Detection of Personal and Family History of Suicidal Thoughts and Behaviors using Deep Learning and Natural Language Processing: A Multi-Site Study.

Objective: Personal and family history of suicidal thoughts and behaviors (PSH and FSH, respectively) are significant risk factors associated with future suicide events. These are often captured in narrative clinical notes in electronic health records (EHRs). Collaboratively, Weill Cornell Medicine (WCM), Northwestern Medicine (NM), and the University of Florida (UF) developed and validated deep learning (DL)-based natural language processing (NLP) tools to detect PSH and FSH from such notes. The tool's performance was further benchmarked against a method relying exclusively on ICD-9/10 diagnosis codes. Materials and Methods: We developed DL-based NLP tools utilizing pre-trained transformer models Bio_ClinicalBERT and GatorTron, and compared them with expert-informed, rule-based methods. The tools were initially developed and validated using manually annotated clinical notes at WCM. Their portability and performance were further evaluated using clinical notes at NM and UF. Results: The DL tools outperformed the rule-based NLP tool in identifying PSH and FHS. For detecting PSH, the rule-based system obtained an F1-score of 0.75 ± 0.07, while the Bio_ClinicalBERT and GatorTron DL tools scored 0.83 ± 0.09 and 0.84 ± 0.07, respectively. For detecting FSH, the rule-based NLP tool's F1-score was 0.69 ± 0.11, compared to 0.89 ± 0.10 for Bio_ClinicalBERT and 0.92 ± 0.07 for GatorTron. For the gold standard corpora across the three sites, only 2.2% (WCM), 9.3% (NM), and 7.8% (UF) of patients reported to have an ICD-9/10 diagnosis code for suicidal thoughts and behaviors prior to the clinical notes report date. The best performing GatorTron DL tool identified 93.0% (WCM), 80.4% (NM), and 89.0% (UF) of patients with documented PSH, and 85.0%(WCM), 89.5%(NM), and 100%(UF) of patients with documented FSH in their notes. Discussion: While PSH and FSH are significant risk factors for future suicide events, little effort has been made previously to identify individuals with these history. To address this, we developed a transformer based DL method and compared with conventional rule-based NLP approach. The varying effectiveness of the rule-based tools across sites suggests a need for improvement in its dictionary-based approach. In contrast, the performances of the DL tools were higher and comparable across sites. Furthermore, DL tools were fine-tuned using only small number of annotated notes at each site, underscores its greater adaptability to local documentation practices and lexical variations. Conclusion: Variations in local documentation practices across health care systems pose challenges to rule-based NLP tools. In contrast, the developed DL tools can effectively extract PSH and FSH information from unstructured clinical notes. These tools will provide clinicians with crucial information for assessing and treating patients at elevated risk for suicide who are rarely been diagnosed.

Association between first-line antidepressant use and risk of dementia in older adults: a retrospective cohort study.

