Publication
Article
The American Journal of Managed Care
Assessing Symptoms, Disease Severity, and Quality of Life in the Clinical Context: A Theoretical Framework
Author(s):
Health-related quality-of-life instruments can yield important health information that is often distinct from objective measures of symptoms and disease severity that clinicians are most attuned to. Comprehensive health assessment can be difficult because there are many available measurement instruments that vary in their scope and content. The conceptual boundaries between symptom, disease severity, and health-related quality-of-life instruments are frequently blurred in practice, and what is measured may not coincide with clinical and research goals. The Assessing the Impact of Disease framework aims to clarify the process of selecting appropriate assessment instruments. Three common diseases are discussed in depth to illustrate the applicability of Assessing the Impact of Disease in distinguishing between symptom, severity, and health-related quality-of-life measurements.
(Am J Manag Care. 2004;10:336-344)
Illness affects more than patients' physical functioning: it may also affect their emotional, social, and occupational functioning. Recognizing that objective clinical indicators are not sufficient to assess the overall effect of disease, interest has focused on including patient-reported measures in disease assessment. Instruments that measure health-related quality of life (HRQOL) are increasingly important to clinicians and researchers conducting disease assessment studies.
Nonetheless, disease severity and symptom assessment are important components of assessing the burden of disease, and such assessments often provide information distinct from HRQOL assessment. Yet, HRQOL instruments vary in scope and content and may incorporate symptom and severity assessment to varying degrees, making it challenging to discern the realms each instrument addresses.
Given these issues and the numerous tools to choose from, instrument selection becomes a complex task. Our framework, Assessing the Impact of Disease (AID) assists clinicians in identifying which realm of assessment– symptoms, disease severity, or HRQOL–is of greatest interest for their clinical or research purposes. It also aims to assist clinicians in determining which corresponding instruments might be most appropriate for their clinical and research goals. Without an understanding of the distinctions between these measurement concepts, clinicians may overlook dimensions of assessment that may prove critical in improving patient care. Therefore, the AID framework was developed to help clinicians clarify the dimensions of greatest interest to their research and improve study design by facilitating decisions regarding the most appropriate measurement tools.
DEFINING SYMPTOMS, DISEASE SEVERITY, AND HRQOL
Symptoms
Symptoms are experienced deviations from an individual's perception of his or her normal, healthy state of being, yet not necessarily an indicator of illness. A symptom can emerge from sensitivity to certain combinations of biological, social, and environmental processes and vary in magnitude, severity, persistence, and character. Symptoms can be subjectively reported or objectively observed. If patients are subjectively surveyed about their symptoms, the findings can be problematic. For example, at different time points, an individual may not perceive the same aberration in health as a symptom or cause for alarm.1 In addition, symptoms can become integrated into the scope of normal experiences or be ignored for various reasons such as fear of stigmatization, embarrassment, or the expectation of other adverse reactions or consequences from society.2
Objective symptom assessment also has drawbacks. Depending on the disease, the scope, intensity, and duration of symptoms can vary over time. In addition, patients may not report all symptoms to their clinicians or realize that physiological changes may be related to illness. Therefore, the presence, absence, or severity of a symptom may be overlooked or attributed to noncausal factors.
Disease Severity
Disease severity refers to the presence and extensiveness of a disease in the body. It is objectively evaluated through diagnostic testing and physiological examination of the impaired biological organs or tissues, in cases in which disease severity can be distinguished from other realms of health, as in heart disease. The amount of plaque buildup in the heart can be measured, and the corresponding severity level of the disease can be determined. In this example, disease severity information can be contrasted with symptom assessment by considering 2 matched patients with moderate plaque buildup. One patient could complain of chest pains and express symptoms indicative of heart disease, but it would not be unusual if someone else with laboratory results indicating an equally severe disease state to experience a different set of functional limitations and symptoms.
Health-Related Quality of Life
quality of life
quality of life
Despite the growing interest in quality of life, debate continues over the appropriate terminology and classification of many quality-of-life concepts and whether they belong under the rubric of HRQOL.3 From the late 1970s to the 1990s, articles that included the concept of appeared in increasing numbers on MEDLINE from health services researchers using instruments as an outcome measure of medical care and in clinical and population studies. In 1974, there were 40 references, but by 1999, more than 12 000 references were cited and today the number is approaching 40 000.4 In 1987, Ware noted that it had become "fashionable to equate health, defined comprehensively, with " (emphasis in original).5 Health-related quality of life is elusive, because it can be defined in a way that includes other abstract and concrete aspects of life or quality of life, such as well-being, pain, and happiness.6,7 However, subjective reports of these concepts are not sufficient for generating HRQOL scores or substituting for HRQOL information.
