COVID-19 vaccine hesitancy in people with migratory backgrounds: a cross-sectional study among Turkish- and German-speaking citizens in Munich

An anonymous survey was conducted in a cross-sectional design. Study reporting was done per the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guideline [19]. The study was approved by the ethics committee for the medical faculty of the Technical University of Munich (January 20, 2021, Number: 37/21 S-EB). An information letter explaining the study details was handed over to all participants. As approved by the ethics committee, no written consent was deemed necessary; filling in the questionnaires was interpreted as an agreement to participate. All methods were implemented in accordance with the ethical guidelines and regulations of the host institution.

The study was conducted within February 2021 in Munich. Participants were German- and Turkish-speaking patients of Turkish-speaking family physicians. Munich is a city with 1.472 million citizens [20], of which around 40,000 are of Turkish origin [18].

A list of the 612 family physicians working in Munich was obtained from the Bavarian Association of Statutory Health Insurance Physicians “Kassenärztliche Vereinigung Bayerns (KVB),” which was reviewed for Turkish names. Additionally, an internet search and snowball inquiry was made by contacting the Turkish-speaking family physicians, which returned a list of 10 such physicians. The principal investigator visited all targeted family doctors in their offices and inquired about their willingness to support the study. As a result, six family physicians agreed to distribute the study questionnaires to their patients. Inclusion criteria were being the age of 18 and above, sufficient knowledge of the German or Turkish language, and willingness to volunteer to participate. Participants who did not disclose their demographic information or had more than 50% missing items in the questionnaire were excluded from the analysis. A flow diagram of physician and patient participation is given in Fig. 1.

The primary outcome measures of the study were COVID-19 knowledge scores and intention to get vaccinated for COVID-19. Other study variables were related to attitudes and behaviors regarding COVID-19 and demographic information (age, sex, citizenship, native language, place of birth, and total years of schooling).

Citizenship/nationality, native language, and place of birth were considered as additional variables for subgroups to take the complex construct of a “migration background” into account. Thus, participants with either citizenship/nationality, country of origin, native language, or place of birth being non-German were categorized as “Having a migratory background.”

The researchers developed the study questionnaire after reviewing the literature for common knowledge and guideline recommendations related to COVID-19. Two team meetings were conducted to refine the questionnaire items. In addition, a pilot test on five Turkish and five German-speaking participants was conducted to further modify the questionnaire. The final questionnaire consisted of a demographic information section (six questions) and three subdomains, including knowledge, attitudes, and behaviors concerning COVID-19.

The knowledge domain contained 25 items, which were scored as true or false. Hence, the COVID-19 knowledge scores were calculated by summing up the correct answers, providing a minimum of 0 and a maximum of 25.

The attitude and behavior domains included both seven items arranged in a five-point Likert scale (1 = disagree/never, thru 5 = agree/very frequent). Item numbers 2, 3, 5, and 6 in the attitude domain and items 4, 5, 6, and 7 in the behavior domain were reversely coded. Attitude and behavior scores were calculated by adding the scores of each item and dividing by 7 (the total number of items), which revealed the minimum and maximum possible scores of 1 and 5. Higher scores indicated more sensitive attitudes and behaviors.

The two-page paper questionnaire was made available in German and Turkish.

The practice personnel was asked to consecutively approach patients speaking German or Turkish and invite them to participate in the questionnaire survey for recruitment.

All interested individuals were given the patient information letter and the study questionnaire. If the individual either spoke Turkish or German (or one language much better than the other), they received the material in the appropriate language. If the individual spoke both languages, they were asked for their preference. Questionnaires were self-administered in a silent place and collected after completion by the practice personnel or the primary author.

Each participant was asked to fill in the questionnaire alone without interference from others. After digitalizing the data, error-checking and debugging were done to eliminate questionnaires with missing data or conflicting information. In addition, an effort was given to minimize selection bias by recruiting concomitant patients. However, although the principal investigator spent time in the practices reassuring data collection according to the protocol, this was not always possible due to the local conditions and the rules and regulations discouraging prolonged stay of patients in practice and accommodating multiple patients at a time.

