Future Health Care Utilization Research Articles

Sleep disturbances across two weeks predict future mental healthcare utilization.

Insufficient sleep costs the U.S. economy over $411 billion per year. However, most studies investigating economic costs of sleep rely on one-time measures of sleep, which may be prone to recall bias and cannot capture variability in sleep. To address these gaps, we examined how sleep metrics captured from daily sleep diaries predicted medical expenditures. Participants were 391 World Trade Center responders enrolled in the World Trade Center Health Program (mean age = 54.97 years, 89% men). At baseline, participants completed 14 days of self-reported sleep and stress measures. Mean sleep, variability in sleep, and a novel measure of sleep reactivity (i.e., how much people's sleep changes in response to daily stress) were used to predict the subsequent year's medical expenditures, covarying for age, race/ethnicity, sex, medical diagnoses, and body mass index. Mean sleep efficiency did not predict mental healthcare utilization. However, greater sleep efficiency reactivity to stress (b=$191.75, p=.027), sleep duration reactivity to stress (b=$206.33, p=.040), variability in sleep efficiency (b=$339.33, p=.002), variability in sleep duration (b=$260.87, p=.004), and quadratic mean sleep duration (b=$182.37, p=.001) all predicted greater mental healthcare expenditures. Together, these sleep variables explained 12% of the unique variance in mental healthcare expenditures. No sleep variables were significantly associated with physical healthcare expenditures. People with more irregular sleep, more sleep reactivity, and either short or long sleep engage in more mental healthcare utilization. It may be important to address these individuals' sleep problems to improve mental health and reduce healthcare costs.

Identifying future high healthcare utilization in patients with multimorbidity – development and internal validation of machine learning prediction models using electronic health record data

PurposeTo develop and internally validate prediction models with machine learning for future potentially preventable healthcare utilization in patients with multiple long term conditions (MLTC). This study is the first step in investigating whether prediction models can help identify patients with MLTC that are most in need of integrated care.MethodsA retrospective cohort study was performed with electronic health record data from adults with MLTC from an academic medical center in the Netherlands. Based on demographic and healthcare utilization characteristics in 2017, we predicted ≥ 12 outpatient visits, ≥ 1 emergency department (ED) visits, and ≥ 1 acute hospitalizations in 2018. Four machine learning models (elastic net regression, extreme gradient boosting (XGB), logistic regression, and random forest) were developed, optimized, and evaluated in a hold-out dataset for each outcome.ResultsA total of 14,486 patients with MLTC were included. Based on the area under the curve (AUC) and calibration curves, the XGB model was selected as final model for all three outcomes. The AUC was 0.82 for ≥ 12 outpatient visits, 0.76 for ≥ 1 ED visits and 0.73 for ≥ 1 acute hospitalizations. Despite adequate AUC and calibration, precision-recall curves showed suboptimal performance.ConclusionsThe final selected models per outcome can identify patients with future potentially preventable high healthcare utilization. However, identifying high-risk patients with MLTC and substantiating if they are most in need of integrated care remains challenging. Further research is warranted investigating whether patients with high healthcare utilization are indeed the most in need of integrated care and whether quantitively identified patients match the identification based on clinicians’ experience and judgment.

A LONGITUDINAL STUDY ON SNAP PARTICIPATION AND SELF-REPORTED HEALTH AMONG DIABETIC AND NON-DIABETIC INDIVIDUALS

Abstract Self-reported health is a strong predictor of future health status, new morbidity, mobility decline, illness recovery, and healthcare utilization. Because self-reported health is rooted in social position and social comparison, it dually reflects individual health perceptions and health inequities, especially among those who experience a social struggle for resources. Food insecurity is associated with higher risk of chronic diseases and higher mortality rates. Interventions such as the supplemental nutrition assistance program (SNAP) were created to help address food insecurity. Although morbidity and mortality risks caused by food insecurity is exacerbated for those with nutrition-centered chronic disease, such as diabetes, limited research has assessed the impact of food insecurity and SNAP participation on health among diabetics. Using Health and Retirement Study data, we explore the effect of SNAP participation and food insecurity on the likelihood of reporting fair or poor health comparing 24,446 diabetes to 69,260 non-diabetics individuals aged 65 and above (2004-2020). Our analysis controls for key socioeconomic factors, including age, gender, race/ethnicity, and wealth, accounting for the complex design of HRS. Our key findings reveal that food insecure diabetics have significantly higher odds of reporting fair/poor health, and diabetic SNAP participants also had significantly higher odds of reporting fair/poor health. Our study demonstrates how food insecurity and diabetes concurrently influence health and how, despite the presence of resources, such as SNAP, there was no improvement in self-reported health. This study is important as it highlights the need for future research on the effects of SNAP participation and self-reported health.

