A Comprehensive Analysis of the Impact of the COVID-19 Pandemic on Lung Cancer in Delaware.

Background: COVID-19 has greatly impacted the U.S. health system. What is not as well-understood is how this has altered specific aspects of lung cancer care. While cancer incidence and screening have been affected, it is not known whether pre-existing racial and socioeconomic disparities worsened or if treatment standards changed. The purpose of this study is to provide a comprehensive analysis of the impact of COVID-19 on lung cancer in the state of Delaware. Methods: Health care claims were analyzed from the Delaware Healthcare Claims Database for the years 2019-2020. Patients with a new lung cancer diagnosis and those who had undergone lung cancer screening were identified. Demographic and socioeconomic variables including gender, age, race, and insurance were studied. Patients were analyzed for type of treatment by CPT code. The intervention of interest in this study was the institution of restrictions at the end of March 2020. An interrupted time series analysis (ITSA) was utilized to evaluate baseline levels and overall trend changes. Results: The incidence of lung cancer diagnoses and lung cancer screenings decreased in the nine-month time period after the initiation of COVID-19 lockdowns. Demographic and socioeconomic variables including gender, race, income, and education level were not affected; however, statistical differences were seen in the most elderly subgroup. Treatment modalities including number of surgeries, chemotherapy, and radiation therapy did not change significantly. Conclusions: COVID-19 has had a significant impact on lung cancer care within the state of Delaware. Lung cancer incidence, screenings, and elderly patients were affected the most.

Neighborhood Environment and Poor Maternal Glycemic Control-Associated Complications of Gestational Diabetes Mellitus.

Introduction: Risk of complications due to gestational diabetes mellitus is increasing in the U.S., particularly among individuals from racial minorities. Research has focused largely on clinical interventions to prevent complications, rarely on individuals' residential environments. This retrospective cohort study aims to examine the association between individuals' neighborhoods and complications of gestational diabetes mellitus. Methods: Demographic and clinical data were extracted from electronic health records and linked to American Community Survey data from the U.S. Census Bureau for 2,047 individuals who had 2,164 deliveries in 2014-2018. Data were analyzed in 2021-2022 using Wilcoxon rank sum test and chi-square test for bivariate analyses and logistic regression for analysis of independent effects. All census tract-based variables used in the model were dichotomized at the median. Results: Bivariate analysis showed that the average percentage of adults earning <$35,000 was higher in neighborhoods where individuals with complications were living than in neighborhoods where individuals without complications were living (30.40%±12.05 vs 28.94%±11.71, p=0.0145). Individuals who lived in areas with ≥8.9% of residents aged >25 years with less than high school diploma had a higher likelihood of complications than those who lived in areas with <8.9% of such residents (33.43% vs 29.02%, p=0.0272). Individuals who lived in neighborhoods that had ≥1.8% of households receiving public assistance were more likely to have complications than those who lived in areas where <1.8% of households received public assistance (33.33% vs 28.97%, p=0.0287). Logistic regression revealed that the odds of deliveries with complications were 44% higher for individuals with obesity (OR=1.44; 95% CI=1.17, 1.77), 35% greater for individuals residing in neighborhoods with higher percentages of households living below the poverty level (OR=1.35; 95% CI=1.09, 1.66), and 28% lower for individuals from neighborhoods where a higher percentage of households had no vehicles available for transportation to work (OR=0.72; 95% CI=0.59, 0.89). Conclusions: Clinical interventions in concert with environmental changes could contribute to preventing maternal and neonatal complications of gestational diabetes mellitus.

Primary care outreach and decision counseling for lung cancer screening.