A sample size of n = 400 in the cross-sectional study was determined to be sufficient for consistent estimation of the coefficients of a multiple logistic regression model, including the factor variable ‘group’ and the demographic variables for adjustment. Further, the sample size is efficient for subgroup analyses (e.g., vaccination hesitancy, migratory background, etc.) [21].

The data were entered into the IBM SPSS Statistics spreadsheets (IBM Corp, Armonk, NY). The data distribution was described within and across the study groups by frequencies, percentages, means, and standard deviations (SD), as appropriate. Cronbach’s alpha was computed to assess the reliability of the items belonging to the knowledge, attitude, and behavior domains. Corresponding hypothesis-testing of univariable group differences was performed by Chi-squared tests, Fisher’s exact tests (or Fisher–Freeman–Halton test), and independent samples t-tests or Mann–Whitney-U tests. Multiple binary logistic regression models were fit to the data to adjust the effect estimation of group differences by potential predefined confounders, such as demographics. Significant variables affecting vaccination intention in the univariate analyses were included in the model. Hypothesis testing was performed at exploratory two-sided 5% significance levels.

Of the invited patients, 452 agreed to participate. Questionnaires with missing demographic data (n = 2) were excluded. Also excluded were a total of 27 participants with more than 50% missing items in the knowledge (n = 9), attitude (n = 23), or behavior (n = 20) domains, and 3 participants with missing migratory background information. Data of 420 participants were analyzed (women: 41.4% (n = 174), mean age: 42.2 ± 15.5 years) (Fig. 1).

Of the participants, 245 (58.3%) preferred the Turkish questionnaire. In addition to these 245 people, 103 (58.8%) of the 175 participants who preferred the German questionnaire had a migratory background, making 348 (82.8%). There were 72 (17.1%) German participants without a migratory background. Of the 348 patients with a migratory history, 90 (25.8%) were German nationals. Of the 71 participants with a migratory background who were born in Germany, only 42 (59.1%) had German nationality. In the case of those with migratory backgrounds, there were significant differences in terms of age, sex, and preferred language. Among the participants with migratory backgrounds, prominent countries of origin were Turkey, Bulgaria, and Iraq, while Turkish, Bulgarian, Kurdish, and Arabic were the most commonly spoken native languages (Table 1).

Responses to the scale items are summarized in Table 2. On average, the knowledge level regarding COVID-19 was high, but looking at attitudes revealed that there was a relatively low fear of death due to the disease. Also, very few participants thought that faith would play a role in protection from the disease. Behavior scores, on the other hand, were comparatively higher (Table 2). Cronbach’s alpha reliability coefficients for the knowledge, attitude, and behavior domains were 0.732, 0.695, and 0.716, respectively.

A total of 197 participants (47.9%) intended to get vaccinated. However, there were significant differences in the mean knowledge, attitude, and behavior scores associated with a migratory background. Most participants with a migratory background were not planning to be vaccinated and had lower scores in all three subscales. However, COVID-19 knowledge scores were relatively high for both groups (Table 3).

Of the 411 patients who indicated their intentions to either accept or refuse vaccination, 253 (61.5%) expressed 1–3 reasons for their thoughts (a total of 294). After categorization of the free texts, the three most common reasons were self-protection (n = 49), concerns about safety or mistrust in vaccines (n = 45), and the perception that vaccines were not sufficiently studied (n = 25) (Fig. 2).

Of the people with a migratory background, 12.6% (n = 44) agreed or somewhat agreed with the statement that COVID-19 actually does not exist. This belief was held by only 5.6% (n = 4) among the non-immigrant participants (Chi-square = 11.623, p = 0.020). Also, the number of participants believing that the COVID-19 was purposely created to control the world was higher among respondents with a migratory background (30.6% of migrant applicants (n = 106) agreed or somewhat agreed vs. 5.6% (n = 4) of non-migrants) (Chi-square = 33.020, p < 0.001).