Prevalence and Risk Factors of Post-Acute Sequelae of COVID-19 Among United States Veterans

PurposeTo better understand Post-Acute Sequelae of COVID-19 (PASC) in the Veteran population, this study aims to determine the prevalence of PASC and identify risk factors associated with its development. MethodsThis retrospective cohort study included 363,825 Veterans that tested positive for COVID-19 between February 1, 2020, and September 30, 2022. The primary outcome was the development of PASC 30 to 180 days following an acute infection with SARS-CoV-2. Multivariate logistic regression was utilized to examine factors associated with PASC. ResultsOf the 363,825 Veterans included in the analysis, 164,315 (45%) displayed symptoms of PASC. The Veterans in this analysis were predominantly male, non-Hispanic White, under the age of 65 years old, and lived in an urban residence. The strongest predictors for PASC included Non-Hispanic Black or African American race compared to Non-Hispanic White race (aOR=1.14), being between the ages of 50 and 64 compared to ages 50 and below (aOR=1.80), diabetes (aOR=8.46), and severe acute infection (aOR=1.42). ConclusionResults demonstrate potential health inequities for vulnerable individuals, as well as increased risk for individuals with pre-existing comorbidities. The prevalence of PASC provides estimates for future health care utilization. The risk factors identified can aid public health interventions to reduce the burden of PASC.

Patterns of Morbidity Across the Lifespan

Background: Classification systems to segment such patients into subgroups for purposes of care management and population analytics should balance administrative simplicity with clinical meaning and measurement precision. Objective: To describe and empirically apply a new clinically relevant population segmentation framework applicable to all payers and all ages across the lifespan. Research Design and Subjects: Cross-sectional analyses using insurance claims database for 3.31 Million commercially insured and 1.05 Million Medicaid enrollees under 65 years old; and 5.27 Million Medicare fee-for-service beneficiaries aged 65 and older. Measures: The “Patient Need Groups” (PNGs) framework, we developed, classifies each person within the entire 0–100+ aged population into one of 11 mutually exclusive need-based categories. For each PNG segment, we documented a range of clinical and resource endpoints, including health care resource use, avoidable emergency department visits, hospitalizations, behavioral health conditions, and social need factors. Results: The PNG categories included: (1) nonuser, (2) low-need child, (3) low-need adult, (4) low-complexity multimorbidity, (5) medium-complexity multimorbidity, (6) low-complexity pregnancy, (7) high-complexity pregnancy, (8) dominant psychiatric/behavioral condition, (9) dominant major chronic condition, (10) high-complexity multimorbidity, and (11) frailty. Each PNG evidenced a characteristic age-related trajectory across the full lifespan. In addition to offering clinically cogent groupings, large percentages (29%–62%) of patients in two pregnancy and high-complexity multimorbidity and frailty PNGs were in a high-risk subgroup (upper 10%) of potential future health care utilization. Conclusions: The PNG population segmentation approach represents a comprehensive measurement framework that captures and categorizes available electronic health care data to characterize individuals of all ages based on their needs.

Abdominal pain is associated with an increased risk of future healthcare resource utilization in inflammatory bowel disease.