INTRODUCTION: Lung cancer screening rates are very low despite a level B recommendation from the United States Preventive Services Task Force since 2013 and clear evidence that lung cancer screening reduces mortality. The Center for Medicare and Medicaid Services requires shared decision-making (SDM) for lung cancer screening reimbursement. The objective of this study was to determine the effect of an SDM intervention on lung cancer screening in primary care. METHODS: The study design was a single-arm clinical trial design. The intervention included phone contact outside of a primary care visit and the use of the Decision Counseling Program ®, an online interactive decision aid focused on determining the factors which influence patients to screen or not screen, prioritizing those factors, and determining a decision preference score. The primary outcome was the completion of low-dose computed tomography scan (LDCT) 1 year after the SDM session compared in participants versus nonparticipants. RESULTS: From six practices, there were 1359 potentially eligible patients in electronic medical record data, and 336 were reached to assess eligibility criteria. A total of 80 patients consented to be in the study, 64 completed a decision counseling session and 16 did not complete a session. Among the 64 people who agreed to have decision counseling, 45% had LDCT, higher than typically seen in routine clinical practice. Although not a comparable group, among the 16 people who declined decision counseling, none had LDCT. CONCLUSIONS: Decision counseling is a promising intervention that might support SDM in the context of improving uptake of lung cancer screening in primary care. However, further, larger studies are needed.

Telemedicine and health disparities: Association between the area deprivation index and primary care telemedicine utilization during the COVID-19 pandemic.

Introduction: The rapid implementation of telemedicine during the COVID-19 pandemic may have exacerbated the existing health disparities. This study investigated the association between the area deprivation index (ADI), which serves as a measure of socioeconomic deprivation within a geographic area, and the utilization of telemedicine in primary care. Methods: The study data source was electronic health records. The study population consisted of patients with at least one primary care visit between March 2020 and December 2021. The primary outcome of interest was the visit modality (office, phone, and video). The exposure of interest was the ADI score grouped into quartiles (one to four, with one being the least deprived). The confounders included patient sociodemographic characteristics (e.g., age, gender, race, ethnicity, insurance coverage, marital status). We utilized generalized estimating equations to compare the utilization of telemedicine visits with office visits, as well as phone visits with video visits. Results: The study population included 41,583 patients with 127,165 office visits, 39,484 phone visits, and 20,268 video visits. Compared to patients in less disadvantaged neighborhoods (ADI quartile = one), patients in more disadvantaged neighborhoods (ADI = two, three, or four) had higher odds of using phone visits vs office visits, lower odds of using video visits vs office visits, and higher odds of using phone visits vs video visits. Conclusions: Patients who resided in socioeconomically disadvantaged neighborhoods mainly relied on phone consultations for telemedicine visits with their primary care provider. Patient-level interventions are essential for achieving equitable access to digital healthcare, particularly for low-income individuals.

Feasibility of a Novel COVID-19 Telehealth Care Management Program Among Individuals Receiving Treatment for Opioid Use Disorder: Analysis of a Pilot Program.

BACKGROUND: The emergence of COVID-19 exacerbated the existing epidemic of opioid use disorder (OUD) across the United States due to the disruption of in-person treatment and support services. Increased use of technology including telehealth and the development of new partnerships may facilitate coordinated treatment interventions that comprehensively address the health and well-being of individuals with OUD. OBJECTIVE: The analysis of this pilot program aimed to determine the feasibility of delivering a COVID-19 telehealth care management program using SMS text messages for patients receiving OUD treatment. METHODS: Eligible individuals were identified from a statewide opioid treatment program (OTP) network. Those who screened positive for COVID-19 symptoms were invited to connect to care management through a secure SMS text message that was compliant with Health Insurance Portability and Accountability Act standards. Care management monitoring for COVID-19 was provided for a period of up to 14 days. Monitoring services consisted of daily SMS text messages from the care manager inquiring about the participant's physical health in relation to COVID-19 symptoms by confirming their temperature, if the participant was feeling worse since the prior day, and if the participant was experiencing symptoms such as coughing or shortness of breath. If COVID-19 symptoms worsened during this observation period, the care manager was instructed to refer participants to the hospital for acute care services. The feasibility of the telehealth care management intervention was assessed by the rates of adoption in terms of program enrollment, engagement as measured by the number of SMS text message responses per participant, and retention in terms of the number of days participants remained in the program. RESULTS: Between January and April 2021, OTP staff members referred 21 patients with COVID-19 symptoms, and 18 (82%) agreed to be contacted by a care manager. Participants ranged in age from 27 to 65 years and primarily identified as female (n=12, 67%) and White (n=15, 83%). The majority of participants were Medicaid recipients (n=14, 78%). There were no statistically significant differences in the demographic characteristics between those enrolled and not enrolled in the program. A total of 12 (67%) patients were enrolled in the program, with 2 (11%) opting out of SMS text message communication and choosing instead to speak with a care manager verbally by telephone. The remaining 10 participants answered a median of 7 (IQR 4-10) SMS text messages and were enrolled in the program for a median of 9 (IQR 7.5-12) days. No participants were referred for acute care services or hospitalized during program enrollment. CONCLUSIONS: These results demonstrate the feasibility of a novel telehealth intervention to monitor COVID-19 symptoms among OTP patients in treatment for OUD. Further research is needed to determine the applicability of this intervention to monitor patients with comorbid chronic conditions in addition to the acceptability among patients and providers using the SMS text messaging modality.

