Health Records Research Articles

Sleep apnea test prediction based on Electronic Health Records

The identification of Obstructive Sleep Apnea (OSA) is done by a Polysomnography test which is often done in later ages. Being able to notify potential insured members at earlier ages is desirable. For that, we develop predictive models that rely on Electronic Health Records (EHR) and predict whether a person will go through a sleep apnea test after the age of 50. A major challenge is the variability in EHR records in various insured members over the years, which this study investigates as well in the context of controls matching, and prediction. Since there are many temporal variables, the RankLi method was introduced for temporal variable selection. This approach employs the t-test to calculate a divergence score for each temporal variable between the target classes. We also investigate here the need to consider the number of EHR records, as part of control matching, and whether modeling separately for subgroups according to the number of EHR records is more effective. For each prediction task, we trained 4 different classifiers including 1-CNN, LSTM, Random Forest, and Logistic Regression, on data until the age of 40 or 50, and on several numbers of temporal variables. Using the number of EHR records for control matching was found crucial, and using learning models for subsets of the population according to the number of EHR records they have was found more effective. The deep learning models, particularly the 1-CNN, achieved the highest balanced accuracy and AUC scores in both male and female groups. In the male group, the highest results were also observed at age 50 with 100 temporal variables, resulting in a balanced accuracy of 90% and an AUC of 93%.

Data-Driven Approach to Development of a Risk Score for Periprosthetic Joint Infections in Total Joint Arthroplasty Using Electronic Health Records

Periprosthetic joint infection (PJI) is an uncommon, but serious complication in total joint arthroplasty. Personalized risk prediction and risk factor management may allow better preoperative assessment and improved outcomes. We evaluated different data-driven approaches to develop surgery-specific PJI prediction models using large-scale data from the electronic health records. A large institutional arthroplasty registry was leveraged to collect data from 58,574 procedures of 41,844 patients who underwent at least one primary and/or revision hip and/or knee arthroplasty between 2000 and 2019. The registry dataset was augmented with additional clinical, procedural, and laboratory data from the electronic health records for more than 100 potential predictor variables. The main outcome was PJI within the first year after surgery. We implemented both traditional and machine learning methods for model development (lasso regression, relaxed lasso regression, ridge regression, random forest, stepwise regression, extreme gradient boosting, neural network) and used 10-fold cross-validation to calculate measures of model performance in terms of discrimination (c-statistic), cross-entropy loss, and calibration. All models performed similarly in predicting PJI risk, with negligible differences of less than 0.08 between the best and worst-performing models. The relaxed and fully relaxed lasso models using the Cox model structure outperformed the other models with concordances of 0.787 in primary hip arthroplasty and 0.722 in revision hip arthroplasty, with the number of predictors ranging from nine to 41. The concordances with the relaxed lasso models were 0.681 in primary and 0.699 in revision knee arthroplasty, with a higher number of predictors in the models. Predictors included in the models varied substantially across the four surgical groups. The incorporation of additional data from the electronic health records offers limited improvement in PJI risk stratification. Furthermore, improvement in PJI risk prediction was modest with the machine learning approaches and may not justify the added complexity.

Effectiveness of EHR systems on decision support in primary healthcare: a technology acceptance model 3 perspective.

This study aims to evaluate healthcare practitioners' perceptions of electronic health record (EHR) systems and their effectiveness in supporting clinical decision-making in Tanzanian Primary Healthcare (PHC) facilities. A mixed-methods approach was employed, combining quantitative data from structured questionnaires and qualitative insights from open-ended responses. The study was conducted in the Dodoma region of Tanzania, focusing on a diverse representation of PHC facilities, including district hospitals, health centres and dispensaries. Data were analysed using multiple linear regression for quantitative data, and thematic analysis was applied to qualitative responses. The results revealed that while EHR systems are widely used in Tanzanian PHC facilities, their impact on clinical decision-making remains limited. Only a moderate portion of practitioners perceived EHR systems as effective in decision support, and frequent system use was negatively correlated with user satisfaction. Challenges such as inadequate training and support, system crashes, slow performance and poor usability and integration into clinical workflows were significant barriers to effectively utilising EHR systems. This study contributes to the limited literature on EHR system implementation in low-resource settings, specifically Tanzania, by focusing on decision-support features within EHR systems. The findings offer valuable insights for healthcare policymakers, system designers and practitioners to optimise EHR implementation and improve healthcare outcomes in resource-constrained environments.