Responses to the question “Will you get vaccinated against COVID-19?” were organized into two categories namely “No + Not sure” and “Yes.” Positive vaccination intentions ranged from 26.9 to 68.5% between different practices.

All investigated variables were related to the vaccination intention. It was also noted that more participants with a preference for a German questionnaire had vaccination intentions than those preferring a Turkish questionnaire (n = 99/57.9% vs. n = 99/40.6%, Chi-square = 12.091, p = 0.001), and men had higher intentions to get vaccinated compared to women (Table 4).

A multiple logistic regression model was fit to the data to estimate and test the relation of predictors of vaccination intention. Independent predictors of the model included sex (male/female), migratory background (yes/no), age (years), duration of schooling (years), a previous infection (yes/no), knowledge score, attitude score, and behavioral score.

The multivariable analysis changed the significance levels. Age, male sex, years of schooling (borderline significant), migratory background, and high attitude scores positively affected the vaccination intention (Table 3). The most critical variable was the migratory background with an odds ratio (OR) of 3.1, followed by attitude scores (OR 2.9) and sex (OR = 2.2).

This study found significant differences among patients with and without migratory backgrounds in the studied sample concerning COVID-19. While 82.8% had a migratory background, 42.1% preferred the German questionnaire. Furthermore, while 42.3% of the participants with a migratory background were considering becoming vaccinated, this proportion was 76.5% for non-immigrant Germans. On the other hand, this group’s mean knowledge, attitude, and behavioral scores were relatively high. After correcting for potential confounders, migratory background, increased attitude scores, male sex, longer schooling, and higher age positively affected vaccination intention.

This study approached patients of Turkish-speaking general practitioners (GPs) in Munich. Hence, its interpretation should consider this context. Slight differences could be expected if the sample had been selected from all GPs, including native German physicians, which might limit the generalizability of the results. As another potential limitation, we may mention that the survey questionnaire was not validated before. However, as described in “Methods” section, an effort was given to standardizing the items. Furthermore, all three subscales revealed reasonable internal consistency findings.

People originating from Turkey constitute the highest proportion of citizens with foreign origin in Germany, followed by Poland and Syria [22]. However, they often find themselves not so highly appreciated and accepted [23, 24]. Considering the participants’ mean age of 42.2 years, attention is needed for those with immigrant backgrounds and who were born in Germany but still did not obtain German citizenship. The relatively high proportion of this group (40.9%) confirms a lack of successful integration even in the third generation.

Although there was a recent influx of immigrants from Turkey (around 40,000 within 5 years) [8], they constitute a minor proportion; the majority are the offspring of the first immigrants after 1961 [7]. Hence, the healthy immigrant effect [10, 11] seems to be negligible here. Our study demonstrated significant shortcomings among people with migratory backgrounds concerning the intention to be vaccinated as well as knowledge, attitudes, and behaviors concerning COVID-19. Thus, interventions are highly required to close the health-related gap in this group of citizens. Besides the efforts of the German state, projects initiated by the Turkish-speaking community may accelerate the integration process [25].

People of Turkish origin make up the largest migration group in Germany, so their different social, political, and health aspects have been widely studied. Therefore, we would like to present some more information on the situation of Turkish migrants and their descendants in Germany to understand the broader social dynamics that could explain the distrust of vaccines. This segment of the German population experiences particular social disadvantages and displays a decline in some aspects of health across generations [26]. Examples of health-related disadvantages include the lack of infrastructure for religious practices, gender-specific treatment options, and nutritional preferences in hospitals [27]. In addition, policies such as the test for Muslim citizenship applicants [28] may have contributed to the introversion of Turkish immigrants and their rejection of Western values.

