67) | 1303 (65) | 1383 (55) |
| Non-hispanic black | 1232 (10) | 706 (9) | 197 (10) | 329 (13) |
| Non-hispanic, other race | 1344 (11) | 860 (11) | 220 (11) | 264 (11) |
| Hispanic, any race | 1844 (15) | 1027 (13) | 295 (15) | 522 (21) |
| Missing | 22 (0) | 15 (0) | 3 (0) | 4 (0) |
| Underlying medical condition other than obesity | 493 (4) | 282 (4) | 73 (4) | 138 (6) |
| Self-reported general health status | ||||
| Excellent/very good | 10272 (83) | 6674 (85) | 1693 (84) | 1905 (76) |
| Good | 1741 (14) | 989 (13) | 266 (13) | 486 (19) |
| Fair/poor | 356 (3) | 189 (2) | 56 (3) | 111 (4) |
| Refused | 11 (0) | 8 (0) | 3 (0) | 0 (0) |
| Vaccinated | 5517 (45) | 3557 (45) | 917 (45) | 1043 (42) |
| < 90 days before onset | 2252 (41) | 1441 (41) | 383 (42) | 428 (41) |
| Received LAIV | 973 (18) | 676 (19) | 159 (17) | 138 (13) |
| Influenza test result | ||||
| Negative | 9055 (73) | 5647 (72) | 1502 (74) | 1906 (76) |
| A/H3N2 | 1842 (15) | 1247 (16) | 281 (14) | 314 (13) |
| A/H1N1pdm09 | 388 (3) | 266 (3) | 58 (3) | 64 (3) |
| B | 1095 (9) | 700 (9) | 177 (9) | 218 (9) |
| Onset to enrollment interval | ||||
| 0–2 days | 5087 (41) | 3180 (40) | 862 (43) | 1045 (42) |
| 3–4 days | 4670 (38) | 2990 (38) | 724 (36) | 956 (38) |
| 5–7 days | 2623 (21) | 1690 (22) | 432 (21) | 501 (20) |
| Season | ||||
| 2011–12 | 1303 (11) | 827 (11) | 226 (11) | 250 (10) |
| 2012–13 | 1802 (15) | 1116 (14) | 272 (13) | 414 (17) |
| 2013–14 | 1246 (10) | 809 (10) | 199 (10) | 238 (10) |
| 2014–15 | 2680 (22) | 1715 (22) | 446 (22) | 519 (21) |
| 2015–16 | 1410 (11) | 890 (11) | 217 (11) | 303 (12) |
| 2016–17 | 1832 (15) | 1153 (15) | 315 (16) | 364 (15) |
| 2017–18 | 2107 (17) | 1350 (17) | 343 (17) | 414 (17) |
BMI body mass index, LAIV live attenuated influenza vaccine.
*BMI categories: Normal (BMI percent 5 to < 85), Overweight (BMI percent 85 to < 95), Obesity (BMI percent ≥ 95).
Table 4.
Characteristics of cases and controls by influenza type in children and adolescents, 2011–12 through 2017–18*.
