Quantifying Health Outcome Disparity in Invasive Methicillin-Resistant Staphylococcus aureus Infection using Fairness Algorithms on Real-World Data

This study quantifies health outcome disparities in invasive Methicillin-Resistant Staphylococcus aureus (MRSA) infections by leveraging a novel artificial intelligence (AI) fairness algorithm, the Fairness-Aware Causal paThs (FACTS) decomposition, and applying it to real-world electronic health record (EHR) data. We spatiotemporally linked 9 years of EHRs from a large healthcare provider in Florida, USA, with contextual social determinants of health (SDoH). We first created a causal structure graph connecting SDoH with individual clinical measurements before/upon diagnosis of invasive MRSA infection, treatments, side effects, and outcomes; then, we applied FACTS to quantify outcome potential disparities of different causal pathways including SDoH, clinical and demographic variables. We found moderate disparity with respect to demographics and SDoH, and all the top ranked pathways that led to outcome disparities in age, gender, race, and income, included comorbidity. Prior kidney impairment, vancomycin use, and timing were associated with racial disparity, while income, rurality, and available healthcare facilities contributed to gender disparity. From an intervention standpoint, our results highlight the necessity of devising policies that consider both clinical factors and SDoH. In conclusion, this work demonstrates a practical utility of fairness AI methods in public health settings.

Clinical Prediction Models for Hospital-Induced Delirium Using Structured and Unstructured Electronic Health Record Data: Protocol for a Development and Validation Study

BACKGROUND

Hospital-induced delirium is one of the most common and costly iatrogenic conditions, and its incidence is predicted to increase as the population of the United States ages. An academic and clinical interdisciplinary systems approach is needed to reduce the frequency and impact of hospital-induced delirium.

OBJECTIVE

The long-term goal of our research is to enhance the safety of hospitalized older adults by reducing iatrogenic conditions through an effective learning health system. In this study, we will develop models for predicting hospital-induced delirium. In order to accomplish this objective, we will create a computable phenotype for our outcome (hospital-induced delirium), design an expert-based traditional logistic regression model, leverage machine learning techniques to generate a model using structured data, and use machine learning and natural language processing to produce an integrated model with components from both structured data and text data.

METHODS

This study will explore text-based data, such as nursing notes, to improve the predictive capability of prognostic models for hospital-induced delirium. By using supervised and unsupervised text mining in addition to structured data, we will examine multiple types of information in electronic health record data to predict medical-surgical patient risk of developing delirium. Development and validation will be compliant to the Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD) statement.

RESULTS

Work on this project will take place through March 2024. For this study, we will use data from approximately 332,230 encounters that occurred between January 2012 to May 2021. Findings from this project will be disseminated at scientific conferences and in peer-reviewed journals.

CONCLUSIONS

Success in this study will yield a durable, high-performing research-data infrastructure that will process, extract, and analyze clinical text data in near real time. This model has the potential to be integrated into the electronic health record and provide point-of-care decision support to prevent harm and improve quality of care.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

DERR1-10.2196/48521.

Application of a practice-based approach in variable selection for a prediction model development study of hospital-induced delirium

BACKGROUND

Prognostic models of hospital-induced delirium, that include potential predisposing and precipitating factors, may be used to identify vulnerable patients and inform the implementation of tailored preventive interventions. It is recommended that, in prediction model development studies, candidate predictors are selected on the basis of existing knowledge, including knowledge from clinical practice. The purpose of this article is to describe the process of identifying and operationalizing candidate predictors of hospital-induced delirium for application in a prediction model development study using a practice-based approach.

METHODS

This study is part of a larger, retrospective cohort study that is developing prognostic models of hospital-induced delirium for medical-surgical older adult patients using structured data from administrative and electronic health records. First, we conducted a review of the literature to identify clinical concepts that had been used as candidate predictors in prognostic model development-and-validation studies of hospital-induced delirium. Then, we consulted a multidisciplinary task force of nine members who independently judged whether each clinical concept was associated with hospital-induced delirium. Finally, we mapped the clinical concepts to the administrative and electronic health records and operationalized our candidate predictors.

RESULTS

In the review of 34 studies, we identified 504 unique clinical concepts. Two-thirds of the clinical concepts (337/504) were used as candidate predictors only once. The most common clinical concepts included age (31/34), sex (29/34), and alcohol use (22/34). 96% of the clinical concepts (484/504) were judged to be associated with the development of hospital-induced delirium by at least two members of the task force. All of the task force members agreed that 47 or 9% of the 504 clinical concepts were associated with hospital-induced delirium.

CONCLUSIONS

Heterogeneity among candidate predictors of hospital-induced delirium in the literature suggests a still evolving list of factors that contribute to the development of this complex phenomenon. We demonstrated a practice-based approach to variable selection for our model development study of hospital-induced delirium. Expert judgement of variables enabled us to categorize the variables based on the amount of agreement among the experts and plan for the development of different models, including an expert-model and data-driven model.

Risk of bias in prognostic models of hospital-induced delirium for medical-surgical units: A systematic review

PURPOSE

The purpose of this systematic review was to assess risk of bias in existing prognostic models of hospital-induced delirium for medical-surgical units.

METHODS

APA PsycInfo, CINAHL, MEDLINE, and Web of Science Core Collection were searched on July 8, 2022, to identify original studies which developed and validated prognostic models of hospital-induced delirium for adult patients who were hospitalized in medical-surgical units. The Checklist for Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modelling Studies was used for data extraction. The Prediction Model Risk of Bias Assessment Tool was used to assess risk of bias. Risk of bias was assessed across four domains: participants, predictors, outcome, and analysis.

