System Configuration Evaluation for a Province-Wide Clinical Information System Using the eSafety Checklist

BACKGROUND

According to Digital Health Canada 2013 eSafety Guidelines, an estimated one-third of patient safety incidents following implementation of clinical information systems (CISs) are technology-related. An eSafety checklist was previously developed to improve CIS safety by providing a comprehensive listing of system-agnostic, evidence-based configuration recommendations.

OBJECTIVES

We sought to use the checklist to support safe initial configuration of a provincial system-wide CIS (Alberta, Canada), referred to as Connect Care.

METHODS

The checklist was applied to 13 Connect Care modules in three successive phases. First, the checklist was adapted to an abbreviated high-priority version. Second, demonstrations of each module were recorded. Finally, independent evaluation of each recording was conducted by two eSafety evaluators using the abbreviated eSafety checklist.

RESULTS

All modules achieved greater than 72% compliance, with an average of 84%. Overall, 273 opportunities for improvement were identified, with four major areas or themes emerging: (1) inconsistent date and time, (2) unclear patient identification, (3) ineffective alert system, and (4) insufficient decision support. These opportunities were forwarded to the appropriate build teams for review and implementation.

CONCLUSION

This work is the first to utilize the eSafety checklist in a real-world CIS, which will become one of the largest in Canada. The checklist has shown clinical applicability in identifying gaps in CIS configuration and should be considered for use in future and pre-existing CISs.

Impact of Disease Activity and Inflammatory Bowel Disease Subtype on Quality of Life in Preconception and Pregnant Patients

BACKGROUND

Inflammatory bowel disease (IBD) and pregnancy both impact health-related quality of life (HRQoL). However, little is known about IBD-related HRQoL around pregnancy.

AIMS

To assess the trajectory and predictors of HRQoL in preconception and pregnant patients with Crohn's disease (CD) and ulcerative colitis (UC).

METHODS

Preconception and pregnant patients with IBD were followed prospectively from preconception to twelve months postpartum at a tertiary referral centre. Participants completed the Short IBD Questionnaire (SIBDQ) and were assessed for clinical disease activity (modified Harvey Bradshaw Index or partial Mayo score) and objective disease activity (C-reactive protein [CRP], fecal calprotectin [FCP]).

RESULTS

A total of 61 patients with IBD (25 CD, 36 UC) were included. During preconception, patients with UC had higher SIBDQ bowel and social sub-scores than those with CD, but this reversed during postpartum. Patients with CD but not UC developed a significant, sustained improvement in SIBDQ upon becoming pregnant, which persisted into 12 months postpartum. In a multivariable linear regression model, clinical disease activity negatively predicted SIBDQ at every pregnancy timepoint and up to 12 months postpartum. SIBDQ was significantly lower in patients with CRP ≥ 8.0 mg/L during trimester 1 (T1), but not later in pregnancy. SIBDQ bowel sub-scores were significantly lower in patients with FCP ≥ 250 mg/kg at T2, T3, and 6 months postpartum.

CONCLUSIONS

Clinical disease activity is a consistent negative predictor of HRQoL from conception to 12 months postpartum. Patients with UC experience better preconception HRQoL but suffer worse postpartum HRQoL than those with CD.

Artificial intelligence enabled automated diagnosis and grading of ulcerative colitis endoscopy images

Endoscopic evaluation to reliably grade disease activity, detect complications including cancer and verification of mucosal healing are paramount in the care of patients with ulcerative colitis (UC); but this evaluation is hampered by substantial intra- and interobserver variability. Recently, artificial intelligence methodologies have been proposed to facilitate more objective, reproducible endoscopic assessment. In a first step, we compared how well several deep learning convolutional neural network architectures (CNNs) applied to a diverse subset of 8000 labeled endoscopic still images derived from HyperKvasir, the largest multi-class image and video dataset from the gastrointestinal tract available today. The HyperKvasir dataset includes 110,079 images and 374 videos and could (1) accurately distinguish UC from non-UC pathologies, and (2) inform the Mayo score of endoscopic disease severity. We grouped 851 UC images labeled with a Mayo score of 0-3, into an inactive/mild (236) and moderate/severe (604) dichotomy. Weights were initialized with ImageNet, and Grid Search was used to identify the best hyperparameters using fivefold cross-validation. The best accuracy (87.50%) and Area Under the Curve (AUC) (0.90) was achieved using the DenseNet121 architecture, compared to 72.02% and 0.50 by predicting the majority class ('no skill' model). Finally, we used Gradient-weighted Class Activation Maps (Grad-CAM) to improve visual interpretation of the model and take an explainable artificial intelligence approach (XAI).

