gms | German Medical Science

Background and aim: The use of complementary therapies (CT) in both medical and non-medical settings have been growing in popularity for prevention and treatment purposes. Although the main factors driving the increased use of CT have remained unknown, recent surveys indicate an increased interest in CT use among patients suffering from psychological and chronic health problems [1], [2], [3]. We aimed to describe the patterns of CT use in the German National Cohort (NAKO), and to classify user groups based on their CT use.

Methods: CT use was collected during the NAKO baseline assessment at NAKO’s Berlin Mitte Study Center between 2016-2019. The baseline assessment included an interview, self-completed questionnaires, and a range of medical examinations [4]. Additionally, a local CT questionnaire assessed the frequency of using seven CT modalities (acupuncture, neural therapy, yoga, tai chi/qigong, meditation/mindful breathing, osteopathic manipulation, manual therapy/ chiropractic treatment). The CT user groups were defined first using a priori definitions (groups were defined by medical experts based on the type and frequency of used modalities). Secondly, groups were classified using exploratory methods, i.e. agglomerative hierarchical clustering utilizing Ward’s minimum variance criteria to link pairwise clusters with most similarity among all points, leading to creation of clusters and agglomerative trees.

Results: We analyzed data of 3,605 NAKO participants (53.5% female). Both a priori and exploratory approaches yielded 5 distinct groups of CT users each. The 5 a priori formed groups consisted of 1) never users (N=941), 2) single category[KS1] , rare users (N=283), 3) multiple categories rare users (N=121), 4) single category frequent users, and 5) mixed categories users (N=426), while the cluster analysis yielded groups of 1) mostly CT non-users (N=1111), 2) only yoga users (N=94), 3) mixed users of active modalities (N=363), 4) mixed users with highest percentage of manual therapies users, and 5) mixed users (N=161).

Both mixed user groups reported the highest percentage of pain in the last 4 weeks, had the highest percentage of mental health diagnosis, especially chronic depression, and the highest percentage of pain disorders (back pain, osteoporosis, arthrosis), while having the lowest percentage of smokers among them compared to all the other groups (a priori and exploratory).

Both classification methods also led to the creation of a group of predominantly chiropractic (a priori) or manual therapy users (clusters), who also reported higher pain levels, less activity on the job market, and who were, on average, older than CT users in other groups.

Overall, when comparing the groups in terms of demographic, socio-economic characteristics, and disease diagnosis, we observed that there was a much higher percentage of men among non-users, while any CT users had higher reported mental health diagnosis.

Conclusion: The employment of a priori and exploratory approaches allows for different ways of looking at the data to analyze patterns of CT use. Both approaches yielded groups with certain similarities, potentially uncovering some generalizable user groups and their characteristics. Most differences among groups were in terms of age, gender, pain related and mental health diagnosis, as well as education and employment activity.

The authors declare that they have no competing interests.

The authors declare that an ethics committee vote is not required.