COVID-Twitter-BERT: A natural language processing model to analyse COVID-19 content on Twitter.

Introduction: This study presents COVID-Twitter-BERT (CT-BERT), a transformer-based model that is pre-trained on a large corpus of COVID-19 related Twitter messages. CT-BERT is specifically designed to be used on COVID-19 content, particularly from social media, and can be utilized for various natural language processing tasks such as classification, question-answering, and chatbots. This paper aims to evaluate the performance of CT-BERT on different classification datasets and compare it with BERT-LARGE, its base model. Methods: The study utilizes CT-BERT, which is pre-trained on a large corpus of COVID-19 related Twitter messages. The authors evaluated the performance of CT-BERT on five different classification datasets, including one in the target domain. The model's performance is compared to its base model, BERT-LARGE, to measure the marginal improvement. The authors also provide detailed information on the training process and the technical specifications of the model. Results: The results indicate that CT-BERT outperforms BERT-LARGE with a marginal improvement of 10-30% on all five classification datasets. The largest improvements are observed in the target domain. The authors provide detailed performance metrics and discuss the significance of these results. Discussion: The study demonstrates the potential of pre-trained transformer models, such as CT-BERT, for COVID-19 related natural language processing tasks. The results indicate that CT-BERT can improve the classification performance on COVID-19 related content, especially on social media. These findings have important implications for various applications, such as monitoring public sentiment and developing chatbots to provide COVID-19 related information. The study also highlights the importance of using domain-specific pre-trained models for specific natural language processing tasks. Overall, this work provides a valuable contribution to the development of COVID-19 related NLP models.

Understanding the vaccine stance of Italian tweets and addressing language changes through the COVID-19 pandemic: Development and validation of a machine learning model.

Social media is increasingly being used to express opinions and attitudes toward vaccines. The vaccine stance of social media posts can be classified in almost real-time using machine learning. We describe the use of a Transformer-based machine learning model for analyzing vaccine stance of Italian tweets, and demonstrate the need to address changes over time in vaccine-related language, through periodic model retraining. Vaccine-related tweets were collected through a platform developed for the European Joint Action on Vaccination. Two datasets were collected, the first between November 2019 and June 2020, the second from April to September 2021. The tweets were manually categorized by three independent annotators. After cleaning, the total dataset consisted of 1,736 tweets with 3 categories (promotional, neutral, and discouraging). The manually classified tweets were used to train and test various machine learning models. The model that classified the data most similarly to humans was XLM-Roberta-large, a multilingual version of the Transformer-based model RoBERTa. The model hyper-parameters were tuned and then the model ran five times. The fine-tuned model with the best F-score over the validation dataset was selected. Running the selected fine-tuned model on just the first test dataset resulted in an accuracy of 72.8% (F-score 0.713). Using this model on the second test dataset resulted in a 10% drop in accuracy to 62.1% (F-score 0.617), indicating that the model recognized a difference in language between the datasets. On the combined test datasets the accuracy was 70.1% (F-score 0.689). Retraining the model using data from the first and second datasets increased the accuracy over the second test dataset to 71.3% (F-score 0.713), a 9% improvement from when using just the first dataset for training. The accuracy over the first test dataset remained the same at 72.8% (F-score 0.721). The accuracy over the combined test datasets was then 72.4% (F-score 0.720), a 2% improvement. Through fine-tuning a machine-learning model on task-specific data, the accuracy achieved in categorizing tweets was close to that expected by a single human annotator. Regular training of machine-learning models with recent data is advisable to maximize accuracy.

Categorizing Vaccine Confidence With a Transformer-Based Machine Learning Model: Analysis of Nuances of Vaccine Sentiment in Twitter Discourse.