| A/H3N2 | A/H1N1pdm09 | B | ||||
|---|---|---|---|---|---|---|
| Controls (n = 6257) | Cases (n = 1809) | Controls (n = 2936) | Cases (n = 325) | Controls (n = 5636) | Cases (n = 1034) | |
| BMI† | ||||||
| Normal | 3890 (62) | 1223 (68) | 1825 (62) | 228 (70) | 3448 (61) | 660 (64) |
| Overweight | 1060 (17) | 278 (15) | 474 (16) | 45 (14) | 949 (17) | 168 (16) |
| Obesity | 1307 (21) | 308 (17) | 637 (22) | 52 (16) | 1239 (22) | 206 (20) |
| Age category | ||||||
| 2–8 years | 3814 (61) | 887 (49) | 1738 (59) | 226 (70) | 3337 (59) | 494 (48) |
| 9–17 years | 2443 (39) | 922 (51) | 1198 (41) | 99 (30) | 2299 (41) | 540 (52) |
| Site | ||||||
| Michigan | 1478 (24) | 473 (26) | 413 (14) | 52 (16) | 757 (13) | 218 (21) |
| Pennsylvania | 708 (11) | 216 (12) | 514 (18) | 87 (27) | 650 (12) | 65 (6) |
| Texas | 1516 (24) | 390 (22) | 926 (32) | 67 (21) | 1931 (34) | 289 (28) |
| Washington | 882 (14) | 167 (9) | 175 (6) | 19 (6) | 941 (17) | 113 (11) |
| Wisconsin | 1673 (27) | 563 (31) | 908 (31) | 100 (31) | 1357 (24) | 349 (34) |
| Male | 3189 (51) | 918 (51) | 1477 (50) | 163 (50) | 2888 (51) | 551 (53) |
| Race | ||||||
| Non-hispanic white | 4001 (64) | 1183 (66) | 1920 (66) | 208 (64) | 3526 (63) | 656 (64) |
| Non-hispanic black | 622 (10) | 208 (12) | 273 (9) | 47 (15) | 509 (9) | 101 (10) |
| Non-hispanic, other race | 694 (11) | 176 (10) | 278 (9) | 31 (10) | 589 (11) | 115 (11) |
| Hispanic, any race | 928 (15) | 239 (13) | 458 (16) | 38 (12) | 1000 (18) | 159 (15) |
| Missing | 12 (0) | 3 (0) | 7 (0) | 1 (0) | 12 (0) | 3 (0) |
| Underlying medical condition other than obesity | 282 (5) | 60 (3) | 125 (4) | 12 (4) | 267 (5) | 37 (4) |
| Self-reported health | ||||||
| Excellent/very good | 5117 (82) | 1552 (86) | 2440 (83) | 277 (85) | 4605 (82) | 874 (85) |
| Good | 937 (15) | 211 (12) | 419 (14) | 38 (12) | 853 (15) | 134 (13) |
| Fair/poor | 201 (3) | 41 (2) | 75 (3) | 10 (3) | 175 (3) | 26 (3) |
| Refused | 2 (0) | 5 (0) | 2 (0) | 0 (0) | 3 (0) | 0 (0) |
| Vaccinated | 3062 (49) | 688 (38) | 1372 (47) | 90 (28) | 2700 (48) | 300 (29) |
| < 90 days before onset | 1314 (43) | 339 (49) | 552 (40) | 33 (37) | 1107 (41) | 96 (32) |
| Received LAIV | 504 (16) | 117 (17) | 163 (12) | 33 (37) | 445 (16) | 44 (15) |
| Calendar time‡ | ||||||
| Pre-peak | 1632 (26) | 306 (17) | 748 (25) | 61 (19) | 2473 (44) | 247 (24) |
| Peak | 1982 (32) | 1141 (63) | 1023 (35) | 201 (62) | 2300 (41) | 628 (61) |
| Post-peak | 2643 (42) | 362 (20) | 1165 (40) | 63 (19) | 863 (15) | 159 (15) |
| Onset to enrollment interval | ||||||
| 0–2 days | 2485 (40) | 934 (52) | 1193 (41) | 150 (46) | 2240 (40) | 475 (46) |
| 3–4 days | 2384 (38) | 605 (33) | 1096 (37) | 108 (33) | 2143 (38) | 395 (38) |
| 5–7 days | 1388 (22) | 270 (15) | 647 (22) | 67 (21) | 1253 (22) | 164 (16) |
| Season | ||||||
| 2011–12 | 698 (11) | 130 (7) | 0 (0) | 0 (0) | 0 (0) | 0 (0) |
| 2012–13 | 1075 (17) | 324 (18) | 0 (0) | 0 (0) | 881 (16) | 390 (38) |
| 2013–14 | 0 (0) | 0 (0) | 997 (34) | 164 (50) | 0 (0) | 0 (0) |
| 2014–15 | 1962 (31) | 525 (29) | 0 (0) | 0 (0) | 1625 (29) | 111 (11) |
| 2015–16 | 0 (0) | 0 (0) | 892 (30) | 82 (25) | 746 (13) | 119 (12) |
| 2016–17 | 1285 (21) | 345 (19) | 0 (0) | 0 (0) | 1250 (22) | 181 (18) |
| 2017–18 | 1237 (20) | 485 (27) | 1047 (36) | 79 (24) | 1134 (20) | 233 (23) |
BMI body mass index, LAIV live attenuated influenza vaccine.
