Assessing the health-related quality of life (Hr-QoL) of patients with atypical parkinsonism disorders is important yet no validated measures are currently available. The atypical parkinsonian disorders are chronic progressive conditions, which not only shorten life expectancy but affect many aspects of patients' and their carers' lives. No curative treatment for these disorders is available, and management of these patients largely has to concentrate on amelioration of symptoms, such as falls, immobility, autonomic features or dysphagia, activities of daily living and (in)dependence, and patients' social and emotional well-being; in short, the improvement of patients' quality of life. Assessment of patients with atypical parkinsonism has concentrated on objective measures such as mortality and clinical evaluation of impairment and physical functioning, supplemented by laboratory test results. However, a large literature shows that patients' own assessments of their health, their preferences, and their views regarding health often differ significantly from physicians' objective assessments (1). Where possible, treatment decisions should focus on health outcomes of value to the individual patient.
Scales to measure Hr-QoL, fully psychometrically tested and validated, are now used in a number of clinical and research contexts. Some types of Hr-QoL scale yield information that, combined with economic data, can be used to assess the cost benefit of health interventions and to inform decisions about the allocation of scarce health care resources. There are currently no validated measures to assess Hr-QoL in patients with atypical Parkinsonian disorders.
This chapter starts with a general overview of some of the conceptual and methodological issues relating to the measurement of subjective health assessment and Hr-QoL. Section two gives an overview of Hr-QoL instruments that have been used in Parkinson's disease (PD) and the impact of PD on Hr-QoL. The third section addresses what is known about the Hr-QoL of patients with atypical par-kinsonism, in particular multiple system atrophy (MSA) and progressive supranculear palsy (PSP). The chapter concludes with some comments about future research in this area.
HEALTH-RELATED QUALITY OF LIFE: CONCEPTUAL AND METHODOLOGICAL ISSUES
What Is Quality of Life?
Although the definition of this somewhat elusive term is still occasionally discussed in the literature, there is general consensus on some fundamental points. First, although the phrases "quality of
From: Current Clinical Neurology: Atypical Parkinsonian Disorders Edited by: I. Litvan © Humana Press Inc., Totowa, NJ
life," "health-related quality of life," and health status are used somewhat interchangeably, there is broad agreement that, in the medical context, Hr-QoL should be regarded as a multidimensional construct (2), comprising physical, psychological, and social well-being. Within these three broad dimensions, most Hr-QoL scales have items on physical health and functioning, activities of daily living, mental health (e.g., perceived stigma, anxiety, depression), social activities, family relationships, and cognitive functioning. Because of these multiple factors, it has been argued that although it may be helpful to derive a summary index of Hr-QoL, the different aspects of Hr-QoL as measured by the scale domains should also be presented separately in order to better understand the precise impact of interventions (3).
Second, since quality of life is highly subjective, any appraisal of Hr-QoL should rely, where possible, on the perception of the individual patient. Many groups of patients cannot, however, assess their own Hr-QoL, e.g., those with severe dementia, and there is a growing literature on the use of proxy ratings.
Third, no quality of life instrument can comprehensively cover all aspects of Hr-QoL. Although some scales attempt to comprehensively assess all aspects of Hr-QoL, such instruments are often lengthy and burdensome and so are not feasible in clinical practice, particularly where patients have disabling conditions. Therefore, most measures focus on a limited number of specific aspects of Hr-QoL. The choice of instrument will be determined by the precise aim of the study.
Finally, in order to provide meaningful data for research and clinical practice, Hr-QoL measures need to be carefully developed and validated and there is now a large literature on the validation and psychometric properties that need to be demonstrated before the scientific community will accept that an instrument has been shown to be appropriately validated. Before considering psychometric testing in more detail, it is useful to consider next why Hr-QoL might be measured.
Why Assess Health-Related Quality of Life?
