Quality Assessment, Improvement, and Patient Safety in Pain Management

This chapter covers challenges in defining, measuring, and improving quality in pain management as well as patient safety issues. These are in some cases distinct and in others inextricably interrelated. They are each debated on national and international levels, and both are fundamental elements of the daily practice of pain management. The first section considers quality assessment and improvement programs and practical steps to measure quality and create a quality improvement (QI) program in pain practice. The second section discusses the patient safety movement, both on a national scale and in terms of how each pain practitioner can expect to be involved.

Section IQuality Assessment and Improvement in Pain Management

Quality is a nebulous concept, especially in the realm of healthcare. Within the field of pain medicine, healthcare quality becomes even more challenging to define because pain itself is a highly individualized, subjective biopsychosocial experience that does not easily lend itself to standardized measurement or treatment. Further, pain medicine itself is practiced in many different ways (e.g. pharmacologic, interventional, rehabilitative, interdisciplinary) and many different settings. Whether one examines healthcare at the micro-level (e.g. encounters between patients and providers) or the macro-system (e.g. processes across large academic healthcare organizations), quality is a property of the system. The healthcare system is laden with quality problems related to underuse, overuse, and misuse of services that impact patient safety. In the ensuing two sections, we discuss the topics of quality and patient safety separately. These sections review the practical efforts each pain care provider can make to address those issues at the level he or she can influence.

What Is “Quality” in Healthcare?

Healthcare quality is generally defined in terms of the attributes and outcomes of care provided by practitioners and received by patients. A more specific definition emphasizes the technical aspects and characteristics of interactions between provider and patient. Important attributes of the provider-patient interaction include communication, trust, empathy, sensitivity, and honesty. Technical quality consists of “doing the right thing right.” In other words, performing the right tests or providing the right services at the right time in the right way to accomplish the desired outcomes. Outcomes refer to the patient’s health status (e.g. changes in pain, physical function, affect, social roles) after treatment concludes. Avedis Donabedian, a physician and founder of the study of quality in healthcare, stated, “quality cannot be judged by technical terms, by healthcare practitioners alone; that the preferences of individual patients and society at large have to be taken into account as well.” Donabedian added, “As such, the definition of quality may be almost anything anyone wishes it to be, although it is, ordinarily, a reflection of values and goals current in the medical care system and in the larger society of which it is a part.”

Donabedian is also known for developing a practical “structure-process-outcome” framework for conceptualizing quality. Structure is defined as the physical and organizational properties of the setting in which care is provided; the process is what is done for patients, and the outcome is what is accomplished for patients. This three-prong model provides a more comprehensive definition of healthcare quality if we add that which is suggested by Bowers and Kiefe: “quality being the extent to which structure and process maximize the likelihood of good outcomes.” Note, the emphasis is on the “likelihood” of good outcomes because high quality care and good outcomes are not necessarily directly linked. As Chassin has described, the vagaries of the human condition mean that good quality medical care can be followed by poor outcomes, and excellent outcomes can occur despite poor care.

In 1990, Donabedian enumerated seven attributes, or “pillars,” of healthcare quality:

1
Efficacy-the ability of care to actually improve health.
2
Effectiveness-how well care achieves improvement in health in the circumstances of “everyday practice.”
3
Efficiency-the cost of any given improvement in health.
4
Optimality-the point at which incremental increases in care begin to diminish in their return on investment, such that health may be improved, but in a less efficient manner.
5
Acceptability of care to patients-accessibility, the practitioner-patient relationship, amenities of care, patient valuations of care outcomes, patient estimation of care’s economic worth.
6
Legitimacy-considerations of the value of care by others than the patient receiving that care, the aspect of societal valuation as mentioned above.
7
Equity-the balance between what individuals and what society considers the appropriate distribution of care and resources.

The degree of attention given to any attribute will vary depending on the aspect of quality being assessed. In any given instance of healthcare delivery, the patient, the provider, and the payer may have very different opinions as to what healthcare quality constitutes. This last lesson should be kept in mind in choosing any single definition of “quality” as most appropriate.

