Semi-supervised learning of the electronic health record for phenotype stratification

Patient interactions with health care providers result in entries to electronic health records (EHRs). EHRs were built for clinical and billing purposes but contain many data points about an individual. Mining these records provides opportunities to extract electronic phenotypes, which can be paired with genetic data to identify genes underlying common human diseases. This task remains challenging: high quality phenotyping is costly and requires physician review; many fields in the records are sparsely filled; and our definitions of diseases are continuing to improve over time. Here we develop and evaluate a semi-supervised learning method for EHR phenotype extraction using denoising autoencoders for phenotype stratification. By combining denoising autoencoders with random forests we find classification improvements across multiple simulation models and improved survival prediction in ALS clinical trial data. This is particularly evident in cases where only a small number of patients have high quality phenotypes, a common scenario in EHR-based research. Denoising autoencoders perform dimensionality reduction enabling visualization and clustering for the discovery of new subtypes of disease. This method represents a promising approach to clarify disease subtypes and improve genotype-phenotype association studies that leverage EHRs.     

‘No “Sane” Person Would Have Any Idea’: Patients’ Involvement in Late Nineteenth-century British Asylum?Psychiatry

In his 1895 textbook, Mental Physiology, Bethlem Royal Hospital physician Theo Hyslop acknowledged the assistance of three fellow hospital residents. One was a junior colleague. The other two were both patients: Walter Abraham Haigh and Henry Francis Harding. Haigh was also thanked in former superintendent George Savage’s book Insanity and Allied Neuroses (1884). In neither instance were the patients identified as such. This begs the question: what role did Haigh and Harding play in asylum theory and practice? And how did these two men interpret their experiences, both within and outside the asylum? By focusing on Haigh and Harding’s unusual status, this paper argues that the notion of nineteenth-century ‘asylum patient’ needs to be investigated by paying close attention to specific national and institutional circumstances. Exploring Haigh and Harding’s active engagement with their physicians provides insight into this lesser-known aspect of psychiatry’s history. Their experience suggests that, in some instances, representations of madness at that period were the product of a two-way process of negotiation between alienist and patient. Patients, in other words, were not always mere victims of ‘psychiatric power’; they participated in the construction and circulation of medical notions by serving as active intermediaries between medical and lay perceptions of madness.     

The utilization of Arabic online drug information among adults in Saudi Arabia

In Saudi Arabia, the utilization of the world wide web has become increasingly popular. However, the exact figure of such use is unknown. This study aimed to determine the percentage of, and experience with, online Arabic drug information by Arabic-speaking adults in Saudi Arabia. A web based questionnaire was used. The questionnaire language was Arabic. Public were invited to participate in the survey through e-mails, Twitter, WhatsApp and Facebook in March 2012. The survey included 17 items examining the types of accessed Arabic drug information, the respondent’s demographics, their ability to easily find and understand Arabic drug-related information, and their trustfulness and dependency on such information websites. Of the 422 Arabic speaking adults who answered the questionnaire, 88% stated that they used Arabic websites to answer drug-related questions. Of the respondents, 50% had a bachelor’s degree, 44% were young adults, over half were female (60%), and 72% of them have a chronic disease. The ease of retrieving online information was the most common reason (69%) for consulting such websites. Google as a search engine was the most frequently (86%) accessible website. Although respondents reported different drug-related topics in their online searching, the search for adverse effects was the most common (68%). Respondents claimed that they could easily find (65%) and understand (49%) the drug-related information. Although a good number of respondents qualified this type of information as good, double-checking of information on other websites was highly recommended. Trustfulness was one of the important parameters to measure and 205 respondents (55%) claimed that they only trusted half of the information cited. Moreover, around 48% of respondents considered that finding the same information on more than one website increased its trustfulness. Surprisingly, 54% of respondents did not depend on Arabic information websites when making decisions on drug use. There are a high proportion of Arabic speaking people in Saudi Arabia using and consulting Arabic drug information websites. This information is easily found and understood. However, the quality and trustfulness of such websites are not high enough to depend on them. A qualified Arabic drug information website is important to meet this need.     

The PaArticular Scales – A new outcome measure to quantify the impact of joint contractures on activities and participation in individuals in geriatric care: Development and Rasch analysis

BackgroundJoint contractures are frequent conditions in individuals in geriatric care settings and are associated with activity limitations and participation restrictions. As such, relevant intervention programs should address these aspects, and the effectiveness of such programs should be determined by assessing improvement in activities and participation. However, no patient-centred and psychometrically sound outcome measures for this purpose are available so far.ObjectivesThe objectives of this study were to develop and to validate a new outcome measure, the PaArticular Scales, to quantify activities and participation in older individuals with joint contractures. Specific aims were (A) to operationalize the content of an International Classification of Functioning, Disability and Health-based standard set towards meaningful questions and to combine them to a questionnaire and (B) to assess the psychometric properties of the developed questionnaire, in detail to evaluate test–retest reliability, objectivity, internal consistency reliability and criterion validity.DesignOperationalization was reached by an expert consensus conference and a subsequent expert Delphi survey. Psychometric properties were assessed in a cross-sectional study.SettingsNursing homes, geriatric rehabilitation facilities.Participants23 experts (nurses, physicians, physical and occupational therapists) participated in the consensus conference and the Delphi survey. A total of 191 individuals with joint contractures (as confirmed by physician, nurse or physical therapist) between 65 and 102 years, living in nursing homes or as patients in geriatric rehabilitation were enrolled in the cross-sectional study.MethodsRasch Partial Credit Modelling.ResultsThe consensus conference and Delphi survey resulted in a questionnaire with 86 items of the International Classification of Functioning, Disability and Health. Test–retest-reliability among those was acceptable (Cohen's weighted kappa: 0.779). The Rasch analysis revealed two independent interval-scaled scales with 24 items for the Activities scale and 11 items for the Participation scale with high internal consistency reliability. Cronbach's alpha was 0.96 for the Activities scale and 0.92 for the Participation scale. Criterion validity was −0.40 and −0.30 for the Activities scale and for the Participation scale, respectively.ConclusionsThe PaArticular Scales, a new patient-centred and psychometric sound outcome measures to comprehensively assess the impact of joint contractures in geriatric care, are available now. These developed scales will serve as primary outcomes in a scheduled evaluation of a complex intervention to improve participation and quality of life in nursing home residents with joint contractures.     

Personas in online health communities

Many researchers and practitioners use online health communities (OHCs) to influence health behavior and provide patients with social support. One of the biggest challenges in this approach, however, is the rate of attrition. OHCs face similar problems as other social media platforms where user migration happens unless tailored content and appropriate socialization is supported. To provide tailored support for each OHC user, we developed personas in OHCs illustrating users’ needs and requirements in OHC use. To develop OHC personas, we first interviewed 16 OHC users and administrators to qualitatively understand varying user needs in OHC. Based on their responses, we developed an online survey to systematically investigate OHC personas. We received 184 survey responses from OHC users, which informed their values and their OHC use patterns. We performed open coding analysis with the interview data and cluster analysis with the survey data and consolidated the analyses of the two datasets. Four personas emerged—Caretakers, Opportunists, Scientists, and Adventurers. The results inform users’ interaction behavior and attitude patterns with OHCs. We discuss implications for how these personas inform OHCs in delivering personalized informational and emotional support.