Management of Inflammatory Bowel Disease Patients With Clinical Care Pathways Reduces Emergency Department Utilization

Background

Standardizing care through pathways has the potential to reduce emergency department (ED) utilization. We developed and evaluated inflammatory bowel disease (IBD) care pathways for that purpose.

Methods

Over 2014-2016, IBD patients were retrospectively stratified into those managed and not managed by pathways. Patient data were extracted, and negative binomial regression used to predict the annual number of ED visits.

Results

There was a difference of 30.7 ED visits/100 patients between managed and nonmanaged at 12 months (P < 0.001). The incidence rate ratio of total ED visits occurring annually was 0.750 (P = 0.008).

Conclusions

Management with IBD care pathways reduces ED utilization.

An overview of clinical decision support systems: benefits, risks, and strategies for success

Computerized clinical decision support systems, or CDSS, represent a paradigm shift in healthcare today. CDSS are used to augment clinicians in their complex decision-making processes. Since their first use in the 1980s, CDSS have seen a rapid evolution. They are now commonly administered through electronic medical records and other computerized clinical workflows, which has been facilitated by increasing global adoption of electronic medical records with advanced capabilities. Despite these advances, there remain unknowns regarding the effect CDSS have on the providers who use them, patient outcomes, and costs. There have been numerous published examples in the past decade(s) of CDSS success stories, but notable setbacks have also shown us that CDSS are not without risks. In this paper, we provide a state-of-the-art overview on the use of clinical decision support systems in medicine, including the different types, current use cases with proven efficacy, common pitfalls, and potential harms. We conclude with evidence-based recommendations for minimizing risk in CDSS design, implementation, evaluation, and maintenance.

Pregnancy-Related Beliefs and Concerns of Inflammatory Bowel Disease Patients Modified After Accessing e-Health Portal

Objective

Poor inflammatory bowel disease (IBD)-specific reproductive knowledge is associated with concerns and medication noncompliance. Having shown an educational portal can improve knowledge, we evaluated its effectiveness for addressing IBD patients’ reproductive and medication concerns.

Methods

Adult IBD participants (aged 18 to 45 years) were invited to access an e-health portal providing information on heritability, fertility, surgery, pregnancy outcomes, delivery, postpartum, and breastfeeding in the context of IBD and IBD medications. At pre-, post-, and 6+-month postintervention, participants completed a questionnaire on IBD-specific pregnancy concerns, medication concerns from the Beliefs About Medicines Questionnaire (BMQ), and medication adherence via the Medication Adherence Rating Scale (MARS). The Wilcoxon signed-rank test was used to compare median differences between scores (95% confidence).

Results

Demographics for 78 (70.3%) participants completing postintervention questionnaires: median age 29.3 (interquartile range: 25.6 to 32.9) years; 54 (69.2%) Crohn’s disease; 21 (26.9%) ulcerative colitis; 63 (80.3%) females, 5 (7.9%) pregnant; and 19 (30.2%) previously pregnant. Postintervention, the median number of reproductive concerns decreased from 3 to 1, and remained stable 6+ months later (P < 0.001*). The median BMQ score decreased from 28 to 25, and remained stable 6+ months later (P = 0.032*). Participants adherent to medications increased from 82.4% to 87.8% postintervention (P = 0.099).

Conclusion

Using an e-health portal may potentially reduce IBD-specific reproductive and medications concerns. An e-health portal is feasible as one component of managing IBD patient’s reproductive and medication concerns during preconception and pregnancy.

