University Of Iowa Hospitals And Clinics Research Articles

Anxiety and depression symptoms among patients with long COVID: a retrospective cohort study.

Patients suffering from post-acute sequelae of COVID-19 (PASC) have a higher prevalence of anxiety and depression than the general population. The long-term trajectory of these sequelae is still unfolding. To assess the burden of anxiety and depression among patients presenting to the University of Iowa Hospitals and Clinics (UIHC) post-COVID-19 clinic, we analyzed how patient factors influenced Generalized Anxiety Disorder-7 (GAD-7) and Patient Health Questionnaire-9 (PHQ-9) scores. In this retrospective cohort study, the GAD-7 and PHQ-9 questionnaire scores of patients presenting to the UIHC post-COVID clinic between March 2021-February 2022 (N = 455) were compared to the scores of a sample of patients presenting to the general internal medicine (GIM) clinic during the same period (N = 94). Our analysis showed that patients with an absent history of depression on their electronic medical record (EMR) problem list scored significantly higher on the GAD-7 (mean difference -1.62, 95% CI -3.12 to -0.12, p = 0.034) and PHQ-9 (mean difference -4.45, 95% CI -5.53 to -3.37, p < 0.001) questionnaires compared to their similar counterparts in the GIM clinic. On the other hand, patients with an absent history of anxiety on their EMR problem list scored significantly higher on the GAD-7 (mean difference -2.90, 95% CI -4.0 to -1.80, p < 0.001) but not on the PHQ-9 questionnaire (p = 0.196). Overall, patients with PASC may have experienced a heavier burden of newly manifest anxiety and depression symptoms compared to patients seen in the GIM clinic. This suggests that the mental health impacts of PASC may be more pronounced in patients with no prior history of anxiety or depression.

1300. Comparison of the Clinical Characteristics and Outcomes of Anaerobic Prosthetic Joint Infections (PJI) and Aerobic PJIs : A Retrospective Review

Abstract Background PJI is an uncommon but serious complication of joint arthroplasties that can result in significant morbidity and cost. There is a paucity of studies in the literature on PJIs caused by anaerobic organisms either as the sole pathogen or part of a polymicrobial infection. The purpose of this study was to compare the clinical presentation and outcomes of PJI caused by anaerobes as opposed to aerobes. Methods This was a retrospective review of 306 patients who met Musculoskeletal Infection Society (MSIS) criteria for PJI from 2014 to 2020 at the University of Iowa Hospitals and Clinics (UIHC). 38 patients with anaerobic PJI were compared to 268 patients with aerobic PJI. Statistical analyses were performed using Pearson’s Chi square, Fisher exact test, t-test, and binary logistic regression. Results PJI with anaerobes represented 12.4% of all PJI cases in our institution, which is higher than the 3-6% reported in literature. Patients with anaerobic PJI had longer duration of symptoms (19.4 weeks vs 10.9 weeks, p = 0.012), were more likely to have a sinus tract on exam (23.7% vs 5.9%, p &amp;lt; 0.001), less fever on presentation ( 13.2% vs 30.2%, p = 0.029), more radiographic abnormalities (50.0% vs 31.1%, p&amp;lt; 0.001), lower erythrocyte sedimentation rate (49.0mm/hr vs 67.7 mm/hr, p = 0.003), lower C-reactive protein (6.6 mg/dL vs 12.1mg/dL, p = 0.004), and more likely to be polymicrobial (42.1%vs 8.0%, p &amp;lt; 0.001) compared to aerobic PJI. There was a higher proportion of anaerobes in shoulder PJIs compared to other joints (39.5% vs 5.5%, p&amp;lt; 0.001). Two-stage exchange was the most common surgical treatment in patients with anaerobic PJI (39.4%) whereas debridement and implant retention was the most common (44.5%) surgical treatment done in patients with aerobic PJI. No significant differences were seen between the two groups in terms of recurrences. Independent factors associated with anaerobic infection were shoulder prosthetic joint (AOR 10.5, CI 3.4 - 32.8) and presence of sinus tract (AOR 10.1, CI 3.1 - 32.6). Conclusion Anaerobic cultures should be performed in patients with PJI, and they should especially be considered on the differential in patients with shoulder PJI, longstanding symptoms, sinus tract on examination and abnormalities noted on radiographs. Disclosures All Authors: No reported disclosures

Improved Visual Outcomes after Severe Open-Globe Injuries Associated with Perioperative Vitreoretinal Evaluation

