Machine learning to develop a predictive model of pressure injury in persons with spinal cord injury.

STUDY DESIGN: A 5-year longitudinal, retrospective, cohort study. OBJECTIVES: Develop a prediction model based on electronic health record (EHR) data to identify veterans with spinal cord injury/diseases (SCI/D) at highest risk for new pressure injuries (PIs). SETTING: Structured (coded) and text EHR data, for veterans with SCI/D treated in a VHA SCI/D Center between October 1, 2008, and September 30, 2013. METHODS: A total of 4709 veterans were available for analysis after randomly selecting 175 to act as a validation (gold standard) sample. Machine learning models were created using ten-fold cross validation and three techniques: (1) two-step logistic regression; (2) regression model employing adaptive LASSO; (3) and gradient boosting. Models based on each method were compared using area under the receiver-operating curve (AUC) analysis. RESULTS: The AUC value for the gradient boosting model was 0.62 (95% CI = 0.54-0.70), for the logistic regression model it was 0.67 (95% CI = 0.59-0.75), and for the adaptive LASSO model it was 0.72 (95% CI = 0.65-80). Based on these results, the adaptive LASSO model was chosen for interpretation. The strongest predictors of new PI cases were having fewer total days in the hospital in the year before the annual exam, higher vs. lower weight and most severe vs. less severe grade of injury based on the American Spinal Cord Injury Association (ASIA) Impairment Scale. CONCLUSIONS: While the analyses resulted in a potentially useful predictive model, clinical implications were limited because modifiable risk factors were absent in the models.

Improving identification of fall-related injuries in ambulatory care using statistical text mining.

OBJECTIVES: We determined whether statistical text mining (STM) can identify fall-related injuries in electronic health record (EHR) documents and the impact on STM models of training on documents from a single or multiple facilities. METHODS: We obtained fiscal year 2007 records for Veterans Health Administration (VHA) ambulatory care clinics in the southeastern United States and Puerto Rico, resulting in a total of 26 010 documents for 1652 veterans treated for fall-related injury and 1341 matched controls. We used the results of an STM model to predict fall-related injuries at the visit and patient levels and compared them with a reference standard based on chart review. RESULTS: STM models based on training data from a single facility resulted in accuracy of 87.5% and 87.1%, F-measure of 87.0% and 90.9%, sensitivity of 92.1% and 94.1%, and specificity of 83.6% and 77.8% at the visit and patient levels, respectively. Results from training data from multiple facilities were almost identical. CONCLUSIONS: STM has the potential to improve identification of fall-related injuries in the VHA, providing a model for wider application in the evolving national EHR system.

Effects of a national safe patient handling program on nursing injury incidence rates.

OBJECTIVE: The objective of the study was to identify which components of a system-wide safe patient handling (SPH) program reduced musculoskeletal injury (MSI) due to patient handling among nurses. METHODS: The 3-year longitudinal study from 2008 to 2011 used a pretest-posttest design. The study was conducted in the Veterans Health Administration, and all medical centers participated. The outcome was 2011 MSI incidence rates due to patient-related handling for nurses, expressed as injuries per 10 000 full-time employees. RESULTS: Three organizational risk factors, bed days of care, facility complexity level, and baseline MSI incidence rate, were significantly associated with MSI incidence rate and explained 21% of its variation. Five SPH components, including deployment of ceiling lifts and other new technologies, peer leader effectiveness, competency in SPH equipment use, facility coordinator link with safety committee, and peer leader training, uniquely accounted for an additional 23% of the total variation. CONCLUSIONS: Findings provide evidence to support the effectiveness of a multicomponent approach to SPH programs given contextual considerations.

Enhancing Veteran Community Reintegration Research (ENCORE): Protocol for a Mixed Methods and Stakeholder Engagement Project.