RESULTS

Thirteen studies were included in the qualitative synthesis, including ten model development and validation studies and three model validation only studies. The methods in all of the studies were rated to be at high overall risk of bias. The methods of statistical analysis were the greatest source of bias. External validity of models in the included studies was tested at low levels of transportability.

CONCLUSIONS

Our findings highlight the ongoing scientific challenge of developing a valid prognostic model of hospital-induced delirium for medical-surgical units to tailor preventive interventions to patients who are at high risk of this iatrogenic condition. With limited knowledge about generalizable prognosis of hospital-induced delirium in medical-surgical units, existing prognostic models should be used with caution when creating clinical practice policies. Future research protocols must include robust study designs which take into account the perspectives of clinicians to identify and validate risk factors of hospital-induced delirium for accurate and generalizable prognosis in medical-surgical units.

Clinical Decision Support Systems for Palliative Care Management: A Scoping Review

CONTEXT

With the expansion of palliative care services in clinical settings, clinical decision support systems (CDSSs) have become increasingly crucial for assisting bedside nurses and other clinicians in improving the quality of care to patients with life-limiting health conditions.

OBJECTIVES

To characterize palliative care CDSSs and explore end-users' actions taken, adherence recommendations, and clinical decision time.

METHODS

The CINAHL, Embase, and PubMed databases were searched from inception to September 2022. The review was developed following the preferred reporting items for systematic reviews and meta-analyses extension for scoping reviews guidelines. Qualified studies were described in tables and assessed the level of evidence.

RESULTS

A total of 284 abstracts were screened, and 12 studies comprised the final sample. The CDSSs selected focused on identifying patients who could benefit from palliative care based on their health status, making referrals to palliative care services, and managing medications and symptom control. Despite the variability of palliative CDSSs, all studies reported that CDSSs assisted clinicians in becoming more informed about palliative care options leading to better decisions and improved patient outcomes. Seven studies explored the impact of CDSSs on end-user adherence. Three studies revealed high adherence to recommendations while four had low adherence. Lack of feature customization and trust in guideline-based in the initial stages of feasibility and usability testing were evident, limiting the usefulness for nurses and other clinicians.

CONCLUSION

This study demonstrated that implementing palliative care CDSSs can assist nurses and other clinicians in improving the quality of care for palliative patients. The studies' different methodological approaches and variations in palliative CDSSs made it challenging to compare and validate the applicability under which CDSSs are effective. Further research utilizing rigorous methods to evaluate the impact of clinical decision support features and guideline-based actions on clinicians' adherence and efficiency is recommended.

Optimizing Dynamic Antibiotic Treatment Strategies against Invasive Methicillin-Resistant Staphylococcus Aureus Infections using Causal Survival Forests and G-Formula on Statewide Electronic Health Record Data

Developing models for individualized, time-varying treatment optimization from observational data with large variable spaces, e.g., electronic health records (EHR), is problematic because of inherent, complex bias that can change over time. Traditional methods such as the g-formula are robust, but must identify critical subsets of variables due to combinatorial issues. Machine learning approaches such as causal survival forests have fewer constraints and can provide fine-tuned, individualized counterfactual predictions. In this study, we aimed to optimize time-varying antibiotic treatment -identifying treatment heterogeneity and conditional treatment effects- against invasive methicillin-resistant Staphylococcus Aureus (MRSA) infections, using statewide EHR data collected in Florida, USA. While many previous studies focused on measuring the effects of the first empiric treatment (i.e., usually vancomycin), our study focuses on dynamic sequential treatment changes, comparing possible vancomycin switches with other antibiotics at clinically relevant time points, e.g., after obtaining a bacterial culture and susceptibility testing. Our study population included adult individuals admitted to the hospital with invasive MRSA. We collected demographic, clinical, medication, and laboratory information from the EHR for these patients. Then, we followed three sequential antibiotic choices (i.e., their empiric treatment, subsequent directed treatment, and final sustaining treatment), evaluating 30-day mortality as the outcome. We applied both causal survival forests and g-formula using different clinical intervention policies. We found that switching from vancomycin to another antibiotic improved survival probability, yet there was a benefit from initiating vancomycin compared to not using it at any time point. These findings show consistency with the empiric choice of vancomycin before confirmation of MRSA and shed light on how to manage switches on course. In conclusion, this application of causal machine learning on EHR demonstrates utility in modeling dynamic, heterogeneous treatment effects that cannot be evaluated precisely using randomized clinical trials.

Validation of the tablet-based Turkish-PAINReportIt® for lung cancer patients after thoracotomy in Turkey

BACKGROUND

Digital pain assessment is advantageous and timely for healthcare priorities in Turkey. However, a multi-dimensional, tablet-based pain assessment tool is not available in the Turkish language.

PURPOSE

To validate the Turkish-PAINReportIt® as a multi-dimensional measure of post-thoracotomy pain.

METHODS

In the first of a two-phased study, 32 Turkish patients (mean age 47.8 ± 15.6 years, 72 % male) participated in individual cognitive interviews as they completed the tablet-based Turkish-PAINReportIt® once during the first four days post-thoracotomy, and 8 clinicians participated in a focus group discussion of implementation barriers. In the second phase, 80 Turkish patients (mean age 59.0 ± 12.7 years, 80 % male) completed the Turkish-PAINReportIt® preoperatively, on postoperative days 1-4, and at the two-week post-operative follow-up visit.

