Implementation of a Pediatric Emergency Telemedicine Program.

OBJECTIVES: Our goal was to describe the experiences after the launch of a pediatric emergency telemedicine program at a large, urban, academic medical center. METHODS: We launched 3 unique pediatric emergency telemedicine programs at an urban, academic medical center: direct-to-consumer pediatric virtual urgent care, pediatric emergency department (PED) telemedicine follow-up, and telemedicine medical screening examination in the PED. RESULTS: We evaluated 84 patients via direct-to-consumer pediatric virtual urgent care with the most common chief complaint related to fever, dermatologic, or respiratory systems; we referred 12% to the PED, and 20% of those required hospital admission. We evaluated 38 patients via PED telemedicine follow-up; we referred 19% back to the PED, and 43% of those required hospital admission. Median duration for a telemedicine encounter was 10 minutes. We screened 3809 patients in the PED using telemedicine medical screening examination. CONCLUSIONS: We offer a description of an innovative and comprehensive new pediatric emergency telemedicine program implemented at a large, urban, academic medical center. Our initial findings demonstrate short visit times, antibiotic stewardship, and low rates of PED referral and subsequent admission for patients who use a telemedicine service. We plan to further examine the impact of pediatric emergency telemedicine on the care of children as our program expands.

Antibiotic Prescribing Practices: Is There a Difference Between Patients Seen by Telemedicine Versus Those Seen In-Person?

Background: Direct-to-consumer telemedicine is becoming part of mainstream medicine, but questions exist regarding the quality of care provided. We assessed antibiotic stewardship, one measure of quality, by comparing antibiotic prescription rates for acute respiratory infections (ARIs) between patients seen by telemedicine and patients seen in-person in two urban emergency departments (EDs). Methods: In two urban EDs where low-acuity patients in the ED have the option of being seen by telemedicine rather than in-person, we analyzed telemedicine and in-person visits of patients ≥18 years who received ARI diagnoses between July 2016 and September 2017. The identified ARI telemedicine visits were matched to in-person visits by diagnosis, treatment hospital, and Emergency Severity Index level. We compared antibiotic prescribing rates for telemedicine and in-person visits. Results: We identified 260 telemedicine visits and compared with 260 matched in-person visits. Antibiotics for ARIs were prescribed for 29% of telemedicine visits and 28% of in-person visits (odds ratio [OR] 1.038; 95% confidence interval [CI] 0.71-1.52; p = 0.846). This finding did not materially change after adjustment for age and gender (adjusted OR 1.034; 95% CI 0.70-1.53; p = 0.86). Conclusions: Antibiotic prescribing rates for ARIs were similar for patients seen by telemedicine and patients seen in-person at two urban EDs. If differences in antibiotic stewardship between telemedicine and in-person encounters are found, contextual factors unrelated to the video-based evaluation should be investigated.

Telemedicine Evaluations for Low-Acuity Patients Presenting to the Emergency Department: Implications for Safety and Patient Satisfaction.

Background:Telemedicine is being rapidly adopted by traditional health care systems. We have used telemedicine in a program we call Express Care to allow a single physician to remotely perform evaluations of low-acuity patients.Materials and Methods:We conducted a retrospective cohort study of quality assurance data comparing low-acuity patients treated by an emergency department (ED) physician through telemedicine (Express Care) with those treated by an ED physician in person between July 16, 2016 and September 30, 2017. We compared patient demographics, length of stay (LOS), visit severity as measured by emergency severity index (ESI), visit diagnosis type, return visits, and patient satisfaction scores.Results:There were 3,266 low-acuity patients seen through telemedicine and 21,129 seen in person during the observation period. Patients receiving evaluation by telemedicine were younger (mean age ± standard deviation [in years]: 42 ± 18 vs. 45 ± 17; p < 0.001) and more likely to be male (51% vs. 46%; p < 0.001). Median ESI was slightly lower for patients treated by telemedicine [4 (4-5) vs. 4 (4-4); p < 0.001], and there were modest differences in diagnosis type between the two groups. Median ED LOS was 63.6 (interquartile range [IQR] 42.6-93.6) min for telemedicine patients and 133.8 (IQR 90.6-196.8) min for patients seen in person (p < 0.001). Seventy-two hour returns (3.4% vs. 3.0%; p = 0.302) and 72-h returns requiring admission (0.2% vs. 0.3%; p = 0.252) were similar between groups. Patient satisfaction scores were also similar between the groups.Conclusion:Telemedicine evaluation for ED patients can be effective and safe when treating low-acuity conditions without compromising patient satisfaction.