Implementation of a clinical pathway to screen and treat medical inpatients for opioid withdrawal.

Background: Opioid-related inpatient hospital stays are increasing at alarming rates. Unidentified and poorly treated opioid withdrawal may be associated with inpatients leaving against medical advice and increased health care utilization. To address these concerns, we developed and implemented a clinical pathway to screen and treat medical service inpatients for opioid withdrawal. Methods: The pathway process included a two-item universal screening instrument to identify opioid withdrawal risk (Opioid Withdrawal Risk Assessment [OWRA]), use of the validated Clinical Opiate Withdrawal Scale (COWS) to monitor opioid withdrawal symptoms and severity, and a 72-h buprenorphine/naloxone-based treatment protocol. Implementation outcomes including adoption, fidelity, and sustainability of this new pathway model were measured. To assess if there were changes in nursing staff acceptability, appropriateness, and adoption of the new pathway process, a cross-sectional survey was administered to pilot four hospital medical units before and after pathway implementation. Results: Between 2016 and 2018, 72.4% (77,483/107,071) of admitted patients received the OWRA screening tool. Of those, 3.0% (2,347/77,483) were identified at risk for opioid withdrawal. Of those 2,347 patients, 2,178 (92.8%) were assessed with the COWS and 29.6% (645/2,178) were found to be in active withdrawal. A total of 49.5% (319/645) patients were treated with buprenorphine/naloxone. Fifty-seven percent (83/145) of nurses completed both the pre- and post-pathway implementation surveys. Analysis of the pre/post survey data revealed that nurse respondents were more confident in their ability to determine which patients were at risk for withdrawal (p = .01) and identify patients currently experiencing withdrawal (p < .01). However, they cited difficulty working with the patient population and coordinating care with physicians. Conclusions: Our study demonstrates a process for successfully implementing and sustaining a clinical pathway to screen and treat medical service inpatients for opioid withdrawal. Standardizing care delivery for patients in opioid withdrawal can also improve nursing confidence when working with this complex population.

Escalation of Opioid Withdrawal Frequency and Subsequent AMA Rates in Hospitalized Patients From 2017 to 2020.

OBJECTIVE: To measure trends for the emergence of opioid withdrawal (OW) and leaving against medical advice (AMA) among hospitalized patients. METHOD: Retrospective time-series of hospitalized patients with OW, defined by a Clinical Opioid Withdrawal score >8, using electronic health record data at a tertiary health system and of patients with a discharge status of AMA from January 1, 2017 to December 31, 2020. RESULTS: The average number of monthly hospitalizations with OW showed a year-to-year increment of 15% in 2018, 21% in 2019, and 34% from 2019 to 2020, whereas the total monthly hospitalizations remained stable. The segmented regression analysis showed that the upward trend in hospitalizations with OW became significant after January 2019 (slope: 1.14, 95% confidence interval [CI]: 0.70, 1.57). After August 2019, Fentanyl was added to the hospital urine drug testing panel and was identified in most OW patients. The monthly proportion of patients who left AMA was significantly higher among the OW patients than among all other admitted patients. There was a significant increase of 0.39 (95% CI: 0.29-0.49, P < 0.001) per month in %AMA among patients with OW. The estimated difference in %AMA among OW patients versus all other patients was 7.25 (95% CI: 5.12-9.38) in January 2017, and 16.92 (95% CI: 14.60-19.24) in December 2020. CONCLUSIONS: The number of hospitalized patients either presenting with or developing OW increased between 2017 and 2020 with a significant rise occurring after January 2019. The percentage of patients who left AMA among those who developed OW steadily worsened during the entire study period.

