Ijaz Rauf is President at Eminent Tech Corporation and an Adjunct Professor of Physics, School of Graduate Studies at York University
Growing evidence of problems in the level of quality and safety of care across healthcare organizations, along with public awareness, has made the quality of health care the talk of town. This has drawn significant government and regulatory attention to healthcare systems across the country. Health Quality Ontario (HQO) was established with the mandate to monitor and report on healthcare performance in Ontario and the mission to bring about meaningful improvement in health care. Besides reporting on the key performance indicators, HQO holds quality rounds to share knowledge and best practices. Recently, I attended one of HQO’s quality rounds, and I left with the impression that quality in health care is not considering the right measures, using the right experts or measuring the right data.
In the first talk the President and CEO of an Ontario hospital presented on the quality improvement achievements of that hospital. One thing that drew my attention was that the accuracy rate in dispensing prescription medication. The CEO reported that the accuracy rate for dispensing prescription medication has risen to 99% through the implementation of robotics and automation. Being a Six Sigma Master Black Belt & Master Trainer (that is, physicist) that I am, I immediately started to calculate the Sigma level and defects per million (DPM) level, and it turns out that this automated and robotics assisted prescription medication dispensing process is operating at only 2.4 Sigma. In layman’s terms, that means that, if the hospital is issuing and filling one million prescriptions in one year, there is a probability of 10,000 wrong prescription per annum.
Now compare this to a restaurant’s billing process that on an average operates at 3 Sigma, the payroll processing system that operates at 3.8 Sigma, the airline baggage handling process that operates at 4 Sigma and the airline fatality rates that are operating at 6.7 Sigma. In monetary terms, generally, an area of an organization operating at 3 Sigma will incur 2.7 million dollars’ cost of poor quality (COPQ) for every 1 billion dollars’ worth of assets. In healthcare, the stakes are much higher as human lives are involved. If systems start to operate at 6 Sigma (as the air aviation system operates) the COPQ comes down to about $2 for every 1 billion in assets.
Another talk at the recent HQO round was delivered by a lead physician at a large family health team, who spoke about the great strides their team has made. I was surprised, though, when the physician presented the composition of the team that worked on the process improvement projects, to learn that there was neither any process engineering expertise nor any person with training in the field of quality on the team. We learned later from the final speaker that a study measuring wait times revealed that one of the factors adding to wait times is that referrals from family physicians sometimes go to the wrong doctors and are simply turned down. The example given was of a knee surgery referral wrongfully sent to a shoulder surgery specialist. It seems wrong to me that in such a specialized practice environment, when it comes to improvements and quality, a need for formal process and quality expertise is neither identified nor felt.
A second family physician spoke about data being collected on wait times. ‘Wait times’ is a parameter that the ministry of health and long term care likes to speak about a lot. But, I’m not sure these are the right data to collect regardless of the practice area. Surely, we should identify the risk priority numbers along with the wait times. Data on wait times, in themselves, are unhelpful unless one also identifies the risks and consequences associated with those wait times. For example, wait times for one kind of surgery may not be associated with the serious clinical consequences associated with a wait times for another kind of surgery. We may also need to begin measuring the dropout rates - and the reasons for dropouts, such as death, lost referrals and more – in order to make sense of wait times. Currently such data are not reported.
While healthcare practitioners may have good and right intentions, I argue that they may not have the right skills and expertise to optimize quality improvements. Health care must collaborate with those who are not skilled in the healthcare practices but are experts in the breakthrough process improvements and quality engineering. Only by doing that can we achieve breakthrough improvements to our healthcare system.