AKI in Hospitalized Children: Poorly Documented (and Underrecognized)

Background: Acute kidney injury (AKI) is common in hospitalized children. We hypothesized that hospital-acquired AKI would be underrecognized and under-reported, with potential implications for prevention of future AKI and CKD risk stratification.Methods: Five hundred thirty-two cases of AKI occurring over a 1 year period in a tertiary children's hospital in the United States were studied. AKI documentation was defined as any mention of AKI in the admission history and physical note, progress notes, or discharge summary. Nephrology follow-up was defined as a completed outpatient clinic visit within 1 year of discharge. Logistic regression was used to assess factors associated with documentation, consultation, and follow-up.Results: AKI developed during 584/7,640 (7.6%) of hospitalizations: 532 cases met inclusion criteria. Documentation was present in 34% (185/532) of AKI cases and 90 (16.9%) had an inpatient nephrology consult. Among 501 survivors, 89 (17.8%) had AKI in their hospital discharge summary and 54 had outpatient nephrology follow up. Stage 3 AKI, peak creatinine >1 mg/dL and longer length of stay were associated with documentation. Stage 3 AKI and higher baseline creatinine were associated with inpatient nephrology consultation. Inpatient nephrology consultation was positively associated with outpatient nephrology follow up, but documentation in the discharge summary was not.Conclusion: Most cases of AKI were not documented and the proportion of children seen by a nephrologist was low, even among those with more severe injury. Increased severity of AKI was associated with documentation and inpatient consultation. Poor rates of documentation has implications for AKI recognition and appropriate management and follow up.

The Healthcare Journey of Women With Advanced Gynecological Cancer From Diagnosis Through Terminal Illness: Qualitative Analysis of Progress Note Data

Objective To examine women’s journeys with gynecologic cancer from before diagnosis through death and identify elements of their healthcare experience that warrant improvement. Methods This exploratory study used longitudinal progress notes data from a multispecialty practice in Northern California. The sample included women with stage IV gynecological cancer diagnosed after 2011 and who died before 2018. Available progress notes from prior to diagnosis to death were qualitatively analyzed. Results We identified 32 women, (median age 61 years) with mostly uterine (n=17) and ovarian (n=9) cancers and median survival of 9.2 months (min:2.9 and max:47.5). Sixteen (50%) received outpatient palliative care and 18 (56%) received hospice care. The analysis found wide variation in documentation about communication about diagnosis, prognosis, goals of care, stopping treatment, and starting hospice care. Challenges included escalating/severe symptoms, repeated urgent care/emergency department/hospital encounters, and lack of or late access to palliative and hospice care. Notes also illustrated how patient background and goals influenced care trajectory and communication. Documentation styles varied substantially, with palliative care notes more consistently documenting conversations about goals of care and psychosocial needs. Conclusion This analysis of longitudinal illness experience of women with advanced gynecological cancer suggests that clinicians may want to (1) prioritize earlier discussion about goals of care; (2) provide supplemental support to patients with higher needs, possibly through palliative care or navigation; and (3) write notes to enhance patient understanding now that patients may access all notes.

Enhancing goals of care communication by oncologists using a pathway-based intervention.

324 Background: Delivering goal-concordant care to patients with advanced cancer requires identifying eligible patients who would benefit from goals of care (GOC) conversations; training clinicians how to have these conversations; conducting conversations in a timely manner; and documenting GOC conversations that can be readily accessed by care teams. We used an existing, locally developed electronic cancer care clinical pathways system to guide oncologists toward these conversations. Methods: To identify eligible patients, pathways directors from 12 oncology disease centers identified therapeutic decision nodes for each pathway that corresponded to a predicted life expectancy of ≤1 year. When oncologists selected one of these pre-identified pathways nodes, the decision was captured in a relational database. From these patients, we sought evidence of GOC documentation within the electronic health record by extracting coded data from the advance care planning (ACP) module—a designated area within the electronic health record for clinicians to document GOC conversations. We also used rule-based natural language processing (NLP) to capture free text GOC documentation within these same patients’ progress notes. A domain expert reviewed all progress notes identified by NLP to confirm the presence of GOC documentation. Results: In a pilot sample obtained between March 20 and September 25, 2020, we identified a total of 21 pathway nodes conveying a poor prognosis, which represented 91 unique patients with advanced cancer. Among these patients, the mean age was 62 (SD 13.8) years old; 55 (60.4%) patients were female, and 69 (75.8%) were non-Hispanic White. The cancers most represented were thoracic (32 [35.2%]), breast (31 [34.1%]), and head and neck (13 [14.3%]). Within the 3 months leading up to the pathways decision date, a total 62 (68.1%) patients had any GOC documentation. Twenty-one (23.1%) patients had documentation in both the ACP module and NLP-identified progress notes; 5 (5.5%) had documentation in the ACP module only; and 36 (39.6%) had documentation in progress notes only. Twenty-two unique clinicians utilized the ACP module, of which 1 (4.5%) was an oncologist and 21 (95.5%) were palliative care clinicians. Conclusions: Approximately two thirds of patients had any GOC documentation. A total of 26 (28.6%) patients had any GOC documentation in the ACP module, and only 1 oncologist documented using the ACP module, where care teams can most easily retrieve GOC information. These findings provide an important baseline for future quality improvement efforts (e.g., implementing serious illness communications training, increasing support around ACP module utilization, and incorporating behavioral nudges) to enhance oncologists’ ability to conduct and to document timely, high quality GOC conversations.