The Effectiveness of Recipe4Health: A Quasi-Experimental Evaluation

IntroductionFood as Medicine is increasingly recognized as an important strategy for addressing the related challenges of food insecurity and nutrition-related chronic conditions. Food as Medicine refers to integration of food-based nutrition interventions into healthcare to prevent and treat disease. However, there is limited evidence to understand the effectiveness of Food as Medicine. MethodsRecipe4Health, a comprehensive Food as Medicine program, was implemented in 4 Federally Qualified Health Centers in California for patients with food insecurity and/or nutrition-related chronic conditions. Patients were referred by a healthcare provider to a ‘Food Farmacy’ (16 weekly produce home deliveries) alone or in combination with a ‘Behavioral Pharmacy’ (16 weekly group visits). A quasi-experimental study with pre/post surveys (4 months) and propensity score matched controls for Electronic Health Record (EHR) outcomes over 12 months was conducted. Participants were 2,643 Recipe4Health patients and 2,643 controls identified from 1/2020 to 12/2022; data were analyzed from 2023-2024. ResultsThere was a significant increase in produce consumption from baseline to four months (0.41 servings/day [0.11, 0.72], p=0.007) in the Food Farmacy in combination with Behavioral Pharmacy. Compared to controls, there were improvements in non-HDL cholesterol for the Food Farmacy alone (-17.1 mg/dl[-26.9, -7.2], p<0.001) and in combination with Behavioral Pharmacy (-17 mg/dl [-28.3, -5.8], p=0.003) at 12 months. Compared to controls, HbA1c significantly decreased in the Food Farmacy alone at 12 months (-0.37%, 95% CI [-0.65, -0.08]; p=0.01), but not the Food Farmacy with Behavioral Pharmacy. ConclusionsRecipe4Health resulted in improvements in diet and multiple clinical health outcomes, such as non-HDL cholesterol and HbA1c.

Electronic health records and data exchange in the WHO European region: A subregional analysis of achievements, challenges, and prospects

Electronic health record (EHR) systems are powerful tools that enhance healthcare quality. They improve efficiency, enable data exchange, and ensure authorized access to patient information. In 2022, the World Health Organization Regional Office for Europe (WHO EURO) conducted a survey to assess the digital health capabilities of the 53 Member States. This article provides a sub-regional analysis of the status of EHR systems and major barriers to their implementation, their readiness for information sharing, and the access and re-use of EHR data.Generally, EHR implementation and national data exchange are at an advanced stage in the region, though achievements and challenges vary across subregions. While more Member States in the Eastern, Western, and Southern Europe subregions reported having centralized national EHR systems, the situation is more diverse in Northern Europe and the Asian subregions, where both centralized and decentralized EHR systems are in use. Significant barriers to EHR implementation, including funding, technical capacity, competing priorities, and lack of interoperability standards are frequently cited, while others like demand, knowledge or acceptance challenges are not reported as significant. Significant barriers were reported the most by the Central Asia subregion, while barriers had least significance in Western Europe. Five out of the six subregions reported a wide adoption of national strategies and have dedicated agencies to ensure interoperability and secure data exchange. However, only 29 Member States have established legal requirements for healthcare providers to adopt EHR systems that conform to national standards for both clinical terminology and electronic messaging, with this being most notable in Western, Eastern, and Northern Europe, and the lowest percentage of Member States in Central Asia. All Member States of the six sub regions have passed privacy and data protection legislation. The use of EHR data is widely regulated, with only five remaining Member States of WHO Europe to develop EHR legislation distributed across subregions (Southern Europe, Western and Central Asia).Looking ahead, Member States are encouraged to define national legislation governing EHR systems and their use, while ensuring the interconnectivity of the local and regional EHR systems. Sustainable funding should be allocated to the development and maintenance of these systems. Efforts should also focus on creating comprehensive roadmaps for the full implementation of health data standards, addressing interoperability at local and regional levels, and developing quality management systems for testing and certification. Additionally, monitoring and evaluation should be conducted to assess whether EHRs are contributing to national health objectives. Finally, engaging patients and intersectoral partners will be key to developing a more patient-centered approach, ensuring that EHR systems meet patient needs and expectations.