Intubation Timing in COVID-19 Based on ROX Index and Association With Patient Outcomes.

BACKGROUND: Timing of intubation in COVID-19 is controversial. We sought to determine the association of the ROX (Respiratory rate-OXygenation) index defined as [Formula: see text] divided by [Formula: see text] divided by breathing frequency at the time of intubation with clinical outcomes. METHODS: We conducted a retrospective cohort study of patients with COVID-19 who were intubated by using a database composed of electronic health record data from patients with COVID-19 from 62 institutions. Multivariable logistic regression was used to evaluate the impact of ROX index score on mortality. We analyzed the ROX index as a continuous variable as well as a categorical variable by using cutoffs previously described as predicting success with high-flow nasal cannula. RESULTS: Of 1,087 subjects in the analysis group, the median age was 64 years, and more than half had diabetes; 55.2% died, 1.8% were discharged to hospice, 7.8% were discharged to home, 27.3% were discharged to another institution, and 7.8% had another disposition. Increasing age and a longer time from admission to intubation were associated with mortality. After adjusting for sex, race, age, comorbidities, and days from admission to intubation, an increasing ROX index score at the time of intubation was associated with a lower risk of death. In a logistic regression model, each increase in the ROX index score by 1 at the time of intubation was associated with an 8% reduction in odds of mortality (odds ratio 0.92, 95% CI 0.88-0.95). We also found an odds ratio for death of 0.62 (95% CI 0.47-0.81) for subjects with an ROX index score ≥ 4.88 at the time of intubation. CONCLUSIONS: Among a cohort of subjects with COVID-19 who were ultimately intubated, a higher ROX index at the time of intubation was positively associated with survival.

Prediction model for COVID-19 patient visits in the ambulatory setting.

Objective: Healthcare systems globally were shocked by coronavirus disease 2019 (COVID-19). Policies put in place to curb the tide of the pandemic resulted in a decrease of patient volumes throughout the ambulatory system. The future implications of COVID-19 in healthcare are still unknown, specifically the continued impact on the ambulatory landscape. The primary objective of this study is to accurately forecast the number of COVID-19 and non-COVID-19 weekly visits in primary care practices. Materials and Methods: This retrospective study was conducted in a single health system in Delaware. All patients' records were abstracted from our electronic health records system (EHR) from January 1, 2019 to July 25, 2020. Patient demographics and comorbidities were compared using t-tests, Chi square, and Mann Whitney U analyses as appropriate. ARIMA time series models were developed to provide an 8-week future forecast for two ambulatory practices (AmbP) and compare it to a naïve moving average approach. Results: Among the 271,530 patients considered during this study period, 4,195 patients (1.5%) were identified as COVID-19 patients. The best fitting ARIMA models for the two AmbP are as follows: AmbP1 COVID-19+ ARIMAX(4,0,1), AmbP1 nonCOVID-19 ARIMA(2,0,1), AmbP2 COVID-19+ ARIMAX(1,1,1), and AmbP2 nonCOVID-19 ARIMA(1,0,0). Discussion and Conclusion: Accurately predicting future patient volumes in the ambulatory setting is essential for resource planning and developing safety guidelines. Our findings show that a time series model that accounts for the number of positive COVID-19 patients delivers better performance than a moving average approach for predicting weekly ambulatory patient volumes in a short-term period.

Cause for Concern: Behavioral and Psychological Symptoms of Dementia in Delaware's Long-Term Care Facilities.