Numerous factors influence healthcare resource utilization (HRU) in inflammatory bowel disease (IBD). We previously demonstrated an association between the presence of certain IBD-related symptoms and HRU. We conducted a longitudinal study to identify the clinical variables and IBD-related symptoms predictive of HRU. This investigation utilized clinical encounters at an IBD center within a tertiary care referral center between 10/29/2015-12/31/2019. Participants were assessed over two time points (index and follow-up office visits) separated by a minimum of 6months. Demographics, endoscopic disease severity, totals and sub-scores of surveys assessing for IBD-related symptoms, HRU, and substance use, and IBD-related medications. HRU was defined as any IBD-related emergency room visit, hospitalization, or surgery during the 6months prior to follow-up appointment. We identified patients exhibiting HRU (at follow-up) and computed descriptive statistics and contingency table analyses of index appointment clinical data to identify predictors of HRU. Multivariable logistic regression models were fit incorporating significant demographic and clinical factors. 162 consecutively enrolled IBD patients (mean age 44.0years; 99f:63m; 115 Crohn's disease [CD], 45 ulcerative colitis [UC], 2 indeterminate colitis) were included. 121 patients (74.7%) exhibited HRU (mean age 43.6years; 73f:48m; 84 CD, 36 UC, 1 IC) preceding follow-up appointment. Abdominal pain (OR = 2.18, 95% CI 1.04-4.35, p = 0.04) at the index appointment was the only study variable significantly associated with HRU on bivariate analysis (Table 1). However, none of the clinical factors evaluated in this study were independently associated with HRU in our multivariable logistic regression model. In this longitudinal study, abdominal pain was the only clinical variable that demonstrated an association with future HRU (even when considering other symptoms and key variables such as disease activity, IBD-medications, and psychiatric comorbidities (i.e., anxious or depressed state). These findings reinforce the importance of regularly screening for and effectively treating abdominal pain in IBD.

First Provider Seen for an Acute Episode of Low Back Pain Influences Subsequent Health Care Utilization.

Costs associated with low back pain (LBP) continue to rise. Despite numerous clinical practice guidelines, the evaluation and treatments for LBP are variable and depend largely on the individual provider. As yet, little attention has been given to the first choice of provider. Early research indicates that the choice of first provider and the timing of interventions for LBP appear to influence utilization. This study sought to examine the association between the first provider seen and utilization. Using 2015-2018 data from a large insurer, this retrospective analysis focused on patients (29,806) seeking care for a new episode of LBP. The study identified the first provider chosen and examined the following year of medical utilization. Cox proportional hazards models were calculated using inverse probability weighting on propensity scores to evaluate the time to event and the relationship to the first choice of provider. The primary outcome was the timing and utilization of health care resources. Total health care utilization was lowest in those who first sought care with chiropractic care or physical therapy. Highest health care utilization was seen in those patients who chose the emergency department. Overall, there appears to be an association between the first choice of provider and future health care utilization. Chiropractic care and physical therapy provide nonpharmacologic and nonsurgical, guideline-based interventions. Their participation appears related to a decrease in immediate and long-term utilization of health care resources. This study expands the existing body of literature and provides a compelling case for the influence of the first provider on an acute episode of LBP. The first provider seen for an acute episode of LBP influences immediate treatment decisions, the trajectory of a specific patient episode, and future health care choices in the management of LBP.

Identifying asthma-related risks during hospitalization using the child asthma risk assessment tool

Objective: The Child Asthma Risk Assessment Tool (CARAT) identifies risk factors for asthma morbidity. We hypothesized that CARAT-identified risk factors (using a CARAT adapted for inpatient use) would be associated with future healthcare utilization and would identify areas for intervention. Methods: We reviewed CARAT data collected during pediatric asthma admissions from 2010–2015, assessing for risk factors in environmental, medical, and social domains and providing prompts for inpatient (specialist consultation or social services engagement) and post-discharge interventions (home care visit or home environmental assessment). Confirmatory factor analysis identified groups of CARAT-identified risk factors with similar effects on healthcare utilization (latent factors). Structural equation models then evaluated relationships between latent factors and future utilization. Results: There were 2731 unique patients admitted for asthma exacerbations; 1015 (37%) had complete CARAT assessments and were included in analyses. Those with incomplete CARAT assessments were more often younger and privately-insured. CARAT-identified risk factors across domains were common in children hospitalized for exacerbations. Risks in the environmental domain were most common. Inpatient asthma consults by pulmonologists or allergists and home care referrals were the most frequent interventions indicated (62%, 628/1015, and 50%, 510/1015, respectively). Two latent factors were positively associated with healthcare utilization in the year after index stay – social stressors and known/suspected allergies (both p < 0.05). Stratified analyses analyzing data just from those children with prior healthcare utilization also indicated known/suspected allergies to be positively associated with future utilization. Conclusions: Inpatient interventions to address social stressors and allergic profiles may be warranted to reduce subsequent asthma morbidity.

