Background
Older age is strongly associated with disease severity and case fatality resulting from
Streptococcus pneumoniae (
s. pneumoniae) [
1], influenza virus infection [
2], and infection with
Clostridium tetani. [
3] Severe disease outcomes from
S. pneumoniae include invasive pneumococcal diseases (IPD) [
4], as sepsis, pneumonia [
5], and meningitis [
6], whereas influenza virus infection is characterized by high fever, aching muscles, headache, cough, sore throat and rhinitis. Tetanus mostly presents as a spastic disease, followed by spastic paralysis and death. In Germany, reported incidence of IPD and influenza infection in 2014 was 1.35/100.000 and 5.92/100.000, respectively in the age group 60 years and older. There was no reported case of tetanus [
7]. Since there is no nationwide mandatory reporting for IPD in Germany, reported incidence is likely to be substantially underestimated [
8,
9]. Vaccines against all three diseases are available and German vaccine recommendations for adults aged ≥60 years include the 10 year interval tetanus and diphtheria vaccination, the annual influenza vaccination and a singular pneumococcal vaccination, using the 23-valent PPV dose, while PCV is also licensed [
10]. The self-reported vaccination rate in the German 60–79 years old population varies by vaccine type and ranges from 31 % for pneumococcal, to 66 % for influenza and 70 % for tetanus vaccination [
11]. While a national campaign promotes influenza vaccination [
12], no comparable strategies are in place for pneumococcal or tetanus vaccination, focusing on the elderly.
Health-related behavior is affected by different aspects of knowledge, attitude and practices (KAP) [
13]. Reviewing the existing evidence from 69 international publications on determinants of vaccine-usage in the elderly, we found the following to be most frequently significantly associated with vaccine uptake: socio-demographic determinants such as living arrangement [
14], low awareness of the vaccine recommendation [
15,
16], attitudes like the perceived low severity of the corresponding disease [
17,
18], and practices including previous uptake of vaccinations [
19]. Little is known about determinants of uptake of vaccinations in the older German population and how they vary by vaccine type. Furthermore, most studies investigated associations of single variables particularly with influenza vaccine uptake, while less attention is paid to officially recommended tetanus and pneumococcal vaccination.
Using data of a representative national survey on infection prevention, the objective of this study was to assess associations between socio-demographic- and KAP-factors and vaccine uptake in order to identify determinants of vaccine uptake in the elderly. Using multivariate and factor analyses, knowledge-, attitude- and practice-related predictors of tetanus, influenza and pneumococcal vaccination were analysed. Given the high information load created when assessing KAP-variables on an individual level, we tested and applied statistical methods to create scores of variables with statistically significant associations.
Discussion
This study is the first one focusing on KAP related to three different vaccinations, recommended for the elderly in Germany. Most vaccine-related surveys concentrate on parents of children [
25] or on adolescents [
26]. Elderly are less frequently the target group of investigations on vaccine-KAP. However, given their vulnerability for severe infectious disease outcomes [
1‐
3] and different exposures to infectious agents (e.g., through their living environment), hygiene behavior and infection prevention including vaccination is of relevance in this age-stratum. We found great variations in predictors of vaccination in the elderly, depending on vaccine type considered, which may allow conclusions on vaccine type-specific public health action needed. Overall, practices such as possessing a vaccination record were predictive for uptake of all vaccines analysed, while the attitude towards vaccination was particularly associated with the influenza vaccine usage. For pneumococcal vaccination, knowledge was positively associated with vaccine utilization. Furthermore, by aggregating individual KAP-variables into three scores, we showed that for any, influenza, and pneumococcal vaccination the generated scores equally reflect the information from the individual variables. For tetanus vaccination, however individual variables of attitude were most important but on a score level, the practice score was the only significant predictor. This could imply that general vaccine attitudes do not influence the uptake of tetanus vaccination, but only those attitudes related to this specific vaccine. On the other hand, general vaccination-related practices as possessing a vaccination record were associated with higher tetanus vaccination uptake. No individual knowledge-related variable was related to the uptake of the tetanus vaccination, which differs from pneumococcal vaccination. As we also found that most elderly in Germany (almost 80 % of the participants) were well informed about the existing recommendation, promotion activities regarding tetanus immunization may rather focus on individual attitudes than on information provision in general. Exclusively for the influenza vaccine, the attitude-score showed a significant influence on uptake, corresponding with the strong association with attitude-related single variables. This may reflect the controversial public discussions regarding the usefulness of this vaccine. As all scores were associated with uptake of the influenza vaccination, all areas of KAP might be equally important targets for promoting influenza vaccination.
Different from influenza vaccination, where attitude-related variables served as best predictors for vaccine uptake, in case of the pneumococcal vaccination, individual attitude-related variables did not have a notable impact on the pneumococcal vaccination and less than 30 % of the participants knew about the recommendation. This contrasts with the almost 80 % who were aware of the influenza vaccination recommendation. Based on these findings, the main reason for the low uptake of the pneumococcal immunization among the elderly in Germany might be the lack of awareness and not a general refusal of this vaccine. Also in other risk groups knowledge about pneumococcal disease and the vaccination was found to be low [
27]. The association of knowledge with uptake of the pneumococcal vaccination has been shown in previous studies. Schneeberg et al. [
28] illustrated that the strongest predictor for pneumococcal vaccination was the recommendation by a health care worker (HCW) and the knowledge of the vaccine [
28]. Similarly, Johnson et al. [
29] found that 79–85 % would get vaccinated with pneumococcal vaccine if a HCW would offer and recommend it. In another study, 91.3 % of the participants who were not vaccinated against pneumococci stated this was because they did not know about the recommendation or their physician did not recommend it [
30].