People’s acceptance of immunization against COVID-19 was a concern even before specific vaccines became available. In particular, nursing staff in Germany have been reported to be hesitant to get vaccinated. Although they are in the highest priority group, only 46.6% had been vaccinated as of April 2021 [29]. It is worth mentioning that 23.2% of these healthcare workers have a migratory background [30].

Presently 70% of the German population is expected to accept the COVID-19 vaccination [31]. Disagreements among health professionals and politicians [32] as well as fake news [33] account for a substantial part of the confusion in public. However, research shows a high degree of hesitation rather than direct opposition to COVID-19 vaccination [33], which is similar to the results from our case: Of the 52% who intended not to be vaccinated or were not sure, 33.3% were hesitant to make a decision, while 18.7% were refusers. At this point, the reasons for vaccine refusal should be noted.

Access to healthcare was proposed as another barrier to vaccination, especially for unregistered immigrants [34]. Lack of access combined with a lack of information about the contents and side effects of the vaccine could lead to a low perceived need, causing vaccine hesitancy. Therefore, hesitation should be addressed with clear, accessible, and individualized information campaigns. Furthermore, ethnic minorities and insecure migrant populations have low intent and uptake of vaccine services [35]. The mistrust might have its origins in the sporadic racist attacks on immigrants, particularly those with Turkish roots. From this perspective, involving trainers from the same community could increase the possibility of acceptance and help people embrace the information that is transferred.

The proportion of participants with an immigrant background who refused vaccination in our study was similar to figures reported from Turkey. In a study conducted on 272 women and 156 men, 66.1% considered not getting vaccinated against COVID-19, and women were less likely to be willing to get a vaccine than men [36]. Lower vaccination rates among women compared to men have been reported by other researchers [37].

As to our findings regarding the motives for accepting or rejecting vaccination, it can be speculated that the rapid developmental process of the COVID-19 vaccines and the global immunization campaigns without satisfying information transfer to the public has contributed to the concerns about the safety of vaccines or raised mistrust in vaccines and the perception that vaccines have not been sufficiently studied. Looking at the distributions of the reasons for accepting or rejecting vaccination in our study, negative news, conflicting information, and conspiracy theories rank in third place. Of the people with a migratory background, 12.6% believed that COVID-19 does not exist, compared to 5.6% among the non-immigrants. An even higher number of immigrants (30.6%) assumed that COVID-19 was purposely created to control the world.

Combining the above information with integration tardiness, we may deduct that Turkish immigrants in Germany are substantially affected by external negative factors. It was reported that 6.17% of adults in Western Turkey received no lifetime vaccination at all and anticipated a further decline in vaccination rates [38]. In fact, vaccine hesitancy proportions in Turkey are growing dramatically. The number of families who signed a vaccine rejection form increased from 183 in 2011 to 12,000 in 2016 and 23,650 in 2017 [39]. As of the 12th of September 2021, the complete (two doses) vaccination rates of Germany and Turkey were 62.3% and 49.6%, respectively [40]. The comparatively lower vaccination readiness of the Turkish immigrants in Germany might be attributed to experiences of discrimination and racism as well as a mistrust in public authorities. Although it is possible that Turkish migrants are exposed to information sources from Turkey rather than Germany, it could also be that they are more exposed to the type of German information sources that spread misinformation on vaccines because they rely more on social media and have less trust in the government.

Bearing in mind that a significant number of participants were critical or suspicious against vaccination, the relatively high knowledge scores suggest an informed decision. Participants scored 21.5 points out of a maximum of 25. In other words, those who are against COVID-19 vaccination or have reservations devised their opinions possibly not because of a lack of knowledge but rather due to their attitudes shaped by factors not addressed in our study. A multivariate analysis endorses this conclusion. After correcting for potential confounders, the most significant two variables affecting the decision to be vaccinated were the migratory background (OR 3.1) and attitude scores (OR 2.9). Qualitative studies could ascertain the reasons behind the negative attitudes concerning the COVID-19 vaccination.