Biosurveillance Of The Healthcare System

A healthcare-associated infection (HCAI) is an infection that develops in a healthcare setting such as a hospital or as a result of medical treatment. HCAIs are also known as nosocomial infections. HCAI is a significant problem in healthcare. In 1992, the CDC estimated that there are at least two million HCAIs in hospitalized patients alone each year in the United States, costing $4.5 billion and causing 90,000 deaths, a third of which are probably preventable (Anonymous, 1992).5 Roughly an equal number of infections occur in long-term care facilities, dialysis centers, clinics and other settings (Martone et al., 1998).

The rate of HCAIs in hospitals has remained steady at approximately 5% of patient admissions for at least the past three decades (Haley et al., 1985a,b; Martone et al., 1998).This lack of improvement does not reflect inattention or lack of improvement in methods of prevention, but rather imperfect implementation of known measures such as hand washing, the relentless evolution of microorganisms, the severity of illness of patients, and the increasing complexity of medical treatment.

To put these statistics in perspective, Florence Nightingale, Ignaz Semmelweiss, Joseph Lister, and Oliver Wendell Holmes lived in eras in which 20% to 30% mortality from

5 These oft referenced statistics stem from data and analyses by Haley RW, Culver DH, Morgan WM, Emori TG, Munn VP, Hooton TP. The efficacy of infection surveillance and control programs in preventing nosocomial infections in U.S. hospitals. Am J Epidemiol 1985; 121:182-205 and Martone, W. J., Jarvis, W.R., Edwards, J.R., Culver, D.H. and Haley, R.W. (1992) In Hospital Infections, Third Edition (Eds, Bennett, J.V. and Brachman, P.S.) Lippincott-Raven Publishers, Philadelphia.

HCAIs was not uncommon.6 These individuals, incidentally, pioneered methods such as hand hygiene that now form the basis of modern infection control.

Outbreaks also occur in hospitals, but they are infrequent and account for only 2% to 3.7% of HCAIs in hospitals (Wenzel et al., 1983; Haley et al., 1985b). Haley et al. (1985b) estimated that a typical community hospital has one nosocomial outbreak per year. The types of outbreaks change over time as organisms and medical technology changes. Contaminated products (e.g., blood) and medical devices are common causes of recent outbreaks investigated by the CDC (Martone et al.,1998).

7.1. Infection Surveillance Control Programs

An infection surveillance control program (ISCP) is a division of a healthcare organization with the mission "to identify and reduce the risks of acquiring and transmitting infections among individuals served, staff, contract service workers, volunteers, students, and visitors.'' (JCAHO, 2004). Among the responsibilities of an ISCP is biosurveillance of the organization.7 In particular, an ISCP is responsible for collection, analysis and interpretation of infection control data, and the investigation and surveillance of suspected outbreaks of infection.

The origins of ISCP can be traced to a pandemic of staphylo-coccal infections in hospitals in the mid 1950s in the United States. In response to this problem, hospitals organized infection control committees. Over the ensuring decade, a few hospitals developed organized infection control programs, initially using physicians and then adding trained infection control nurses. By the mid 1970s, most hospitals had adopted this practice.

A typical ISCP consists of one or more doctors and nurses or medical technologists with specialized training related to epidemiology of hospital infections and disease prevention. These individuals have expertise in the recognition of disease in individual patients as well as recognition of outbreaks in the hospital population and in their prevention and control. The APIC created the Certification Board of Infection Control, which certifies infection control practitioners (Sheckler, 1998).

Approximately one-fourth to one-half of hospital HCAIs come to the attention of ISCP as a result of laboratory testing.

The rest are identified by a variety of formal and informal surveillance activities. A typical ISCP identifies patients via a daily printout of "positive cultures'' from an electronic laboratory system. The ISCP may also obtain a list of new prescriptions for antibiotics. A good ISCP also requires "shoe-leather" epidemiology (i.e., daily ward rounds to speak with personnel providing direct patient care) and some form of "post-discharge surveillance" to detect infections in patients who have already left the hospital. The combined result of all these processes is a list of potential patients to investigate that day. The staff reviews this list to organize and prioritize the work for the day. The staff collects additional information for each patient from hospital information systems, for example, accessing a single system or multiple systems to review radiology reports, physician dictations, medication records, and other results of laboratory testing. In addition, the staff may read the paper record of a patient or speak with physicians and nurses caring for a particular patient.