Real-World Evidence of Indapamide-Induced Rhabdomyolysis: A Retrospective Analysis of Electronic Health Records.

Previous research and pharmacovigilance monitoring activities have suggested a potential association between indapamide use and rhabdomyolysis. This study aims to investigate the potential causal relationship between the use of indapamide and rhabdomyolysis. A case-control study conducted using electronic health records data, between July 1, 2016 and December 31, 2022. Patients who have rhabdomyolysis event (cases) were matched to four controls bases on age, gender, and date. We examined the odds for indapamide exposure through three risk periods: current use, recent use, and former. The study outcome was ascertained through the presence of CK level over 1000 U/L (i.e., rhabdomyolysis event). Subsequently, a multivariable conditional logistic regression analysis was utilized to assess the causal association of indapamide exposure on the likelihood of developing rhabdomyolysis, while accounting for potential confounding variables. The study population consisted of 2965 cases and 11 860 controls. The results of the conditional logistic regression analysis indicated a lack of association between exposure to indapamide for the current users with an odds ratio (OR) of 0.6 (95% confidence intervals (CI): 0.39-1.05). The odds of recent indapamide use among cases was lower than controls (OR 0.2; 95% CI: 0.14-0.47). Lastly, the OR for former use of indapamide was 0.1 (95% CI: 0.07-0.23). In this study, we did not find association between indapamide use and rhabdomyolysis regardless timing of exposure.

The implementation of a pain navigator program in the department of Veterans Affairs' (VA) health care systems: a cluster randomized pragmatic clinical trial.

This manuscript describes the uptake of the AIM-Back Pain Navigator Pathway (PNP) designed to encourage use of non-pharmacologic care options within the Veterans Health Administration (VHA). This manuscript describes the implementation of a telehealth intervention from one arm of a multisite, embedded, cluster-randomized pragmatic trial comparing the effectiveness of two novel clinical care pathways that provide access to non-pharmacologic care for Veterans with low back pain (LBP). Ten VHA clinics. 19 pain navigators, >200 primary care physicians, and over 1000 Veterans were involved in the PNP implementation. Data were generated within the VHA electronic health record (EHR) for the ongoing AIM-Back trial to describe PNP implementation for system-level findings in terms of number of visits, and type of care received. Over a 3-year period, 9 of 10 clinics implemented the PNP within the context of the AIM-Back trial. The most frequent care recommended in the PNP included physical therapy, chiropractic, acupuncture, and yoga/tai chi. During follow-up at six-weeks, ∼50% of Veterans elected to receive a different care choice than what was initially prescribed. Notable variation across clinics was documented for PNP based on time to initiation of care and follow-up rates. Implementation of the telehealth delivered PNP provides a nuanced understanding of the introduction of novel care programs within diverse clinical settings. These findings are most applicable to care programs that are delivered remotely and involve facilitation of existing care options.

