Chapter 8_ The Safety and Quality of Health Care

INTRODUCTION

Safety and quality are two of the central dimensions of health care. In recent years, it has become easier to measure safety and quality, and it is increasingly clear that performance in both dimensions could be much better. The public is—with good justification—demanding measurement and accountability, and payment for services will increasingly be based on performance in these areas. Thus, physicians must learn about these two domains, how they can be improved, and the relative strengths and limitations of the current ability to measure them.

Safety and quality are closely related but do not completely overlap. The Institute of Medicine has suggested that safety is the first part of quality and that the health care system must first and foremost guarantee that it will deliver safe care, although quality is also pivotal. Ultimately, more net clinical benefit may come from improving quality than safety, though both are important and safety is often more tangible to the public. The chapter addresses safety first, then quality.

SAFETY IN HEALTH CARE

Safety Theory and Systems Theory

Safety theory points out that individuals make errors all the time. Examples include:

Slips: Low-level, semiautomatic behavior errors (e.g., forgetting to write an order due to distraction).
Mistakes: Higher-level errors in conscious decision-making, especially in new or complex situations (e.g., unfamiliar medication dosing).

Systems theory suggests most accidents result from a series of small system failures aligning (“Swiss cheese” model). Most health care workers intend to deliver safe care, so errors usually result from system defects rather than individual negligence.

FIGURE 8-1: Swiss cheese diagram

Accidents occur when latent system failures line up, e.g., busy unit + wet floor → fall.

Factors That Increase the Likelihood of Errors

Fatigue: Increases error risk; e.g., 24-h shifts increase error likelihood by one-third.
Stress: High-pressure situations (e.g., cardiac arrest) increase errors; protocols help reduce these.
Interruptions: Common in care delivery, cause forgotten tasks.
Complexity: Multiple data streams may overwhelm providers; tools that highlight important abnormalities help.
Transitions: Provider handoffs are vulnerable points; structured communication tools mitigate risk.

The Frequency of Adverse Events in Health Care

Large inpatient studies (e.g., Harvard Medical Practice Study) show adverse event rates of ~3.7% to 10%, with 58% preventable.
Most common adverse events: Adverse Drug Events (ADEs) (19%), wound infections (14%), technical complications (13%).
ADEs occur in 6–10% of hospital admissions; about one-third preventable.
Post-discharge period is risky; adverse event rate ~19%, with ADEs leading.

Prevention Strategies

Nosocomial infection rates reduced by checklists (e.g., ICU).
Computerized Physician Order Entry (CPOE) with clinical decision support reduces serious medication errors by 55%.
Bar coding and smart pumps improve medication safety.
National organizations (e.g., National Quality Forum) recommend practices such as readback of verbal orders and standardized abbreviations.

Measurement of Safety

Difficult due to rarity of adverse events.
Reliance on spontaneous reporting is low sensitivity (~1 in 20 ADEs reported).
Future use of electronic record signals and claims data may improve detection.
Providers must report safety problems proactively for improvement.

Conclusions about Safety

Safety can be substantially improved, with most knowledge in inpatient settings.
Effective strategies include checklists, IT leverage, human factors, team training, and a culture of safety.

QUALITY IN HEALTH CARE

Quality Theory

Donabedian's framework:
Structure: Resources and settings (e.g., catheterization lab availability).
Process: How care is delivered (e.g., aspirin use in MI).
Outcome: Results (e.g., mortality rate).
Good structure/process do not guarantee good outcome.
Continuous quality improvement through ongoing evaluation and small changes is essential.
Plan-Do-Check-Act (PDCA) cycle is a key tool for process improvement.

FIGURE 8-2: Plan-Do-Check-Act cycle

Cycle involves planning, testing changes, measuring effects, and acting on results.

Factors Relating to Quality

Stress, production pressure (too high or low), and poor systems degrade quality.
Even dedicated providers cannot achieve high-quality care in poor systems.

Data about the Current State of Quality

RAND 2006 study: patients received only 55% of recommended care.
No regional correlation between utilization and quality, but providers with higher volume have better outcomes.

Strategies for Improving Quality and Performance

Individual-level: rationing, education, feedback, incentives, penalties.
System-level: reminders, bundles (e.g., ventilator-associated pneumonia bundle), SCAMPs (Standardized Clinical Assessment and Management Plans).
Chronic Care Model emphasizes team-based care and patient engagement.

FIGURE 8-3: Chronic Care Model

Combines self-management support, delivery system design, decision support, and information systems for chronic disease care.

National State of Quality Measurement

Quality measurement widespread for MI, heart failure, pneumonia, surgical infection prevention.
Data publicly reported via Hospital Compare.
Lower performers improving more than higher performers over time.

Public Reporting

Increasingly common but patients underuse data, relying more on provider reputation.
Quality metrics can be gamed; many come from claims data.
More EHR-based metrics needed for accuracy.

Pay-for-Performance

Incentives for higher-quality care are under study.
UK example: up to 40% GP salary at risk linked to performance → improved reported quality.
Risks: patient selection bias, focusing only on incentivized measures.

CONCLUSIONS

Safety and quality in health care can improve substantially.
Value-based care and pay-for-performance will encourage investments.
Patient involvement and team care essential.
More robust measures of safety and quality are needed.
Providers must address these domains directly.