BACKGROUND: Social media has become an established platform for individuals to discuss and debate various subjects, including vaccination. With growing conversations on the web and less than desired maternal vaccination uptake rates, these conversations could provide useful insights to inform future interventions. However, owing to the volume of web-based posts, manual annotation and analysis are difficult and time consuming. Automated processes for this type of analysis, such as natural language processing, have faced challenges in extracting complex stances such as attitudes toward vaccination from large amounts of text. OBJECTIVE: The aim of this study is to build upon recent advances in transposer-based machine learning methods and test whether transformer-based machine learning could be used as a tool to assess the stance expressed in social media posts toward vaccination during pregnancy. METHODS: A total of 16,604 tweets posted between November 1, 2018, and April 30, 2019, were selected using keyword searches related to maternal vaccination. After excluding irrelevant tweets, the remaining tweets were coded by 3 individual researchers into the categories Promotional, Discouraging, Ambiguous, and Neutral or No Stance. After creating a final data set of 2722 unique tweets, multiple machine learning techniques were trained on a part of this data set and then tested and compared with the human annotators. RESULTS: We found the accuracy of the machine learning techniques to be 81.8% (F score=0.78) compared with the agreed score among the 3 annotators. For comparison, the accuracies of the individual annotators compared with the final score were 83.3%, 77.9%, and 77.5%. CONCLUSIONS: This study demonstrates that we are able to achieve close to the same accuracy in categorizing tweets using our machine learning models as could be expected from a single human coder. The potential to use this automated process, which is reliable and accurate, could free valuable time and resources for conducting this analysis, in addition to informing potentially effective and necessary interventions.

"Vaccines for pregnant women ?! Absurd" - Mapping maternal vaccination discourse and stance on social media over six months.

OBJECTIVE: To understand the predominant topics of discussion, stance and associated language used on social media platforms relating to maternal vaccines in 15 countries over a six-month period. BACKGROUND: In 2019, the World Health Organisation prioritised vaccine hesitancy as a top ten global health threat and recognized the role of viral misinformation on social media as propagating vaccine hesitancy. Maternal vaccination offers the potential to improve maternal and child health, and to reduce the risk of severe morbidity and mortality in pregnancy. Understanding the topics of discussion, stance and language used around maternal vaccines on social media can inform public health bodies on how to combat vaccine misinformation and vaccine hesitancy. METHODS: Social media data was extracted (Twitter, forums, blogs and comments) for six months from 15 countries (Australia, Brazil, Canada, France, Germany, India, Italy, Korea, Mexico, Panama, South Africa, Spain, United Kingdom and United States). We used stance, discourse and topic analysis to provide insight into the most frequent and weighted keywords, hashtags and themes of conversation within and across countries. RESULTS: We exported a total of 19,192 social media posts in 16 languages obtained between 1st November 2018 and 30th April 2019. After screening all posts, 16,000 were included in analyses, while excluding retweets, 2,722 were annotated for sentiment. Main topics of discussion were the safety of the maternal influenza and pertussis vaccines. Discouraging posts were most common in Italy (44.9%), and the USA (30.8%). CONCLUSION: The content and stance of maternal vaccination posts from November 2018 to April 2019 differed across countries, however specific topics of discussion were not limited to geographical location. These discussions included the promotion of vaccination, involvement of pregnant women in vaccine research, and the trust and transparency of institutions. Future research should examine the relationship between stance (promotional, neutral, ambiguous, discouraging) online and maternal vaccination uptake in the respective regions.

Patient Use and Experience With Online Access to Electronic Health Records in Norway: Results From an Online Survey.

BACKGROUND: The electronic health record (EHR) has been fully established in all Norwegian hospitals. Patient-accessible electronic health records (PAEHRs) are available to citizens aged 16 years and older through the national health portal Helsenorge. OBJECTIVE: This study aimed at understanding how patients use PAEHRs. Three research questions were addressed in order to explore (1) characteristics of users, (2) patients' use of the service, and (3) patient experience with the service. METHODS: We conducted an online survey of users who had accessed their EHR online at least once through the national health portal. Patients from two of the four health regions in Norway were invited to participate. Quantitative data were supplemented by qualitative information. RESULTS: A total of 1037 respondents participated in the survey, most of whom used the PAEHR regularly (305/1037, 29.4%) or when necessary (303/1037, 29.2%). Service utilization was associated with self-reported health, age, gender, education, and health care professional background. Patients found the service useful to look up health information (687/778, 88.3%), keep track of their treatment (684/778, 87.9%), prepare for a hospital appointment (498/778, 64.0%), and share documents with their general practitioner (292/778, 37.5%) or family (194/778, 24.9%). Most users found it easy to access their EHR online (965/1037, 93.1%) and did not encounter technical challenges. The vast majority of respondents (643/755, 85.2%) understood the content, despite over half of them acknowledging some difficulties with medical terms or phrases. The overall satisfaction with the service was very high (700/755, 92.7%). Clinical advantages to the patients included enhanced knowledge of their health condition (565/691, 81.8%), easier control over their health status (685/740, 92.6%), better self-care (571/653, 87.4%), greater empowerment (493/674, 73.1%), easier communication with health care providers (493/618, 79.8%), and increased security (655/730, 89.7%). Patients with complex, long-term or chronic conditions seemed to benefit the most. PAEHRs were described as useful, informative, effective, helpful, easy, practical, and safe. CONCLUSIONS: PAEHRs in Norway are becoming a mature service and are perceived as useful by patients. Future studies should include experimental designs focused on specific populations or chronic conditions that are more likely to achieve clinically meaningful benefits. Continuous evaluation programs should be conducted to assess implementation and changes of wide-scale routine services over time.

