1
|
Xingwei W, Huan C, Mengting L, Lv Q, Jiaying Z, Enwu L, Jiuqun Z, Rongsheng T. A machine learning-based risk warning platform for potentially inappropriate prescriptions for elderly patients with cardiovascular disease. Front Pharmacol 2022; 13:804566. [PMID: 36034817 PMCID: PMC9402906 DOI: 10.3389/fphar.2022.804566] [Citation(s) in RCA: 4] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/29/2021] [Accepted: 07/11/2022] [Indexed: 11/17/2022] Open
Abstract
Potentially inappropriate prescribing (PIP), including potentially inappropriate medications (PIMs) and potential prescribing omissions (PPOs), is a major risk factor for adverse drug reactions (ADRs). Establishing a risk warning model for PIP to screen high-risk patients and implementing targeted interventions would significantly reduce the occurrence of PIP and adverse drug events. Elderly patients with cardiovascular disease hospitalized at the Sichuan Provincial People’s Hospital were included in the study. Information about PIP, PIM, and PPO was obtained by reviewing patient prescriptions according to the STOPP/START criteria (2nd edition). Data were divided into a training set and test set at a ratio of 8:2. Five sampling methods, three feature screening methods, and eighteen machine learning algorithms were used to handle data and establish risk warning models. A 10-fold cross-validation method was employed for internal validation in the training set, and the bootstrap method was used for external validation in the test set. The performances were assessed by area under the receiver operating characteristic curve (AUC), and the risk warning platform was developed based on the best models. The contributions of features were interpreted using SHapley Additive ExPlanation (SHAP). A total of 404 patients were included in the study (318 [78.7%] with PIP; 112 [27.7%] with PIM; and 273 [67.6%] with PPO). After data sampling and feature selection, 15 datasets were obtained and 270 risk warning models were built based on them to predict PIP, PPO, and PIM, respectively. External validation showed that the AUCs of the best model for PIP, PPO, and PIM were 0.8341, 0.7007, and 0.7061, respectively. The results suggested that angina, number of medications, number of diseases, and age were the key factors in the PIP risk warning model. The risk warning platform was established to predict PIP, PIM, and PPO, which has acceptable accuracy, prediction performance, and potential clinical application perspective.
Collapse
Affiliation(s)
- Wu Xingwei
- Personalized Drug Therapy Key Laboratory of Sichuan Province, Department of Pharmacy, Sichuan Provincial People’s Hospital, School of Medicine, University of Electronic Science and Technology of China, Chengdu, Sichuan, China
- Chinese Academy of Sciences Sichuan Translational Medicine Research Hospital, Chengdu, Sichuan, China
| | - Chang Huan
- Personalized Drug Therapy Key Laboratory of Sichuan Province, Department of Pharmacy, Sichuan Provincial People’s Hospital, School of Medicine, University of Electronic Science and Technology of China, Chengdu, Sichuan, China
| | - Li Mengting
- Personalized Drug Therapy Key Laboratory of Sichuan Province, Department of Pharmacy, Sichuan Provincial People’s Hospital, School of Medicine, University of Electronic Science and Technology of China, Chengdu, Sichuan, China
| | - Qin Lv
- Department of Pulmonary and Critical Care Medicine, Sichuan Academy of Medical Sciences and Sichuan Provincial People’s Hospital, Chengdu, China
| | - Zhang Jiaying
- Department of Western Pharmacy, Chengdu First People’s Hospital, Chengdu, China
| | - Long Enwu
- Personalized Drug Therapy Key Laboratory of Sichuan Province, Department of Pharmacy, Sichuan Provincial People’s Hospital, School of Medicine, University of Electronic Science and Technology of China, Chengdu, Sichuan, China
- Chinese Academy of Sciences Sichuan Translational Medicine Research Hospital, Chengdu, Sichuan, China
| | - Zhu Jiuqun
- Personalized Drug Therapy Key Laboratory of Sichuan Province, Department of Pharmacy, Sichuan Provincial People’s Hospital, School of Medicine, University of Electronic Science and Technology of China, Chengdu, Sichuan, China
- Chinese Academy of Sciences Sichuan Translational Medicine Research Hospital, Chengdu, Sichuan, China
- *Correspondence: Zhu Jiuqun, ; Tong Rongsheng,
| | - Tong Rongsheng
- Personalized Drug Therapy Key Laboratory of Sichuan Province, Department of Pharmacy, Sichuan Provincial People’s Hospital, School of Medicine, University of Electronic Science and Technology of China, Chengdu, Sichuan, China
- Chinese Academy of Sciences Sichuan Translational Medicine Research Hospital, Chengdu, Sichuan, China
- *Correspondence: Zhu Jiuqun, ; Tong Rongsheng,
| |
Collapse
|
2
|