Though Hr-QoL measures have been developed for a number of reasons, two basic aspects of health care underlie most of the questions that Hr-QoL appraisals set out to answer: outcome of treatment and cost. As discussed above, Hr-QoL has emerged as an important outcome that incorporates patients' views of their health. Also, since no country in the world can afford to do all that it is technically possible to do to improve the health of its citizens, the need has arisen for some system of setting priorities. The assessment of the Hr-QoL of patients with atypical parkinsonian disorders will become increasingly important if and when new drug treatments and other therapies for these disorders are developed. Trials will need to address the benefit of therapeutic interventions and measure change of symptoms in relation to Hr-QoL.
HEALTH-RELATED QUALITY OF LIFE MEASURES: DEVELOPMENT AND VALIDATION
All clinical assessment measures need to be shown to be valid and reliable. In addition, self-completed measures for patients with a disabling disease need to be short and feasible. Instrument developers must test the psychometric properties of a new instrument, which is a labor-intensive exercise, involving a series of studies to obtain data on the performance of the measure in different situations. For a comprehensive review of the statistical procedures, see Streiner and Norman (4). In brief, validity is how well the instrument measures what it purports to measure. There are various statistical procedures for testing different aspects of an instrument's validity. The terminology is somewhat confusing but Streiner and Norman provide a useful guide to the various types (e.g., face validity, construct validity, criterion validity, concurrent validity, and predictive validity). Reliability assesses whether the same measurement can be obtained on other occasions and concerns the amount of error inherent in any measurement. Two basic tests are the internal consistency of a test, measured by coefficient alpha, and test-retest reliability where scores taken on two occasions are compared. Sensi tivity or responsiveness to change is concerned with how sensitive the measure is to detecting clinically relevant changes in Hr-QoL. This is important for monitoring benefits of treatment. Newer methods that allow further improvement of scales include Rasch analysis and Item Response Theory, a discussion of which is beyond the scope of this chapter.
This psychometric testing has not uniformly been conducted with all instruments, particularly older instruments.
There is no "gold standard" for measuring Hr-QoL and there is a wide range of instruments available, or in development. The categories of Hr-QoL measures have been comprehensively reviewed elsewhere (5). In brief, generic instruments cover a broad range of Hr-QoL domains in a single instrument. Their chief advantage is in facilitating comparisons among different disease groups. Disease-specific instruments reduce patient burden by including only relevant items for a particular illness but their main disadvantage is the lack of comparability of results with those from other disease groups. Health profiles provide separate scores for each of the dimensions of Hr-QoL, whereas a health index, a type of generic instrument, gives a single summary score, usually from 0 (death) to 1 (perfect health). A further category, developed within the economic tradition, is that of utility measures, which are based on preferences for health states. Preference weighted measures are required when the focus is on society as a whole and the societal allocation of scarce resources. The choice of measure will depend upon the goal of the study.
Preference-based outcome measures are a particular type of measure used in economic analyses, such as cost-utility analyses. Cost-utility analysis is a technique that uses the quality adjusted life year (QALY) as an outcome measure. For its calculation, the QALY requires well-being or Hr-QoL to be expressed as a single index score. The three most commonly used preference measurement techniques are visual analog scales, time trade-off, and standard gamble. A review of the literature on the use of Hr-QoL life data in economic studies is beyond the scope of this chapter, but interested readers can consult a series of chapters on this topic in ref. 6. As treatments for atypical parkinsonisan disorders become available, they will undoubtedly be subject to economic appraisal and robust, pro-spectively collected Hr-QOL data will be important for the calculation of QALYs and for other economic analyses.
Which Outcome Measure to Use?
The choice of instrument depends on the purpose of the study. A common recommendation is to include both disease-specific and generic measures in an investigation. The generic measure facilitates comparisons of the target group with the normal population and/or other patient groups whereas the disease-specific instrument provides more sensitivity and is therefore usually more responsive to change in health status. If pharmaco-economic evaluation or a comparison of two or more treatment options is the aim of the study, incorporation of an additional utility measure is recommended. In atypical parkinsonism no disease-specific instruments are available to date but disease-specific instruments for MSA and PSP are currently being developed.