To satisfy reporting requirements, the staff may notify governmental public health when a patient with a reportable disease is found. To satisfy JCAHO requirements (discussed below), they compile periodic reports.

Prevention of infections in the healthcare setting requires cooperation of virtually all divisions and individuals. A list of related departments created by JCAHO identifies central sterile processing, environmental services, equipment maintenance personnel, facilities management (including engineering), housekeeping, information management, laboratory, medical staff, nursing, and pharmacy.

7.2. JCAHO Infection Control Guidelines

JCAHO establishes guidelines for patient safety that include guidelines for infection control. Infection control is one of JCAHO's 14 priority focus areas.

JCAHO is widely recognized for its leadership role in developing standards and performance measures and for the adaptability of its rigorous evaluation processes to emerging new forms of healthcare delivery. JCAHO evaluates and accredits more than 15,000 healthcare organizations and programs in the United States. An independent, not-for-profit organization, JCAHO is the nation's predominant

6 Ignaz Philipp Semmelweis (1818-65), a Hungarian obstetrician, introduced antiseptic prophylaxis into medicine. In the 1840s, puerperal or childbirth fever, a bacterial infection of the female genital tract after childbirth, was taking the lives of up to 30% of women who gave birth in hospitals. Women who gave birth at home remained relatively unaffected. Semmelweis observed that women examined by student doctors who had not washed their hands after leaving the autopsy room had very high death rates. When a colleague who had received a scalpel cut died of infection, Semmelweis concluded that puerperal fever was septic and contagious. He ordered students to wash their hands with chlorinated lime before examining patients; as a result, the maternal death rate in his hospital was reduced from 12% to 1% in two years. Source: Copyright 1999,2000.

7 The other responsibilities of a ISCP include (1) planning, implementation and evaluation of infection prevention and control measures; (2) education of individuals about infection risk, prevention and control; (3) development and revision of infection control policies and procedures; (4) management of infection prevention and control activities; (5) provision of consultation on infection risk assessment, prevention and control strategies. Source:

standards-setting and accrediting body in health care. Since 1951, JCAHO has maintained state-of-the-art standards that focus on improving the quality and safety of care provided by healthcare organizations. Infection control is a critical component of safe, quality health care.

JCAHO is becoming more active, perhaps even militant, as a result of increasing awareness of the impact of HCAIs on the cost and quality of health care. Effective January l, 2005, JCAHO established a New Patient Safety Goal (the seventh) in the area of infection control, which it promulgates as a set of standards that includes the following: accountability of the CEO of a healthcare organization for compliance and fiscal support of an ISCP; staffing and training of ISCP; communication and coordination with health departments and other community organizations; clear command and control (delegated authority); and surveillance, and monitoring of efficacy if its infection control programs (JCAHO, 2004).

JCAHO is very influential. As previously discussed, its influence derives from the Medicare Act of 1965, which included JCAHO accreditation as one basis for Medicare reimbursement. The most recent Accreditation Manual for Hospitals issued by JCAHO includes a set of scoring guidelines on which the compliance of a hospital will be judged. To obtain the highest score, a hospital must provide evidence of having switched from processes such as surveillance of antibiotic use and nosocomial infections as ends in themselves to measures of patient outcomes as indicators of hospital per-formance.To link patient outcomes such as length of stay, days or morbidity, mortality, and costs will require substantial collection of data, or integration of data, already being collected by registration systems and billing systems.

7.3. Information Systems in Infection Control

There is and will be an increasing trend to support ISCP with IT, although the barriers to comprehensive support of all functions are high.

At present, ISCPs use computers to manage surveillance data in two ways: the most common use is to store and analyze surveillance data that are collected manually. We refer to such systems as free-standing. Far less commonly, ISCPs computerize the actual collection of surveillance data. We refer to these ISCP systems as embedded because they receive data directly from clinical information systems.