Lohr crafted a definition that the Institute of Medicine (IOM), now termed the National Academies of Medicine, included in its discussion of quality: “quality is the degree to which health services for individuals and populations increase the likelihood of desired health outcomes and are consistent with current professional knowledge.” The IOM National Roundtable on Healthcare Quality revisited this statement in 1998, redefining the phrase “desired health outcomes” to specifically include the health outcomes that patients desire, ushering in the consideration of patient and family satisfaction as a factor in evaluating the quality of healthcare services. However, in the context of pain management, interpreting patient satisfaction is complex. High satisfaction scores are not consistently correlated with traditional outcome and safety indicators and may be associated with increased costs of care. ^, Concern over the potential receipt of low patient satisfaction scores impacting pay for performance formulas may have contributed to the overprescribing of opioids in the United States even though opioid prescribing has not been correlated with consumer satisfaction. Patient satisfaction surveys are almost always skewed toward the positive and may be more representative of the quality of the interpersonal relationship between patient and caregiver than the quality of the actual services or their outcomes. Interviews of primary care patients with chronic pain revealed that high patient satisfaction was driven by concrete behaviors including listening, maintaining communication with patients, acting as an access point to comprehensive pain care, providing an honest assessment of the possibilities of pain care, and taking time during consultations with patients.

Drivers of Attention to Quality in Healthcare

Several noteworthy historical influences have led to sustained attention on quality assessment and management in healthcare.

Development of Standards of Care and Accreditation Programs

The debate about the quality of care began more than 100 years ago when a surgeon named Ernest Codman (1869-1940) first proposed standardized outcomes data collection to determine whether care was effective. Codman’s ideas swayed the American College of Surgeons to establish a Committee on the Standardization of Hospital Care, ultimately leading to the formation of The Joint Commission (TJC). Interest in care standards grew among public and private insurers as the basis for better and presumably more cost effective coverage policies and quality measurement. Care standards began to be used to shape policy, assess utilization of services, and define and identify inappropriate medical care. In the 1960s, Medicare and Medicaid were established to broaden the population that received healthcare coverage, improve access to care, expand the array of healthcare providers, and decrease cost while improving the quality of care. To help ensure delivery of healthcare in a manner consistent with available standards, accreditation programs such as TJC and the Commission on Accreditation of Rehabilitation Facilities (CARF) became prerequisites for participation in the Medicare program.

Growth of Epidemiology, Information Technology, and Outcomes Research

Growth in clinical epidemiology, information technology, and outcomes research has strongly accelerated the progress of healthcare quality. Use of information and communications garnered from these technological advances helped identify wide variation in the practice and outcomes of care among patients treated for the same healthcare problems in different places and healthcare settings. The awareness of practice variations led to the need to better understand their effect on outcomes. Geographic variations have been uncovered in cost and care, often without linkage between more care and better quality. ^, This information fueled interest in outcomes research resulted in new measures of quality, especially in the treatment of chronic illness such as pain, where patient reported outcomes that impact physical, emotional, and social function are important.

The Triple Aim Campaign and Related Concerns over the Cost of Healthcare

In 2007, the Institute for Healthcare Improvement (IHI) created a framework for optimizing health system performance termed the “triple aim” focused on the health of a population, experience of care for individuals within that population, and per capita cost of providing that care. The IHI triple aim entails improvement at all levels of the system and includes identification of target populations, the definition of system aims, and rapid scale testing of models of care and scale up adapted to local needs and conditions. The Patient Protection and Affordable Care Act, passed in March 2010, was designed to address triple pain concerns through a myriad of reforms in healthcare insurance markets. Though the legislation remains controversial, one important objective is to better orient the nation’s healthcare system toward health promotion and disease prevention.

The triple aim supports efforts to focus on value for money invested rather than solely on reducing growth or cost reduction. The healthcare spending in the United States grew 4.6 percent in 2018 and accounts for 17.7 percent of the nation’s gross domestic product. Ninety percent of the national $3.5 trillion in annual healthcare expenditures are for people with chronic and mental health conditions, many of which include a significant component of pain. It is the perception of many in business, government, and among the public that there is much waste of money in these figures. One example of increased expenditures being disabling low back pain that is partly iatrogenic because of harmful consequences associated with exposure to healthcare.

