Background and objectives Drug addiction is a chronic and relapsing brain disease. Practicing self-care is one of the rules offered by therapists to improve the withdrawal process and prevent relapse. Based on the evidence, health beliefs, including Pender’s proposed model, are significantly effective in adopting self-care behaviors in patients. On the other hand, in order to evaluate preventive interventions regarding drug use and relapse of drug use; Having a good and appropriate tool is a special priority. Therefore, the present study aims to design and determine the psychometric characteristics of the questionnaire for measuring health-promoting self-care behaviors in patients recovered from drug addiction has been carried out in the west of Mazandaran province. Methods The current study is a cross-sectional descriptive methodological research. In the first stage, the items and primary classes of the new tool were extracted based on the review of studies consistent with Pender’s self-care promotion and a questionnaire was designed. In the second stage, the psychometric characteristics of the designed questionnaire were examined using face validity, content validity, construct validity and reliability. In the construct validity, the number of participants was 245 for exploratory factor analysis and 203 for confirmatory factor analysis. In the reliability test, 25 people participated to check internal consistency and test-retest. Random sampling was done from 12 drug addiction treatment clinics in two cities of Tonkabon and Ramsar in the west of Mazandaran province during 2021–2022. Data were analyzed using SPSS and AMOS software version 23. Results The exploratory factor analysis revealed 7 factors containing 29 item explained more than 61% of the total observed variance. The results of confirmatory factor analysis showed that the data fit the model. In the reliability test, Cronbach’s alpha coefficient indicated the appropriate internal consistency of the tool and retesting the tool with an interval of 2 weeks showed acceptable stability of the tool. Conclusion The present study showed that the researcher-made questionnaire has good validity and reliability and can be used to measure self-care behaviors of patients who have recovered from addiction in order to provide appropriate solutions to prevent the relapse of drug use.