BACKGROUND: Veteran community reintegration (CR) has been defined as participation in community life, including employment or other productive activities, independent living, and social relationships. Veteran CR is a Veterans Health Administration priority, as a substantial proportion of veterans report difficulties with veteran CR following discharge from military service. OBJECTIVE: Enhancing Veteran Community Reintegration Research (ENCORE) is a project funded by Veterans Health Administration's Health Service Research and Development Service. The goal of ENCORE is to maximize veteran and family reintegration by promoting innovative research and knowledge translation (KT) that informs and improves equitable Department of Veterans Affairs (VA) policies, programs, and services. Overall, 2 strategic objectives guide ENCORE activities: mobilize veteran CR research and promote innovation, relevance, and acceleration of veteran CR research and KT. METHODS: ENCORE uses a mixed methods and stakeholder-engaged approach to achieve objectives and to ensure that the KT products generated are inclusive, innovative, and meaningful to stakeholders. Project activities will occur over 5 years (2019-2024) in 5 phases: plan, engage, mobilize, promote, and evaluate. All activities will be conducted remotely owing to the ongoing COVID-19 pandemic. Methods used will include reviewing research funding and literature examining the gaps in veteran CR research, conducting expert informant interviews with VA program office representatives, and assembling and working with a Multistakeholder Partnership (MSP). MSP meetings will use external facilitation services, group facilitation techniques adapted for virtual settings, and a 6-step group facilitation process to ensure successful execution of meetings and accomplishment of goals. RESULTS: As of December 2022, data collection for ENCORE is ongoing, with the team completing interviews with 20 stakeholders from 16 VA program offices providing veteran CR-related services. ENCORE developed and assembled the MSP, reviewed the VA funding portfolio and veteran CR research literature, and conducted a scientific gap analysis. The MSP developed a veteran CR research agenda in 2021 and continues to work with the ENCORE team to prepare materials for dissemination. CONCLUSIONS: The goal of this program is to improve the impact of veteran CR research on policies and programs. Using a stakeholder-engaged process, insights from key stakeholder groups are being incorporated to set a research agenda that is more likely to result in a relevant and responsive veteran CR research program. Future products will include the development of an effective and relevant dissemination plan and the generation of innovative and relevant dissemination products designed for rapid KT. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/42029.

Mental Health Diagnoses are Not Associated With Indicators of Lower Quality Pain Care in Electronic Health Records of a National Sample of Veterans Treated in Veterans Health Administration Primary Care Settings.

Prior research has demonstrated disparities in general medical care for patients with mental health conditions, but little is known about disparities in pain care. The objective of this retrospective cohort study was to determine whether mental health conditions are associated with indicators of pain care quality (PCQ) as documented by primary care clinicians in the Veterans Health Administration (VHA). We used natural language processing to analyze electronic health record data from a national sample of Veterans with moderate to severe musculoskeletal pain during primary care visits in the Fiscal Year 2017. Twelve PCQ indicators were annotated from clinician progress notes as present or absent; PCQ score was defined as the sum of these indicators. Generalized estimating equation Poisson models examined associations among mental health diagnosis categories and PCQ scores. The overall mean PCQ score across 135,408 person-visits was 8.4 (SD = 2.3). In the final adjusted model, post-traumatic stress disorder was associated with higher PCQ scores (RR = 1.006, 95%CI 1.002-1.010, P = .007). Depression, alcohol use disorder, other substance use disorder, schizophrenia, and bipolar disorder diagnoses were not associated with PCQ scores. Overall, results suggest that in this patient population, presence of a mental health condition is not associated with lower quality pain care. PERSPECTIVE: This study used a natural language processing approach to analyze medical records to determine whether mental health conditions are associated with indicators of pain care quality as documented by primary care clinicians. Findings suggest that presence of a diagnosed mental health condition is not associated with lower quality pain care.

Complementary and Integrative Health Approaches and Pain Care Quality in the Veterans Health Administration Primary Care Setting: A Quasi-Experimental Analysis.