To investigate the impact of perioperative evaluation by a vitreoretinal surgeon on outcomes after severe open-globe injury (OGI). Retrospective, comparative study. Open-globe injury cohorts from 2 academic United States ophthalmology departments with disparate OGI management protocols and vitreoretinal referral patterns. Patients with severe OGI (presenting visual acuity [VA] of counting fingers or worse) from the University of Iowa Hospitals and Clinics (UIHC) were compared with patients with severe OGI from the Bascom Palmer Eye Institute (BPEI). At UIHC, almost all cases of OGI were repaired by anterior segment surgeons, with postoperative vitreoretinal referral at the surgeon's discretion. In contrast, at BPEI, all OGIs were both repaired and managed postoperatively by a vitreoretinal surgeon. Rate of vitreoretinal surgeon evaluation, rate of pars plana vitrectomy (PPV) (either primary or secondary), and VA at the last follow-up. Overall, 74 subjects from UIHC and 72 subjects from BPEI met the inclusion criteria. There were no differences in preoperative VA or rates of vitreoretinal pathology. The rate of vitreoretinal surgeon evaluation was 100% at BPEI and 65% at UIHC (P < 0.001), and the rate of PPV was 71% at BPEI and 40% at UIHC (P < 0.001). Median VA at the last follow-up in the BPEI cohort was 1.35 logarithm of the minimum angle of resolution (logMAR) (interquartile range [IQR], 0.53-2.30, corresponding to a Snellen VA of ∼ 20/500) compared with a median VA of 2.70 logMAR (IQR, 0.93-2.92, corresponding to a Snellen VA of light perception) in the UIHC cohort (P= 0.031). Overall, 68% of patients in the BPEI cohort had an improvement in VA from presentation to last follow-up versus 43% in the UIHC cohort (P= 0.004). Automatic perioperative evaluation by a vitreoretinal surgeon was associated with a higher rate of PPV and improved visual outcomes. Prompt evaluation by a vitreoretinal surgeon, either preoperatively or in the early postoperative period, should be considered, when logistically feasible, in severe OGIs because PPV is frequently indicated and may result in significant visual improvement. Proprietary or commercial disclosure may be found after the references.

Prevalence of maternal complications and neonatal outcomes at a Midwest academic health center

Introduction: The University of Iowa Hospitals and Clinics (UIHC) serves as the main referral center for Iowa and has seen increased delivery volumes following many hospital closures across the state. Maternal characteristics are also shifting as pre-pregnancy obesity of reproductive-age women is increasing nationally and in Iowa; average maternal age at first delivery has also increased over time. The aim of the current study was to characterize the outcomes of the delivering population at UIHC over a one-year period to better understand the population we serve, as well as compare our population to state and national delivering populations. Methods: This study is a secondary analysis of a pre-existing prospectively gathered database related to COVID-19 in pregnancy. The current study included all pregnant patients who delivered or underwent procedures for a spontaneous or induced second-trimester abortion on the labor and delivery unit at UIHC between May 1, 2020, and April 30, 2021. Demographic and clinical data including maternal characteristics, delivery information, maternal complications, and neonatal outcomes were obtained from the electronic medical record and double entered in a Research Electronic Data Capture (REDCap) database. Continuous and categorical variables were summarized using medians and ranges. Results: The median maternal age among the 2497 delivering patients was 30 (IQR 26, 33). The median pre-delivery body mass index (BMI) was 32 (IQR 28, 37), which included 439 patients (17.6%) with BMI &amp;gt;40 at time of delivery. 1769 (70.8%) patients listed White as their self-defined race and 2275 (91.0%) listed English as their preferred language. 23% had hypertensive disease of pregnancy. The patients delivered a total of 2603 neonates, with 907 (34.8%) born by cesarean section. 18.7% of deliveries were preterm and 16.6% of neonates had low birth weight. Discussion: The delivering population at UIHC has a high rate of obesity and preeclampsia, both of which place patients at higher risk for other obstetric complications. The cesarean section rate was significantly higher than the national rate of 31.8% (p=0.0017). Understanding common complications in our changing population will better equip providers to care for these patients; providing these data also creates a resource for other researchers.

Gaussian process regression and classification using International Classification of Disease codes as covariates

AbstractIn electronic health records (EHRs) data analysis, nonparametric regression and classification using International Classification of Disease (ICD) codes as covariates remain understudied. Automated methods have been developed over the years for predicting biomedical responses using EHRs, but relatively less attention has been paid to developing patient similarity measures that use ICD codes and chronic conditions, where a chronic condition is defined as a set of ICD codes. We address this problem by first developing a string kernel function for measuring the similarity between a pair of primary chronic conditions, represented as subsets of ICD codes. Second, we extend this similarity measure to a family of covariance functions on subsets of chronic conditions. This family is used in developing Gaussian process (GP) priors for Bayesian nonparametric regression and classification using diagnoses and other demographic information as covariates. Markov chain Monte Carlo (MCMC) algorithms are used for posterior inference and predictions. The proposed methods are tuning free, so they are ideal for automated prediction of biomedical responses depending on chronic conditions. We evaluate the practical performance of our method on EHR data collected from 1660 patients at the University of Iowa Hospitals and Clinics (UIHC) with six different primary cancer sites. Our method provides better sensitivity and specificity than its competitors in classifying different primary cancer sites and estimates the marginal associations between chronic conditions and primary cancer sites.

