Stabilized sparse ordinal regression for medical risk stratification

The recent wide adoption of electronic medical records (EMRs) presents great opportunities and challenges for data mining. The EMR data are largely temporal, often noisy, irregular and high dimensional. This paper constructs a novel ordinal regression framework for predicting medical risk stratification from EMR. First, a conceptual view of EMR as a temporal image is constructed to extract a diverse set of features. Second, ordinal modeling is applied for predicting cumulative or progressive risk. The challenges are building a transparent predictive model that works with a large number of weakly predictive features, and at the same time, is stable against resampling variations. Our solution employs sparsity methods that are stabilized through domain-specific feature interaction networks. We introduces two indices that measure the model stability against data resampling. Feature networks are used to generate two multivariate Gaussian priors with sparse precision matrices (the Laplacian and Random Walk). We apply the framework on a large short-term suicide risk prediction problem and demonstrate that our methods outperform clinicians to a large margin, discover suicide risk factors that conform with mental health knowledge, and produce models with enhanced stability. © 2014 Springer-Verlag London.

Challenges of managing medications for older people at transition points of care

In clinical practice, pharmacists play a very important role in identifying and correcting medication discrepancies as older patients move across transition points of care. With increasing complexity of health care needs of older people, these discrepancies are likely to increase. The major concern with identifying and correcting medication discrepancies is that medication reconciliation is considered a retrospective problem - that is, dealing with medication discrepancies after they have occurred. It is argued here that a more proactive stance should be taken where doctors, nurses and pharmacists collectively work together to prevent medication discrepancies from happening in the first place. Improved involvement of patients and family members will help to facilitate better management of medications across transition points of care. Efficient use of information technology aids, such as electronic medication reconciliation tools, should also assist with organizational systems problems associated with the working culture, heavy workloads, and staff and skill mix of health professionals.

Learning vector representation of medical objects via EMR-driven nonnegative restricted Boltzmann machines (eNRBM)

Electronic medical record (EMR) offers promises for novel analytics. However, manual feature engineering from EMR is labor intensive because EMR is complex - it contains temporal, mixed-type and multimodal data packed in irregular episodes. We present a computational framework to harness EMR with minimal human supervision via restricted Boltzmann machine (RBM). The framework derives a new representation of medical objects by embedding them in a low-dimensional vector space. This new representation facilitates algebraic and statistical manipulations such as projection onto 2D plane (thereby offering intuitive visualization), object grouping (hence enabling automated phenotyping), and risk stratification. To enhance model interpretability, we introduced two constraints into model parameters: (a) nonnegative coefficients, and (b) structural smoothness. These result in a novel model called eNRBM (EMR-driven nonnegative RBM). We demonstrate the capability of the eNRBM on a cohort of 7578 mental health patients under suicide risk assessment. The derived representation not only shows clinically meaningful feature grouping but also facilitates short-term risk stratification. The F-scores, 0.21 for moderate-risk and 0.36 for high-risk, are significantly higher than those obtained by clinicians and competitive with the results obtained by support vector machines.

An evaluation of an action research nursing documentation project

AIMS AND OBJECTIVES: The aim of this study is to describe healthcare professionals' experiences and perceptions of an intervention implemented in an action research project conducted to improve nursing documentation practices in four municipalities in Norway.

BACKGROUND: Documentation of individualized patient care is a continuing concern in healthcare services and could impacts the quality and safety of healthcare. Use of electronic systems has made some aspects of documentation more comprehensive, but creation of an individualized care plan remains a pressing issue.

DESIGN: A qualitative descriptive design was used.

METHODS: An action research project was conducted between 2010 and 2012 to improve the content and quality of nursing documentation in community healthcare services in four municipalities. One year after the project was completed four focus group interviews were conducted with healthcare professionals, one for each involved municipality. Two unit managers were interviewed individually. Qualitative content analysis was used.

RESULTS: Three themes emerged: healthcare professionals perceived competing interest; they experienced that they had to manage complexity and changes; and they highlighted a clear and visible leader as important for success.

CONCLUSIONS: Quality improvement activities are essential. Healthcare professionals experience a complicated situation when electronic health record systems do not support workflow. Further research is recommended to focus on the functionality and user interface of EHR systems, and on the role of leadership when implementing changes in clinical practice. This article is protected by copyright. All rights reserved.

Nurses' role in medication safety

Aim To explore the nurses role in the process of medication management and identify the challenges associated with safe medication management in contemporary clinical practice.
Background Medication errors have been a long-standing factor affecting consumer safety. The nursing profession has been identified as essential to the promotion of patient safety.
Evaluation A review of literature on medication errors and the use of electronic prescribing in medication errors.
Key issues Medication management requires a multidisciplinary approach and interdisciplinary communication is essential to reduce medication errors. Information technologies can help to reduce some medication errors through eradication of transcription and dosing errors. Nurses must play a major role in the design of computerized medication systems to ensure a smooth transition to such as system.
Conclusion The nurses roles in medication management cannot be over-emphasized. This is particularly true when designing a computerized medication system.
Implication for nursing management The adoption of safety measures during decision making that parallel those of the aviation industry safety procedures can provide some strategies to prevent medication error. Innovations in information technology offer potential mechanisms to avert adverse events in medication management for nurses.

Implementing scanned medical record systems in Australia : a structured case study on envisioned changes to elective admissions process in a Victorian hospital

In recent years, influenced by the pervasive power of technology, standards and mandates, Australian hospitals have begun exploring digital forms of keeping this record. The main rationale is the ease of accessing different data sources at the same time by varied staff members. The initial step in this transition was implementation of scanned medical record systems, which converts the paper based records to digitised form, which required process flow redesign and changes to existing modes of work. For maximising the benefits of scanning implementation and to better prepare for the changes, Austin Hospital in the State of Victoria commissioned this research focused on elective admissions area. This structured case study redesigned existing processes that constituted the flow of external patient forms and recommended a set of best practices at the same time highlighting the significance of user participation in maximising the potential benefits anticipated. In the absence of published academic studies focused on Victorian hospitals, this study has become a conduit for other departments in the hospital as well as other hospitals in the incursion.

Medication adherence self-report instruments : implications for practice and research

Background: After an acute cardiac event, adhering to recommendations for pharmacologic therapy is important in achieving optimal health outcomes. Considering the impressive evidence base for cardiovascular pharmacotherapy, strategies for promoting adherence are less well developed. Furthermore, accessing reliable, valid, and cost-effective mechanisms of monitoring adherence in the research and clinical settings is challenging. Aim: The aim of this article was to review published self-report measures assessing and monitoring medication adherence in cardiovascular disease and provide recommendations for research into medication adherence. Methods: The electronic databases CINAHL, Medline, and Science Direct were searched using the key search terms medication adherence and/or compliance, cardiovascular, self-report measures, and questionnaires. The World Wide Web was searched using the Google and Google Scholar search engines, and reference lists of retrieved documents were reviewed. The search strategy was verified by a health librarian. Instruments were included if they specifically addressed medication adherence as a discrete construct rather than a disease-specific or a generic health status measurement. Findings: Despite of the problems with medication adherence identified in the literature, only 7 instruments met the search criteria. There was limited use of instruments across studies and settings to enable comparison across populations and extensive psychometric evaluation. Conclusions: Medication adherence is a complex, multifaceted construct dependent on a range of physical, social, economic, and psychological considerations. In spite of the importance of adherence in ensuring optimal cardiovascular outcomes, conceptual underpinnings and methods of assessing medication adherence require further discussion and debate.

Medication communication during handover interactions in specialty practice settings

AIMS AND OBJECTIVES: To investigate what and how medication information is communicated during handover interactions in specialty hospital settings. BACKGROUND: Effective communication about patients' medications between health professionals and nurses at handover is vital for the delivery of safe continuity of care. DESIGN: An exploratory qualitative design and observational study. METHODS: Participant observation was undertaken at a metropolitan Australian public hospital in four specialty settings: cardiothoracic care, intensive care, emergency care and oncology care. A medication communication model was applied to the data and thematic analysis was performed. RESULTS: Over 130 hours of observational data were collected. In total, 185 (predominately nursing) handovers were observed across the four specialty settings involving 37 nurse participants. Health professionals communicated partial details of patients' medication regimens, by focusing on auditing the medication administration record, and through the handover approach employed. Gaps in medication information at handover were evident as shown by lack of communication about detailed and specific medication content. Incoming nurses rarely posed questions about medications at handover. CONCLUSIONS: Handover interactions contained restricted and incomplete medication information. Improving the transparency, completeness and accuracy of medication communication is vital for optimising patient safety and quality of care in specialty practice settings. RELEVANCE TO CLINICAL PRACTICE: For nurses to make informed and rapid decisions regarding appropriate patient care, information about all types of prescribed medications is essential, which is communicated in an explicit and clear way. Jargon and assumptions related to medication details should be minimised to reduce the risk of misunderstandings. Disclosure of structured medication information supports nurses to perform accurate patient assessments, make knowledgeable decisions about the appropriateness of medications and their doses, and anticipate possible adverse events associated with medications. In addition, benefits of patient and family member contributions in communicating about medications at handover should also be considered.

