The Cost of Unsafe Medical Care

Unsafe care is a significant global public health issue, profoundly impacting millions of patients annually and incurring substantial economic costs. More than one in ten patients experience harm in medical care settings, with approximately half of these incidents considered preventable. This preventable harm leads to millions of deaths and places a considerable financial burden on health systems and society, consuming valuable resources. Beyond the direct health consequences, patient harm also causes reputational damage to health care systems, erodes trust, negatively affects the morale and well-being of health workers, and influences public opinion about investing societal resources in healthcare.

Here’s a detailed breakdown of the burden of harm in healthcare:

Burden of Harm to Patients

Overall Impact and Preventability

• Around 12% of patients experience harm across different medical care settings, meaning more than one in ten patients are harmed by adverse events due to unsafe care.
• The severity of about half of patient harm extends beyond mild injuries, with up to 12% of adverse events causing permanent disability or patient death.
• Recent estimates indicate that unsafe care causes more than 3 million deaths every year.
• Improving patient safety can prevent half of the harm in healthcare settings, potentially saving over 3 million lives annually.

Geographic Distribution of Harm

• The burden of unsafe care disproportionately affects low- and middle-income countries (LMICs), where the majority of patient harm and associated deaths occur.
• Approximately two-thirds of all patient harm due to unsafe care, and the resulting years lost to disability and death, occur in LMICs.
• It is estimated that 134 million adverse events occur in hospitals in LMICs each year, contributing to roughly 2.6 million deaths.
• An analysis of 2016 data estimated that around 5 million people died in LMICs due to poor quality care for Sustainable Development Goal (SDG)-related conditions.
• The highest per capita death rates were observed in central and west Africa and in South Asia.

Burden of Harm by Demographic Distribution

• Age:
  • While most evidence on patient harm comes from studies involving adults aged 18 to 65 years, older adults, children, and ethnic minorities are increasingly viewed as vulnerable populations.
  • The prevalence of in-hospital adverse drug reactions (ADRs) is reported at 16% among older adults (≥65 years).
  • Factors such as clinical complexity, co-morbidities, illness severity, and reduced functional ability contribute to higher levels of patient harm in older adults.
  • For hospitalized children, the incidence of adverse events varies significantly, ranging from 3.8% to 53.8% in general care and 6.9% to 91.6% in intensive care patients using the ‘Trigger tool’ methodology.
  • Children with special medical needs or dependence on medical technology have higher rates of medical errors.
• Sex:
  • In 2019, medical treatments harmed over 1.3 million people, with women slightly more affected than men (prevalence rate of 18.1 per 100,000 in females vs. 17.6 per 100,000 in males).
  • An analysis of 15 million ADR reports showed that more ADRs were reported for female than male patients across all regions, although male patients had a higher proportion of serious and fatal ADRs.
  • Overall, current evidence indicates that females experience greater levels of patient harm compared to males, highlighting the necessity of addressing gender disparities in healthcare.
• Race and Ethnicity:
  • Stark health inequities affect people of African descent, Roma, other ethnic minorities, and indigenous peoples.
  • Studies reveal that Black adult patients experienced significantly worse patient safety compared to white patients in similar age and gender groups treated in the same hospital.
  • Ethnic minority backgrounds have higher rates of healthcare-associated infections (HCAIs), complications, adverse drug events (ADEs), and dosing errors.
  • Key contributing factors to increased risk of patient harm among ethnic minorities include language proficiency, beliefs about illness and treatment, interpreter use, patient engagement, and interactions with health professionals.
  • Addressing racial and ethnic disparities in healthcare is critical, as minority groups face increased risks of patient harm.
• Patient Complexity:
  • Patient complexity, whether clinical, biological, psychological, or social, is a key risk factor for lapses in healthcare safety.
  • Patients with clinical risk factors like hypertension (32%), diabetes (18%), obesity (14%), dyslipidaemia (13%), and depression (11%) frequently experience harm.
  • Multimorbidities, psychiatric conditions, diabetes, polypharmacy, and being immunocompromised are considered some of the most important clinical risk factors for patient harm.