A Novel Emergency Department-Based Telemedicine Program: How Do Older Patients Fare?

Introduction:When we started using telemedicine to treat low acuity patients in the emergency department (ED), we assumed that this voluntary treatment pathway would primarily be used by younger patients. We were surprised to find that a significant portion of patients evaluated by telemedicine were older adults.Materials and Methods:We conducted a retrospective cohort study of quality assurance data. Adult ED patients at an urban academic medical center who had their care provided by telemedicine from July 2016 to September 2017 were included. We measured demographic characteristics, ED length of stay (LOS), triage severity score, X-ray orders placed, ED revisit within 72 h, need for change in treatment plan or admission on 72-h return, and patient satisfaction.Results:Of 1,592 patients evaluated, 18% were age 65 and older. Older patients were more likely to be evaluated for wound care and less likely to be evaluated for nontraumatic connective tissue illnesses. Older patients also had shorter median LOS (59 min vs. 63 min). Unplanned 72-h return (2% vs. 2%), likelihood to have a change in treatment on return (1% vs. 0.2%), and patient satisfaction were similar between age groups. The percentage of patients who returned in 72 h requiring admission were similar between age groups (0.4% vs. 0.1% p = 0.325). Sensitivity analysis with an age threshold at 75 years did not change primary results.Conclusion:These findings suggest that among low acuity patients there are groups of older adults for whom an ED telemedicine evaluation can provide safe and effective medical care that is satisfactory to patients.

Predicting frequent emergency department use among children with epilepsy: A retrospective cohort study using electronic health data from 2 centers.

OBJECTIVE: Among children with epilepsy, to develop and evaluate a model to predict emergency department (ED) use, an indicator of poor disease control and/or poor access to care. METHODS: We used electronic health record data from 2013 to predict ED use in 2014 at 2 centers, benchmarking predictive performance against machine learning algorithms. We evaluated algorithms by calculating the expected yearly ED visits among the 5% highest risk individuals. We estimated the breakeven cost per patient per year for an intervention that reduced ED visits by 10%. We estimated uncertainty via cross-validation and bootstrapping. RESULTS: Bivariate analyses showed multiple potential predictors of ED use (demographics, social determinants of health, comorbidities, insurance, disease severity, and prior health care utilization). A 3-variable model (prior ED use, insurance, number of antiepileptic drugs [AEDs]) performed as well as the best machine learning algorithm at one center (N = 2730; ED visits among top 5% highest risk, 3-variable model, mean = 2.9, interquartile range [IQR] = 2.7-3.1 vs Random Forest, mean = 2.9, IQR = 2.7-3.1), and superior at the second (N = 784; mean = 2.5, IQR = 2.2-2.9 vs mean = 1.9, IQR = 1.6-2.5). The per-patient-per-year breakeven point using this model to identify high-risk individuals was $958 (95% confidence interval [CI] = $568-$1390) at one center and $1086 (95% CI = $886-$1320) at the second. SIGNIFICANCE: Prior ED use, insurance status, and number of AEDs, taken together, predict future ED use for children with epilepsy. Our estimates suggest a program targeting high-risk children with epilepsy that reduced ED visits by 10% could spend approximately $1000 per patient per year and break even. Further work is indicated to develop and evaluate such programs.

Care coordination at a pediatric accountable care organization (ACO): A qualitative analysis.