Health Professionals' Experience with Patients Accessing Their Electronic Health Records: Results from an Online Survey.

The aim of this study was to investigate hospital professionals' experience and attitude with patients accessing their own electronic health records. The study was conducted one year after service establishment. Data was collected through an online survey. In total, 457 replies were received. The results revealed a quarter of the administrative staff received feedback from patients or relatives regarding mistakes or missing information in their EHR. In addition, 67.5% of health professionals expected more patients to have basic knowledge of their health status in the future, and 21.4% found patients already gained better knowledge about diagnosis, treatment, or follow-up. The results also revealed some challenges with the service, especially for health professionals working in psychiatry, with some scepticism on whether the service is suitable for the sickest and most vulnerable patients.

Differences in Emotional and Pain-Related Language in Tweets About Dentists and Medical Doctors: Text Analysis of Twitter Content.

BACKGROUND: Social media provides people with easy ways to communicate their attitudes and feelings to a wide audience. Many people, unfortunately, have negative associations and feelings about dental treatment due to former painful experiences. Previous research indicates that there might be a pervasive and negative occupational stereotype related to dentists and that this stereotype is expressed in many different venues, including movies and literature. OBJECTIVE: This study investigates the language used in relation to dentists and medical doctors on the social media platform Twitter. The purpose is to compare the professions in terms of the use of emotional and pain-related words, which might underlie and reflect the pervasive negative stereotype identified in relation to dentists. We hypothesized that (A) tweets about dentists will have more negative emotion-related words than those about medical doctors and (B) pain-related words occur more frequently in tweets about dentists than in those about medical doctors. METHODS: Twitter content ("tweets") about dentists and medical doctors was collected using the Twitter application program interface 140Dev over a 4-week period in 2015, scanning the search terms "dentist" and "doctor". Word content of the selected tweets was analyzed using Linguistic Inquiry and Word Count software. The research hypotheses were investigated using nonparametric Wilcoxon-Mann-Whitney tests. RESULTS: Over 2.3 million tweets were collected in total, of which about one-third contained the word "dentist" and about two-thirds contained the word "doctor." Hypothesis A was supported since a higher proportion of negative words was used in tweets about dentists than in those about medical doctors (z=-10.47; P<.001). Similarly, tests showed a difference in the proportions of anger words (z=-12.54; P<.001), anxiety words (z=-6.96; P<.001), and sadness words (z=-9.58; P<.001), with higher proportions of these words in tweets about dentists than in those about doctors. Also, Hypothesis B was supported since a higher proportion of pain-related words was used in tweets about dentists than in those about doctors (z=-8.02; P<.001). CONCLUSIONS: The results from this study suggest that stereotypes regarding dentists and dental treatment are spread through social media such as Twitter and that social media also might represent an avenue for improving messaging and disseminating more positive attitudes toward dentists and dental treatment.

Controversial Ebola vaccine trials in Ghana: a thematic analysis of critiques and rebuttals in digital news.

