Quality Improvement

Betiel Fesseha, M.D., M.P.H., Nestoras Mathioudakis, M.D., MHS



  • The US Agency for Healthcare Research and Quality defines quality health care as “doing the right thing, at the right time, in the right way, for the right person—and having the best possible results.”[5]
  • Quality improvement (QI) is a "systematic, formal approach to the analysis of practice performance and efforts to improve performance in health care."[10]

History of QI

  • QI in healthcare dates back to 19th century when Ignaz Semmelweis, an obstetrician, first advocated for the importance of hand washing in medical care.[4]
  • Over past 50 years, modern medicine embraced a formalized approach in identifying errors and improving the way healthcare was delivered to ensure patient safety.
  • Avedis Donabedian, considered the father of quality measurement, described in 1966 a way to evaluate the quality of health care, focusing on 3 areas:[5][8][9]
    • Structure: the physical and staffing characteristics of caring for patients
    • Process: the method of delivery
    • Outcome: the results of care
  • National Academies of Science established the Institute of Medicine (IOM) in 1970, which has since launched numerous concerted efforts focused on evaluating, informing, and improving the quality of healthcare delivered.[4]
  • Agency for Health Care Policy and Research—currently known as the Agency for Healthcare Research and Quality (AHRQ)—was created in 1989.
  • Patient Protection and Affordable Care Act (PPACA) signed into law by President Obama on March 23, 2010[4]
    • Contains multiple provisions designed to modify the manner in which care is delivered to Medicare and Medicaid patients, and the system by which provider payment is determined, with a central objective of improving quality while lowering healthcare costs and expanding access[4]
    • Key provision was creation of the nonprofit Patient-Centered Outcomes Research Institute (PCORI) to conduct research that compares the clinical effectiveness of medical treatments.
  • Diabetes-specific QI:
    • Diabetes Quality Improvement Project (DQIP), which was founded jointly by CMS (Centers for Medicare and Medical Services), the National Committee for Quality Assurance (NCQA), and the American Diabetes Association (ADA) in 1997[7]
      • Core features of the DQIP include the interplay between evidence, feasibility and variability for accountability measures.[7]
      • With those core values, the DQIP has successfully developed and implemented a comprehensive set of national measures for evaluation and QI, for example, using HbA1c as a measure for glycemic control.

Basics of QI

  • Currently several methodologies are used to perform QI.
  • Continuous QI (CQI) is the idea that improvement opportunities exist in every process; emphasizes the view of health care as a process and focuses on the system rather than the individual when considering improvement prospects.[5]
  • Most common QI methodologies used in health care include plan-do-study-act (PDSA), six-sigma, and lean strategies and less commonly, the RADAR matrix

PDSA Cycle

  • Also known as the Deming circle, as it was made popular by Dr. Edwards Deming
    • This model is based on the scientific method, as developed from the work of Francis Bacon and advocated by the Institute for Healthcare Improvement.[5]
  • Most commonly used QI approach
  • Involves a “trial-and-learning” approach where a hypothesis for improvement is made and carried out in a small scale before any changes are made to the whole system.[5]
  • Includes 4 phases/steps as the name suggests[5]:
    • Plan phase
      • Where objectives and ideas for improvement are discussed and established with the expected target/goal
    • Do phase
      • Where the plan is implemented and any deviation from the plan is documented (these are called defects)
      • The data are collected and analyzed in the next phase.
    • Study phase (also known as the Check phase)
      • Results from the Do phase are reviewed and compared with the expected results as outlined in the Plan phase.
      • Further questions and ideas are also discussed so as to be implemented in the next phase.
    • Act phase
      • Lessons learned from the Study phase are incorporated into the test of change.
      • Decisions are made about whether to continue with further test cycles or to go ahead and apply changes to the whole system.


  • Method first introduced by Motorola in 1986[5]
    • Designed to improve quality of output by identifying variability in manufacturing, to reduce costs and eliminate defects
    • Sigma is a statistical unit reflecting the number of standard deviations a given process is from perfection[5]
  • Six-sigma includes a series of steps: define, measure, analyze, improve and control (also known as DMAIC)[5]
  • 5-phase version known as DMADV (define, measure, analyze, design and verify) or DFSS (Design For Six Sigma)

Lean Methodology

  • Based on the revolutionary thinking of a Toyota Motor Corporation, this methodology is based on the need to identify what is valuable to the customer (or patient) and aiming to improve process by removing those “non-value-added activities.”[5]
    • One of the most commonly used tools in the lean methodology is value stream mapping (VSM).
      • Tool that graphically displays the process of services or product delivery with use of inputs, throughput, and outputs[5]
      • Usually implemented via the 5S strategy[5]:
        • Sort: prioritizing and sorting items in the immediate work area
        • Shine: cleaning the workplace for any abnormal wear
        • Straighten: setting work items in order after efficiency of the workflow has been optimized through VSM
        • Systemize: standardization of workflow processes
        • Sustain: sustaining gains made from the above steps

RADAR matrix

  • RADAR stands for results, approach, deploy, assess and refine.
    • Used to assess organizational performance and is based on the EFQM excellence model[11]
    • EFQM, formerly known as the European Foundation for Quality Management, is a non-prescriptive business excellence framework for organization management systems[12] and has been adapted to do QI in the health care setting.