OBJECTIVE: Care coordinators may help manage care for children with chronic illness. Their role in pediatric epilepsy care is understudied. We aimed to qualitatively describe the content of a care coordination intervention for children with epilepsy. METHODS: We conducted nine semi-structured interviews and one focus group with care coordinators at a pediatric accountable care organization (ACO) in Ohio. The care coordinators used a modified version of a published care coordination checklist for children with epilepsy (Patel AD, 2014). We analyzed transcripts using thematic analysis. We focused on (1) the content of the intervention; and (2) perceptions of facilitators and barriers to improve outcomes, with an emphasis on epilepsy specific facilitators and barriers. RESULTS: Care coordinators interacted with children and families in multiple contexts (phone calls, physician visits, home visits), and included relationship building (developing rapport and trust between families and the health system), communication (transmission of information between the child, family, physician, and other care providers), and service (help with housing, transportation, scheduling, liaison with community resources, etc.). Facilitators and barriers of care coordination included factors related to parents, physicians, health system, payers, and community. Epilepsy-specific barriers included stigma (felt & enacted) and the anxiety associated with clinical uncertainty. Epilepsy related facilitators included a seizure action plan, written educational materials, and an epilepsy specific care coordination checklist. CONCLUSION: In addition to facilitators and barriers common to many care coordination programs, pediatric epilepsy care coordinators should be particularly aware of epilepsy stigma and clinical uncertainty. A care coordination checklist and epilepsy focused educational materials written to accommodate people with low health literacy may provide additional benefit. Further research is required to understand the effect of care coordination on costs, use of health services, seizure control, and quality of life for children with epilepsy.

"You Get Reminded You're a Sick Person": Personal Data Tracking and Patients With Multiple Chronic Conditions.

BACKGROUND: Consumer health information technologies (HIT) that encourage self-tracking, such as diet and fitness tracking apps and disease journals, are attracting widespread interest among technology-oriented consumers (such as "quantified self" advocates), entrepreneurs, and the health care industry. Such electronic technologies could potentially benefit the growing population of patients with multiple chronic conditions (MCC). However, MCC is predominantly a condition of the elderly and disproportionately affects the less affluent, so it also seems possible that the barriers to use of consumer HIT would be particularly severe for this patient population. OBJECTIVE: Our aim was to explore the perspectives of individuals with MCC using a semistructured interview study. Our research questions were (1) How do individuals with MCC track their own health and medical data? and (2) How do patients and providers perceive and use patient-tracked data? METHODS: We used semistructured interviews with patients with multiple chronic diseases and providers with experience caring for such patients, as well as participation in a diabetes education group to triangulate emerging themes. Data were analyzed using grounded theory and thematic analysis. Recruitment and analysis took place iteratively until thematic saturation was reached. RESULTS: Interviews were conducted with 22 patients and 7 health care providers. The patients had an average of 3.5 chronic conditions, including type 2 diabetes, heart disease, chronic pain, and depression, and had regular relationships with an average of 5 providers. Four major themes arose from the interviews: (1) tracking this data feels like work for many patients, (2) personal medical data for individuals with chronic conditions are not simply objective facts, but instead provoke strong positive and negative emotions, value judgments, and diverse interpretations, (3) patients track for different purposes, ranging from sense-making to self-management to reporting to the doctor, and (4) patients often notice that physicians trust technologically measured data such as lab reports over patients' self-tracked data. CONCLUSIONS: Developers of consumer health information technologies for data tracking (such as diet and exercise apps or blood glucose logs) often assume patients have unlimited enthusiasm for tracking their own health data via technology. However, our findings potentially explain relatively low adoption of consumer HIT, as they suggest that patients with multiple chronic illnesses consider it work to track their own data, that the data can be emotionally charged, and that they may perceive that providers do not welcome it. Similar themes have been found in some individual chronic diseases but appeared more complex because patients often encountered "illness work" connected to multiple diseases simultaneously and frequently faced additional challenges from aging or difficult comorbidities such as chronic pain, depression, and anxiety. We suggest that to make a public health impact, consumer HIT developers should engage creatively with these pragmatic and emotional issues to reach an audience that is broader than technologically sophisticated early adopters. Novel technologies are likely to be successful only if they clearly reduce patient inconvenience and burden, helping them to accomplish their "illness work" more efficiently and effectively.