7.3.1. Free-Standing ISCP Systems

An ISCP may utilize general purpose software such as Microsoft Excel, SAS, Microsoft Access, Microsoft SQL Server, and Oracle to store and analyze surveillance data. It may use software specifically designed for infection control such as AICE, NNIS-IDEAS, QLOGIC II, Epidemic Information Exchange (Epi-X), and WHOCARE

The practice of analyzing ISCP data by using free-standing computers is nearly ubiquitous because of the low cost of computers and their ability to facilitate statistical analysis and report generation. However, ISCPs still rely heavily on paperand card-based systems. An ISCP that collects surveillance data on paper may subsequently enter the data into a computer system simply for analytic purposes. Manual collection of data with subsequent entry into computers for storage, analysis, and report generation is by far the more common use of computers in ICSP.

7.3.2. Embedded ISCP Systems

A small but growing number of health systems have deployed embedded ISCP systems, motivated by research goals and or the potential to improve the cost-efficiency and efficacy of ICSP (although the initial cost is high).

In the 1990s, a group of researchers at the University of Utah developed a program called Antibiotic Assistant, which was a module in the HELP clinical information system operating at LDS hospital (Evans et al., 1985,1986,1992,1998; Burke et al., 1991; Evans, 1991; Gaunt, 1991; Classen et al., 1992; Rocha et al., 1994; Chizzali-Bonfadin et al., 1995; Classen and Burke, 1995; Fiszman et al., 2000b). This research demonstrated new types of hospital infection control functionality that access to clinical information systems made possible (e.g., reminders to administer preoperative antibiotics) as well as their efficacy. An offshoot of this research was Theradoc, Inc., which has commercialized this technology.

Also in the 1990s, researchers at Barnes Jewish Christian (BJC) Hospital in St. Louis developed the GermWatch and GermAlert systems (Kahn et al., 1993,1995,1996a,b). Similar to Antibiotic Assistant, these systems collect surveillance data automatically from clinical information systems.They use a rule-based expert system (see Chapter 13) to detect patients of interest to ISCP. These systems are still in use at BJC Health System.

Brossette and colleagues explored the use of computers to perform brute-force search through routinely collected data (also known as data mining) to detect changes in rates of infection in subpopulations (e.g., patients in intensive care units) (Brossette et al., 1998; Moser et al., 1999; Brossette et al., 2000). An offshoot of this research was MedMined, Inc., which has commercialized this technology.

Although the above systems have demonstrated the feasibility of automatic data collection, their market penetration remains low owing to their cost and lack of definitive economic data showing direct benefit to healthcare systems.

7.4. Challenges to Automating ISCP

The challenges to integrating patient data for biosurveillance in the hospital are identical or greater than are those for integrating biosurveillance data for public health surveillance. The difficulty is slightly greater because the set of diseases of concern in ISCP are a superset of those of concern in public health practice. Not only must the healthcare system report notifiable conditions to governmental public health, but it is encouraged by JCAHO to monitor for urinary tract infections, pneumonia, multiple drug-resistant organisms (e.g., methicillin-resistant Staphylococcus aureus and vancomycin resistant enterococci), surgical site infections, infections related to implanted devices, needle-stick injuries in staff, and infections within immunocompromised patient populations. JCAHO also encourages healthcare organizations to monitor health outcomes in addition to infection control processes. To automate this type of analysis, an organization must integrate data about processes (e.g., which surgeon performed which procedure on which patient on which date) with data about outcomes (e.g., infection rate, length of stay, and hospital costs).

There are healthcare systems that have accomplished such integration. However, they are the exception rather than the rule. These healthcare systems had already achieved a high level of information system integration for other reasons. They had sufficient medical informatics expertise and grant funding to integrate the systems. They are to some degree the same organizations that we discuss in the next section on regional health information organizations (RHIOs).This overlap is not coincidental. The same IT infrastructure that is a prerequisite for automated sharing of clinical information among hospitals is required for automation of hospital infection control.