Background: Complementary and integrative health (CIH) approaches have been recommended in national and international clinical guidelines for chronic pain management. We set out to determine whether exposure to CIH approaches is associated with pain care quality (PCQ) in the Veterans Health Administration (VHA) primary care setting. Methods: We followed a cohort of 62,721 Veterans with newly diagnosed musculoskeletal disorders between October 2016 and September 2017 over 1-year. PCQ scores were derived from primary care progress notes using natural language processing. CIH exposure was defined as documentation of acupuncture, chiropractic or massage therapies by providers. Propensity scores (PSs) were used to match one control for each Veteran with CIH exposure. Generalized estimating equations were used to examine associations between CIH exposure and PCQ scores, accounting for potential selection and confounding bias. Results: CIH was documented for 14,114 (22.5%) Veterans over 16,015 primary care clinic visits during the follow-up period. The CIH exposure group and the 1:1 PS-matched control group achieved superior balance on all measured baseline covariates, with standardized differences ranging from 0.000 to 0.045. CIH exposure was associated with an adjusted rate ratio (aRR) of 1.147 (95% confidence interval [CI]: 1.142, 1.151) on PCQ total score (mean: 8.36). Sensitivity analyses using an alternative PCQ scoring algorithm (aRR: 1.155; 95% CI: 1.150-1.160) and redefining CIH exposure by chiropractic alone (aRR: 1.118; 95% CI: 1.110-1.126) derived consistent results. Discussion: Our data suggest that incorporating CIH approaches may reflect higher overall quality of care for patients with musculoskeletal pain seen in primary care settings, supporting VHA initiatives and the Declaration of Astana to build comprehensive, sustainable primary care capacity for pain management. Future investigation is warranted to better understand whether and to what degree the observed association may reflect the therapeutic benefits patients actually received or other factors such as empowering provider-patient education and communication about these approaches.

Guideline-recommended management of community-acquired pneumonia in veterans with spinal cord injury.

BACKGROUND: Pneumonia is a leading cause of death in persons with spinal cord injuries and disorders (SCI/D), but little is known about guideline-based management for this disease in persons with SCI/D. OBJECTIVES: The goal of this study was to describe guideline-based medical care for community-acquired pneumonia (CAP) in veterans with SCI/D. METHODS: A retrospective medical record review was conducted at 7 Department of Veterans Affairs (VA) medical centers where veterans with SCI/D and CAP between 2005 and 2008 were included. Outcomes assessed were receipt of blood or sputum culture, antibiotic timeliness, appropriateness of empiric antibiotic treatment, and vaccination. RESULTS: In 70 patients, 77 CAP episodes occurred and 83.1% were treated in the inpatient setting. The average age was 70.0 years and 64.9% had tetraplegia. Sputum culture was completed in 24.7% and blood culture in 59.7% of cases. Of inpatients, 79.7% had antibiotic treatment within 8 hours and 45.1% received guideline-recommended empirical antibiotics. More than 90% of inpatients received antibiotic treatment within 3 days of presentation and 78.1% received recommended treatment. The rates of pneumococcal pneumonia (89.9%) and influenza (79.7%) vaccinations were high in CAP cases. CONCLUSIONS: Diagnostic testing and early guideline-recommended treatment is variable in veterans with SCI/D and CAP. However vaccination against influenza and pneumococcal pneumonia is high. Specific guidelines for management of CAP in SCI/D patients may be needed, which reflect the unique risk factors and needs of this population. These data can be used as a benchmark for improvements in care processes for treating and managing CAP in persons with SCI/D.

Measuring pain care quality in the Veterans Health Administration primary care setting.

ABSTRACT: The lack of a reliable approach to assess quality of pain care hinders quality improvement initiatives. Rule-based natural language processing algorithms were used to extract pain care quality (PCQ) indicators from documents of Veterans Health Administration primary care providers for veterans diagnosed within the past year with musculoskeletal disorders with moderate-to-severe pain intensity across 2 time periods 2013 to 2014 (fiscal year [FY] 2013) and 2017 to 2018 (FY 2017). Patterns of documentation of PCQ indicators for 64,444 veterans and 124,408 unique visits (FY 2013) and 63,427 veterans and 146,507 visits (FY 2017) are described. The most commonly documented PCQ indicators in each cohort were presence of pain, etiology or source, and site of pain (greater than 90% of progress notes), while least commonly documented were sensation, what makes pain better or worse, and pain's impact on function (documented in fewer than 50%). A PCQ indicator score (maximum = 12) was calculated for each visit in FY 2013 (mean = 7.8, SD = 1.9) and FY 2017 (mean = 8.3, SD = 2.3) by adding one point for every indicator documented. Standardized Cronbach alpha for total PCQ scores was 0.74 in the most recent data (FY 2017). The mean PCQ indicator scores across patient characteristics and types of healthcare facilities were highly stable. Estimates of the frequency of documentation of PCQ indicators have face validity and encourage further evaluation of the reliability, validity, and utility of the measure. A reliable measure of PCQ fills an important scientific knowledge and practice gap.