Differences in Outcomes in Obese (≥30), Morbidly Obese (≥40), and Super Morbidly Obese (≥50) Pregnancies

BackgroundObesity in pregnancy is an increasing problem in the United States. Obesity in pregnancy can result in dangerous complications such as premature birth, cesarean delivery, gestational diabetes, and preeclampsia. Little is currently known about outcomes in the morbidly (body mass index (BMI) ≥40 kg/m2) and super morbidly obese (BMI ≥50 kg/m2) in pregnant patients. Consequently, it is difficult to adequately counsel patients and plan for delivery care. Most current data groups outcomes for obese patients in general, however, some aspects such as reductions in hormone concentrations in pregnancy vary markedly with increasing weight. Obesity can alter hormone concentrations in pregnancy. For example, significantly lower beta human chorionic gonadotropin concentrations and progesterone are observed with increasing weight. Thus, we hypothesized that there are differences outcomes between obese, morbidly obese, and super morbidly obese pregnancies. The purpose of this study (IRB #202004178) is to analyze pregnancy outcomes of women with body mass index (BMI) ≥40 kg/m2 in comparison to obese women with a BMI of 30‐35 kg/m2 with a more rigorous BMI classification system than has been previously used. Our hypothesis is that morbidly obese and super morbidly obese pregnant women will have significant differences in adverse outcomes compared to obese women.MethodsA retrospective cohort study was performed comparing pregnancy outcomes of women with a BMI ≥ 30kg/m2 who delivered at the University of Iowa Hospitals and Clinics (UIHC) between 2009 and 2020. Data were extracted from the electronic health record (Epic) using the Intergenerational Health Knowledgebase, an enterprise data warehouse for research containing validated data from all pregnancies receiving care at UIHC. Additional data was extracted directly from the medical record. Women were included if they were ≥18 years old, had a singleton pregnancy and BMI ≥ 30 kg/m2, and delivered at UIHC (N=6445). Women were stratified into cohorts based on BMI (35‐39.9, 40‐49.9, 50‐59.9, 60‐69.9, 70‐79.9, ≥80 kg/m2) and compared to women with a BMI of 30‐35 kg/m2. Categorical variables were analyzed by chi square and continuous variables by t‐test. Regression models were created to control for confounding.ResultsStatistically significant increased incidences for gestational diabetes, preeclampsia, newborn hypoglycemia, number of days in the hospital post‐delivery, birthweight, and a composite poor neonatal outcome (P&lt;0.001 for each outcome) were associated with increasing BMI. Both the 1‐ and 5‐minute APGAR scores worsened with increasing BMI (P&lt; 0.001). However, there were decreased rates of caesarean section (P&lt;0.001) and pre‐term birth (P=0.02) associated with increasing BMI. There were no statistically significant differences in neonatal ICU admission, death, respiratory distress, and pulmonary hypoplasia.ConclusionBy identifying pregnancy outcomes using classifications with a smaller BMI range, it will be possible to develop more specific counseling and effective care plans for obese, morbidly obese, and super morbidly obese patients in the future.

Patients’ experiences and compliance with preoperative screening and decolonization

BackgroundTo improve adherence with pre-surgical screening for Staphylococcus aureus nasal carriage and decolonization, we need more information about patients’ experiences with these protocols. MethodsWe surveyed patients undergoing orthopedic, neurosurgical, or cardiac operations at Johns Hopkins Hospitals (JHH), the University of Iowa Hospitals and Clinics (UIHC) at MercyOne Northeast Iowa Neurosurgery (MONIN) to assess patients’ experiences with decolonization protocols. ResultsFive hundred thirty-four patients responded. Respondents at JHH were significantly more likely than those at the UIHC to report using mupirocin and were significantly more likely than those at the UIHC and MONIN to feel they received adequate information about surgical site infection (SSI) prevention and decolonization. Respondents at JHH were the least likely to not worry about SSI and they were more willing to do anything they could to prevent SSI. Few patients reported barriers to adherence and side effects of mupirocin or chlorhexidine. ConclusionRespondents did not report either major side effects or barriers to adherence. Patients varied in their level of concern about SSI, their willingness to invest effort in preventing SSI, and their assessments of preoperative information. To improve patients' adherence, clinicians and hospitals should assess their patients’ needs and desires and tailor their preoperative processes, education, and prophylaxis accordingly.