Challenges and opportunities of undertaking a video ethnographic study to understand medication communication

AIMS AND OBJECTIVES: To examine the challenges and opportunities of undertaking a video ethnographic study on medication communication among nurses, doctors, pharmacists and patients. BACKGROUND: Video ethnography has proved to be a dynamic and useful method to explore clinical communication activities. This approach involves filming actual behaviours and activities of clinicians to develop new knowledge and to stimulate reflections of clinicians on their behaviours and activities. However, there is limited information about the complex negotiations required to use video ethnography in actual clinical practice. DESIGN: Discursive paper. METHOD: A video ethnographic approach was used to gain better understanding of medication communication processes in two general medical wards of a metropolitan hospital in Melbourne, Australia. This paper presents the arduous and delicate process of gaining access into hospital wards to video-record actual clinical practice and the methodological and ethical issues associated with video-recording. CONCLUSIONS: Obtaining access to clinical settings and clinician consent are the first hurdles of conducting a video ethnographic study. Clinicians may still feel intimidated or self-conscious in being video recorded about their medication communication practices, which they could perceive as judgements being passed about their clinical competence. By thoughtful and strategic planning, video ethnography can provide in-depth understandings of medication communication in acute care hospital settings. Ethical issues of informed consent, patient safety and respect for the confidentiality of patients and clinicians need to be carefully addressed to build up and maintain trusting relationships between researchers and participants in the clinical environment. RELEVANCE TO CLINICAL PRACTICE: By prudently considering the complex ethical and methodological concerns of using video ethnography, this approach can help to reveal the unpredictability and messiness of clinical practice. The visual data generated can stimulate clinicians' reflexivity about their norms of practice and bring about improved communication about managing medications.

Financial model for investment recovery period in electronic health records implementations

Technology usage for better healthcare delivery is being emphasised in the USA and other advanced nations. Electronic health records (EHR) are being widely seen as improving operational efficiency and reducing medication errors in clinic practices and hospitals. Further, hospitals and clinics stand to gain incentives from the federal government if they implement EHRs and demonstrate meaningful use of EHRs. While numerous other aspects of HER implementations is found in literature, financial models have not been well studied. Before implementing EHR, one must take into consideration investment recovery period considering the costs, savings and possible tax incentives. In this paper, we develop financial model for computing investment recovery period in EHR implementations assuming constant patient visits. We further develop required growth rate formula if investments need to be recovered in fixed number of years. The model is illustrated with numerical example.

Stressors and coping resources of Australian kidney transplant recipients related to medication taking: a qualitative study

AIM: To understand the stressors related to life post-kidney transplantation, with a focus on medication adherence, and the coping resources people use to deal with these stressors. BACKGROUND: Although kidney transplantation offers enhanced quality and years of life for patients, the management of a kidney transplant post-surgery is a complex process. DESIGN: A descriptive exploratory study. METHOD: Participants were recruited from five kidney transplant units in Victoria, Australia. From March to May 2014, patients who had either maintained their kidney transplant for ≥ 8 months or had experienced a kidney graft loss due to medication non-adherence were interviewed. All audio-recordings of interviews were transcribed verbatim and underwent Ritchie and Spencer's framework analysis. RESULTS: Participants consisted of fifteen men and ten women aged 26 - 72 years old. All identified themes were categorised into: 1) Causes of distress and 2) Coping resources. Post-kidney transplantation, causes of distress included the regimented routine necessary for graft maintenance, and the everlasting fear of potential graft rejection, contracting infections and developing cancer. Coping resources utilised to manage the stressors were firstly, a shift in perspective about how easy it was to manage a kidney transplant than to be dialysis-dependent and secondly, receiving external help from fellow patients, family members and healthcare professionals in addition to utilising electronic reminders. CONCLUSION: An individual well-equipped with coping resources is able to deal with stressors better. It is recommended that changes, such as providing regular reminders about the lifestyle benefits of kidney transplantation, creating opportunities for patients to share their experiences and promoting the utilisation of a reminder alarm to take medications, will reduce the stress of managing a kidney transplant.

Nursing documentation of pressure ulcers in nursing homes: comparison of record content and patient examinations

AIM: The aim of this study was to describe the accuracy and quality of nursing documentation of the prevalence, risk factors and prevention of pressure ulcers, and compare retrospective audits of nursing documentation with patient examinations conducted in nursing homes.

DESIGN: This study used a cross-sectional descriptive design.

METHOD: A retrospective audit of 155 patients' records and patient examinations using the European Pressure Ulcer Advisory Panel form and the Braden scale, conducted in January and February 2013.

RESULTS: The prevalence of pressure ulcers was 38 (26%) in the audit of the patient records and 33 (22%) in patient examinations. A total of 17 (45%) of the documented pressure ulcers were not graded. When comparing the patient examinations with the patient record contents, the patient records lacked information about pressure ulcers and preventive interventions.