RESILIENT BUT NOT RESTORABLE? An Empirical Test of the Resilience–Restorability Distinction in Canadian Healthcare
Year: 2026
Type: preprint
Abstract: RESILIENT BUT NOT RESTORABLE? — Empirical Test of the Resilience–Restorability Distinction This manuscript is the first empirical adversarial test of the distinction between Resilience and Restorability inside Structural Restoration Dynamics (SRD). It asks whether systems that maintain or adapt function under disturbance necessarily rebuild durable slack, retire accumulated deficit, and reduce extraordinary dependence afterward. The paper uses Canadian healthcare as a within‑system testbed, avoiding cross‑domain confounding and allowing multiple service trajectories to be compared under the same institutional environment, the same pandemic shock, and the same national reporting architecture. The manuscript contributes by demonstrating descriptive separability between resilience and restorability. Radiation therapy remained near benchmark performance throughout the pandemic, providing a High Resilience–High Restorability case. Cataract surgery deteriorated sharply in 2020 but returned to pre‑pandemic benchmark performance by 2023–2024, providing a Low Resilience–High Restorability case. Hip and knee replacements recovered volume but not benchmark attainment, providing a Low Resilience–Low/Incomplete Restorability case. Hospital‑wide continuity remained high through large increases in overtime and purchased staffing while access pressures persisted, providing a High Resilience–Low/Incomplete Restorability case. These off‑diagonal cases show that resilience and restorability do not collapse into a single axis. The manuscript’s strongest contribution is clarifying what restorability must measure. It cannot be defined as “resilience later.” It must use variables not contained in the resilience classification: Restoration Deficit (RD), Cumulative Restoration Deficit (CRD), Recovery Slack, Depleting Compensation, and Restoration Lead Time (τ_R). The paper freezes the evidentiary standard for durable restorability: deficit closure, slack rebuilding, declining extraordinary dependence, and restoration feasibility within remaining buffer endurance. The manuscript contributes by specifying the decisive falsification test. Resilience must be measured independently using validated instruments such as the Resilience Assessment Grid (RAG) or ED‑RAG, not inferred from continuity. Restorability must be measured independently using SRD variables. The nested comparison is then:M_R (resilience‑based prediction) versus M_SRD (resilience + restoration variables).SRD survives only if restoration variables improve prospective, out‑of‑sample prediction of future restoration difficulty, backlog closure, slack rebuilding, or compensation normalization beyond resilience potentials. The manuscript also contributes by identifying strong negative controls. Cataract surgery demonstrates that low resilience during shock does not imply low restorability. Radiation therapy demonstrates that high resilience does not imply hidden deterioration. These controls prevent SRD from overdiagnosing restoration deficit based on continuity alone. The paper’s most important contribution is narrowing SRD’s scientific identity. It shows that SRD is not a theory of hidden failure or a replacement for resilience engineering. Its surviving core is requirement‑adjusted restoration sufficiency: the question of whether a system that adapts successfully under disturbance can rebuild durable capacity fast enough to meet its changing structural requirement. This reframing aligns SRD with functional‑recovery engineering while preserving a distinct empirical burden. Finally, the manuscript contributes by clarifying what is not yet established. Statistical independence between resilience and restorability is not yet demonstrated because independent resilience‑potential measures are not available nationally. Incremental predictive value of SRD variables is not yet demonstrated because a matched resilience‑restoration panel has not been run. The causal loop linking persistent compensation to future Canadian system‑level capacity loss is plausible but not yet identified. In unified form, the manuscript establishes that resilience and restorability are descriptively separable, that restoration trajectories cannot be inferred from continuity alone, and that SRD’s scientific future depends on whether restoration variables improve prospective prediction beyond validated resilience measures. It is the first empirical narrowing of SRD and the foundation for the next prospective validation study.
Author: Rupture, Signal
Language: English
Subfield: Health Information Management
Field: Health Professions
Domain: Health Sciences
Open Access status: green
OpenAlex ID: W7204147307
Cites:
Cited by:
Related to: