This section summarizes the epidemiological data and relevant issues related to the global burden of health care-associated infection (HCAI) and emphasizes the importance of preventing HCAI by giving priority to the promotion of hand hygiene best practices in health care. When available, national or multicentre surveys were preferred to single hospital surveys, and only studies or reports published in English were considered. This overview of available data on HCAI is therefore not to be considered exhaustive, but rather as an informative, evidence-based introduction to the topic of hand hygiene in health care.
HCAI is a major problem for patient safety and its surveillance and prevention must be a first priority for settings and institutions committed to making health care safer. The impact of HCAI implies prolonged hospital stay, long-term disability, increased resistance of microorganisms to antimicrobials, massive additional financial burden, high costs for patients and their families, and excess deaths. Although the risk of acquiring HCAI is universal and pervades every health-care facility and system around the world, the global burden is unknown because of the difficulty of gathering reliable diagnostic data. Overall estimates indicate that more than 1.4 million patients worldwide in developed and developing countries are affected at any time.2 Although data on the burden of diseases worldwide that are published in WHO’s World Health Reports inform HCWs, policy-makers, and the public of the most important diseases in terms of morbidity and mortality, HCAI does not appear on the list of the 136 diseases evaluated.3 The most likely reason is that the diagnosis of HCAI is complex, relying on multiple criteria and not on a single laboratory test. In addition, although national surveillance systems exist in many industrialized countries,4 e.g. the National Nosocomial Infection Surveillance (NNIS) system in the United States of America (USA) (http://www.cdc.gov/ncidod/dhqp/nnis.html), they often use different diagnostic criteria and methods, which render international comparisons difficult due to benchmarking obstacles. In developing countries, such systems are seldom in place. Therefore, in many settings, from hospitals to ambulatory and long-term care, HCAI appears to be a hidden, cross-cutting concern that no institution or country can claim to have solved as yet.
For the purpose of this review on the HCAI burden worldwide, countries are ranked as “developed” and “developing” according to the World Bank classification based on their estimated per capita income (http://siteresources.worldbank.org/DATASTATISTICS/Resources/CLASS.XLS).
3.1. Health care-associated infection in developed countries
In developed countries, HCAI concerns 5–15% of hospitalized patients and can affect 9–37% of those admitted to intensive care units (ICUs).2,5 Recent studies conducted in Europe reported hospital-wide prevalence rates of patients affected by HCAI ranging from 4.6% to 9.3%.6–14 According to data provided by the Hospital in Europe Link for Infection Control through Surveillance (HELICS) (http://helics.univ-lyon1.fr/helicshome.htm), approximately 5 million HCAIs are estimated to occur in acute care hospitals in Europe annually, representing around 25 million extra days of hospital stay and a corresponding economic burden of €13–24 billion. In general, attributable mortality due to HCAI in Europe is estimated to be 1% (50 000 deaths per year), but HCAI contributes to death in at least 2.7% of cases (135 000 deaths per year). The estimated HCAI incidence rate in the USA was 4.5% in 2002, corresponding to 9.3 infections per 1000 patient-days and 1.7 million affected patients; approximately 99 000 deaths were attributed to HCAI.7 The annual economic impact of HCAI in the USA was approximately US$ 6.5 billion in 2004.15
In the USA, similar to the position in other industrialized countries, the most frequent type of infection hospitalwide is urinary tract infection (UTI) (36%), followed by surgical site infection (SSI) (20%), bloodstream infection (BSI), and pneumonia (both 11%).7 It is noteworthy, however, that some infection types such as BSI and ventilator-associated pneumonia have a more severe impact than others in terms of mortality and extra-costs. For instance, the mortality rate directly attributable to BSIs in ICU patients has been estimated to be 16–40% and prolongation of the length of stay 7.5–25 days.16,17 Furthermore, nosocomial BSI, estimated to account for 250 000 episodes every year in the USA, has shown a trend towards increasing frequency over the last decades, particularly in cases due to antibiotic-resistant organisms.18
The HCAI burden is greatly increased in high-risk patients such as those admitted to ICUs. Prevalence rates of infection acquired in ICUs vary from 9.7–31.8% in Europe19 and 9–37% in the USA, with crude mortality rates ranging from 12% to 80%.5 In the USA, the national infection rate in ICUs was estimated to be 13 per 1000 patient-days in 2002.7 In ICU settings particularly, the use of various invasive devices (e.g. central venous catheter, mechanical ventilation or urinary catheter) is one of the most important risk factors for acquiring HCAI. Device-associated infection rates per 1000 device-days detected through the NNIS System in the USA are summarized in .20
Device-associated infection rates in ICUs in developing countries compared with NNIS rates.