Outcomes Following Acute Kidney Injury Requiring Dialysis: A Cohort Study

ObjectiveRetrospective study to examine the outcomes of acute kidney injury requiring dialysis (AKI-D) patients that received outpatient hemodialysis as part of continued AKI-D care and explore factors associated with recovery of kidney function and discontinuation of dialysis. MethodsRecords of all admissions to Emory Dialysis centers between January 2010 to December 2021 were reviewed to include patients with confirmed diagnosis of AKI-D. Basic demographics, comorbidities, duration of hospitalization and cause of AKI were extracted from hospital records and cross-referenced with the dialysis center electronic health record. Patients were followed starting from the day of the first outpatient hemodialysis up to 180 days. All hemodialysis sessions and laboratory data were analyzed. Logistic regression models were used to examine factors associated with recovery of kidney function, defined as survival free of dialysis. Results132 patients were analyzed, corresponding to 12,662 patient-day of outpatient AKI-D care. Among those, 19 (14%) patients died during the observation period. 54% of patients were male and 70% were Black. Median Age was 64 years (IQR 49-79). 42 patients (32%) recovered enough kidney function to discontinue dialysis. Median time to kidney recovery was 31 days (IQR: 19-75), and the cumulative probability of kidney recovery at 6 months was 39%. Patients who discontinued dialysis were younger (58 vs 66 years), had higher estimated glomerular filtration rate (eGFR) at time of admission (69 vs 46 ml/min/1.73m2) to the hospital, and were less likely to have a history of hypertension (61% vs 82%). Intra-dialytic hypotension was more common in patients who did not recover kidney function. Conclusions39% of patients with AKI-D recovered kidney function within 180 days of outpatient HD start. The median time to recovery was 31 days. Younger age, higher e-GFR at time of hospital admission, and absence of hypertension were predictors of kidney recovery. Patients who recover kidney function experienced episodes of intradialytic hypotension less frequently.

Uncovering specific genetic-respiratory disease endotypes for rheumatoid arthritis risk.

We aimed to identify specific genetic-respiratory disease endotypes for rheumatoid arthritis (RA) risk. This case-control study used the Mass General Brigham (MGB) and Mayo Clinic (MC) Biobanks for discovery and replication, respectively. We matched criteria-confirmed incident RA cases to four non-RA controls on age, sex and health record history. Genetic exposures included the top 11 RA risk alleles, and a validated human leucocyte antigen (HLA) genetic risk score (GRS). We identified seven respiratory diseases by codes. Using logistic regression models adjusting for potential confounders, we estimated Rs with 95% CIs for the interactions between genetic and respiratory exposures for RA risk. We identified 653 RA cases and 2607 controls in MGB, and 428 incident RA cases and 1712 non-RA controls in MC (mean age 64, 69% female). Respiratory diseases were associated with an increased risk of RA (OR 1.34, 95% CI 1.05, 1.71). Six out of 11 non-HLA RA risk alleles interacted strongly with specific respiratory diseases for RA risk, including NFKBIE and sinusitis (OR 5.49, 95% CI 1.56, 19.4 MGB; 5.26, 95% CI 2.00, 13.86 MC) and FAM167A and acute sinusitis for seronegative RA (OR 6.00, 95% CI 2.09, 17.24 MGB; 4.90, 95% CI 1.71, 14.1 MC). The RA HLA GRS interacted synergistically with interstitial lung disease for RA risk (OR 5.41, 95% CI 2.71, 10.8 in MC), with DPB1*02:01, DRB1*16:01 and DRB1*04:04 best predicting RA (positive predictive value 61%). Several genetic-respiratory disease interactions strongly drive RA onset. If confirmed, these novel associations may reflect RA endotypes that can facilitate individualised prevention, diagnosis and treatment.

Electronic Health Record Serious Illness Conversation Dashboards: An Implementation Case Series

IntroductionDashboards are used to track and visualize quality indicators within health systems to improve clinical performance. Structured serious illness conversation (SIC) documentation templates within electronic health records (EHR) have enabled the development of SIC dashboards for quality improvement. Little is known about the successes and challenges of SIC dashboards. MethodsThis implementation case series describes findings from semi-structured interviews and surveys with nine palliative care leaders in eight health systems that implemented SIC dashboards. Interviews and case information were analyzed to identify themes. ResultsFive themes were identified. First, dashboards focus on process metrics. By creating transparent and efficient access to data on EHR-documented SICs, dashboards facilitated monitoring of the results of clinician training and quality improvement efforts. Second, palliative care leaders used dashboard data on documented SICs to implement clinician practice change strategies (e.g. data feedback; quality incentives), but clinicians had mixed reactions to data. Third, dashboards facilitated leadership investment in SIC efforts and required financial and technical resources to build and maintain. Fourth, while dashboards streamlined data collection for implementation teams, participants noted challenges with data reliability, including inadequate clinician use of structured SIC documentation templates (which most dashboards rely on for measurement). Fifth, needs and tensions arose with integrating patient-centered outcome measures as part of dashboards. ConclusionDashboards can be powerful tools for identifying gaps in SIC and driving interventions for clinician practice change. However, challenges related to clinician adoption of structured templates for SIC documentation and mixed clinician receptivity to data feedback may limit their reliability and use.