Introduction: Much of the suffering and expense associated with treatment of persons with dementia (Major Neurocognitive Disorder) arises from associated noncognitive behavioral and psychological symptoms of dementia (BPSD). Although a consensus on the prevalence of BPSD is lacking, evidence suggests that most people with dementia will manifest one or more of these symptoms during the disorder's progression. BPSD raise the cost of care by leading to more frequent emergency room visits, more and longer hospitalizations, and earlier admission to long-term care facilities (LTCF). Treatment of BPSD presents a stressful challenge in LTCFs. We sought to investigate the care burden of BPSD in Delaware's LTCFs and to gather data that can inform management approaches. Methods: Using REDCap, we created an anonymous cross-sectional survey designed for completion by LTCF administrators. The Delaware Health Care Facilities Association (DHCFA) and Delaware's Division of Services for Aging and Adults with Physical Disabilities (DSAAPD) encouraged participation. A link to the survey was emailed to the administrators of 81 facilities in Delaware. The resulting data were evaluated using descriptive statistics. Results: Forty-four of the 81 facilities surveyed opened the survey link. Thirty-eight facilities answered at least some of the questions, and 19 surveys were fully completed. The reported average prevalence of BPSD among Delaware LTCF residents with dementia was 49.3% (SD 28.9). The five most frequently reported BPSD symptoms were anxiety, agitation, wandering, dysphoria/depression, and appetite/eating abnormalities. All facilities reported employing a spectrum of pharmacologic and non-pharmacologic management strategies. Twenty-two of 24 respondents (91.7%) reported that behavioral health consultation was available at their facilities and 18 of 20 respondents (90.0%) indicated that they provided training on how to manage residents with BPSD. Conclusion: BPSD are a pervasive concern among Delaware's LTCFs. LTCFs may benefit from the development of training programs and dissemination of treatment guidelines incorporating evidence-based interventions and their implementation in managing BPSD to improve care, decrease stress on residents and caregivers, and reduce some avoidable health care costs.

A Social Network Analysis Approach for Contact Tracing in the Hospital Setting.

Since the beginning of the COVID-19 pandemic, the State of Delaware has implemented various strategies including a stay-at-home order, mask-wearing requirements in public places, and community-based testing to control the spread of the disease. Health systems across the U.S. have taken actions including symptom monitoring and screening for visitors and healthcare workers, providing personal protection equipment (PPE), and contact tracing of confirmed infected individuals to provide maximum possible protection for healthcare workers. Despite such efforts, there remains a significant risk of intra-hospital transmission of COVID-19. Healthcare workers who contact patients with COVID-19 or were exposed to the disease in the community may transmit the infection to coworkers in the inpatient setting. In addition to universal and case-based precautions to prevent exposure and disease transmission, contact tracing is essential to minimizing the impact of outbreaks among healthcare workers and the community. A rapid increase in cases can quickly diminish hospital infection control and prevention program capacity to perform high-quality contact tracing. This article will describe an approach using the application of social network analysis (SNA) and Electronic Medical Records (EMR) to enhance the current efforts in COVID-19 contact tracings.

Telephone-Based Shared Decision-making for Lung Cancer Screening in Primary Care.

The national rate of  lung cancer screening, approximately 3-5%, is too low and strategies which include shared decision-making and increase screening are needed. A feasibility study in one large primary care practice of telephone-based delivery of decision support via an online tool, the Decision Counseling Program© (DCP) was administered to patients eligible for lung cancer screening according to USPSTF screening guidelines. We collected data on demographics, decisional conflict, and conducted chart audits to ascertain screening. From electronic medical record data, we identified 829 age-eligible current or former smokers. Of the 297 individuals reached, 54 were eligible and 28 were recruited to the study and 20 underwent the DCP© intervention. Participants in the intervention were more likely to complete low-dose CT scans at 90 days. Current smokers were less likely to complete the DCP. Women were less likely to complete LDCT. This non-persuasive, high-quality shared decision-making intervention significantly increased lung cancer screening and was feasible in real-world clinical care. This intervention offers a promising model whereby patients can be supported in a decision, based on their values and beliefs while also supporting gains in lung cancer screening.