RESULTS

Patients generally interpreted accurately the Turkish-PAINReportIt® instructions and items. We eliminated some items unnecessary for daily assessment based on focus-group suggestions. In the second study phase, pain scores (intensity, quality, pattern) were low pre-thoracotomy for lung cancer and high postoperatively high on day 1, decreasing on days 2, 3 and 4, and back down to pre-surgical levels at 2-weeks. Over time, pain intensity decreased from post-operative day 1 to post-operative day 4 (p < .001) and from post-operative day 1 to post-operative week 2 (p < .001).

CONCLUSIONS

The formative research supported proof of concept and informed the longitudinal study. Findings showed strong validity of the Turkish-PAINReportIt® to detect reduced pain over time as healing occurs after thoracotomy.

Clinical Decision Support for Fall Prevention: Defining End-User Needs

BACKGROUND AND SIGNIFICANCE

Falls in community-dwelling older adults are common, and there is a lack of clinical decision support (CDS) to provide health care providers with effective, individualized fall prevention recommendations.

OBJECTIVES

The goal of this research is to identify end-user (primary care staff and patients) needs through a human-centered design process for a tool that will generate CDS to protect older adults from falls and injuries.

METHODS

Primary care staff (primary care providers, care coordinator nurses, licensed practical nurses, and medical assistants) and community-dwelling patients aged 60 years or older associated with Brigham & Women's Hospital-affiliated primary care clinics and the University of Florida Health Archer Family Health Care primary care clinic were eligible to participate in this study. Through semi-structured and exploratory interviews with participants, our team identified end-user needs through content analysis.

RESULTS

User needs for primary care staff (n = 24) and patients (n = 18) were categorized under the following themes: workload burden; systematic communication; in-person assessment of patient condition; personal support networks; motivational tools; patient understanding of fall risk; individualized resources; and evidence-based safe exercises and expert guidance. While some of these themes are specific to either primary care staff or patients, several address needs expressed by both groups of end-users.

CONCLUSION

Our findings suggest that there are many care gaps in fall prevention management in primary care and that personalized, actionable, and evidence-based CDS has the potential to address some of these gaps.

Application of persuasive systems design principles to design a self-management application user interface for Hispanic informal dementia caregivers: user preferences and perceptions

OBJECTIVE

We designed an mHealth application (app) user interface (UI) prototype informed by participatory design sessions, persuasive systems design (PSD) principles, and Lorig and Holman's self-management behavior framework to support self-management activities of Hispanic informal dementia caregivers and assessed their perceptions and preferences regarding features and functions of the app.

MATERIALS AND METHODS

Our observational usability study design employed qualitative methods and forced choice preference assessments to identify: (1) the relationship between user preferences for UI features and functions and PSD principles and (2) user preferences for UI design features and functions and app functionality. We evaluated 16 pairs of mHealth app UI prototype designs. Eight paper-based paired designs were used to assess the relationship between PSD principles and caregiver preferences for UI features and functions to support self-management. An Apple iPad WIFI 32GB was used to display another 8 paired designs and assess caregiver preferences for UI functions to support the self-management process.

RESULTS

Caregivers preferred an app UI with features and functions that incorporated a greater number of PSD principles and included an infographic to facilitate self-management. Moreover, caregivers preferred a design that did not depend on manual data entry, opting instead for functions such as drop-down list, drag-and-drop, and voice query to prioritize, choose, decide, and search when performing self-management activities.

CONCLUSION

Our assessment approaches allowed us to discern which UI features, functions, and designs caregivers preferred. The targeted application of PSD principles in UI designs holds promise for supporting personalized problem identification, goal setting, decision-making, and action planning as strategies for improving caregiver self-management confidence.

A Stress and Pain Self-management m-Health App for Adult Outpatients with Sickle Cell Disease (RADIANCE Study): Protocol for a Randomized Controlled Study (Preprint)

Background

This paper describes the research protocol for a randomized controlled trial of a self-management intervention for adults diagnosed with sickle cell disease (SCD). People living with SCD experience lifelong recurrent episodes of acute and chronic pain, which are exacerbated by stress.

Objective

This study aims to decrease stress and improve SCD pain control with reduced opioid use through an intervention with self-management relaxation exercises, named You Cope, We Support (YCWS). Building on our previous findings from formative studies, this study is designed to test the efficacy of YCWS on stress intensity, pain intensity, and opioid use in adults with SCD.

Methods

A randomized controlled trial of the short-term (8 weeks) and long-term (6 months) effects of YCWS on stress, pain, and opioid use will be conducted with 170 adults with SCD. Patients will be randomized based on 1:1 ratio (stratified on pain intensity [≤5 or >5]) to be either in the experimental (self-monitoring of outcomes, alerts or reminders, and use of YCWS [relaxation and distraction exercises and support]) or control (self-monitoring of outcomes and alerts or reminders) group. Patients will be asked to report outcomes daily. During weeks 1 to 8, patients in both groups will receive system-generated alerts or reminders via phone call, text, or email to facilitate data entry (both groups) and intervention use support (experimental). If the participant does not enter data after 24 hours, the study support staff will contact them for data entry troubleshooting (both groups) and YCWS use (experimental). We will time stamp and track patients’ web-based activities to understand the study context and conduct exit interviews on the acceptability of system-generated and staff support. This study was approved by our institutional review board.

Results

This study was funded by the National Institute of Nursing Research of the National Institutes of Health in 2020. The study began in March 2021 and will be completed in June 2025. As of April 2022, we have enrolled 45.9% (78/170) of patients. We will analyze the data using mixed effects regression models (short term and long term) to account for the repeated measurements over time and use machine learning to construct and evaluate prediction models. Owing to the COVID-19 pandemic, the study was modified to allow for mail-in consent process, internet-based consent process via email or Zoom videoconference, devices delivered by FedEx, and training via Zoom videoconference.