Malpractice Litigation and For-Profit Healthcare

As an indicator of poor quality, medical malpractice litigation is viewed by others as a driving force for the defense of over utilization of services, leading to increased spending and exposure to unnecessary risk for the patient. Malpractice litigation might also inhibit error disclosure and be counterproductive to efforts to improve quality. Findings from a review of 37 studies in 2020 found no association between measures of malpractice liability risk and healthcare quality and outcomes. The authors concluded that malpractice liability risk might not be an effective deterrent in preventing substandard care.

Additionally, the increase in for-profit healthcare delivery systems, specialty (“boutique”) hospitals, and increased office-based surgery is a concern for some regarding cost and waste in the system. There is a fear that such practice modes “skim” the most profitable segment of income from the traditional hospitals, leaving the larger nonprofit and public entities at risk for financial ruin. There is also a potential conflict of interest when the providers are stakeholders in these healthcare entities. However, a cross-sectional study of 31 provider-led health plans (PLHPs) reported higher Medicare Advantage program star ratings, effectiveness, access, and patient satisfaction among PLHPs compared with non-PLHPs, and lower procedure rates. The association between PLHPs and outcomes did differ by plan size, nonprofit status, and region. This heterogeneity is likely attributed to multiple factors, including population needs, enrollment policies, and leveraging of health system resources.

Politics and the National Pain Strategy

Finally, the politicization of decisions affecting healthcare through the expanding role of government and big business in scrutinizing and regulating its practice is a huge force impacting quality and cost. Examples include potential implications of United States Supreme Court appointments, political influence from industry campaign donations that influence decision makers on a range of issues, including coverage for new technology, payment levels for a care sector, interpretation of a Medicaid regulation, classification of a hospital as “rural” or “critical access,” measurement and reporting of quality, or even coverage for physician counseling on end-of-life care options. The toxicity of politics perpetuates enormous waste.

In 2010 the National Institutes of Health (NIH) contracted with the IOM to make recommendations to increase the recognition of pain as a significant public health problem. In response to the final report, the Department of Health and Human Services asked the Interagency Pain Research Coordinating Committee to oversee the creation of a National Pain Strategy (NPS). Evidence from this report suggests that wide variations in clinical practice, inadequate tailoring of pain therapies to individuals, and reliance on relatively ineffective and potentially high risk treatments such as inappropriate prescribing of opioid analgesics or certain surgical interventions that contribute to poor quality care for people with pain and increase healthcare costs. The NPS aligns with the triple aim and recommends improvements in standard methods and metrics that can reveal the pattern of health services, including over and under treatment, costs, and quality of care. The NPS provides a roadmap with substantial broad and specific political implications, including concrete steps to transform pain education, models of care, and payment structures. However, transitions in political power, available data and funding, and the opioid epidemic continue to impact implementation progress.

What Is “Quality” in Pain Management?

Few explicit definitions of quality pain management can be found in the literature. This is partly because, in the history of medicine, views on pain have only relatively recently changed from that of a mere symptom to a field of study and practice specialty. As well, definitions of quality have historically been couched in quality assurance (QA) paradigms popular in the 1970s and 80s that relied mainly on inspection and audit activities aimed at identifying and isolating performance problems. In QA, standards of care are used as yardsticks, focused on identifying adverse outcomes, reducing providers’ errors in the care of individual patients, and censuring responsible parties. The difficulty with the QA model is that feedback comes too late in the process and does little to explain differences in practice or outcomes. Also, medical records, a primary source of this type of quality review data, are notoriously incomplete and are often missing information important to evaluate the quality of pain management. In 1990, Mitchell Max, then chairman of the American Pain Society’s (APS) Quality of Care Committee, suggested that the traditional formula of QA and education was insufficient to improve pain management and treatment outcomes. Citing the failure of these efforts to change practice, Max suggested that a series of background factors must be addressed in the context of the clinical setting bringing attention to QI in the field of pain management.

