Quality Care in the ICU1

Just a few years ago, physicians could be confident that they alone had a mandate to judge the quality of care they provided. Now, that mandate is questioned almost daily in boardrooms, legal hearing rooms and even at homes of ordinary people. The very language of current discussions about the quality of care leaves many physicians tongue-tied and bewildered. Observed and expected mortality, case mix, process measures, critical paths and appropriateness criteria belong to the modern jargon that is very poorly understood by doctors. None of these terms showed up in textbooks when most physicians now in practice attended medical college. Few, if any relate to the actual provision of care to an individual patient. No wonder, many physicians react to this conundrum over quality of healthcare with disinterest or skepticism or sometimes even with anger. This is a luxury that doctors in our country, or for that matter anywhere in the world can no longer afford.

Experts have struggled for years to formulate a meaningful definition of quality. In 1980, Donabedian¹ defined care of high quality as “that kind of care which is expected to maximize an inclusive measure of patient welfare, after one has taken account of the balance of expected gains and losses that attend the process of care in all its parts.” In 1984, American Medical Association² defined high quality care as one “which consistently contributes to the improvement or maintenance of quality and /or duration of life.” The association identified specific attributes of healthcare that should be examined to determine quality including health promotion, disease prevention, timeliness, participation of patients, scientific basis of medical practice and efficient use of resources. One of the most widely cited definitions, formulated by the Institute of Medicine in 1990³ holds that quality consists of the “degree to which health services for individuals and populations increase the likelihood of desired health outcomes and are consistent with current knowledge.”4

Deming^4,5 was probably the pioneer of modern quality movement and his work influenced profoundly the Japanese industrial production after Second World War. The Deming philosophy comprises of three key elements: customer satisfaction, employee involvement and continuous quality improvement. He condensed his convictions into a well-chiseled list of 14 points of management or into an easy to remember concept of ‘Plan-Do-Study-Act’ (PDSA) cycle. The healthcare industry unfortunately utilizes only two of Deming's fourteen principles⁶ of quality improvement and standardization. The remaining twelve principles that emphasize the importance of building safe and collaborative environments and using the skill and knowledge of workers to lead improvement are sadly, never applied. Individual doctors on their parts, tend to have an even more narrow focus on the overall quality of healthcare delivery by merely concentrating on the technical quality of their work. This too, doctors seem to do badly.

Although the amount of medical evidence increases daily, many core questions remain unanswered. For instance, an analysis of Dutch College of general practitioners' guidelines identified 875 relevant clinical questions with no answer in published works.⁹ Clinical questions trigger a wide variety of methodological approaches including clinical trials, meta-analysis, decision analysis and systematic reviews. Evidence based medicine relies heavily on randomised clinical trials that merely answer questions of efficacy: In other words, how an intervention works for a specific group of patients with a particular disease. A basic assumption of evidence based medicine is that clinical 5research follows clinical relevance. In reality, that is not the case. More evidence is published on pharmacological treatments than on the effects of simple interventions aimed at changing health-related behaviour. Negative trials are not even published at all. As a result, there is danger of modern day research to pursue what is possible and available rather than what is relevant. Individual doctors find it difficult to translate information from clinical trials across to clinical practice. In part, this is due to the fact that doctors treat a large number of patients with indistinct clinical syndromes in their practice. Research does not include a representative sample of patients with respect to age, ethnic origin, co-morbidity or in the atypical presentation of non specific symptoms. Another drawback is that conclusions from trials are presented to a doctor who looks for guidance in a definite yes-no decision paradigm for a particular patient. Disaggregating research data from a group of patients to an individual requires a thorough understanding of statistics and probabilities. Patients expect certainty from their doctor and want to know whether a treatment will be successful or not. Extrapolation from research to practice presumes that patients are open to a rational understanding about their disease and can make decisions for their own health based upon the available evidence. Many patients in fact, view disease as being due to factors beyond their control and seek solutions beyond what doctors can offer. Even while promoting quality care, doctors can thus be caught in a conflict between their obligation to promote best evidence care and respect for patient autonomy. To pursue the medical benefits over patient's wishes can be viewed as paternalism. On the other hand, respect for patient autonomy can be viewed as overprotection. To bridge the gap between efficacy–what works in isolation under ideal circumstances, and effectiveness–what works in routine practice is sometimes really difficult.

The most cost effective opportunity to improve patient outcomes over the next quarter century is likely to come not from new therapies but from discovering ways to deliver those that are already known to be effective. Result of studies from the USA and Europe^10,11 have suggested that about 30–40% of patients do not receive care according to the available evidence and another 20–30% are provided care that is not needed or is potentially harmful. In his 1972 report, Cochrane¹² concluded that to improve quality, the clinical sector of the NHS should be encouraged with a scientific enquiry in the belief that dissemination of results from clinical trials would directly affect patient care. Thirty years on, Cochrane's suggestion has enhanced the volume of high quality papers published in literature but it has had a rather limited effect on the quality of clinical work.6

In dealing with quality, Intensive Care units (ICU) have a head start over all other areas of a modern healthcare organization. In most ICUs, elements of a quality system have existed for a long time, if only in a more or less informal way.7

Usually several “ground rules” have been written down and appear as policies, good practice guidelines or resident manuals. They are further enhanced by informal or unwritten rules that are often more strictly enforced than the written ones. Outcomes research, that has profoundly influenced our current approach to quality in healthcare was pioneered in ICUs more than 20 years ago.¹⁴ Refinements of outcome prediction models in the ICU continue to drive the research agenda and stimulate the growth of a comprehensive performance appraisal system in healthcare.

A quality control programme must first set a goal and then remain focussed on while developing a pathway for implementation based on local expertise and needs (Table 1.2).

The need for improvement in any clinical environment usually serves one of two purposes;

The same diagram also portrays the possibility that a change in the process of care (ie change in sedation practices) can profoundly influence the output. In addition, the environment within which these processes take place also affects the interaction of the variables. There are changes that develop from internal or external sources that make us review the process of care and also the effectiveness of our outcomes. Leaders are typically responsible for scanning the environment for these external forces and their interpretation has a large impact on whether or not a change is needed. Examples of the external environment that force a transformation includes, amendment in government policy, new research, advances in technology or changes in the demography of the population. More importantly, the internal environment plays a major role. Internal forces include issues like the availability of ICU beds, admissions policy, model of care giving, “ownership” of patients and remuneration schemes.

Quality initiatives encompass three areas: structures, processes and outcomes. For each area, key instruments assess quality as depicted in Table 1.3. Since outcome will be discussed in some detail in another part of this book, the present chapter will focus on the structure and process of intensive care. Of course some overlap is inevitable as the sheer volume of research and the overwhelming influence of outcome models is hard to ignore.

Structure denotes the attributes of the settings in which care occurs (Table 1.4). This includes the attributes of material resources such as facilities and equipment and of human resources such as the number of personnel available and their individual professional characteristics.10

For the healthcare system, organizational aspects such as accessibility and continuity of care affect quality, as do characteristics of staff organization, methods of peer review and reimbursement procedures. The only aspects of care that a clinician controls directly in an ICU, is the structures of care giving and to a certain extent the processes. Therefore to effect a positive change in practice, medical care must focus upon the organizational structure and processes of ICU care giving.

Process refers to all interventions and interactions between patients and providers (Fig. 1.4). Process quality is the sum of all measures taken or not taken during patient care, including the timing of all activities, total number of incidents, delays, misunderstandings, confusion and unforeseen events. Process quality largely depends on good decision making, communication skills and a good work ethic. Guidelines or protocols or algorithms that underpin process quality are increasingly based on scientific evidence and are now available widely. Indicators are representative markers of performance, looking mainly at processes or outcomes and not the totality.

In spite of being surrogate markers of overall quality, indicators are particularly powerful and widely used. In order to be useful, indicators must evaluate a frequent or a troublesome situation and give clear answers.

Outcomes The issue of ICU outcomes and its relevance to quality is covered in another chapter of this book. Suffice to add that a focus on outcomes is more useful in a situation wherein an audit of processes is difficult. Since an optimum treatment process requires a strong scientific base, the applicability of a process-based audit in the ICU is limited. The evidence base for intensive care is fairly small and the assumption that best practice is associated with good outcomes remains an untested hypothesis. Consequently, an outcome feedback model of audit is more useful. Variation in outcomes across different ICUs may arise due to a problem in the delivery of care in a given setting and can spur an investigation into its cause. Used in this way, outcomes research can be a powerful quality control tool.

Recently, Berenholdtz et al¹⁵ systematically reviewed the literature and selected 66 articles published from 1991 onwards in order to list quality indicators that can be applied to the ICU. The authors identified six outcome measures, six process measure and six access measures including a few complication measures.

Standards and accreditation Accrediting bodies play an important role in assessing the infrastructure of healthcare and can do the same for an ICU. Standards related to space and physical layout as well as staffing of an intensive care unit have been laid down by the American, Australian and European societies. It is therefore possible to have a mechanism by which different units are assessed, graded and accredited to perform at different levels of intensive care. It is desirable that this should be done in order to match the resources of an ICU to the acuity of an individual patient. In other words, a sicker patient should be managed in a tertiary care ICU where facilities exist for advanced life supportive therapy.

Guidelines and Audit Guidelines and protocols eliminate unnecessary variation in performing patient care routines in the ICU. Guidelines are also based on the most current available evidence and their implementation in an ICU, ensures that all healthcare workers are at least aware of the latest standards related to patient care. In addition, safety of the patient is not compromised during the process of ICU care. 12Audits can be developed to make sure that protocols are followed correctly within an ICU. An audit of the process and procedure for elective insertion of a central line is an example of how audits can help in improvement of quality.

