Continuous variables are given as mean ± standard deviation (SD), or median and interquartile range, and categorical variables as frequencies (percentages). Normality of distribution was assessed by the Shapiro–Wilk test. Continuous variables were tested for differences with the Mann–Whitney U test or the Wilcoxon signed-rank test, and categorical variables with the Pearson’s χ² or the Fisher’s exact test. Crosstab analysis revealed patterns, correlations, and trends among categorical (nominal or ordinal) variables. Decision trees (DTs) were used as a non-parametric supervised learning method for classification and regression. The goal is to create a model that predicts the value of a target variable by learning simple decision rules inferred from the data features. We applied Exhausted Chi-squared Automatic Interaction Detection (Exhausted CHAID) and Classification and Regression Trees (CART). Variable importance for predictors in decision trees, such as CHAID and CART, are based on measures of sensitivity. The sensitivity of a variable is a measure of the amount of output variance that is removed when we learn the true value of the predictor. Normalized importance is simply the importance values divided by the largest importance values and expressed as percentages. Variable (normalized) importance as measured by the sensitivity analysis does not predict the order in which predictors appear in the decision tree. In this study, we do not report demographic weighting (age, gender, education, federal state, and city size). The bias would only be minimally reduced, and we found no statistically significant differences between weighted and unweighted results. All statistical analyses were performed using R (R Foundation, version 4.3) and SPSS (IBM^® SPSS^® Statistics, version 29).

A small percentage (1.6%) of the whole study population reported having used Ayurveda services more than once, while 7.6% had used them at least once. The majority (85.1%) had never utilized Ayurveda services (Figure 1).

Participants’ associations with Ayurveda predominantly include Indian Medicine (27.7%) or wellness (18%), a large group (25.4%) had a lack of specific associations. Esoteric concepts, on the other hand, were less frequently linked, with only 7.3% mentioning them. Keywords like massage, nutrition, spirituality, or spices were rated lower than 7%. Among Ayurveda users (n = 377), the most frequent association was also Indian medicine (36.3%), followed by wellness (20.4%), massage (15.4%) and nutrition (13%). Among non-Ayurveda users the most rated categories were Indian medicine (26.8%), followed by wellness (17.8%), esotericism (7.8%) and massage (5.7%). The association with Ayurveda differs highly significantly between the participants with or without Ayurveda services (p < 0.001; Figure 2).

The 377 Ayurveda users stated that they had used a total of 666 Ayurveda services. 42.2% had used Ayurveda products, 39.8% non-medical Ayurveda treatments, 24.9% had had Ayurveda nutritional consultation and 22.3% had experienced medical Ayurvedic treatments, another 22.3% had undergone Ayurveda lifestyle consultation (Table 2).

The Ayurveda users reported accessing Ayurveda services most frequently at wellness resorts (spa; 48.3%), in an outpatient practice (physician or alternative practitioner; 27.1%), and hotels (23.6%). A smaller percentage (9.5% or lower) experienced Ayurveda services online, via educational institutions or other (Figure 3).

6.3% of all participants “definitely” believed in the medical therapeutic benefits of Ayurveda and 23.9% responded “probably.” Additionally, 27.8% responded with “neutral” and 26.6% had “no opinion” on it, while 10.8% thought of potential benefits as “not likely” to exist or materialize. Another 4.5% believed that Ayurveda “definitely” had no benefits (Figure 4).

Significantly (p < 0.001) more participants who had already used Ayurveda believed in its therapeutic benefits than participants who had no previous experience with Ayurveda. Thus, Ayurveda users answered “definitely” (28.4%) or “probably” (40.1%) in comparison to non-Ayurveda users (4.1% or and 22.3%, respectively; Supplementary Figure 1).

Looking at subgroups who believed that Ayurveda had major therapeutical benefits, 47.4% of the participants who had a “mostly positive” or “very positive” overall attitude toward TCIM also rated the therapeutic benefits of Ayurveda as positive (definitely/probably). If these participants in addition stated that they were spiritual and older than 20 years, then the percentage of these participants who viewed the therapeutic benefits of Ayurveda positively (definitely/probably) increased to 68.1%. Conversely, if the participants’ global attitude toward TCIM was negative (mostly or very negative) or if they said they “did not know,” then only 7.6% had a positive attitude toward the benefits of Ayurveda. If these participants also stated that they were not spiritual and younger than 40 years, then this percentage was reduced to 5.9%. Further details are provided in Supplementary Figures 2A–C.