Conclusions

We expect the intervention group to report reductions in pain intensity (primary outcome; 0-10 scale) and in stress intensity (0-10 scale) and opioid use (Wisepill event medication monitoring system), which are secondary outcomes. Our study will contribute to advancing the use of nonopioid therapy such as guided relaxation and distraction techniques for managing SCD pain.

Trial Registration

ClinicalTrials.gov NCT04484272; https://clinicaltrials.gov/ct2/show/NCT04484272

International Registered Report Identifier (IRRID)

PRR1-10.2196/33818

Identification of important factors in an inpatient fall risk prediction model to improve the quality of care using EHR and electronic administrative data: A machine-learning approach

BACKGROUND

Inpatient falls, many resulting in injury or death, are a serious problem in hospital settings. Existing falls risk assessment tools, such as the Morse Fall Scale, give a risk score based on a set of factors, but don't necessarily signal which factors are most important for predicting falls. Artificial intelligence (AI) methods provide an opportunity to improve predictive performance while also identifying the most important risk factors associated with hospital-acquired falls. We can glean insight into these risk factors by applying classification tree, bagging, random forest, and adaptive boosting methods applied to Electronic Health Record (EHR) data.

OBJECTIVE

The purpose of this study was to use tree-based machine learning methods to determine the most important predictors of inpatient falls, while also validating each via cross-validation.

MATERIALS AND METHODS

A case-control study was designed using EHR and electronic administrative data collected between January 1, 2013 to October 31, 2013 in 14 medical surgical units. The data contained 38 predictor variables which comprised of patient characteristics, admission information, assessment information, clinical data, and organizational characteristics. Classification tree, bagging, random forest, and adaptive boosting methods were used to identify the most important factors of inpatient fall-risk through variable importance measures. Sensitivity, specificity, and area under the ROC curve were computed via ten-fold cross validation and compared via pairwise t-tests. These methods were also compared to a univariate logistic regression of the Morse Fall Scale total score.

RESULTS

In terms of AUROC, bagging (0.89), random forest (0.90), and boosting (0.89) all outperformed the Morse Fall Scale (0.86) and the classification tree (0.85), but no differences were measured between bagging, random forest, and adaptive boosting, at a p-value of 0.05. History of Falls, Age, Morse Fall Scale total score, quality of gait, unit type, mental status, and number of high fall risk increasing drugs (FRIDs) were considered the most important features for predicting inpatient fall risk.

CONCLUSIONS

Machine learning methods have the potential to identify the most relevant and novel factors for the detection of hospitalized patients at risk of falling, which would improve the quality of patient care, and to more fully support healthcare provider and organizational leadership decision-making. Nurses would be able to enhance their judgement to caring for patients at risk for falls. Our study may also serve as a reference for the development of AI-based prediction models of other iatrogenic conditions. To our knowledge, this is the first study to report the importance of patient, clinical, and organizational features based on the use of AI approaches.

Nutrition Support Therapies on the Medication Administration Record: Impacts on Staff Perception of Nutrition Care

BACKGROUND

It has been reported that many hospitals in the United States have fragmented and ineffective ordering, administration, documentation, and evaluation/monitoring of nutrition therapies. This paper reports on a project to investigate if perceived hospital staff awareness and documentation of nutrition support therapies (NSTs) improves by including them as part of the medication administration record (MAR).

METHODS

Surveys were conducted with nursing staff, physicians, and dietitians before and after adding NSTs to the MAR to evaluate the perceived impact on the outcome of interest. The outcomes of interest include nurses' perception of ease of finding information, awareness of an order, and ability to assess administration and documentation and dietitian, nurse, and physician staff perceptions of impact of intervention on aspects of the nutrition care process.

RESULTS

After adding NST to the MAR, nursing staff perceived improvement in knowing that their patient had an oral nutritional supplement (ONS) order (P = .01), when and how much product was last administered (P = .01), and documentation of the type of product consumed (P = .01) and volume of product consumed (P = .01). The majority of dietitian and nurses surveyed reported perceived improvement in placing and finding ONS orders, in administration of ONS, in ability to evaluate patient nutrition status, and in ONS intake and a positive impact on clinical practice.

CONCLUSION

Inclusion of NST in the MAR presents an innovative solution to enhance staff awareness of ordered therapies and perception of improved documentation of nutrition interventions for hospitalized patients.

Fall Prevention Decision Making of Acute Care Registered Nurses

OBJECTIVE

The aim of this study was to examine acute care registered nurses' (RNs') fall prevention decision-making.

BACKGROUND

The RN decision-making process related to fall prevention needs to be investigated to ensure that hospital policies align with nursing workflow and support nursing judgment.

METHODS

Qualitative semistructured interviews based on the Critical Decision Method were conducted with RNs about their planning and decision making during their last 12-hour shift worked.

RESULTS

Data saturation was achieved with 12 RNs. Nine themes emerged related to the RN decision-making process and included hospital-level (eg, fear of discipline), unit-level (eg, value of bed alarm technology), and nurse-level (eg, professional judgment) factors that could influence fall prevention.

CONCLUSIONS

Nursing administrators should consider a multilevel approach to fall prevention policies that includes promoting a practice environment that embraces self-reporting adverse events without fear of shame or being reprimanded, evaluating unit-level practice and technology acceptance and usability, and supporting autonomous nursing practice.

Language equivalence of the modified falls efficacy scale (MFES) among English- and Spanish-speaking older adults: Rasch analysis

BACKGROUND

To investigate item-level measurement properties of the Modified Falls Efficacy (MFES) Scale among English- and Spanish-speaking urban-dwelling older adults as a means to evaluate language equivalence of the tool.

