2.1. Study Design

Between March 2012 and December 2015, the study was conducted in 68 MMT clinics in five provinces (Sichuan, Guangdong, Hunan, Shaanxi, and Jiangsu) of China. The intervention trial, entitled MMT CARE, utilized a two-arm design. The participating clinics in each province were matched into pairs based on the number of patients and the affiliation of the clinic (hospital affiliated vs. CDC affiliated). In China, CDC affiliated MMT clinics and hospital affiliated clinics differ in funding resources and staff credentials (Lin et al., 2009). Therefore we considered clinic affiliation in matching to ensure comparability of the clinics in the intervention and the control conditions. Each pair of clinics were randomized into either an intervention or a control condition after baseline assessment, which resulted in 34 clinics in each condition. The short-term and sustained perceived improvement in provider-patient interaction were evaluated at the 6-month follow-up and the end of the study, respectively. The study protocol and materials were approved by the Institutional Review Broads of the participating agencies in China and the United States. This trial was registered in the ClinicalTrials.gov protocol registration system ({"type":"clinical-trial","attrs":{"text":"NCT01760720","term_id":"NCT01760720"}}NCT01760720).