In 1996, Sanders and colleagues forwarded guidelines for a uniform approach to a quality assessment of programs that treat persons with chronic non-cancer pain syndromes hinting at a definition of chronic pain care quality. The guidelines were meant to be used as a complimentary document to those published by the CARF and to evaluate interdisciplinary chronic pain programs based on eight basic outcomes objectives including:

1
Reduction in the misuse of medications (if present)
2
Increase in physical function
3
Increase in productive activity at home, socially, and/or at work
4
Improvement in overall mood
5
Reduction in patients’ subjective pain intensity
6
Reduction in use of and expenditures for healthcare
7
Where applicable, achievement of case settlement
8
Minimizing pain treatment program costs without compromising the quality of care.

Given the evolution of the healthcare system, frequent updates were recommended, but to date, there have been no visible coordinated efforts to further specify quality definitions for chronic pain programs until the development of the NPS.

Under the leadership of Max, the APS provided QI guidelines for acute and cancer pain. The recommendations specify that all care settings formulate structured, multilevel systems approaches that are sensitive to the type of pain, population served and setting of care that ensure prompt recognition and treatment of pain, involvement of patients and families in the pain treatment plan, improved treatment patterns, regular reassessment and adjustment of the pain management plan as needed, and measurement of processes and outcomes of pain management. This work helped shaped one explicit definition of quality pain management to include appropriate assessment (screening for the presence of pain, completion of a comprehensive initial evaluation when pain is present, and frequent reassessment of patient response to treatment); interdisciplinary, collaborative care planning that includes patient input; appropriate treatment that is efficacious, cost conscious, culturally and developmentally appropriate, and safe; and access to specialty care as needed. A more recent concept evaluation of quality pain management in adult hospitalized patients concluded that pain management quality is a multidimensional concept that encompasses several subconcepts ( Box 6.1 ), and valid, reliable, and operational measures of structure, process, and outcomes are needed.

BOX 6.1

Characteristics of Quality Pain Management According to the Donabedian Model

Structure

system-wide efforts and organizational commitment
policies and standards supportive of effective pain management
monitoring effectiveness and appropriateness of pain management
staff education and training
interprofessional care
competent staff
access to specialized care
accountability

Process

prompt recognition of pain
screening, standardized assessment, and reassessment of pain, based on self-report
documentation of pain assessment
appropriate, safe, individualized, evidence-based pain treatment
patient and family education
identification of patient groups at risk for inadequate pain treatment
patient participation
caring attitude and trust in patient/provider relationship
continuity of care

Outcomes

reduced pain severity
acceptable pain relief
minimal pain interference on physical and psychosocial function, and quality of life
minimal suffering from pain
minimal side effects/adverse effects from treatment
satisfaction with pain management

Quality Assessment and Measurement

Demand for measurement of healthcare quality continues to grow from all fronts, including managed care plans, healthcare consumers, accreditors, and clinicians. The national quality forum (NQF) was created in 1999 by a coalition of public and private sector healthcare leaders with representatives from providers, purchasers, and consumers as well as the agency for healthcare research and quality (AHRQ), the Centers for Disease Control and Prevention (CDC), the Centers for Medicare & Medicaid Services (CMS), and the Health Resources and Services Administration. Under legislated authority, the NQF-convened Measure Applications Partnership advises the federal government and private sector payers on the optimal measures for use in specific payment and accountability programs. There are five major hospital quality measurement sets endorsed by the NQF, that were developed for a variety of purposes, including QI, pay for performance, and public health monitoring, including:

1
AHRQ Quality Indicators
2
National Hospital Quality Measures
3
Consumer Assessment of Healthcare Providers and Systems Hospital Survey
4
TJC’s performance initiative ORYX®
5
Leapfrog Group’s Measures

These measurement sets focus on quality and safety and utilize administrative data, patient medical records, and patient surveys.

Recognition of the differences in measures for performance (accountability), research , and QI is important because the measurement is not a neutral activity, meaning there can be unintended consequences. One example being attempts to quantify pain using the zero to ten scale combined with the infamous campaign to make pain a “ fifth vital sign” that contributed to the overprescribing of opioids .

Performance Measures

Accountability measures are performance indicators used by purchasers of healthcare to compare services. Performance indicators are typically public and intended to measure variations and improvements in performance and reflect the concerns of patients, providers, regulators, accreditation bodies, and other stakeholders. These indicators are often rate-based and reported as fractions or percentages of a total number of eligible events. The theory is that performance measures create competition in the marketplace, which will lead to improvements in healthcare quality and service. However, success in improving quality and safety through the use of public performance reports has been mixed . ^– This is partly because of the choice of comparison measures with insufficient evidence of validity and how the measures are used. Measures should be carefully chosen for attributes ( Box 6 . 2 ) that make them appropriate for the purpose used and support the intended goal . ^, Use of poorly designed measures may lead to misinterpretation of findings.