BACKGROUND: Prior studies suggested that antidepressant use is associated with an increased risk of dementia compared to no use, which is subject to confounding by indication. We aimed to compare the dementia risk among older adults with depression receiving first-line antidepressants (i.e., SSRI/SNRI) versus psychotherapy, which is also considered the first-line therapy for depression. METHODS: This retrospective cohort study was conducted using the US Medical Expenditure Panel Survey from 2010 to 2019. We included adults aged ≥ 50 years diagnosed with depression who initiated SSRI/SNRI or psychotherapy. We excluded patients with a dementia diagnosis before the first record of SSRI/SNRI use or psychotherapy. The exposure was the patient's receipt of SSRI/SNRI (identified from self-report questionnaires) or psychotherapy (identified from the Outpatient Visits or Office-Based Medical Provider Visits files). The outcome was a new diagnosis of dementia within 2 years (i.e., survey panel period) identified using ICD-9/ICD-10 codes from the Medical Conditions file. Using a multivariable logistic regression model, we reported adjusted odds ratios (aORs) with 95% confidence intervals (CIs). We also conducted subgroup analyses by patient sex, age group, race/ethnicity, severity of depression, combined use of other non-SSRI/SNRI antidepressants, and presence of underlying cognitive impairment. RESULTS: Among 2,710 eligible patients (mean age = 61 ± 8, female = 69%, White = 84%), 89% used SSRIs/SNRIs, and 11% received psychotherapy. The SSRI/SNRI users had a higher crude incidence of dementia than the psychotherapy group (16.4% vs. 11.8%), with an aOR of 1.36 (95% CI = 1.06-1.74). Subgroup analyses yielded similar findings as the main analyses, except no significant association for patients who were aged < 65 years (1.23, 95% CI = 0.93-1.62), male (1.34, 95% CI = 0.95-1.90), Black (0.76, 95% CI = 0.48-1.19), had a higher PHQ-2 (1.39, 95% CI = 0.90-2.15), and had underlying cognitive impairment (1.06, 95% CI = 0.80-1.42). CONCLUSIONS: Our findings suggested that older adults with depression receiving SSRIs/SNRIs were associated with an increased dementia risk compared to those receiving psychotherapy.

Association between Antidepressants and Dementia Risk in Older Adults with Depression: A Systematic Review and Meta-Analysis.

Depression, commonly treated with antidepressants, is associated with an increased risk of dementia, especially in older adults. However, the association between antidepressant use and dementia risk is unclear. We searched for randomized controlled trials and observational studies from PubMed, Embase, and Cochrane on 1 February 2022, restricting to full texts in English. Since dementia is a chronic disease requiring a long induction time, we restricted studies with ≥1 year follow-up. We extracted the relative risk (RR) adjusted for the most variables from each study and evaluated the heterogeneity using I square (I2). The protocol was registered in the PROSPERO International Register of Systematic Reviews (CRD42022338038). We included six articles in the systematic review, of which the sample size ranged from 716 to 141,740, and the median length of follow-up was 5 years. The pooled RR was 1.21 (95% CI = 1.12-1.29) with an I2 of 71%. Our findings suggest that antidepressant use was associated with an increased risk of dementia in older adults with depression, yet moderate to high heterogeneity existed across studies. Future work accounting for the depression progression is needed to differentiate the effect of depression and antidepressants on dementia risk.

Modifiable statin characteristics associated with potential statin-related prescribing cascades.

STUDY OBJECTIVE: Clinicians may prescribe new medications (marker drug) to treat statin-related (index drug) adverse events, constituting a prescribing cascade. We aimed to identify modifiable statin characteristics (intensity and individual statin agents) associated with lower risk of prescribing cascades to inform clinical decisions in the presence of statin-related adverse events. DESIGN: A secondary analysis based on our previous work, a high-throughput sequence symmetry analysis screening for potential statin-related prescribing cascades. DATA SOURCE: MarketScan Commercial and Medicare Supplemental Insurance claims databases between 2005 and 2019. PATIENTS: Adults who initiated a statin between 2007 and 2018, and who were continuously enrolled in the same healthcare plan for at least 720 days before and 360 days after statin initiation. INTERVENTION: Among the previously identified 57 potential prescribing cascades, 42 statin-marker class dyad with a sample size of ≥ 500 were assessed in this study. MEASUREMENTS: We measured patients' baseline characteristics within -360 days of statin initiation and reported by modifiable statin characteristics. We also performed logistic regression and reported the adjusted odds ratios (aOR) with 95% confidence intervals (CI) of modifiable statin characteristics after adjusting for baseline characteristics. MAIN RESULTS: We identified 1,307,867 statin initiators who met the study criteria (21% elderly, 52% female). Compared with patients initiating low-intensity statins, those initiating moderate- or high-intensity statins had significantly greater odds to develop 29 (69%) prescribing cascades, including antidiabetic drugs such as dipeptidyl peptidase 4 (DPP-4) inhibitors (aOR 1.22; 95% CI, 1.11-1.35) and glucagon-like peptide-1 (GLP-1) analogs (aOR 1.31; 95% CI, 1.16-1.47), and opioids (aOR 1.18; 95% CI, 1.13-1.23). Individual statin agent selection also had a differential effect on 34 (81%) of the prescribing cascades. For example, compared with simvastatin initiators, the probability of initiating osmotically acting laxatives was significantly higher for lovastatin initiators (aOR 1.09; 95% CI, 1.03-1.15) and significantly lower in atorvastatin initiators (aOR 0.92; 95% CI, 0.89-0.94). CONCLUSION: Compared with low-intensity statins, high-intensity statins are associated with increased risk in many potential prescribing cascades, while the choice of individual statin agents affects the risk of prescribing cascades bidirectionally.