Although there is no gold standard or agreed-on definition for the scope of HRQOL, several prominent HRQOL conceptualizations include the physical, psychological, social, spiritual, role functioning, and general well-being dimensions of health. These conceptualizations are based on the World Health Organization (WHO) definition that "health is complete physical, mental, and social well-being, and not merely the absence of disease and infirmary."8 The WHO definition has become a common starting point for research and discussion of overall health status and HRQOL. Moyer and colleagues9 provide a review of commonly used generic HRQOL instruments, and additional studies10-14 describe some important aspects to consider when choosing HRQOL instruments.
Different conceptualizations of HRQOL may not assess a patient's overall quality of life or yield information about all the dimensions and domains contributing to HRQOL. Schipper et al offer a clinical definition of HRQOL based on physical and occupational function, psychological function, social interaction, and somatic sensation domains: "Quality of life in clinical medicine represents the functional effect of an illness and its consequent therapy upon a patient, as perceived by the patient."15 (p16) In this definition, the functional effect of an illness is important yet difficult to define because it is closely related to the concepts of symptom and disease severity.
HRQOL MODELS
Two frequently consulted conceptualizations of HRQOL assessment are the pyramid model by Spilker3 and the model of patient outcomes by Wilson and Cleary.16 However, these models do not address the differences between assessing symptoms, disease severity, and HRQOL, or how to select appropriate instruments once specific measurement goals are identified.
Spilker3 presents the definition and scope of HRQOL in 3 levels of a pyramid. A patient's overall assessment of well-being is at the top of the pyramid (level 1). The middle section (level 2) contains the broad domains that contribute to HRQOL, and the many possible components of each domain comprise the base of the pyramid (level 3). This model was intended to enable researchers or clinicians to include different domains or components of interest in HRQOL assessment, providing the ability to approach HRQOL assessment from a bottom-up or top-down perspective.
However, the pyramid model does not adequately represent the relationships among HRQOL and other related health concepts. According to the model, symptoms and physical functioning measures exist as subsets that contribute to an overall HRQOL score. Spilker notes that, although the number and identity of domains vary, "each cuts the overall pie (level 1) into different pieces of domains (level 2)."3 (p3) The levels are related in a hierarchical fashion, with level 1 including all of the contents of levels 2 and 3. This characterization does not allow for components of different domains to overlap. The model also fails to consider the effects of other health measurements on HRQOL, such as severity of disease. Disease severity can affect health and overall well-being, yet is not included in the pyramid model. Often, disease severity information is useful for contextualizing HRQOL scores and explaining changes over time, and should be part of a comprehensive assessment of the effect of disease. Spilker's pyramid is useful as an educational tool for understanding HRQOL and its component realms, but may not be ideal when applied to all clinical situations.
In the causal pathway of patient outcomes presented by Wilson and Cleary,16 disease severity is accounted for at the first level of the model by biological and physiological variables reflecting health and functioning at the genetic level. The next 4 levels include symptom status, functional status, general health perceptions, and overall quality of life. Each of the 5 outcome levels accounts for various indirect and direct inputs and factors out of the control of clinicians and the health system that affect health outcomes. This model maps causal relationships between patient outcome measures to increase understanding about the pathogenesis of impairment and enable the development of more effective interventions. This model is most useful for examining many health factors and HRQOL, yet it does not offer clinicians guidance in choosing the specific tools to measure each factor.
The AID framework that we propose incorporates the effect of disease severity on HRQOL and offers relevant guidance in disease assessment. Assessing the Impact of Disease presents the relationships among measures of symptoms, disease severity, and HRQOL in a model that allows the measurement of health concepts to overlap (Figure 1). Our figure is symmetrical for simplicity in illustration; however, it does not reflect the relationships among these concepts for every disease. The AID framework recognizes that, in some disease states, symptom measurement cannot be separated in practice from disease severity measurement. We draw attention to the various instruments and underlying health concepts being assessed for different classes of diseases. This information can increase a clinician's understanding of which health domain he or she is measuring and its effect on the individual's overall well-being.