QI in Diabetes Management

  • As diabetes is a complex, life-long disorder requiring continuous self-management and adherence to several therapies for glucose control, prevention of complications and reducing cardiovascular risk factors, implementing QI strategies using methodologies as discussed above to improve diabetes care is imperative.[2]
  • Optimal diabetes care includes measurement of HbA1c several times a year for glycemic control assessment; blood pressure and lipid monitoring and management; use of statins, antiplatelet agents, and ACE inhibitors/ARBs as indicated and tobacco cessation if smoking is present; and preventive screening with routine eye exams, foot exams, urine evaluation for albuminuria as necessary, and vaccinations
  • Implementing metrics to evaluate the attainment and feasibility of the above by monitoring provider performance and patient characteristics can help identify where change needs to be made to attain better diabetes care.
  • Most common targets of current QI strategies in diabetes involve addressing barriers at the health system, healthcare provider, and patient level.[2]

QI Diabetes Targets

  • Interventions focused on reorganizing and optimizing the health system include[2]:
    • Use of multidisciplinary teams
    • Expansion of professional roles (for example, allowing pharmacists to have more active role in monitoring or adjusting drug regimens)
  • Strategies that target healthcare providers include[2]:
    • Providing individual feedback on clinical performance
    • Using clinical reminder systems for specific tasks or decision support assistance
    • Adding pay-for-performance or bonus pay for achieving certain screening or care targets
  • Patient-level interventions involve[2]:
    • Strategies promoting better self-management through diabetes education, behavior changes, and/or using reminder systems for patients regarding their care

Effectiveness of QI Strategies on Diabetes Management

  • According to a recent systemic review and meta-analysis, QI interventions targeted at HbA1c reduction, blood pressure improvement, and LDL reduction were successful[3]:
    • HbA1c was reduced by a mean difference of 0.37% (95% CI 0.28 – 0.45), LDL cholesterol was reduced by 3.86 mg/dl (CI 95% 1.93 – 5.41), systolic blood pressure by 3.13 mmHg (95% CI 2.19 – 4.06) and diastolic blood pressure by 1.55 mmHg (95% CI 0.95 – 2.15).
      • Larger effects were noted when baseline levels were greater than 8% for HbA1c, 100 mg/dl for LDL cholesterol, 140 mmHg for systolic blood pressure and 80 mmHg for diastolic blood pressure.
    • QI strategies also increased the likelihood of patients receiving aspirin, antihypertensive drugs and screening for retinopathy, renal function, and foot abnormalities.[3]
      • However, statin use, hypertension control, and smoking cessation were not significantly increased.
    • Of note, for this review, the authors looked at studies that included QI strategies that targeted health systems or provider performance--QI strategies aimed solely at the patient were excluded.[3]
      • Strategies that intervened upon the entire system of chronic disease management were associated with largest effects irrespective of baseline HbA1c.
      • However, interventions solely targeting health care professionals seemed to be beneficial only if baseline HbA1c control was poor (> 8%).
  • QI strategies need to address health disparities in diabetes care as diabetes disproportionately affects racial and ethnic minorities and those of low socioeconomic status[2]:
    • Incidence of diabetes is 18% higher in Asian Americans, 66% higher in Hispanics, and 77% higher in non-Hispanic blacks.
    • African Americans are more likely to suffer from complications of diabetes.
  • Among QI interventions targeting socially disadvantaged populations, interventions that were culturally-tailored, had involvement of community educators and lay people in addition to having individualized interventions were successful in diabetes care improvement and were able to be sustained for > 6 months.[2][6]
    • Improving overall quality of care can also help reduce racial disparities in diabetes outcomes.
    • Programs at state and national levels such as the Alliance to Reduce Disparities in Diabetes and REACH (Racial and Ethnic Approaches to Community Health) have been shown to reduce healthcare disparities in chronic disease management and improve population health.[2]


  • Several continuous QI methodologies exist, and it is important to fully understand and properly execute the chosen QI model to achieve results that improve patient care, clinical outcomes, and efficiency.
  • From 2002 – 2013, the quality of outpatient care delivered to American adults was thought to not have consistently improved, though patient experience has improved.[1]
    • Thus, deficits in care continue to exist and ongoing QI efforts are needed now more than ever.
  • There is no conclusive agreement on the most effective target or strategy for diabetes-specific QI, and the strategy used should be dependent on context.[2]

There's more to see -- the rest of this topic is available only to subscribers.

Last updated: June 2, 2018