METHODS

Secondary analysis of survey data from 170 English (n = 83) and Spanish (n = 87) speaking older adults who reported to the emergency department of a quaternary medical center in New York City between February 2010 and August 2011. The Rasch rating scale model was used to investigate item statistics and ordering of items, item and person reliability, and model performance of the Modified Falls Efficacy Scale.

RESULTS

The Modified Falls Efficacy Scale, for English- and Spanish-speakers, demonstrated acceptable fit to the Rasch model of a unidimensional measure. While the range of the construct is more limited for the Spanish group, the interval between tasks are much closer, reflecting little to no construct under-representation.

CONCLUSION

There is rationale for continued testing of a unidemsional English- and Spanish-MFES among urban community-dwelling older adults. Large-scale international studies linking the unidemsional MFES to patient outcomes will support the validity of this tool for research and practice.

Randomized Pilot Study: A Mobile Technology-based Self-management Intervention for Sickle Cell Pain

Little is known about the effects of self-managed relaxation interventions on pain, stress, and autonomic responses in patients with sickle cell disease (SCD). This pre-post randomized controlled pilot study was conducted to determine the feasibility of using computer tablets for relaxation intervention delivery; acceptability of study procedures; and intervention effects on pain, stress, and indicators of relaxation. The 30 research participants ranged in age from 22 years to 59 years. All were African American; 53% were male. They were randomized to an experimental group that watched a relaxation video or a control group that discussed their disease. All participants completed the study, indicating feasibility. Acceptability rates were also high. Data were obtained for the intervention's immediate effect on pain, stress, respiration, pulse, finger skin temperature, and self-reported relaxation. These preliminary findings will guide future, higher-powered studies to determine the intervention's efficacy and mechanism in SCD.The ClinicalTrials.gov Identifier: NCT02729363.

Hispanic Dementia Family Caregiver's Knowledge, Experience, and Awareness of Self-Management: Foundations for Health Information Technology Interventions

PURPOSE

As a first step toward developing a web-based Family-Health Information Management System intervention, we explored Hispanic dementia family caregiver's knowledge, use, and awareness of self-management principles and skills to address health and health care needs for themselves and the person with dementia (PWD).

METHOD

Twenty caregivers and 11 caregiver counselors attended an English or Spanish language focus group ranging from 4 to 6 participants. We conducted a directed content analysis informed by Lorig and Holman's conceptualization of self-management.

RESULTS

A complement of six skills (i.e., problem solving, decision making, resource utilization, patient-provider partnership, action planning, and self-tailoring) to achieve one of three tasks (i.e., emotional, medical, and role management) can fully represent Hispanic dementia family caregivers' ability to self-manage health and health care needs. While not prominent in our study, caregivers and caregiver counselors pointed out existing and potential uses of personal consumer technology to schedule reminders and search for resources.

DISCUSSION

A broad conceptualization of self-management may be necessary to understand Hispanic dementia family caregiver's ability and needs to address emotional, medical, and role challenges of caregiving.

CONCLUSIONS

These findings and advances in the use of consumer health information technology support the development of self-management caregiver interventions.

The effects of information and communication technologies on informal caregivers of persons living with dementia: A systematic review

Introduction

Information and communication technology (ICT) has emerged as promising to support health care consumers, including informal caregivers. This systematic review seeks to evaluate the state of the science of ICT interventions on the health of informal dementia caregivers.

Methods

We searched PubMed, CINAHL, Web of Science, and PsycINFO using concepts associated with ICT, dementia, and caregiver. Studies were assessed using the Quality Assessment Tool for Quantitative Studies.

Results

We identified 657 full-text publications. After removal of duplicates and title, abstract, and full-text screening, the quality of 12 studies was assessed. Studies varied in technology, implementation, results, and intervention evaluation.

Discussion

The methodological quality of the ICT intervention studies among dementia family caregivers was moderate to strong, yet outcome measurement was not uniform. The evidence is strongest for various forms of telephone-based interventions. However, there is a need for research that includes heterogeneous participants based on gender, race, and ethnicity.

What Can We Learn about Fall Risk Factors from EHR Nursing Notes? A Text Mining Study

INTRODUCTION

Hospital falls are a continuing clinical concern, with over one million falls occurring each year in the United States. Annually, hospital-acquired falls result in an estimated $34 billion in direct medical costs. Falls are considered largely preventable and, as a result, the Centers for Medicare and Medicaid Services have announced that fall-related injuries are no longer a reimbursable hospital cost. While policies and practices have been implemented to reduce falls, little sustained reduction has been achieved. Little empirical evidence supports the validity of published fall risk factors. While chart abstraction has been used to operationalize risk factors, few studies have examined registered nurses' (RNs') narrative notes as a source of actionable data. Therefore, the purpose of our study was to explore whether there is meaningful fall risk and prevention information in RNs' electronic narrative notes.

METHODS

This study utilized a natural language processing design. Data for this study were extracted from the publicly available Medical Information Mart for Intensive Care (MIMIC-III) database. The date comprises deidentified EHR data associated with patients who stayed in critical care units between 2001 and 2012. Text mining procedures were performed on RN's narrative notes following the traditional steps of knowledge discovery.