The next section presents case studies of several diseases. Each case study illustrates how the focus of measurement instruments draws on different realms of health, leading to the need for careful instrument selection in clinical practice.
DISEASE CASE STUDIES
We identify 3 classes of diseases with contrasting degrees of conceptual overlap. We chose urinary incontinence (UI), obstructive sleep apnea (OSA), and gastroesophageal reflux disease (GERD) as examples of the differences between symptom assessment, disease severity, and HRQOL using the AID framework. These case studies were selected to illustrate varying degrees of conceptual overlap, not to describe the clinical scenarios most commonly associated with HRQOL assessment.
Case Study 1: Urinary Incontinence
The International Continence Society defines UI as a "condition where involuntary loss of urine becomes a social or hygienic problem and is objectively demonstrable." 17 (p17) Urinary incontinence is characterized by a lack of bladder control and can be further dichotomized as stress UI if control problems are related to storing and emptying or as urge UI if problems are related to outlet functioning. Social interactions and psychological well-being may be adversely affected by persons with UI, typically older persons and women. Individuals with UI often alter their activities and lifestyles to minimize the occurrence of episodes of uncontrollable urinary discharge.18
Urinary incontinence represents a class of diseases with tremendous overlap across the instruments measuring the realms of symptoms, disease severity, and HRQOL (Figure 2). Urinary incontinence exists as a disease state because of the presence and acuteness of experienced symptoms. Although symptom and disease severity measures are independent concepts analytically, they are virtually equivalent in practice.
Two commonly used objective and reliable diagnostic tests measure the volume of leakage. A pad test tabulates the number of sanitary pads used and the weight of each before and after use during a specified period. This is often a 24-hour test, although 1- or 2-hour or 48-hour spans have also been used in studies.19-21 Cystometry is one component of the urodynamic diagnostic evaluation, which includes measuring the volume at which the patient initially feels the desire to micturate; flow rate; pressure; and volume of liquid voided during "stress" tests of forceful coughing and straining, while the external urethral meatus is observed for gross urine loss.22 This test can discriminate between stress UI, which is the "result of an incompetent urethral sphincter mechanism"23 (p375) and urge UI, a form of urinary leakage that is caused by detrusor contraction.
Several subjective indexes have been developed for patients to self-monitor and record the occurrence of symptoms and experiences associated with episodes of urinary leakage. Tools such as the 24-hour micturition or frequency-volume diary,19,24,25 UI-specific case history,26 UI score,27 severity index for epidemiological surveys of female urinary incontinence,28 and leakage index29 document the number of UI episodes by type, diurnal and nocturnal voiding episodes, overall number of episodes, and mean daily voided volume. The diary is sometimes used as the standard for psychometric comparison for these indexes. Generally, these tools are more popular, less invasive, and have a lower patient burden than the pad test or cystometry, because patients remain in their natural environment and normal living conditions, instead of undergoing catheterization or instigating leakage in front of an observer.
Most UI-specific HRQOL instruments measure the psychosocial effects of UI in women. Several of these instruments are reviewed by Kelleher et al.18 Multiple versions of the 30-item Incontinence Impact Questionnaire have been developed to assess the effect of UI on various activities, roles, and emotional states in women.30-32 A 26-item Incontinence Impact Questionnaire investigates the effect of UI on mental and physical health; performing chores or job and leisure activities; transportation; social relationships with family, spouse, and friends; and self-perception.25 The Psychosocial Consequences of Urinary Incontinence Questionnaire contains 25-items focusing on the 3 domains of mental distress, practical inconvenience, and social restrictions of UI in women.20 The 33-item Bristol Female Lower Urinary Tract Symptoms questionnaire33 quantifies the effects of UI symptoms on HRQOL in terms of sexual functioning and inquires about the bothersomeness of symptoms. One scale for middle-aged and older men, the International Continence Society Quality of Life instrument, contains questions about daily leakage; changing volume of liquid intake; level of symptom interference with daily activities; and persistence, worries, and opinions regarding symptoms.34
In cases like UI, instruments that measure disease and symptom severity overlap considerably. Diseasespecific HRQOL instruments for this class of diseases examine the symptoms and the perceived effects of those symptoms on multiple realms of life. A study designed to assess quality of life in women with UI could focus on evaluating the presence and severity of symptoms objectively and subjectively, using an HRQOL survey with specific items from the psychosocial domain of health. Specifically, a 24-hour pad test19-21 combined with responses from the leakage index29 and Psychosocial Consequences of Urinary Incontinence Questionnaire20 could be one set of instruments selected to provide the desired data for a comprehensive study.