Projections and Epidemiology of Revision Hip and Knee Arthroplasty in the United States to 2040-2060

National projections of future joint arthroplasties are useful in understanding the changing burden of surgery and related outcomes on the health system. The aim of this study is to update the literature by producing Medicare projections for revision total joint arthroplasty procedures from 2040 through 2060.The study uses 2000-2019 data from the CMS Medicare Part-B National Summary and combines procedure counts using CPT codes for revision total joint arthroplasty procedures. In 2019, revision total knee arthroplasty (rTKA) and revision total hip arthroplasty (rTHA) procedures totaled 53,217 and 30,541, respectively, forming a baseline from which we generated point forecasts between 2020 and 2060 and 95% forecast intervals (FI).On average, the model projects an annual growth rate of 1.77% for rTHAs and 4.67% for rTKAs. By 2040, rTHAs were projected to be 43,514 (95% FI = 37,429-50,589) and rTKAs were projected to be 115,147 (95% FI = 105,640-125,510). By 2060, rTHAs was projected to be 61,764 (95% FI = 49,927-76,408) and rTKAs were projected to be 286,740 (95% FI = 253,882-323,852).Based on 2019 total volume counts, the log-linear exponential model forecasts an increase in rTHA procedures of 42% by 2040 and 101% by 2060. Similarly, the estimated increase for rTKA is projected to be 149% by 2040 and 520% by 2060. An accurate projection of future revision procedure demands is important to understand future healthcare utilization and surgeon demand. This finding is only applicable to the Medicare population and demands further analysis for other population groups.

Neurocognitive and emotional factors predict pain-related healthcare utilization in children with sickle cell disease.

Youth with sickle cell disease (SCD) experience increased rates of neurocognitive and emotional difficulties. Cross-sectional studies suggest neurocognitive and emotional functioning are associated with health outcomes in SCD. We investigated whether neurocognitive and emotional factors predicted future pain-related healthcare utilization in children with SCD. Total 112 youth with SCD between ages 7 and 16years reported sociodemographics and completed measures of neurocognitive functioning and emotional well-being. The number of emergency department (ED) visits and hospitalizations for pain 1 and 3years after enrollment were determined by chart review. The mean age of participants was 10.61years (standard deviation=2.91), with most being female (n=65; 58%). Eighty-three (74%) participants had either HbSS or HbSβ0 thalassemia. Regression analyses showed that attention significantly predicted ED visits and hospitalizations for pain at 1 and 3years after enrollment (all p-values≤.017), such that poorer attention was associated with higher healthcare utilization. Lower emotional quality of life also predicted more ED visits for pain at 3years (b=-.009, p=.013) and hospitalizations for pain at 3years (b=-.008, p=.020). Neurocognitive and emotional factors are associated with subsequent healthcare use in youth with SCD. Poor attentional control might limit implementation of strategies to distract from pain or could make disease self-management behaviors more challenging. Results also highlight the potential impact of stress on pain onset, perception, and management. Clinicians should consider neurocognitive and emotional factors when developing strategies to optimize pain-related outcomes in SCD.

Quantitative CT Lung Imaging and Machine Learning Improves Prediction of Emergency Room Visits and Hospitalizations in COPD.

Predicting increased risk of future healthcare utilization in chronic obstructive pulmonary disease (COPD) patients is an important goal for improving patient management. Our objective was to determine the importance of computed tomography (CT) lung imaging measurements relative to other demographic and clinical measurements for predicting future health services use with machine learning in COPD. In this retrospective study, lung function measurements and chest CT images were acquired from Canadian Cohort of Obstructive Lung Disease study participants from 2010 to 2017 (https://clinicaltrials.gov, NCT00920348). Up to two follow-up visits (1.5- and 3-year follow-up) were performed and participants were asked for details related to healthcare utilization. Healthcare utilization was defined as any COPD hospitalization or emergency room visit due to respiratory problems in the 12 months prior to the follow-up visits. CT analysis was performed (VIDA Diagnostics Inc.); a total of 108 CT quantitative emphysema, airway and vascular measurements were investigated. A hybrid feature selection method with support vector machine classifier was used to predict healthcare utilization. Performance was determined using accuracy, F1-measure and area under the receiver operating characteristic curve (AUC) and Matthews's correlation coefficient (MC). Of the 527 COPD participants evaluated, 179 (35%) used healthcare services at follow-up. There were no significant differences between the participants with or without healthcare utilization at follow-up for age (p=0.50), sex (p=0.44), BMI (p=0.05) or pack-years (p=0.76). The accuracy for predicting subsequent healthcare utilization was 80% ± 3% (F1-measure=74%, AUC=0.80, MC=0.6) when all measurements were considered, 76% ± 6% (F1-measure=72%, AUC=0.77, MC=0.55) for CT measurements alone and 65% ± 5% (F1-measure=60%, AUC=0.67, MC=0.34) for demographic and lung function measurements alone. The combination of CT lung imaging and conventional measurements leads to greater prediction accuracy of subsequent health services use than conventional measurements alone, and may provide needed prognostic information for patients suffering from COPD.