Characterizing head and neck cancer survivors' discontinuation of survivorship care.

Little is known about cancer survivors who discontinue survivorship care. The objective of this study was to characterize patients with head and neck cancer who discontinue survivorship care with their treating institution and identify factors associated with discontinuation. This was a retrospective cohort study of patients diagnosed with head and neck cancer between January 1, 2014, and December 31, 2016, who received cancer-directed therapy at the University of Iowa Hospitals and Clinics (UIHC). Eligible patients achieved a cancer-free status after curative-intent treatment and made at least 1 visit 90+ days after treatment completion. The primary outcome was discontinuation of survivorship care, which was defined as a still living survivor who had not returned to a UIHC cancer clinic for twice the expected interval. Demographic and oncologic factors were examined to identify associations with discontinuation. Ninety-seven of the 426 eligible patients (22.8%) discontinued survivorship care at UIHC during the study period. The mean time in follow-up for those who discontinued treatment was 15.4 months. Factors associated with discontinuation of care included an unmarried status (P = .036), a longer driving distance to the facility (P = .0031), and a single-modality cancer treatment (P < .0001). Rurality was not associated with discontinuation (24.3% vs 21.6% for urban residence; P = .52), nor was age, gender, or payor status. The study results indicate that a sizeable percentage of head and neck cancer survivors discontinue care with their treating institution. Both demographic and oncologic factors were associated with discontinuation at the treating institution, and this points to potential clinical and care delivery interventions.

Bat Intrusions at a Tertiary Care Center, Iowa 2018–2020

Background: Bats are recognized as important vectors in disease transmission. Frequently, bats intrude into homes and buildings, increasing the risk to human health. We describe bat intrusions and exposure incidents in our hospital over a 3-year period. Methods: The University of Iowa Hospitals and Clinics (UIHC) is an 811-bed academic medical center in Iowa City, Iowa. Established in 1928, UIHC currently covers 209,031.84 m2 (~2,250,000 ft2) and contains 6 pavilions built between 1928 and 2017. We retrospectively obtained bat intrusion calls from the infection prevention and control program call database at UIHC during 2018–2020. We have also described the event management for intrusions potentially associated with patient exposures. Results: In total, 67 bat intrusions occurred during 2018–2020. The most frequent locations were hallways or lounges 28 (42%), nonclinical office spaces 19 (14%), and stairwells 8 (12%). Most bat intrusions (65%) occurred during the summer and fall (June–November). The number of events were 15 in 2018, 28 in 2019, and 24 in 2020. We observed that the number of intrusions increased with the age of each pavilion (Figure 1). Of 67 intrusions, 2 incidents (3%) were associated with potential exposure to patients. In the first incident, reported in 2019, the bat was captured in a patient care area and released before an investigation of exposures was completed and no rabies testing was available. Also, 10 patients were identified as having had potential exposure to the bat. Among them, 9 patients (90%) received rabies postexposure prophylaxis. In response to this serious event, we provided facility-wide education on our bat control policy, which includes the capture and safe handling of the bat, assessment of potential exposures, and potential need for rabies testing. We also implemented a bat exclusion project focused on the exterior of the oldest hospital buildings. The second event, 1 patient was identified to have potential exposure to the bat. The bat was captured, tested negative for rabies, no further action was needed. Conclusions: Bat intrusions can be an infection prevention and control challenge in facilities with older buildings. Hospitals may need animal intrusion surveillance systems, management protocols, and remediation efforts.Funding: NoDisclosures: None

Suspected COVID-19 Reinfections at a Tertiary Care Center, Iowa 2020

Background: Coronavirus disease 2019 (COVID-19) is caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). SARS-CoV-2 RNA can be detected by real-time reverse-transcription polymerase chain reaction (RT-PCR) for several weeks after infection. Discerning persistent RT-PCR positivity versus reinfection is challenging and the frequency of COVID-19 reinfections is unknown. We aimed to determine the frequency of clinically suspected reinfection in our center and confirm reinfection using viral whole-genome sequencing (WGS). Methods: The University of Iowa Hospitals and Clinics (UIHC) is an 811-bed academic medical center. Patients with respiratory complaints undergo COVID-19 RT-PCR using nasopharyngeal swabs. The RT-PCR (TaqPath COVID-19 Combo kit) uses 3 targets (ORF1ab, S gene, and N gene). We identified patients with previous laboratory-confirmed COVID-19 who sought care for new respiratory complaints and underwent a repeated SARS-CoV-2 test at least 45 days from their first positive test. We then identified patients with median RT-PCR cycle threshold (Ct) values. Results: During the study period, 13,603 patients had a SARS-CoV-2– positive RT-PCR. Of these, 296 (2.2%) had a clinical visit for new onset of symptoms and a repeated RT-PCR assay &gt;45 days from the first test. Moreover, 29 patients (9.8%) had a positive RT-PCR assay in the repeated testing. Ct values were available for samples from 25 patients; 7 (28%) had Ct values. Conclusions: In patients with a recent history of COVID-19 infection, repeated testing for respiratory symptoms was infrequent. Some had a SARS-CoV-2–positive RT-PCR assay on repeated testing, but only 1 in 4 had Ct values suggestive of a reinfection. We confirmed 1 case of reinfection using WGS.Funding: NoDisclosures: None