*Cases and controls included in this table are from seasons with 50 or more cases of the specific virus and sites that had 10 or more cases of the specific virus.
†BMI categories: Normal (BMI percent 5 to < 85), Overweight (BMI percent 85 to < 95), Obesity (BMI percent ≥ 95).
‡Peak defined as the weeks during which the middle 50% of all cases of the type/subtype occurred during the season at each site.
By influenza virus type/subtype and BMI category, VE estimates in adults against A/H3N2 varied from 16% (95% confidence interval (CI) 2, 28%) to 31% (95% CI 21, 39%), against A/H1N1pdm09 from 46% (95% CI 30, 58%) to 54% (95% CI 42, 63%), and against type B from 44% (95% CI 31, 55%) to 57% (48, 64%) (Fig. 1). Results were similar in a sensitivity analysis that excluded adults with vaccination based on self-report. In children and adolescents, VE estimates by influenza virus type/subtype and BMI category varied from 29% (95% CI 7, 45%) to 34% (95% CI 13, 50%) against A/H3N2, from 57% (95% CI 19, 77%) to 65% (31, 83%), against A/H1N1pdm09, and from 50% (95% CI 29, 65%) to 55% (37, 68%) against type B. (Fig. 1). Results were similar in a sensitivity analysis that excluded children < 9 years old who were partially vaccinated. VE in children was also similar across BMI categories in a sensitivity analysis that excluded LAIV recipients. Interaction terms for modification of VE by BMI category were not statistically significant (p > 0. 10) for adults or children.
Fig. 1.
Adjusted vaccine effectiveness against medically attended influenza A/H3N2, A/H1N1pdm09, and B by BMI category in adults and children/adolescents in the US Flu VE Network, 2011–12 through 2017–18. BMI body mass index, OR odds ratio, CI confidence interval, aVE adjusted vaccine effectiveness. *BMI categories: Adults: Normal (18.5 to < 25 kg/m2), Overweight (25 to < 30 kg/m2), Obesity (≥ 30 kg/m2); Children/adolescents: Normal (BMI percent 5 to < 85), Overweight (BMI percent 85 to < 95), Obesity (BMI percent ≥ 95). †Adjusted vaccine effectiveness—calculated as 100*(1—adjusted odds ratio). Adjusted odds ratio of vaccination status—Adjusted for site, season, age (as spline in adults; as category (2– < 9 years, 9–17 years) in children), BMI, and interaction of BMI and vaccination status.
When the 2014–15 season was excluded for A/H3N2, there was evidence that BMI category modified VE among adults (interaction p = 0.06). Compared to the main analysis, among both adults and children, A/H3N2 VE estimates increased for all BMI categories, and the trend was the same with the highest VE among adults in the overweight category, and among children in the normal category. (Supplemental Table 1, 2). Among adults, there was evidence that BMI category may modify VE against A/H3N2 when class 3 obesity is categorized separately from obesity (interaction p = 0.09), but the VE point estimates do not indicate a consistent trend of increased risk for higher BMI. Across all flu types/subtypes, the VE trends observed for distinct obesity and class 3 obesity categories had wide confidence intervals and were generally similar to the results presented in the main findings (Supplemental Table 1). VE estimates across all types/subtypes decreased when restricting the adult population to ≥ 65 years of age. None of the estimates for A/H3N2 remained statistically significant, but VE among older adults with BMI in the overweight and obesity categories were significant for A/H1N1pdm09 and B (Supplemental Table 1).