RESULTS

The corpus of data extracted from MIMIC-III database was comprised of 1,046,053 RNs' notes from 36,583 unique patients. We identified 3,972 notes (0.4 percent) representing 1,789 (5 percent) patients with explicit documentation related to fall risk/prevention. Around 10 percent of the notes (103,685) from 23,025 patients mentioned intrinsic (patient-related) factors that have been theoretically associated with risk of falling. An additional 1,322 notes (0.1 percent) from 692 patients (2 percent) mentioned extrinsic risk factors, related to organizational design and environment. Moreover, 7672 notes (0.7 percent) from 2,571 patients (7 percent) included information on interventions that could theoretically impact patient falls.

CONCLUSIONS

This exploratory study using a NLP approach revealed that meaningful information related to fall risk and prevention may be found in RNs' narrative notes. In particular, RNs' notes can contain information about clinical as well as environmental and organizational factors that could affect fall risk but are not explicitly recorded by the provider as a fall risk factors. In our study, potential fall risk factors were documented for more than half of the sample. Further research is needed to determine the predictive value of these factors.

IMPLICATIONS FOR POLICY OR PRACTICE

This study highlights a potentially rich but understudied source of actionable fall risk data. Furthermore, the application of novel methods to identify quality and safety measures in RNs' notes can facilitate inclusion of RNs' voices in patient outcomes and health services research.

Impact of the CMS No-Pay Policy on Hospital-Acquired Fall Prevention Related Practice Patterns

Background and Objectives

In October 2008, the Centers for Medicare & Medicaid Services (CMS) stopped reimbursing hospitals for costs related to patient falls. This study aimed to examine whether the CMS no-pay policy influenced four fall prevention practices: bed alarms, sitters, room changes, and physical restraints.

Research Design and Methods

Using electronic medical record data collected from four hospitals between 2005 and 2010, this secondary observational analysis examined the associations between the CMS no-pay policy and nursing interventions and medical orders related to fall prevention. Multivariable generalized linear mixed models with logit link function and accommodation for matching was used to assess the associations between the CMS no-pay policy and nursing interventions and medical orders.

Results

After the CMS policy change, nurses were more likely to perform one or more fall-related interventions (adjusted odds ratio (aOR): 1.667; 95% confidence interval (CI): 1.097–2.534). Of the four prevention practices, the use of bed alarms (aOR: 2.343; 95% CI: 1.409–3.897) increased significantly after the CMS policy change.

Discussion and Implications

The CMS no-pay policy increased utilization of fall prevention strategies despite little evidence that these measures prevent falls.

Interest in using mobile technology to help self-manage alcohol use among persons living with the human immunodeficiency virus: A Florida Cohort cross-sectional study

BACKGROUND

Alcohol consumption at hazardous levels is more prevalent and associated with poor health outcomes among persons living with the human immunodeficiency virus (HIV; PLWH). Although PLWH are receptive to using technology to manage health issues, it is unknown whether a cell phone app to self-manage alcohol use would be acceptable among PLWH who drink. The objectives of this study were to determine factors associated with interest in an app to self-manage drinking and to identify differences in baseline mobile technology use among PLWH by drinking level.

METHODS

The study population included 757 PLWH recruited from 2014 to 2016 into the Florida Cohort, an ongoing cohort study investigating the utilization of health services and HIV care outcomes among PLWH. Participants completed a questionnaire examining demographics, substance use, mobile technology use, and other health behaviors. Multivariable logistic regression was used to identify factors significantly associated with interest in an app to self-manage drinking. We also determined whether mobile technology use varied by drinking level.

RESULTS

Of the sample, 40% of persons who drink at hazardous levels, 34% of persons who drink at nonhazardous levels, and 19% of persons who do not drink were interested in a self-management app for alcohol use. Multivariable logistic regression analysis indicated that nonhazardous drinking (adjusted odds ratio [AOR] = 1.78; confidence interval [CI 95%]: 1.10-2.88) and hazardous drinking (AOR = 2.58; CI: 1.60-4.16) were associated with interest, controlling for age, gender, education, and drug use. Regarding mobile technology use, most of the sample reported smartphone ownership (56%), text messaging (89%), and at least one cell phone app (69%).

CONCLUSIONS

Regardless of drinking level, overall mobile technology use among PLWH was moderate, whereas PLWH who consumed alcohol expressed greater interest in a cell phone app to self-manage alcohol use. This indicates that many PLWH who drink would be interested in and prepared for a mobile technology-based intervention to reduce alcohol consumption.

Associations between hyponatraemia, volume depletion and the risk of falls in US hospitalised patients: a case-control study

OBJECTIVE

We aimed to determine if abnormal laboratory values which may indicate volume depletion are associated with increased odds of experiencing a hospital-acquired fall.

DESIGN

Matched case-control study.

SETTING

Four hospitals located in the Southeast USA.

PARTICIPANTS

Data from 699 adult fallers and 1189 matched controls (non-fallers) were collected via chart review from 2005 to 2010. Controls were matched to cases by nursing unit, time of fall and length of stay.

OUTCOME MEASURES

The primary exposures included serum sodium, blood urea nitrogen (BUN), creatinine, BUN/creatinine ratio and haematocrit. Conditional logistic regression with m:n matching was used to determine adjusted and unadjusted ORs.