A multicenter analysis of treatment and outcomes in neuroendocrine carcinoma of the uterine cervix (NCUC).

e17518 Background: Morbidity and mortality for patients with cervical cancer has improved significantly over the past few decades with modern multimodal therapy. However, neuroendocrine carcinoma of the uterine cervix (NCUC), which accounts for 1-2% of cervical cancers, remains a deadly subtype. In this study, we combine data from Mayo Clinic (MC) and the University of Iowa Hospitals and Clinics (UIHC) to provide information on tumor characteristics, treatment, and outcomes. Methods: The electronic medical record was reviewed for patients with NCUC from MC and UIHC. Data on diagnosis, treatment, and outcomes were collected through chart review. Primary endpoints included progression-free survival (PFS) and overall survival (OS). Secondary endpoints included median survival, survival at 1 year after surgery, and survival at 1 year by first line chemotherapy agent. Kaplan-Meier survival analysis was used to estimate median PFS, median survival, and OS. Fisher’s test analysis was used to calculate survival at 1 year after surgery and by first line chemotherapy agent. Results: There were 62 patients (MC: 26, UIHCC: 36) with NCUC stage I-IV (stage I: 29, stage II: 9, stage III: 7, stage IV: 14, unknown: 3). Median age of diagnosis was 47 years (range 21-77 years). By subtype, 47 were small cell (76%), 9 were large cell (15%), and 6 were unknown/undetermined (9%). The initial treatment modalities for each patient are outlined in the table. 28 patients had complete/partial response or stable disease from first line treatment, while 10 patients had disease progression. Of the patients who initially responded or had stable disease, 16 later progressed (57%) with a median time to progression of 15 months. Median follow up was 65.1 months with a median OS of 28.5 months. Median survival for those with stage I was 40.9 months, stage II: 54.6 months, stage III: 8.75 months, and stage IV: 11.7 months. There was a significant difference in overall survival at 1 year between those who received surgery and those who did not in stage I/II ( p = 0.01). There was no significant difference in overall survival at 1 year for those who received surgery in stage III/IV. There was no statistical difference in survival at 1 year for carboplatin or cisplatin in combination with etoposide as first line chemotherapy agent. Conclusions: NCUC is an aggressive malignancy that is usually progressive despite multimodal therapy. Our study demonstrated a median overall survival of 28.5 months and 5-year survival rate of 21%. Our study showed a survival benefit at 1 year for those who receive surgery with stage I/II NCUC. There was no significant survival benefit at 1 year between carboplatin or cisplatin in combination with etoposide as first line agent.[Table: see text]

Evaluating predictive equations for deriving PaO 2 from SpO 2 in neonatal intensive care patients Joseph McDonell 1 , Trassanee Chatmethakul 3 , Phillip Levy 4 , Patrick J. McNamara 3 , Michael Tomasson 2 , Craig Wheeler 4 , Melissa L. Bates 1,2,3 Departments of

The oxygen-hemoglobin dissociation curve can be used to calculate the arterial partial pressure of oxygen (PaO2) from the peripheral saturation (SpO2)in adults. Our objective was to evaluate whether this calculation could be performed in neonates. We obtained blood gas measurements from 104 de-identified infants at the University of Iowa Hospitals and Clinics (UIHC) neonatal intensive care unit. We plotted PaO2 against SaO2, measured by co-oximetry, and compared the resultant curve to published curves for adults and neonates. The UIHC curve was leftward shifted compared to published curves for neonates and adults. We tested the additional hypothesis that this because of the large proportion of premature infants (72.3%) in our cohort, but gestational age was not a predictor of the shift. Next, we quantified the degree to which pH, PaCO2, and gestational age contributed to sources of error in using this curve to estimate PaO2 from SpO2. We found that deviations in pH and PaCO2 from the physiological ideal contribute up to 3 mmHg and 4 mmHg of error, respectively, to the calculation of PaO2 using our cohort's curve. We demonstrate that data from previously published cohorts may not be universally useful for the calculation of PaO2. In future studies, we will validate the UIHC curve in additional patients with well characterized clinical phenotypes. Additional simulations are needed that quantify the additive contribution of pH, PaCO2, and gestational age to the error in calculated PaO2 in neonates.