Among vaccinated adults, the adjusted odds of laboratory-confirmed influenza for all types/subtypes were significantly lower in participants with obesity vs normal weight (Table 5). Significantly lower odds of influenza among participants with obesity remained when excluding participants with plausible self-reported vaccination. Among vaccinated children, odds of infection with A/H3N2 were significantly lower among children with obesity compared to children with normal weight, but odds of infection with A/H1N1pdm09 or type B did not differ by BMI category (Table 5). The significantly lower odds of infection with A/H3N2 remained among children with obesity when excluding partially vaccinated children < 9 years of age and LAIV recipients.
Table 5.
Association between obesity and laboratory confirmed influenza illness in vaccine recipients, 2011–12 through 2017–18.
| A/H3N2 aOR* (95% CI) | A/H1N1pdm09 aOR* (95% CI) | B aOR* (95% CI) | |
|---|---|---|---|
| Adult BMI category† | n = 10061 | n = 5646 | n = 9119 |
| Normal | Ref | Ref | Ref |
| Overweight | 0.88 (0.77, 1.01) | 0.77 (0.59, 1.00) | 0.82 (0.67, 1.02) |
| Obesity | 0.83 (0.73, 0.94) | 0.62 (0.49, 0.79) | 0.74 (0.61, 0.90) |
| Children/adolescent BMI category‡ | n = 3750 | n = 1462 | n = 3000 |
| Normal | Ref | Ref | Ref |
| Overweight | 0.81 (0.64, 1.03) | 0.73 (0.39, 1.39) | 1.0 (0.73, 1.43) |
| Obesity | 0.70 (0.55, 0.88) | 0.78 (0.42, 1.43) | 0.75 (0.54, 1.05) |
BMI body mass index, aOR adjusted odds ratio, CI confidence interval, Ref Reference category.
*From logistic regression models adjusted for season, site, time from vaccination to illness onset (< 90 days, ≥ 90 days) and age (spline for adults; categories 2 to < 9 years, 9–17 years for children/adolescents).
†Adult BMI categories: Normal (18.5 to < 25 kg/m2), Overweight (25 to < 30 kg/m2), Obesity (≥ 30 kg/m2).
‡Children/adolescent BMI categories: Normal (BMI percent 5 to < 85), Overweight (BMI percent 85 to < 95), Obesity (BMI percent ≥ 95).
Discussion
In this analysis over multiple seasons using consistent methods from a test-negative design study of influenza VE, participants with BMI categorized as overweight or obesity at the time of illness had similar VE against medically attended influenza compared to individuals with normal weight BMI. Findings were similar in adults and children, and across all three influenza types/subtypes (A/H3N2, A/H1N1pdm09, and B). With high prevalence of obesity in the US population, even a modest reduction in VE due to obesity could have a substantial public health impact. The results from this analysis provide evidence that influenza vaccination remains effective at preventing medically attended influenza illness across a range of BMI.
Obesity is often accompanied by metabolic syndrome and other comorbidities, and these conditions may contribute to immune dysregulation and increased severity of respiratory infections31. During the 2009 influenza A/H1N1pdm09 pandemic and the COVID-19 pandemic, obesity was associated with more severe illness 32,33. The mechanism is not well-understood, but increased production of inflammatory cytokines in the lungs may be important during acute infection34. Obesity can also influence the antibody response to vaccine antigens. Impaired vaccine response in individuals with obesity was first reported in 1985 with hepatitis B vaccination of health care workers35. Few studies have assessed influenza vaccine immunogenicity and obesity, and results have been inconsistent. During the 2009–10 season, the HI response to each of the vaccine antigens was similar in a convenience sample of individuals with obesity and normal weight15. However, individuals with obesity were more likely to have a fourfold reduction in HI titer by 12 months, suggesting more rapid waning. CD8 + T cell activation was also impaired in the participants with obesity. A more recent prospective cohort study in children and adolescents in 2019–20 found that pre- and post-vaccination geometric mean HI titers were similar in participants with obesity and normal weight36.