The invisible work of personal health information management among people with multiple chronic conditions: qualitative interview study among patients and providers.

BACKGROUND: A critical problem for patients with chronic conditions who see multiple health care providers is incomplete or inaccurate information, which can contribute to lack of care coordination, low quality of care, and medical errors. OBJECTIVE: As part of a larger project on applications of consumer health information technology (HIT) and barriers to its use, we conducted a semistructured interview study with patients with multiple chronic conditions (MCC) with the objective of exploring their role in managing their personal health information. METHODS: Semistructured interviews were conducted with patients and providers. Patients were eligible if they had multiple chronic conditions and were in regular care with one of two medical organizations in New York City; health care providers were eligible if they had experience caring for patients with multiple chronic conditions. Analysis was conducted from a grounded theory perspective, and recruitment was concluded when saturation was achieved. RESULTS: A total of 22 patients and 7 providers were interviewed; patients had an average of 3.5 (SD 1.5) chronic conditions and reported having regular relationships with an average of 5 providers. Four major themes arose: (1) Responsibility for managing medical information: some patients perceived information management and sharing as the responsibility of health care providers; others­particularly those who had had bad experiences in the past­took primary responsibility for information sharing; (2) What information should be shared: although privacy concerns did influence some patients' perceptions of sharing of medical data, decisions about what to share were also heavily influenced by their understanding of health and disease and by the degree to which they understood the health care system; (3) Methods and tools varied: those patients who did take an active role in managing their records used a variety of electronic tools, paper tools, and memory; and (4) Information management as invisible work: managing transfers of medical information to solve problems was a tremendous amount of work that was largely unrecognized by the medical establishment. CONCLUSIONS: We conclude that personal health information management should be recognized as an additional burden that MCC places upon patients. Effective structural solutions for information sharing, whether institutional ones such as care management or technological ones such as electronic health information exchange, are likely not only to improve the quality of information shared but reduce the burden on patients already weighed down by MCC.

Focus Section Health IT Usability: Applying a Task-Technology Fit Model to Adapt an Electronic Patient Portal for Patient Work.

OBJECTIVES: Although electronic patient portals are offered by most health care organizations, poor usability and poor fit to patient needs may pose barriers to adoption. We collaborated with an academic hospital to conduct iterative user evaluation of a newly deployed portal designed to deliver inpatient data upon hospital discharge. METHODS: Three evaluators applied heuristic usability evaluation and conducted 23 individual user testing sessions with patients with chronic disease or managing the care of family members with chronic disease. Evaluation and development/improvement were conducted iteratively. User testing and analysis of qualitative data were both conducted from the perspective of a task-technology fit framework, to assess the degree of fit between the portal and patient work. RESULTS: Ability to complete health information management tasks, perceived usability, and positive comments from users improved over the course of the iterative development. However, patients still encountered significant difficulties accomplishing certain tasks such as setting up proxy accounts. The problems were most severe when patients did not start with a clear understanding of tasks that they could accomplish. In exploring the portal, novice users frequently described anecdotes from their own medical history or constructed fictional narratives about a hypothetical patient. CONCLUSION: Chronic illness imposes a significant workload on patients, and applying a task-technology framework for evaluation of a patient portal helped improve the portal's fit to patient needs. However, it also revealed that patients often lack a clear understanding of tasks that would help them accomplish personal health information management. Portal developers may need to educate patients about types of patient work involving medical centers, in a way that developers of clinical information systems do not need to do. An approach to doing this might be to provide narratives about hypothetical patients.

Common terms for rare epilepsies: Synonyms, associated terms, and links to structured vocabularies.

