Construction and validation of a medication deviation prediction model for hospital-to-home transition period in coronary heart disease patients with initial treatment
|更新时间:2026-02-27
|
Construction and validation of a medication deviation prediction model for hospital-to-home transition period in coronary heart disease patients with initial treatment
LI Yushuang,LI Shu,ZHANG Qianying,et al.Construction and validation of a medication deviation prediction model for hospital-to-home transition period in coronary heart disease patients with initial treatment[J].ZHONGGUO YAOFANG,2026,37(04):491-496.
LI Yushuang,LI Shu,ZHANG Qianying,et al.Construction and validation of a medication deviation prediction model for hospital-to-home transition period in coronary heart disease patients with initial treatment[J].ZHONGGUO YAOFANG,2026,37(04):491-496. DOI: 10.6039/j.issn.1001-0408.2026.04.14.
Construction and validation of a medication deviation prediction model for hospital-to-home transition period in coronary heart disease patients with initial treatment
To develope a predictive model for medication deviation risks during the hospital-to-home transition period in coronary heart disease (CHD) patients with initial treatment, aiming to assist medical staff in rapidly identifying high-risk groups for medication deviation.
METHODS
2
A total of 462 CHD patients with initial treatment from the Affiliated Hospital of North China University of Science and Technology (hereinafter referred to as “our hospital”) between January and July 2024 were enrolled. The patients were randomly divided into a modeling group and an internal validation group. The modeling group was further categorized into a medication deviation group and a non-medication deviation group based on whether medication deviations occurred. Similarly, 57 CHD patients with initial treatment from the cardiology department of our hospital between June and September 2025 were collected as an external validation group. Univariate analysis was used to screen predictive factors, followed by multivariate Logistic regression to construct the predictive model. Internal validation methods were employed to evaluate model performance, while external validation methods were used to test the model’s generalizability.
RESULTS
2
The 462 patients were divided into a modeling group (319 cases) and an internal validation group (143 cases). In the modeling group, the medication deviation group (192 cases, 60.19%) and the non-medication deviation group (127 cases, 39.81%) were identified. Multivariate Logistic regression analysis revealed that age, medication type, medication adherence, and self-efficacy in rational medication use were predictive factors for medication deviations in CHD patients with initial treatment (
P
<0.05). The predictive model equation was logit
P
=ln[
P
/(1-
P
)
]
=1.321+1.732×age+4.091×medication type -4.360×medication adherence -3.081×self-efficacy in rational medication use. The model demonstrated good discrimination, with a Hosmer-Lemeshow goodness-of-fit test
P
-value of 0.439, an area under the receiver operating characteristic curve (AUC) of 0.870, sensitivity of 0.970, and specificity of 0.607. A risk nomogram with a total score of 350 points and a cutoff value of 110 points was plotted. The internal validation group showed an AUC o
f 0.787 and a prediction accuracy of 77.6%, while the external validation group exhibited an AUC of 0.802 and a prediction accuracy of 73.7%.
CONCLUSIONS
2
This study successfully developed a predictive model for medication deviation risks during the hospital-to-home transition period in CHD patients with initial treatment. The model demonstrates excellent discrimination and predictive accuracy, effectively identifying high-risk populations for medication deviations. Age (>70 years), number of drug types≥5, poor medication adherence, and poor self-efficacy in rational medication use are independent risk factors for medication deviations.
关键词
Keywords
references
HENRIQUES J , AMARO AM , PIEDADE A P . Biomi-micking atherosclerotic vessels:a relevant and (yet) sub-explored topic [J ] . Biomimetics (Basel) , 2024 , 9 ( 3 ): 135 .
LIU X , TANG L , TANG Y , et al . Closing gaps in medication taking for secondary prevention of coronary heart disease patients among US adults [J ] . Heliyon , 2022 , 8 ( 11 ): e11530 .
GALLITTO G , ENGLERT R , KINCSES B , et al . External validation of machine learning models-registered models and adaptive sample splitting [J ] . Gigascience , 2025 , 14 : giaf036 .
SMITH J D , COLEMAN E A , MIN S J . A new tool for identifying discrepancies in postacute medications for community-dwelling older adults [J ] . Am J Geriatr Pharmacother , 2004 , 2 ( 2 ): 141 - 147 .
RISSER J , JACOBSON T A , KRIPALANI S . Development and psychometric evaluation of the Self-efficacy for Appropriate Medication Use Scale (SEAMS)in low-literacy patients with chronic disease [J ] . J Nurs Meas , 2007 , 15 ( 3 ): 203 - 219 .
VAN DER NAT D J , TAKS M , HUISKES V J B , et al . Risk factors for clinically relevant deviations in patients’ medication lists reported by patients in personal health records:a prospective cohort study in a hospital setting [J ] . Int J Clin Pharm , 2022 , 44 ( 2 ): 539 - 547 .
SAMOILENKO M , LEFEBVRE G . An exact regression-based approach for the estimation of natural direct and indirect effects with a binary outcome and a continuous mediator [J ] . Stat Med , 2023 , 42 ( 3 ): 353 - 387 .
LIU Y H , JIN L J . Research progress of drug treatment adherence in patients with coronary heart disease [J ] . Adv Cardiovasc Med , 2019 , 40 ( 1 ): 113 - 116 .
BAKAN G , INCI F H . Predictor of self-efficacy in individuals with chronic disease:stress-coping strategies [J ] . J Clin Nurs , 2021 , 30 ( 5/6 ): 874 - 881 .
Predictive model for severe adverse reaction associated with bevacizumab based on the global trigger tool and machine learning
Analysis of influencing factors for successful detoxification in patients undergoing methadone maintenance therapy
Analysis of factors influencing the achievement of target vancomycin plasma concentration and construction of a predictive model in patients from high-altitude regions: a single-center retrospective study
MeDIP-Seq 法检测冠心病患者全基因组甲基化谱
硝苯地平联合氯沙坦对高血压合并冠心病患者血压和肾功能的影响
Related Author
FU Yongfei
LONG Xin
XU Hongzhen
TANG Jian
LI Xiangqing
LONG Yucheng
QIN Dong
MO Guodong
Related Institution
College of Pharmacy, Guilin Medical University
Dept. of Pharmacy, Guilin People’s Hospital
Data Center, Guilin People’s Hospital
Zhongshan Second People’s Hospital
School of Public Health, Guangdong Pharmaceutical University