دورية أكاديمية

Contextualizing the revised Patient Perception of Patient-Centeredness (PPPC-R) scale in primary healthcare settings: a validity and reliability evaluation study

التفاصيل البيبلوغرافية
العنوان: Contextualizing the revised Patient Perception of Patient-Centeredness (PPPC-R) scale in primary healthcare settings: a validity and reliability evaluation study
المؤلفون: Yiyuan Cai, Pengfei Guo, Jiong Tu, Mengyao Hu, Lingrui Liu, Bridget L. Ryan, Jing Liao, Rubee Dev, Yiran Li, Tianyu Huang, Ruilin Wang, Li Kuang, Ruonan Huang, Xinfang Li, Edmundo Roberto Melipillán, Shuaixiang Zhao, Wenjun He, Xiaohui Wang, Nan Zhang, Dong (Roman) Xu
المصدر: BMC Primary Care, Vol 25, Iss 1, Pp 1-13 (2024)
بيانات النشر: BMC, 2024.
سنة النشر: 2024
المجموعة: LCC:Medicine (General)
مصطلحات موضوعية: Patients-Centeredness Care, Quality of health care, Factor analysis, Localization and Validation, Medicine (General), R5-920
الوصف: Abstract Background An English version of the Patient Perception of Patient-Centeredness (PPPC) scale was recently revised, and it is necessary to test this instrument in different primary care populations. Aim This study aimed to assess the validity and reliability of a Chinese version of the PPPC scale. Design A mixed method was used in this study. The Delphi method was used to collect qualitative and quantitative data to address the content validity of the PPPC scale by calculating the Content Validity Index, Content Validity Ratio, the adjusted Kappa, and the Item Impact Score. Confirmatory factor analysis (CFA) and exploratory factor analysis (EFA) were used to assess the construct validity of the PPPC scale through a cross-sectional survey. The internal consistency was also assessed. Setting/participants In the Delphi consultation, seven experts were consulted through a questionnaire sent by email. The cross-sectional survey interviewed 188 outpatients in Guangzhou city and 108 outpatients in Hohhot City from community health service centers or stations face-to-face. Results The 21 items in the scale were relevant to their component. The Item-level Content Validity Index for each item was higher than 0.79, and the average Scale-level content validity index was 0.97 in each evaluation round. The initial proposed 4-factor CFA model did not fit adequately. Still, we found a 3-factor solution based on our EFA model and the validation via the CFA model (model fit: $${\chi }^{2}=294.573$$ χ 2 = 294.573 , P
نوع الوثيقة: article
وصف الملف: electronic resource
اللغة: English
تدمد: 2731-4553
العلاقة: https://doaj.org/toc/2731-4553Test
DOI: 10.1186/s12875-023-02227-x
الوصول الحر: https://doaj.org/article/7176ff0ce135451e91fb09dbca37fddaTest
رقم الانضمام: edsdoj.7176ff0ce135451e91fb09dbca37fdda
قاعدة البيانات: Directory of Open Access Journals
الوصف
تدمد:27314553
DOI:10.1186/s12875-023-02227-x