Parkview Health Research Repository
Not a member yet
774 research outputs found
Sort by
Parents’ Feelings about School Reopening during COVID-19: Survey Results from Northeast Indiana
A report compiled by a team from the Health Services and Informatics Research Group. Assess concerns and attitudes related to child(ren)’s return to school amid the COVID-19 pandemic among parents in Northeast Indiana. Method, results, and conclusion of survey results
Making a Difference in Infant Mortality
Making a Difference in Infant Mortality is a short health feature in the September 2020 issue of Fort Wayne\u27s GLO magazine.
Ethel Massing, innovation project specialist at the Parkview Mirro Center for Research & Innovation and Paige Wilkins, executive director of Healthier Moms and Babies are featured with their work in reducing Allen County Infant Mortality
Pneumothoraces Prevented With Use of Electromagnetic Device to Place Feeding Tubes.
BACKGROUND: A US Food and Drug Administration safety letter warned about the risk for pneumothoraces during feeding tube insertion despite the use of electromagnetic placement devices that provide real-time visualization of feeding tube position.
OBJECTIVES: To systematically assess pulmonary placement and pneumothoraces in CORTRAK-assisted feeding tube insertions.
METHODS: CINAHL, MEDLINE, and Cochrane databases were searched for studies of CORTRAK-assisted feeding tube insertion. Thirty-two studies documenting pulmonary placement and/or complications of feeding tube insertion were found.
RESULTS: Operators recognized pulmonary placement on insertion tracings during 202 CORTRAK-assisted feeding tube insertion procedures, resulting in the immediate withdrawal of 199 feeding tubes. One pneumothorax was identified later by radiography. Seven pulmonary placements were not recognized by CORTRAK operators at the time of feeding tube insertion, resulting in 2 pneumothoraces. The incidence of pneumothorax for CORTRAK-assisted feeding tube insertions was 0.02% (3 of 17039). Of the feeding tubes inserted into the pulmonary system - either found during or after the procedure -1.4% (3 of 209) resulted in pneumothoraces (as opposed to the 19% to 28% incidence of pneumothorax for blind feeding tube insertions. Operators recognizing pulmonary placement on CORTRAK insertion tracings may have prevented 97% (202 of 209) of feeding tubes from being inserted farther into the respiratory tract.
CONCLUSIONS: Feeding tube insertion with an electromagnetic placement device is advantageous over blind feeding tube insertion because the operator can recognize pulmonary placement early and withdraw the feeding tube, thus decreasing the risk of pulmonary complications
Weighted Blankets: Anxiety Reduction in Adult Patients Receiving Chemotherapy.
BACKGROUND: Weighted blankets have been used as a deep touch pressure (DTP) tool to alleviate anxiety in many patient populations. Patients with cancer can experience anxiety related to diagnosis and treatments, such as chemotherapy infusions. Research on the effectiveness of weighted blankets as a DTP tool to alleviate anxiety in patients receiving chemotherapy is limited.
OBJECTIVES: This study assessed the effectiveness of a medical-grade therapeutic weighted blanket on anxiety in patients receiving their first and second chemotherapy infusions at an outpatient infusion center.
METHODS: A randomized controlled trial with a crossover design was performed with new patients in the outpatient chemotherapy infusion center. Patients were randomized to one of two study arms (given the weighted blanket during their first infusion or given the weighted blanket during their second infusion). Anxiety was measured using an anxiety inventory and a visual analog scale.
FINDINGS: Anxiety was reduced among patients receiving chemotherapy when the weighted blanket was used. Based on the results, a standard-weight, medical-grade therapeutic weighted blanket can be safely used to reduce anxiety in patients of various weights, and a visual analog scale can be a reliable indicator of patients\u27 state anxiety
Effectiveness of Multimodal Pain Therapy on Reducing Opioid Use in Surgical Geriatric Hip Fracture Patients.
BACKGROUND: There is a need for appropriate pain control in the geriatric hip fracture population to prevent diminished function, increased mortality, and opioid dependence. Multimodal pain therapy is one method for reducing pain postoperatively while also decreasing opioid use in the geriatric hip fracture patient. This study aimed to determine whether multimodal pain therapy could decrease opioid use without increasing pain scores in surgical geriatric hip fracture patients.
