Prevalence of and factors associated with receipt of provider recommendation for influenza vaccination and uptake of influenza vaccination during pregnancy: cross-sectional study

Background Seasonal influenza vaccination has been recommended for pregnant women in Germany since 2010. The aim of this study was to examine prevalence and determinants of receipt of provider recommendation for influenza vaccination as well as influenza vaccination uptake during pregnancy. Methods We analysed data from the “KUNO Kids Health Study”, a prospective birth cohort. During the study period (5th July 2015 to 27th June 2018) data were collected from participating mothers by interview and questionnaire. According to Andersen’s behavioural model of health services use potential influencing factors describing the circumstances and characteristics of the mothers and their pregnancies which are potentially affecting whether women receive a recommendation for a vaccination or whether they utilize influenza vaccination were classified into three domains: ‘predisposing characteristics’, ‘enabling resources’ and ‘need’. Using multivariable logistic regression models odds ratios (OR) and corresponding 95% confidence intervals (95% CI) were calculated. Results As a combined result across three flu seasons, 368 of 1814 (20.3%) women received an influenza vaccination recommendation during pregnancy. Having had a high-risk pregnancy increased the odds of receiving a vaccination recommendation (OR = 1.3; 95% CI = 1.0–1.6; p = 0.045). In contrast, pregnancy onset in summer (OR = 0.7; 95% CI = 0.5–1.0; p = 0.027), autumn (OR = 0.4; 95% CI = 0.3–0.5; p < =0.001) or winter (OR = 0.5; 95% CI = 0.3–0.6; p < =0.001) (compared to spring) as well as mother’s birthplace outside Germany (OR = 0.6; 95% CI = 0.4–0.9; p = 0.023) reduced the chance of getting a vaccination recommendation. Two hundred forty-two of one thousand eight hundred sixty-five (13%) women were vaccinated against influenza during pregnancy. Having received a vaccination recommendation was strongly associated with vaccination uptake (OR = 37.8; 95% CI = 25.5–55.9; p < =0.001). Higher health literacy status was also associated with a higher chance of vaccination uptake (OR = 1.7; 95% CI = 1.2–2.6; p = 0.008), whereas pregnancy onset in autumn (compared to spring) reduced the chance (OR = 0.5; 95% CI = 0.3–0.8; p = 0.008). Conclusions At 13% the uptake rate of influenza vaccination is low. Having received a recommendation to vaccinate was strongly associated with uptake but only one fifth of all mothers report such a recommendation. Raising awareness in physicians regarding vaccinating during pregnancy seems to be of essential importance to increase vaccine uptake and to prevent influenza-related complications in pregnant women. Supplementary Information The online version contains supplementary material available at 10.1186/s12884-021-04182-w.


Background
Pregnant women are at increased risk for influenza-associated complications like pneumonia and, less commonly, myositis, rhabdomyolysis, encephalitis or myocarditis. This leads to higher rates of hospitalization and increased morbidity and mortality among pregnant women [1][2][3]. In addition, the unborn child is in danger of being affected by maternal influenza infection, potentially leading to preterm delivery, spontaneous abortion and congenital anomalies [4][5][6][7][8].
Influenza vaccination is available to and approved for pregnant women. Not only does it protect pregnant women from influenza infection, but the vaccination also benefits the child. Maternal immunization is associated with a reduced rate of preterm birth and a reduced rate of suboptimal birth weight [9][10][11][12]. Via placental transfer of maternally-derived antibodies nest protection leads to a reduced risk of influenza illness in infants under 6 months of age, who are too young to be vaccinated themselves [13,14]. The fact that influenza vaccination is not licensed for children under the age of 6 months stresses the importance of maternal immunization during pregnancy.
Seasonal influenza vaccination has been recommended for pregnant women in Germany since 2010. The recommendations in Germany state that all pregnant women from the 2nd trimester onwards should be vaccinated with the quadrivalent inactivated influenza vaccine, pregnant women with underlying chronic diseases even from the 1st trimester [15,16].
Although the efficacy [17,18] and safety [19][20][21][22] of influenza vaccination has been shown by many studies, influenza vaccination uptake during pregnancy remained quite low so far, as several studies have shown [23][24][25][26][27]. Possible reasons frequently suggested for low uptake are maternal concerns about the safety of the vaccine to the unborn child, lacking recommendations by health care workers (HCWs) and the perception that the vaccination is not necessary [26][27][28][29][30].
Our study was conducted to examine the current prevalence of receipt of provider recommendation for influenza vaccination as well as influenza vaccination uptake during pregnancy among all participating women in the KUNO Kids study [31]. A further aim of our study was to identify factors affecting receipt of provider recommendation for influenza vaccination and uptake of vaccination. This distinguishes our study from most comparable studies where the focus was mainly on reasons for accepting or rejecting the influenza vaccination named by the pregnant women themselves.

