Monitoring healthcare processes poses unique challenges due to the substantial variability in patient risk profiles, which can significantly influence surgical outcomes. Traditional control charts often neglect these individual differences, leading to potentially biased and misleading performance assessments. To overcome these limitations, risk-adjusted control charts have been developed to incorporate patient-specific covariates for more equitable monitoring. This study extends previous approaches by proposing a risk-adjusted cumulative sum (RA-CUSUM) control chart that accommodates ordinal surgical outcomes and incorporates random effects to model unobserved heterogeneity among healthcare providers. The proposed RA-CUSUM chart employs dynamic probability control limits (DPCLs) to maintain a constant conditional false alarm rate, enabling consistent performance across heterogeneous patient populations. Through extensive simulation studies, we demonstrate its efficacy in detecting shifts in surgical performance stability, particularly in response to changes in location and scale. A real-world case study using cardiac surgery data demonstrates the practical applicability of the method. This work provides a more refined and fair framework for evaluating surgical quality and lays the groundwork for integrating adaptive techniques in future healthcare monitoring systems. In addition to healthcare monitoring, the method can be extended to other domains where ordinal outcomes and case heterogeneity are relevant, such as education and finance. This adaptability makes it a valuable decision-support tool for quality improvement programs and real-time risk management.