In the total population, 17.9% considered themselves somewhat spiritual and 4.1% as very spiritual, while 38.9% claimed they are not at all spiritual (Supplementary Figure 3). When examining the 9.3% participants who used Ayurveda services, 50.4% were identified as very (13.5%) or somewhat spiritual (36.9%) and 9.8% as not spiritual (Figure 5B). Conversely, among the 90.7% participants who had not previously used Ayurveda services, 19.2% described themselves as very (3.2%) or somewhat (16.0%) spiritual and 41.9% as non-spiritual. Further details in Figures 5A,B.

Moreover, if we compare the religious affiliation, we get a significant difference (p < 0.001) between Ayurveda users and Ayurveda non-users (Supplementary Figure 4).

Ayurveda users and Ayurveda non-users can be classified using a set of rules or equivalently by decision trees based on nine predictors with an accuracy over 90%. These predictors are in descending order of importance (Figure 6).

Other possible predictors like “gender” (normalized importance 0.5%) or “global attitude to TCIM” (1.7%) are no longer significant in this multivariate model in contrast to univariate analyses. The following example from this decision tree shows the increasing percentage of Ayurveda users by gradually adding various predictors (Figure 7). Further details of this decision tree are shown in the Supplementary Figure 5.

76.7% of Ayurveda users found healthy nutrition important or very important. This proportion is significantly (p < 0.001) higher than in comparison to non-Ayurveda users (Supplementary Figure 6). There is also a significant (p < 0.001) difference in the choice of diet depending on the use of Ayurvedic services (Supplementary Figure 7). 21.5% of the Ayurveda-users characterized their nutrition as vegetarian, vegan or raw food-vegan/based. 35.3% characterized themselves as flexitarians.

Sinus-Milieus^® (available for more than 50 countries) are a social model, which summarizes participants with similar values, a similar lifestyle and a comparable social situation in groups of “like-minded people.” The transitions between the milieus are fluid. The Sinus-Milieus^® are defined by the social situation (ranging from low to high) and the value orientation (ranging from traditional to postmodern) (44). 39.5% of the Ayurveda users belonged to the Society’s Leading Milieus which consist mostly of following three milieus: Post-Materialist (13.5%), Conservative-Upscale (13.3%) and Performer (12.7%). 25.2% belonged to the Modern Mainstream [Adaptive-Pragmatic Middle Class (11.1%), Consumer-Hedonistic (9.5) and Precarious (4.5%)]. Only 3.7% of Ayurveda users came from a traditional milieu (Figures 8A,B).

The distribution of the Sinus Milieus^® differed significantly (p = 0.001) between Ayurveda users and Ayurveda non-users. The distribution of Sinus Main Milieus^®, on the other hand, is borderline between both groups (p = 0.058).

This is the first study on the use and perception of Ayurveda in Germany including a large study population. Limitations of the study include a relatively low response rate at 21.5%, raising minor concerns about the applicability of the findings to a broader population. To enhance generalizability, the data utilized for analysis underwent weighting based on factors such as age, gender, education, federal state, and city size (42). However, no statistically significant differences between the weighted and unweighted data could be found, which suggests a sufficient quota system (42).

The phenomenon of low response in surveys is multifaceted and warrants thorough examination. Research has shed light on various factors that contribute to low response rates. For instance, Dillman et al. emphasized the impact of survey design and question quality on participant willingness to engage (47). Additionally, Singer and Kulka highlighted the significance of adequate incentives in boosting response rates (48). Demographic characteristics also play a role with factors such as age, education level, and employment status influencing survey participation (49). These findings underscore the importance of conducting a comprehensive analysis of factors influencing response rates to ensure accurate interpretation of survey results and to inform strategies for improvement. The study utilized an online access panel for surveying, chosen for its high-quality standards in participant selection and maintenance, as well as the implementation of a quota system (50). This approach ensures population-representative insights into the utilization and acceptance of TCIM in the German population. The online approach was selected due to the sensitivity of personal health-related questions. It is important to note that the use of an access panel led to the exclusion of certain populations, like those without online access or with low online affinity. However, this exclusion is not unique to access panels or online surveys in general. The potential under-representation of older people in this study had only a minor impact given the lower likelihood of them having internet access, as this age group was excluded from the analysis sample.