METHODS: This was a before-and-after cohort study. The hospital implemented multimodal pain control order sets with a standardized pain regimen and performed retrospective chart review pre- and postorder set implementation for analysis.
RESULTS: A total of 248 patients were enrolled in the study: 131 in the preorder set group and 117 in the postorder set group. The mean postoperative oral morphine equivalent (OME) was significantly lower in the postorder set group than in the preorder set group (45.1 mg vs. 63.4 mg, respectively, p = .03). Compared with the preorder set group, total OME and postoperative OME were decreased by 22.6% (95% confidence interval [CI] -44.9, -3.8), 1-tailed p \u3c .01, and 53.6% (95% CI -103.4, -16.1), 1-tailed p
CONCLUSION: Implementing a multimodal approach to pain management may help reduce opioid use and may be a critical maneuver in averting the national opioid epidemic
Untold Stories in User-Centered Design of Mobile Health: Practical Challenges and Strategies Learned From the Design and Evaluation of an App for Older Adults With Heart Failure.
BACKGROUND: User-centered design (UCD) is a powerful framework for creating useful, easy-to-use, and satisfying mobile health (mHealth) apps. However, the literature seldom reports the practical challenges of implementing UCD, particularly in the field of mHealth.
OBJECTIVE: This study aims to characterize the practical challenges encountered and propose strategies when implementing UCD for mHealth.
METHODS: Our multidisciplinary team implemented a UCD process to design and evaluate a mobile app for older adults with heart failure. During and after this process, we documented the challenges the team encountered and the strategies they used or considered using to address those challenges.
RESULTS: We identified 12 challenges, 3 about UCD as a whole and 9 across the UCD stages of formative research, design, and evaluation. Challenges included the timing of stakeholder involvement, overcoming designers\u27 assumptions, adapting methods to end users, and managing heterogeneity among stakeholders. To address these challenges, practical recommendations are provided to UCD researchers and practitioners.
CONCLUSIONS: UCD is a gold standard approach that is increasingly adopted for mHealth projects. Although UCD methods are well-described and easily accessible, practical challenges and strategies for implementing them are underreported. To improve the implementation of UCD for mHealth, we must tell and learn from these traditionally untold stories
Prognostic impact of single and multiple descriptors in pathologically staged T3N0M0 non-small cell lung cancer
Background T components of the current 8th edition of lung cancer American Joint Commission on Cancer (AJCC) staging assignment include size of primary tumor and others such as chest wall invasion. The role of the presence of multiple T3 descriptors in prognosis remains unknown. Materials and methods Using National Cancer Database (NCDB) and of AJCC 7th edition, pathologically staged (R0) N0M0 non-small cell lung cancer (NSCLC) cases diagnosed in 2010-2016 were analyzed. Selected cases had primary size larger than 5cm or staged as T3 by AJCC 7th edition despite size \u3c5cm. T3 descriptor status by 8th edition was defined as single descriptor (“T3-single”) as primary size 5-7cm or size \u3c5cm and T3 per 7th edition (“T3-other”), or multiple descriptor (“T3-multi”) as presence of both descriptors. Survival analysis was performed with validation of multivariate analyses. Results Of the 108,632 surgically resected pathologically staged N0M0R0 NSCLC cases, 9,931 met the following criteria: 8,955 as T3 Single (4,381 as T3 Size, 4,574 as T3 Other), and 884 as T3 Multi. Univariate and multivariate analyses demonstrated that T3 Multi had significantly worse overall survival than T3 Single with median survival of 37.3 vs. 69.3 months, respectively. Propensity score matching analysis validated statistical significance. Exploratory analysis also demonstrated that survival of T3 Multi group is similar to that of T4 groups. Conclusion Our retrospective analysis using NCDB suggests that prognosis of patients with multiple T3 descriptors is significantly worse than those with single descriptors. Further research may be required to accurately define prognosis of NSCLC for future staging update
Migration of Feeding Tubes Assessed by Using an Electromagnetic Device: A Cohort Study.
BACKGROUND: Bedside methods to verify placement of a feeding tube are not accurate for detecting placement within the gastrointestinal tract, increasing risk of pulmonary aspiration. Current guidelines recommend verifying placement every 4 hours, yet the rationale for this recommendation is unknown.