Disaster preparedness and response practices among providers from the Veterans Health Administration and Veterans with spinal cord injuries and/or disorders.

OBJECTIVES: Few empirical studies have examined the disaster preparedness and response practices of individuals with spinal cord injuries and/or disorders (SCI/D) and the healthcare providers who serve them. This study was conducted to understand the experiences of Veterans Health Administration (VHA) providers and Veterans with SCI/D in recent natural disasters, and to identify lessons learned for disaster preparedness and response in the context of SCI/D. DESIGN: Semi-structured interviews were conducted with providers and Veterans recruited through seven VHA facilities that had sustained a disaster since 2003. Audio recordings of the interviews were transcribed; transcripts were analyzed using constant comparative techniques. RESULTS: Forty participants completed an interview, including 21 VHA SCI/D providers and 19 Veterans with SCI/D. Disasters experienced by participants were weather related. While many Veterans were evacuated or admitted to nearby VHA facilities, others chose to stay in their communities. All facilities had formal disaster plans and engaged in related training; however, participants explained that many aspects of a response take shape 'in the moment,' and must address both provider and Veteran needs. Dispersion of resources hindered well-coordinated care, but effective communication, teamwork, advanced warnings, and VHA's electronic medical record facilitated efforts. CONCLUSIONS: Even in the case of thorough planning, Veterans with SCI/D and their healthcare providers are faced with pressing needs during disasters, and identifying strategies to coordinate care is critical. The lessons learned are intended to inform the efforts of healthcare providers who may be involved in the care of individuals with SCI/D in future disasters.

Examining the Relationship Between Environmental Factors and Inpatient Hospital Falls: Protocol for a Mixed Methods Study.

BACKGROUND: Patient falls are the most common adverse events reported in hospitals. Although it is well understood that the physical hospital environment contributes to nearly 40% of severe or fatal hospital falls, there are significant gaps in the knowledge about the relationship between inpatient unit design and fall rates. The few studies that have examined unit design have been conducted in a single hospital (non-Veterans Health Administration [VHA]) or a small number of inpatient units, limiting generalizability. The goal of this study is to identify unit design factors contributing to inpatient falls in the VHA. OBJECTIVE: The first aim of the study is to investigate frontline and management perceptions of and experiences with veteran falls as they pertain to inpatient environmental factors. An iterative rapid assessment process will be used to analyze the data. Interview findings will directly inform the development of an environmental assessment survey to be conducted as part of aim 2 and to contribute to interpretation of aim 2. The second aim of this study is to quantify unit design factors and compare spatial and environmental factors of units with higher- versus lower-than-expected fall rates. METHODS: We will first conduct walk-through interviews with facility personnel in 10 medical/surgical units at 3 VHA medical centers to identify environmental fall risk factors. Data will be used to finalize an environmental assessment survey for nurse managers and facilities managers. We will then use fall data from the VA Inpatient Evaluation Center and patient data from additional sources to identify 50 medical/surgical nursing units with higher- and lower-than-expected fall rates. We will measure spatial factors by analyzing computer-aided design files of unit floorplans and environmental factors from the environmental assessment survey. Statistical tests will be performed to identify design factors that distinguish high and low outliers. RESULTS: The VA Health Services Research and Development Service approved funding for the study. The research protocol was approved by institutional review boards and VA research committees at both sites. Data collection started in February 2018. Results of the data analysis are expected by February 2022. Data collection and analysis was completed for aim 1 with a manuscript of results in progress. For aim 2, the medical/surgical units were categorized into higher- and lower-than-expected fall categories, the environmental assessment surveys were distributed to facility managers and nurse managers. Data to measure spatial characteristics are being compiled. CONCLUSIONS: To our knowledge, this study is the first to objectively identify spatial risks for falls in hospitals within in a large multihospital system. Findings can contribute to evidence-based design guidelines for hospitals such as those of the Facility Guidelines Institute and the Department of Veterans Affairs. The metrics for characterizing spatial features are quantitative indices that could be incorporated in larger scale contextual studies examining contributors to falls, which to date often exclude physical environmental factors at the unit level. Space syntax measures could be used as physical environmental factors in future research examining a range of contextual factors-social, personal, organizational, and environmental-that contribute to patient falls. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/24974.