The study provide interesting data on Ayurveda use in Germany, however a limitation was that we did not ask specific details, e.g., regarding the types of Ayurvedic treatments and therapies used by participants. Without this information, it is challenging to determine which Ayurvedic interventions are more or less popular, limiting the insights into the aspects of Ayurveda that are in demand. While the survey mentions common locations for accessing Ayurvedic services, it does not provide a detailed breakdown of the preferences for these locations. Understanding why individuals choose one location over another could provide further context. These and other detailed questions would help us to better understand Ayurveda user demands in Germany. The data did also not contain information about which health conditions or diseases participants sought Ayurvedic treatments for. Understanding the specific health issues for which Ayurveda is being utilized in Germany could offer valuable insights into its effects and potential areas for further research. Moreover, side effects of Ayurveda were not asked. However, these limitations of our study can be addressed in future cross-sectional studies.

Finally, this data-set from a cross-sectional survey does not allow any causal statements. Also, the survey does not offer an in-depth comparison between Ayurveda and other traditional or complementary healthcare options, making it difficult to assess its relative popularity or effects. Those limitations highlight the need for more comprehensive and specific research to fully understand the role and efficacy as well as effectiveness of Ayurveda in the German context.

Ayurvedic interventions used in wellness resorts/spa facilities could presumably reflect the pattern that health-conscious people without an acute medical need are more likely to use low-threshold entry options and use Ayurveda for health-maintenance, prevention and well-being. While the use of Ayurveda as a wellness service could be a way to reach a wider audience, it could serve the medical visibility and credibility of Ayurveda to differentiate these services from explicit medical and therapeutic Ayurveda-options for specific health needs (51).

Almost one-third of the population (30.2%) believes in Ayurveda’s therapeutic potential, while an almost equivalent portion remains neutral or has no opinion on this. Moreover, participants which used Ayurveda in the past rated its potential benefit considerably higher. Scientific evidence in the field of Ayurveda is still at an infant stage in Germany and the European Union; only few methodologically high-quality trials which proof to live up to high (−quality) standards concerning the implemented methods were published so far (52–55). More methodologically high-quality research, professional communication with the public and public support (e.g., governmental, financial) would be necessary to define potential roles of Ayurveda as a complementary treatment option within the German health care system. Overall, there is a need for more clinical evidence, particularly effectiveness studies, to collect real-world data on Ayurveda interventions outside its countries of origin (56). Although available studies show promising results for Ayurveda in the field of clinical research for some diagnoses, the current state of research is still insufficient, especially against the background of the growing utilization of Ayurvedic services in Germany and the EU. Foremost, more high-quality clinical research is needed at German and European universities to be able to answer questions about the effectiveness of Ayurvedic therapies under local conditions. In this context, AYUSH exchange programs with foreign universities to promote Ayurveda, yoga and other traditional Indian medicine could be a relevant building block for the further development of the academic infrastructure required for this (12).