In this study we did not assess vaccine immune response, but we observed reduced odds for laboratory-confirmed breakthrough influenza among vaccinated adults with obesity vs. those with normal weight. In vaccinated children, we observed a similar association for A/H3N2 but not for A/H1N1pdm09 or type B, which also remained with the exclusion of partially vaccinated children. The reduced odds of medically attended breakthrough influenza infection in adult vaccine recipients with obesity vs normal weight were statistically significant and ranged from 0.62 (for A/H1N1pdm09) to 0.83 (for A/H3N2). In contrast, other studies have reported elevated risk of influenza (or A/H1N1pdm09) in adults with obesity vs normal weight8,37. Given these previous findings and the accumulating evidence for obesity-related immune dysregulation, our finding was unexpected. This may be the result of residual confounding due to differences in health care seeking behaviors or differential risk of infection between individuals with obesity and normal weight. Individuals with obesity may have additional underlying medical conditions such as chronic respiratory conditions that we were unable to measure in this analysis. Higher body mass index has been associated with asthma and chronic cough38,39. Despite requiring acute cough of less than 7 days duration for enrollment, individuals with obesity may have enrolled more frequently with a cough without an infectious cause. Additionally, pre-existing respiratory conditions may increase the severity of influenza virus infection causing patients with obesity to bypass the outpatient settings from which we enrolled to seek higher levels of care.
Strengths of this study include enrollment based on symptoms rather than infection status in multiple states over seven seasons, use of the test-negative design to minimize confounding due to health care seeking behavior, and separate analyses by influenza type/subtype in adults and children/adolescents. Two different sensitivity analyses yielded results that were consistent with the primary analysis. This study also has some limitations. We were unable to assess VE against influenza emergency department visits or hospitalization and other severe outcomes. Misclassification of BMI category may have occurred because height and weight were not measured during study enrollment. Instead, BMI categories were determined from height and weight variables in EMRs. Misclassification of vaccination status is possible since we included plausible self-report without documentation of vaccine receipt in EMR.
In conclusion, this large multi-season study provides evidence that influenza vaccination reduces influenza illness and outpatient healthcare visits across a spectrum of BMI. Further research to assess the association between influenza vaccine effectiveness and obesity in the hospital setting may be beneficial since influenza related illness may be more severe in people with obesity. Our findings from the outpatient setting reinforce current recommendations for universal influenza vaccination in all adults and all children greater than 6 months of age, to protect against influenza and its complications.
Supplementary Information
Acknowledgements
The authors would like to acknowledge the following individuals for their contributions to this work: Deanna Cole, Lynn Ivacic, Sarah Kopitzke, Jennifer Meece, Madalyn Palmquist, Carla Rottscheit, Jackie Salzwedel, Sandra Strey, Maria Sundaram, Michael Smith, Chandni Raiyani, Lydia Clipper, Kempapura Murthy, Wencong Chen, Michael Reis, Teresa Ponder, Todd Crumbaker, Iosefo Iosefo, Patricia Sleeth, Virginia Gandy, Kelsey Bounds, Mary Kylberg, Arundhati Rao, Robert Fader, Kimberley Walker, Marcus Volz, Jeremy Ray, Deborah Price, Jennifer Thomas, Hania Wehbe-Janek, Madhava Beeram, John Boyd, Jamie Walkowiak, Robert Probe, Glen Couchman, Shahin Motakef and Alejandro Arroliga, Alan Aspinall, GK Balasubramani, Todd Bear, Rina Chabra, Edward Garofolo, Robert Hickey, Richard Hoffmaster, Philip Iozzi, Monika Johnson, Christopher Olbrich, Dina Perry, Jonathan Raviotta, Evelyn Reis, Brett Rosenblum, Theresa M. Sax, Michael Susick, Joe Suyama, Rachel Taber, Leonard Urbanski, Sara Walters, Alexandra Weissman, John V. Williams, Richard K. Zimmerman.
Author contributions
JPK, HQN, and EAB designed the study. JPK, HQN, ELK, CHP, MG, ETM, KMG, MPN, and EAB supervised data collection at their respective sites. JRC and BF compiled the data and provided critical feedback to shape the analysis. JPK performed the analyses and drafted the manuscript. All authors contributed to the interpretation of the data and provided critical feedback on the manuscript.