The impact of knowledge and awareness on pneumococcal vaccine uptake shown from our study could warrant nationwide campaigns for pneumococcal vaccine in Germany, or implementation of public health measures as done for influenza vaccine. The potential need for vaccine information and promotion activities is additionally reflected by the relatively low vaccine coverage shown in our study, particularly for pneumococcal vaccine. The coverage in our study was also lower compared to those reported from previous investigations in Germany. For example, Poethko-Müller et al. [
11] reported coverage of about 70 % for tetanus vaccine, about 66 % for influenza vaccine and about 31 % for pneumococcal vaccine in individuals aged 60–79 years [
11] versus 56.3 % for tetanus vaccine, 50 % for influenza vaccine and 11.5 % for pneumococcal vaccine in individuals aged 60–85 years in our study. These differences might result from the fact that in our study it was only asked for vaccinations received during the last 5 years. As tetanus vaccination is generally given in a 10 year interval and pneumococcal vaccinations just once, the answers from our study may not reflect “coverage” as such.
The main strength of our study is the representativeness of the sample, providing information on vaccine-related KAP among the older population of Germany, for the first time. Generating all three KAP-scores based on factor analyses has, to our knowledge, not been applied in the area of vaccination-related KAP-surveys in the elderly but may assist in drawing conclusions from numerous KAP-specific variables without losing important information. Similar results obtained from the variable- versus score-specific analyses furthermore suggest that the scores adequately reflect individual variables in each of the three KAP categories. Up to our knowledge, only one published study used scores based on factor analysis to investigate KAP related to vaccination against Human Papilloma Virus [
31]. Despite the similar approach used in this investigation, attitude- and belief-related determinants but no variables on knowledge and practice were included. There was no direct comparison made between score- and variable-based results, which is, in our opinion, crucial to determine the validity of results. Having said this, we assessed each multivariate model by calculating Nagelkerke’s r
2 and found a high degree of explained variance from 20.6 to 64.7 %, meaning that a great amount of the outcome can be explained by the determinants in our models.
As a limiting factor and inherent in the design of a cross-sectional study, we were not able to assess the timing of the vaccinations and its direct predictors. Thus, it is possible that immunization status itself has influenced knowledge or attitudes, which both might have been different before getting vaccinated. Since public awareness activities mainly target influenza vaccination in Germany, this possibly also affected knowledge and attitudes towards this vaccination but also the self-reported vaccine-uptake for this vaccination. Although the attitude-related questions targeted some personal beliefs, for example the trust in official vaccine recommendations, there was no question precisely assessing risk perceptions such as vaccine safety concerns or perceived seriousness of infection in detail. This information may have provided additional information for interpreting our results and should be included in follow-up investigations of this kind.
Despite the availability of safe vaccines, the proportion of vaccinated elderly is low and beyond recommended coverage in Germany. This is furthermore remarkable with regard to demographic changes and the impact vaccine-preventable diseases have on the health outcomes of the elderly. In our study population vaccination was recommended to only 29 % of the participants by their physician. Thus, the result that vaccine uptake in the elderly is strongly influenced by practices like getting a vaccination consultation, is of high importance to health care providers. The significance of the physician- patient- interaction based on knowledge and attitudes of the health care professionals for the uptake of vaccination in this age-strata, has also been shown in another study [
32].
Conclusions
We analysed associations of KAP with the uptake of specific vaccinations in the elderly in Germany, using logistic regressions and score building within a representative sample. Our findings suggest that there are different patterns for the different vaccine types, leading to the necessity of specific public health actions for each vaccine. While for the pneumococcal vaccine the knowledge about the recommendation needs to be addressed, for the influenza vaccine mainly the attitudes towards the vaccination are of interest and for the tetanus vaccination general vaccination-related practices, as possessing a vaccination record, should be tackled. In general, many elderly have not received advice regarding recommended vaccinations by their physician. This results in a need for integrating health care workers in the research of determinants of vaccine uptake in the elderly in Germany and in the design of related promotion activities, taking into consideration implications of varying vaccine performance, i.e., efficacy of each of the vaccines in general and in the elderly [
33‐
38].
Ethics approval and consent to participate
All procedures performed in this study involving human participants were in accordance with the institutional and national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. The original collection of the data was performed on behalf of the Federal Centre for health education (BZgA), in accordance to the ESOMAR-Code. As data used for these analyses did not include individual personal data or personal identifiers, ethical approval was not required.
Availability of data and materials
Data and questionnaires used were available from the Federal Centre for Health Education (BZgA) (Head 1985–2015, Professor Pott. Dataset and questionnaires were sent on behalf of Professor Pott). In timely alignment with this study, the data set has been made available as a public use file in November 2015 on GESIS - Leibniz-Institut für Sozialwissenschaften: Datenbestandskatalog.
Competing interests
All authors declare that they have no competing interests.
Authors’ contributions
CJKT initiated and performed the statistical analyses. JJO coordinated the study and analyses and provided comments and consultation on all aspects of the study. GK provided technical expertise and advice on data interpretation. LS provided relevant background information on the original dataset and the survey methods used. CJKT composed the initial manuscript with contributions from JJO; GK and LS commented. All authors are equally responsible for the content of the manuscript and have read and approved the final manuscript.