The public impact of academic and print media portrayals of TMS: shining a spotlight on discrepancies in the literature – BMC Medical Ethics – BMC…

We used qualitative methods to study the ethics of the presentation of TMS in the literature. In order to assess the publicly available information regarding TMS, we performed an extensive database search that allowed us to analyze the full body of relevant texts that have been published from 2014 to 2019 in both the academic and print medialiterature. Focusing first on the academic literature, we used the PubMed database to search for Transcranial Magnetic Stimulation and the relevant MESH termsFootnote 2 in the period from January 1, 2014 to December 31, 2019. From this, we yielded 2273 articles for our retrieved sample. We then applied the following exclusion criteria to the retrieved sample: (1) papers without an English abstract or available full text in English; (2) papers detailing techniques other than TMS; (3) papers reporting the use of TMS during surgery or on patients under anesthesia; (4) TMS applied to nonhuman models (e.g., rat models); (5) TMS applied to areas other than the brain/outside of typical TMS scope; and (6) unpublished proofs. These criteria led to the exclusion of 641 articles, leaving a final count of 1632 relevant papers.

Applying similar techniques, we used the NexisUni database to gather a comparable print media sample. Print media sources were selected exclusively for three primary reasons. First, focusing on print media allowed us to create a manageable dataset for analysis. For a simple comparison, as of 2018, there are approximately 1300 daily print newspapers in the United States [28] compared to the ever-increasing volume of news sources online, which range from the online platform of The New York Times to social media accounts [29]. Second, print media remains more conducive to systematic study owing to the refined filtering capacities of databases at university libraries and platforms like NexisUni. Third, print media remains better preserved in accessible archives. A recent study showed that a quarter of links of a major media outlet like The New York Times were corruptedmeaning the links were dead and the linked pages were either deleted, changed, or moved without HTML redirection [30]. Focusing on print media afforded the opportunity to create a more stable and more replicable archive, which will benefit future studies.

We used the search terms transcranial magnetic stimulation and (enhancement or therapy) from the period of January 1, 2014 to December 31, 2019, and limited our search to the categories of newswires & press releases, newspapers, and magazines & journals in order to gather the full body of print media articles. This search yielded 1420 print media articles. Here, our exclusion criteria were as follows: (1) sources targeting medical professionals; (2) irrelevant article types (e.g., obituaries, clinical trial reports, economic/market reports); (3) articles primarily detailing another neurostimulation/treatment technique; (4) duplicates; (5) online-only articles; and (6) articles unavailable in English. After applying these criteria, the final print media sample was 468 relevant papers (952 excluded). Our final sample thus consisted of 2100 articles in total, which we kept separated into an academic sample and a print media sample.

We organized both the academic and print media samples by publication date, sorting them from oldest to most recent, and selected every tenth article from these two compilations to create two pilot samples. Doing so provided us with a sample that was 10% of the size of the overall body of data (163 academic articles and 47 print media articles), making a more manageable sample for analysis that would allow us to establish inter-coder reliability while still allowing us to draw some conclusions about the larger sample as a whole.

The articles were coded independently by two coders (AS and JM for the academic sample, AS and LO for the print media sample), with a third coder (VD) consulted to settle any discrepancies. Our coding structure included the identification of (1) type of source, (2) year of publication, (3) purpose of TMS application, (4) population, (5) overall tone, and (6) specification of TMS parameters. If they were specified in the article, we further coded for what specific parameters were given (i.e., specific forms of TMS, like repetitive TMS (rTMS); frequency of TMS being applied; shape of the TMS coil; and brain region targeted by TMS). The coders then convened to discuss edits that needed to be made to the methods and finalized the coding structure for the larger project sample.

Ultimately, our coding structure consisted of seven mandatory coding categories for the entire 2100 article sample: (1) type of source, (2) year of publication, (3) purpose of TMS application, (4) age of subjects (5) population, (6) overall tone, and (7) specification of TMS parameters. Figure1 shows a more extensive breakdown of these code classes and the subnodes that we coded for under each one. These codes were selected because they each pinpointed some point of variability in either the application or the presentation of the technology, best leading us to where the discrepancies in the TMS universe might exist. Namely, since TMSs reach as a technology is broad, the categories of type of source, year of publication, purpose of TMS application, age of subjects, and population help narrow the specific areas of interest to scientists and the public. Overall tone and specification of TMS parameters allow for examination of the attitudes and specificity the contributing authors take in their writings. Each of the seven codes was marked in every article of the 2100 article sample, and each code was assigned only one subnode per article. Additionally, each of the subnodes was assigned a numerical valueFootnote 3 (also shown in Fig.1 in bold) to allow us to perform additional analyses on our results across the whole sample.

Breakdown of the main codes with quantitative identifiers. Dark blue bubbles represent the seven main codes. Light blue outlined bubbles branching down from the dark blue bubbles represent the subnodes that may be assigned in each main coding category

In the subsections below, we provide a brief description of each code and their respective subnodes for the purpose of clarity before presenting our results.

The code for type of source, comprised of the subnodes academic or print media, corresponds to the broad bodies of literature we sought to evaluate in our search. Articles obtained from PubMed were marked as academic, while articles obtained from NexisUni were marked as print media. This code was used to divide our sample into the two subsamples to be analyzed.