Xingwei W, Huan C, Mengting L, Lv Q, Jiaying Z, Enwu L, Jiuqun Z, Rongsheng T. A machine learning-based risk warning platform for potentially inappropriate prescriptions for elderly patients with cardiovascular disease. Front Pharmacol 2022. [PMID: 36034817 DOI: 10.3389/fphar.2022.804566.ecollection] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Key Words] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 05/05/2023] Open
Abstract
Potentially inappropriate prescribing (PIP), including potentially inappropriate medications (PIMs) and potential prescribing omissions (PPOs), is a major risk factor for adverse drug reactions (ADRs). Establishing a risk warning model for PIP to screen high-risk patients and implementing targeted interventions would significantly reduce the occurrence of PIP and adverse drug events. Elderly patients with cardiovascular disease hospitalized at the Sichuan Provincial People's Hospital were included in the study. Information about PIP, PIM, and PPO was obtained by reviewing patient prescriptions according to the STOPP/START criteria (2nd edition). Data were divided into a training set and test set at a ratio of 8:2. Five sampling methods, three feature screening methods, and eighteen machine learning algorithms were used to handle data and establish risk warning models. A 10-fold cross-validation method was employed for internal validation in the training set, and the bootstrap method was used for external validation in the test set. The performances were assessed by area under the receiver operating characteristic curve (AUC), and the risk warning platform was developed based on the best models. The contributions of features were interpreted using SHapley Additive ExPlanation (SHAP). A total of 404 patients were included in the study (318 [78.7%] with PIP; 112 [27.7%] with PIM; and 273 [67.6%] with PPO). After data sampling and feature selection, 15 datasets were obtained and 270 risk warning models were built based on them to predict PIP, PPO, and PIM, respectively. External validation showed that the AUCs of the best model for PIP, PPO, and PIM were 0.8341, 0.7007, and 0.7061, respectively. The results suggested that angina, number of medications, number of diseases, and age were the key factors in the PIP risk warning model. The risk warning platform was established to predict PIP, PIM, and PPO, which has acceptable accuracy, prediction performance, and potential clinical application perspective.
Collapse
Affiliation(s)
- Wu Xingwei
- Personalized Drug Therapy Key Laboratory of Sichuan Province, Department of Pharmacy, Sichuan Provincial People's Hospital, School of Medicine, University of Electronic Science and Technology of China, Chengdu, Sichuan, China
- Chinese Academy of Sciences Sichuan Translational Medicine Research Hospital, Chengdu, Sichuan, China
| | - Chang Huan
- Personalized Drug Therapy Key Laboratory of Sichuan Province, Department of Pharmacy, Sichuan Provincial People's Hospital, School of Medicine, University of Electronic Science and Technology of China, Chengdu, Sichuan, China
| | - Li Mengting
- Personalized Drug Therapy Key Laboratory of Sichuan Province, Department of Pharmacy, Sichuan Provincial People's Hospital, School of Medicine, University of Electronic Science and Technology of China, Chengdu, Sichuan, China
| | - Qin Lv
- Department of Pulmonary and Critical Care Medicine, Sichuan Academy of Medical Sciences and Sichuan Provincial People's Hospital, Chengdu, China
| | - Zhang Jiaying
- Department of Western Pharmacy, Chengdu First People's Hospital, Chengdu, China
| | - Long Enwu
- Personalized Drug Therapy Key Laboratory of Sichuan Province, Department of Pharmacy, Sichuan Provincial People's Hospital, School of Medicine, University of Electronic Science and Technology of China, Chengdu, Sichuan, China
- Chinese Academy of Sciences Sichuan Translational Medicine Research Hospital, Chengdu, Sichuan, China
| | - Zhu Jiuqun
- Personalized Drug Therapy Key Laboratory of Sichuan Province, Department of Pharmacy, Sichuan Provincial People's Hospital, School of Medicine, University of Electronic Science and Technology of China, Chengdu, Sichuan, China
- Chinese Academy of Sciences Sichuan Translational Medicine Research Hospital, Chengdu, Sichuan, China
| | - Tong Rongsheng
- Personalized Drug Therapy Key Laboratory of Sichuan Province, Department of Pharmacy, Sichuan Provincial People's Hospital, School of Medicine, University of Electronic Science and Technology of China, Chengdu, Sichuan, China
- Chinese Academy of Sciences Sichuan Translational Medicine Research Hospital, Chengdu, Sichuan, China
| |
Collapse
|
3
|
Ioffe M, Kremer A, Nachimov I, Swartzon M, Justo D. Mortality associated with stopping statins in the oldest-old - with and without ischemic heart disease. Medicine (Baltimore) 2021; 100:e26966. [PMID: 34664827 PMCID: PMC8448040 DOI: 10.1097/md.0000000000026966] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.7] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 12/13/2019] [Accepted: 08/01/2021] [Indexed: 12/15/2022] Open