OBJECTIVE: To assess spontaneous migration of small-bore feeding tubes in critically ill adults.
METHODS: A prospective, repeated-measures cohort study was performed in 2 intensive care units. An electromagnetic placement device was used to assess distal feeding tube location every 24 hours for 7 days. Tube migration between zones-esophageal, gastric, and postpyloric- was considered clinically significant.
RESULTS: Feeding tubes were analyzed in 20 patients. Interrater agreement was substantial for round 2 of a blinded analysis of insertion tracings (g = 0.78); 100% agreement was achieved after unblinding. Among 62 outcomes (migration assessments), 4 feeding tubes migrated 8 times (3 forward and 5 retrograde). All migrations occurred in the postpyloric zone and none were clinically significant. Within 24 hours of insertion, 50% of feeding tubes had migrated forward. Repeated-measures analysis showed a greater likelihood of migration in patients with an endotracheal tube (relative risk, 3.46 [95% CI, 1.14-10.53]; P = .03).
CONCLUSIONS: No tubes migrated retrograde into the stomach or esophagus, challenging the practice of verifying placement every 4 hours. Verification every 24 hours may be adequate if migration is not suspected. Also, lack of visible anatomical structures on insertion tracings from an electromagnetic placement device make subtle changes in postpyloric placement difficult to identify accurately
Reaching consensus on a home infusion central line-associated bloodstream infection surveillance definition via a modified Delphi approach.
BACKGROUND: A consensus on a central line-associated bloodstream infection (CLABSI) surveillance definition in home infusion is needed to standardize measurement and benchmark CLABSI to provide data to drive improvement initiatives METHODS: Experts across fields including home infusion therapy, infectious diseases, and healthcare epidemiology convened to perform a 3-step modified Delphi approach to obtain input and achieve consensus on a candidate home infusion CLABSI definition.
RESULTS: The numerator criterion was identified by participants as involving one of the 2 following: (1) recognized pathogen isolated from blood culture and pathogen is not related to infection at another site, or (2) one of the following signs or symptoms: fever of 38°C (100.4°F), chills, or hypotension (systolic blood pressure ≤90 mm Hg), and one of the 2 following: (A) common skin contaminant isolated from 2 blood cultures drawn on separate occasions and organism is not related to infection at another site, or (B) common skin contaminant isolated from blood culture from patient with intravascular access device and provider institutes appropriate antimicrobial therapy. The criteria for a denominator included days from the day of admission with a central venous catheter to day of removal of central venous catheter. In addition, 11 inclusion criteria and 4 exclusion criteria were included.
DISCUSSION: Home infusion therapy and healthcare epidemiology experts developed candidate criteria for a home infusion CLABSI surveillance definition.
CONCLUSIONS: Home care and home infusion agencies can use this definition to monitor their own CLABSI rates and implement preventative strategies
Validation Methods to Promote Real-world Applicability of Machine Learning in Medicine
DMIP \u2720: 2020 3rd International Conference on Digital Medicine and Image Processing, Kyoto, Japan, November 2020
Abstract
The impact of Artificial Intelligence (AI) on health care has been dramatic; however, there is a considerable degree of skepticism among clinicians about the real-world applicability of advanced predictive models; for this reason, it is particularly important to emphasize the need for proper model validation in machine learning. Often model skepticism is well-placed as modelers may overclaim the real-world replicability for their models, understate the known limitations, or simply not be aware of the hidden limits of the modeling approach. Educational approaches limited to rigorous and thorough justification of all model design decisions may not be practical given model complexity. This also becomes more challenging as state-of-the-art models with the highest benchmark accuracy are becoming less interpretable, e.g. ensemble methods or deep learning. However, in the same way that test-driven development has been a successful paradigm to navigate the complex coding landscape through a focus on testable results, we have observed a similar improvement in modeling strategy when the focus of a predictive model is driven by validation targets rather than more abstract, theoretical concerns. In this study, we provide an overview of the common limitations of model validation methods in medicine. We then present solutions to address such limitations, with a focus on strengthening the validity of predictive models.
Full Text through Lin