Funding
This work was supported by the US Centers for Disease Control and Prevention (CDC) (Grant numbers: 1U01IP000471 and 1U01IP001038) and CDC Grant Number IP000467 and the National Institutes of Health (NIH) grant number 1UL1 TR001857 (Pittsburgh).
Data availability
The datasets generated during and/or analysed during the current study are not publicly available because they contain coded, personally identifiable information, but may be made available from the corresponding author on reasonable request.
Competing interests
EAB and HQN receive research support from Seqirus. MPN receives research support from Sanofi. MJG received research support from MedImmune during some of the analyzed seasons and was the Texas Pediatric Society, Texas Chapter of American Academy of Pediatrics, Co-Chair, Infectious Diseases and Immunization Committee (Sep 2016–Aug 2022). ETM receives research support from Merck. JPK, ELK, CHP, KMG, JRC, and BF report no competing interests.
Ethics approval
The institutional review boards at the Centers for Disease Control and Prevention and all participating sites (Marshfield Clinic Health System IRB, Group Health Cooperative Human Subjects Review Committee/Kaiser Permanente-Washington Region IRB, Baylor Scott & White Research Institute IRB, University of Michigan IRBMED, and University of Pittsburgh Human Research Protection Office) approved the study.
Footnotes
Publisher's note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
The online version contains supplementary material available at 10.1038/s41598-024-72081-z.
References
- 1.World Health Organization. Obesity and overweight, <https://www.who.int/news-room/fact-sheets/detail/obesity-and-overweight> (2021).
- 2.Hales, C. M., Carroll, M. D., Fryar, C. D. & Ogden, C. L. Prevalence of obesity among adults and youth: United States, 2015-2016. NCHS Data brief, no 288. Hyattsville, MD: National Center for Health Statistics. 2017. [PubMed]
- 3.Centers for Disease Control and Prevention. Overweight & Obesity-Adult Obesity Facts, <https://www.cdc.gov/obesity/data/adult.html> (2022).
- 4.Coleman, B. L., Fadel, S. A., Fitzpatrick, T. & Thomas, S. M. Risk factors for serious outcomes associated with influenza illness in high- versus low- and middle-income countries: Systematic literature review and meta-analysis. Influ. Other Respir. Virus.12, 22–29. 10.1111/irv.12504 (2018). 10.1111/irv.12504 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Vitoratou, D. I., Milas, G. P., Korovessi, P., Kostaridou, S. & Koletsi, P. Obesity as a risk factor for severe influenza infection in children and adolescents: A systematic review and meta-analysis. Eur. J. Pediatr.182, 363–374. 10.1007/s00431-022-04689-0 (2023). 10.1007/s00431-022-04689-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Cocoros, N. M., Lash, T. L., DeMaria, A. Jr. & Klompas, M. Obesity as a risk factor for severe influenza-like illness. Influ. Other Respir. Virus.8, 25–32. 10.1111/irv.12156 (2014). 10.1111/irv.12156 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Moser, J. S. et al. Underweight, overweight, and obesity as independent risk factors for hospitalization in adults and children from influenza and other respiratory viruses. Influ. Other Respir. Virus.10.1111/irv.12618 (2018). 10.1111/irv.12618 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Karki, S. et al. Association between body mass index and laboratory-confirmed influenza in middle aged and older adults: A prospective cohort study. Int. J. Obes.42, 1480–1488. 10.1038/s41366-018-0029-x (2018). 10.1038/s41366-018-0029-x [DOI] [PubMed] [Google Scholar]
- 9.Martin, E. T. et al. Epidemiology of severe influenza outcomes among adult patients with obesity in Detroit, Michigan, 2011. Influ. Other Respir. Virus.7, 1004–1007. 10.1111/irv.12115 (2013). 10.1111/irv.12115 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Tempia, S. et al. Risk factors for influenza-associated severe acute respiratory illness hospitalization in South Africa, 2012–2015. Open Forum Infect. Dis.4, ofw262. 10.1093/ofid/ofw262 (2017). 10.1093/ofid/ofw262 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Zhou, Y. et al. Adiposity and influenza-associated respiratory mortality: A cohort study. Clin. Infect. Dis.60, e49-57. 10.1093/cid/civ060 (2015). 10.1093/cid/civ060 [DOI] [PubMed] [Google Scholar]
