Objectives: This study aims to develop the Directive and Nondirective Support Scale for Patients with Type 2 Diabetes (DNSS-T2DM) to measure diabetes-specific support and patients' preference as well as evaluate the construct validity and reliability of the DNSS-T2DM. Methods: A cross-sectional study was conducted in Tongzhou District, Beijing, China from July to September 2015. A total of 474 participants who had been diagnosed as type 2 diabetes by physicians and completed the DNSS-T2DM were included. The original 11-item DNSS-T2DM contains five items on nondirective support (Items 1-5) and six items on directive support (Items 6-11). There were two parallel questions for each item with one to measure the preference for support (Preference part) and the other to measure the perception of support in reality (Reality part). The final DNSS-T2DM was determined based on the results of the exploratory factor analysis (EFA). The construct validity of the final DNSS-T2DM was evaluated by the confirmatory factor analysis (CFA). The reliability was evaluated by internal consistency with Cronbach's alpha coefficients. Results: A final 7-item DNSS-T2DM loaded on 2 factors with four items representing nondirective support and three items representing directive support was determined based on the EFA. The CFA indicated a satisfactory construct validity. The internal consistency of the 7-item DNSS-T2DM as well as the nondirective support items was satisfactory with Cronbach's alpha > 0.70. Conclusions: Our study supported the validity and reliability of the 7-item DNSS-T2DM. Further studies on the application of the DNSS-T2DM in different settings and population are needed. (C) 2020 The authors. Published by Elsevier B.V. on behalf of the Chinese Nursing Association.