Reducing serious injury from falls in two veterans' hospital medical-surgical units.

A large veteran's hospital participated in a year-long collaborative project across 9 hospitals to reduce serious injury from falls in acute care, targeting medical-surgical units. The primary objective of this project was to develop and test a set of interventions (bundles) to prevent serious physical injury (fractures and hemorrhagic bleeds) from patient falls. The interventions were implemented using tests of change on 2 medical-surgical units focused on engaging unit-based staff and combining innovations for vulnerable populations at greatest risk for injury if they fall.

Action Ethnography of Community Reintegration for Veterans and Military Service Members With Traumatic Brain Injury: Protocol for a Mixed Methods Study.

BACKGROUND: Numerous studies of community reintegration (CR) in traumatic brain injury (TBI) have been conducted in civilian populations, but research is limited in veteran and military service member populations. Little is known about how knowledge from civilian studies translates into veterans' experiences and needs. The US Department of Veterans Health Administration (VHA) recognizes the distinctive health care needs of post-9/11 veteran and military service members, particularly with TBI, including the need to bridge health and rehabilitation-related services from acute care and inpatient settings to veteran and military service members' homes and communities to facilitate CR. OBJECTIVE: The goal of this study is to better understand the experiences of veterans with complicated mild, moderate, or severe TBI; their families; and CR workers as veterans and servicemembers transition to and sustain living in communities. This paper describes the rationale, design, and methods used to reach this goal. METHODS: This five-year longitudinal mixed methods study uses both a community-engaged research (CEnR) approach and an ethnographic approach. The sample includes 30 veterans and service members with TBI, 13 family caregivers, 11 CR specialists, 16 key stakeholders, and 82 community events. Interviews and observations are coded and analyzed using hierarchical coding schemes and thematic analysis. Analyses include data from surveys, interviews, and participant observations. Content analysis is used to highlight the complex social context of reintegration and to triangulate quantitative data. Egocentric (personal) social network analysis is used to examine the support system a veteran or service member has in place to facilitate reintegration. RESULTS: Study enrollment and data collection are completed. Data analyses are underway. CONCLUSIONS: The results of this study may provide a heightened understanding of environmental factors affecting CR in complicated mild, moderate, or severe TBI. Veteran, servicemember and family voices and insights provide VHA clinicians and policy makers with an ecological view of CR that is grounded in the life experiences of veterans, military service members, and families. The results of this study provide a roadmap for designing and testing interventions to maximize CR in a variety of domains. The longitudinal ethnographic approach allows for capturing detailed experiences within the naturalistic context. CEnR allows collaborative assessment of the social context of reintegration with community members. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/14170.

Sociotechnical probabilistic risk modeling to predict injurious falls in community living centers.

The goal of this study was to apply sociotechnical probabilistic risk assessment to prioritize risks and prevention strategies for serious injurious falls of residents in nursing homes. Risk modeling teams consisted of 26 clinical and nonclinical staff from three Department of Veterans Affairs community living centers and one state Veteran's nursing home. Participants met in groups several times to identify and assign probabilities to provider and resident at-risk behaviors and equipment failures. They identified prevention strategies for the failures that accounted for the highest levels of risk. Six scenarios were modeled: (1) transferring from bed to wheelchair, (2) propelling from bedside to bathroom, (3) transferring from wheelchair to toilet, (4) transferring from toilet to wheelchair, (5) propelling from bathroom to bedside, and (6) transferring from wheelchair to bed. The greatest paths of risk were for residents with impaired mobility and high fragility. A 26% reduction in injurious falls could be achieved by (1) reducing the number of unassisted transfers through a modest improvement in response time to alarms, (2) installing automatic brake locks on 90% of wheelchairs, (3) making the wheelchair maintenance process highly reliable, and (4) decreasing improper transfer techniques by 10%.