This code, comprised of the subnodes 2014, 2015, 2016, 2017, 2018, and 2019, indicates the year that each article was published. The code was intended to illuminate the volume of articles being published each year regarding TMS, which we then take as a quantifiable proxy for measuring popular interest in TMS technology. For consistency in our coding, the academic articles are coded based on the year that they were published in print rather than the year they were published online.

The code for purpose of TMS application, comprised of the subnodes enhancement, diagnostic/therapeutic, technical, and investigative, was designed to show how TMS is being employed in application (cf. [1]). With it, we sought to understand which applications of TMS are of most interest to both scientists and the general public. Articles coded as enhancement indicate that TMS was being used in healthy populations for the purpose of giving people physical, mental, or emotional abilities beyond their natural baseline state (e.g., enhanced memory capabilities). Articles coded as diagnostic/therapeutic indicate that TMS was being used either to diagnose or to treat a disease or ailment in non-healthy populations, such as for the treatment of major depressive disorder. Articles coded as technical indicate scenarios where TMS was being used or operated to change or better the technology itself. For example, several articles reported on simulated TMS on computer models to compare the effectiveness of different coil shapes in TMS. Finally, articles coded as investigative indicate that TMS was being used as a measurement device or probing tool for the sake of understanding different biological processes. Investigative uses include mapping the localization of brain function, understanding disease progressions without the goal of incorporating TMS into treatment, and other such uses.

This code, comprised of the subnodes N/A, adult, and children/adolescents, refers to the age range of subjects for whom TMS was being used. The age of subjects helped us determine the general age population that is represented in academic or print media literature on TMS technology in its various applications to help us understand the technologys current target audience. The subnode N/A includes both articles in which no population was used, such as those that demonstrated TMS with a computer model, and review articles, which demonstrated TMS broadly and did not focus on any one particular age range of people. The adult subnode refers to articles which either specified a population of 18years old or older or else centered around a condition that necessitated an older population (e.g., Alzheimers disease). The children/adolescents subnode, in contrast to this, refers to articles that specified a focus on children, adolescents, or otherwise young people in their discussion.

The population code, comprised of the subnodes N/A, healthy subjects, mental health, motor function/chronic condition, neurodevelopmental/neurodegenerative disorders, addiction, and miscellaneous, underscores the broad classes of issues being addressed by TMS. This further narrows the group of people who interact with or benefit from TMS technology in its various applications by illuminating which categories the technology served during this period according to the articles in our sample. Articles coded as N/A indicate articles that, as stated above, used no population or reviewed several populations. Articles coded as healthy subjects indicate those articles in which no underlying condition was being targeted and TMS was being used on healthy individuals. Articles coded as mental health indicate that TMS was being used in populations suffering from some form of mental illness or psychiatric condition, such as depression or OCD. Articles given the subnode motor function/chronic condition refer to populations suffering from some kind of physical deficit (e.g., stroke) or who are afflicted with an ongoing condition (e.g., fibromyalgia or Parkinsons disease). Articles coded as neurodevelopmental/neurodegenerative disorders indicate TMS was being used to address one of those two types of conditions, including attention-deficit/hyperactivity disorder (ADHD), autism, and dementia or Alzheimers disease. Articles coded as addiction indicate TMS was being used to address both addictive social behaviors (e.g., gambling) and substance addictions (e.g., smoking, heroin), and finally, the miscellaneous subnode covered articles that addressed any condition which did not fit into the prior five codes (e.g., obesity). Articles coded miscellaneous did not receive their own subnode because they occurred in such small numbers (less than 0.5% each) with no discernable link to one another such that their occurrence could not be reported as a trend.

This code, comprised of optimistic, critical, and neutral/balanced subnodes, contains our assessment of the attitude of each article in our sample. We shifted our focus to what attitude on the whole authors took towards the subject of TMS in order create a top-level understanding of the sentiments of both the print media community and scientific communities. Optimistic indicates an overtly positive stance towards TMS technology and refers to articles which either presented only the benefits of TMS and not the drawbacks (e.g., side-effects), or used an abundance of hype language, such as calling TMS a miracle cure or life changing (see, e.g., [31] in the academic sample and [32] in the print sample). Critical indicates an overtly negative stance towards TMS and refers to articles that either overly stressed the drawbacks or dangers of TMS and failed to present the benefits of the technology, or used doom and gloom (cf. [21]) language about the technology, such as calling TMS a hoax (see, e.g., [33] in the academic sample and [34] in the print media sample). Finally, neutral/balanced articles adequately presented the benefits of TMS as well as the drawbacks and used no hyperbolic language, contributing to an overall realistic picture of the TMS technology.

Our final code, specification of TMS parameters, was comprised of only two subnodes, unspecified and specified, and it targeted the degree of specificity authors were using with regards to different TMS paradigms. By ascertaining the degree of specificity, we hoped to understand how well-founded the authors conclusions were in scientific backing. As mentioned earlier, we set aside four key pieces of TMS paradigms that we coded where applicable (general TMS paradigm, TMS frequency, shape/orientation of TMS coil, and brain region targeted). Here, articles coded as unspecified indicate that none of the four parameters we outlined were mentioned in the article. In contrast, articles coded as specified mentioned at least one of the parameters.

Visit link:

The public impact of academic and print media portrayals of TMS: shining a spotlight on discrepancies in the literature - BMC Medical Ethics - BMC...