Case Study 2: Obstructive Sleep Apnea
Obstructive sleep apnea is characterized by difficulties in breathing during sleep, which can have detrimental effects on physical and mental functioning. This condition afflicts 4% of men and 2% of women, who typically experience loud, heavy, and repetitive snoring; chronic fatigue; disrupted sleep; nocturnal arousals; and excessive daytime sleepiness.35,36 Many may be unaware that they are experiencing these symptoms during their sleep and are not diagnosed with OSA. This condition is most common among overweight individuals, who often have a short, thick neck, or individuals with craniofacial abnormalities of the bony and soft tissue structure in the head and neck that complicate breathing ability.36,37
Obstructive sleep apnea represents a class of diseases with moderate overlap between the instruments measuring the realms of symptoms, disease severity, and HRQOL (Figure 3 ). Most OSA disease severity measures rely on objective, physiological data from a nocturnal polysomnography test, although no gold standard exists. The polysomnography test measures nose and mouth airflow levels (minimum oxygen saturation in apnea measure), respiratory flow and effort, body position, heart rate, and pulse oximetry. This test can include other monitors as well (ie, electrocardiogram and electro-oculogram), as in the setup by Chervin and Aldrich,38 and produces summary indexes of apneic episodes per hour of sleep, such as the apnea or hypopnea index and respiratory disease index.
The Epworth Sleepiness Scale is the most commonly used39-41 short, standardized questionnaire subjectively asking patients to report on the prominent symptoms of OSA, including loud snoring and excessive daytime sleepiness.35 Subjects are asked about the probability of falling asleep while sitting and reading, watching television, sitting inactively, sitting in a car on a long drive, sitting in a stopped car, lying down, talking with someone, and sitting quietly after lunch.42 Other evaluations objectively assess these symptoms through the Maintenance of Wakefulness Test43,44 and the Multiple Sleep Latency Test.44,45 For these tests, disease severity is essentially synonymous with symptom severity. Subjects wearing street clothing are placed in a dim, sleep-inducing, controlled- environment room on a semireclined bed 2 hours after waking up, and their propensity to fall asleep in a 30-minute interval is monitored 4 times.43 The Multiple Sleep Latency Test uses different recording devices than the Maintenance of Wakefulness Test, but both are typically performed after a polysomnography test and after 1 or 2 weeks of keeping a sleep diary.44
Obstructive sleep apnea—specific HRQOL instruments are beginning to be developed and validated and have been reviewed and discussed elsewhere.9 For diseases like OSA with moderately overlapping health measures, recognizing when symptoms and disease severity are treated as identical may be more difficult. A comprehensive study for monitoring OSA in adults likely would want to rely on objective polysomnography test data to determine the presence and severity of the disease, plus self-reported information from the Epworth Sleepiness Scale,39-41 with verification from the Maintenance of Wakefulness Test.43-44 An HRQOL questionnaire such as the Functional Outcomes of Sleep Questionnaire that addresses the domains of daily functioning, social interaction, emotional functioning, and symptoms could also provide useful information.45
Case Study 3: Gastroesophageal Reflux Disease
Gastroesophageal reflux disease includes dyspepsia and other related reflux diseases that are characterized by recurrent pain or discomfort in the upper abdominal region. Reflux, or several manifestations of the movement of gastric or duodenal contents consisting of acid, pepsin, bile salts, and trypsin into the esophagus, is typical of GERD, although it may result from physiological causes and not necessarily reflect the presence of this disease.46 The main complaints of persons with GERD are a bitter taste concomitant with heartburn and acid regurgitation, and less commonly include recurrent pneumonia, chronic cough or hiccup, hoarseness, asthma, dental erosion, and chest pain.47 This condition is common among middle-aged and older adults, but anyone can have GERD.