RESULTS

Serum sodium levels were strongly associated with falls. In models controlling for demographic and other fall risk factors, patients with serum sodium levels of 125 mEq/L or less were associated with increased odds of experiencing a fall as compared with those with serum sodium levels of greater than 134 mEq/L (adjusted OR (aOR)=5.08, 95% CI 1.43 to 18.08). Conversely, elevated BUN, creatinine and elevated BUN/creatinine ratios were not associated with increased odds of experiencing a fall (aOR=0.64, 95% CI 0.49 to 0.84; aOR=0.70, 95% CI 0.54 to 0.92 and aOR=0.77, 95% CI 0.58 to 1.04, respectively.) CONCLUSIONS: Laboratory indices that may indicate volume depletion appear to be unrelated to falls. However, hyponatraemia does appear to be a risk factor for falls, and those with serum sodium levels below 126 mEq/L are at especially high risk. It may be that other deficits associated with hyponatraemia, like altered mental status, are associated with risk of experiencing a hospital-acquired fall. These results indicate that abnormal laboratory values, like low sodium, can be useful for identifying hospitalised patients at risk of falling. Therefore, further investigation into abnormal laboratory values as predictors of hospital-acquired falls is warranted.

Urban-Dwelling Community Members' Views on Biomedical Research Engagement

In this study, we explore community members' overall understanding and experience with biomedical research engagement. We conducted a qualitative analysis to explore a concept that emerged but was not specifically addressed in a pre-existing dataset obtained using four focus group sessions with 30 urban-dwelling community members. Transcripts were read in an iterative process, and an emergent content analysis was performed. Five main themes were identified: (a) engaging in research to contribute to personal or greater good, (b) hierarchy of trust, (c) the importance of disclosure and transparency, (d) practical barriers to research engagement, and (e) fear of research procedures. Community members view research engagement as a collaborative process whereby community members and researchers are involved in all stages of the investigation. Focusing on research engagement, and not merely participation, may enhance community knowledge of the research process and advance scientific knowledge.

Initial Usability and Feasibility Evaluation of a Personal Health Record-Based Self-Management System for Older Adults

Introduction:

Electronic personal health record-based (ePHR-based) self-management systems can improve patient engagement and have an impact on health outcomes. In order to realize the benefits of these systems, there is a need to develop and evaluate heath information technology from the same theoretical underpinnings.

Methods:

Using an innovative usability approach based in human-centered distributed information design (HCDID), we tested an ePHR-based falls-prevention self-management system—Self-Assessment via a Personal Health Record (i.e., SAPHeR)—designed using HCDID principles in a laboratory. And we later evaluated SAPHeR’s use by community-dwelling older adults at home.

Results:

The innovative approach used in this study supported the analysis of four components: tasks, users, representations, and functions. Tasks were easily learned and features such as text-associated images facilitated task completion. Task performance times were slow, however user satisfaction was high. Nearly seven out of every ten features desired by design participants were evaluated in our usability testing of the SAPHeR system. The in vivo evaluation suggests that older adults could improve their confidence in performing indoor and outdoor activities after using the SAPHeR system.

Discussion/Conclusion:

We have applied an innovative consumer-usability evaluation. Our approach addresses the limitations of other usability testing methods that do not utilize consistent theoretically based methods for designing and testing technology. We have successfully demonstrated the utility of testing consumer technology use across multiple components (i.e., task, user, representational, functional) to evaluate the usefulness, usability, and satisfaction of an ePHR-based self-management system.

Benefits and Risks in Secondary Use of Digitized Clinical Data: Views of Community Members Living in a Predominantly Ethnic Minority Urban Neighborhood

BACKGROUND

There is potential to increase the speed of scientific discovery and implement personalized health care by using digitized clinical data collected on the patient care experience. The use of these data in research raises concerns about the privacy and confidentiality of personal health information. This study explored community members' views on the secondary use of digitized clinical data to (1) recruit participants for clinical studies; (2) recruit family members of persons with an index condition for primary studies; and (3) conduct studies of information related to stored biospecimens.

METHODS

A qualitative descriptive design was used to examine the bioethical issues outlined from the perspective of urban-dwelling community members. Focus groups were used for data collection, and emergent content analysis was employed to organize and interpret the data.

RESULTS

Thirty community members attended one of four focus groups ranging in size from 4 to 11 participants. Five critical themes emerged from the focus-group material: (1) perceived motivators for research participation; (2) objective or "real-life" barriers to research participation; (3) a psychological component of uncertainty and mistrust; (4) preferred mechanisms for recruitment and participation; and (5) cultural characteristics that can impact understanding and willingness to engage in research.

CONCLUSIONS

The overriding concern of community members regarding research participation and/or secondary clinical and nonclinical use of digitized information was that their involvement would be safe and the outcome would be meaningful to them and to others. According to participants, biospecimens acquired during routine clinical visits or for research are no longer possessions of the participant. Although the loss of privacy was a concern for participants, they preferred that researchers access their personal health information using a digitized clinical file rather than through a paper-based medical record.

The Northern Manhattan Caregiver Intervention Project: a randomised trial testing the effectiveness of a dementia caregiver intervention in Hispanics in New York City

OBJECTIVES

Dementia prevalence and its burden on families are increasing. Caregivers of persons with dementia have more depression and stress than the general population. Several interventions have proven efficacy in decreasing depression and stress in selected populations of caregivers. Hispanics in New York City tend to have a higher burden of dementia caregiving compared to non-Hispanic whites (NHW) because Hispanics have a higher prevalence of dementia, tend to have high family involvement, and tend to have higher psychosocial and economic stressors. Thus, we chose to test the effectiveness of a dementia caregiving intervention, the New York University Caregiver Intervention (NYUCI), with demonstrated efficacy in spouse caregivers in Hispanic relative caregivers of persons with dementia. Including the community health worker (CHW) intervention in both arms alleviates general psychosocial stressors and allows the assessment of the effectiveness of the intervention. Compared to two original efficacy studies of the NYUCI, which included only spouse caregivers, our study includes all relative caregivers, including common law spouses, children, siblings, a nephew and nieces. This study will be the first randomised trial to test the effectiveness of the NYUCI in Hispanic caregivers including non-spouses.