- 12.Callahan, S. T. et al. Impact of body mass index on immunogenicity of pandemic H1N1 vaccine in children and adults. Journal Infect. Dis.210, 1270–1274. 10.1093/infdis/jiu245 (2014). 10.1093/infdis/jiu245 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Kainth, M. K. et al. Obesity and metabolic dysregulation in children provide protective influenza vaccine responses. Viruses10.3390/v14010124 (2022). 10.3390/v14010124 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Yang, W. H. et al. Long-term immunogenicity of an AS03-adjuvanted influenza A(H1N1)pdm09 vaccine in young and elderly adults: An observer-blind, randomized trial. Vaccine31, 4389–4397. 10.1016/j.vaccine.2013.07.007 (2013). 10.1016/j.vaccine.2013.07.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Sheridan, P. A. et al. Obesity is associated with impaired immune response to influenza vaccination in humans. Int. J. Obes.36, 1072–1077. 10.1038/ijo.2011.208 (2012). 10.1038/ijo.2011.208 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Talbot, H. K. et al. Association between obesity and vulnerability and serologic response to influenza vaccination in older adults. Vaccine30, 3937–3943. 10.1016/j.vaccine.2012.03.071 (2012). 10.1016/j.vaccine.2012.03.071 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Paich, H. A. et al. Overweight and obese adult humans have a defective cellular immune response to pandemic H1N1 influenza A virus. Obesity21, 2377–2386. 10.1002/oby.20383 (2013). 10.1002/oby.20383 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Smit, M. A. et al. Influenza Vaccine is Protective Against Laboratory-confirmed Influenza in Obese Children. Pediatr. Infect. Dis. J.35, 440–445. 10.1097/INF.0000000000001029 (2016). 10.1097/INF.0000000000001029 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Neidich, S. D. et al. Increased risk of influenza among vaccinated adults who are obese. Int. J. Obes.41, 1324–1330. 10.1038/ijo.2017.131 (2017). 10.1038/ijo.2017.131 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Centers for Disease Control and Prevention. US Flu VE Network, <https://www.cdc.gov/flu/vaccines-work/us-flu-ve-network.htm> (2023).
- 21.Jackson, M. L. et al. Influenza vaccine effectiveness in the United States during the 2015–2016 season. N. Engl. J. Med.377, 534–543. 10.1056/NEJMoa1700153 (2017). 10.1056/NEJMoa1700153 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.McLean, H. Q. et al. Influenza vaccine effectiveness in the United States during 2012–2013: Variable protection by age and virus type. J. Infect. Dis.211, 1529–1540. 10.1093/infdis/jiu647 (2015). 10.1093/infdis/jiu647 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.U.S. Department of Health and Human Services, Public Health Service, National Institutes of Health & National Heart Lung and Blood Institute. The Practical Guide. Identification, Evaluation, and Treatment of Overweight and Obesity in Adults, <https://www.nhlbi.nih.gov/files/docs/guidelines/prctgd_c.pdf> (2000).
- 24.Flegal, K. M., Kit, B. K. & Graubard, B. I. Body mass index categories in observational studies of weight and risk of death. Am. J. Epidemiol.180, 288–296. 10.1093/aje/kwu111 (2014). 10.1093/aje/kwu111 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Centers for Disease Control and Prevention & Division of Nutrition Physical Activity and Obesity. A SAS Program for the 2000 CDC Growth Charts (ages 0 to <20 years), <https://www.cdc.gov/nccdphp/dnpao/growthcharts/resources/sas.htm> (2019).
- 26.Centers for Disease Control and Prevention. BMI Frequently Asked Questions, <https://www.cdc.gov/bmi/faq/?CDC_AAref_Val=https://www.cdc.gov/healthyweight/assessing/bmi/childrens_bmi/about_childrens_bmi.html> (2024).