HEALTH-RELATED QUALITY OF LIFE IN PARKINSON'S DISEASE
Health-Related Quality of Life Instruments Used in Parkinson's Disease
A number of studies have assessed Hr-QoL in idiopathic PD. The authors of the first of these studies used generic instruments including the Sickness Impact Profile (SIP) (7), the Nottingham Health Profile (NHP) (8), the Medical Outcomes Short Form (SF 36) (9), and EQ-5D (10). These measures were shown to be valid to varying degrees. However, the older instruments, such as the SIP and the NHP have been criticized for their content and their psychometric properties. For example, the NHP is skewed toward the severe end of disability and worse functional status and is therefore less likely to capture subtle changes in early stages of disease. The questions in the SIP have been felt to be offensive to patients by some (11), and the SF 36 may have limited feasibility and validity in patients with parkinsonism (12). However, the SF 36, a widely used Hr-QoL instrument, which has been translated in several languages and been validated for use in many cultures, enables comparisons across cultures and disease groups. The EQ-5D has also been shown to be valid in patients with PD (12), and its brevity of five questions and a visual analog scale is an advantage in disabled patients. In addition, its summary index yields a utility score that can be used in pharmaco-economic analyses. The EQ-5D has also been translation into many languages.
More recently, PD-specific Hr-QoL measures have been developed. Table 1 briefly describes the PD-specific Hr-QoL measures, showing the scale domains. All of the PD-specific instruments have been shown to have good psychometric properties, but only the Parkinson's Disease Questionnaire (PDQ 39; ref. 13) and the Parkinson's Disease Quality of Life Questionnaire (PDQL; ref. 14) have been validated by researchers independent of the developers (15). The PDQ 39 is the most widely used Hr-QoL instrument in Parkinson's disease. It has been translated in several languages, and has been shown to be valid, reliable and sensitive to change. An abbreviated format, the PDQ 8, which has been shown to have comparable validity (16), is also available. Although PDQ 39 has been validated in some cultures, including Britain, the United States, Spain, France, China, and Japan, its validity in other cultures needs to be established. The PDQL is similar to the PDQ 39 in content and format, but includes some questions that are missing in the PDQ 39, e.g., on sexuality. It has been validated in The Netherlands and Britain, but it has not been translated into other languages and no validation studies in other cultures are currently available. Its psychometric properties are less well tested than those of the PDQ 39 and some issues such as self-care, role functions, and close relationships are not addressed (15).
The Parkinson's Impact Scale (PIMS) was developed to identify the major problems in patients' lives in a clinical setting. It is based on consensus rather than testing in a patient sample and its content validity has been criticized (15). However, it is the only instrument that distinguishes between on and off periods and has been reported to be valid, reliable, and sensitive to change (17). The Parkinson's Disease Quality of Life Scale (PDQUALIF) includes questions on fatigue and driving ability, concentrates on the nonmotor symptoms of Parkinson's disease, and has more emphasis on social functioning than other scales (18). The Parkinson's Disease Symptom Inventory (PDSI; ref. 19) has a larger number of questions (51 items) and asks patients to indicate the frequency as well as the distress caused by each item. There is also a German questionnaire, the ParkinsonLebensqualitat (PQL), which has been psychometrically tested in a German population (20). The differences between some of these scales are discussed in an excellent review by Marinus et al. (15).
Finally, a number of measures that assess only the psychosocial aspects of Hr-QoL, excluding items relating to physical impairment, have recently been developed and validated (21,22). The choice of instrument in each setting will be guided by the differences between the content of the questionnaires, published data on the psychometric testing, and, if relevant to the study, the availability of translations and cultural adaptation. For specific interventions different aspects of Hr-QoL will be important and as no instrument can be both completely comprehensive and feasible, the selection of the instrument will be based on the particular aim of the study. As discussed above, whereas generic instruments can be used in patients with Parkinson's disease, PD-specific instruments are likely to be more valid, sensitive, and responsive to change.
Disease-Specific Hr-QoL Measures Developed for Parkinson's Disease
Parkinson' Disease Questionnaire 39-item version (PDQ-39)
Was this article helpful?