Cost-effectiveness analysis of monthly, 3-monthly, and 6-monthly long-acting injectable and oral paliperidone in adults with schizophrenia.

BACKGROUND: Paliperidone is among the most cost-effective antipsychotics in adults with schizophrenia, and it has different formulations, including oral paliperidone extended-release (ER) and long-acting injectable (LAI) paliperidone formulations administered every month (PP1M), 3 months (PP3M), or 6 months (PP6M). However, cost-effectiveness analyses comparing different paliperidone formulations were limited. OBJECTIVE: To compare the cost-effectiveness across different paliperidone formulations. METHODS: A Markov model was developed to simulate 1,000 adults aged 40 years with stable schizophrenia transitioning among stable disease-medication adherent, stable disease-medication nonadherent, relapse with hospitalization, relapse with ambulatory care, and death states every 3 months for 5 years. Drug costs were estimated using the prices listed in the Veterans Affairs Federal Supply Schedule, and costs for treating complications were estimated from published studies. All costs were estimated from the US health care system perspective and standardized to 2022 US dollars using the Consumer Price Index Inflation Calculator. Quality-adjusted life-years (QALYs) were estimated using relapse rates from randomized clinical trials and health-related quality of life scores from observational studies. The estimated future costs and QALYs were discounted at 3%. We reported incremental net monetary benefits between alternative formulations at the $50,000 willingness-to-pay (WTP) threshold with a positive value indicating cost-effectiveness. The impact of parameter uncertainty on study outcomes was assessed using 1-way deterministic and probabilistic sensitivity analyses. RESULTS: In adults with schizophrenia stabilized with paliperidone ER, switching to LAI formulations was associated with increased QALY (PP1M = 0.05, PP3M = 0.14, PP6M = 0.15) and increased cost (PP1M = 49,433, PP3M = 26,698, PP6M = 26,147), leading to a negative incremental net monetary benefit (PP1M = -$46,804, PP3M = -$19,508, PP6M = -$18,886) compared with continuing ER. Among LAI formulations, PP6M was cost-saving with the most QALYs gained (cost = $63,277, QALY = 3.731), followed by PP3M (cost = $63,828, QALY = 3.729) and PP1M (cost = $86,563, QALY = 3.638). At the $50,000 WTP threshold, the probabilities for PP1M, PP3M, and PP6M being cost-effective compared with paliperidone ER were 0.4%, 10.2%, and 9.8%, respectively. The probability of PP6M being cost-effective was 92.6% for the PP6M-PP1M pair and 55.2% for the PP6M-PP3M pair, and 91.1% of PP3M use was cost-effective in the PP3M-PP1M pair. The results were generally robust in the sensitivity analyses, even at the $190,000 WTP threshold. CONCLUSIONS: For patients with schizophrenia stabilized with paliperidone ER, switching to LAI formulations was not cost-effective, suggesting the high drug costs for LAI may not justify the improved quality of life within 5 years. Among LAI formulations, PP6M was cost-effective over PP1M and PP3M.

Concomitant use of levothyroxine and interacting medications in U.S. ambulatory care visits.