Evaluation of a Systems-Based Tobacco Cessation Program Using Bedside Volunteers.

INTRODUCTION: Hospitalization and post-discharge provide an opportune time for tobacco cessation. This study tested the feasibility, uptake, and cessation outcomes of a hospital-based tobacco cessation program, delivered by volunteers to the bedside with post-discharge referral to Quitline services. Patient characteristics associated with Quitline uptake and cessation were assessed. METHODS: Between February and November 2016, trained hospital volunteers approached inpatient tobacco users on six pilot units. Volunteers shared a cessation brochure and used the ASK-ADVISE-CONNECT model to connect ready to quit patients to the Delaware Quitline via fax-referral. Volunteers administered a follow-up survey to all admitted tobacco users via telephone or email at 3-months post-discharge. RESULTS: Of the 743 admitted tobacco users, 531 (72%) were visited by a volunteer, and 97% (531/547) of those approached, accepted the visit. Over one-third (201/531; 38%) were ready to quit and fax-referred to the Quitline, and 36% of those referred accepted Quitline services. At 3 months post-discharge, 37% (135/368) reported not using tobacco in the last 30 days; intent-to-treat cessation rate was 18% (135/743). In a multivariable regression model of Quitline fax-referral completion, receiving nicotine replacement therapy (NRT) during hospitalization was the strongest predictor (odds ratios [OR] = 1.97; 95% confidence interval [CI] = 1.34 to 2.90). In a model of 3-month cessation, receiving Quitline services (OR = 3.21, 95% CI = 1.35 to 7.68) and having coronary artery disease (OR = 2.28; 95% CI = 1.11 to 4.68) were associated with tobacco cessation, but a volunteer visit was not. CONCLUSIONS: An "opt-out" tobacco cessation service using trained volunteers is feasible for connecting patients to Quitline services. IMPLICATIONS: This study demonstrates the feasibility of a systems-based approach to link inpatients to evidence-based treatment for tobacco use. This model used trained bedside volunteers to connect inpatients to a state-funded Quitline after discharge that offers free cessation treatment of telephone coaching and cessation medications. Receiving NRT during hospitalization positively impacted Quitline referral, and engagement with Quitline resources was critical to tobacco abstinence post-discharge. Future work is needed to evaluate the cost-effectiveness and sustainability of this volunteer model.

Development and Validation of a Pragmatic Electronic Phenotype for CKD.

BACKGROUND AND OBJECTIVES: Poor identification of individuals with CKD is a major barrier to research and appropriate clinical management of the disease. We aimed to develop and validate a pragmatic electronic (e-) phenotype to identify patients likely to have CKD. DESIGN, SETTING, PARTICIPANTS, & MEASUREMENTS: The e-phenotype was developed by an expert working group and implemented among adults receiving in- or outpatient care at five healthcare organizations. To determine urine albumin (UA) dipstick cutoffs for CKD to enable use in the e-phenotype when lacking urine albumin-to-creatinine ratio (UACR), we compared same day UACR and UA results at four sites. A sample of patients, spanning no CKD to ESKD, was randomly selected at four sites for validation via blinded chart review. RESULTS: The CKD e-phenotype was defined as most recent eGFR <60 ml/min per 1.73 m2 with at least one value <60 ml/min per 1.73 m2 >90 days prior and/or a UACR of ≥30 mg/g in the most recent test with at least one positive value >90 days prior. Dialysis and transplant were identified using diagnosis codes. In absence of UACR, a sensitive CKD definition would consider negative UA results as normal to mildly increased (KDIGO A1), trace to 1+ as moderately increased (KDIGO A2), and ≥2+ as severely increased (KDIGO A3). Sensitivity, specificity, and diagnostic accuracy of the CKD e-phenotype were 99%, 99%, and 98%, respectively. For dialysis sensitivity was 94% and specificity was 89%. For transplant, sensitivity was 97% and specificity was 91%. CONCLUSIONS: The CKD e-phenotype provides a pragmatic and accurate method for EHR-based identification of patients likely to have CKD.