Knowledge translation dataset: An e-health intervention for pregnancy in inflammatory bowel disease

This article presents data collected from a cohort of patients with inflammatory bowel disease, who expressed interest in family planning and reproductive health in their clinical context. They were randomized (1:1, text-only vs. multimedia content) to access an online e-health portal containing educational information on the topic. The data collected includes baseline demographics, medication history, reproductive history, as well as standardized, validated questionnaires on knowledge ('CCPKnow'), reproductive concerns, beliefs about medications ('BMQ'), and medication adherence ('MARS-5'). These questionnaires were administered prior to the intervention, immediately after accessing the materials, and a minimum of 6 months later (without re-accessing the online material). Two publications have been generated from analysis and aggregation of the CCPKnow data ("Pregnancy-related Beliefs and Concerns of Inflammatory Bowel Disease Patients are Modified After Accessing e-Health Portal" (Sutton et al., in press), "Innovative Online Educational Portal Improves Disease-Specific Reproductive Knowledge Among Patients With Inflammatory Bowel Disease" (Sutton et al., 2018) however this is an extensive dataset that could be analyzed or combined with others' datasets for further insights.

Picture archiving and communications systems (PACS) in private practice: St. Paul Radiology PACS project

While much of diagnostic radiology is practiced in a private setting, many of the papers and presentations describing picture archiving and communications systems (PACS) implementation to date have been sponsored by university and government health facilities. We will present a 4-year retrospective review of a private practice PACS project. The challenges, benefits, and cost analysis of the PACS project will be presented.

Diagnosis of osteomyelitis of the foot in diabetic patients: value of 111In-leukocyte scintigraphy

The noninvasive diagnosis of osteomyelitis of the foot in diabetic patients with currently available radiologic and radionuclide imaging techniques is often difficult. Recently, 111In-labeled leukocyte scintigraphy has been proposed as an attractive alternative. Accordingly, we retrospectively reviewed 51 111In-labeled leukocyte scans, 49 technetium-99m bone scans, and 49 plain radiographs obtained in 51 adults with diabetes in whom osteomyelitis of the foot was suspected. The sensitivity and specificity of these techniques were evaluated in all patients, as well as in a subgroup of 11 patients with neuroarthropathy. Results with 111In-labeled leukocyte scans were also examined in subsets of patients with soft-tissue ulcers (n = 35) and those receiving antibiotics during investigation (n = 20). Confirmation or exclusion of osteomyelitis was made surgically in 28 patients and clinically in 23. Fourteen patients had osteomyelitis. Bone scans were most sensitive (93%) but least specific (43%); plain radiographs were most specific (83%) but least sensitive (43%). 111In-labeled leukocyte scans were both sensitive (79%) and specific (78%), and remained useful in patients with neuroarthropathy, soft-tissue ulcers, and antibiotic treatment. Poor spatial resolution contributed to the false-negative and false-positive 111In-labeled leukocyte scans, suggesting that this technique should not be interpreted independent of other tests. 111In-labeled leukocyte scans are a valuable diagnostic tool for the diagnosis of pedal osteomyelitis in diabetic patients.

US-guided biopsy of neck masses in postoperative management of patients with thyroid cancer

High-frequency (10-MHz) sonography demonstrated a cervical mass or lymphadenopathy, or both, during postoperative follow-up of 52 patients who had undergone surgery for thyroid cancer. Percutaneous biopsy with ultrasonographic (US) guidance was performed in all 52 masses, 44 of which were nonpalpable. Malignant cells were obtained in 29 biopsies, and the results of 20 biopsies were negative, yielding benign lymphocytes only. Results in three biopsies were nondiagnostic due to hypocellular specimens. Therefore, 94% of biopsy results (49) of 52) were confidently assigned as either positive (56%) or negative (38%) for malignancy. There were no complications. High-frequency sonography can demonstrate clinically occult thyroid bed tumor recurrence and lymph node metastases. US-guided biopsy is an accurate and safe technique to confirm or exclude malignancy in patients at high risk of recurrence of thyroid cancer.