Health care quality can be defined as providing safe and effective care that improves outcomes, optimizes health, and results in high patient satisfaction and quality of life. As such, delivering high quality of care should be a top priority for all rehabilitation professionals. In the past, the focus of practitioner and payer alike has been on efficiency of care and timely and appropriate reimbursement for services provided. However, in the current environment of health care reform in the United States, the focus has shifted from the fee-for-services model, in which volume of health care services rendered was rewarded, to emphasis on quality, safety, and outcomes as top priorities. Thus, the evaluation, measurement, and reporting of quality and outcomes have become the business of not only every health care institution and hospital but also every individual health care provider. This chapter discusses health care quality, evidence-based medicine (EBM), clinical practice guidelines (CPGs), outcome measures, performance measures, practice improvement, patient safety, and accreditation.
Access, affordability, and high quality in health care are the main objectives of the Patient Protection and Affordable Care Act of 2010. To accomplish these goals, the U.S. Department of Health and Human Services (HHS) has outlined a National Quality Strategy (NQS) that was developed through a collaborative and participatory process including input from a wide variety of stakeholders from all over the health care industry. Each agency with the HHS, including the Centers for Medicare and Medicaid Services (CMS), is required to develop a quality strategy and report on its progress and development to the HHS. The quality strategy of the CMS is depicted in Figure 7-1 . One of the three key aims of the NQS is “better care,” which is about improving quality of care and safety. To achieve these goals, the CMS and other health care insurance providers rely on simple, meaningful, efficient, and accurate measurements of health care quality and safety delivered by providers.
In addition to the CMS, several other governmental and many nongovernmental agencies monitor and oversee health care quality and safety. For example, the National Committee for Quality Assurance (NCQA) is a nonprofit organization focusing on health care quality. Health insurance plans all over the United States seek and obtain the NCQA seal of approval, which is a reflection of their high-quality health care and service. According to the NCQA, accredited plans cover 109 million Americans, or 70.5% of all Americans enrolled in health insurance plans. The quality standards and performance measures used by the NCQA and other similar agencies are tools that individual clinicians and organizations use to self-report the quality and safety of health care they delivered. These measures are then used (1) to compare against local, regional, or national standards; (2) by consumers (patients), media, and other members of the public to compare clinicians, plans, and institutions against each other; and (3) compared with similar scores over time to identify opportunities for improvement, such as designing practice and performance improvement projects.
An important concept in health care quality is the concept of value. Despite a common objective of improving performance in health care delivery, there are several stakeholders in health care, sometimes with conflicting goals. This can lead to divergent approaches to improving health care quality, resulting from a lack of clarity and cohesion among stakeholders, resulting in slow rate of progress. Porter makes a case for achieving high value to be the overarching goal of health care delivery, regardless of your perspective or role. Value, in this instance, was defined as health outcomes achieved per dollar spent. Health care outcomes were divided into three tiers: tier 1, health status achieved or retained; tier 2, process of recovery; and tier 3, sustainability of health. Porter recommended that measuring “valuable” medical outcomes in any medical condition or disease must include at least one measure in each tier.
Quality of care is a measure of the performance on the six Institute of Medicine’s (IOM) specified health care aims: (1) safety, (2) timeliness, (3) effectiveness, (4) equity, (5) efficiency, and (6) patient centeredness. Safety refers to “freedom from accidental injury” and it entails reducing errors. An error is the failure of a planned action to be completed as intended or the use of a wrong plan to achieve an aim. Based on performance metrics, providers, individual, organizations, and health plans are incentivized to improve quality through accountability, thereby achieving the IOM’s six aims.