BACKGROUND: Levothyroxine (LT4) is the third most commonly prescribed medication in the United States. It is a narrow therapeutic index medication, and thus can be impacted by drug-drug interactions, which are primarily available over-the-counter. The prevalence and associated factors with concomitant interacting drugs with LT4 is limited since over-the-counter products are not routinely captured in many drug databases. OBJECTIVE: This study aimed to characterize the concomitant use of LT4 with interacting drugs at ambulatory care visits in the United States. DESIGN: A cross-sectional analysis of the National Ambulatory Medical Care Survey (NAMCS) from 2006 to 2018 was completed. SETTING AND PARTICIPANTS: Ambulatory care visits in the United States involving adult patients with a LT4 prescription were included in the analysis. OUTCOME MEASURES: The primary outcome was initiation or continuation of a selected concomitant interacting drug which impacts LT4 absorption (e.g., proton pump inhibitor) in a patient visit in conjunction with LT4. RESULTS: The authors analyzed 372,942,000 visits (weighted from a sample of 14,880) with a reported LT4 prescription. Concomitant use of interacting drugs with LT4 occurred in 24.4% of visits in which 80% of interacting drugs were proton pump inhibitors. Ages 35-49 years (adjusted odds ratio [aOR], 1.59), 50-64 years (aOR, 2.27), and ≥65 years (aOR, 2.87) compared to 18-34 years, female (aOR 1.37) versus males, and visits in 2014 or later (aOR, 1.27) versus 2006-2009 were associated with increased odds of concomitant interacting drug use in multivariable analysis. CONCLUSION: At ambulatory care visits between 2006 and 2018, concomitant use of LT4 and interacting drugs impacted one-quarter of patient visits. Increased age, females, and visits later in the study period were associated with increased odds for concomitant interacting drugs. Additional work is needed to identify downstream consequences of concomitant use.

Global Epidemiology of Hip Fractures: Secular Trends in Incidence Rate, Post-Fracture Treatment, and All-Cause Mortality.

In this international study, we examined the incidence of hip fractures, postfracture treatment, and all-cause mortality following hip fractures, based on demographics, geography, and calendar year. We used patient-level healthcare data from 19 countries and regions to identify patients aged 50 years and older hospitalized with a hip fracture from 2005 to 2018. The age- and sex-standardized incidence rates of hip fractures, post-hip fracture treatment (defined as the proportion of patients receiving anti-osteoporosis medication with various mechanisms of action [bisphosphonates, denosumab, raloxifene, strontium ranelate, or teriparatide] following a hip fracture), and the all-cause mortality rates after hip fractures were estimated using a standardized protocol and common data model. The number of hip fractures in 2050 was projected based on trends in the incidence and estimated future population demographics. In total, 4,115,046 hip fractures were identified from 20 databases. The reported age- and sex-standardized incidence rates of hip fractures ranged from 95.1 (95% confidence interval [CI] 94.8-95.4) in Brazil to 315.9 (95% CI 314.0-317.7) in Denmark per 100,000 population. Incidence rates decreased over the study period in most countries; however, the estimated total annual number of hip fractures nearly doubled from 2018 to 2050. Within 1 year following a hip fracture, post-hip fracture treatment ranged from 11.5% (95% CI 11.1% to 11.9%) in Germany to 50.3% (95% CI 50.0% to 50.7%) in the United Kingdom, and all-cause mortality rates ranged from 14.4% (95% CI 14.0% to 14.8%) in Singapore to 28.3% (95% CI 28.0% to 28.6%) in the United Kingdom. Males had lower use of anti-osteoporosis medication than females, higher rates of all-cause mortality, and a larger increase in the projected number of hip fractures by 2050. Substantial variations exist in the global epidemiology of hip fractures and postfracture outcomes. Our findings inform possible actions to reduce the projected public health burden of osteoporotic fractures among the aging population. © 2023 The Authors. Journal of Bone and Mineral Research published by Wiley Periodicals LLC on behalf of American Society for Bone and Mineral Research (ASBMR).

Continued potassium supplementation use following loop diuretic discontinuation in older adults: An evaluation of a prescribing cascade relic.