Electronic Health Record Documentation Burden Crowds Out Health Information Exchange Use By Primary Care Physicians.

Although electronic health record (EHR) documentation burden is known to be associated with reduced clinician well-being and burnout, it may have even worse unintended consequences if documentation work also crowds out other high-value EHR tasks. We examined this possibility by assessing the relationship between documentation burden and a high-value but optional EHR task: the use of health information exchange (HIE) to view patient records from outside organizations. Our study took advantage of an exogenous shock to documentation time: appointment no-shows. We found that documentation time had a strong impact on HIE use, with each additional hour spent documenting resulting in a 7.1percent reduction in the proportion of patients with an outside record viewed by the primary care physician seeing them that day. Our results point to the urgent need for policy makers to do more to reduce documentation burden.

Who transitions to bipolar disorder? A comparison of major depressive disorder, anxiety, and ADHD

BackgroundDiagnostic delays in Bipolar Disorder (BD) are common and can contribute to worse outcomes. While most studies focus on depression as a primary precursor, both anxiety and attention deficit disorders are also frequent initial diagnoses. In the current study, we utilized a large, diverse electronic health record (EHR) dataset to quantify the rates and correlates of conversion to BD from these major precursor diagnoses. MethodsOur study analyzed a comprehensive ten-year EHR dataset from Johns Hopkins Medicine, a diverse urban medical center, to assess and compare the rates and correlates of conversion to BD from Major Depressive Disorder (MDD), anxiety disorders, and ADHD. Risk factors for transition were assessed as time-varying variables in proportional hazards models. ResultsOf the 21,341 patients initially included, 1232 later transitioned to a diagnosis of BD. Adjusted-one-year conversion rates for patients with MDD, anxiety disorders, and ADHD were 4.2 %, 3.4 %, and 4.0 %, respectively, with ten-year rates at 11.4 %, 9.4 %, and 10.9 %, respectively. Age (19–29 years), treatment setting (emergency and inpatient), and psychotropic medications were associated with conversion to BD across all precursor diagnoses. Severe and psychotic forms of MDD were among the strongest risk factors for transitioning to BD. Although transition rates were lower in children, risk factors were similar, with the exception of ADHD, which showed a stronger association with BD conversion in adults. ConclusionsThe highest risk of transitioning to BD was observed in patients initially diagnosed with MDD, though significant risk was also noted in those with initial diagnoses of anxiety disorders and adult ADHD.

Behavioural economics to improve and motivate vaccination in primary care using nudges through the electronic health record: rationale and design of the BE IMMUNE randomised clinical trial

IntroductionAnnual influenza vaccination reduces disease burden but vaccination rates are suboptimal, with persistent disparities among subpopulations. The purpose of this trial is to evaluate multicomponent behavioural economic nudge interventions to clinicians and patients to increase influenza vaccination. This trial also includes an intensification nudge to reduce disparities in vaccination among older adult, primary care patients.MethodsThis is a two-part, multisite cluster randomised, pragmatic clinical trial. In the first part, a multicomponent nudge intervention will be tested over approximately 6 months (September 2023–February 2024). The second part consists of a replication trial conducted at an additional site during the following influenza season (September 2024–February 2025). Primary care clinics will be randomised to the nudge intervention or usual care. Eligible clinicians and patients at intervention clinics will receive the intervention, and patients deemed high risk for not receiving a vaccine will be further randomised to receive an intensification nudge. The primary outcome is vaccine completion during the eligible visit and the secondary outcome is vaccine completion within 3 months of the eligible visit.AnalysisThe effect of the clinic-level nudge intervention on the primary and secondary outcomes will be evaluated using generalised estimating equations (GEEs) with a clinic-level exchangeable working correlation to account for clustering of observations within the clinic. GEE models with an independent working correlation will be used to evaluate the impact of the additional intensification nudge on the primary and secondary outcomes.Ethics and disseminationThe University of Pennsylvania Institutional Review Board (IRB) approved this trial and serves as the single IRB of record (IRB #851838). Results will be disseminated via peer-reviewed publication and conference presentations.Trial registration numberNCT06057727.

