A Large-Scale, International Cross-Sectional Survey of Published Pediatrics Authors: Perceptions of Complementary, Alternative, and Integrative Medicine

Transparency Statement

Clearance was obtained from the University Hospital Tübingen Research Ethics Board before commencing this research project (REB Number: 389/2023BO2). The study protocol was registered and made accessible via a platform called the Open Science Framework (OSF) before participant recruitment began [25]. OSF was also used to share study materials and raw data, which can be found at: https://doi.org/10.17605/OSF.IO/Z6U4X.

Study Design

This study was an anonymous, online, cross-sectional survey of published pediatrics authors who have published in medical journals indexed in MEDLINE [26].

Sampling Framework

A sample of corresponding authors from all articles published between December 1, 2018 and July 1, 2023 were selected from a sample of pediatrics journals as found here: https://journal-reports.nlm.nih.gov/broad-subjects/. National Library of Medicine (NLM) IDs associated with the selected journals were extracted [27]. A search strategy was generated using these NLM IDs and subsequently, run on OVID Medline. The list of PMIDs compiled from articles identified in the search was exported as a .csv file and inputted into an R script to extract the authors’ names and email addresses. The R script has been designed in accordance with the easyPubMed package [28]. Authors who had published any type of manuscript were included.

Participant Recruitment

Published pediatrics authors were identified using the sampling framework to participate in this study by completing the survey. Duplicate email addresses were removed. The survey was closed, meaning that only invitees had access to this survey, and they were instructed not to share the survey link with anyone else. Emails were sent out to the prospective participants via SurveyMonkey [29]. Numbers of invalid and non-functional emails which bounced back were recorded by SurveyMonkey. The authors in our sample received an initial email on either October 30, 2023 or October 31, 2023. This initial email included a description of the study, the objectives, and a link to access the survey. Clicking on the survey link led the participants to an informed consent form. After the participants provided consent, participants were taken to the survey questions. The survey contained 33 questions in a one-by-one order, spanning over 12 pages. Reminder emails were sent out after the first, second, and third weeks following the initial invitation. The survey was closed on December 19, 2023, exactly 4 weeks after the last reminder was sent out. Financial compensation was not provided for participation. Moreover, there was no requirement for those who were emailed a survey link to partake in this study. Participants who did not wish to answer any given question were able to skip it and no personal identifying information was collected.

Survey Design

The survey began with a screening question. Respondents were asked several questions covering demographic information and their perceptions of CAIM. The survey questions primarily followed a multiple-choice format and an open-response question. Respondents were able to use the back button to change answers prior to submitting the survey. The survey was pilot tested prior to distribution by two independent CAIM researchers and their feedback was integrated into the survey. Since these researchers are not affiliated with this project and are highly experienced with CAIM, their feedback serves to improve the face validity of this survey. A copy of the survey can be found on the following link: https://osf.io/4d23w.

Data Management and Analysis

Descriptive statistics were generated from the analysis of the quantitative data and reported. Qualitative data was analyzed using thematic content analysis. A data driven approach was taken to analyze the responses to the open-ended question, “do you have anything else to share about your perceptions of CAIM?”. Open-ended responses were interpreted and assigned a code. These codes were representations of respondent responses and then were further grouped into themes. Themes were created by the authors based on the patterns and commonalities observed in the data. An inductive coding method was used with no specific, underpinning theories to guide analysis. The Checklist for Reporting Results of Internet E-Surveys (CHERRIES) was used to inform the reporting of this survey study [30].