Abstract
The association between stopping statins and 1-year mortality in the general population of the oldest-old - with or without ischemic heart disease (IHD) - has been studied herein for the first time.This was a retrospective study. Included were all consecutive patients (n = 369) aged 80 years or more (mean age 87.8 years) hospitalized in a single Geriatrics department during 1 year. The study group included 140 patients in whom statins were stopped upon admission (statin stoppers). The control group included 229 patients who did not use statins in the first place (statin non-users). All-cause 1-year mortality rates were studied in both groups following propensity score matching and in IHD patients separately.Overall, 110 (29.8%) patients died during the year following admission: 38 (27.1%) statin stoppers and 72 (31.4%) statin non-users (P = .498). Cox regression analysis showed no association between stopping statins and 1-year mortality in the crude analysis (hazard ratio [HR] 0.976, 95% confidence interval [CI] 0.651-1.463, P = .907) and following propensity score matching (HR 1.067, 95%CI 0.674-1.689, P = .782). Among 108 IHD patients, 38 (35.2%) patients died during the year following admission: 18 (27.7%) statin stoppers and 20 (46.5%) statin non-users (P = .059). Cox regression analysis showed a nearly significant association between stopping statins (rather than not using statins) in IHD patients and lower 1-year mortality (HR 0.524, 95%CI 0.259-1.060, P = .072).Hence, stopping statins in the general population of the oldest-old - with or without IHD - is possibly safe. Future studies including the oldest-old statin continuers are warranted to confirm this observation.
Collapse
Affiliation(s)
- Marina Ioffe
- Geriatrics Division, Sheba Medical Center, Tel-Hashomer, Israel
| | - Anjelika Kremer
- Geriatrics Division, Sheba Medical Center, Tel-Hashomer, Israel
| | | | | | - Dan Justo
- Geriatrics Division, Sheba Medical Center, Tel-Hashomer, Israel
- Sackler School of Medicine, Tel-Aviv University, Israel
| |
Collapse
|
4
|
Kuo TT, Huang YB, Hsieh CJ. Consumption and market share of cholesterol-lowering drugs in high-risk patients before and after the release of the 2013 ACC/AHA cholesterol guidelines: a retrospective observational study. BMJ Open 2020; 10:e036769. [PMID: 33444173 PMCID: PMC7682459 DOI: 10.1136/bmjopen-2020-036769] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 12/03/2022] Open
Abstract
OBJECTIVE We examined the use of cholesterol-lowering drugs in Taiwan in high-risk patients before and after the release of the 2013 American College of Cardiology and the American Heart Association (ACC/AHA) cholesterol guidelines. DESIGN Retrospective observational study. SETTING Kaohsiung Chang Gung Memorial Hospital database, Kaohsiung City, Taiwan. PARTICIPANTS Outpatients aged ≥20 years with atherosclerosis cardiovascular disease, familial hypercholesterolaemia and diabetes. PRIMARY AND SECONDARY OUTCOME MEASURES Data on brand and generic names, use and dosage of cholesterol-lowering drugs in 2012 and 2015 were compiled and the total amount used was calculated. Differences in usage and market share were compared. Usage rates of single and fixed-dose combination (FDC) products were compared. RESULTS The number of patients receiving ambulatory care increased from 36 367 in 2012 to 41 807 in 2015. Single (3 679 979-4 568 086 tablets) and FDC (540 522-572 954 tablets) product use increased from 2012 to 2015, respectively. Statins were the most commonly prescribed medications in 2012 (71.14%) and 2015 (72.91%). The average monthly consumption of statin among high-risk patients in 2012 was 269 948.8 tablets, and it increased significantly to 343 975.3 tablets in 2015. The average monthly consumption of pitavastatin was 34 113.4 tablets in 2015, which was significantly higher than 0 in 2012. Conversely, the highest decline was observed for fluvastatin use, with the average monthly consumption being 38 754.3 tablets in 2015, which was significantly lower than 45 929.8 tablets consumed in 2012. Regarding FDC therapy for cholesterol-lowering drugs, Vytorin (ezetimibe 10 mg + simvastatin 20 mg) use was the highest among all FDCs in 2015. CONCLUSIONS The 2013 ACC/AHA cholesterol guidelines likely promoted the use of fixed-dose, high-intensity and moderate-intensity monotherapy and FDC therapy statins in high-risk groups, and this was consistent with the use of high-intensity or moderate-intensity statins in the present study. Furthermore, these changes were associated with increased effectiveness and reduced adverse effects.
