bims-librar Biomed News
on Biomedical librarianship
Issue of 2020–03–15
eleven papers selected by
Thomas Krichel, Open Library Society



  1. J Biomed Inform. 2020 Mar 06. pii: S1532-0464(20)30027-7. [Epub ahead of print] 103399
       OBJECTIVE: The centrality of data to biomedical research is difficult to understate, and the same is true for the importance of the biomedical literature in disseminating empirical findings to scientific questions made on such data. But the connections between the literature and related datasets are often weak, hampering the ability of scientists to easily move between existing datasets and existing findings to derive new scientific hypotheses. This work aims to recommend relevant literature articles for datasets with the ultimate goal of increasing the productivity of researchers. Our approach to literature recommendation for datasets is a part of the dataset reusability platform developed at the University Texas Health Science Center at Houston for datasets related to gene expression. This platform incorporates datasets from Gene Expression Omnibus (GEO). An average of 34 datasets were added to GEO daily in the last five years (i.e. 2014 to 2018), demonstrating the need for automatic methods to connect these datasets with relevant literature. The relevant literature for a given dataset may describe that dataset, provide a scientific finding based on that dataset, or even describe prior and related work to the dataset's topic that is of interest to users of the dataset.
    MATERIALS AND METHODS: We adopt an information retrieval paradigm for literature recommendation. In our experiments, distributional semantic features are created from the title and abstract of MEDLINE articles. Then, related articles are identified for datasets in GEO. We evaluate multiple distributional methods such as TF-IDF, BM25, Latent Semantic Analysis, Latent Dirichlet Allocation, word2vec, and doc2vec. Top similar papers are recommended for each dataset using cosine similarity between the dataset's vector representation and every paper's vector representation. We also propose several novel re-ranking and normalization methods over embeddings to improve the recommendations.
    RESULTS: The top-performing literature recommendation technique achieved a strict precision at 10 of 0.8333 and a partial precision at 10 of 0.9000 using BM25 based on a manual evaluation of 36 datasets. Evaluation on a larger, automatically-collected benchmark shows small but consistent gains by emphasizing the similarity of dataset and article titles.
    CONCLUSION: This work is the first step toward developing a literature recommendation tool by recommending relevant literature for datasets. This will hopefully lead to better data reuse experience.
    Keywords:  Cosine similarity; Gene Expression Omnibus (GEO); Literature recommendation; Re-ranking; Vector space model
    DOI:  https://doi.org/10.1016/j.jbi.2020.103399
  2. Nature. 2020 03;579(7798): 193
      
    Keywords:  Publishing; Research data; Virology
    DOI:  https://doi.org/10.1038/d41586-020-00694-1
  3. Fam Med Community Health. 2019 ;7(4): e000142
      
    Keywords:  health literacy; health promotion; information dissemination; patient medication knowledge; practice guidelines as topic
    DOI:  https://doi.org/10.1136/fmch-2019-000142
  4. J Med Internet Res. 2020 Mar 10. 22(3): e16685
       BACKGROUND: When searching for health information, many people use the internet as their first source of information. In online health forums, for example, users post their questions and exchange health advice. In recent years, information givers from various professions have begun to use positive language (indicated by the frequent use of positively valenced adjectives) to communicate their information and persuade their audiences.
    OBJECTIVE: The goal of the current study was to answer the following research questions: (1) How does positive language, in comparison to neutral language, influence the trustworthiness of a person arguing in an online health forum and the credibility of their health claims; (2) How does working for a university, compared to working for a lobbying organization, influence the trustworthiness of a person arguing in an online health forum and the credibility of their health claims; and (3) Do the two factors of language style and professional affiliation interact with each other to influence trustworthiness and credibility judgments?
    METHODS: In a 2 × 2 between-subject experiment, 242 participants read a post from an online health forum and subsequently rated the trustworthiness of the forum post author and the credibility of their information. Within the post, the professional affiliation (scientist vs lobbyist) and language style (neutral vs positive) of the forum post author was varied.
    RESULTS: When the forum post author used a positive language style, they were perceived as less trustworthy (high Machiavellianism [P<.001; η2p=.076], low Integrity [P=.001; η2p=.045], and low Benevolence [P=.02; η2p=.025]) and their information was perceived as less credible (low Message Credibility [P=.001; η2p=.045]). The professional affiliation of the forum post author did not influence their trustworthiness or the credibility of their information.
    CONCLUSIONS: When searching for health information, information seekers evaluate the language style of forum posts to decide whether forum post authors are trustworthy and their information is credible.
    Keywords:  health communication; information-seeking behavior; language; occupations; trust
    DOI:  https://doi.org/10.2196/16685
  5. Curr Pharm Teach Learn. 2020 Feb;pii: S1877-1297(18)30466-0. [Epub ahead of print]12(2): 127-131
       INTRODUCTION: The literature has limited information on pharmacists working in the alternate career path of a medical information specialist in the pharmaceutical industry. The purpose of this study was to gather the background, functions, and skills of pharmacists working in the role of a medical information specialist (MIS) and their preparation for the role through their pharmacy education.
