Hospitals are not car plants - stop trying to improve them as if they are

Summary - hospitals are complex systems and trying to impose linear metrics on them leads to unexpected consequences and failures to actually improve. They need complex adaptive metrics to monitor and improve outcomes and cost across the variability of hospital, staff, case mix, practice, social deteminants of health, socio-economic status etc. Currently, up to 90% of issues are not found and so cannot be resolved. In the face of $2trillion of waste in healthcare, we all need to do better. Our approach supports optimizing patient outcomes and equity at lowest cost/highest margin – delivering and maintaining improvements that are unobtainable without C2-Ai – e.g. finding and helping resolve the 90% of cost/variation issues that hospital systems miss today.

__

Back in the day, I was fascinated by factories. Honestly, I still am. But it meant I ended up helping automate everything from dog food production to aluminium production and beyond. I nearly got killed falling off a crane (think opening scene of Casino Royale but with someone white knuckling onto the crane not leaping from it to the ground), being hit by molten steel and died of fright when a large glass section shattered next to me (oh look, it's bending, bending, bending... SMASH!).

A car plant is a linear thing. Bits come in one end, cars come out the other. The processes are repeated for every car. The only variation is in the mass customisation of cars. You can improve key metrics that make sense to everyone. Fewer defects. You can drive these down but you realise that each reduction in defects can cost you more money. That's why a Rolls Royce is slightly more expensive than a Dacia.

As my colleague Mike Roberts says, hospitals are complex adaptive systems. People come in and people leave (a lower number than arrive) but the variation in care and treatment is extraordinary. That means metrics you might use in a car plant don't work in hospitals. Discharge more patients and you risk higher readmissions. Drive for lower length of stay and you can get the above.

1. Complexity of Services: As Mike says, hospitals are complex adaptive systems providing multifaceted services. Each hospital is different (ask them - they'll tell you in very strong terms), every patient presents unique health conditions, comorbidities, and personal circumstances that affect their outcomes. Car plants are pretty simple in comparison, producing variations within well-defined specifications. Bottom line - car plants allow for straightforward quality control and measurement - hospitals don't
2. Variability of Inputs and Outputs: In hospitals, what comes through the door (e.g., patient demographics, medical history, severity of illness) and outputs (e.g., patient outcomes, quality of life improvements) are highly variable and influenced by numerous factors beyond the control of healthcare providers - genetics, socio-economic status and Social Determinants of Health (SDoH) etc.. In contrast, factories deal with standardized inputs (e.g., raw materials, components, pre-built assemblites) and outputs (e.g., finished products) that are more predictable and uniform. We're decades away from a car being a lemon and creating myriad issues for the owner for years after purchase. However, healthcare can create those longstanding issues for patients with avoidable harm delivering long lived, quality of life issues.
3. Human Element: Healthcare delivery involves people. Do I need to say more? You have complex interactions among healthcare professionals, patients, and their families, each with unique needs. Empathy, communication, and shared decision-making should be integral parts of patient care but the squeeze on resources and time spent staring at screens rather than walking the wards/nursing floor mean this is harder. It also cannot be easily quantified or standardised like the processes in factories. In contrast, factories primarily deal with machines, materials, and processes, with minimal human variability in production tasks.
4. Long-Term Outcomes and Quality of Life: Hospital outcomes extend beyond immediate treatment results to include long-term health outcomes, quality of life improvements, and patient satisfaction. These outcomes are often subjective and multifaceted, encompassing factors such as pain relief, functional ability, mental well-being, and social support. Factory metrics typically focus on short-term performance indicators like production efficiency, defect rates, and cost-effectiveness, which may not capture the full spectrum of outcomes relevant to healthcare.
5. Ethical Considerations: Healthcare involves ethical considerations and obligations to prioritize patient welfare, autonomy, and dignity. Unlike manufacturing processes, medical decision-making is guided by ethical principles such as beneficence, non-maleficence, autonomy, and justice. Patient-centered care should emphasize respect for individual values, preferences, and rights, which may conflict with purely efficiency-driven metrics commonly used in factory settings.
6. Choice of metrics: People respond to how they are measured. If the heat they are getting is on a particular thing, they will work to improve that (or game it in some cases). But those metrics don't always make sense. Sometimes they are politically motivated to improve a headline making number but actually that's not the right thing to improve. The resulting behaviour is therefore doing the right thing by responding to the metric but doing the wrong thing overall and patients are harmed.
7. What can't be measured, doesn't get improved: With apologies to the old saying, 90% of issues in hospitals across all of inpatient care go undetected and quality teams are left chasing 'false positives' driven by trying to use complication rates or mortality rates as indicators of performance. You wouldn't use Rolls Royce defect rates to measure a Ford plant. You certainly can't use population averages to dig into that complex system that is a hospital.

While hospitals and factories have some commonality, attempting to improve hospital outcomes with factory-style metrics completely misses the complexities of healthcare delivery and fails to capture the unique aspects of patients and their care, their well-being and the ethical considerations inherent in medical practice. Instead, healthcare performance measurement requires a comprehensive, multidimensional approach that considers clinical effectiveness, patient-centeredness, safety, equity, and efficiency.

A little about C2-Ai

C2-Ai’s AI-backed digital analytics platform delivers a new level of precision on clinical performance, outcomes and cost improvement that is essential for transforming healthcare:

• drives significant improvements even in exceptional hospitals/systems
• precision patient-level clinical risk-adjustment at scale enables sustainable improvements in margins, monetization, differentiation, revenue generation and product innovation
• comprehensively addresses the biggest issues in healthcare and more than 80% of healthcare costs ($6 trillion+)
• supports optimizing patient outcomes and equity at lowest cost/highest margin – delivering and maintaining improvements that are unobtainable without C2-Ai – e.g. can find and help resolve the 90% of cost/variation issues that hospital systems miss today
• delivers a 3-5+ year quantum leap and lasting strategic advantage
• high-velocity deployments with days to results, no integration required and no clinical disruption