Impact of Telehealth and Process Virtualization on Healthcare Utilization

Telehealth has emerged as a tool to improve patient access by virtualizing healthcare services, particularly during the COVID-19 pandemic. However, concerns have been raised that telehealth may actually increase healthcare spending by leading to new types of utilization. Our research provides empirical evidence that this concern is unfounded based on a state-wide study of patient visit-level data of telehealth use in 58 hospitals in Maryland from 2012 to 2021. On average, telehealth use can reduce future outpatient visits by 13.6% within 30 days after a telehealth visit, leading to a cost reduction of $239. The benefits of telehealth are most apparent for diseases with high potential for process virtualization, such as mental health, skin disorders, metabolic, and musculoskeletal diseases. Although telehealth has a substitution effect on future healthcare utilization, this effect is not observed among rural patients who use telehealth as a gateway to utilize more primary care and specialist services. Our findings suggest that policymakers should promote the use of telehealth in a value-based healthcare environment by providing monetary incentives to expand telehealth use among patients and providers, and expand the scope of telehealth services to include consultation with specialists especially among rural patients.

Projections and Epidemiology of Primary Hip and Knee Arthroplasty in Medicare Patients to 2040-2060.

The present study used data from the Centers for Medicare & Medicaid Services (CMS) Medicare/Medicaid Part B National Summary and combined procedure counts with use of Current Procedural Terminology (CPT) codes to identify whether the procedure was a primary total hip arthroplasty (THA) or total knee arthroplasty (TKA) procedure. In 2019, the annual volume of primary TKA was 480,958 and that of primary THA was 262,369. These values formed a baseline from which we generated point forecasts for 2020-2060 and 95% forecast intervals (FIs). Between 2000 and 2019, the estimated annual volume of THA increased by 177% and that of TKA increased by 156% on average. Regression analysis projected an annual growth rate of 5.2% for THA and 4.44% for TKA. Based on these yearly projected increases, an estimated increase of 28.84% and 24.28% is expected for each 5-year period after 2020 for THA and TKA, respectively. By 2040, the number of THAs is projected to be 719,364 (95% FI, 624,766 to 828,286) and the number of TKA is projected to be 1,222,988 (95% FI, 988,714 to 1,512,772). By 2060, the number of THAs is projected to be 1,982,099 (95% FI, 1,624,215 to 2,418,839) and the number of TKAs is projected to be 2,917,959 (95% FI, 2,160,951 to 3,940,156). In 2019, Medicare data showed that THA constituted approximately 35% of TJA procedures performed. Based on 2019 total volume counts, our model forecasts an increase in THA procedures of 176% by 2040 and 659% by 2060. The estimated increase for TKA is projected to be 139% by 2040 and 469% by 2060. An accurate projection of future primary TJA procedure demands is important in order to understand future health-care utilization and surgeon demand. This finding is only applicable to a Medicare population and demands further analysis to see if this extends to other population groups. Prognostic Level III. See Instructions for Authors for a complete description of levels of evidence.

Developing a Model to Predict High Health Care Utilization Among Patients in a New York City Safety Net System.

Health care facilities use predictive models to identify patients at risk of high future health care utilization who may benefit from tailored interventions. Previous predictive models that have focused solely on inpatient readmission risk, relied on commercial insurance claims data, or failed to incorporate social determinants of health may not be generalizable to safety net hospital populations. To address these limitations, we developed a payer-agnostic risk model for patients receiving care at the largest US safety net hospital system. We transformed electronic health record and administrative data from 833,969 adult patients who received care during July 2016-July 2017 into demographic, utilization, diagnosis, medication, and social determinant variables (including homelessness and incarceration history) to predict health care utilization during the following year.We selected the final model by developing and validating multiple classification and regression models predicting 10+ acute days, 5+ acute days, or continuous acute days. We compared a portfolio of performance metrics while prioritizing positive predictive value for patients whose predicted utilization was among the top 1% to maximize clinical utility. The final model predicted continuous number of acute days and included 17 variables. For the top 1% of high acute care utilizers, the model had a positive predictive value of 47.6% and sensitivity of 17.3%. Previous health care utilization and psychosocial factors were the strongest predictors of future high acute care utilization. We demonstrated a feasible approach to predictive high acute care utilization in a safety net hospital using electronic health record data while incorporating social risk factors.