BACKGROUND: Communication is of paramount importance in responding to health crises. We studied the media messages put forth by different stakeholders in two Ebola vaccine trials that became controversial in Ghana. These interactions between health authorities, political actors, and public citizens can offer key lessons for future research. Through an analysis of online media, we analyse stakeholder concerns and incentives, and the phases of the dispute, to understand how the dispute evolved to the point of the trials being suspended, and analyse what steps might have been taken to avert this outcome. METHODS: A web-based system was developed to download and analyse news reports relevant to Ebola vaccine trials. This included monitoring major online newspapers in each country with planned clinical trials, including Ghana. All news articles were downloaded, selecting out those containing variants of the words "Ebola," and "vaccine," which were analysed thematically by a team of three coders. Two types of themes were defined: critiques of the trials and rebuttals in favour of the trials. After reconciling differences between coders' results, the data were visualised and reviewed to describe and interpret the debate. RESULTS: A total of 27,460 articles, published between 1 May and 30 July 2015, were collected from nine different newspapers in Ghana, of which 139 articles contained the keywords and met the inclusion criteria. The final codebook included 27 themes, comprising 16 critiques and 11 rebuttals. After coding and reconciliation, the main critiques (and their associated rebuttals) were selected for in-depth analysis, including statements about the trials being secret (mentioned in 21% of articles), claims that the vaccine trials would cause an Ebola outbreak in Ghana (33%), and the alleged impropriety of the incentives offered to participants (35%). DISCUSSION: Perceptions that the trials were "secret" arose from a combination of premature news reporting and the fact that the trials were prohibited from conducting any publicity before being approved at the time that the story came out, which created an impression of secrecy. Fears about Ebola being spread in Ghana appeared in two forms, the first alleging that scientists would intentionally infect Ghanaians with Ebola in order to test the vaccine, and the second suggesting that the vaccine might give trial participants Ebola as a side-effect - over the course of the debate, the latter became the more prominent of the two variants. The incentives were sometimes criticised for being coercively large, but were much more often criticised for being too small, which may have been related to a misperception that the incentives were meant as compensation for the trials' risks, which were themselves exaggerated. CONCLUSION: The rumours captured through this research indicate the variety of strong emotions drawn out by the trials, highlighting the importance of understanding the emotional and social context of such research. The uncertainty, fear, and distrust associated with the trials draw from the contemporary context of the Ebola outbreak, as well as longstanding historical issues in Ghana. By analysing the debate from its inception, we can see how the controversy unfolded, and identify points of concern that can inform health communication, suggesting that this tool may be valuable in future epidemics and crises.

A Proposed Framework to Enrich Norwegian EHR System with Health-trusted Information for Patients and Professionals.

In the era of social media, semantic web and big data, a huge amount of health-related information, knowledge and resources exist on the Web. Patients and healthcare professionals should spend enormous effort and time in order to find health-trusted information, while the appropriate technologies to interlink and retrieve this type of information already exist. In this paper we propose a framework to enrich DIPS, the most deployed Norwegian EHR System, with health-trusted information for patients and state-of-the-art resources for healthcare professionals. The framework based upon the new architecture of DIPS, namely DIPS Arena, and upon the interlinking with the semantic web, social media and open linked data cloud information, knowledge and resources through well-established medical thesauri like SNOMED CT and MeSH.

Media preferences in scenarios involving relationship closeness and information valence: evidence of strategic self-presentation and sex differences.

The study investigated strategic self-presentation (relationship closeness, information valence, and sex) on hypothetical choice of media used. 145 participants (73 women, 72 men; M age = 22.3 yr.) were randomly assigned to experimental conditions where they indicated their preference for communicating with either a friend or a stranger using Short Messaging Service (SMS), e-mail, or telephone, compared to face-to-face. Information valence was manipulated as a within-subjects variable by scenarios where information was self-referential and either negative or positive. Preference to mediated channels in the two scenarios was measured as an average of self-reported scores on a scale from 1-5 where 3 indicated face-to-face communication. Relationship closeness and scenario affected media preferences. Participants had higher preference scores for mediated channels when communicating with strangers than with friends and when sharing self-referential and negative information. Only women's preferences appeared to be affected by the manipulation of relationship closeness.

Using noninferiority tests to evaluate telemedicine and e-health services: systematic review.

BACKGROUND: An increasing number of studies within the field of telemedicine and e-health are designed as noninferiority studies, aiming to show that the telemedicine/e-health solution is not inferior to the traditional way of treating patients. OBJECTIVE: The objective is to review and sum up the status of noninferiority studies within this field, describing advantages and pitfalls of this approach. METHODS: PubMed was searched according to defined criteria, and 16 relevant articles were identified from the period 2008-June 2011. RESULTS: Most of the studies were related to the fields of psychiatry and emergency medicine, and most were published in journals relating to these fields or in general scientific or general medicine journals. All the studies claimed to be noninferiority studies, but 7 out of 16 tested for statistical differences as a proxy of noninferiority. CONCLUSIONS: The methodological quality of the studies varied. We discuss optimal procedures for future noninferiority studies within the field of telemedicine and e-health and situations in which this approach is most appropriate.