    METHODS: An electronic survey was sent to medical information departments and obtained information on educational background, prior job experience, and skills required for the MIS role. The survey also asked for suggestions on how pharmacy schools can prepare students for this role.
    RESULTS: The survey was sent to 40 medical information departments from September to December 2017 and a total of 33 pharmacists responded. Approximately half of the responders did not complete post graduate training. The role of a MIS includes answering medical information questions and collecting adverse events and product complaints. Skills rated as important to the role as a MIS were communication skills, medical writing, and literature evaluation. Four pharmacists thought their pharmacy education definitely prepared them for role in medical information. The pharmacists suggested pharmacy curriculum focus more on drug information skills, structure of the pharmaceutical industry, and providing opportunities to improve emotional intelligence and critical thinking.
    CONCLUSION: The information collected in this survey can provide valuable insight for students interested in this career path. In addition, pharmacy schools may include more opportunities during the didactic and experiential years to help prepare students for this alternate career path.
    Keywords:  Medical information; Pharmaceutical industry; Pharmacist's role; Pharmacy education
    DOI:  https://doi.org/10.1016/j.cptl.2019.11.005
  6. J Geriatr Oncol. 2020 Mar 10. pii: S1879-4068(19)30462-X. [Epub ahead of print]
       OBJECTIVES: The internet and social media provide information and support to cancer survivors, and adolescent and young adults (AYA, age < 40 years), adults, and older (age 65+ years) cancer survivors may have different needs. We evaluated the impact of age on cancer-related internet and social media use and confidence in evaluating online information for cancer-care decision making.
    MATERIALS AND METHODS: Cancer survivors completed a convenience cross-sectional survey evaluating their cancer-related internet and social media use and their confidence in using these resources for decision making. Multivariable regression models evaluated the impact of age on usage patterns and confidence.
    RESULTS: Among 371 cancer survivors, 58 were older adults and 138 were AYA; 74% used the internet and 39% social media for cancer care; 48% felt confident in using online information for cancer-care decisions. Compared to adult survivors, there was a non-significant trend for older survivors to be less likely to use the internet for cancer-care information(aOR = 0.49, 95% CI[0.23-1.03], P = .06), while AYA were more likely to use social media for cancer-care (aOR = 1.79[1.08-2.99], P = .03). Although confidence at using online information for cancer-care decision making did not differ between age groups, increasing age had a non-significant trend towards reduced confidence (aOR = 0.99 per year [0.97-1.00], P = .09). Most commonly researched and desired online information were causes/risk factors/symptoms, treatment options, and prognosis/outcomes.
    CONCLUSIONS: Age may influence the use of internet and social media for cancer-care, and older cancer survivors may be less confident at evaluating online information for cancer-care decision making. Future research should explore other strategies at meeting the informational needs of older cancer survivors.
    Keywords:  Adolescent and young adults; Cancer survivorship; Geriatric oncology; Internet use; Patient education; Social media
    DOI:  https://doi.org/10.1016/j.jgo.2020.02.011
  7. Int J Med Inform. 2020 Mar 03. pii: S1386-5056(19)31270-5. [Epub ahead of print]137 104107
       BACKGROUND: Evaluation of the accuracy, quality and reliability of written and visual health contents on the internet by healthcare professionals.
    AIMS: The objective of this study was to evaluate gastroesophageal reflux health contents in Google search engine and videos on YouTube, in terms of quality, actuality, treatment diversity, accuracy and reliability.
    METHODS: A total of 77 YouTube videos and 57 Google contents related to gastroesophageal reflux disease were included in the study. The contents were assessed by two independent reviewers. Quality of the content included in this study was evaluated using DISCERN, which is commonly used for the assessment of health information on the internet, JAMA and video power index scores.
    RESULTS: Among YouTube videos; 44.7 % (n = 34) were uploaded by experts, 44.7 % (n = 34) by non-experts, 9.0 % (n = 7) by media and 2.7 % (n = 2) by academic institutions, while 10.5 % (n = 6) of the Google contents were uploaded by experts, 59.6 % (n = 34) by non-experts and 22.4 % (n = 17) by academic institutions. The mean total DISCERN score (5-80 points) of reflux related YouTube was found as 37.2 ± 16.5. The mean total DISCERN scores of the evaluated Google contents was found as 42.1 ± 15.6. The mean JAMA score (0-4 points) was found as 2 ± 0.9 for YouTube videos and 2 ± 0.9 for Google contents. There was a strong correlation between all scores given by the two reviewers.
    CONCLUSIONS: Overall quality of YouTube videos and Google contents about reflux was low. Healthcare professionals and organizations should be encouraged to provide more beneficial material and animated videos to people who seek reliable information on the internet.