Providers' Note-Writing Practices for Post-traumatic Stress Disorder at Five United States Veterans Affairs Facilities.

The capacity of electronic health records (EHRs) to capture desired information depends on the practices of health care providers. These practices have not been well studied in relation to post-traumatic stress disorder (PTSD). This qualitative study investigated how providers write EHR notes on PTSD through 38 interviews with providers working at five Veterans Affairs (VA) hospitals across the United States of America. Two overarching themes were prominent in the results. Providers used progress notes primarily to remember and access details for direct patient care, but only rarely for care coordination. Providers infrequently recorded information not judged to directly contribute to improved care, sometimes deliberately omitting information perceived to jeopardize patients' access to, or quality of, care. Omitted information frequently included sexual or non-military trauma. Understanding providers' thought processes can help clinicians be aware of the limitations of EHR notes as a tool for learning the histories of new patients. Similarly, researchers relying on EHR data for PTSD research should be aware of likely areas of missing data.

Processes and outcomes of the veterans health administration safe patient handling program: study protocol.

BACKGROUND: Health care workers, such as nurses, nursing aides, orderlies, and attendants, who manually move patients, are consistently listed in the top professions for musculoskeletal injuries (MSIs) by the Bureau of Labor Statistics. These MSIs are typically caused by high-risk patient caregiving activities. In 2008, a safe patient handling (SPH) program was implemented in all 153 Veterans Administration Medical Centers (VAMCs) throughout the United States to reduce patient handling injuries. OBJECTIVE: The goal of the present study is to evaluate the effects associated with the national implementation of a comprehensive SPH program. The primary objectives of the research were to determine the effectiveness of the SPH program in improving direct care nursing outcomes and to provide a context for understanding variations in program results across sites over time. Secondary objectives of the present research were to evaluate the effectiveness of the program in reducing direct and indirect costs associated with patient handling, to explore the potential mediating and moderating mechanisms, and to identify unintended consequences of implementing the program. METHODS: This 3-year longitudinal study used mixed methods of data collection at 6- to 9-month intervals. The analyses will include data from surveys, administrative databases, individual and focus group interviews, and nonparticipant observations. For this study, a 3-tiered measurement plan was used. For Tier 1, the unit of analysis was the facility, the data source was the facility coordinator or administrative data, and all 153 VAMCs participated. For Tier 2, frontline caregivers and program peer leaders at 17 facilities each completed different surveys. For Tier 3, six facilities completed qualitative site visits, which included individual interviews, focus groups, and nonparticipant observations. Multiple regression models were proposed to test the effects of SPH components on nursing outcomes related to patient handling. Content analysis and constant comparative analysis were proposed for qualitative data analysis to understand the context of implementation and to triangulate quantitative data. RESULTS: All three tiers of data for this study have been collected. We are now in the analyses and writing phase of the project, with the possibility for extraction of additional administrative data. The focus of this paper is to describe the SPH program, its evaluation study design, and its data collection procedures. This study evaluates the effects associated with the national implementation of a comprehensive SPH program that was implemented in all 153 VAMCs throughout the United States to reduce patient handling injuries. CONCLUSIONS: To our knowledge, this is the largest evaluation of an SPH program in the United States. A major strength of this observational study design is that all VAMCs implemented the program and were included in Tier 1 of the study; therefore, population sampling bias is not a concern. Although the design lacks a comparison group for testing program effects, this longitudinal field study design allows for capturing program dose-response effects within a naturalistic context. Implementation of the VA-wide SPH program afforded the opportunity for rigorous evaluation in a naturalistic context. Findings will guide VA operations for policy and decision making about resources, and will be useful for health care, in general, outside of the VA, in implementation and impact of an SPH program.