Identifying individuals with rare epilepsy syndromes in electronic data sources is difficult, in part because of missing codes in the International Classification of Diseases (ICD) system. Our objectives were the following: (1) to describe the representation of rare epilepsies in other medical vocabularies, to identify gaps; and (2) to compile synonyms and associated terms for rare epilepsies, to facilitate text and natural language processing tools for cohort identification and population-based surveillance. We describe the representation of 33 epilepsies in 3 vocabularies: Orphanet, SNOMED-CT, and UMLS-Metathesaurus. We compiled terms via 2 surveys, correspondence with parent advocates, and review of web resources and standard vocabularies. UMLS-Metathesaurus had entries for all 33 epilepsies, Orphanet 32, and SNOMED-CT 25. The vocabularies had redundancies and missing phenotypes. Emerging epilepsies (SCN2A-, SCN8A-, KCNQ2-, SLC13A5-, and SYNGAP-related epilepsies) were underrepresented. Survey and correspondence respondents included 160 providers, 375 caregivers, and 11 advocacy group leaders. Each epilepsy syndrome had a median of 15 (range 6-28) synonyms. Nineteen had associated terms, with a median of 4 (range 1-41). We conclude that medical vocabularies should fill gaps in representation of rare epilepsies to improve their value for epilepsy research. We encourage epilepsy researchers to use this resource to develop tools to identify individuals with rare epilepsies in electronic data sources.

Health IT Usability Focus Section: Adapting EHR-Based Medication Instructions to Comply with Plain Language Guidance-A Randomized Experiment.

Objective Patient instructions are generally written by clinicians. However, clinician-centered language is challenging for patients to understand; in the case of pediatric medication instructions, consequences can be serious. Using examples of clinician-written medication instructions from an electronic health record, we conducted an experiment to determine whether parental misinterpretations would be reduced by instructions that followed best practices for plain language. Methods We selected examples of dosing instructions from after-visit summaries in a commercial electronic health record. A demographically diverse sample of parents and adult caregivers was recruited from an online panel to participate in an English-language experiment, in which they received a comprehension questionnaire with either original after-visit summary instructions or instructions revised to comply with federal and other sources of plain-language guidance. Results Nine-hundred and fifty-one respondents completed the experiment; 50% were women, the mean age was 36 years, and 38% had less than a 4-year college education. The revisions were associated with an 8 percentage point increase in correct answers overall (from 55% to 63%, p < 0.001), although revisions were not equally effective for all instructions. Health literacy and health numeracy were strong and independent predictors of comprehension. Overall, mistakes on comprehension questions were common, with respondents missing an average of 41% (6.1 of 15) of questions. Conclusion In this experimental study, a relatively simple intervention of revising text was associated with a modest reduction in frequency of misinterpretations of medication instructions. As a supplement to more intensive high-touch interventions, revising electronic health record output to replace complex language with patient-centered language in an automated fashion is a potentially scalable solution that could reduce medication administration errors by parents.

Barriers to Genetic Testing for Pediatric Medicaid Beneficiaries With Epilepsy.

BACKGROUND: Children with public insurance (Medicaid) have increased barriers to specialty care in the United States. For children with epilepsy, the relationship between public insurance and barriers to genetic testing is understudied. METHODS: We surveyed a sample of US child neurology clinicians. We performed quantitative and qualitative analysis of responses. RESULTS: There were 302 responses (of 1982 surveyed; response rate 15%) from clinicians from 46 states, the District of Columbia, and Puerto Rico, including board-certified child neurologists (82%), resident physicians (6%), nurses (3%), and nurse practitioners (3%). Clinicians felt it was more difficult to get genetic testing for patients with Medicaid insurance compared with commercial insurance, (43% vs 12%, P < 0.05), although many felt it was about the same degree of difficulty (25%) or were not sure (20%). Increased availability of testing was associated with less complex testing (P < 0.001), in-house testing (P < 0.001), and no preauthorization requirements (P < 0.001). Qualitative responses described barriers related to cost, clinician familiarity and comfort, commercial laboratories, health care organization, payer, and patient concerns. Descriptions of facilitators included lowered cost, availability of clinical genetics expertise, clinician knowledge, commercial laboratory assistance, health care organizational changes, improved payer coverage, and increased interest by parents. CONCLUSIONS: Pediatric Medicaid beneficiaries with epilepsy have barriers to genetic testing, compared with children with commercial insurance, particularly for more advanced testing. Potential strategies to improve access include broader coverage, lower co-pays, increased capacity for testing outside of specialty laboratories, fewer preauthorization requirements, improved clinician education, ongoing development and dissemination of guidelines, improved availability of clinical genetics services, and continued assistance programs from commercial laboratories.

