The objective of this GOALI project is to revolutionize outpatient clinic performance by developing optimal or near-optimal clinical scheduling methods that accommodate the environmental complexities faced by clinic schedulers. Clinical operations are driven by the patient schedule, which determines the arrival time of patients to the clinic. Clinical managers are always quick to identify inadequate patient scheduling as a major source of operational inefficiency and patient dissatisfaction. Although clinical scheduling has a long research history, practical impact has been limited for at least three reasons. First, the expertise to implement advanced scheduling methods does not exist in most clinics. Further, clinical information technologies do not easily support the computational needs of these methods. And finally, published scheduling methods inadequately address many routine and important factors that complicate clinical scheduling such as sequential patient call-in, patient no-shows (missed appointments), and patient walk-ins. Thus, there is a great need for developing and implementing new clinical scheduling methods that account for these various complications when attempting to optimize clinic effectiveness in terms of resource efficiency, patient care, and costs/revenue.
To assure applicability of the research, we will develop implementations in two large clinics located at Wishard Hospital in Indianapolis, Indiana. Wishard is a county-tax-funded hospital that provides care to an inner-city population with a large proportion of indigent and elderly patients. Wishard's clinics provide 24,000 consultations per year and are in a constant struggle to maximize capacity and minimize costs while serving a safety-net function for under- and un-insured patient population with complex healthcare needs. Thus, the work will be directly impacting the needs of the most vulnerable in our society. Specifically, we will 1) develop analytical patient scheduling methods that incorporate complex environmental factors, particularly patient no-show behavior, while provably optimizing clinic efficiency and quality of care; 2) develop predictive models of patient no-show, walk-in, and cancellation behavior; 3) implement these methods in two primary care clinics; 4) validate the impact of these methods by performing pre- and post-implementation studies; and 5) develop cross-disciplinary internship and education programs for graduate engineering students and medical residents with the two participating clinics. Given that there are approximately 200,000 non-psychiatric outpatient clinics in the US, the potential impact of this research is very high.