BACKGROUND: The use of a new medication (e.g., potassium supplementation) for managing a drug-induced adverse event (e.g., loop diuretic-induced hypokalemia) constitutes a prescribing cascade. However, loop diuretics are often stopped while potassium may be unnecessarily continued (i.e., relic). We aimed to quantify the occurrence of relics using older adults previously experiencing a loop diuretic-potassium prescribing cascade as an example. METHODS: We conducted a prescription sequence symmetry analysis using the population-based Medicare Fee-For-Service data (2011-2018) and partitioned the 150 days following potassium initiation by day to assess the daily treatment scenarios (i.e., loop diuretics alone, potassium alone, combination of loop diuretics and potassium, or neither). We calculated the proportion of patients developing the relic, proportion of person-days under potassium alone, the daily probability of the relic, and the proportion of patients filling potassium after loop diuretic discontinuation. We also identified the risk factors of the relic. RESULTS: We identified 284,369 loop diuretic initiators who were 8 times more likely to receive potassium supplementation simultaneously or after (i.e., the prescribing cascade), rather than before, loop diuretic initiation (aSR 8.0, 95% CI 7.9-8.2). Among the 66,451 loop diuretic initiators who subsequently (≤30 days) initiated potassium, 20,445 (30.8%) patients remained on potassium after loop diuretic discontinuation, and 9365 (14.1%) patients subsequently filled another potassium supplementation. Following loop diuretic initiation, 4.0% of person-days were for potassium alone, and daily probability of the relic was the highest after day 90 of loop diuretic initiation (5.6%). Older age, female sex, higher diuretic daily dose, and greater baseline comorbidities were risk factors for the relic, while patients having the same prescriber or pharmacy involved in the use of both medications were less likely to experience the relic. CONCLUSIONS: Our findings suggest the need for clinicians to be aware of the potential of relic to avoid unnecessary drug use.

Association between gabapentinoids and oedema treated with loop diuretics: A pooled sequence symmetry analysis from the USA and Denmark.

AIMS: To assess the gabapentinoid-oedema-loop diuretic prescribing cascade in adults using large administrative health care databases from the USA and Denmark. METHODS: This study used a sequence symmetry analysis to assess loop diuretic initiation before and after the initiation of gabapentinoids among patients aged 20 years or older without heart failure or chronic kidney disease. Data from MarketScan Commercial and Medicare Supplemental Claims databases (2005 to 2019) and Danish National Prescription Register (2005 to 2018) were analyzed. Use of loop diuretics associated with initiation of selective norepinephrine reuptake inhibitors (SNRI) was used as a negative control. We assessed the pooled temporality of loop diuretic initiation relative to gabapentinoid or SNRI initiation across the 2 countries. Secular trend-adjusted sequence ratios (aSRs) with 95% confidence intervals (CIs) were calculated using data from 90 days before and after initiation of gabapentinoids. Pooled ratio of aSRs were calculated by comparing gabapentinoids to SNRIs. RESULTS: Among the 1 511 493 gabapentinoid initiators (Denmark [n = 338 941]; USA [n = 1 172 552]), 20 139 patients had a new loop diuretic prescription 90 days before or after gabapentinoid initiation, resulting in a pooled aSR of 1.33 (95% CI 1.06-1.67). The pooled aSR for the negative control (i.e., SNRI) was 0.84 (95% CI 0.75-0.94), which resulted in a pooled ratio of aSRs of 1.58 (95% CI 1.23-2.04). Pooled estimated incidence of the gabapentinoid-loop diuretic prescribing cascade was 8.14 (95% CI, 1.92-34.49) events per 1000 patient-years. CONCLUSION: We identified evidence of the gabapentinoid-oedema-loop diuretic prescribing cascade in 2 countries.

Use of antipsychotic drugs and cholinesterase inhibitors and risk of falls and fractures: self-controlled case series.