Health information accessed on the internet: the development in 5 European countries.

The aim of this study was to summarize and analyse findings from four prior studies on the use of the Internet as a source of health information in five European countries (Norway, Denmark, Germany, Greece, and Portugal). A cross-study comparison of data was performed. All the studies included fit with a trend of a sharp and continuous growth in the use of the Internet for health information access in the major part of the last decade. Importantly, the Internet has become an important mass media source of health information in northern Europe. While the use of the Internet for health information is somewhat less common in the south European countries, its use is also clearly increasing there. We discuss the advantages of cross-study comparisons of data and methodological challenges. As the use of the Internet for health information is likely to peak in some countries in the near future, new population surveys on health information access should focus more on the details of information that is accessed and which sites that are most used and trusted.

Physician response time when communicating with patients over the Internet.

BACKGROUND: Patients want to use electronic communication to access health services more easily. Health authorities in several countries see this as a way to improve health care. Physicians appear to have conflicting opinions regarding the suitability of electronic communication in clinical settings. OBJECTIVES: The aim of our study was to measure how long it actually takes physicians to answer questions from patients through an electronic communication channel, and whether some of the questions are especially time consuming. METHODS: We monitored electronic patient-physician communication. A total of 1113 messages from 14 participating physicians from 7 medical offices were analyzed. The length of questions and answers, and the time physicians spent answering the questions were recorded and analyzed. RESULTS: Physicians spent an average of 2.3 minutes (median 2 minutes) answering questions from patients. The patients' questions had an average length of 507.1 characters (95% CI 487.4-526.9, SD 336.2), while physicians' answers averaged 119.9 characters (95% CI 189.8-210.0, SD 172.6). The results show that the influence of patient question length on time spent responding was negligible. For the shortest 25% of the questions the answer time was 2.1 minutes (95% CI 1.9-2.3), while it was 2.4 minutes (95% CI 2.2-2.7) for the longest 25%. Even extremely long questions had a minimal impact on the time spent answering them. A threefold increase in question length from patients resulted in only an 18% increase in physician response time. CONCLUSIONS: The study shows the potential clinical usefulness of electronic communication between patients and health care services by demonstrating the potential for saving time.

Online communication between doctors and patients in Europe: status and perspectives.

BACKGROUND: Use of the Internet for health purposes is steadily increasing in Europe, while the eHealth market is still a niche. Online communication between doctor and patient is one aspect of eHealth with potentially great impact on the use of health systems, patient-doctor roles and relations and individuals' health. Monitoring and understanding practices, trends, and expectations in this area is important, as it may bring invaluable knowledge to all stakeholders, in the Health 2.0 era. OBJECTIVE: Our two main goals were: (1) to investigate use of the Internet and changes in expectations about future use for particular aspects of communication with a known doctor (obtaining a prescription, scheduling an appointment, or asking a particular health question), and (2) to investigate how important the provision of email and Web services to communicate with the physician is when choosing a new doctor for a first time face-to-face appointment. The data come from the second survey of the eHealth Trends study, which addressed trends and perspectives of health-related Internet use in Europe. This study builds on previous work that established levels of generic use of the Internet for self-help activities, ordering medicine or other health products, interacting with a Web doctor/unknown health professional, and communicating with a family doctor or other known health professional. METHODS: A representative sample of citizens from seven European countries was surveyed (n = 7022) in April and May of 2007 through computer-assisted telephone interviews (CATI). Respondents were questioned about their use of the Internet to obtain a prescription, schedule an appointment, or ask a health professional about a particular health question. They were also asked what their expectations were regarding future use of the Internet for health-related matters. In a more pragmatic approach to the subject, they were asked about the perceived importance when choosing a new doctor of the possibility of using email and the Web to communicate with that physician. Logistic regression analysis was used to draw the profiles of users of related eHealth services in Europe among the population in general and in the subgroup of those who use the Internet for health-related matters. Changes from 2005 to 2007 were computed using data from the first eHealth Trends survey (October and November 2005, n = 7934). RESULTS: In 2007, an estimated 1.8% (95% confidence interval [CI], 1.5 - 2.1) of the population in these countries had used the Internet to request or renew a prescription; 3.2% (95% CI 2.8 - 3.6) had used the Internet to schedule an appointment; and 2.5% (95% CI 2.2 - 2.9) had used the Internet to ask a particular health question. This represents estimated increases of 0.9% (95% CI 0.5 - 1.3), 1.7% (95% CI 1.2 - 2.2), and 1.4% (95% CI 0.9 - 1.8). An estimated 18.0% (95% CI 17.1 - 18.9) of the populations of these countries expected that in the near future they would have consultations with health professionals online, and 25.4% (95% CI 24.4 - 26.3) expected that in the near future they would be able to schedule an appointment online. Among those using the Internet for health-related purposes, on average more than 4 in 10 people considered the provision of these eHealth services to be important when choosing a new doctor. CONCLUSIONS: Use of the Internet to communicate with a known health professional is still rare in Europe. Legal context, health policy issues, and technical conditions prevailing in different countries might be playing a major role in the situation. Interest in associated eHealth services is high among citizens and likely to increase.