Gastroesophageal reflux disease represents the clearest distinctions between the measurement instruments for realms of disease severity, symptom assessment, and HRQOL (Figure 4 ). In contrast to UI and OSA, GERD diagnostic tools assess only the presence and severity of disease and not the associated symptoms. Most of these instruments are quantitative, physiologically based, and administered before and after laparoscopic fundoplication or related procedures. One GERD indicator is the 24-hour pH test, which monitors acid levels in the esophagus. During an upper gastrointestinal series, barium is swallowed to allow x-ray examination of the esophagus, stomach, and small bowel. An esophageal motility study examines changes in the stomach and upper small intestinal muscle contractions via a small swallowed tube after a test meal or medication. A similar test, esophageal manometry, requires a tube through the patient's nostril and esophagus that enables the monitoring of movement and pressure in the esophagus while the patient drinks water, and tests for acid from the stomach. The patient is then injected with a drug, and any movements, pressure, or acid changes are documented.
Subjective questionnaires inquire about the presence, frequency, and severity of a wide range of GERD symptoms. The Gastrointestinal Symptom Rating Scale measures patients' self-reported discomfort due to abdominal pain, reflux, diarrhea, indigestion, and constipation syndrome and offers a comprehensive evaluation of patients with irritable bowel syndrome (IBS) and peptic ulcer disease.48 The Leeds Dyspepsia Questionnaire49 has 8 items asking patients to report on epigastric and retrosternal pain, regurgitation, nausea, vomiting, belching, early satiety, and dysphagia and to identify the most troublesome symptom during the previous 6 months. Other questionnaires focus on symptom changes and interference in normal daily activity, such as the 15-item self-administered Dyspepsia Questionnaire.50 More detailed instruments add items querying about the use of medication and medical visits, such as the Glasgow Dyspepsia Severity Score.51
Selected HRQOL instruments for upper gastrointestinal disease are reviewed elsewhere by Moyer and Fendrick52 and include the Gastrointestinal Quality of Life Index,53 which is a hybrid instrument of general and disease-specific items, and the Gastroesophageal Reflux Data Sheet addressing heartburn.54 Of GERDspecific HRQOL instruments, the sensitive inflammatory bowel disease questionnaire is commonly used in outcome assessments to survey bowel disturbance, systemic symptoms, and social and emotional functioning.55-57 Most other GERD HRQOL instruments inquire about the same topics. The IBS—specific quality-of-life instrument58 and the quality-of-life instrument for GERD and dyspepsia59 measure the emotional, mental health, sleep, energy, physical functioning, diet, social role, physical role, and sexual relations domains of GERD. The Nepean Dyspepsia Index60 adds measures that address medications, physician visits, control over and attitude of the disease, finance, recreation, religious or spiritual issues, emotions, self-confidence, thinking or concentrating, and cognition.
The instruments for GERD assessment can be classified within well-defined underlying realms of health. Diagnostic and disease severity tests include physiological monitoring of internal digestive structures and functioning. One set of GERD questionnaires evaluates the frequency and effects of various symptoms, and another set of GERD-specific HRQOL instruments addresses symptoms in conjunction with multidimensional quality-of-life items. Selecting assessment instruments for diseases like GERD is a straightforward task, and there is little tendency to confuse the health concepts addressed by a given instrument. A comprehensive GERD study focusing on the effects of medical therapies and medication patterns could use the objective esophageal motility study, then administer the Glasgow Dyspepsia Severity Score51 test and Nepean Dyspepsia Index60 to gauge the range of a patient's experienced symptoms and the effects across multiple domains of life.
INSTRUMENT SELECTION
When planning a research study or implementing clinic-based assessments, it is important for researchers and clinicians to be cognizant of the different types of assessment tools and the primary purpose of those tools. Table 1 illustrates some examples of the types of research and clinical situations that might indicate a need for symptom, disease severity, or quality-of-life assessment.
Health-related quality-of-life instruments vary in content. It is necessary to understand what aspects of HRQOL are being evaluated and whether survey items overlap with severity or symptom measures also being obtained. A global question or series of questions related to overall HRQOL will yield different information than questions focused on a single dimension, multiple dimensions, or aspects of 1 dimension.