Topological data analysis (TDA) enhances bispectral EEG (BSEEG) algorithm for detection of delirium

Current methods for screening and detecting delirium are not practical in clinical settings. We previously showed that a simplified EEG with bispectral electroencephalography (BSEEG) algorithm can detect delirium in elderly inpatients. In this study, we performed a post-hoc BSEEG data analysis using larger sample size and performed topological data analysis to improve the BSEEG method. Data from 274 subjects included in the previous study were analyzed as a 1st cohort. Subjects were enrolled at the University of Iowa Hospitals and Clinics (UIHC) between January 30, 2016, and October 30, 2017. A second cohort with 265 subjects was recruited between January 16, 2019, and August 19, 2019. The BSEEG score was calculated as a power ratio between low frequency to high frequency using our newly developed algorithm. Additionally, Topological data analysis (TDA) score was calculated by applying TDA to our EEG data. The BSEEG score and TDA score were compared between those patients with delirium and without delirium. Among the 274 subjects from the first cohort, 102 were categorized as delirious. Among the 206 subjects from the second cohort, 42 were categorized as delirious. The areas under the curve (AUCs) based on BSEEG score were 0.72 (1st cohort, Fp1-A1), 0.76 (1st cohort, Fp2-A2), and 0.67 (2nd cohort). AUCs from TDA were much higher at 0.82 (1st cohort, Fp1-A1), 0.84 (1st cohort, Fp2-A2), and 0.78 (2nd cohort). When sensitivity was set to be 0.80, the TDA drastically improved specificity to 0.66 (1st cohort, Fp1-A1), 0.72 (1st cohort, Fp2-A2), and 0.62 (2nd cohort), compared to 0.48 (1st cohort, Fp1-A1), 0.54 (1st cohort, Fp2-A2), and 0.46 (2nd cohort) with BSEEG. BSEEG has the potential to detect delirium, and TDA is helpful to improve the performance.

Highly Local Clostridioides difficile Infection (CDI) Pressure as Risk Factors for CDI

Background. Colonization pressure at the unit level is known to be a risk factor for Clostridioides difficile infections in hospitals. Because C. difficile colonization is not routinely detected in clinical practice, only patients identified as having C. difficile infection (CDI) are included in these pressure calculations. We used data from the University of Iowa Hospitals and Clinics (UIHC) to determine whether highly local CDI pressure, due to patients in nearby rooms, is more strongly correlated with CDI than unit-level CDI pressure. Methods: We designed a base logistic regression model using variables known to be risk factors for CDI: age, antibiotic/gastric acid suppressor use, low albumin, prior hospitalization, comorbidities. To the base model, we add 2 measures, mean colonization pressure (MCP) and sum colonization pressure (SCP) of CDI at the unit level to obtain new models. To the base model, we also added CDI colonization pressure by considering CDI cases at different distance thresholds from the focal patient. Distances between patient rooms were extracted from hospital floor plans. Results: Adding unit-level CDI colonization pressures to the base model improved performance. However, adding CDI colonization pressures due to roommates and due to patients at different distances improved the model much more (Table 1). The top (resp. bottom) row shows in-sample (resp. out-of-sample) C-statistics for the base model, the base model with unit-level MCP, the base model with roommate MCP, and the base model with MCP from patients are different distances added as separate features. C-statistics for the base model and the base model with unit CDI pressure (SCP and MCP) are compared in Fig. 1 with C-statistics from the base model with CDI pressure from patients at distances D = 0, 1, 2, 3, 4, 5, 10, 15, 20 hops (1 hop = 5–6 meters). Conclusions: Our results support the hypothesis that unit CDI colonization pressure is a risk factor for CDI. However, by incorporating spatially granular notions of distances between patients in our analysis, we were able to demonstrate that the true source of CDI pressure at the UIHC is almost exclusively attributable to roommates and patients in adjacent rooms.Funding: NoneDisclosures: None

Naturally Emerging Cohorting Behavior of Healthcare Workers and Its Implications for Disease Spread

