Section 3 (continued)

Cost of Poor Quality or Waste in Integrated Delivery System Settings

3.3. Estimating Waste at the Patient Care Level—TPS and Lean Analysis

The Toyota Production System (TPS) offers a problem solving approach for process improvement and operational excellence. Growing interest in TPS and lean production thinking in health care, coupled with the relevance of its direct focus on waste/inefficiency, led us to explore this approach for our study.

Eiji Toyoda and Taiichi Ohno of Toyota Motors developed TPS as part of a lean production approach when Japan faced critical threats to production (Womack et al., 1991). The underlying premise of lean production is to make things flow as close as possible to ideal manufacturing conditions, where there is no waste of any kind. Waste, therefore, is ideally eliminated between machine, equipment, and personnel such that these factors can work together to produce added value. Three key components are as follows:

  • A just-in-time system that provides all needed materials (i.e., inventory).
  • Kanban, a method for process control.
  • Kaizen, or continuous improvement, via trained workers who build quality into the production process to increase the reliability of product.

Lean production combines the best features of craft and mass production and has amassed nearly 70 years of trial and error experience across multiple industrial sectors. Researchers have empirically examined the relationship between lean production and performance in the automobile industry, ending the debate over relative value added by lean production and TPS principles where significant process discipline and control have been observed (Oliver et al., 1994, 1996). Although no systematic validation and/or evaluation of lean production and TPS principles have been conducted to date in health care, substantial value added experience is documented for a growing number of health care facilities ( Spear, 2004). Most notably, a large-scale effort by the Pittsburgh Regional Health Initiative to implement TPS principles across health care facilities in Southwestern Pennsylvania has been a success (Thompson et al., 2003). AHRQ has recently supported an exploratory demonstration of lean production as part of a larger health care redesign study DenverHealth through the Integrated Delivery System Research Network (Gabow, 2006).

As previously stated, trained workers accountable for continuous QI are a core feature of lean production and TPS principles. The ability to embed problem-solving techniques at the point of care by enabled staff offers high operational utility in health care. For this reason, we embraced this approach to examine waste/inefficiency at the Patient Care Level, with potential extension to the Episode Level, and we developed a set of data collection tools that are based on lean production and TPS concepts (go to Appendix D). We note that structured observations incorporate a traditional, operations research, time-motion study approach. Developed tools were used by a trained observer at Intermountain Healthcare, with reliability testing of waste estimates at UNC Health Care. This allowed us to crudely validate the generalizability of waste estimates (a more detailed description of our approach and waste estimates are reported in a manuscript that has been submitted for publication; go to Appendix C). Overall, these efforts were shown to produce actionable results that could be used by hospital management within current financial/operational structures.

We adopted six classification categories from the manufacturing literature: operations, clarifying, defect/error management, processing, motion, and other. Twelve new subcategory clarifications emerged from our data, as well as "interruptions" and "location changes" as new categories to describe workflow fragmentation. Definitions, intended to create mutually exclusive activity categories and explicit measurement rules, are documented in Exhibit 13. Activity categories were reviewed with the observed workers. Three clinicians with research experience reviewed the definitions for language clarity. 

Exhibit 13. Activity categories and definitions for TPS observation

CategoryDefinition/Description/Examples
1. OperationsBedside caregivers: Caregiver is with the patient or family performing physical, mental, or emotional care. Nonbedside staff: Worker is engaged in operations specific to their job (e.g., phlebotomist drawing blood, scrub tech assisting surgeon).
2. ClarifyingDiscussion (direct or by telephone) of day-to-day operations, workload, staffing, work processes. Meetings, report, rounds, teaching, "huddles", looking through medical records, locating information, paging.
3. Defect/ErrorMistakes or interruptions in work that require a corrective response
DefectEquipment, computer, or supply-related problem that requires time to correct (e.g., missing supplies)
ErrortFailure of a planned action to be completed as intended (e.g., mislabeled lab specimen).
A wrong action is taken or a wrong plan is used to achieve an aim (deviation from policy, procedure, orders, or accepted standards).
Medication error: A preventable event that may cause or lead to inappropriate medication use or patient harm while the medication is in the control of the health care professional (prescribing, order communication, product labeling, compounding, dispensing, administration, education, monitoring, and use).
4.  ProcessingRedundant work or activities that do not fundamentally change service delivery
DocumentationRecording patient care actions or patient information (e.g. test results, vital signs, notes) in the medical record; includes dictating
PaperworkRecording non-patient care actions, including writing/taking off orders (clerk taking off orders is operations), filling out forms, requisitions, care plans, work lists, registration/billing data entry, copying information to alternate forms, filing/ organizing/printing paperwork.
Prep timeEquipment/room/procedure setup, running quality control tests, etc.
StockingCounting, stocking, organizing inventory
5. MotionMoving from place to place or waiting
TravelWalking/moving from place to place (more than 10 steps; see locating)
LocatingSearching for missing items or people-if travel is required, log activity as locating, if searching for information, log as clarifying
WaitingIdle time created when people, information, materials, or work are not available
6. OtherAll other activities not categorized above, (e.g., cleaning the work area, talking to the observer)
7. InterruptionsSocial conversation, breaks, personal phone calls, etc. (excluded from waste estimates)
 All unanticipated external (to the worker) requests from people or other external events that take attention away from work, including pages, telephone calls, monitor alarms, etc.
8. Location ChangesLocation changes that require movement from one work area to another and more than 10 steps.

Source: Patient Safety Resources: Definitions, National Patient Safety Foundation

The TPS-Lean substudy was designed to estimate the cost of waste in a cross-section of acute hospital worker activities and provide a qualitative description of observed problems. The study grew out of ongoing operational TPS activities at Intermountain Healthcare. An observation tool with explicit definitions for categorizing worker activities and rules for estimating the hourly cost of waste was constructed and the reliability was verified. A single observer shadowed 61 health caregivers for 72 hours in tertiary academic and community hospital settings using structured, nonparticipant observation of worker activities across multiple care processes. Data yielded estimates of waste and a qualitative description of problems encountered.