Young Adult Substance Use and Healthcare Use Associated With Screening, Brief Intervention and Referral to Treatment in Pediatric Primary Care

PurposeScreening, brief intervention, and referral to treatment (SBIRT) may impact future comorbidity and healthcare utilization among adolescents screening positive for substance use or mood problems. MethodsIn a randomized trial sample, we compared an SBIRT group to usual care for substance use, mental health, medical diagnoses, and healthcare utilization over 7 years postscreening. ResultsIn logistic regression models adjusting for patient characteristics, the SBIRT group had lower odds of any substance (Odds Ratio[OR] = 0.80, 95% Confidence Interval [CI] = 0.66–.98), alcohol (OR = 0.69, 95% CI = 0.51–0.94), any drug (OR = 0.73, 95% CI = 0.54–0.98), marijuana (OR = 0.70, 95% CI = 0.50–0.98), and tobacco (OR = 0.83, 95% CI = 0.69–1.00) diagnoses, and lower odds of any inpatient hospitalizations (OR = 0.59, 95% CI = 0.41–0.85) compared with usual care. Negative binomial models examining number of visits among adolescents with at least one visit of that type found that those in the SBIRT group had fewer primary care (incidence rate ratio[iRR] = 0.90, p < .05) and psychiatry (iRR = 0.64, p < .01) and more addiction medicine (iRR = 1.52, p < .01) visits over 7 years compared with usual care. In posthoc analyses, we found that among Hispanic patients, those in the SBIRT group had lower odds of any substance, any drug and marijuana use disorder diagnoses compared with usual care, and among Black/African American patients, those in the SBIRT group had lower odds of alcohol use disorder diagnoses compared with usual care. DiscussionBeneficial effects of adolescent SBIRT on substance use and healthcare utilization may persist into young adulthood.

Potentially Inappropriate Prescribing in Hospitalized Older Adult High-Cost Health Care Users: A Pilot Study.

High-cost health care users use disproportionate amounts of health care resources relative to the typical patient. It is unclear to what extent poor-quality prescribing, including potentially inappropriate prescribing (PIP), may be contributing to their adverse outcomes and health utilization costs. To evaluate the prevalence of PIP and to explore its impact in older adult high-cost health care users. The charts of older adult high-cost health care users admitted to 2 academic hospitals in Ontario, Canada, in fiscal year 2015/16 were reviewed. Eligible patients were at least 66 years old with at least 5 emergency department visits and 3 hospital admissions in the previous year. A total of 243 patients met these criteria, of whom 100 were randomly selected for review. Cases of PIP were identified using explicit prescribing quality indicators, including the STOPP/START criteria. Types of PIP included potentially inappropriate medications (PIMs) and potential prescribing omissions (PPOs). Log-linear regression was used to characterize the relationship between PIP and future health care utilization. Medications were reconciled to determine the proportion of PIP addressed by the time of discharge. Eighty-nine of the 100 patients had at least 1 instance of PIP. In total, 276 PIMs and 54 PPOs were identified. Of the 271 instances of PIP identified on admission, only 38 (14%) were resolved by the time of hospital discharge. Each additional PPO was associated with a 1.43-fold increase in the rate of future emergency department visits (p < 0.001). The rate of PIP among older adult high-cost health care users was high. Despite frequent interactions with the health care system, many opportunities to improve the quality of prescribing for this vulnerable population were missed. Greater attention to medication optimization is needed.

Prediction of Future Health Care Utilization.

Psychosocial factors are known to have adverse health impacts, but are rarely measured; using natural language processing, we extracted factors that identified a higher risk segment of older adults with multimorbidity. We find these extracted features are highly predictive of future emergency department visits and hospitalizations, although only marginal prediction gains are seen compared to other models without these factors. Combining these extraction techniques with other measures of social determinants may help catalyze population health efforts to mitigate these health impacts.

Prediction of Future Health Care Utilization Through Note-extracted Psychosocial Factors.

