| Objectives:The aim of the study was to achieve external validation of the prediction model for coronary heart disease(CHD)patients’ secondary medication adherence,and verify the effects in improving patients’ medication adherence by using this prediction model to give target intervention on CHD patients.Methods:This study was conducted in two general hospitals in Baoding from June to August 2018.Through continuous fixed-point sampling,165 CHD patients who met the inclusion and exclusion criteria were recruited in this study.Participants were assigned to compliance group(66 cases)and non-compliance(99 cases)by the predicted scores from the prediction model for coronary heart disease(CHD)patients’ secondary medication adherence.Then,noncompliance group was randomly divided into intervention group(50 cases)and control group(49 cases).The compliance group and control group only accepted the routine health education during hospitalization.The intervention group received standardized interventions for influencing factors identified from the prediction model,except for routine health education during hospitalization.All participants’ medication adherence was assessed by telephone follow up at third-month and sixth-month after discharge.The analysis was conducted with SPSS 21.0.The medication adherence of patients was described in mean and standard.The differences between the predictive value and the actual value at the 3rd and the 6th month after discharge both of the compliance group and the control group were compared by the Kappa coefficient.The normal mean square error(NMSE)was also calculated to verify the accuracy of the prediction model by R statistical software.The effect of the intervention was examined by comparing the medication adherence and the constituent ratio of the medication adherence between the control group and intervention group.P<0.05 was considered statistically significant.Results:1.At the 3rd month after discharge,15 patients were drop in this study,and the completion rate of follow-up was 90.91%.At the 6th month after discharge,10 patients were drop,and the completion rate of follow-up was 84.85%.After the end of the 6th month follow-up,53 patients were in the compliance group,44 were in the intervention group,and 43 were in the control group.2.The socio-characteristics and clinical characteristics between the intervention group and the control group patients before intervention was not statistically significant(P > 0.05).3.For compliance group patients,the predicted value of medication adherence was(16.111±0.793),and 53 participants were compliance.The actual scores of medication adherence was(17.15±0.744)at 3rd month after discharge,and all were compliance.Then,the scores of medication adherence was(17.08 ± 1.124)at 6th month,and 51 were compliance,and 2 were non-compliance.For control group patients,the predicted value of medication adherence was(12.910±1.925),and 43 were non-compliance.The scores of medication adherence was(14.72±2.197)at 3rd month after discharge,and 26 were compliance,and 17 were non-compliance.The scores of medication adherence was(14.88±3.935)at 6rd month,and 31 were compliance,and 12 were non-compliance.The Kappa coefficient was examined to comparing the predicted value and the actual value of medication adherence both of the compliance group and the control group.And the Kappa coefficient in 3rd month and in 6th month was 0.419 and 0.258 respectively.After patients who had review were excluded,the Kappa coefficient was examined.The Kappa coefficient was 0.635 and 0.308 respectively.4.The NMSE of the compliance group and the control group at the 3rd month and the 6th month were 0.497 and 0.681,respectively.5.The predicted medication adherence of the intervention group was(13.320±1.741),and the 3rd month was(16.20±1.850)and the 6th month was(16.09±3.277).The actual medication adherence was significantly higher than the predicted value at the 3rd month and the 6th month(P<0.05).6.The constituent ratio of compliance was 93.2% at the 3rd month and 90.9% at the 6th month in the intervention group after the prediction.The ratio was 60.5% at the 3rd month and 72.1% at the 6th month in the control group after the prediction.The differences were significant in the compliance component ratio between the intervention group and the control group at the 3rd month and the 6th month(P<0.05).Conclusions:The prediction model for coronary heart disease(CHD)patients’ secondary medication adherence could effectively predict medication adherence of the 3rd month and the 6th month.Larger sample and multi-center data need to be collected to improve the prediction performance,in order to construct a medication adherence prediction model with higher prediction performance.CHD patients were given individualized and targeted interventions on influencing their medication adherence risk factors based on this prediction model.It does significantly guide in improving CHD patients’ actual medication adherence. |
PDF Full Text Request