Energy balance in obese, mechanically ventilated intensive care unit patients.

OBJECTIVES: The aims of this study were, first, to compare the predicted (calculated) energy requirements based on standard equations with target energy requirement based on indirect calorimetry (IC) in critically ill, obese mechanically ventilated patients; and second, to compare actual energy intake to target energy requirements. METHODS: We conducted a prospective cohort study of mechanically ventilated critically ill patients with body mass index ≥30.0 kg/m2 for whom enteral feeding was planned. Clinical and demographic data were prospectively collected. Resting energy expenditure was measured by open-circuit IC. American Society of Parenteral and Enteral Nutrition (APSPEN)/Society of Critical Care Medicine (SCCM) 2016 equations were used to determine predicted (calculated) energy requirements. Target energy requirements were set at 65% to 70% of measured resting energy expenditure as recommended by ASPEN/SCCM. Nitrogen balance was determined via simultaneous measurement of 24-h urinary nitrogen concentration and protein intake. RESULTS: Twenty-five patients (mean age: 64.5 ± 11.8 y, mean body mass index: 35.2 ± 3.6 kg/m2) underwent IC. The mean predicted energy requirement was 1227 kcal/d compared with mean measured target energy requirement of 1691 kcal/d. Predicted (calculated) energy requirements derived from ASPEN/SCCM equations were less than the target energy requirements in most cases. Actual energy intake from enteral nutrition met 57% of target energy requirements. Protein intake met 25% of target protein requirement and the mean nitrogen balance was -2.3 ± 5.1 g/d. CONCLUSIONS: Predictive equations underestimated target energy needs in this population. Further, we found that feeding to goal was often delayed resulting in failure to meet both protein and energy intake goals.

Chronic Kidney Disease stratification using office visit records: Handling data imbalance via hierarchical meta-classification.

BACKGROUND: Chronic Kidney Disease (CKD) is one of several conditions that affect a growing percentage of the US population; the disease is accompanied by multiple co-morbidities, and is hard to diagnose in-and-of itself. In its advanced forms it carries severe outcomes and can lead to death. It is thus important to detect the disease as early as possible, which can help devise effective intervention and treatment plan. Here we investigate ways to utilize information available in electronic health records (EHRs) from regular office visits of more than 13,000 patients, in order to distinguish among several stages of the disease. While clinical data stored in EHRs provide valuable information for risk-stratification, one of the major challenges in using them arises from data imbalance. That is, records associated with a more severe condition are typically under-represented compared to those associated with a milder manifestation of the disease. To address imbalance, we propose and develop a sampling-based ensemble approach, hierarchical meta-classification, aiming to stratify CKD patients into severity stages, using simple quantitative non-text features gathered from standard office visit records. METHODS: The proposed hierarchical meta-classification method frames the multiclass classification task as a hierarchy of two subtasks. The first is binary classification, separating records associated with the majority class from those associated with all minority classes combined, using meta-classification. The second subtask separates the records assigned to the combined minority classes into the individual constituent classes. RESULTS: The proposed method identifies a significant proportion of patients suffering from the more advanced stages of the condition, while also correctly identifying most of the less severe cases, maintaining high sensitivity, specificity and F-measure (≥ 93%). Our results show that the high level of performance attained by our method is preserved even when the size of the training set is significantly reduced, demonstrating the stability and generalizability of our approach. CONCLUSION: We present a new approach to perform classification while addressing data imbalance, which is inherent in the biomedical domain. Our model effectively identifies severity stages of CKD patients, using information readily available in office visit records within the realistic context of high data imbalance.

Co-occurrence of medical conditions: Exposing patterns through probabilistic topic modeling of snomed codes.