BOX 6.2

National Quality Forum (NQF) Criteria for Evaluating Measures Suitable for Public Reporting

Standardization : The measures are standardized at the national level, which means that all healthcare providers will be reporting the same kind of data in the same way.
Comparability : If appropriate, the results are adjusted for external factors that could make a healthcare organization’s performance appear better or worse than it really is; such factors include age, education, gender, income, and health status.
Availability : Data will be available for the majority of healthcare organizations that you are profiling.
Timeliness : The results will be available in time for you to produce and distribute a report when it is most needed by consumers.
Relevance : The measures address the concerns of your audience.
Validity : The measures have been adequately tested to ensure that they consistently and accurately reflect the performance of healthcare organizations.
Experience : Healthcare organizations have experience with these measures to be confident that the measure reflects actual performance and not shortcomings in information systems.
Stability : The measures are not scheduled to be “retired,” e.g. removed from a measurement data set to make room for better measures.
Evaluability : The results can be evaluated as either better or worse than other results, in contrast to descriptive information that merely shows how healthcare organizations may be different from each other. For example, a complication rate is an evaluable measure because we know that a lower rate is always better; in contrast, a Caesarian-section rate is not evaluable because we do not necessarily know whether a higher rate or a lower rate is desirable.
Distinguishable : The measures reveal significant differences among healthcare organizations.
Credibility : The measures are either audited or do not require an audit.

The use of performance measures and the report cards that derive from them to rate institutional, practice, and individual physician performance continues to grow and be tied to payment models. Financial incentives to reward performance measures include accountable care organizations and bundled payments, value-based purchasing, or pay-for-performance (P4P) programs. All are complex, with limited evidence from which to draw conclusions. Participation has been somewhat voluntary, and the financial incentives have been either paltry, easily gained, or both. Regardless, legislation has accelerated a shift to value-based payments, as evidenced in 2015 when the Department of Health and Human Services announced their intent to tie 85% of all traditional Medicare payments to quality or value by 2016 and 90% by 2018.

Physicians have been reported to support the use of evidence-based measures that are congruent with provider expectations for clinical quality, targeting areas of poor performance and de-emphasizing areas that have achieved high performance. Thus measures targeting process-of-care or clinical outcomes that are evidence-based and viewed as clinically important may inspire more positive change than programs using measures targeted to efficiency or productivity or those that do not explicitly engage providers from the outset. As mentioned, a challenge is that the correlation between process measures that tend to be those most easily measured, and outcome measures that are more difficult to measure, remains controversial. In addition, the use of composite indicators that combine performance measures presents challenges to interpretation distortions and the ability to guide improvement efforts.

Research Measures

In contrast to performance measures, the purpose of measurement for research on quality is to generate new knowledge, though often with limited generalizability. Research on quality may focus on methods of implementation for new processes or outcomes. Measurement for research requires control of all possible variables, yet the study of quality of care must often rely on pragmatic and hybrid study designs because of the complexity of the interventions and context of the clinical arena.

Quality outcomes research has resulted in new measures of quality to help improve clinical practice, especially in the treatment of chronic illnesses such as back pain, where improved function is a primary objective. ^, Methods required for selecting or developing a psychometrically validated measure for quality research are beyond the scope of this chapter. Much has been written about the validity and reliability of the current pool of clinical outcome measures, including debate about use as a reference standard in responsiveness studies or as an accurate overall assessment of change. Greater consistency is needed in research measures, hence the development of the NIH Pain Consortium PROMIS® (patient reported outcomes measurement information system), a set of person-centered measures that evaluates and monitors physical, mental, and social health in adults and children. Drawing heavily on the PROMIS methodology, the NIH pain consortium has developed research standards for chronic low back pain, including recommendations for definitions, a minimum dataset, reporting outcomes, and future research.