Efforts to accurately and efficiently measure quality continue to grow rapidly. These measures are introduced to and adopted by health plans and clinicians based on several factors. Quality measures need to be meaningful and relevant to patients and society, as well as valid and based on sound evidence. The measurement of quality needs to be transparent, fair, and reliable. Reliability prevents accidents by good design and management to reduce conditions that cause errors. Transparency supports safety by making the reliability of organizations visible by offering comparisons through public reporting of performance measures. Quality measurement must not place too much additional burden on clinicians, their institutions, or health plans. And, providing high quality and safe care has to be appropriately rewarded and incentivized.
Figure 7-2 illustrates a model of how evidence-based clinical research translates into quality measures that eventually lead to health improvement. Ideally, all quality measures are based on the highest possible quality of scientific evidence but, in reality, some measures are adopted because they are in the best interest of consumers and the society at large, even if they are based on best available evidence and not the best possible evidence.
The ultimate objective is to deliver high-quality health care and rehabilitation and to improve the health of the population we serve. What one can do to turn the gears of the quality machine depends on the role they plan in the quality arena. One way to conceptualize the interplay between how efforts at providing high-quality health care can vary depending on one’s perspective and role in the health care system is illustrated in Figure 7-3 . Each aspect of quality and outcomes pertains to the practice of physical medicine and rehabilitation (PM&R).
Outcome Measures
Broadly speaking, an outcome is the result or end product that follows an activity, intervention, or situation. In health care, an outcome is the health consequence resulting from health care provided. The National Quality Measures Clearinghouse of the Agency of Healthcare Research and Quality (AHRQ) defines a clinical outcome as “a health state of a patient resulting from health care.” Thus, an outcome measure is the assessment of clinical outcomes through qualitative or quantitative means. Outcome measurement assesses what is important to the patient, provider, and health system and includes measurement of positive outcomes such as improved function and negative outcomes such as hospital-acquired skin breakdown. There are other outcomes that are the result of health care such as financial burden or family impact that are secondary to the primary clinical outcomes.
Types of Outcome Measures
Outcome measures should be distinguished from process measures. In 1966, Donabedian conceptualized how information regarding health care quality may be gathered and modeled three dimensions of health care: structure, process, and outcome. The structure is the context in which health care is delivered; the buildings, equipment, resource, and the organizational characteristics. The structure of health care is often implicated as an upstream cause of a problem. For example, the current payment structure for health care services rewards interventions (higher payments for injections than regular office visits), which may incentivize providers to recommend and perform interventions. The process of health care is the additive sum of all actions and activities involved in the delivery of health care services. This includes how patients access health care, how patients are assessed and diagnoses made, the way in which treatments are rendered, and the activities involved in counseling, health education, and health promotion. For example, the process of making a diagnosis involves information gathering, conducting an examination, synthesizing data, eliciting additional information in the form of testing or additional consultations, and making an assessment. Finally, from the Donabedian perspective, the outcome is the effect of health care on a patient or a population. Because it is difficult to ascertain if an outcome is the direct result of health care, it is often challenging to accurately measure if, or how much of, an outcome is attributable to an intervention or health care process. Nonetheless, achieving a specific outcome is the reason why specific health care processes take place and therefore is essential to measure. Figure 7-4 illustrates the Donabedian concept with some rehabilitation relevant examples included in each dimension.
When considering the use of outcome measure(s), it is important to determine if the measure has been subject to evaluation of validity (how well the test measures what it intends to measure) and reliability (the degree to which the measure produces consistent and stable results). Consider the following example: a patient with spinal cord injury is diagnosed with a urinary tract infection (UTI) based on the presence of greater than 100,000 colony-forming units (CFU) on urine culture. The patient is prescribed a course of antibiotics and on repeat urine culture 2 weeks later, greater than 100,000 CFU were again found. The goal of the treatment with antibiotics was to eliminate the UTI. Based on the outcome measure of choice in this example, that is, repeat urine culture (a reliable and valid way to measure UTI), the outcome was not achieved. Thus, why was the desired outcome not achieved? It could be because of a deficit in the health care process, such as not prescribing an antibiotic to which the bacteria was susceptible or intending to call in a prescription to a pharmacy but not actually doing it. Alternatively, the patient may have decided to take the prescribed antibiotic for a day and then stop. Or, as is often the case, it is a combination of health care factors and patient factors that impact outcomes. Even in the situation where the patient prematurely stopped taking the prescribed medication, there may have been health care process factors contributing to the poor outcome such as not explaining how to take the medication or not labeling the medication in the patient’s native language. Similarly, other health care processes could have been put in place to improve adherence, such as a telephone or a text check-in with the patient. As evidenced by this example, achieving a particular health care outcome, even when the management plan is clear and the science behind the treatment is solid, is challenging.