- 27.Cheng, F. W. et al. Body mass index and all-cause mortality among older adults. Obesity24, 2232–2239. 10.1002/oby.21612 (2016). 10.1002/oby.21612 [DOI] [PubMed] [Google Scholar]
- 28.Jackson, M. L. & Nelson, J. C. The test-negative design for estimating influenza vaccine effectiveness. Vaccine31, 2165–2168. 10.1016/j.vaccine.2013.02.053 (2013). 10.1016/j.vaccine.2013.02.053 [DOI] [PubMed] [Google Scholar]
- 29.Foppa, I. M., Haber, M., Ferdinands, J. M. & Shay, D. K. The case test-negative design for studies of the effectiveness of influenza vaccine. Vaccine31, 3104–3109. 10.1016/j.vaccine.2013.04.026 (2013). 10.1016/j.vaccine.2013.04.026 [DOI] [PubMed] [Google Scholar]
- 30.Centers for Disease Control and Prevention. Child Immunization Schedule Notes - Influenza Vaccination, <https://www.cdc.gov/vaccines/schedules/hcp/imz/child-schedule-notes.html#note-flu> (2023).
- 31.Tagliabue, C., Principi, N., Giavoli, C. & Esposito, S. Obesity: Impact of infections and response to vaccines. Eur. J. Clin. Microbiol. Infect. Dis.35, 325–331. 10.1007/s10096-015-2558-8 (2016). 10.1007/s10096-015-2558-8 [DOI] [PubMed] [Google Scholar]
- 32.Sun, Y. et al. Weight and prognosis for influenza A(H1N1)pdm09 infection during the pandemic period between 2009 and 2011: A systematic review of observational studies with meta-analysis. Infect. Dis. (Lond. Engl.)48, 813–822. 10.1080/23744235.2016.1201721 (2016). 10.1080/23744235.2016.1201721 [DOI] [PubMed] [Google Scholar]
- 33.Popkin, B. M. et al. Individuals with obesity and COVID-19: A global perspective on the epidemiology and biological relationships. Obes. Rev.21, e13128. 10.1111/obr.13128 (2020). 10.1111/obr.13128 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Chandrasekaran, R. et al. Obesity exacerbates influenza-induced respiratory disease via the arachidonic acid-p38 MAPK pathway. Front. Pharmacol.14, 1248873. 10.3389/fphar.2023.1248873 (2023). 10.3389/fphar.2023.1248873 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Weber, D. J., Rutala, W. A., Samsa, G. P., Santimaw, J. E. & Lemon, S. M. Obesity as a predictor of poor antibody response to hepatitis B plasma vaccine. JAMA254, 3187–3189 (1985). 10.1001/jama.1985.03360220053027 [DOI] [PubMed] [Google Scholar]
- 36.Clarke, M. et al. A prospective study investigating the impact of obesity on the immune response to the quadrivalent influenza vaccine in children and adolescents. Vaccines10.3390/vaccines10050699 (2022). 10.3390/vaccines10050699 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Maier, H. E. et al. Obesity is associated with increased susceptibility to influenza A (H1N1pdm) but Not H3N2 Infection. Clin. Infect. Dis.73, e4345–e4352. 10.1093/cid/ciaa928 (2021). 10.1093/cid/ciaa928 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Beuther, D. A. & Sutherland, E. R. Overweight, obesity, and incident asthma: a meta-analysis of prospective epidemiologic studies. Am. J. Respir. Crit. Care Med.175, 661–666. 10.1164/rccm.200611-1717OC (2007). 10.1164/rccm.200611-1717OC [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Morales-Estrella, J. L., Ciftci, F. D., Trick, W. E. & Hinami, K. Physical symptoms screening for cardiopulmonary complications of obesity using audio computer-assisted self-interviews. Qual. Life. Res.26, 2085–2092. 10.1007/s11136-017-1549-x (2017). 10.1007/s11136-017-1549-x [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The datasets generated during and/or analysed during the current study are not publicly available because they contain coded, personally identifiable information, but may be made available from the corresponding author on reasonable request.