7.5. Hospital Biosurveillance as a Model of an Ideal Biosurveillance System

It is interesting to note that biosurveillance in hospitals is more intensive than in the general population. In fact, many of the ingredients of an ideal biosurveillance system (see Chapter 13) are already in place in modern hospitals: highly trained clinicians evaluate every patient every day, patient's temperatures are taken regularly, and surveillance data are available electronically in real time. The few missing ingredients include surveillance of the staff and visitors (who are part of the population) and real-time information about patterns of disease in the community, including other hospitals and long-term care facilities from which patients are transferred. Nevertheless, the ideal biosurveillance system that we will discuss in Chapter 13 has its most complete realization in the modern hospital.


Two important trends in clinical computing are (1) the use of application service providers (ASPs), and (2) regional integration of healthcare data for the improvement of clinical care.

8.1. Application Service Providers

ASPs are companies that are in the business of hosting computer applications in central locations. A healthcare system may contract with an ASP to outsource some or all of its data processing.

Clinicians interact with the server-based applications over private or public networks. The relevance of an ASP for regional or national biosurveillance is that an ASP may, after obtaining appropriate legal and administrative permission, provide data collected by many healthcare systems.The physical colocation of hundreds of clinical information systems in a single location is helpful but it represents a 20% solution, with the residual 80% comprising unaddressed confidentiality, organizational, vocabulary, and other data integration issues.

8.2. National Health Information Infrastructure

The NHII is an initiative of the U.S. government whose goal is to promote the use of IT by the healthcare system. The government, in particular, hopes to improve the quality of medical care while also reducing its cost.8

Importantly, NHII understands the importance of biosurveillance (National Committee on Vital and Health Statistics, 2001, Thompson and Brailer, 2004). The objectives of NHII relevant to biosurveillance are (1) increasing the adoption of electronic medical records, (2) promoting the exchange of data among various healthcare organizations, and (3) improving public health (Thompson and Brailer, 2004).

8.2.1. Increasing Adoption of Electronic Medical Records

The relevance of the first objective to biosurveillance is that data that currently exist only on paper would become available electronically. The federal government, as part of its NHII initiative, has taken several actions to promote adoption of electronic medical records.

The Center for Medicare and Medicaid Services (CMS) announced in July 2005 that it will make available to physicians a free electronic medical record called Office-VistA, which is based on the electronic medical record, VistA, used by VA hospitals throughout the United States. Because the VA also provides outpatient care in clinics located in its facilities, VistA has significant functionality for outpatient offices. CMS is working with the VA to create Office-VistA from VistA by removing inpatient functionality and making it easy to install.

In 2004, CMS initiated a pilot program called Doctor's Office Quality (DOQ)-IT.9 As part of the DOQ-IT pilot, four Quality Improvement Organizations (QIO)10 in four states received contracts to assist physicians with selecting and implementing EMRs. Based on the pilot, CMS has subsequently funded the QIO in every state to assist physicians with adoption of EMRs, with the sum of all QIO contracts totaling $120 million (Monegain, 2005).

8 Similar initiatives exist in other countries, inculding the United Kingdom, Australia, and Canada.

9 See

10 The Peer Review Improvement Act of 1982 created Quality Improvement Organizations to improve the quality of care received by Medicine beneficiaries, ensure that beneficiaries receive only medically necessary care, and handle individual beneficiary issues such as complaints about care received.

In 2004, the Secretary of Health and Human Services (HHS) created an exception to the Stark law for the development of a "community-wide health information system.'' The Stark law is a federal statute that prohibits physicians from referring Medicare patients to a facility with which they have a financial relationship.11 It has been an obstacle for hospitals that wished to provide associated outpatient practices with EMRs because providing a practice with an EMR creates a financial relationship under the law. However, the Stark law includes a provision that permits the Secretary of HHS to exempt a specific financial relationship if he or she determines that the relationship does not pose a risk for abuse. Thus, the Secretary of HHS in 2004 made an exception for provision of an EMR when the EMR is necessary to connect to a community-wide health information system.12 This action by the Secretary of HHS may encourage the development of community-wide efforts to exchange patient data from the outpatient setting and the provision of EMRs to physicians by organizations that participate in the community-wide effort.13 We discuss some of these community-wide efforts next.