Cross-modal similar clinical case retrieval using a modular model based on contrastive learning and k-nearest neighbor search

ObjectiveElectronic health record systems have made it possible for clinicians to use previously encountered similar cases to support clinical decision-making. However, most studies for similar case retrieval were based on single-modal data. The existing studies on cross-modal clinical case retrieval were limited. We aimed to develop a CRoss-Modal Retrieval (CRMR) model to retrieve similar clinical cases recorded in different data modalities. Materials and methodsThe publically available Medical Information Mart for Intensive Care-Chest X-ray (MIMIC-CXR) dataset was used for model development and testing. The CRMR model was designed as a modular model containing two feature extraction models, two feature transformation models, one feature transformation optimization model, and one case retrieval model. The ability to retrieve similar clinical cases recorded in different data modalities was facilitated by the use of contrastive deep learning and k-nearest neighbor search. ResultsThe average retrieval precision, denoted as AP@k, of the developed CRMR model, were 76.9 %@5, 76.7 %@10, 76.5 %@20, 76.3 %@50, and 77.9 %@100, respectively. Here k is the number of similar cases returned after retrieval. The average retrieval time varied from 0.013 ms to 0.016 ms with k varying from 5 to 100. Moreover, the model can retrieve similar cases with the same multiple radiographic manifestations as the query case. DiscussionThe CRMR model has shown promising cross-modal retrieval performance in clinical case analysis, with the potential for future scalability and improvement in handling diverse disease types and data modalities. The CRMR model has promising potential to aid clinicians in making optimal and explainable clinical decisions.

Nurse leaders’ recommendations for implementing LGBTQ+ inclusive practices in health systems: A qualitative descriptive study

BackgroundDiscrimination against lesbian, gay, bisexual, transgender, queer or questioning, and other sexual and gender diverse (LGBTQ+) individuals in healthcare settings remains a priority issue. Nurse leaders, who oversee the largest healthcare workforce, are critical in implementing inclusive practices to address LGBTQ+ discrimination, yet implementing LGBTQ+ inclusive practices and effective strategies is inconsistent. This qualitative descriptive study explored nurse leaders’ perceptions of LGBTQ+ inclusive practices across the United States to inform recommended strategies within health systems. MethodsTwenty-one nurse leaders (e.g., chief nurse executives) from diverse healthcare settings (e.g., children's hospitals and public health centers) in 12 states across the United States participated in semi-structured interviews from October to December 2023. Interview transcripts were analyzed using content analysis to (1) describe nurse leaders’ definitions of LGBTQ+ inclusive practices and (2) identify effective strategies for their implementation. ResultsNurse leaders defined LGBTQ+ inclusive practices to include delivering culturally responsive and competent care, fostering safe and affirming environments, integrating inclusivity into daily interactions and organizational culture, and supporting the needs of both LGBTQ+ staff and patients. Key implementation strategies included fostering a flexible organizational culture through open communication; a diverse workforce; and inclusive mission statements, policies, and strategic plans. Supportive infrastructure involved effective LGBTQ+ training and adapting electronic health record systems to include information on sexual orientation and gender identity. Active and interdisciplinary engagement from staff and leadership, as well as systematic data collection on patient and staff experiences, were essential for driving inclusivity. The advocacy highlighted included legislative support, discourse within the nursing profession, and the incorporation of LGBTQ+ inclusion into healthcare regulations and accreditation standards. Additionally, multisectoral partnerships with LGBTQ+ health experts and external organizations, along with multilevel evaluations, were recommended to develop tailored interventions and promote LGBTQ+ inclusive practices. ConclusionImplementing LGBTQ+ inclusive practices in healthcare is a multifaceted effort influenced by strategies at various levels. Promoting a supportive organizational culture, establishing comprehensive support structures, encouraging active engagement, and fostering community partnerships can address the needs of LGBTQ+ individuals more effectively. Future implementation efforts should tailor strategies to accommodate variability across healthcare settings and regions, ensuring that practices are both adaptable and sustainable.