Table 2 illustrates some examples of symptom, disease severity, and HRQOL instruments for each of the case studies presented, UI, OSA, and GERD. Depending on the level of overlap between symptoms, disease severity, and quality of life, as illustrated by the AID framework in Figures 1 through 4, the same instruments can often be used to assess 1 or more of these dimensions. The HRQOL instruments presented herein are generally disease-specific instruments, yet deciding between using a disease-specific and a generic instrument is another important issue. Disease-specific instruments are sensitive to particular aspects of health status and change over time among narrowly defined populations with specific disorders. One potential advantage is that specific instruments can provide richer, more detailed information compared with generic ones. If the measurement of specific HRQOL domains is essential, a disease-specific instrument addressing the domains of interest should be used. Although validated HRQOL instruments are not available for every disease, generic HRQOL indexes also have merit and may be more useful depending on the rationale underlying their application.64 Generic HRQOL instruments that are a single global indicator, health profile, or use measure, such as the Medical Outcomes Study 36-Item Short-Form Health Survey, are ideal for comparisons across studies or populations or to provide the basic core of HRQOL information.
The point of interest underlying the research question should also affect the choice of instruments used. Instruments are often designed to discriminate between patients, predict future outcomes, or evaluate changes over time. Some discriminative indexes provide information that distinguishes among individuals or groups with respect to a domain without a clear gold standard.7 Others classify individuals based on an available gold standard and estimate their likelihood of developing or contracting a specific disease, or they predict particular outcomes. Evaluative measures focus on assessing the magnitude of longitudinal change in an individual or group. The original intent of an instrument's developers will affect the applicability of that instrument to various clinical and research situations; therefore, it is important to be aware of how the instrument was originally intended to be used.
Assessing the chosen instrument's psychometric properties and use is another important component of selecting an appropriate instrument. Maunder et al62 published a list of 7 criteria used in a literature review of studies that measured HRQOL for inflammatory bowel disease, including reproducibility, reliability, validity, ease of use, responsiveness to change, meaningfulness of results, and sampling of the patient's perspective. These criteria offer a starting point for judging whether it is worthwhile to use a particular instrument in practice.
There is rarely one instrument that is suited to all situations. The choice of appropriate instruments is an individual one, based on such factors as the goals of the study, characteristics of the population, time available for patients to complete instruments, and type of data analysis required to answer the research questions of interest. Therefore, there is no prescription for choosing the perfect set of instruments. The point of the AID framework is to assist clinicians in understanding some of the many factors that may affect the manner in which they choose the instruments that will best suit their needs.
CONCLUSIONS
The conceptual boundaries between symptom assessment, disease severity, and HRQOL evaluation are not always clear, especially for clinicians and researchers unaccustomed to measuring aspects of health that go beyond traditional clinical end points. By definition, many clinicians are most attuned to the biological and physiological aspects of disease. These data often are obtained from clinical and laboratory information, yet may not convey the full effect of the disease from a patient's viewpoint. A patient's experience of disease can range from asymptomatic to debilitating, and the presence of any form of disease can diminish quality of life in multiple realms.
Yet, experiencing more symptoms, or having greater disease severity, does not necessarily translate to lower quality of life. Therefore, these concepts are not always interchangeable. This is critical when planning research studies or clinical assessments in which quality-of-life measures may show different outcomes than symptom assessment or disease severity assessment. Understanding how assessment tools may differ and how concepts may overlap for one type of illness, yet be discrete for another, is important to planning and implementing research studies that reflect patients' realities for various disease states.
Urinary incontinence is one example of a condition in which symptoms and disease severity are one and the same, given that disease severity is defined by the presence and frequency of observed or patient-reported symptoms. As a result, if symptoms or disease severity were related to quality of life, it would be difficult to determine any independent relationship. Obstructive sleep apnea represents modest overlap between symptoms, disease severity, and HRQOL. Several disease severity measures include assessments of the frequency of symptoms, but most assessments rely on objective, physiological data, such as oxygen saturation. In this case, it would be easier to determine the independent effects of disease severity or symptoms on quality of life. Finally, GERD represents the clearest distinction between symptoms, disease severity, and HRQOL. Disease severity is assessed through biological markers, symptoms are assessed through patient recall, and HRQOL assesses the effect of GERD on multiple dimensions of a patient's life.