METHODS AND ANALYSIS

The design of the study is a randomised controlled trial (RCT). Participants are randomised to two arms: case management by a CHW and an intervention arm including the NYUCI in addition to case management by the CHW. The duration of intervention is 6 months. The main outcomes in the trial are changes in the Geriatric Depression Scale (GDS) and the Zarit Caregiver Burden Scale (ZCBS) from baseline to 6 months.

ETHICS AND DISSEMINATION

This trial is approved by the Columbia University Medical Center Institutional Review Board (AAAI0022), and funded by the National Institute on Minority Health and Health Disparities. The funding agency has no role in dissemination.

TRIAL REGISTRATION

 www.ClinicalTrials.gov NCT01306695.

Developing Self-Management Tools with Vulnerable Populations for use in Personal Health Information Management Systems

Vulnerable populations have potential to be significant partners and informants in the development of health information technology. We describe our experience in conducting human-centered participatory design methods with community-dwelling elders in the development of a computer-based falls prevention self-management tool for use in a personal health information management system. Community-dwelling elders contributed significantly to understanding appropriate content and functions; task performance; and graphical representations that should be considered in designing our self-management tool. Design participants should include those who have and have not experienced the clinical condition being considered during the process of system development. Knowledge transfer between system developers and community members about health and personal safety issues can be facilitated through human-centered participatory design methods.

Using health information technology to engage communities in health, education, and research

The August 2011 Clinical and Translational Science Awards conference "Using IT to Improve Community Health: How Health Care Reform Supports Innovation" convened four "Think Tank" sessions. Thirty individuals, representing various perspectives on community engagement, attended the "Health information technology (HIT) as a resource to improve community health and education" session, which focused on using HIT to improve patient health, education, and research involvement. Participants discussed a range of topics using a semistructured format. This article describes themes and lessons that emerged from that session, with a particular focus on using HIT to engage communities to improve health and reduce health disparities in populations.

Information technology, nurse staffing, and patient needs

As health care organizations increasingly adopt health information technology, time-sensitive data that track patients' requirements for nursing care and nurses' responsiveness to these needs might be available to support evidence-based nurse staffing decisions. care information technologies available in hospitals and on nursing units may provide valuable sources of information that can be translated into usable data. In this study, the usefulness of electronic data obtained from a nurse tracking call light system as a source of information for quality measurement was explored. The findings point to what might be under-utilization of existing health information technology to track patients' needs and nurses' responsiveness, patient census, and patient movements. The authors recommend health information technology be used less as support for other organizational systems and more as an administrative resource that can allow nurse executives to be more actively engaged within and across nursing environments.

Nursing care quality and adverse events in US hospitals

AIM

To examine the association between nurses' reports of unmet nursing care needs and their reports of patients' receipt of the wrong medication or dose, nosocomial infections and patient falls with injury in hospitals.

BACKGROUND

Because nursing activities are often difficult to measure, and data are typically not collected by health care organisations, there are few studies that have addressed the association between nursing activities and patient outcomes.

DESIGN

Secondary analysis of cross-sectional data collected in 1999 from 10,184 staff nurses and 168 acute care hospitals in the US.

METHODS

Multivariate linear regression models estimated the effect of unmet nursing care needs on adverse events given the influence of patient factors and the care environment.

RESULTS

The proportion of necessary nursing care left undone ranged from 26% for preparing patients and families for discharge to as high as 74% for developing or updating nursing care plans. A majority of nurses reported that patients received the wrong medication or dose, acquired nosocomial infections, or had a fall with injury infrequently. However, nurses who reported that these adverse events occurred frequently varied considerably [i.e. medication errors (15%), patient falls with injury (20%), nosocomial infection (31%)]. After adjusting for patient factors and the care environment, there remained a significant association between unmet nursing care needs and each adverse event.

CONCLUSION

The findings suggest that attention to optimising patient care delivery could result in a reduction in the occurrence of adverse events in hospitals.

RELEVANCE TO CLINICAL PRACTICE

The occurrence of adverse events may be mitigated when nurses complete care activities that require them to spend time with their patients. Hospitals should engage staff nurses in the creation of policies that influence human resources management to enhance their awareness of the care environment and patient care delivery.

Variations in nursing care quality across hospitals

TITLE

Variations in nursing care quality across hospitals.

AIMS

The aim of the study was to describe Registered Nurses' reports of unmet nursing care needs and examine the variation of nursing care quality across hospitals.

BACKGROUND

Large proportions of Registered Nurses have reported leaving necessary care activities undone because they lacked the time to complete the activities. Nursing care left undone can be expected to adversely affect the quality of care. However, little is known about the degree of variation in the quality of nursing care across hospitals.

METHODS

In 2008, a secondary analysis of a 1999 survey of Registered Nurses (n = 10,184) was conducted using descriptive and comparative statistics. Data were derived from inpatient staff nurses working in acute care hospital settings (n = 168). A hospital-level measure (i.e. unmet nursing care needs) of the quality of nursing care was developed from care needs left undone among all nurses.

RESULTS

Across hospitals there was a wide range in the proportion of Registered Nurses who reported leaving each nursing care need undone. They reported leaving two of seven necessary nursing care activities undone during their last shift. After controlling for nurses' demographic information, we found statistically significant variations in the quality of nursing care across hospitals.

CONCLUSION

Differences in nursing care quality across hospitals appear to be closely associated with variations in the quality of care environments. Understanding the determinants of unmet nursing care needs can support policy decisions on systems and human resources management to enhance nurses' awareness of their care practices and the care environment.