Natural language processing to identify social determinants of health in Alzheimer's disease and related dementia from electronic health records.

OBJECTIVE: To develop a natural language processing (NLP) algorithm that identifies social determinants of health (SDoH), including housing, transportation, food, and medication insecurities, social isolation, abuse, neglect, or exploitation, and financial difficulties for patients with Alzheimer's disease and related dementias (ADRD) from unstructured electronic health records (EHRs). DATA SOURCES AND STUDY SETTING: We leveraged 1000 medical notes randomly selected from 7401 emergency department and inpatient social worker notes generated between 2015 and 2019 for 231 unique patients diagnosed with ADRD at Michigan Medicine. STUDY DESIGN: We developed a rule-based NLP algorithm for the identification of seven domains of SDoH noted above. We also compared the rule-based algorithm with deep learning and regularized logistic regression approaches. These models were compared using accuracy, sensitivity, specificity, F1 score, and the area under the receiver operating characteristic curve (AUC). All notes were split into 700 notes for training NLP algorithms, and 300 notes for validation. DATA COLLECTION/EXTRACTION METHODS: Social worker notes used in this study were extracted from the Michigan Medicine EHR database. PRINCIPAL FINDINGS: Of the 700 notes for training, F1 and AUC for the rule-based algorithm were at least 0.94 and 0.95, respectively, for all SDoH categories. Of the 300 notes for validation, F1 and AUC were at least 0.80 and 0.97, respectively, for all SDoH except housing and medication insecurities. The deep learning and regularized logistic regression algorithms had unsatisfactory performance. CONCLUSIONS: The rule-based algorithm can accurately extract SDoH information in all seven domains of SDoH except housing and medication insecurities. Findings from the algorithm can be used by clinicians and social workers to proactively address social needs of patients with ADRD and other vulnerable patient populations.

Epidemiology of hydrogen phosphide exposures in humans reported to the poison center in Mainz, Germany, 1983-2003.

BACKGROUND: Poisonings with rodenticides containing hydrogen phosphide-releasing compounds may lead to deleterious organ dysfunction and death. Since data of hydrogen phosphide poisonings is limited to case reports/series, this study was intended to elucidate hydrogen phosphide poisonings based on a 20-year data collection. METHODS: Explorative data analysis of the Poison Center Mainz database looking for route of exposure, symptoms, and severity using the Poisoning Severity Score. RESULTS: From 1983-2003, 188 hydrogen phosphide poisonings were reported. Sixty-five percent of these were unintentional residential, 28% attempts to commit suicide (intentional), 5% occupational, and 2% undetermined. In the majority of intentional poisonings the poison was ingested, whereas in unintentional poisoning of adults inhalation exposure dominated, caused by inappropriate self-protection from the released hydrogen phosphide gas during usage. Frequently observed symptoms in unintentional poisonings were nausea, vomiting, pain, coughing, and dizziness with no further worsening of symptoms. In intentional poisonings frequent symptoms were vomiting, somnolence, seizures, coma, and shock with two initially fatal poisonings. Follow-up on these cases showed a significant worsening of symptoms and a two-fold increase in fatal poisonings. CONCLUSION: Route of exposure, severity of symptoms, and the necessary treatment differs substantially between unintentional and intentional poisonings. In this study, two initially symptomatic intentional poisonings were later reported fatal. Careful monitoring is recommended in symptomatic intentional poisonings.

Experiences of a poison center network with renal insufficiency in acetaminophen overdose: an analysis of 17 cases.

OBJECTIVE: Renal insufficiency is less common than liver failure in acetaminophen overdose but renal tubular damage occurs even in the absence of hepatotoxicity. Data published on this topic are rare consisting mostly of case reports or reports in a small number of patients. Presently, a larger number of patients with renal insufficiency associated with acetaminophen overdose should be analyzed using a multicenter approach. STUDY DESIGN: Retrospective analysis of patients with acetaminophen-related nephrotoxicity reported to a poison center network from 1995 to 2003. Renal insufficiency was defined as elevated serum creatinine of more than double of the normal range (>2.4 mg/dL [212 micromol/L]). Patients were classified into 4 groups (A: creatinine 2.4-5.0 mg/dL, B: creatinine>5.0 mg/dL requiring no dialysis, C: creatinine>5.0 mg/dL requiring dialysis, D: creatinine>5.0 mg/dL with fatal outcome). RESULTS: Seventeen patients were included (8 female, 9 male, average age 31.7 +/- 21.1 yrs) with 6 patients in group A (B: 7, C: 2, D: 2). In 5 patients renal insufficiency occurred without elevation of liver enzymes. Regarding possible risk factors 5 patients concomitantly ingested nephrotoxic substances, 4 presented with dehydration due to vomiting, 4 with chronic excessive dosing (overdose) of acetaminophen, 3 showed pre-existing renal insufficiency, 2 pre-existing liver disease and 2 died with multiple organ failure. CONCLUSIONS: Renal insufficiency in acetaminophen overdose mostly resolved without dialysis and occurred isolated without hepatotoxicity in less than one-third of the investigated patients. Conditions which might play a role as influencing factors for renal complications included concomitant ingestion of nephrotoxic drugs, dehydration, chronic excessive dosing (overdose) of acetaminophen, pre-existing renal or liver disease and multiple organ failure. Renal function should be monitored in acetaminophen overdose particularly in patients showing the latter comorbidity.

[Deliberate self-poisoning with paracetamol (acetaminophen): an analysis from 1995 to 2002]. / Suizidale und parasuizidale Intoxikationen mit Paracetamol - Eine Analyse von 1995 bis 2002.

BACKGROUND AND OBJECTIVE: Paracetamol is frequently used in deliberate self-poisoning resulting in a major risk for the patients due to its dose-dependent hepatotoxicity. In the present study the cases of intoxications consulting our Poison Center should be analysed illustrating recent results and trends. PATIENTS AND METHODS: From 38 065 patients (25 098 female, 12447 male, 520 sex unknown, average age 36.8 years) registered during the study period from 1.1.1995 until 31.5.2002 4021 with paracetamol intoxication were analysed with respect to the ingested dose, concomitant substances, the degree of observed symptoms and the length of hospital stay. RESULTS: The use of paracetamol in deliberate self-poisoning continuously increased during the study period from 8.9 % in the year 1995 to 12.4 % in 2002. Paracetamol was mainly used from female patients and patients in the age group between 10 and 29 years. 88.6 % of cases were reported to the poison center within the first twelve hours after ingestion. Concomitantly non-steroidal antirheumatics (38.0 %), ethanol (20.6 %) and antibiotics (15.0 %) were ingested. Monointoxications as well as mixed intoxications with paracetamol caused severe intoxications and deaths less frequently as compared to the group of the remainder substances with the degree of symptoms tended to be dose-dependent. In 73.7 % of cases the length of hospital stay did not exceed 3 days. CONCLUSIONS: In recent years, an increasing importance of paracetamol in deliberate self-poisoning was determined particularly concerning female and younger patients. The dose of paracetamol and the duration of exposition are crucial for prognosis. With an early antidote therapy hospitalization is usually observed for a few days only. A risk for complications are frequently used concomitant substances requiring a detailed registration.