Collapse
Affiliation(s)
- Tzu-Tsen Kuo
- School of Pharmacy, Kaohsiung Medical University, Kaohsiung, Taiwan
- Department of Pharmacy, Kaohsiung Chang Gung Memorial Hospital, Kaohsiung, Taiwan
| | - Yaw-Bin Huang
- School of Pharmacy, Kaohsiung Medical University, Kaohsiung, Taiwan
| | - Ching-Jung Hsieh
- Division of Endocrinology and Metabolism, Department of Internal Medicine, Pao Chien Hospital, Pingtung, Taiwan
| |
Collapse
|
5
|
Abstract
Statin drugs reduce low-density lipoprotein (LDL)-cholesterol (LDL-C) and cardiovascular risk. Ezetimibe may be used to supplement statin therapy, or used alone in cases of statin intolerance. Statin-associated side effects do occur, especially muscle symptoms and new onset diabetes, but they do not detract from the benefits of statin therapy. Inhibitors of proprotein convertase subtilisin/kexin type 9 (PCSK9) reduce LDL-C and cardiovascular risk. Evolocumab is subsidised in Australia for patients with familial hypercholesterolaemia when LDL-C is not adequately controlled with maximum doses of statin or ezetimibe or when statin therapy is contraindicated. Fenofibrate reduces triglycerides and cardiovascular risk in patients with type 2 diabetes when triglycerides are elevated and high-density lipoprotein (HDL) is low. A role for dietary omega-3 fatty acids and esters in reducing cardiovascular risk remains controversial. All cases of secondary cardiovascular disease prevention merit intensive lipid therapy, unless a contraindication exists. Lipid therapy is justified in cases of primary prevention when absolute risk is high, especially when lipids are highly elevated or when multiple risk factors are present. Clinical management requires a focus on the predominant lipid disorder present, namely hypercholesterolaemia, hypertriglyceridaemia or combined hyperlipidaemia. There is an ongoing problem of poor long term persistence on lipid therapy, as well as reduced awareness by practitioners of poor risk factor control.
Collapse
Affiliation(s)
- Leon A Simons
- UNSW Sydney, Sydney, NSW.,St Vincent's Hospital, Sydney, NSW
| |
Collapse
|
6
|
Duntas LH, Brenta G. A Renewed Focus on the Association Between Thyroid Hormones and Lipid Metabolism. Front Endocrinol (Lausanne) 2018; 9:511. [PMID: 30233497 PMCID: PMC6129606 DOI: 10.3389/fendo.2018.00511] [Citation(s) in RCA: 86] [Impact Index Per Article: 14.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 05/24/2018] [Accepted: 08/15/2018] [Indexed: 12/19/2022] Open
Abstract
Thyroid dysfunction, manifesting as either overt or subclinical hypothyroidism, negatively affects lipid metabolism: this leads to hypercholesterolemia which progressively increases the risk for cardiovascular disease and, potentially, mortality. Hypercholesterolemia in hypothyroidism is mainly due to a reduction in low-density lipoprotein (LDL) receptor activity, this accompanied by concomitant diminishing control by triiodothyronine (T3) of sterol regulatory element-binding protein 2 (SREBP-2), which modulates cholesterol biosynthesis by regulating rate-limit degrading enzyme 3-hydroxy-3-methylglutaryl-coenzyme A reductase (HMG-CoA) activity. Recently, 3,5-diiodothyronine (T2), a natural thyroid hormone derivative, was found to repress the transcription factor carbohydrate-response element-binding protein (ChREBP) and also to be involved in lipid catabolism and lipogenesis, though via a different pathway than that of T3. While thyroid hormone could therapeutically reverse the dyslipidemic profile commonly occurring in hypothyroidism, it should be borne in mind that the potency of the effects may be age-and sex-dependent. Thyroid hormone administration possibly also sustains and enhances the efficacy of hypolipidemic drugs, such as statins, ezetimibe and proprotein convertase subtilisin/kexin type 9 (PCSK9), in patients with dyslipidemia and hypothyroidism.
