Losing ground at the wrong time: Trends in self-reported influenza vaccination uptake in Switzerland, Heath Survey 2007-2017

Objectives We studied time trends in seasonal influenza vaccination and associations with socioeconomic and health-related determinants in Switzerland, overall and in people aged [≥]65 years. Design Three cross-sectional surveys. Participants Individuals who participated in the Swiss Health Surveys 2007, 2012, and 2017. We calculated the proportion reporting influenza vaccination in the last 12 months, and performed multivariable logistic regression analyses. Results: The proportion of reporting a history of influenza vaccination overall was 31.9% (95% confidence intervals [95% CI] 31.4-32.4); and dropped from 34.5% in 2007 to 28.8% in 2017. The uptake of vaccination within the past 12 months was 16% in 2007 and similar in 2012 and 2017 (around 14%). In people with chronic disease, uptake dropped from 43.8% in 2007 to 37.1% in 2012 and to 31.6% in 2017 (p<0.001). In people aged [≥]65 years, uptake dropped from 47.8% in 2007 to 38.5% in 2012 to 36.2% in 2017 (p<0.001). Similarly, a decrease in vaccine uptake was seen in people with poor self-reported health status (39.4%, 33.1%, and 27.0%). In logistic regression, self-reported vaccination coverage decreased in the 65 to 75 years old (adjusted odds ratio (aOR) aOR 0.56, 95% Cl 0.48-0.66 between 2007 and 2012; aOR 0.89, 95% CI 0.77-1.03). Uptake was positively associated with the [≥]65 age group, living in French-speaking and urban areas, history of smoking, bad self-reported health status, private/semiprivate health insurance, having a medical profession, and having any underlying chronic disease. Use of any alternative medicine therapy was negatively associated with influenza vaccination (aOR 0.72, 95% CI 0.67-0.80). Conclusion: Influenza vaccination coverage was low in older and chronically ill persons. Significant efforts are required in preparing for the flu season 2020/21 to reduce the double burden of COVID-19 and seasonal influenza. These efforts should include campaigns but also novel approaches using social media.


Introduction
(including acupuncture, traditional Chinese medicine, homoeopathy, and osteopathy) 129 in the past 12 months was recorded as well. 130 Pregnancy was recorded as current pregnancy among women 15 to 49 years 131 old. Current chronic conditions included asthma, chronic bronchitis and emphysema. 132 Diabetes was defined by the use of any diabetic drug, cardiovascular disease by the 133 use of any heart medication, and all other chronic diseases were recorded as self-134 reported. We defined any chronic disease as the presence of at least one of the 135 mentioned diseases. Health care workers were defined as individuals reporting 136 profession in the health care system. 137 Cantons are the administrative subdivisions of Switzerland (see Figure 1 ). 138 139

Statistical analysis 140
For each of the three survey years we calculated the proportions (overall and ≥ 65 141 years) that reported having been vaccinated within the last 12 months. We estimated 142 associations between vaccination status and socio-demographic and health-related 143 factors. We included an interaction term with the year of survey and the variable of 144 interest in multivariable logistic regression models. We used the SFSO's survey 145 weights and reported all proportions and unadjusted and adjusted odds ratio (ORs 146 and aOR) with the corresponding 95% confidence intervals (95% CI) derived from 147 robust standard error calculations (Table 2). All analyses were performed in Stata 148 (version 15.1,Corporation,College Station,Texas,USA). 149 We visualized changes in the frequency of vaccination uptake, and 150 geographical distributions of the population that reported vaccination for influenza at 151 the cantonal level using ArcGIS version 10.5 (Redlands, CA, USA). 152 153 154 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted June 7, 2020. . https: //doi.org/10.1101//doi.org/10. /2020 Ethics statement 155 Data were collected anonymized and ethical approval was not required but we 156 obtained permission to analyze and publish the data through a contract with the 157 . 158 159 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

Trends of influenza vaccinations status over time 161
The proportion of survey participants reporting a history of influenza vaccination 162 overall was 31.9% (95% CI 31.4-32.4), having dropped from 34.5% in 2007 to 28.8% 163 in 2017. The proportion reporting vaccination within the past 12 months was 16.4% 164 (95% CI 15.6-17.2) in 2007, dropped to 14.1% (95% CI 13.5-14.8) Figure 2). The 202 regional differences in the vaccination status in the last 12 months in the overall 203 population and the In all three surveys, influenza vaccination in the last 12 months was positively 209 associated with age 65 years or older, living in French-speaking and urban areas, 210 having a tertiary education, history of smoking, bad self-reported health status, 211 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted June 7, 2020. . https://doi.org/10.1101/2020.06.05.20123026 doi: medRxiv preprint private/semiprivate hospital stay insurance, being a health care worker, and having 212 any underlying chronic disease. Use of any alternative medicine therapy was 213 negatively associated with influenza vaccination (aOR 0.72, 95% CI 0.67-0.80). We 214 found no association with sex, body mass index, or non-Swiss citizenship. (Table 2). 215 Similar findings were observed when analyzing only participants' ≥ 65 years old. 216 Supplementary Table 2 shows unadjusted and adjusted ORs of the associations 217 between self-reported vaccination for influenza in the last 12 months and socio-218 demographic characteristics and health-related factors. 219 220 221 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted June 7, 2020. .