The average cost of waste (i.e., the cost per hour per worker) ranged from USD $7.40 to USD $18.98 across all roles and functions. Overall, workers encountered an average of two problems per hour. Results are provided for specific roles and functions.

Sixty-one workers were observed for 72 hours (36 morning hours and 36 afternoon hours). Professionals included 8 physicians, 26 nurses, and 8 other roles. Of the registered nurses (RNs), 5 were ICU/emergency department staff, 10 were non-ICU medical/surgical staff, 5 were operating room/postanesthesia care unit nurses, 2 were house supervisors, 2 were patient care managers, 1 was a labor/delivery nurse, and 1 was an endoscopy lab nurse. The laboratory workers included two phlebotomists, two medical technologists, and two specimen processors; these workers were grouped together with other technical staff (n = 13). In general, the workers were experienced; only 5 (8%) had less than 1 year of experience in their role and 44 (72%) had more than 3 years.

The average, overall cost of waste (i.e., cost per hour per worker) across all staffing groups was USD $7.40 (low), USD $13.20 (medium), and USD $18.98 (high). Interruptions and location changes occurred at an average (standard deviation, range) rate of 8 (11, 0-80) and 13 (11, 0-58) times per hour, respectively (one technical worker assisting with a cardiac catheterization was uninterrupted during a 30-minute observation). Even though our sample size was limited, subgroup analyses of clarification activities suggested differences between roles. A more detailed description of the methods, results, discussion, and conclusions is provided in our manuscript (go to Appendix C).

Our study suggests that the cost of waste for frontline health care worker activities is substantial. Given our data, the lowest cost of waste in caregiver activities for a single-day shift on one 46-bed medical unit (staffed with eight RNs, eight patient care technicians, two care managers, one social worker, one physical therapist, one pharmacist, one respiratory therapist, two clerks, and two hospitalists) is USD $2,309 (12 hours x 26 workers x $7.40 per hour); the annual cost for the same unit is USD $843,000 (USD $2,309 x 365). Because of our conservative assumptions, these estimates almost certainly represent an underestimate. During the observations, workers spent less than half of their time engaged in operations. Nonoperational activities were almost evenly split between clarifying (20%), processing (19%), and motion (17%). The subgroup analyses suggested physicians and supervisory RNs spent more time and technicians spent less time than the overall average in clarification activities. Given the nature of their functions, these data are not surprising, and one might argue that time spent by more senior health care workers in clarification activities is important. We note that some of the clarification may be necessary aspects of training given the current methods used in academic teaching hospitals. Outside of training, one could argue that clarification activities indicate a lack of specified processes and a high tolerance for ambiguity, resulting in greater waste of the most experienced workers in our system. It is easy to make the case that redundant documentation or paperwork is an unproductive use of workers' time. Despite the advanced information technology available at Intermountain Healthcare, redundant documentation and paperwork were not infrequent. Other investigators who have included observation methods to evaluate information technology in clinical work reported unanticipated results that would not have been uncovered without qualitative data. It is self-evident that motion (i.e., traveling, locating, and waiting) is wasteful and should be minimized whenever possible.

Increased attention to operational quality in health care is needed and could potentially decrease costs while increasing patient safety. Implications for poor operational quality and recommendations for action are presented in our manuscript, which is under review.

Return to Contents

3.4. Conclusions

We have seen that each of our three methods can provide a perspective on the amount of quality waste present in the U.S. health system. The Dartmouth Atlas approach is perhaps the most refined at this point and the only approach we feel comfortable using to derive synthetic estimates for the national/policy level. The other methods have important limitations.

The process improvement efforts, for example, often struggle with isolating a process for improvement. Health care is extremely complex, some would say chaotic, and the health care industry simply has not done the work to more clearly define its work in terms of an organizational structure with established processes that are universally consistent across facilities. This is actually a major source of quality waste in that we do not understand our work processes. Theoretically, the steps involved would be:

  1. Quality control, developing a management system that allows for monitoring, managing, and stabilizing a process.
  2. Quality improvement (QI).
  3. E-design with innovations.

The health care industry jumped to Step 2 (QI) and never did Step 1 (quality control); most care delivery facilities have never built the measurement and management infrastructure necessary to systematically manage care processes. Thus, QI is disjointed, unfocused, uncoordinated, and much less effective than it could be. Attempts to identify processes for improvement have found processes embedded in complex systems that may resist efforts to change. As quality control-based measurement and management systems advance, our ability to directly measure, then eliminate, waste within health care delivery should increase rapidly. While we were able to find synthetic estimates of health care only at the Population Level, those alone were massive, representing 32 percent of total health care expenditures. Our work suggests, but does not demonstrate, that waste at the Episode and Patient Care Levels is of similar scale. If so, the total amount of waste within the health care delivery system will total well over 50 percent of all health care expenditures.

As important, our approach supplies the tools not just to identify waste but to eliminate waste through process management. As integrated delivery systems with real health care operations challenges, both Intermountain Healthcare and Providence Health System plan to vigorously continue this line of investigation. In particular, we hope to strengthen waste estimates at the Patient Care Level by more broadly applying our TPS observation tools, and we plan to create additional savings estimates from successful projects in clinical management.

Current as of September 2008
Internet Citation: Section 3 (continued): Cost of Poor Quality or Waste in Integrated Delivery System Settings. September 2008. Agency for Healthcare Research and Quality, Rockville, MD. http://www.ahrq.gov/research/findings/final-reports/costpqids/costpqids3a.html