763Validity of algorithms for identifying five chronic conditions in MedicineInsight, Australian national primary care data

Background MedicineInsight is a database containing de-identified electronic health records (EHRs) from over 700 Australian general practices. To support the trust placed in analyses of MedicineInsight data, additional evidence regarding the accuracy of the data is needed. Methods This study measures the validity of algorithms available in MedicineInsight that identify patients with depression, anxiety, asthma, type 2 diabetes and osteoporosis. Fifty practices met eligibility criteria regarding patient load and location, five were randomly selected and four agreed to participate. Within each practice, 250 patients aged ≥ 40 years were randomly selected. This age restriction increased the prevalence of the evaluated conditions, thereby optimising statistical power. Trained staff review the full EHR for these patients, including progress notes and correspondence, which are not available in MedicineInsight because they may contain identifiable information. Results With data collection almost complete, the target sample size will not be attainable. Power calculations indicate the current sample of 479 should provide adequate precision. For each condition of interest, the sensitivity, specificity, positive predictive value and negative predictive value of the algorithm is calculated. The full EHR review is the gold standard against which the algorithms are benchmarked. Conclusions The findings will indicate whether these algorithms demonstrate adequate accuracy to be used for research and decision-making. Key messages This additional understanding regarding the accuracy of MedicineInsight data will facilitate the interpretation of analyses of MedicineInsight data and guide improvements to the algorithms.

Open notes sounds great, but will a provider’s documentation change? An exploratory study of the effect of open notes on oncology documentation

Objective The effects of shared clinical notes on patients, care partners, and clinicians (“open notes”) were first studied as a demonstration project in 2010. Since then, multiple studies have shown clinicians agree shared progress notes are beneficial to patients, and patients and care partners report benefits from reading notes. To determine if implementing open notes at a hematology/oncology practice changed providers’ documentation style, we assessed the length and readability of clinicians’ notes before and after open notes implementation at an academic medical center in Boston, MA, USA. Materials and Methods We analyzed 143 888 notes from 60 hematology/oncology clinicians before and after the open notes debut at Beth Israel Deaconess Medical Center, from January 1, 2012 to September 1, 2016. We measured the providers’ (medical doctor/nurse practitioner) documentation styles by analyzing character length, the number of addenda, note entry mode (dictated vs typed), and note readability. Measurements used 5 different readability formulas and were assessed on notes written before and after the introduction of open notes on November 25, 2013. Results After the introduction of open notes, the mean length of progress notes increased from 6174 characters to 6648 characters (P < .001), and the mean character length of the “assessment and plan” (A&P) increased from 1435 characters to 1597 characters (P < .001). The Average Grade Level Readability of progress notes decreased from 11.50 to 11.33, and overall readability improved by 0.17 (P = .01). There were no statistically significant changes in the length or readability of “Initial Notes” or Letters, inter-doctor communication, nor in the modality of the recording of any kind of note. Conclusions After the implementation of open notes, progress notes and A&P sections became both longer and easier to read. This suggests clinician documenters may be responding to the perceived pressures of a transparent medical records environment.

Effects of pharmacogenomic (PGx) testing on clinical pain management prescriptions, a retrospective study.

Current deficits in effectively utilizing PGx testing in clinical practice include limited awareness and training of healthcare professionals, routine ordering of assays investigating up to 5 genes and lack of concise reporting of dosing guidelines and drug-drug-interactions. A novel deep sequencing (>1000X) PGx panel is described encompassing 23 genes and 141 SNPs or indels combined with PGx dosing guidance, drug-gene-interaction (DGI) and drug-drug-interaction (DDI) reporting to prevent adverse drug reaction events. During a 2-year period, patients (n = 171) were monitored in a pain management clinic. Urine toxicology, PGx reports, and progress notes were studied retrospectively for changes in prescription regimens before and after the PGx report was made available to the provider. Among patient PGx reports with medication lists provided (n = 146) 57.5% showed one or more moderate and 5.5% at least one serious pharmacogenetic interaction. 66% of patients showed at least one moderate and 15% one or more serious drug-gene or drug-drug-interaction. A significant number of active changes in prescriptions based on the PGx reports provided was observed for 85 patients (83%) for which a specific drug was either discontinued, switched within the defined drug classes of the report or a new drug added. Preventative action was observed for all serious interactions and only moderate interactions were tolerated for lack of other alternatives. This study demonstrates a successful implementation of PGx testing utilizing an extended PGx panel combined with a customized, informational report to help improve clinical outcomes.

Re-Envisioning School-Based Counseling: Sports-Based Group Therapy for Elementary School Students Exposed to Trauma

Mixed-methods were used to investigate treatment outcomes for youth enrolled in an innovative sports-based group therapy intervention. Chart review was conducted for 65 youth, aged 5 to 13 years over one academic year at an elementary school serving high-needs youth in an urban district. Analysis revealed symptom reduction for 72% of participants. For a subset of students ( n = 34) with available Child and Adolescent Needs and Strengths (CANS) Assessments, the average number of adverse experiences was 2.74 ( SD: 2.11). Qualitative coding of progress notes and discharge summaries was used to assess symptom presentation and treatment outcomes. This sports-based group therapy intervention promotes access to services while minimizing stigma.