Use of the internet for health purposes: trends in Norway 2000-2010.

AIM: The aim of this work is to study the 10 year trends in the use and projected use of internet for health purposes in Norway. This includes more detailed analyses of changes in latter years, current uses, valuation as source of health information, reported effects and projected developments. METHOD: Surveys on the use of the internet for health purposes have been conducted in Norway in the years 2000, 2001, 2003, 2005 and 2007. Representative samples were drawn from the population, and interviews were conducted by poll agencies using telephone (CATI). RESULTS: Internet use for health purposes in the Norwegian population has increased dramatically over the last 7 years from 19% in 2000 to 67% in 2007. If this trend continues, we estimate that 84% of the Norwegian population will be using internet for health purposes by the year 2010. Reading about health or illness is the most common activity, and the rated importance of the internet as a source of health information is increasing. The internet is also used more for ordering medicines and other health-related products. Forty-four per cent of those having used the internet for health purposes reported having searched for lifestyle-related information, and 40% reported having felt inspired to change health behaviour. CONCLUSIONS: The internet is an important source of health information, which is likely to increase in importance in the years to come. This study further underlines the potential of internet use for health promoting purposes, as well as the potential for exacerbating social disparities in health.

eHealth trends in Europe 2005-2007: a population-based survey.

BACKGROUND: In the last decade, the number of Internet users worldwide has dramatically increased. People are using the Internet for various health-related purposes. It is important to monitor such use as it may have an impact on the individual's health and behavior, patient-practitioner roles, and on general health care provision. OBJECTIVES: This study investigates trends and patterns of European health-related Internet use over a period of 18 months. The main study objective was to estimate the change in the proportion of the population using the Internet for health purposes, and the importance of the Internet as a source of health information compared to more traditional sources. METHODS: The survey data were collected through computer-assisted telephone interviews. A representative sample (N = 14,956) from seven European countries has been used: Denmark, Germany, Greece, Latvia, Norway, Poland, and Portugal. The European eHealth Consumer Trends Survey was first conducted in October-November 2005 and repeated in April-May 2007. In addition to providing background information, respondents were asked to rate the importance of various sources of health information. They were also queried as to the frequency of different online activities related to health and illness and the effects of such use on their disposition. RESULTS: The percentage of the population that has used the Internet for health purposes increased from an estimated 42.3% (95% CI [Confidence Interval] 41.3 - 43.3) in 2005 to an estimated 52.2% (95% CI 51.3 - 53.2) in 2007. Significant growth in the use of the Internet for health purposes was found in all the seven countries. Young women are the most active Internet health users. The importance of the Internet as a source of health information has increased. In 2007, the Internet was perceived as an important source of health information by an estimated 46.8% (95% CI 45.7 - 47.9) of the population, a significant increase of 6.5 % (95% CI 4.9 - 8.1) from 2005. The importance of all the traditional health information channels has either decreased or remained the same. An estimated 22.7% (95% CI 21.7 - 23.6) are using it for more interactive services than just reading health information. CONCLUSION: The Internet is increasingly being used as a source of health information by the European population, and its perceived importance is rising. Use of the Internet for health purposes is growing in all age groups and for both men and women, with especially strong growth among young women. We see that experienced Internet health users are also using the Internet as an active communication channel, both for reaching health professionals and for communicating with peers.

Communicating textual health information to the mobile phones of visually-impaired users.

