FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2006775401> ?p ?o ?g. }

• W2006775401 endingPage "64" @default.
• W2006775401 startingPage "60" @default.
• W2006775401 abstract "Medical librarians have often explored the conditions under which their services to health-care professionals positively impact clinical outcomes.1 However, providing medical library services directly to patients is less studied. Online links that use patient data to automate retrieval of information to serve the needs of clinicians have been developed,2 but no such links for patient online discussion groups have been developed. For the purposes of this study, a member of an online group: uses a software system that provides a searchable archive of previous messages, and; may create a message and post it to the group via the system. The software system may interface to the member via an e-mail client or a website. The groups to be studied are devoted to discussions from the patient's perspective about dealing with a chronic disease. Each patient group is typically focused on one disease. The groups include patients, caregivers and others. Caregivers are those people who help a patient, not as members of the health-care profession but as significant others of the patient3 (a medical professional is by contrast called a ‘care provider’). Members of a group are not obligated to identify themselves as patient, caregiver or other, and the term ‘patients’ will be used to refer to any member of a group. In these groups, patterns of online participation tend to be highly skewed, with most participants saying very little.4 Unlike face-to-face groups, online groups are asynchronous and a participant has time to find references to include in a response to a message.5 While Internet users often search for health information,6 patients in online groups also want to share empathy.7 A common topic is whether physicians are giving patients the tests or treatments that they should receive.8 Patients benefit from interactive health communication applications9 and, in particular, they benefit from their participation in online groups.10 Patient experiences are also a key to reducing errors.11 The aim of this small-scale study was to explore the question: by augmenting patient discussions with references to the literature, might patients identify when their data could help fill a gap in the literature?12 Two case studies are discussed: In case study 1, the author reflects on student experiences with providing information to patients. In case study 2, the author reflects on his own experience in online patient groups. Twenty-seven students of Information Systems were enrolled in a health-care information systems class and given an option to explore online patient groups. Students were encouraged to use Google, medlinePlus,13 or PubMed for information to help a patient. Google, medlinePlus, and PubMed cover a spectrum of web portal types whose interfaces range from easy-to-use to expert, and content ranges from general purpose to highly specialized. Other researchers have catalogued or evaluated health-care web portals from other perspectives.14 In case study 2, the author was a participant in patient groups. The author is a professor who was diagnosed with stage III, head-and-neck cancer in 2003. He participated in head-and-neck cancer online groups because of concerns about his personal health. Studying online groups via ethnography is in some ways easier than studying face-to-face groups15 and is established as a method to elicit requirements for new systems.16 The term ‘netnography’ has been coined by Kozinets to apply to such ethnography, and Kozinets said17: ‘As a method, netnography is faster, simpler, and less expensive than traditional ethnography and more naturalistic and unobtrusive than focus groups or interviews.’ The author presents an ethnographic perspective on his direct experiences and qualitatively interprets his students’ experiences. The results are presented in two subsections: (i) Strategy and (ii) Patient Reaction. The ‘Strategy’ subsection covers the types of groups, the web portals that were used and the search strategy. The ‘Patient Reaction’ subsection covers the response of the patients to the information provided. Within each subsection, the experiences of both the students and the author are presented. The students were invited to join a group on any disease. Groups for cancer and diabetes together accounted for over half the groups. Some groups, such as the American Cancer Society breast cancer group, which one student joined, had tens of thousands of members, while most groups were much smaller. The students tended to find a message that asked for information, identify keywords within that message and pose a query with those keywords to another web portal. The students consistently reported that the content of PubMed was too medically technical for them to understand and that the PubMed interface was difficult to traverse. Students were most often comfortable with medlinePlus and sometimes with Google. The students explained that Google too often led them to useless websites, whereas medlinePlus tended to take them to high-quality and relevant web pages. Ultimately, the appropriate choice of a web portal for which to retrieve information in response to a patient question would depend on the kind of question raised by the patient. The author's strategy in his head-and-neck cancer groups was to identify a message with an information question and to parse the question into MeSH terms. His primary reference source was PubMed. The query to PubMed was then a conjunction of MeSH terms flagged to be ‘major’ in the indexing of the article. If this query retrieved too few citations, then the query was broadened.18 In constructing a reply based on the retrieved information, one must consider the amount of: material brought from the web page (at one extreme the response only contains the web address of the relevant page) and; emotional support and personal involvement to manifest (for instance, one could include a relevant, personal anecdote). Students repeatedly suggested that effectively replying requires contributing information in a personalized way. The patients want to read about personal experiences from other group members. Students typically reported receiving a positive reply to the information that the student had retrieved and presented. However, if the student replied to an old message, then the value of the student reply was less. In the author's head-and-neck cancer patient groups, the discussions helped reveal gaps in the literature. The medical details of two discussions are briefly described next: This author and other patients complained of fatigue. One journal article said that this fatigue is likely to be secondary to obstructive sleep apnoea, that was in turn secondary to head-and-neck cancer treatment. This relationship between the treatment and apnoea is poorly understood by doctors and poorly documented in the literature.19 A high proportion of patients were misdiagnosed with a branchial cleft cyst before the correct diagnosis of head-and-neck cancer was obtained at a later date. The literature shows that, for such patients, what might be a branchial cleft