Spirituality, which appears to be associated with Ayurveda use, indicated by the fact that half of Ayurveda users appears to have a spiritual attitude (57, 58) is notable and requires further attention. Especially as the conception of “spirituality” may vary enormously on a social cultural level between the cultures of origin (India/Germany) and even within the respective milieus. The connection between Ayurveda and spirituality can be traced back to the close historical relationship between Ayurveda and Buddhism, Hinduism and Indian culture in general (59, 60), first described in the Caraka Samhita [see Gupta (61); Caraka Saṃhitā Śārīrasthāna chapters 1 and 5 (62)], which is one of the oldest systematic text collections of on Ayurveda. Previous publications explored the role of religion and spirituality in medical contexts, emphasizing that these elements may influence attitudes and choices regarding the use of Ayurvedic healthcare services. Spirituality plays a crucial role in how Ayurveda is perceived, and in India it coexists with modern medicine (63–66). In a paper of our working group associations were identified between individuals’ religious or spiritual affiliations and their choices to either provide or seek access to Ayurveda (67). Importantly, the utilization of Ayurveda did not preclude concurrent use of both modern medicine and TCIM (67). Also, in combination with yoga, Ayurveda has a long tradition in India of mind (−fullness) training, breath work or ethical-philosophical recommendations for everyday life, which also has found a “mainstream-compatible approach” e.g., into fitness- and yoga-studios (11, 68). In a systematic review by Jeserich et al. on the correlation between religion and spirituality (R/S) and the sense of coherence (SOC), a connection was found between R/S and mental health (64). Relevant effect sizes were found in relation to the potential resource of spirituality. These relationships with a sense of coherence, which is an important indicator of the capacity of resilient coping with difficult situations (such as coping with illness), was stronger the more semantically open and non-institutional the spiritual belief system was. The results support the link between R/S and SOC and point to different religious/spiritual pathways to a strong SOC that are influenced by individual and cultural factors (64).

Individuals who utilize Ayurveda services mostly belong to the Society’s Leading Milieu (35.5%) with basic values being shaped by either tradition, modernization or re-orientation (69). At the same time, it is noteworthy that Ayurveda utilization can be found across all milieus, with a relatively consistent percentage ranging from 9 to 13.5%. However, the Traditional and Precarious Milieus may be apparent as individuals into whose lives Ayurveda has not yet made significant inroads (44). While milieu-concepts might have been a useful tool for understanding social groups, there could be valid concerns regarding their relevance and potential drawbacks. One of these concerns is the risk of oversimplification and stereotyping, as milieus tend to categorize individuals based on specific characteristics or behaviors (70). India’s pluralistic healthcare system does not explicitly promote TCIM, but the various traditional health systems covered by the Ministry of AYUSH (Ayurveda, Yoga and Naturopathy, Unani, Siddha, Sowa Rigpa and Homeopathy) (71). Ayurveda is widespread in India as a folk medicine and is practiced alongside Western medicine with diverse Ayurvedic trends and currents (18, 72). Ayurveda also faces “medicalisation,” meaning standardization, professionalization and pharmaceuticalization, which may impact the education, knowledge, practice and narrow the holistic view of traditional Ayurveda approaches (73, 74). The Ministry of AYUSH seeks to integrate diverse local health traditions, as well as traditional and complementary medicine into a modern healthcare system (12, 75). The European research network CAMbrella for complementary and alternative medicine (CAM) that operated between 2010 and 2012 and aimed to assess the situation of CAM in Europe, addressed that more research is needed and the integration of evidence-based or -informed CAM treatments into the Western healthcare systems. The findings emphasize the high demand for CAM, its heterogeneity, and the challenges in evaluating its effects due to insufficient integration and lack of validated data. Traditional Indian medicine is only mentioned in passing in one work package to provide a global perspective (76, 77).

The World Health Organization (WHO) is actively promoting the integration of traditional medicine into global health systems, emphasizing evidence-based and scientifically validated methods to guarantee the safety, qualification, and effectiveness of traditional, complementary, and integrative medicine (TCIM) services (20). Key points of the WHO vision include the development of norms and standards, the use of data and analysis to shape policies and regulatory frameworks, the promotion of sustainability and the integration of TCIM. This vision could be an impulse that could trigger corresponding measures at legal and political level for Ayurveda in Germany and Europe (20).

More high-quality data are needed to gain a deeper understanding of the use and impact of Ayurveda in Germany, given its growing popularity and potential to complement modern medicine. Furthermore, differentiation between therapy and wellness, and health policy measures are needed to establish scientific credibility, translate Ayurvedic concepts into mainstream western medical thinking and provide access to a wider audience. In addition, potential risks and side effects of Ayurveda should be investigated with adequate methodology. In addition, further methodologically high-quality studies are required to evaluate efficacy and effectiveness of Ayurveda in and outside its countries of origin.