Quality Improvement (QI) Measures

Measurement for QI is intended to understand a process, the attitudes of key process workers and customers, and to motivate all relevant personnel to improve and to evaluate changes that have been made. Measures chosen for QI must be few and easy to collect because there are often limited resources and time to collect them. Measurements should be made over a short period so that data can be collected periodically and, most importantly, be available to provide timely insight into a problem. A criticism of QI has been the lack of rigor in the methodologies used to generate data with resultant doubts about the validity and reliability of QI measures. However, there are no gold standard methodologic approaches to QI measurement development. A more thorough description of QI approaches follows.

Continuous Quality Improvement (CQI) or Total Quality Management Approaches

Used for years in the manufacturing industry, QI is a compilation of methods adapted from psychology, statistics, and operations research to avert predictable human errors, eliminate unnecessary and harmful variations in practice, and improve the production of goods and services. Of note, many assessment techniques developed in engineering and used in QI (e.g. statistical process control, time series analysis) have more power to inform about processes and context than randomized controlled trials. Continuous QI (CQI) is a philosophy and involves enlisting an entire organization to work toward a goal of continuous improvement in quality by carefully studying the process one is attempting to improve and changes the responsibilities and power of frontline workers. There are numerous approaches to QI, but the general principles are similar. All focus on reducing unnecessary variation in production processes, standardization, and continuous improvement in outcomes rather than on the identification and elimination of “defects.” The following four major frameworks for CQI share the philosophy that quality problems are process problems rather than people problems and base decisions on data, not hunches or opinions.

1
The PDSA approach (an acronym that stands for Plan, Do, Study, and Act) was first popularized by Edwards Deming (1900 to 1993), an electrical engineer, and is a rephrasing of the scientific method. This method is driven by three primary questions: (1) what are we trying to accomplish, (2) how will we know that a change is an improvement, and (3) what changes can we make that will result in improvement. Continual improvement is made through repetitive iterations of the PDSA cycle. Deming is also known for his “profound system of knowledge,” composed of four interrelated parts:
- 1
  Appreciation for a system
- 2
  Understanding variation
- 3
  Theory of knowledge or how people’s view of what is meaningful impacts their learning and decision making
- 4
  Psychology
  
  Deming’s work was influenced by Walter A. Shewhart, a physicist known for his developments in using statistics such as upper and lower control limits set at three standard errors above and below the mean to identify major causes of variation.
2
Total Quality Management (TQM ), attributed to Joseph M. Juran (1904 to 2008), builds on the work of Deming to prioritize customer expectations and team-based approaches to improve products and customer satisfaction. Juran is attributed with the development of the Pareto principle (the 80-20 rule), which states that 80% of the effects in a system arise from 20% of the causes. TQM integrates all employees into the common goal of QI. Adapted by the IHI, team-based improvements are accomplished by using analytical tools (e.g. flowcharts, statistical charts) and process techniques (e.g. brainstorming, nominal groups, consensus decision making).
3
Six Sigma is a management system developed by Motorola in 1986 to reduce variability within work processes by reducing error rates. A five step methodology known as DMAIC (define, measure, analyze, improve, control) is followed using histograms, run charts and Pareto charts, and Ishikawa or “fishbone diagrams” most relevant to the variables to be addressed by the QI team. An example of a pain management fishbone diagram can be seen in Figure 6.1 .
4
Finally, Lean emerged from the Toyota Production System in the 1990s and is aimed at reducing waste or non-value-adding activities. Value stream mapping is used to visualize materials and information in a process and identify excess steps and bottlenecks such as wait times, motion, and cost of quality. Value stream mapping helps analyze the current state and design a future state for the series of events that take a product or service from the beginning of the specific process until it reaches the customer.

TJC has constructed a framework utilizing Lean and Six Sigma methods along with a systematic approach to change management to help healthcare organizations create a culture of high reliability for quality and safety. Medical students from the Columbia-Bassett Medical School used a Lean Six Sigma approach to pain management when Hospital Consumer Assessment of Healthcare Providers and Systems survey data revealed opportunities for improvement. Analysis of patient survey data revealed timeliness of addressing pain and feeling the care team understood their pain significantly correlated with being “very satisfied.” These two variables were identified in a Pareto analysis as accounting for 71% of defects. Process mapping helped identify root causes, and a successful procedure was created to focus on response time for pain-related calls from patient call to nursing contact.