    Keywords:  DISCERN; Gastroesophageal reflux disease treatment; Google; JAMA; Reflux; YouTube
    DOI:  https://doi.org/10.1016/j.ijmedinf.2020.104107
  8. Cell Tissue Bank. 2020 Mar 11.
      The use of umbilical cord blood (UCB) holds promise for the treatment of a wide spectrum of diseases. However, information on UCB donation is not widespread or accurate among expectant women. The aim of this study is to evaluate pregnant women's knowledge of UCB donation, their main sources of information and their satisfaction with the knowledge possessed. Women (N = 375) in the last semester of pregnancy completed questionnaires evaluating sociodemographics, knowledge of UCB donation, past donation, sources of information, satisfaction with the information and the desire to have received more information. Women were aware of the possibility of donating UCB (97.5%) although, on average, they reported not having enough knowledge of donation possibilities, procedures to be followed and uses of UCB (2.51, on a 5-point scale). Considering knowledge satisfaction, 28% were not at all satisfied. Only 2.8% felt fully prepared and the great majority (75.2%) would have liked to have received more information. The main source of information was the Internet (51.2%). Gynecologists and midwives were indicated by only 24.4% and 18.6% of women, respectively. Age and education were significantly correlated with UCB knowledge. Chi-square tests evidenced that those who reported professional sources of information (gynecologists, obstetricians, prenatal courses) did not need additional information. Conversely, mothers who turned to other mothers for information were more likely to desire further information. Most mothers report the Internet as the main source of information. Providing accurate information through official sources may result in a more positive attitude toward donation.
    Keywords:  Cord blood donation; Donation; Pregnant women; Public cord blood banking; Umbilical cord blood (UCB)
    DOI:  https://doi.org/10.1007/s10561-020-09820-9
  9. Clin Rheumatol. 2020 Mar 09.
       INTRODUCTION: Patients are increasingly using internet-based information to inform healthcare utilization and treatment decisions. Our aim was to examine the quality and readability of internet-based information relating to Raynaud's phenomenon (RP) and systemic sclerosis (SSc).
    MATERIAL AND METHODS: A systematic review of three commonly used search engines (Google®, Yahoo®, and Bing®) using the terms "Raynaud's phenomenon" and "systemic sclerosis" separately. The first 30 websites per search engine were examined. Quality was assessed using the DISCERN questionnaire and readability by the Flesch-Kincaid Grade Level, SMOG Index, Coleman-Liau index, and Flesch Reading Ease score.
    RESULTS: Fifty-two studies (30 RP and 22 SSc) were included after duplicates and exclusion criteria were applied. The overall quality of information was low for both SSc and RP (1.99 & 2.21), including in relation to reliability of the literature and information on treatment choices. Readability for RP and SSc was also poor (i.e., the texts were difficult to read) across all of the four methods examined.
    CONCLUSION: Overall, RP and SSc internet-based information is of low quality and inadequate readability. The RP and SSc international community should strongly consider developing an information standard for internet-based resources for healthcare users.
    KEY POINTS: • Patients with SSc and RP are increasingly using internet/online sources of information and support.• RP represents an important opportunity for the early diagnosis of SSc.• The overall quality and readability of internet-based RP and SSc information is poor.• Internet-based RP and SSc information requires improvement to facilitate early diagnosis and inform shared decision-making.
    Keywords:  Information; Internet; Quality; Raynaud’s phenomenon; Readability; Systemic sclerosis
    DOI:  https://doi.org/10.1007/s10067-020-05023-5
  10. World Neurosurg. 2020 Mar 05. pii: S1878-8750(20)30424-1. [Epub ahead of print]
       OBJECTIVE: YouTube is currently the second most popular website in the world and thus it is often used by patients to access health information regarding their condition(s). Our aim was to evaluate the content-quality of YouTube videos relating to hydrocephalus.
    METHODS: We chose the first 35 videos for four different search phrases: "water on the brain," "hydrocephalus," "pediatric hydrocephalus" and "adult hydrocephalus." Video contents were evaluated by two independent final year medical students with more than 5 years of experience using the DISCERN criteria (scoring system from 1-5 per question). Qualitative data, quantitative data and the upload source about each video was recorded for quality and optimization analysis.
    RESULTS: Out of the total 140 videos, 63 videos met our inclusion criteria and were evaluated. The mean DISCERN score was 29.9 out of a total of 75 possible points. This indicates that the quality of YouTube videos on hydrocephalus is currently poor. Reliability between the two raters was excellent (intraclass correlation coefficient = 0.96). Most videos had clear information (90%), a doctor speaking (70%), and described the symptoms (62%). Videos were most commonly uploaded by hospitals (44%) or by educational channels (43%). Our study found that videos that contained the results of treatment had a much higher average daily view (P=0.0229) than videos that did not.
    CONCLUSION: The quality of YouTube videos on hydrocephalus is poor, however, we indicated the top-quality videos in our paper as they may be effective tools for patient education. Our optimization analysis found that including diagrams and explaining the results of hydrocephalus treatment results in a higher audience engagement (in the form of likes, comments and views).
    Keywords:  Hydrocephalus; Internet; Neurosurgery; Online; Quality; YouTube
    DOI:  https://doi.org/10.1016/j.wneu.2020.02.149