Patients associated with multiple co-occurring health conditions often face aggravated complications and less favorable outcomes. Co-occurring conditions are especially prevalent among individuals suffering from kidney disease, an increasingly widespread condition affecting 13% of the general population in the US. This study aims to identify and characterize patterns of co-occurring medical conditions in patients employing a probabilistic framework. Specifically, we apply topic modeling in a non-traditional way to find associations across SNOMED-CT codes assigned and recorded in the EHRs of >13,000 patients diagnosed with kidney disease. Unlike most prior work on topic modeling, we apply the method to codes rather than to natural language. Moreover, we quantitatively evaluate the topics, assessing their tightness and distinctiveness, and also assess the medical validity of our results. Our experiments show that each topic is succinctly characterized by a few highly probable and unique disease codes, indicating that the topics are tight. Furthermore, inter-topic distance between each pair of topics is typically high, illustrating distinctiveness. Last, most coded conditions grouped together within a topic, are indeed reported to co-occur in the medical literature. Notably, our results uncover a few indirect associations among conditions that have hitherto not been reported as correlated in the medical literature.

Association Between Enteral Feeding, Weight Status, and Mortality in a Medical Intensive Care Unit.

BACKGROUND: Clinical practice guidelines recommend enteral nutrition for most patients receiving mechanical ventilation. However, recently published evidence on the effect of enteral nutrition on mortality, particularly for patients who are well nourished, is conflicting. OBJECTIVES: To examine the association between enteral feeding and hospital mortality in critically ill patients receiving mechanical ventilation and to determine if body mass index mediates this relationship. METHODS: A retrospective cohort study of patients receiving mechanical ventilation admitted to a medical intensive care unit in 2013. Demographic and clinical variables were collected. Cox proportional hazards regression was used to examine the relationship between an enteral feeding order and hospital mortality and to determine if the relationship was mediated by body mass index. RESULTS: Of 777 patients who had 811 hospitalizations requiring mechanical ventilation, 182 (23.4%) died in the hospital. A total of 478 patients (61.5%) received an order for enteral tube feeding, which was associated with a lower risk of death (hazard ratio, 0.41; 95% CI, 0.29-0.59). Body mass index did not mediate the relationship between mortality and receipt of an order for enteral feeding. Median stay in the unit was 3.6 days. Most deaths (72.0%) occurred more than 48 hours after admission. CONCLUSION: The finding of a positive association between an order for enteral feeding and survival supports enteral feeding of patients in medical intensive care units. Furthermore, the beneficial effect of enteral feeding appears to apply to patients regardless of body mass index.

Characteristics and Outcomes of Adult Inpatients With Malnutrition.

BACKGROUND: The diagnosis of malnutrition remains controversial. Furthermore, it is unknown if physician diagnosis of malnutrition impacts outcomes. We sought to compare outcomes of patients with physician diagnosed malnutrition to patients recognized as malnourished by registered dietitians (RDs), but not physicians, and to describe the impact of each of 6 criteria on the diagnosis of malnutrition. METHODS: We conducted a retrospective cohort study of adult patients identified as meeting criteria for malnutrition. Pediatric, psychiatric, maternity, and rehabilitation patients were excluded. Patient demographics, clinical data, malnutrition type and criteria, nutrition interventions, and outcomes were abstracted from the electronic medical record. RESULTS: RDs identified malnutrition for 291 admissions during our study period. This represents 4.1% of hospital discharges. Physicians only diagnosed malnutrition on 93 (32%) of these cases. Physicians diagnosed malnutrition in 43% of patients with a body mass index <18.5 but only 26% of patients with body mass index higher than 18.5. Patients with a physician diagnosis had a longer length of stay (mean 14.9 days vs 7.1 days) and were more likely to receive parenteral nutrition (PN) (20.4% vs 4.6%). Of the patients, 62% had malnutrition due to chronic illness. Of the 6 criteria used to identify malnourished patients, weight loss and reduced energy intake were the most common. CONCLUSIONS: Malnutrition is underrecognized by physicians. However, further research is needed to determine if physician recognition and treatment of malnutrition can improve outcomes. The most important criteria for identifying malnourished patients in our cohort were weight loss and reduced energy intake.