Though the evidence supporting these CQI methodologies in healthcare is positive, it remains limited and may only have modest effects on outcomes at best.

Reducing Unnecessary Variation by Understanding Root Cause Analysis

The process of reducing variation or eliminating misuse, underuse, or overuse of treatments is inherent in the standards of evidence-based medicine. It is important to recognize that variation may be good or bad, or neither. It may indicate significant overuse or underuse of a therapy and thereby signal that either the science of the therapy is poor or that it is being applied haphazardly. However, before practice data can truly be defined as significant variation, it must be assayed for any underlying reasons. This includes exploring potential root causes and associations of multiple factors influencing diverse practices, such as dissimilarities in supply and demand because of geographic economics and climate. For instance, findings in the variation of inappropriate use of MRI for occupational low back pain showed significant associations based on gender, state workman’s compensation policy not limiting initial treating provider choice, higher state orthopedic surgeon density, and lower state MRI facility density. The reader is reminded that associations are not causation. In reality, a deeper inquiry into actual root causes is needed focusing on systems and processes in an organization to identify potential targets for improvements.

Statistical control charts are often used to determine if a variation is because of chance or special causes. Common-cause variation is chance or random variation where no one or combination of factors is affecting the process variation, whereas special-cause variation is when one or more factors influence the process variation in a non-random way. Five root cause analysis tools ( Box 6.3 ) are commonly used in QI efforts, each appropriate for different situations. The value in finding variation in care is to provide a signal that there may be best practices that can be emulated, but best practices can only be determined if variation is understood. Again, not all variation is misuse, overuse, or underuse.

BOX 6.3

Common Root Cause Analysis Tools ^,

1. Pareto Chart

A Pareto chart is a histogram or bar chart combined with a line graph that groups the frequency or cost of different problems to show their relative significance. The bars show frequency in descending order, while the line shows cumulative percentage or total as you move from left to right.

Hypothetical example

2. 5 Whys

This method uses a series of questions to examine and peel away surface layers of a problem to get to the root cause. Each time you ask why the answer becomes the basis of the next why. This tool is most useful for problems where advanced statistics are not needed. A deeper analysis may incorporate the results of a Pareto analysis. It may take more than five times to uncover the problem.

For example

Why is the patient not seeing the right provider in a multidisciplinary pain clinic? According to a Pareto chart of contributing causes, the biggest factor is the scheduling center’s routine to calendar a patient for the earliest available appointment.
Why does the call center schedule for earliest available appointments? Because they have no triage or screening criteria or information to do otherwise.
Why is there no triage or screening criteria? Because the pain clinic has not created triage criteria that can easily be used by a scheduler.
Why hasn’t the clinic created triage criteria? Because historically, the workflow involved a nurse triage of incoming referrals.
Why is the nurse triage no longer happening? Because the need to clarify triage criteria with the appointment scheduler was not captured in a reorganization of the pain clinic and change in nursing roles.

3. Fishbone Diagram (see Fig. 6.1 )

Also called an Ishakawa or “cause-and-effect” diagram, a fishbone sorts the possible causes into various categories that branch off from the original problem. Multiple sub-causes branch off each identified category.

4. Scatter Plot Diagram

A scatter diagram plots your suspected cause (or independent variable) on the x-axis and the effect (or your dependent variable) on the y-axis. This scatter plot or diagram helps uncover relationships. If the patterns show a clear line or curve, you know the variables are correlated and can proceed to correlation or regression analysis.

Legend: Correlation between RMDQ scores and pain intensity. Scatter plot showing the relationship between disability (RMDQ scores) and pain intensity scores measured with visual analog scale. The many small circles represent the plotted values obtained for each of the variables, while the line represents the best fit for the correlation between them.

RMDQ = Roland Morris Disability Questionnaire

Reproduced with permission. The final publication is available at http://link.springer.com .

Scatter Plot Source

Doualla M, Aminde J, Aminde LN, et al. Factors influencing disability in patients with chronic low back pain attending a tertiary hospital in sub-Saharan Africa. BMC Musculoskelet Disor. 2019;29(1):25.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here