Collapse
Affiliation(s)
- Leonidas H. Duntas
- Unit of Endocrinology Diabetes and Metabolism, Thyroid Section, Evgenideion Hospital, University of Athens, Athens, Greece
- *Correspondence: Leonidas H. Duntas
| | - Gabriela Brenta
- Unit of Endocrinology and Metabolism, Thyroid Section, Dr. Cesar Milstein Hospital, Buenos Aires, Argentina
| |
Collapse
|
7
|
Gazzola K, Vigna GB. Hypolipidemic drugs in elderly subjects: Indications and limits. Nutr Metab Cardiovasc Dis 2016; 26:1064-1070. [PMID: 27522161 DOI: 10.1016/j.numecd.2016.07.008] [Citation(s) in RCA: 4] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 03/31/2016] [Revised: 06/30/2016] [Accepted: 07/13/2016] [Indexed: 01/10/2023]
Abstract
AIMS Cardiovascular disease is a major cause of death worldwide. Safety and efficacy of lipid lowering therapy have been clearly established for either primary and secondary prevention of cardiovascular events in adults. Nevertheless, the use of hypolipidemic drugs in elderly individuals, especially in the oldest ones, still raises some concerns. Aim of this paper is to review indications and limits of lipid lowering in advanced age, furnishing a practical medical attitude tempered by clinical and geriatric competences. DATA SYNTHESIS While figures from randomized controlled trials and from observational studies seem to support the use of lipid lowering drugs for secondary prevention in the elderly, drawing inferences from primary prevention in old populations is far more challenging. Although these pharmacological agents seem to reduce the incidence of cardiovascular events, they do not prolong survival. In addition, there is some doubt about the cost-effectiveness of treatment because of a more delicate balance between benefit and potential adverse reactions. However, lipid-lowering drugs seem largely underutilized in older age, mainly due to safety concerns that must be reconsidered, at least in part, given the somewhat reassuring results deriving from specific cohort surveys. CONCLUSIONS Data on the use and on the effects of lipid lowering drugs in elderly populations are incomplete, especially those concerning very old subjects without established cardiovascular disease. Comprehensive guidelines for the management of dyslipidemias in this rapidly-growing population is a urgent need, and treatment should be based, besides the aforementioned considerations, on patient preferences, cognitive function and life expectancy.
Collapse
Affiliation(s)
- K Gazzola
- Dipartimento Medico, Azienda Ospedaliero-Universitaria di Ferrara, Italy
| | - G B Vigna
- Dipartimento Medico, Azienda Ospedaliero-Universitaria di Ferrara, Italy.
| |
Collapse
|
8
|
Intensive LDL Reduction Post Acute Coronary Syndromes: A Catalyst for Improved Outcomes. Heart Lung Circ 2016; 25:1051-1054. [PMID: 27725093 DOI: 10.1016/j.hlc.2016.09.005] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/03/2016] [Accepted: 09/16/2016] [Indexed: 11/23/2022]
|
9
|
Petersen JF, Larsen BS, Sabbah M, Nielsen OW, Kumarathurai P, Sajadieh A. Long-term prognostic significance of homocysteine in middle-aged and elderly. Biomarkers 2016; 21:490-6. [PMID: 27008914 DOI: 10.3109/1354750x.2016.1160288] [Citation(s) in RCA: 9] [Impact Index Per Article: 1.1] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/13/2022]
Abstract
OBJECTIVE We investigated the association among increased levels of plasma homocysteine (Hcy), all-cause mortality, and cardiovascular events. METHODS Hcy was measured in 670 middle-aged and elderly subjects with no previous manifest cardiovascular disease. The follow-up period was 15 years. RESULTS Subjects with Hcy ≥ 10.8 μmol/l (n = 231) had a significant higher incidence of all-cause mortality (p < 0.001) and CV events (p < 0.001) compared with subjects with Hcy < 10.8 μmol/l (n = 439). However, there was no association on high levels of Hcy and VTE events or stroke. CONCLUSION Increased levels of Hcy are associated with all-cause mortality and CV events.
Collapse
Affiliation(s)
| | | | - Muhammad Sabbah
- a Copenhagen University Hospital of Bispebjerg , Copenhagen , NV , Denmark
| | | | | | - Ahmad Sajadieh
- a Copenhagen University Hospital of Bispebjerg , Copenhagen , NV , Denmark
| |
Collapse
|
10
|
Development of a new highly standardized and granulated extract from Monascus purpureus with a high content of monacolin K and KA and free of inactive secondary monacolins and citrinin. ACTA ACUST UNITED AC 2015. [DOI: 10.1007/s13749-015-0047-4] [Citation(s) in RCA: 8] [Impact Index Per Article: 0.9] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/26/2022]
|