Study design and data collection
Methods and design of KUNO Kids health study have been described in detail before [31]. It is an ongoing prospective birth cohort study conducted at the clinic St. Hedwig in Regensburg, Bavaria, Germany and aims to explore a broad range of factors influencing the development and health of children. Briefly, all mothers giving birth at St. Hedwig hospital in Regensburg were asked to take part in the study within 48 h after birth. After mothers have given informed consent a computer-assisted interview and a self-report questionnaire with the mother were performed. Exclusion criteria were maternal age less than 18 years and insufficient German language skills. Also, women who had already participated in the study with a previous child were not allowed to take part once again [31,32]. Data from 5th July 2015 to 27th June 2018 were used for the purposes of our study.

Variables and measurements Outcome variables
To examine the frequency of receipt of provider recommendation for influenza vaccination and uptake rate of influenza vaccination the mothers were asked whether their gynaecologist/obstetrician had advised them to get vaccinated against influenza during pregnancy and whether they received influenza vaccination during pregnancy. The information on vaccination uptake was missing for 21 participants.

Predictor variables: predisposing characteristics, enabling resources and need factors
Variables potentially related to receipt of provider recommendation for influenza vaccination or influenza vaccination uptake were chosen according to Andersen's behavioural model of health services use which distinguished three domains: 'predisposing characteristics' , 'enabling resources' and 'need' [33].
Predisposing characteristics ▪ mother's age (years), which was collected through interview. ▪ country of birth, which was collected through interview. ▪ marital status, which was collected through interview. ▪ being primipara or multipara, which was collected through interview. ▪ highest educational level, which was collected through interview. As a measurement instrument for educational level the "CASMIN" [34] (Comparative Analysis of Social Mobility in Industrial Nations) educational classification was applied in this study. This classification scheme contains nine levels of educations qualification, which can be summarized in three main categories: lowest, medium and highest educational level. ▪ occupational status, which was collected through interview. ▪ type of employment, which was collected through interview. ▪ type of insurance, which was collected through interview. ▪ current living space per person, which was collected through interview. ▪ healthy lifestyle, which was collected through interview. In this context healthy lifestyle was investigated by the factors body mass index (BMI), sporting activities and healthy eating. According to the WHO criteria [35] women with a BMI < 18,5 kg/m 2 were thereby classified as underweight, women with a BMI 18,5-24,9 kg/m 2 as normal weight and women with a BMI > 25,0 kg/ m 2 as overweight. Maintaining a healthy eating was supposed when women stated that they eat fruits and vegetables nearly daily. ▪ smokers living in the household, which was collected through interview. ▪ strength of social support, which was collected through interview. Information about social support was gained with the help of the questionnaire F-SozU-K14 [36]. Thereby, the mothers were, for instance, asked whether they experience much understanding and security from others or whether they have a person, they can go to when they are feeling depressed.

Enabling resources
▪ mother's health literacy, which was collected through interview. Health literacy was assessed with the health care scale (16 items) of the questionnaire HLS-EU-Q47 [37]. The questions refer to an individual's ability to access, understand, appraise, and apply health information in the field of health care (in contrast to prevention and health promotion) and could be answered on a four-point Likert scale from "very easy", "fairly easy", "fairly difficult" to "very difficult". ▪ satisfaction with gynaecological/obstetric treatment during pregnancy, which was collected through questionnaire. ▪ number of doctor visits during pregnancy, which was collected through questionnaire. ▪ claiming of additional prenatal diagnostics, such as amniocentesis, 3D ultrasound or nuchal translucency measurement. ▪ For analysing possible influencing factors on vaccination uptake, the receipt of vaccination recommendation by HCW was also considered an enabling resource, which was collected through questionnaire.

Need factors
▪ having a high-risk pregnancy, which was collected through interview. It describes a situation in which conditions are present which could threaten the life or health of mother or foetus. ▪ presence of maternal pre-existing conditions (including allergic rhinitis/conjunctivitis, bronchial asthma, neurodermitis, Crohn's disease, ulcerative colitis, psoriasis arthritis, rheumatoid arthritis, other autoimmune disease, Diabetes mellitus type 1 or 2, hepatobiliary diseases, kidney disease, thyroid disease, tumour disease, coagulation disorder, heart rhythm disorder, heart attack, high blood pressure, other metabolic disease, multiple sclerosis, epilepsy), which was collected through questionnaire. ▪ season of pregnancy onset, which was collected through interview.