In surveillance studies conducted in developed countries, HCAI diagnosis relies mostly on microbiological and/or laboratory criteria. In large-scale studies conducted in the USA, the pathogens most frequently detected in HCAI are reported by infection site both hospitalwide and in ICUs.21,22
Furthermore, in high-income countries with modern and sophisticated health-care provision, many factors have been shown to be associated with the risk of acquiring an HCAI. These factors can be related to the infectious agent (e.g. virulence, capacity to survive in the environment, antimicrobial resistance), the host (e.g. advanced age, low birthweight, underlying diseases, state of debilitation, immunosuppression, malnutrition), and the environment (e.g. ICU admission, prolonged hospitalization, invasive devices and procedures, antimicrobial therapy).
3.2. Burden of health care-associated infection in developing countries
While HCAI surveillance is already a challenging task in highly resourced settings, it may often appear an unrealistic goal in everyday care in developing countries. In addition to the usual difficulties to define the diagnosis of HCAI must be added the paucity and unreliability of laboratory data, lack of standardized information from medical records, and scarce access to radiological facilities. Limited data on HCAI from these settings are available from the literature. This is well demonstrated by an electronic search of the period 1995–2008, which allowed the retrieval of around 200 scientific papers published in English and approximately 100 in other languages.23 Overall, no more than 80 of these papers featured rigorous, high quality, methodological characteristics.
The magnitude of the problem is particularly relevant in settings where basic infection control measures are virtually nonexistent. This is the result of the combination of numerous unfavourable factors such as understaffing, poor hygiene and sanitation, lack or shortage of basic equipment, and inadequate structures and overcrowding, almost all of which can be attributed to limited financial resources. In addition to these specific factors, an unfavourable social background and a population largely affected by malnutrition and other types of infection and/or diseases contribute to increase the risk of HCAI in developing countries.24,25 Under these conditions, thousands of infections – in particular due to hepatitis B and C viruses and human immunodeficiency virus (HIV) transmission – are still acquired from patients, but also from HCWs through unsafe use of injections, medical devices and blood products, inadequate surgical procedures, and deficiencies in biomedical waste management.24
When referring to endemic HCAI, many studies conducted in developing countries report hospitalwide rates higher than in developed countries. Nevertheless, it is important to note that most of these studies concern single hospitals and therefore may not be representative of the problem across the whole country.26–36 For example, in one-day prevalence surveys recently carried out in single hospitals in Albania,36 Morocco,35 Tunisia,34 and the United Republic of Tanzania,33 HCAI prevalence rates were 19.1%, 17.8%, 17.9%, and 14.8%, respectively. Given the difficulties to comply with the USA Centers for Disease Control and Prevention (CDC) definitions of nosocomial infection,37 the most frequently surveyed type of infection is SSI, which is the easiest to define according to clinical criteria. The risk for patients to develop SSI in developing countries is significantly higher than in developed countries (e.g. 30.9% in a paediatric hospital in Nigeria,38 23% in general surgery in a hospital in the United Republic of Tanzania,33 and 19% in a maternity unit in Kenya39).
The burden of HCAI is also much more severe in high-risk populations such as adults housed in ICUs and neonates, with general infection rates, particularly device-associated infection rates, several-fold higher than in developed countries. As an example, in , device-associated infection rates reported from multicentre studies conducted in adult and paediatric ICUs are compared with the USA NNIS system rates.20,40,41 In a systematic review of the literature, neonatal infections were reported to be 3–20 times higher among hospital-born babies in developing than in developed countries.42
A very limited number of studies from developing countries assessed HCAI risk factors by multivariate analysis. The most frequently identified were prolonged length of stay, surgery, intravascular and urinary catheters, and sedative medication.27,30,33–35,43–47
The magnitude and scope of the HCAI burden worldwide appears to be very important and greatly underestimated. Methods to assess the size and nature of the problem exist and can contribute to correct monitoring and to finding solutions. Nevertheless, these tools need to be simplified and adapted so as to be affordable in settings where resources and data sources are limited. Similarly, preventive measures have been identified and proven effective; they are often simple to implement, such as hand hygiene. However, based on an improved awareness of the problem, infection control must reach a higher position among the first priorities in national health programmes, especially in developing countries.