Because attribution of outcomes to particular processes is challenging, many providers and hospitals prefer to measure processes instead of outcomes. Processes are easier to control and often are easier to measure. Often a process measure is used as a proxy for an outcome. Measuring 30-day readmission rates is an assessment of process but tends to be understood as representative of poor health outcomes. Thirty day readmission rates indicate that the health status of the population served was not stable enough and that the hospitalization process did not achieve its intended goal of assuring health. The reasons why patients are readmitted are many, including some that are well outside of the control of the hospital process during the original admission. In addition, readmission may be purposeful. Therefore it is necessary to consider the underlying reasons why patients may return to the hospital when setting process measure goals.
In general, when selecting a process measure it is important to determine that there is strong scientific evidence linking the process to the outcome. As an example, hand washing reduces nosocomial infections and therefore measuring hand hygiene policy adherence is an appropriate process measure. Additionally, there is no contraindication to hand hygiene and thus the process measure goal for achievement would be 100%. This is in contrast to the above example of 30-day readmission rates in which a target goal of 0% would be unrealistic and inappropriate. Another important consideration when picking a process measure is the ease of data collection. In the era of electronic health care records, queries of process are much easier to conduct. Electronic health care records allow for keyword searching, categorization of patients by diagnostic codes to determine if protocols were followed, tracking of orders, and monitoring the compliance of providers. Overall, the use of process measures helps guide quality improvement in a more straightforward manner than focusing on outcome measures.
Adjust for risk factors or stratify the population (risk stratification) when using outcome measures because it is challenging to account for all of the factors that may influence health outcomes. In a large number of patients or cases, it is often necessary to allow for the statistical adjustment of risk. Nonetheless, it is essential to measure outcomes of health care delivery to assure that the system is optimized to promote the health of the population served.
International Classification of Functioning, Disability, and Health
The PM&R community is well versed in outcome measurement, especially as it relates to functional outcomes and pain reduction. The International Classification of Functioning, Disability and Health (ICF) was released in 2001 as an update by the World Health Organization to their International Classification of Impairments, Disabilities and Handicaps. In the ICF, function is understood as occurring at three levels: (1) at the level of the body part or system, (2) at the level of the person (activities), and (3) at the level of the person in society (participation). Function is the result of an interaction between an individual’s health conditions and their environment and personal factors. Environmental and personal factors mediate function at each level in positive and negative ways. In PM&R, functional outcomes can be measured at each of the levels described in the ICF and interventions may be geared at any one or more of these levels. The goal of a particular treatment may be to alleviate pain or improve range of motion (body functions and structures), or the goal may be to improve walking (an activity), or to improve the patient’s attendance at work so that they do not lose their job (participation); or, as is often the case, the goal is to improve outcomes at all three levels. An intervention such as a joint injection to the knee may directly impact the patient’s body function, but hopefully also impacts ambulation and therefore also the ability to participate in work. Prescribing a walker may directly impact ambulation and allow for improved work attendance, but does not impact knee function. Recommending work site adaptations may improve participation but not the patient’s body function or their ability to ambulate. There is not a clear linear causal relationship between each level of functioning; therefore there is no guarantee that a prescribed treatment that was successful at impacting one level of functioning will impact another. Thus, when considering the outcome of interest to measure, it is important to attend to the influences of the patient’s environment and preferences because an optimal outcome measure will evaluate all the results of health care delivery, which are hopefully minimally influenced by other factors.