Statistical analysis
Statistical analysis was performed using IBM SPSS Statistics 24. At first, descriptive analyses of the characteristics of the study population were performed. Differences in characteristic features between those who received a recommendation by HCW and those who did not and respectively between those who had received influenza vaccination during pregnancy and those who had not were analysed.
Furthermore, univariable logistic regression analyses were conducted to identify variables related to receipt of provider recommendation for influenza vaccination or uptake of influenza vaccination during pregnancy. Odd Ratios (OR) with 95% confidence intervals (95% CI) and p-values were calculated. Variables with p < 0.20 in univariable analysis were included in the multivariable logistic regression models.
For analysing trend of receipt of provider recommendation for influenza vaccination and uptake rate of influenza vaccination, percentage rates of each examined season in this study, particularly 2015/2016, 2016/2017 and 2017/2018, were calculated and subsequently compared.

Characteristics of the study population
During the study period (5th July 2015 to 27th June 2018) 2620 participants were included in the KUNO Kids study. Of these, 1886 also completed the basic questionnaire and could therefore be considered in further statistical evaluation. This corresponds to more than a 25% loss from the total study enrolment.
Regarding the analytical sample, the median age was 34.33 years (SD = 4.46). 12.2% were born in a country other than Germany, 80.5% were married and living together, 54.8% were primipara. Further characteristics of the study population are shown in Table 1.

Receipt of provider recommendation for influenza vaccination
In total, 368 women (20.3%) responded that they had received an influenza vaccination recommendation by HCW during their pregnancy. Results of univariable analyses are presented in Table 2 and revealed country of birth other than Germany and season of pregnancy onset as statistically significant factors associated with vaccination recommendation (p < 0.05, compare Table 2).
Being born in a country other than Germany, decreased the odds of getting a vaccination recommendation by almost the half (OR = 0.590; 95% CI = 0.391-0.888; p = 0.012). Pregnant women who had their pregnancy onset in autumn (OR = 0.365; 95% CI = 0.262-0.509 p < =0.001) or winter (OR = 0.508; 95% CI = 0.367-0.703; p < =0.001) had significantly lower odds of receiving vaccination recommendation by HCW compared to women whose pregnancy onset was in spring.

Uptake of influenza vaccination
Out of 1865 women 242 (13.0%) were vaccinated against influenza during pregnancy. Results of univariable analyses are shown in Table 3. In univariable analysis country of birth, health literacy, vaccination recommendation by HCW, high-risk pregnancy and season of pregnancy onset reached statistical significance (p < 0.05).
The odds of women with pregnancy onset in winter (OR = 0.531; 95% CI = 0.365-0.774; p = 0.001) or autumn (OR = 0.307; 95% CI = 0.204-0.462; p < =0.001) was significantly lower compared to women whose pregnancy onset was in spring. Multivariable analysis showed a significant (p < 0.05) association between vaccination status and health literacy, respective season of pregnancy onset and vaccination recommendation by HCW. Results are shown in Table 4. Increasing health literacy went along with a higher chance of getting influenza vaccination (OR = 1.736; 95% CI = 1.151-2.618; p = 0.008). Women who had their pregnancy onset in autumn showed halved odds of getting influenza vaccination compared to women whose pregnancy onset was in spring (OR = 0.487; 95% CI = 0.287-0.828; p = 0.008). The chance of influenza vaccination uptake during pregnancy was around 38-fold increased when women had received a vaccination recommendation by HCW (OR = 37.767; 95% CI = 25.515-55.903; p < =0.001).

Prevalence of receipt of recommendation and uptake of influenza vaccination
This study showed a low (13%) overall uptake rate of influenza vaccination by pregnant women over the study period 2015-2018 and also a low proportion of pregnant women who reported to have received a recommendation for the vaccination (20%). Considering the efficacy and safety of the vaccine, as well as the possible reduction of societal cost by preventing influenza infection in pregnant women [38,39], an increase of this uptake rate should be urgently sought.

