Operational performance and delivery structure of hospital facilities: empirical relationship analyses in Ontario, Canada

While expending large amounts of scarce public funds on social infrastructure, including hospitals, governments globally apply increasingly diverse delivery mechanisms. Across Canada, particularly the province of Ontario, policy shifts favour the use of Public-Private Partnerships (P3s). Yet little empirical understanding exists of the implications these delivery structures have on operational outcomes like energy and labour use, major contributors to Facility Management (FM) budgets and climate change, as well as facility user satisfaction. In order to further investigate concepts surrounding a positive theory of P3s (Vining and Boardman, 2008), the three articles within this thesis pragmatically employ qualitative and quantitative methods to explore diverse aspects of the relationship between variations in facility delivery structure and operational performance among hospital facilities in Ontario, Canada. Article 1 applies a cross-sectional analysis of 2016 energy use data for all 286 operational hospitals in Ontario. Results show facilities delivered through P3 contracts including an operating term had lower energy intensity than those delivered conventionally or via P3 contracts without an operating term. Article 2 uses a comparative, longitudinal Case Study of two P3 facilities with an operating term to further investigate preliminary findings in Article 1 of a strong relationship between energy intensity and contractual approach to energy management. The facilities studied were identified in Article 1 as having among the highest and lowest energy intensities of all Ontario hospitals. Initially both facilities had poor energy performance. Then, while the facility applying a Regression Analysis Model to set annual targets and payments maintained only modest improvements over the operating period, the facility applying a Bid Energy Model achieved dramatic improvement in energy use and associated environmental performance over the same period. Supporting findings of Article 1, the facility using a Bid Energy Model capitalised on risk transfer mechanisms unique to delivery structures bundling design, construction and operating phases. In Article 3, a second Case Study investigates energy use and FM labour utilisation at a conventionally procured and operated hospital and an early P3 hospital with an operating term. Both facilities are part of the same public health care corporation in Ontario and studied in Article 1. The P3 facility was also included in Article 2 as the site applying a Bid Energy Model for energy management. Using empirical data on labour and energy use - supplemented by facility user surveys, interviews of key stakeholders and contractual study - findings show consistently better performance at the P3 facility, though after a short bedding-in period for energy use. Stakeholders reported that, despite transferring many best practices from the P3 facility to improve data tracking and reporting across the organisation, internal accountability remained a challenge at the conventionally procured facility and budgeting factors negatively affected its operations. Strong financial penalties and incentives within the P3 contract are identified as drivers for high performance for energy use and maintenance directly, and facility user satisfaction levels indirectly. The evidence from this study reveals that delivery structure is a potentially useful tool to drive operational performance, and thereby address recognised performance gaps in public buildings, but that effectiveness is highly subject to contractual details with respect to risk transfer. The studies within are unique in their use of empirical data collected from mature operational facilities under a variety of delivery models. The work helps to describe how key actors interact and respond to policy decisions that have intended and unintended consequences to hospital facilities, their stakeholders, regional economies and the broader environment. To date the author has presented findings in multiple conferences, published in conference proceedings, submitted one article for journal publication and applied these findings within his professional duties.