Background: Mobility patterns of healthcare workers (HCWs) (ie, the spatiotemporal distribution of patient rooms they visit) have a significant impact on the spread of healthcare acquired infections (HAIs). Objective: In this project, we used fine-grained data from a sensor deployment at the medical intensive care unit (MICU) in the University of Iowa Hospitals and Clinics (UIHC) to study the mobility patterns of HCWs and their impact on HAI spread. Methods: We analyzed 10 days of data from a 20-bed MICU sensor deployment. For parameters t1 and t2, each pair of rooms i and j is assigned a weight W(i, j) representing the number of times an HCW spends at least t1 seconds in room i followed by at least t1 seconds in room j, within t2 seconds of each other. W(i, j) is a measure of HCW traffic going from room i to room j; we study the correlation between W(i, j) and the distance between rooms i and j. Additionally, we perform 2 disease-spread simulations: (1) a base simulation, obtained by replaying observed HCW mobility traces and (2) a perturbed simulation, which is the same as the base simulation, except that we replace each HCW who visits a room by a random available HCW. Thus, the perturbed simulation removes correlations in the observed HCW mobility traces. Results: We computed W(i, j) for all room pairs i, j for parameters t1 = 30 seconds and t2 = 1,800 and 3,600 seconds. For nurses, there was a strong negative correlation of between pairwise room distance and the weights W(i, j) (−0.768 for t2 = 1,800; −0.711 for t2 = 3,600), The more distant 2 rooms were, the less they shared nurse traffic. This was not true for physicians (correlation = −0.027 for t2 = 1,800; −0.014 for t2 = 3,600). Figure 1 shows a weight versus distance scatter plot for nurses for t1 = 30 and t2 = 1,800. This spatial correlation has positive implications for disease spread; the base simulation, which preserves these spatial correlations, has between 12% and 55% fewer mean infected patients (&gt;100 replicates) for different simulation parameters compared to the perturbed simulation. Conclusions: Our results, based on fine-grained data, show a “naturally emerging” cohorting behavior of nurses, where nurses are more likely to visit rooms close to each other within a 30–60 minute time window, than rooms further away. Through simulations, this behavior provides substantial protection against disease spread.Funding: NoneDisclosures: None

Mining Camera Traces to Estimate Interactions Between Healthcare Workers and Patients

Background: Simulations based on models of healthcare worker (HCW) mobility and contact patterns with patients provide a key tool for understanding spread of healthcare-acquired infections (HAIs). However, simulations suffer from lack of accurate model parameters. This research uses Microsoft Kinect cameras placed in a patient room in the medical intensive care unit (MICU) at the University of Iowa Hospitals and Clinics (UIHC) to obtain reliable distributions of HCW visit length and time spent by HCWs near a patient. These data can inform modeling efforts for understanding HAI spread. Methods: Three Kinect cameras (left, right, and door cameras) were placed in a patient room to track the human body (ie, left/right hands and head) at 30 frames per second. The results reported here are based on 7 randomly selected days from a total of 308 observation days. Each tracked body may have multiple raw segments over the 2 camera regions, which we “stitch” up by matching features (eg, direction, velocity, etc), to obtain complete trajectories. Due to camera noise, in a substantial fraction of the frames bodies display unnatural characteristics including frequent and rapid directional and velocity change. We use unsupervised learning techniques to identify such “ghost” frames and we remove from our analysis bodies that have 20% or more “ghost” frames. Results: The heat map of hand positions (Fig. 1) shows that high-frequency locations are clustered around the bed and more to the patient’s right in accordance with the general medical practice of performing patient exams from their right. HCW visit frequency per hour (mean, 6.952; SD, 2.855) has 2 peaks, 1 during morning shift and 1 during the afternoon shift, with a distinct decrease after midnight. Figure 2 shows visit length (in minutes) distribution (mean, 1.570; SD, 2.679) being dominated by “check in visits” of &lt;30 seconds. HCWs do not spend much time at touching distance from patients during short-length visits, and the fraction of time spent near the patient’s bed seems to increase with visit length up to a point. Conclusions: Using fine-grained data, this research extracts distributions of these critical parameters of HCW–patient interactions: (1) HCW visit length, (2) HCW visit frequency as a function of time of day, and (3) time spent by HCW within touching distance of patient as a function of visit length. To the best of our knowledge, we provide the first reliable estimates of these parameters.Funding: NoneDisclosures: None

Abstract TP291: Developing an Electronic Medical Record Based Data Abstraction Tool to Allow for Real Time Clinical Quality Review in Stroke Patients