Functional Independence Measure
Translating individual patients’ outcome measurements to patient populations is relatively straightforward. The aggregation of individual patient outcome measurements allows providers and hospitals to benchmark and can help identify areas for improvement in health care delivery. The most familiar example of this in PM&R is the use of functional assessments for inpatient rehabilitation. Most inpatient rehabilitation programs submit their patients’ Functional Independence Measure (FIM) scores to the Uniform Data System (UDS). The FIM instrument ( Box 7-1 ) for adults and the WeeFIM instrument for children measure function in self-care, sphincter control, transfers, locomotion, communication, and social cognition with 18 different items. These items are scored on a 1 to 7 ordinal scale based on how much help is needed from another person to complete the tasks with a score of 1 indicating total assistance needed and a score of 7 indicating complete independence. Tracking functional progress is a key activity in inpatient rehabilitation and aggregate data can speak volumes about how the facility and its providers are doing to promote function. The UDS provides report cards that detail FIM change (discharge FIM–admission FIM) and FIM efficiency (FIM change/length of stay) for patients by impairment grouping. It also tracks the percentage of patients that are successfully discharged to home. Therefore it is possible to track functional outcomes across the population served. With the data submitted to the UDS, it is possible to stratify by impairment type, age, gender, and insurance status. It is important to note that the change in function that a population of patients experience during rehabilitation is attributable to a host of factors outside of the control of the rehabilitation physician. Therefore FIM change and FIM efficiency may better outcome measures for the institution instead of the physician. However, there are statistical methods of controlling for other factors that would allow for the patients’ FIM change and efficiency scores of one physician to be compared with other physicians’ scores. This would allow for physician level benchmarking, but it requires a large sample size.
| LEVELS | Independence | NO HELPER |
| ||
| Modified Dependence | HELPER NEEDED | |
| ||
| Complete Dependence | ||
| ||
| DOMAINS | Self-Care | |
| ||
| Sphincter Control | ||
| ||
| Transfers | ||
| ||
| Locomotion | ||
| ||
| Motor Subtotal Rating | ||
| Communication | ||
| ||
| Social Cognition | ||
| ||
| Cognitive Subtotal Rating | ||
Patient-Reported Outcomes
As described earlier, outcome measures may be classified by level of function. In the rest of medicine most of those measures focus on body function and structures, but in PM&R outcome measures frequently focus on activities with the use of assessment instruments such as the FIM. Additionally, outcome measures can be categorized by whether or not they are reported by the patient. Some outcome measures are self-reported and others require a clinician to administer. For example, health-related quality of life is patient-reported (patient-reported outcome [PRO]), whereas pulmonary function is provider-assessed (provider-based assessment). Both types of measures are important, valuable, and can be rigorously applied.
The Patient-Reported Outcomes Measurement Information System (PROMIS) is a system created by the National Institutes of Health that provides several reliable and precise PRO measures. PROMIS measures are valid and dynamic; can efficiently study the self-reported health status of patients’ physical, mental, and social well-being; and are particularly useful in chronic disease states and across a wide range of demographics. PROMIS has a core set of questions to assess common patient-relevant outcomes across a wide range of conditions in addition to specific item banks that focus on areas such as pain, fatigue, physical functioning, etc. The 10-item, 29-item, and 57-item PROMIS instruments are available to choose from depending on the time available and level of detail desired.
However, a particularly attractive feature of PROMIS is that it uses item response theory and computerized adaptive testing (CAT) that allows for statistical modeling to determine the most appropriate follow-up question to ask a patient based on their response to an initial question. Unlike a traditional questionnaire that is the same for all patients, a CAT instrument tailors questions to the specific patient. If a patient’s response to an initial question indicates that he/she is able to complete a task with little help, the CAT selects the next question with a more difficult task than the first task. If that response indicates higher difficulty in completing the more difficult task, the CAT selects the following question with an easier task than in the second question. This process continues until the computer identifies that the preprogrammed confidence interval has been met or the patient has answered the maximum number of items used to estimate the score.