Demographics

A total of 39 434 emails were sent out of which 22 412 remained unopened, 12082 were opened and 4394 bounced. The response rate was 1.85% including bounced emails, 2.12% excluding bounced emails, and 6.05% excluding bounced and unopened emails. There were 731 responses which were broken down into 604 fully completed surveys and 127 partially completed surveys, with a completion rate of 82.6%. Respondents took an average of 9 minutes and 37 seconds to complete the survey. Approximately 55.64% (n=365) of the respondents were female and the most common age of respondents was between 35 to 44 (n=180, 27.44%) or 45 to 54 (n=178, 27.13%). Most respondents identified as both a clinician and a researcher (n=391, 55.23%). From the respondents who specified their roles, respondents were primarily faculty members/principal investigators (n=368, 56.10%), clinicians (n=285, 43.45%) and/or scientists in academia (n=162, 24.70%). Most respondents did not identify as being a part of a visible minority (n=545, 82.95%) and were most commonly from the following World Health Organization regions: Americas (n=305, 46.56%) and Europe (n=200, 30.53%). In terms of the respondent’s stage of career and type of research conducted, most respondents had been working for more than 10 years (n=406, 61.89%) and conducted clinical research (n=439, 76.61%). Table 1 contains a detailed breakdown of respondent demographics. Raw survey data is available at: https://osf.io/43xz5. Crosstabs for key demographic variables can also be found on OSF: https://osf.io/z6u4x/.

Complementary Alternative and Integrative Medicine

To learn more about CAIM, clinicians and researchers stated they would turn to academic literature (n=544, 88.74%), conferences (n= 263, 42.90%) and/or colleagues (n=215, 35.07%). See Figure 1 for breakdown of where pediatrics clinicians and researchers would source CAIM information. In a clinical setting, respondents have practiced or recommended the following therapies (in descending order): mind-body (n=181, 45.14%), none (n=148, 36.91%) and biologically based (n=135, 33.67%). Most respondents had received no formal (medical school, residency, etc.) or supplemental training (webinar, conference, etc.) on CAIM therapies. See Figure 2 for a breakdown of the types of CAIM therapies in which respondents had formal or supplemental training. The percentage of respondents that believed one particular CAIM modality was the most promising for pediatric populations varied: mind-body (n=382, 62.01%), biologically based (n=275, 44.64%), manipulative and body-based (n=145, 23.54%). Patients that were treated by respondents sought counselling for and disclosed the use of biologically based (n=300, 74.26%), mind-body (n= 210, 51.98%) and manipulative and body-based therapies (n= 204, 50.50%). Approximately 55% of respondents (n=220) said that 0-10% of their patient population sought counselling for or disclosed use of CAIM with the majority of respondents having been asked about CAIM occasionally (n=370, 59.97%) as opposed to often (n=97, 15.72%) and never (n=150, 24.31%). Table 2 contains a detailed breakdown of CAIM research and clinical experiences. Respondents see several benefits and harms to using CAIM. Respondents perceived a lack of scientific evidence for safety and efficacy (n=519, 93.01%) and a lack of standardization in product quality and dosing (n=482, 86.38%) as the greatest challenges with CAIM. Perceived benefits of CAIM included a holistic approach to health and wellness (n=352, 64.83%), expanded treatment options for patients (n=344, 63.35%) and a focus on prevention and lifestyle changes (n=344, 63.35%). Table 3 contains a detailed breakdown of respondent perceptions as they relate to the challenges and benefits of CAIM.

• Download figure
• Open in new tab

Figure 1:

Sources of Information for Pediatrics Clinicians and Researchers

• Download figure
• Open in new tab

Figure 2:

Formal and Supplementary CAIM training in Pediatrics Clinicians and Researchers

Participants were asked about their opinions on various prompts and were requested to indicate their level of agreement on a scale ranging from strongly disagree to strongly agree. With regards to CAIM therapies in general, approximately one in three respondents disagreed in feeling comfortable recommending (n=123, 33.79%) and counselling (n=124, 34.07%) patients. Approximately one in three respondents expressed neutrality (neither agreed or disagreed) when asked whether CAIM therapies should be covered by insurance companies (n=178, 31.50%), whether CAIM therapies should be integrated into mainstream medical practice (n=188, 33.39%), if CAIM therapies are safe (n=214, 37.88%) and if CAIM therapies are effective (n=238, 42.20%). Respondents most commonly agreed that clinicians should receive training on CAIM therapies through supplementary (n=292, 51.87%) and formal (n=250, 44.17%) education, more research funding should be allocated to study CAIM therapies (n=218, 38.58%) and there is value in conducting research on CAIM therapies (n=284, 50.18%). Respondents most commonly strongly disagreed with feeling comfortable recommending most CAIM therapies to patients (n=113, 31.04%) and most commonly strongly agreed with ensuring that there is value in conducting research on CAIM therapies (n=200, 35.34%). See Figure 3 for a detailed breakdown of pediatrics clinicians and/or researchers’ perceptions of CAIM. Please see Figures 4-8 for detailed breakdown of results by specific CAIM modalities.

• Download figure
• Open in new tab

Figure 3:

Agreement with the Following Statements Regarding CAIM in General

• Download figure
• Open in new tab

Figure 4:

Agreement with the Following Statements Regarding Mind-Body Therapies

• Download figure
• Open in new tab

Figure 5:

Agreement with the Following Statements Regarding Whole Medical Systems

• Download figure
• Open in new tab

Figure 6:

Agreement with the Following Statements Regarding Biofield Therapies

• Download figure
• Open in new tab

Figure 7:

Agreement with the Following Statements Regarding Manipulative and Body-Based Practices

• Download figure
• Open in new tab

Figure 8:

Agreement with the Following Statements Regarding Biologically Based Practices

Thematic Analysis

A total of 129 responses were grouped into 52 codes that were further grouped into 7 themes. These themes represent distinct patterns found within the data [31]. The most common theme across all responses was the “need for further research on CAIM”. This was presented as a lack of quality and quantity of current CAIM research. Next, “integration and regulation of CAIM within the conventional healthcare system” included the opinions of many respondents who felt CAIM needs more regulation, and that CAIM practices should be integrated into the health system. Next, “patient safety, legal and ethical concerns” reflects respondent concerns about deceiving patients, spreading misinformation, fostering false hope, replacing conventional medical therapies and compromising safety in infants, and believing that those who provide CAIM should face legal consequences. Next, “diverse perspectives of CAIM” encompassed the varying perspectives respondents presented such as supporting or opposing certain CAIM practices. Next, “observational insights and outcomes of CAIM use” included insights and outcomes respondents had observed in a clinical or personal setting. For example, some respondents observed patient improvement following CAIM use, interest in CAIM, and increase in CAIM use following the COVID-19 pandemic. Next, the theme of “education and training for clinicians” was reflected by respondent beliefs that healthcare clinicians and researchers need to be familiar with CAIM practices and support for physicians receiving adequate, appropriate, and evidence-based training on CAIM. Lastly, “cultural and personal considerations’’ were presented as the support or opposition of the integration of religious practices and the consideration of cultural support of CAIM. Please see Table 4 for themes and examples. Coding and thematic analysis data are available at: https://osf.io/729cv

Comparative Literature

Our findings align with preexisting literature on healthcare clinicians’ and researchers’ perceptions of CAIM. There are three similar studies which investigate clinician and researcher perceptions of CAIM within the fields of oncology, neurology, and psychiatry [32–34]. Respondents in these fields generally were not comfortable recommending and counselling patients about CAIM but agreed that clinicians should receive more formal and supplemental research on CAIM and that more research funding should be allocated for CAIM. Specifically, mind-body therapies were the most favourably viewed CAIM therapies by clinicians and researchers across these three fields. The respondents were most comfortable recommending and counselling patients on these mind-body therapies. Interestingly, our study found that pediatrics clinicians and researchers were neutral (neither agreed nor disagreed) towards the safety of CAIM practices as opposed to oncology, neurology and psychiatry clinicians and researchers in which respondents were most likely to agree that CAIM therapies were safe [32–34].