cyst should always be first considered to be a malignancy,20 but the literature does not address the problem of reducing the rate of misdiagnosis. These and other such experiences suggest that gaps in the literature tend to relate to adverse events and to involve multiple experts (such as a primary-care physician and a specialist). The students and professor studied a patient question, did an information search and presented the results to the patient group. Based on the author's experiences and his observations of his students’ experiences, the following further guidelines for this process are suggested. The message chosen should have: been posted recently rather than having been sent to the group many days earlier; expressed a clear desire for further information, and; should not have already received a reply which satisfies the information need. The response should: include a synopsis of the retrieved information, and not only a pointer, and; be personalized to both relate to the question and to indicate the empathy of the person providing the response. These characteristics of the response would be difficult to automate. However, some steps might be supported by automatic tools. Discussion system software packages are freely available with source code that can be augmented. One such tool was acquired and tailored. A hyperlink was placed next to each message that would open the medlinePlus website and retrieve information based on the heading of the message (see Fig. 1). The link says: The discussion system illustrated here was built in free software to which was added a link called ‘MoreInfo?’. When the user selects ‘MoreInfo?’, then he or she is taken to a new window that contains a list of medlinePlus information that matches the query generated from the subject heading of the particular message <a href=‘http://search.nlm.nih.gov/medlineplus/query?FUNCTION=search&PARAMETER=&adoRS.Fields(headerStr)>More Info?</a> where ‘More Info?’ is the name of the link and ‘headerStr’ contains the message heading. However, the message subject line is often too vague to support the generation of an adequately precise query for extra information. For example, one message subject said: ‘My dad was told he has no chance’. By reading the message body, one could determine that the dad has been newly diagnosed with ‘non-small cell lung cancer’ and that the daughter is looking for chemotherapy information. The author's simulations of algorithms to parse a patient message show that the query should be automatically augmented with the general topic of the group, which in the preceding example was ‘lung cancer’. Because a group exists to explore a particular disease, messages may leave that disease unmentioned because members will infer that. The information retrieval software could include that inference. In online patient groups, a message that provides information from the web to help another patient is appreciated. If a student's goal is to reply to a patient's request for general medical knowledge, then the student might prefer a web portal like medlinePlus or Google over PubMed. The patients prefer to receive information digested and personalized than to simply receive a pointer to the information. The process of reading patient messages and providing relevant information from the literature is labour intensive. Some of the work might be semi-automated by using natural language processing techniques in a tailored information retrieval link. Patient experiences might also contribute to insights about gaps in the literature. However, if the goal is to discover gaps in the literature, then a medically knowledgeable person should be involved and PubMed should be used. Implications for Policy Librarians might consider visiting online patient–patient discussion groups and helping patients with queries. Health-care quality assurance or research organizations should consider looking at patient–patient online discussion groups to find indications of problems that patients experience which are inadequately covered in the literature. Implications for Practice Information systems students who try to answer questions in online patient discussion groups find medlinePlus more useful than PubMed or Google, while a medical expert in the same situation will find PubMed most useful. A response to a patient message is most useful when the response is immediate and empathetic. A link could be automatically generated beside a patient message that retrieves information from a web portal based on keywords in the patient message." @default.
• W2006775401 created "2016-06-24" @default.
• W2006775401 creator A5050142728 @default.
• W2006775401 date "2006-03-01" @default.
• W2006775401 modified "2023-09-23" @default.
• W2006775401 title "Information retrieval for online patient groups" @default.
• W2006775401 cites W1724546338 @default.
• W2006775401 cites W1981160791 @default.
• W2006775401 cites W1983895294 @default.
• W2006775401 cites W2022072995 @default.
• W2006775401 cites W2025976977 @default.
• W2006775401 cites W2030015411 @default.
• W2006775401 cites W2126860484 @default.
• W2006775401 doi "https://doi.org/10.1111/j.1471-1842.2006.00626.x" @default.
• W2006775401 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/16466500" @default.
• W2006775401 hasPublicationYear "2006" @default.
• W2006775401 type Work @default.
• W2006775401 sameAs 2006775401 @default.
• W2006775401 citedByCount "2" @default.
• W2006775401 crossrefType "journal-article" @default.
• W2006775401 hasAuthorship W2006775401A5050142728 @default.
• W2006775401 hasConcept C136764020 @default.
• W2006775401 hasConcept C17744445 @default.
• W2006775401 hasConcept C199539241 @default.
• W2006775401 hasConcept C23123220 @default.
• W2006775401 hasConcept C2522767166 @default.
• W2006775401 hasConcept C2779473830 @default.
• W2006775401 hasConcept C41008148 @default.
• W2006775401 hasConceptScore W2006775401C136764020 @default.
• W2006775401 hasConceptScore W2006775401C17744445 @default.
• W2006775401 hasConceptScore W2006775401C199539241 @default.
• W2006775401 hasConceptScore W2006775401C23123220 @default.
• W2006775401 hasConceptScore W2006775401C2522767166 @default.
• W2006775401 hasConceptScore W2006775401C2779473830 @default.
• W2006775401 hasConceptScore W2006775401C41008148 @default.
• W2006775401 hasIssue "1" @default.
• W2006775401 hasLocation W20067754011 @default.
• W2006775401 hasLocation W20067754012 @default.
• W2006775401 hasOpenAccess W2006775401 @default.
• W2006775401 hasPrimaryLocation W20067754011 @default.
• W2006775401 hasRelatedWork W2115485936 @default.
• W2006775401 hasRelatedWork W2119135658 @default.
• W2006775401 hasRelatedWork W2119214692 @default.
• W2006775401 hasRelatedWork W2146184373 @default.
• W2006775401 hasRelatedWork W2153015554 @default.
• W2006775401 hasRelatedWork W2357241418 @default.
• W2006775401 hasRelatedWork W2366644548 @default.
• W2006775401 hasRelatedWork W2376314740 @default.
• W2006775401 hasRelatedWork W2384888906 @default.
• W2006775401 hasRelatedWork W2748952813 @default.
• W2006775401 hasVolume "23" @default.
• W2006775401 isParatext "false" @default.
• W2006775401 isRetracted "false" @default.
• W2006775401 magId "2006775401" @default.
• W2006775401 workType "article" @default.