8.2.2. Promoting Exchange of Healthcare Data

Central to the NHII effort is the concept of a RHIO,14 A RHIO is typically a nonprofit organization founded by a mul-tistakeholder group in a single metropolitan region or state. Its mission is typically to develop electronic exchange of patient data (both clinical and administrative) among its member organizations. The NHII concept also includes inter-RHIO data exchange so that when patients travel or move from one region to another, their medical records are available to treating physicians and other authorized parties.

The organizations that participate in a RHIO vary, but most often they include health plans, hospitals, and physicians. Other organizations that participate less frequently include pharmacies, commercial laboratories, diagnostic imaging centers, nursing homes, and government agencies such as health departments.

The RHIO movement can be traced to the Community Health Information Network (CHIN) movement that began and largely ended in the first half of the 1990s. CHINs had similar goals as today's RHIOs: electronic exchange of health care data to support patient care. The CHIN movement largely collapsed because of lack of trust among competing organizations, concerns about privacy of data, failure of the technological approach of creating a centralized database, and the cost of technology at that time ( Appleby, 1995; Starr, 1997; Payton and Ginzberg, 2001; MacDonald and Metzger, 2004).

The RHIO movement has better prospects for success because the federal government is providing incentives and addressing the aforementioned problems that CHINs encountered. At present, federal incentives to RHIOs have mostly come in the form of grant funding. The Agency for Healthcare Research and Quality (AHRQ) has provided nearly $150 million in grant funding to support healthcare data exchange. Federal efforts also include promoting healthcare IT standards (for details on federal efforts to promote the creation and adoption of IT standards necessary for data exchange, see Chapter 32).

The RHIO movement is in its infancy. Overhage et al. (2005) report the results of a recent survey of RHIOs that identified only nine operational RHIOs out of 134 that responded. A majority of RHIOs that provided information to the survey did not yet have substantial commitment from the leadership of the various organizations involved. Nearly one-third of these RHIOs had no funding. The most common technological approach was a centralized database, which was a cause of failure during the CHIN movement (MacDonald and Metzger, 2004). The report notes that a federated database is a characteristic of successful RHIOs.15 The report judged the RHIOs' plans for implementing data exchange as overly ambitious in general.

Table 6.3 summarizes states in which a single RHIO is attempting to integrate clinical data across a whole state. These statewide RHIOs are also in a state of infancy, with only four of the RHIOs actively exchanging data (three are exchanging clinical data). Two-thirds of the RHIOs are new, having formed only in the past two years.

11 The rationale for this law is that physicians might refer patients to facilities with which they have a financial relationship (such as an ownership relationship) even when it is not in the best interest of the patient.

12 There are other qualifications on the exception, such as the community-wide health information system must be available to all physicians who wish to participate, the party providing the EMR cannot take referrals into account when deciding which physicians to give an EMR, and the arrangement cannot violate the Anti-Kickback Law (another law that regulates hospital-physician relationships).

13 Many observers have noted that this exception, although important, may not be sufficient to spur the provision of EMRs to physicians because it does not define "community-wide health information system" or change the Anti-Kickback Law, which is another legal barrier to hospitals providing physicians with EMRs.

14 Also known as a local or regional health information infrastructure.

15 A centralized database is a single collection of patient data from all the healthcare organizations in the community. A federated database, on the other hand, is a system of sending data about patients from one organization to another only when there is a legitimate request.

TABLE 6.3 Regional Healthcare Organizations



Pregnancy And Childbirth

Pregnancy And Childbirth

If Pregnancy Is Something That Frightens You, It's Time To Convert Your Fear Into Joy. Ready To Give Birth To A Child? Is The New Status Hitting Your State Of Mind? Are You Still Scared To Undergo All The Pain That Your Best Friend Underwent Just A Few Days Back? Not Convinced With The Answers Given By The Experts?

Get My Free Ebook

Post a comment