The AID framework presented herein aims to help clarify the differences among measures of symptom assessment, disease severity, and HRQOL. By highlighting differences in assessment tools, the AID framework assists clinicians and researchers in identifying the outcomes of interest and choosing appropriate instruments for measuring those outcomes. This article does not include a list of all possible disease severity measures, symptom assessments, or HRQOL instruments, nor does it offer a "one size fits all" prescription for which instruments to use in all situations. Instead, it provides clinicians with a method for conceptualizing health measurement that can be applied across disease states. Ensuring the inclusion of appropriate measures will not only provide valuable information for clinicians but also allow clinicians to communicate health information to their patients that resonates with their daily lives and priorities. Clinicians will also gain contextual knowledge about their patients' objective test results and healthrelated behaviors and beliefs. Although designing comprehensive health assessment batteries that reflect the experiences of one's patients is never easy, the AID framework provides a model for clinicians and researchers to consider when designing these studies.
From the Department of Health Management and Policy, School of Public Health (TLF), and Global REACH, School of Medicine (CAM), University of Michigan, Ann Arbor; and the Department of Surgery, University of Pennsylvania, Philadelphia (SSS).
Address correspondence to: Seema S. Sonnad, PhD, Department of Surgery, University of Pennsylvania, 3400 Spruce Street, 4 Silverstein, Philadelphia, PA 19104. Email: seema.sonnad@uphs.upenn.edu.
Psychol Med.
1. Mechanic D. The concept of illness behavior: culture, situation and personal predisposition. 1986;16:1-7.
N Engl J Med.
2. Mechanic D. Social psychologic factors affecting the presentation of bodily complaints. 1972;286:1132-1139.
Quality of Life and Pharmacoeconomics in Clinical Trials.
3. Spilker B, ed. Introduction. In: 2nd ed. Philadelphia, Pa: Lippincott-Raven; 1996:1-10.
J Clin Epidemiol.
4. Wood-Dauphinee S. Assessing quality of life in clinical research: from where we come and where we are going? 1999;52:355-363.
J Chronic Dis.
5. Ware JE. Standards for validating health measures: definition and content. 1987;40:473-480.
Research on the Quality of Life.
6. Andrews FM, ed. Ann Arbor: Institute of Social Research, University of Michigan; 1986.
Quality of Life and Pharmacoeconomics in Clinical Trials.
7. Guyatt GH, Jaeschke R, Feeny DH, Patrick DL. Measurements in clinical trials: choosing the right approach. In: Spilker B, ed. 2nd ed. Philadelphia, Pa: Lippincott-Raven; 1996:41-48.
Basic Documents.
8. World Health Organization. Constitution of the World Health Organization. In: Geneva, Switzerland: World Health Organization; 1948.
Sleep Med.
9. Moyer CA, Sonnad SS, Garetz SL, Helman JI, Chervin RD. Quality of life in obstructive sleep apnea: a systematic review of the literature. 2001;2:477-491.
Quality of Life Res.
10. Scientific Advisory Committee of the Medical Outcomes Trust. Assessing health status and quality of life instruments: attributes and review criteria. 2002;11:193-205.
Qual Life Res
11. Terwee CB, Dekker FW, Wiersinga WM, Prummel MF, Bossuyt PMM. On assessing responsiveness of health-related quality of life instruments: guidelines for instrument evaluation. . 2003;12:349-362.
Med Care.
12. Osoba D. A taxonomy of the uses of health-related quality-of-life instruments in cancer care and the clinical meaningfulness of results. 2002;40(suppl):III31-III38.
BMJ.
13. Sanders C, Egger M, Donovan J, Tallon D, Frankel S. Reporting on quality of life in randomised controlled trials: bibliographic study. 1998;17:1191-1194.
Eur J Cancer.
14. Fayers PM. Interpreting quality of life data: population-based reference data for the EORTC QLQ-C30. 2001;37:1331-1334.
Quality of Life and Pharmacoeconomics in Clinical Trials
15. Schipper HL, Clinch JJ, Olweny CLM. Quality of life studies: definitions and conceptual issues. In: Spilker B, ed. . 2nd ed. Philadelphia, Pa: Lippincott-Raven; 1996:11-23.
JAMA.
16. Wilson IB, Cleary PD. Linking clinical variables with health-related quality of life: a conceptual model of patient outcomes. 1995;273:59-65.
Scand J Urol Nephrol Suppl.
17. Abrams P, Blaivas JG, Stanton SL, Andersen JT, International Continence Society Committee on Standardisation of Terminology. The standardisation of terminology of lower urinary tract function. 1988;114:5-19.
Curr Opin Obstet Gynecol.
18. Kelleher CJ, Cardozo LD, Toozs-Hobson PM. Quality of life and urinary incontinence. 1995;7:404-408.
Neurourol Urodyn.