Outcome indicators and Benchmarking One method of comparing ICU performance is benchmarking in which the rates of certain events are compared across ICUs with a similar style of governance and patient acuity. Although benchmarking is gaining ground, the level of acceptable risk (or rate) and its range are difficult to define. Furthermore, no two ICU systems work identically and the process of recording events also vary widely. A simple way to start a quality control programme in an ICU is to benchmark the ICU mortality rate to a known standard. This can then be extended to involve the average stay of a patient in an ICU, number of days on the ventilator and even the rate of adverse events.

Incident reporting can become a key element of process control if combined with a proper ICU management structure. Incident reporting can be practiced for very specific processes of intensive care. Adverse drug events, medication errors can be identified and prevented by a systematic analysis of weaknesses in the system. In addition, critical incidents like unplanned extubations, failed extubation, readmission of ICU patients or death of a patient within 24 hours of ICU discharge are indicators of ICU quality. Reporting of incidents is covered in greater detail in the chapter on errors in medicine.

Infection Control Intuition suggests that nosocomial infection rate may be a good indicator of quality care in the ICU and the occurrence of resistant organisms in ICU patients may be an marker of deficiencies in the process of care. Some infection related complications in ICU patients are due to factors associated with a patient's illness while others are clearly due to care related factors. Amongst the latter, some are preventable and others are not. Although there is some controversy about which ICU infection is a robust indicator of quality, in large parts of the world, colonization or infection of a patient with a multi drug resistant strain is considered a quality tool. Likewise, catheter related blood stream infection is also a good indicator of ICU performance.

Employee satisfaction is the cornerstone of a high quality system. There are no ways of assessing overall staff satisfaction levels but some indirect indicators can be used. Frequent absence is usually a clear hint of dissatisfaction and employee turnover or burnout is an indicator of a badly performing organization. The amount of time on sick leave, nursing turnover and physician burnout are clearly indicators of process quality.

A great deal of anecdotal evidence suggests that many quality assurance programmes with their emphasis on finding error have been counterproductive. In part, the impetus for uncovering medical errors has its roots in litigation that has now become a global 13phenomenon and threatens the very survival of the medical profession. The imposition of punitive and sometimes humiliating sanctions accompanied by enthusiastic media reporting has made it difficult for the practicing doctor to squarely face the problem of medical error. As a result, many in the profession remain convinced for the need to improve error rate but remain apprehensive of how to bring about this change without exposing themselves and others to the risk of a law suit. Although courts and consumer organizations handle genuine grievances, the pendulum has now swung too far in certain countries where the consumer and the lawyers see litigation as a means of making money out of the deep pockets of the healthcare industry.

As Berwick has noted,¹⁶ the development and widespread application of quality management methods has paralleled the rise in the US, of commercialization in the medical marketplace. Much of what passes for quality improvement strategy can justifiably viewed as thinly veiled attempts at cost containment. When hospital teams for instance, engage the attention of doctors to focus on reducing hospital stay, physicians may understandably feel that quality of care is truly not a priority. At other times, those promoting quality management initiatives may seem overtly interested in using them to reduce expenditure and thereby improve profitability. Thus, in their efforts to reduce costs, some healthcare organizations may pursue options that pose unacceptable threats to the quality of care. Generally speaking, doctors are unwilling to be part of such arrangements. There is nothing wrong with efficient use of resources to sustain the financial viability of an organization; in fact, in most circumstances it is desirable. To extend the argument and impose all round cost saving reforms that unduly compromise welfare of patients is clearly undesirable and erroneous.

The international standards organization (ISO) developed quality improvement standards, initially for industrial production and later for service industries. ISO certification awarded after a certification process is held in high esteem in industrial settings. Increasingly, there are attempts to apply the ISO 9000 concept to hospitals and healthcare institutions under the unproven assumption that it will work. So far, ISO 900 applications in hospital have only addressed hotel, housekeeping and other functions, thus scratching at the surface of a complex problem. The core issue of healthcare and quality in the provision of services has not even been tackled. Doctors believe that such programmes rarely deal with issues regarded as important for patient care. Traditionally, efforts have focussed on checking documentation, studying safety at the work place, reviewing the role of committees and the like. Although important, they rarely answer real questions about health care.

Intensive Care is part of a continuum of progressive patient care and only one component of overall care provided in a healthcare organization. The real value of a hospital (or an ICU) is the effectiveness with which it either improves outcomes or decreases costs for a particular disease episode. In raw economic terms, a high performance system is one that admits and discharges patients speedily, has low severity adjusted mortality rates, good functional outcomes in survivors and an efficient use of resources. Whereas, administrators emphasize these essential elements of a new “healthcare culture”, patient view a healthcare organization differently.