Implementing Patient Health Questionnaire Depression Screening in Inflammatory Arthritis Clinic: Insights From a Quality Improvement Project at a Tertiary Rheumatology Center.

This quality improvement project evaluates the feasibility and sustainability of adopting the Patient Health Questionnaire (PHQ) depression screening tool into routine clinical care at a rheumatology fellows' inflammatory arthritis (IA) clinic at a large tertiary center. The aim was to achieve 50% compliance in documentation of PHQ after five months. Providers received a 30-minute education on the importance of depression screening in patients with IA. A week after the education, two-step depression screening with the PHQ-2 followed by the PHQ-9 was implemented. Nurses performed PHQ-2 at each IA clinic visit verbally and documented the results while rooming patients using an electronic health record (EHR) dotphrase. Patients completed paper forms of the PHQ-9 only if the PHQ-2 score was positive for depression. Fellows then reviewed the PHQ-9 during the clinic visit and documented it using a separate EHR dotphrase. We tracked both PHQ-2 and PHQ-9 documentation rates as the key outcome measures. Before to the intervention, depression documentation rate was only 2%. After initial poor participation, nurses achieved the aim after repeated reminders and ongoing education by the senior nurse. Fellows failed to achieve the aim despite repeated reminders and education. Lack of time during clinic visit was found to be the biggest challenge. Sustained adoption of the PHQ was difficult to achieve. Additional support at the health systems level that prioritizes depression screening may need to take place. Additional research demonstrating improved IA outcomes in screened patients may also be helpful to gain more buy-in from providers.

Formulating a Dual Prediction ML Model for Patient Outcomes and Hospital Resource Use Through Electronic Health Records

Formulating a Dual Prediction ML Model for Patient Outcomes and Hospital Resource Use Through Electronic Health Records

Development of a Practical Prediction Model for Adverse Neonatal Outcomes at the Start of the Second Stage of Labor.

To develop a prediction model for adverse neonatal outcomes using electronic fetal monitoring (EFM) interpretation data and other relevant clinical information known at the start of the second stage of labor. This was a retrospective cohort study of individuals who labored and delivered at two academic medical centers between July 2016 and June 2020. Individuals were included if they had a singleton gestation at term (more than 37 weeks of gestation), a vertex-presenting, nonanomalous fetus, and planned vaginal delivery and reached the start of the second stage of labor. The primary outcome was a composite of severe adverse neonatal outcomes. We developed and compared three modeling approaches to predict the primary outcome using factors related to EFM data (as interpreted and entered in structured data fields in the electronic health record by the bedside nurse), maternal comorbidities, and labor characteristics: traditional logistic regression, LASSO (least absolute shrinkage and selection operator), and extreme gradient boosting. Model discrimination and calibration were compared. Predicted probabilities were stratified into risk groups to facilitate clinical interpretation, and positive predictive values for adverse neonatal outcomes were calculated for each. A total of 22,454 patients were included: 14,820 in the training set and 7,634 in the test set. The composite adverse neonatal outcome occurred in 3.2% of deliveries. Of the three modeling methods compared, the logistic regression model had the highest discrimination (0.690, 95% CI, 0.656-0.724) and was well calibrated. When stratified into risk groups (no increased risk, higher risk, and highest risk), the rates of the composite adverse neonatal outcome were 2.6% (95% CI, 2.3-3.1%), 6.7% (95% CI, 4.6-9.6%), and 10.3% (95% CI, 7.6-13.8%), respectively. Factors with the strongest associations with the composite adverse neonatal outcome included the presence of meconium (adjusted odds ratio [aOR] 2.10, 95% CI, 1.68-2.62), fetal tachycardia within the 2 hours preceding the start of the second stage (aOR 1.94, 95% CI, 1.03-3.65), and number of prior deliveries (aOR 0.77, 95% CI, 0.60-0.99).