Burden of Harm by Medical Setting and Clinical Domain

• Highly Specialized Care Settings: These settings, including intensive care units (ICUs), emergency departments, and surgical units, are associated with the highest rates of patient harm.
  • ICUs: Experience high estimates of overall harm (~34%) and preventable harm (~18%). Up to one in five ICU patients suffer patient harm, with 13% of events being lethal or life-threatening. Patient harm in ICUs increases the length of ICU stays by an average of 6.8 days and hospital stays by 8.9 days.
  • Emergency Units: Acknowledged as high-risk due to factors like high patient volume, complex cases, time constraints, and varying physician training. The incidence of patient harm ranges from 0.2% to 6%, with preventability from 36% to 71%. The most frequent types of harm are related to management, diagnosis, and medication.
  • Surgical Units: Pose a high risk for patient harm due to complex procedures and high-risk patients. Approximately 14.4% of surgical patients experience harms, with 5.2% being potentially preventable. Unsafe surgical care contributes to approximately 13% of the world’s total patient DALYs (disability-adjusted life years) lost to care-related harms. Up to 7 million surgical patients globally suffer significant harm annually, with 1 million dying during or immediately following surgery. At least half of surgical harm is preventable.
• General Hospitals: The prevalence of all patient harm is approximately 10% of all patient interactions, and preventable harm is around 5%. Preventable harm in hospitalized patients may lead to as many as 400,000 deaths per year globally.
• Primary Care: The reported prevalence of all harm is 7% and preventable harm is 3%. However, other global estimates suggest that as many as 4 in 10 patients may be harmed in primary and outpatient care, and up to 85% of this harm is preventable. Common problems include diagnosis (60.8%), medication-related issues (25.7%), and delayed referrals (10.8%).
• Other Settings:
  • Long-term care: Over half of the harm is preventable, and over 40% of admissions to hospitals from long-term care are avoidable.
  • Mental health settings: Patients with psychiatric diagnoses face nearly double the risk of preventable harm compared to others.
  • Palliative care: Patients are vulnerable to inadvertent harm, with common incidents including pressure ulcers, medication errors, falls, and HCAIs.
  • Radiotherapy: The overall incidence of radiotherapy errors is around 1500 per million treatment courses, with 30% of errors occurring in the planning phase and 29% in the treatment step.
  • Paediatric ICUs: Harm occurs with an incidence as high as 74 per 100 admissions, and one in six patients experience ADEs, with over half preventable.
  • Telemedicine and digital health: While enhancing access to care, they present unique patient safety concerns, including diagnostic errors due to inadequate history-taking or physical examinations, and gaps in medication safety.
  • Dentistry: The prevalence of patient harm was found to be 1.8%, with main sources including treatment delays (23.6%), procedural errors (15.6%), and adverse drug events (11.1%).
  • Obstetrics services: Patient harm is estimated at between 2% and 4%, with approximately half considered preventable. Common preventable harms include peripartum therapy delay, diagnostic errors, and organizational errors.
  • Trauma (Orthopaedics): Trauma patients face higher risks, with an incidence of patient harm at 11.4% in trauma cases compared to 4.1% in general orthopaedic cases, effectively doubling the probability of harm.

Burden of Harm by Source

• Medication Errors: These are the single most important source of patient harm in healthcare systems.
  • Globally, at least 1 in 20 patients (5%) experience preventable medication-related harm.
  • More than half (53%) of this preventable harm arises at the prescribing stage, highlighting a crucial need for improving medication safety practices. In LMICs, almost 80% occurs at this stage.
  • Common error categories include omitted/delayed medicine (16%) and wrong dose (15%).
• Diagnostic Errors: Increasingly recognized as a key source, affecting 5% of adults in outpatient environments in the United States.
  • Globally, 16% of preventable patient harm across the health system may be due to diagnostic errors.
  • Common contributing processes include problems in patient–practitioner encounters such as history taking, examination, ordering and interpreting tests, and follow-up.
• Healthcare-Associated Infections (HCAIs): These are among the most common complications in hospital care, causing significant disability and premature mortality.
  • 7% of hospital patients in high-income countries (HICs) and 15% in LMICs will acquire HCAIs.
  • Globally, an estimated 136 million hospital-associated infections resistant to antibiotics occur every year.
  • Up to 30% of patients in intensive care can be affected by HCAIs, with incidence 2 to 20 times higher in LMICs.
• Venous Thromboembolism (VTE): This is a common and preventable cause of patient harm, with an annual incidence of 5 to 12 people per 10,000.
  • VTE is a leading cause of adverse events in LMICs and globally is the biggest source of lost DALYs.
• Other Common Sources:
  • Patient falls: Occur at a rate of 3 to 5 per 1000 bed-days, with over one-third causing injury.
  • Pressure ulcers (bedsores): Affect more than 10% of adult hospital patients and are common and preventable.
  • Patient identification errors: Can lead to severe adverse events such as operating on the wrong patient or incorrect site.
  • Unsafe transfusion and injection practices: Pose significant risks of adverse reactions and transmission of infections (e.g., hepatitis B, C, HIV).