Background: University of Iowa Hospitals and Clinics (UIHC) participates in American Heart Association’s Get with the Guidelines® - Stroke (GWTG) registry to help drive quality improvement. As the only Comprehensive Stroke Center in Iowa, UIHC discharges many stroke patients and experiences exponential stroke patient growth each year. Due to ever-growing patient volumes and limited staff resources, UIHC identified a need to be able to abstract more quickly and efficiently to assess patient level clinical quality and adherence to evidence-based stroke guidelines. Purpose: UIHC engaged their Information Technology (IT) Department to assist in developing a process to expedite the process and reduce the workload associated with collecting stroke data for quality improvement, with the aim of collecting data as close as possible to the episode of care. Methods: UIHC pioneered the connection between GWTG - Stroke and Epic, their electronic medical record, via a re-abstraction tool embedded in Epic. The IT team worked closely with the stroke coordinator and quality improvement team to identify efficient workflows and time saving strategies. Gaps in discrete data collection were identified and collaboration between interdisciplinary care teams commenced to standardize processes for improved charting. Results: Historically, a self-reported average of 30 minutes was spent on each patient chart in GWTG. As of August 2019, that time declined to an average of 10 minutes per chart, representing a 66% reduction in manual labor required. Prior to project implantation, lag time from patient discharge to data abstraction averaged 4 weeks. After implementing the tool, quality data is abstracted and chart review to compare the stroke care episode against current guidelines occurs while the patient is still an inpatient. Conclusions: Investing in and fostering collaboration between IT, stroke, and quality departments at UIHC led to substantial, sustainable reduction in manual work required to collect stroke quality data. Hospitals should explore their ability to create an EMR based re-abstraction tool to not only save time but improve the quality and timeliness of data collection.

Impact of pharmacy services on cystic fibrosis transmembrane conductance regulator modulator prescribing at a pediatric cystic fibrosis center.

This study was undertaken to determine if the presence of a clinical pharmacy team impacted patients' access to cystic fibrosis transmembrane conductance regulator (CFTR) modulators. A retrospective chart review of electronic medical records from the University of Iowa Hospitals and Clinics (UIHC) was conducted. Data were collected regarding the timing of prior authorization (PA) submissions and approvals from 2012 to 2018. The Wilcoxon rank-sum test was used to compare the meantime (days) between prescription and PA submission dates, and PA submission and approval date for all patients included in the analysis. Comparisons were made for pre- and postpharmacy services eras as well as the UIHC Specialty Pharmacy versus a non-UIHC Specialty Pharmacy. Sixty-three patients were included in the final analysis. The average time between prescription date and PA submission was 12.5 days (standard deviation [SD] = 17.4 days) in the preclinical pharmacy services era and 3.5 days (SD = 5.8 days; P = .028) in the postclinical pharmacy services era. The average time to PA submission significantly decreased from 9.8 days (SD = 13.1 days) to 1.3 days (SD = 4.2 days; P < .0001) when prescriptions were filled by the UIHC Specialty Pharmacy vs a non-UIHC Specialty Pharmacy. There was a significant benefit to CFTR modulator prescribing when clinical pharmacy services were incorporated in our cystic fibrosis (CF) care team, which will become increasingly important with the anticipation of new CF medications in the near future.

College Palliative Care Volunteers: Too Early to Feed the Pipeline for Palliative Care Clinicians?

Background: The University of Iowa Hospitals and Clinics (UIHC) has a palliative care (PC) volunteering program that has recruited college students since 2010. There is little research on the effects of PC volunteering on collegiate volunteers. Objective: The objective is to determine the impact of PC volunteering on college students' professional lives and on their interest in PC. Design/Setting/Measurements: The UIHC Volunteer Services office sent a 25-question survey with closed- and open-ended items to previous and current PC college volunteers. We used descriptive statistics to characterize the sample. Free text responses were analyzed using a descriptive qualitative approach with three independent coders. Results: Seventy-one percent of respondents (23/33) reported they were more likely to pursue PC after volunteering. PC volunteering helped change views of patient care and abilities to discuss end-of-life situations. Ninety-one percent served as informal ambassadors by discussing PC with family and peers. Major themes identified include motivation to volunteer for patient contact and interest in learning about PC. Respondents described meaningful patient interactions, lessons in empathy, and the power of listening. Conclusions: These results suggest that PC volunteering affects career choices and helps volunteers gain needed listening skills for patient care in the future. The unique exposure and interactions with PC patients and their families have changed volunteers' understanding of health care. PC volunteers speak in their social networks about PC. This experience may increase the likelihood of student volunteers to pursue careers in PC.

Team orienteering with uncertain rewards and service times with an application to phlebotomist intrahospital routing

AbstractThis study focuses on the intrahospital routing of phlebotomists at the University of Iowa Hospitals and Clinics (UIHC). Phlebotomists are responsible for drawing specimens from patients based on doctors' orders. The results of the analysis of these specimens play an important role in determining patient treatment. However, the demand for phlebotomists is likely to outpace supply over the next few years. Therefore, it is important to improve the efficiency of phlebotomists. In this paper, we formulate the phlebotomist intrahospital routing problem as a team orienteering problem with stochastic rewards and service times. The rewards and service times are particularly interesting as they are the result of a queueing process. We present an a priori solution approach and derive a method for efficiently sampling the value of a solution, a value that cannot be determined analytically. Finally, we demonstrate that our proposed approach outperforms the current practice at UIHC.