222
Progress toward the WHO target of 75% vaccination coverage among high-risk 223 groups has not only stalled in Switzerland, some of the gains made in earlier years 224 were lost by 2017. Of particular concern is that the self-reported seasonal influenza The decline in coverage and an increase in vaccine hesitancy is of concern. In 239 the last years, several countries of the WHO European Region including Switzerland, 240 had large outbreaks of vaccine-preventable diseases, including measles, rubella and 241 influenza. Switzerland and other countries identified steps to improve vaccination 242 coverage for influenza and other infectious diseases (12, 13). The relationship 243 between vaccination uptake, knowledge, attitudes, and awareness is complex (14). 244 Reasons for people not getting vaccinated against influenza include underestimation 245 of disease severity, fear of side effects, and the cost of vaccines (15-17). Often 246 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted June 7, 2020. .
people are unaware that they should receive the vaccination (15). Of note, the 247 coverage of measles vaccination increased to almost 90% in young adults in 248 Switzerland, probably due to the campaigns and a national measles strategy (18). 249 The experience with measles could serve as a model for influenza. 250 The advent of the novel coronavirus SARS-Cov-2 has profoundly changed 251 everyday life in Switzerland and elsewhere. It is unclear how the COVID-19 252 pandemic will affect attitudes toward vaccines. The director of the US Centers for 253 Disease Control has warned that a possible second wave of Covid-19 could be worse 254 than the first (19). This has already been seen in the "Spanish flu" pandemic of 255 1918/19. The Spanish flu affected Switzerland in two waves. The first one occurred in 256 July 1918, the second, more severe one, in October-November 1918 (20). However, 257 even in the absence of a second wave, a resurgence of Covid-19 that coincides with 258 the start of the flu season could significantly stress health care systems. An effective 259 and safe vaccine against COVID 19 is unlikely to become widely available in 2020. 260 Therefore, low influenza vaccination rates at around half the 75% coverage 261 recommended for high-risk groups constitute a hazard that merits prompt, focused 262 attention by public health authorities. 263 A concerted effort to increase influenza vaccination coverage in 2020/21, 264 when the COVID-19 pandemic to continue to pose a threat to the public's health, is 265 urgently needed. Influenza vaccination prevents deaths, morbidity, hospital 266 admissions, and other adverse health outcomes, in target populations such as older 267 people, chronically ill people (21-26), and also children (27) and pregnant women 268 (28). The continued promotion of infection control measures such as avoiding close 269 contact with sick people, covering one's nose and mouth while coughing or sneezing, 270 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted June 7, 2020. . social distancing and hand hygiene will contribute to reducing the spread of both 271 influenza and Covid-19 (3). 272 Our study has several limitations. Influenza vaccination status is self-reported 273 in the Swiss Health Survey, and the reliability of the data unclear. For example, 274 vaccination coverage could be lower if social desirability bias led to an overestimation 275 of uptake. Conversely, incomplete recall of vaccinations could bias coverage 276 downwards. Individuals younger than 15 years are excluded from the survey, but 277 coverage in this age group is probably even lower than in the 15 to 19-year-olds. A 278 strength of our analysis is the fact that the survey is a nationwide and representative, 279 and repeated every five years using the same methodology. Also, the analyses were 280 weighted and adjusted for a range of potential confounders, which did not 281 substantially change the results. 282 In conclusion, we need to increase influenza vaccination uptake, particularly 283 in the elderly and chronically ill, who are also the risk groups most heavily affected by 284 COVID-19. These efforts should include classic information campaigns, but novel 285 approaches using social media should also be considered (29,30). 286 Recommendations by health care professionals are essential to improve influenza 287 vaccination coverage, such as client reminder/recall and standing orders (31). The 288 preparation of influenza season 2020/21 must start now to address the double 289 burden of COVID-19 and seasonal influenza. 290 291 292 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted June 7, 2020. There was no specific funding for this project. ME was supported by special project 309 funding (grant 17481) from the Swiss National Science Foundation. 310 311 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted June 7, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted June 7, 2020. .

9.
Swiss Federal Statistical Office. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted June 7, 2020. .

17.
Wheelock A, is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted June 7, 2020. .

25.
Liu CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted June 7, 2020. . . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted June 7, 2020. . is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted June 7, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted June 7, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted June 7, 2020. .

23
3 Current chronic diseases vs. no disease (reference); lung diseases: asthma bronchiale, chronic bronchitis and emphysema; diabetes: use of any diabetic drug; cardiovascular disease: use of any heart medication; all other diseases: self-reported 4 Any profession in the health care system (in the 2007 survey, the veterinary professions were also included in this category) vs. any other profession (reference) . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted June 7, 2020.  . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 7, 2020. . traditional Chinese medicine, homeopathy, non-medical practitioner, osteopathy) 3 Any current chronic disease versus no chronic disease (reference); lung diseases: asthma bronchiale, chronic bronchitis and emphysema vs. no lung disease (reference); diabetes: use of any diabetic drug vs. no diabetes (reference); cardiovascular disease: use of any heart medication vs. no cardiovascular disease (reference); all other diseases: self-reported disease vs. no disease (reference) 4 Estimates from separate models which included the variable "any chronic disease" instead of specific chronic diseases 5 Any profession in the health care system (in the 2007 survey, the veterinary professions were also included in this category) vs. any other profession (reference) . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 7, 2020. . https://doi.org/10.1101/2020.06.05.20123026 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.