Our thematic analysis found a key theme of cultural and personal considerations of CAIM. A study exploring the perspectives of pediatrics healthcare clinicians and researchers (HCCRs) about CAIM found that Indonesian HCCRs were more likely to recommend most CAM therapies as compared to Dutch HCCRs [39]. Our demographic data shows only 14.65% of respondents were from South-East Asia and the Western Pacific regions. These regions may see CAIM therapies in a much more positive perception and this can influence the comfort levels that respondents from these regions have in recommending CAIM therapies [35]. Additionally, individuals from South-East Asia are more likely to seek out CAIM training and this can further enhance their positive perceptions of CAIM [35]. In previous literature assessing attitudes of Swiss, American, and German pediatricians and pediatrics subspecialists such as pediatrics oncologists, a common theme of preferring to receive more CAIM education was found in 82%, 84% and 86% of respondents, respectively [36–38]. These results were seen in pediatrics residents as well, where 88% of respondents reported preferring to increase their knowledge of CAIM [39]. Similarly, our results demonstrate that respondents agree clinicians should receive formal and supplementary education across almost all CAIM therapies. These survey responses align with thematic findings which highlight a theme of a ‘lack of education and training” regarding CAIM. Further research should focus on tailored educational resources specific to pediatrics clinicians and researchers to be reflective of the pediatric patient CAIM interest and use.

Our results highlighted that some perceived challenges with CAIM include a lack of integration within mainstream healthcare systems. Pediatricians have reported that they rarely ask patients about their CAIM use [40]. As CAIM therapies can interact, antagonistically or synergistically, with conventional medications [41,42], physicians must be open to initiating a conversation about CAIM therapies with their patients to ensure they are aware of all the therapies that the patient may be pursuing. A previous study exploring pediatrician experiences with CAIM therapies show that pediatricians most commonly referred patients to mind-body therapies [18]. This could help explain why the most preferred CAIM therapy in our study was mind-body therapies and why our respondents were most comfortable recommending and counselling patients about mind-body therapies such as biofeedback. Pediatricians are significantly less likely to recommend manipulative and body-based practices such as chiropractic therapy as compared to general physicians [43]. This could explain why respondents in our study were unsure about the safety and efficacy of manipulative and body-based practices and why they were hesitant to recommend and counsel patients about these practices.

Strengths and Limitations

A strength of this study is that by exploring the perceptions of pediatrics researchers and clinicians using a large, international sample of participants with varying opinions on the topic, the findings of this study will be generalizable. Moreover, emails were sent to researchers and clinicians who published an article recently. By doing this, we limited the number of invalid or inactive email addresses. There was also a high survey completion rate among those who initiated the survey. This study also has some limitations. First, our sample may only include individuals who can understand English, as the survey and email were sent in English. Therefore, this study might not capture the perceptions of non-English-speaking researchers and clinicians. Secondly, given the self-reported, cross-sectional nature of the survey, our analysis is subject to recall bias, which refers to the varying accuracy and completeness with which different participants remember information and experiences, in this case regarding CAIM [44]. Thirdly, the categorization of CAIM therapies on the survey was broad, allowing for the grouping of various therapies. While this allowed for a generalized approach, many respondents expressed that the categorization needs to be narrower. For example, manipulative and body-based practices cover massage, chiropractic therapy, reflexology etc. As such, future work should explore perceptions of specific therapies. It is also important to note that perceptions towards CAIM modalities may be due to a variety of concerns ranging from lack of evidence and safety to a lack of knowledge or training and future research should explore concerns surrounding specific perceptions. Unfortunately, as responses were collected anonymously, it is not possible to analyze the characteristics of those who participated versus did not participate in our survey. Additionally, the phrasing "most CAIM therapies" in some survey questions may have introduced a bias toward more negative responses. Instead, if participants were asked about "CAIM therapies with a certain level of evidence", responses might have been more positive. This highlights the influence of wording on survey results and should be considered when interpreting our findings. Finally, the study also contains non-response bias, especially with a low response rate, as authors in our initial sample who declined participation may have answered the questions in our survey differently, resulting in a potential difference between participants and non-participants [44].