RTS, S malaria vaccination among children aged 24–59 months in the Sunyani Municipality, Ghana; 2023

BackgroundFollowing the successful pilot of RTS,S malaria vaccine, it has been recommended by the WHO for prevention of Plasmodium falciparum malaria in children living in endemic areas. Despite the proven benefits of the malaria vaccine, uptake especially of the fourth dose, remains relatively low. We assessed complete uptake of RTS,S malaria vaccine among children 24–59 months and associated factors in the Sunyani Municipality of Ghana. MethodsA cross-sectional study was conducted among 361 caregivers of children aged 24–59 months in the Sunyani Municipality, who had received at least one dose of RTS,S malaria vaccine. Structured questionnaires were administered to caregivers and relevant vaccination information was extracted from the maternal and child health record books. Binary logistic regression analysis was used to determine factors associated with complete uptake of the vaccine at 95 % confidence interval (CI). ResultsApproximately 45 % (164/361) of the children had completed the RTS,S malaria vaccination schedule. Children aged 48–59 months (AOR:0.26, 95 %CI:0.09–0.76), those with caregivers who believed that children were being used for experiment (AOR:0.29, 95 %CI:0.16–0.54) or doubted the safety of the vaccine (AOR:0.27, 95 %CI:0.14–0.52) were less likely to complete the four doses. On the other hand, children of caregivers who reported vaccine shortages and travelled long distances to access immunization services (AOR:3.15, 95 %CI:1.17–8.47), and those with adequate knowledge on the malaria vaccine (AOR:1.89, 95 %CI:1.08–3.29) were more likely to complete the vaccination schedule. ConclusionLess than half of the children who started RTS,S malaria vaccination completed the full schedule. Age of child, caregiver knowledge and perception of the malaria vaccine, and access to immunization service were associated with complete uptake of the RTS,S malaria vaccine. Strengthening caregiver education and access to immunization services could improve uptake of the subsequent doses and coverage of the fourth dose.

Evaluating the Implementation and Clinical Effectiveness of an Innovative Digital First Care Model for Behavioral Health Using the RE-AIM Framework: Quantitative Evaluation.

In the United States, innovation is needed to address the increasing need for mental health care services and widen the patient-to-provider ratio. Despite the benefits of digital mental health interventions (DMHIs), they have not been effective in addressing patients' behavioral health challenges as stand-alone treatments. This study evaluates the implementation and effectiveness of precision behavioral health (PBH), a digital-first behavioral health care model embedded within routine primary care that refers patients to an ecosystem of evidence-based DMHIs with strategically placed human support. Patient demographic information, triage visit outcomes, multidimensional patient-reported outcome measure, enrollment, and engagement with the DMHIs were analyzed using data from the electronic health record and vendor-reported data files. The RE-AIM (Reach, Effectiveness, Adoption, Implementation, Maintenance) framework was used to evaluate the implementation and clinical effectiveness outcomes of PBH. PBH had a 47.58% reach rate, defined as patients accepting the PBH referral from their behavioral health integrated clinician. PBH patients had high DMHI registration rates (79.62%), high activation rates (76.54%), and high retention rates at 15 days (57.69%) and 30 days (44.58%) compared to literature benchmarks. In total, 74.01% (n=168) of patients showed clinical improvement, 22.47% (n=51) showed no clinical change, and 3.52% (n=8) showed clinical deterioration in symptoms. PBH had high adoption rates, with behavioral health integrated clinicians referring on average 4.35 (SD 0.46) patients to PBH per month and 90%-100% of clinicians (n=12) consistently referring at least 1 patient to PBH each month. A third (32%, n=1114) of patients were offered PBH as a treatment option during their triage visit. PBH as a care model with evidence-based DMHIs, human support for patients, and integration within routine settings offers a credible service to support patients with mild to moderate mental health challenges. This type of model has the potential to address real-life access to care problems faced by health care settings.