When considering outcome measures for children, it is important to keep in mind that outcome measures for adults may not be appropriate for children. Pediatric outcome measures should be age appropriate and ideally acknowledge the life course health perspective. Pediatric PROs should allow for child report and parental report. Note that both the child’s and parents’ perspectives are valuable and may differ from one another. The measurement of functional outcomes should be developmentally appropriate because the self-care, cognitive, and mobility expectations change substantially over the course of childhood. Because children live within the context of their families, it is also important to consider the measurement of family function, health, and well-being.
The AHRQ has quality indicators for children receiving care in hospitals, a few of which are particularly relevant to rehabilitation care such as the number of stage III or stage IV ulcers as a secondary diagnosis per 1000 discharges or the number of central venous catheter–related bloodstream infections per 1000. Similarly, for the rehabilitation population, the outcome of ventilator-associated pneumonia is an important outcome of interest because hospital complications can lead to worsening debility and a delay in rehabilitative care.
Choosing Outcome Measures
The Rehabilitation Measures Database has numerous measures listed that can be used to track outcomes in a patient population of interest. The database intends to provide clinicians with a list of instruments that can be used to screen and monitor patient progress. Outcome measures from this database can be geared at the individual patient or at groups of patients to assure overall quality of care. Some representative examples of outcome measures in various topic areas within PM&R are shown in Table 7-1 .
| Measure and Description | ICF Domain(s) | |
|---|---|---|
| Inpatient rehabilitation | Functional Independence Measure (FIM): contains 18 items assessing self-care, mobility, bowel and bladder management, cognition, language and social interaction | Activities/participation |
| Brain injury | Community Integration Questionnaire: contains 15 items assessing home and social integration and productive activity | Activities/participation |
| Ashworth Scale/Modified Ashworth Scale: measures spasticity. Can be used to assess responsiveness to medication | Body functions/structures | |
| Spinal cord injury | Spinal Cord Independence Measure: assesses self-care, respiratory and sphincter management, and mobility | Body functions/structures Activities/participation |
| Spinal Cord Injury Functional Ambulation Inventory: observational gait assessment evaluating parameters of gait, assistive device use, and temporal distance | Activities/participation | |
| Musculoskeletal medicine/pain | Medical Outcomes Study Short Form 36 (SF-36): physical and mental composite domains | Activities/participation |
| Numeric Pain Rating Scale: 10-point scale; 0 = no pain and 10 = the most intense pain imaginable | Body functions/structures | |
| Pediatric rehabilitation | WeeFIM: companion measure to the FIM | Activities/participation |
| Glasgow Outcome Scale Extended Pediatric Version (GOS-E Peds): used in pediatric traumatic brain injury and evaluates consciousness, independence in and out of the home, school, social and leisure activities, family and friendships, and return to normal life | Body functions/structures Activities/participation | |
| Stroke | Rivermead Mobility Index: assesses mobility after stroke with 14 self-report measures and one direct observation measure | Activities/participation |
| Four Square Step Test: assesses dynamic balance | Activities/participation | |
| Amputation/limb deficiency | 10-Meter Walk Test: measures time to walk set distance | Activities/participation |
| Medicare K Levels: estimates functional ability and which prosthesis would be most appropriate | Activities/participation | |
| Pulmonary rehabilitation | Saint George’s Respiratory Questionnaire: addresses pulmonary symptoms, activities, and impact of disease on daily life | Body functions/structures Activities/participation |
| 2-Minute or 6-Minute Walk Test: measure of endurance | Activities/participation | |
| Spine | Oswestry Disability Index: addresses functional disabilities commonly seen in spine patients | Activities/participation |
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