OBJECTIVE: To evaluate the association between the use of antipsychotic drugs and cholinesterase inhibitors and the risk of falls and fractures in elderly patients with major neurocognitive disorders. DESIGN: Self-controlled case series. SETTING: Taiwan's National Health Insurance Database. PARTICIPANTS: 15 278 adults, aged ≥65, with newly prescribed antipsychotic drugs and cholinesterase inhibitors, who had an incident fall or fracture between 2006 and 2017. Prescription records of cholinesterase inhibitors confirmed the diagnosis of major neurocognitive disorders; all use of cholinesterase inhibitors was reviewed by experts. MAIN OUTCOME MEASURES: Conditional Poisson regression was used to derive incidence rate ratios and 95% confidence intervals for evaluating the risk of falls and fractures for different treatment periods: use of cholinesterase inhibitors alone, antipsychotic drugs alone, and a combination of cholinesterase inhibitors and antipsychotic drugs, compared with the non-treatment period in the same individual. A 14 day pretreatment period was defined before starting the study drugs because of concerns about confounding by indication. RESULTS: The incidence of falls and fractures per 100 person years was 8.30 (95% confidence interval 8.14 to 8.46) for the non-treatment period, 52.35 (48.46 to 56.47) for the pretreatment period, and 10.55 (9.98 to 11.14), 10.34 (9.80 to 10.89), and 9.41 (8.98 to 9.86) for use of a combination of cholinesterase inhibitors and antipsychotic drugs, antipsychotic drugs alone, and cholinesterase inhibitors alone, respectively. Compared with the non-treatment period, the highest risk of falls and fractures was during the pretreatment period (adjusted incidence rate ratio 6.17, 95% confidence interval 5.69 to 6.69), followed by treatment with the combination of cholinesterase inhibitors and antipsychotic drugs (1.35, 1.26 to 1.45), antipsychotic drugs alone (1.33, 1.24 to 1.43), and cholinesterase inhibitors alone (1.17, 1.10 to 1.24). CONCLUSIONS: The incidence of falls and fractures was high in the pretreatment period, suggesting that factors other than the study drugs, such as underlying diseases, should be taken into consideration when evaluating the association between the risk of falls and fractures and use of cholinesterase inhibitors and antipsychotic drugs. The treatment periods were also associated with a higher risk of falls and fractures compared with the non-treatment period, although the magnitude was much lower than during the pretreatment period. Strategies for prevention and close monitoring of the risk of falls are still necessary until patients regain a more stable physical and mental state.

Global epidemiology of hip fractures: a study protocol using a common analytical platform among multiple countries.

INTRODUCTION: Hip fractures are associated with a high burden of morbidity and mortality. Globally, there is wide variation in the incidence of hip fracture in people aged 50 years and older. Longitudinal and cross-geographical comparisons of health data can provide insights on aetiology, risk factors, and healthcare practices. However, systematic reviews of studies that use different methods and study periods do not permit direct comparison across geographical regions. Thus, the objective of this study is to investigate global secular trends in hip fracture incidence, mortality and use of postfracture pharmacological treatment across Asia, Oceania, North and South America, and Western and Northern Europe using a unified methodology applied to health records. METHODS AND ANALYSIS: This retrospective cohort study will use a common protocol and an analytical common data model approach to examine incidence of hip fracture across population-based databases in different geographical regions and healthcare settings. The study period will be from 2005 to 2018 subject to data availability in study sites. Patients aged 50 years and older and hospitalised due to hip fracture during the study period will be included. The primary outcome will be expressed as the annual incidence of hip fracture. Secondary outcomes will be the pharmacological treatment rate and mortality within 12 months following initial hip fracture by year. For the primary outcome, crude and standardised incidence of hip fracture will be reported. Linear regression will be used to test for time trends in the annual incidence. For secondary outcomes, the crude mortality and standardised mortality incidence will be reported. ETHICS AND DISSEMINATION: Each participating site will follow the relevant local ethics and regulatory frameworks for study approval. The results of the study will be submitted for peer-reviewed scientific publications and presented at scientific conferences.