Persons with multimorbidity (≥2 chronic conditions) face an increased risk of poor health outcomes, especially as they age. Psychosocial factors such as social isolation, chronic stress, housing insecurity, and financial insecurity have been shown to exacerbate these outcomes, but are not routinely assessed during the clinical encounter. Our objective was to extract these concepts from chart notes using natural language processing and predict their impact on health care utilization for patients with multimorbidity. A cohort study to predict the 1-year likelihood of hospitalizations and emergency department visits for patients 65+ with multimorbidity with and without psychosocial factors. Psychosocial factors were extracted from narrative notes; all other covariates were extracted from electronic health record data from a large academic medical center using validated algorithms and concept sets. Logistic regression was performed to predict the likelihood of hospitalization and emergency department visit in the next year. In all, 76,479 patients were eligible; the majority were White (89%), 54% were female, with mean age 73. Those with psychosocial factors were older, had higher baseline utilization, and more chronic illnesses. The 4 psychosocial factors all independently predicted future utilization (odds ratio=1.27-2.77, C -statistic=0.63). Accounting for demographics, specific conditions, and previous utilization, 3 of 4 of the extracted factors remained predictive (odds ratio=1.13-1.86) for future utilization. Compared with models with no psychosocial factors, they had improved discrimination. Individual predictions were mixed, with social isolation predicting depression and morbidity; stress predicting atherosclerotic cardiovascular disease onset; and housing insecurity predicting substance use disorder morbidity. Psychosocial factors are known to have adverse health impacts, but are rarely measured; using natural language processing, we extracted factors that identified a higher risk segment of older adults with multimorbidity. Combining these extraction techniques with other measures of social determinants may help catalyze population health efforts to address psychosocial factors to mitigate their health impacts.

Development and internal validation of prediction models for future hospital care utilization by patients with multimorbidity using electronic health record data.

To develop and internally validate prediction models for future hospital care utilization in patients with multiple chronic conditions. Retrospective cohort study. A teaching hospital in the Netherlands (542 beds). All adult patients (n = 18.180) who received care at the outpatient clinic in 2017 for two chronic diagnoses or more (including oncological diagnoses) and who returned for hospital care or outpatient clinical care in 2018. Development and validation using a stratified random split-sample (n = 12.120 for development, n = 6.060 for internal validation). ≥2 emergency department visits in 2018, ≥1 hospitalization in 2018 and ≥12 outpatient visits in 2018. Multivariable logistic regression with forward selection. Evaluation of the models' performance showed c-statistics of 0.70 (95% CI 0.69-0.72) for the hospitalization model, 0.72 (95% CI 0.70-0.74) for the ED visits model and 0.76 (95% 0.74-0.77) for the outpatient visits model. With regard to calibration, there was agreement between lower predicted and observed probability for all models, but the models overestimated the probability for patients with higher predicted probabilities. These models showed promising results for further development of prediction models for future healthcare utilization using data from local electronic health records. This could be the first step in developing automated alert systems in electronic health records for identifying patients with multimorbidity with higher risk for high healthcare utilization, who might benefit from a more integrated care approach.

The prevalence of long-term rehabilitation following motor-vehicle crashes in Saudi Arabia: a multicenter study

IntroductionIn Saudi Arabia, motor-vehicle crashes (MVC) are the leading cause of disability-adjusted life years (DALYs). There is limited information locally on the magnitude and need for rehabilitation following MVC. This study examined the prevalence of MVC patients requiring long-term rehabilitation and the epidemiology of associated injuries.MethodsA retrospective study was conducted at four hospitals of the National Guard Hospitals Affairs from January 2016 to March 2019. The study used data from an institutional trauma registry of all MVC admissions. Chi-square tests, bivariate and multivariate analyses were conducted to compare patients requiring long-term and short-term rehabilitation.ResultsThe study included 506 patients. The study population was relatively young, with an average age was 32.8 ± 15.5 years, and the majority were males. Over two-thirds (71.3%) of patients required long-term rehabilitation. Half the patients sustained multiple fractures, and 17.0% sustained traumatic brain injuries. Overall, 53.1 and 61.8% of patients required occupational and physiotherapy, respectively. Those admitted to the intensive care unit were four times more likely to need long-term rehabilitation.ConclusionsWe found a significant burden of long-term rehabilitation following MVC. Patients were relatively young, thus posing a significant burden on future healthcare utilization. Policymakers should use these findings to guide primary, secondary, and tertiary prevention to improve health outcomes.