Measurement Considerations and Challenges

• Despite varying figures, even seemingly small proportions of harm equate to hundreds of thousands of people potentially harmed each year.
• The burden of patient harm in LMICs is very likely underestimated due to a lack of high-quality studies and inadequate infrastructure for data collection and record-keeping.
• Currently, there is no internationally agreed measurement strategy to reliably identify and analyze the burden of patient harm.
• There is a need to establish a minimum set of appropriate and feasible standards for measuring patient harm globally and to move from non-systematic methods like voluntary reporting to coordinated systematic measurement.

Financial and Economic Burden of Unsafe Care

Unsafe care incurs considerable financial and economic costs, extending beyond direct medical expenses to societal burdens such as lost productivity and foregone income.

Direct Costs Imposed on Health Care Systems and Budgets

• Unsafe care diverts resources to ameliorate patient harm, including additional diagnostic testing, acute and non-acute care, and administrative actions that would otherwise not be needed. This creates an opportunity cost, as these resources cannot be used for other priority care needs.
• In selected HICs, 12.6% of total health expenditure (approximately US$878 billion annually) is allocated to managing the consequences of patient harm across inpatient, primary, and long-term care settings. This is equivalent to about 1.4% of their combined gross domestic product (GDP).
• The direct financial cost of avoidable harm is estimated to be 8.7% of total health expenditure, or US$606 billion across OECD countries.
• While studies on LMICs are limited, available evidence suggests that direct costs are likely similar to HICs estimates. For example, a study in Thailand found that adverse events in acute care cost 5.5% of the national health budget.
• Costs by setting and source of harm:
  • In acute care, HCAIs can significantly increase patient length of stay and treatment costs. Managing sepsis consumes 2.7% of global healthcare budgets. Hospital-acquired VTEs are estimated to cost US$7–10 billion annually in the United States. Surgical harm is associated with a 2-fold increase in length of stay and a 1.5-fold increase in direct hospital costs.
  • In primary/ambulatory settings, patient harm from ADEs and wrong/delayed diagnosis/treatment leads to significant costs, including additional non-acute care, emergency department visits, and hospital admissions. The combined cost of ADEs across all healthcare settings in OECD countries has been estimated at US$54 billion annually, or 1% of total health expenditures.
  • In long-term care, common adverse events like pressure ulcers and falls account for significant in-patient expenditures.

Indirect Costs of Unsafe Care

• Indirect costs comprise the burden of patient harms on people’s productivity, labor participation, and associated income loss.
• These costs can exceed direct costs by orders of magnitude. For example, a study on surgical site infections in Australian public hospitals found indirect costs (AUD 3 billion / US$1.9 billion) to be nine times higher than direct costs.
• Disadvantaged people are more likely to be disproportionately impacted by the indirect costs of harm, exacerbating inequalities in incomes and poverty.
• An indicative estimate suggests that patient harm may reduce global economic output by up to 0.7% each year.
• If all unsafe care had been eliminated in 2000, the gross world product (GWP) would have been 15% higher two decades later, resulting in a cumulative GWP gain of about US$120 trillion over that timespan.
• Using a willingness-to-pay approach, the indirect cost of patient harm in the United States approaches US$1 trillion per annum. The annual global societal cost of unsafe care is estimated to be US$1.17 trillion, based on a conservative willingness-to-pay value.

Investing in Strategies to Reduce Harm Can Pay High Dividends

• Many existing strategies and interventions to improve patient safety are very cost-effective compared to medical services, offering a high return on investment (ROI).
• Interventions targeting the most costly and harmful events, such as HCAIs, VTEs, medication errors, pressure injuries, and falls, offer significant returns. For instance, interventions targeting HCAIs can deliver a saving-to-cost ratio of 7:1.
• The WHO Surgical Safety Checklist has proven to be an effective and highly efficient tool for reducing surgical harms.
• Preventing pressure ulcers and patient falls also offers excellent value in terms of financial savings and health outcomes.
• Technological interventions like barcodes or computerized provider order entry systems are cost-effective ways to reduce medication errors.
• A comprehensive crew resource management program in a US medical center, costing US$3.6 million, saved between US$12.6 million and US$28 million, demonstrating an ROI of US$3.5 to US$6.8 per dollar over four years. This highlights that systemic patient safety strategies are worthwhile.
• Patient engagement and health literacy are key factors that can decrease harm by up to 15%, offering significant benefits for both patients and health systems.

Reference: Global patient safety report 2024. Geneva: World Health Organization; 2024