Increasing Ketamine Use for Refractory Status Epilepticus in US Pediatric Hospitals.

Ketamine is an emerging therapy for pediatric refractory status epilepticus. The circumstances of its use, however, are understudied. The authors described pediatric refractory status epilepticus treated with ketamine from 2010 to 2014 at 45 centers using the Pediatric Hospital Inpatient System database. For comparison, they described children treated with pentobarbital. The authors estimated that 48 children received ketamine and pentobarbital for refractory status epilepticus, and 630 pentobarbital without ketamine. Those receiving only pentobarbital were median age 3 [interquartile range 0-10], and spent 30 [18-52] days in-hospital, including 17 [9-28] intensive care unit (ICU) days; 17% died. Median cost was $148 000 [81 000-241 000]. The pentobarbital-ketamine group was older (7 [2-11]) with longer hospital stays (51 [30-93]) and more ICU days (29 [20-56]); 29% died. Median cost was $298 000 [176 000-607 000]. For 71%, ketamine was given ≥1 day after pentobarbital. Ketamine cases per half-year increased from 2 to 9 ( P < .05). Ketamine is increasingly used for severe pediatric refractory status epilepticus, typically after pentobarbital. Research on its effectiveness is indicated.

Socioeconomic disparities in adoption of personal health records over time.

OBJECTIVES: Since 2009, federal policies have incentivized medical organizations to provide medical record access to patients. We sought to track personal health record (PHR) adoption and differences by sociodemographic group over time. STUDY DESIGN: Random-digit-dial survey conducted for 4 consecutive years in New York state. METHODS: The Empire State Poll is a random digit-dial survey conducted every year in New York state, with an annual sample size of 800 individuals, weighted to create a representative state sample. We analyzed 4 consecutive years of poll data to examine trends. RESULTS: The proportion of New Yorkers using PHRs rose sharply, from 11% in 2012 to 27% in 2015. By 2015, there were no significant differences in PHR use between blacks and other races, but Hispanics and low-income respondents were less likely to use PHRs. CONCLUSIONS: During a 4-year period in which federal policies incentivized medical organizations to give medical record access to patients through PHRs and electronic portals, rates of PHR use increased rapidly in all sociodemographic groups. However, a digital divide remains evident, linked to Hispanic ethnicity and lower income.

Feasibility of Automatic Extraction of Electronic Health Data to Evaluate a Status Epilepticus Clinical Protocol.

Status epilepticus is a common neurologic emergency in children. Pediatric medical centers often develop protocols to standardize care. Widespread adoption of electronic health records by hospitals affords the opportunity for clinicians to rapidly, and electronically evaluate protocol adherence. We reviewed the clinical data of a small sample of 7 children with status epilepticus, in order to (1) qualitatively determine the feasibility of automated data extraction and (2) demonstrate a timeline-style visualization of each patient's first 24 hours of care. Qualitatively, our observations indicate that most clinical data are well labeled in structured fields within the electronic health record, though some important information, particularly electroencephalography (EEG) data, may require manual abstraction. We conclude that a visualization that clarifies a patient's clinical course can be automatically created using the patient's electronic clinical data, supplemented with some manually abstracted data. Future work could use this timeline to evaluate adherence to status epilepticus clinical protocols.