Health information systems (HIS) serve as the cornerstone of modern healthcare, seamlessly weaving data into actionable insights and empowering professionals to make informed decisions and elevate patient care. Decision support systems became a prominent research area in the discipline of HIS, where clinical decision support systems (CDSSs) with the ability to diagnose and prognosis of disease are recognised as the most common and beneficial information systems. Moreover, the high prevalence and adverse effects of diabetes all over the globe evidently make it vital to predict diabetes and its complications. The current study aims to resolve an issue at Te Whatu Ora, by generating a CDSS which can predict the complications of diabetes mellitus (CoDM) while answering the knowledge gap. Although a rich data set of diabetes patients’ is maintained at Te Whatu Ora, their involvement in decision-making is unsatisfactory. This study created a CDSS to resolve the issue at Te Whatu Ora while considering two perspectives of the question: design and data analysis. The system design followed design science research methodologies (DSRM) while selecting suitable techniques in the steps of the empirical cycle to confirm their applicability in the domain. The data analysis perspective of the study focused on survival analysis methods due to their appropriateness in fulfilling the identified research gaps. The created CDSS is the primary outcome of the research, which resolves the real-world issue while addressing the recognised research gaps. The solution's design perspective confirms the applicability of adopting design science research approaches in the context of a systematic solution-design process. The data analytics perspective confirms the appropriateness of survival techniques in the domain while validating the system’s performance. The outcome of this research has significant academical, and managerial implications. The implemented CDSS is capable of providing a chronological risk percentage for 10 CoDM in a cohort of New Zealand. The systematic procedure adopted in the research contributes to the existing knowledge gaps while answering the design, implementation, and evaluation stages. The managerial implications of the study expand through policy-makers, resource-allocators, iii and healthcare administrators to doctors, nurses, and patients. The predicted risk of CoDM may be beneficial in managing the patients by issuing early warnings, starting treatment plans, conducting diagnosis tests, recommending dietary/exercise routines and more from the perspective of patient care. The visualisation of the statistical details and the survival curves in the system may assist in managing diabetes data repositories efficiently. This study creates a cohort-specific risk prediction model based on the New Zealand cohort. The interoperability challenges in the CDSS could occur due to the variety of practices in real life. Future research studies in this domain can concentrate on increasing the accuracy of the models with more features of a rich dataset while protecting the patents’ confidentiality. Additionally, the cohort specificity of the CDSS can be avoided with the engagement of a global dataset.