Factors associated with receipt of provider recommendation
Being born in a country other than Germany was associated with a decreased chance of getting a vaccination recommendation by HCW. Providers might anticipate vaccination hesitancy in this group of women. The association between receiving a vaccination recommendation and pregnancy onset in spring might be explained by the fact these women's second or third trimester of pregnancy (i.e. when influenza vaccination is recommended for every pregnant woman) is likely between November and February -months with increased risk of influenza infection and resulting greater provider awareness for vaccinations. At the same time providers treating women with high-risk pregnancy might be more aware of possible hazards during pregnancy and therefore more willing to recommend influenza vaccination.
In contrast to our findings a study which was conducted in the context of the national influenza immunization campaign in Germany in 2014 [40] found that 93% of gynaecologists/obstetricians advised their patients to get vaccinated against influenza during pregnancy. This discrepancy could result from an overrepresentation of HCWs with a positive attitude toward influenza vaccination in that study. At the same time women may be equally likely to underreport vaccination if the provider did not make a strong recommendation, if the woman was not receptive to the recommendation, or simply due to recall problems. Discrepancies between provider's and patients' reports regarding influenza vaccination recommendation and uptake during pregnancy were also found in a study from the U.S. [41] Since influenza vaccination for pregnant women is officially recommended by the "German Standing Committee on Vaccination" (STIKO) [15,16] in Germany, all gynaecologists/obstetricians are in charge to educate their pregnant patients about the possibility and need of being vaccinated against influenza. Given that there is no evidence that influenza vaccination is teratogenic [42], a change to the recommendations to vaccinate during the second and third trimester to make them consistent with the any-trimester recommendation for women at increased risk could result in better maternal vaccination rates in German women.
Reasons why providers currently do not adhere to the STIKO recommendation were investigated by Bödeker et al. The most frequently mentioned reasons reported by  gynaecologists/obstetricians were safety concerns about the unborn child, high time consumption in informing patients about the recommendation, safety concerns about the pregnant women, and doubts about the efficiency of the vaccination [40]. Another recent study in Germany listed high requirement of time and effort to inform pregnant patients about vaccination as a barrier to recommending seasonal influenza vaccination [43]. A further study pointed out that physicians do not want to decide for pregnant women [44], possibly because they do not want to be liable for any adverse events.

Factors associated with uptake of influenza vaccination
We found that increasing health literacy goes along with a higher chance of getting influenza vaccination and women whose pregnancy onset was in autumn showed lower odds of receiving influenza vaccination compared to women with pregnancy onset in spring. However, with an OR of 38 the strongest association with receipt of influenza vaccination was seen for having received vaccination recommendation by HCW. The presence of recommendations by HCW was found to be a strong influencing factor on influenza vaccination uptake during pregnancy also in studies from Australia and the United States [30,45,46].

Comparison of associated factors based on Andersen's model
Our study was based on the theoretical framework of Andersen's behavioural model of health services. While the receipt of vaccination recommendation was mainly influenced by 'predisposing characteristics' , with the variable country of birth and 'need' with the variables high-risk pregnancy and season of pregnancy onset, the uptake of vaccination was affected by the domains 'need' with the season of pregnancy onset and 'enabling resources' with the variables health literacy and receipt of vaccination recommendation.

Strengths and limitations of the study
A strength of the present study is its large sample size leading to a higher statistical power. Another strength of this study is the consideration of many different variables and thereby the availability of a comprehensive data record, out of which many potential influencing factors could be analysed. Our study has some limitations that need to be addressed. Out of the total study enrolment with a number of 2620 participants, only 1886 women could be included in final statistical analysis. This might weaken the representativeness of the study population. Furthermore, women with high school degree are overrepresented. This phenomenon is known from other surveys, where women with higher education level or socioeconomic standing are more likely willing to participate in studies like that [47]. Moreover, the exclusion of women with insufficient German language skills might reduce its external validity. However, the study team tried to include all women whose German language skills were sufficient to participate. Another limitation is the fact that we had to rely on self-reported data, both for predictor and outcome variables. We were not able to verify them with medical records or similar objective data collection with a thus arising potential of recall bias. Possibly, recommendation was only reminded by women, who got vaccinated against influenza subsequently and socially desirable responses are given preferentially. However, previous studies suggest that self-reported information about having received or not received influenza vaccination can be considered reliable [48,49]. As a further limitation we need to address that we might have missed women who were vaccinated prior to pregnancy onset, but within the same season. This might have led to an underestimation of number of women who got vaccinated, as these women were not eligible for vaccination during pregnancy. Since there is no recommendation for influenza vaccination for young healthy women in Germany, unless they are working in the medical field, we assume that the vast majority of women did not receive influenza vaccination before pregnancy.

Conclusions
This study showed that influenza vaccination coverage for pregnant women is low. Lack of vaccination recommendations by HCW can be considered the main reason, as only 20,3% of women received the recommendation while it was simultaneously found to be the strongest influencing factor on vaccination uptake by far.
Efforts should be made to increase the number of HCWs who recommend the influenza vaccination to their pregnant clients. On the one hand, awareness among gynaecologists/obstetricians towards the importance of influenza vaccination during pregnancy must be raised. On the other hand, gynaecologists/ obstetricians should also be enabled and supported to inform their patients properly about vaccination, e.g., in terms of providing them with information material or guidelines for communication with vaccination scepticism.