We investigated how short messages communicating health information would best be distributed to people with vision difficulties using mobile phones. Twelve visually-impaired persons who were unable to read short message service (SMS) messages directly compared three methods of presenting text messages as speech: (1) ordinary SMS messages were sent to the users and converted into speech by the mobile phone; (2) multimedia messages were sent to the users with prerecorded speech-synthesized information; and (3) mobile phone calls were placed to the users and prerecorded speech-synthesized messages were streamed to them. The latter two approaches used server-generated sound files. Over a three-month trial period, we sent a total of 88 SMS messages to the subjects, 111 multimedia messaging service (MMS) messages and 104 telephone calls. All of the SMS messages, 88% of the MMS messages and 69% of the telephone calls were received. In subsequent interviews, we asked the users which presentation method they preferred. SMS scored significantly better than both MMS (P = 0.033) and telephones (P = 0.006). All three methods had serious drawbacks. However, the study suggests that it might be possible to develop suitable technology for communicating with people with vision difficulties by mobile phone.

Patients who use e-mediated communication with their doctor: new constructions of trust in the patient-doctor relationship.

The introduction of information and communication technology (ICT) into the patient-doctor relationship represents a significant change in modern health care. Communication via computers-e-mediated communication-is affecting the context of patient-doctor interaction, touching core elements of the relationship. Based on data from a qualitative study conducted among Norwegian patients who had used ICT to communicate with their doctors, the authors argue that patients' use of ICT and the element of trust in the patient-doctor relationship influence each other. Furthermore, they contend that patients' constructions of trust in this relationship can be understood in light of basic mechanisms in modern society. The study sheds light on some potential concerns and benefits as communication technology increasingly is integrated into the patient-doctor relationship.

Electronic patient-provider communication: will it offset office visits and telephone consultations in primary care?

BACKGROUND AND AIM: Electronic patient-provider communication promises to improve efficiency and effectiveness of clinical care. This study aims to explore whether a secure web-based messaging system is an effective way of providing patient care in general practices. METHOD: We conducted a randomised controlled trail and recruited 200 patients from the waiting area in one primary clinic in Norway. Participants were randomised to either the intervention group, which received access to a secure messaging system, or the control group receiving standard care without such access. Primary outcome measures were number of online consultations, telephone consultations and office visits in the two groups. Data were derived from patient records and collected 1 year prior to (baseline), and 1 year after the intervention. RESULTS: Forty-six percent of the patients who were given access to the messaging system (n=99) used the online communication system on at least one occasion (ranging from 1 to 17 messages per patient per year). A total of 147 electronic messages were sent to six general practitioners during a 1-year trial period. Eleven percent of the messages were to schedule an appointment. In 10% of the messages, the GP was unable to respond adequately and recommended an office visit. The reduction in office visits over time was greater for the intervention group than for the control group (P=0.034). There was however no significant difference in the number of telephone consultations between the groups during the study (P=0.258). CONCLUSION: The use of a secure electronic messaging system reduced the number of office visits at the general practice, but not phone consultations.

"Is Cybermedicine Killing You?"--The story of a Cochrane disaster.

This editorial briefly reviews the series of unfortunate events that led to the publication, dissemination, and eventual retraction of a flawed Cochrane systematic review on interactive health communication applications (IHCAs), which was widely reported in the media with headlines such as "Internet Makes Us Sick," "Knowledge May Be Hazardous to Web Consumers' Health," "Too Much Advice Can Be Bad for Your Health," "Click to Get Sick?," and even "Is Cybermedicine Killing You?". While the media attention helped to speed up the identification of errors, leading to a retraction of the review after only 13 days, a paper published in this issue of JMIR by Rada shows that the retraction, in contrast to the original review, remained largely unnoticed by the public. We discuss the three flaws of the review, which include (1) data extraction and coding errors, (2) the pooling of heterogeneous studies, and (3) a problematic and ambiguous scope and, possibly, some overlooked studies. We then discuss "retraction ethics" for researchers, editors/publishers, and journalists. Researchers and editors should, in the case of retractions, match the aggressiveness of the original dissemination campaign if errors are detected. It is argued that researchers and their organizations may have an ethical obligation to track down journalists who reported stories on the basis of a flawed study and to specifically ask them to publish an article indicating the error. Journalists should respond to errors or retractions with reports that have the same prominence as the original story. Finally, we look at some of the lessons for the Cochrane Collaboration, which include (1) improving the peer-review system by routinely sending out pre-prints to authors of the original studies, (2) avoiding downplay of the magnitude of errors if they occur, (3) addressing the usability issues of RevMan, and (4) making critical articles such as retraction notices open access.