Hospitals Need IE Data to Drive Staffing Decisions
by Mark Graban
How do hospitals set staffing levels? Often, they are not using time-tested industrial engineering methods. When hospital managers are asked, for example, “Why do you have four people working this area?” the responses include:
• “That’s what the benchmarking data indicated.”
• ‘That’s what the consultant told us.”
• “We couldn’t get the budget for five!”
Benchmarking data is often suspect, as it can be difficult to find a true apples-to-apples comparison between two hospitals. Aspects of the organization’s specific context, such as medical equipment, patient mix, and information systems all contribute to that particular hospital’s productivity. Without solid IE methods such as work measurement, the hospitals end up with staffing levels that are too high as to be excessively costly or too low so that quality, morale or safety suffers.
In my years of working with hospitals, managers and staff will almost certainly make two definitive statements, usually without the data to back them up:
1. “We need more space.” and
2. “We need more people.”
Through the application of IE methods and practices, it is often proven that both statements are untrue. Improving layouts and flow reduces space requirements. Removing process waste and inefficiency increases productivity and allows hospitals to reduce labor costs responsibly(preferably through reassignment or natural attrition).
I have worked with three hospitals that used IE methods to find the elusive middle ground between too much staff and not enough. The three cases were a microbiology laboratory, a radiology department and the front desk operations of a family practice clinic.
In these three hospitals, a few staff members learned to implement lean as part of a 13-week project. The employees were trained on the eight wastes of lean and taught how to observe a process in the gemba (meaning the place where the work is done). The first few weeks were dedicated to detailed observation of their peers, but not using the stereotypical “stopwatch and clipboard.” In most cases, they used videotape with time-date stamps so they could start the tape and then work with their peers to design better methods in a blame-free context.
Each team combined their old knowledge of the process and their new lean eyes. They first identified and eliminated waste in the way the work was done so that the new staff levels would not be based on the old, inefficient methods. They also looked for quality problems or “short cuts” that were forced into the process due to a lack of time. The experienced team members had to observe the work as-is but also project what work would be done in an ideal system, where quality standards were being fully met. Sometimes that meant taking longer than was current practice. They asked, “How long would the work take if it were done properly?”
In the microbiology lab, the question was, “How many plate readers do we need?” Data was collected on current work volumes and variation. From the team’s observation and time study analysis, they determined how long it took to read plates and the associated variation. The team noticed some instances where the ideal standardized work was not being followed, meaning some employees were working faster than desired (with quality being the top priority). The team determined that, to meet demand on most days, that five plate readers were needed, not four. The basic equation (oversimplified) was:
(# of plates per day) x (time to read each plate) = hours of plate reading staff required
While staff members and managers had been complaining for years of the need for more staff, this was always an opinion – not a statement backed by data. “Our workload has doubled,” was their lament. The data actually showed how workloads had increased 50 percent without any increase in staffing levels (leading to stress, overtime, and short cuts). By using IE principles and eliminating waste, the department was able to get by with only a 25 percent staffing increase, while also feeling assured that quality would improve.
A similar pattern followed in the radiology (MRI) and office settings. Before, in the absence of hard data, the opinion of, “We need more staff,” was seen as complaining – a statement easily refuted by the opinion of upper management that, instinctively, said, “No you don’t.” The team learned how to collect time study data, eliminate waste in the process and make a business case to management for how increasing staff could reduce overtime, improve quality and increase volume and revenue.
In the radiology department of a children’s hospital, staff had long asked for more sedation nursing staff (as children often must be sedated to hold still for radiology studies). Through the collection of time study data and some basic spreadsheet simulation, the team confirmed that the sedation room was indeed the bottleneck that delayed patients. We were able to model, through the data and process definition, precisely how assigning an additional radiologist and nursing staff would lead to an increase in MRI machine utilization, which both increased MRI throughput and reduced outpatient scheduling waiting time by over 80 percent.
In the medical office, although an EMR system was in place, stacks of paper had to be scanned into the system each day. This paper included faxes that arrived into the office, such as lab results and patient histories. A large backlog of paper waiting to be scanned (stacks measured in feet) caused waste, as physicians and staff would have to search for information that was not yet entered into the EMR. Again, with process observation and time study, the team was able to determine that the 0.5 FTE who ran the scanner was enough capacity in the steady state (to scan what comes in each day). While the office had long asked for help in getting the backlog down, the team was able to calculate that a temp would be needed for about three weeks to get the backlog down. The office got approval from senior leadership for the staffing expense because data drove the discussion.
In all of these cases, basic IE methods -- gathering process data through time studies -- are best combined with lean concepts of waste reduction and “respect for people.” Armed with lean concepts, time and some coaching, hospital personnel can provide the correct staffing answers.
Mark Graban is author of the book, Lean Hospitals: Improving Quality, Patient Safety, and Employee Satisfaction. He has a BSIE from Northwestern University and an MSME and MBA from the Massachusetts Institute of Technology.