19. Groutz A, Blaivas JG, Rosenthal JE. A simplified urinary incontinence score for the evaluation of treatment outcomes. 2000;19:127-135.
Qual Life Res.
20. Seim A, Hermstad R, Hunskaar S. Management in general practice significantly reduced psychosocial consequences of female urinary incontinence. 1997;6:257-264.
Acta Obstet Gynecol Scand.
21. Molander U, Milsom I, Ekelund P, Arvidsson L, Eriksson O. A health care program for the investigation and treatment of elderly women with urinary incontinence and related urogenital symptoms. 1991;60:137-142.
Urol Clin N Am.
22. Dupont MC, Albo ME, Raz S. Diagnosis of stress urinary incontinence: an overview. 1996;23:407-415.
Br J Nurs.
23. Harris A. Impact of urinary incontinence on the quality of life of women. 1999;8:375-380.
Br J Urol.
24. Donovan JL, Abrams P, Peters TJ, et al. The ICS-"BPH" Study: the psychometric validity and reliability of the ICSmale questionnaire. 1996;77:554-562.
Obstet Gynecol.
25. Wyman JF, Harkins SW, Choi SC, Taylor JR, Fantl JA. Psychosocial impact of urinary incontinence in women. 1987;70(pt 1):378-381.
Br J Urol.
26. Langro-Janssen A, Debruyne F, Van Weel C. Value of the patient's case history in diagnosing urinary incontinence in general practice. 1991;67:569- 572.
Int J Gynaecol Obstet.
27. Ishiko O, Hirai K, Sumi T, Nishimura S, Ogita S. The urinary incontinence score in diagnosis of female urinary incontinence. 2000;68: 131-137.
Neurourol Urodyn.
28. Sandvik H, Seim A, Vanvik A, Hunskaar S. A severity index for epidemiological surveys of female urinary incontinence: comparison with 48-hour pad-weighing tests. 2000;19:137-145.
Scand J Urol Nephrol.
29. Bo K. Reproducibility of instruments designed to measure subjective evaluation of female stress urinary incontinence. 1994;28:97-100.
Int Rehab Med.
30. Norton C. The effects of urinary incontinence in women. 1982;4:9-14.
Qual Life Res.
31. Shumaker SA, Wyman JF, Uebersax JS, McClish D, Fantl JA, Continence Program in Women (CPW) Research Group. Health-related quality of life measures for women with urinary incontinence: the Incontinence Impact Questionnaire and the Urogenital Distress Inventory. 1994;3:291-306.
Neurourol Urodyn.
32. Uebersax JS, Wyman JF, Shumaker SA, McClish DK, Fantl JA, Continence Program for Women Research Group. Short forms to assess life quality and symptom distress for urinary incontinence in women: the Incontinence Impact Questionnaire and the Urogenital Distress Inventory. 1995;14:131-139.
Br J Urol.
33. Jackson S, Donovan J, Brookes S, Eckford S, Swithinbank L, Abrams P. The Bristol Female Lower Urinary Tract Symptoms questionnaire: development and psychometric testing. 1996;77:805-812.
Br J Urol.
34. Donovan JL, Kay HE, Peters TJ, et al, International Continence Society–Benign Prostatic Hyperplasia. Using the ICSQOL to measure the impact of lower urinary tract symptoms on quality of life: evidence from the ICS-"BPH" Study. 1997;80:712-721.
N Engl J Med.
35. Young T, Palta M, Dempsey J, Skatrud J, Weber S, Badr S. The occurrence of sleep-disordered breathing among middle-aged adults. 1993;328: 1230-1235.
Am Fam Physician.
36. Victor LD. Obstructive sleep apnea. 1999;60:2279-2286.
Ear Nose Throat J.
37. Redline S, Young T. Epidemiology and natural history of obstructive sleep apnea. 1993;72:20-26.
Neurology.
38. Chervin RD, Aldrich MS. The Epworth Sleepiness Scale may not reflect objective measures of sleepiness or sleep apnea. 1999;52:125-131.
Am J Respir Crit Care Med.
39. Engleman HM, Kingshott RN, Wraith PK, Mackay TW, Deary IJ, Douglas NJ. Randomized placebo-controlled crossover trial of continuous positive airway pressure for mild sleep apnea/hypopnea syndrome. 1999;159:461-467.
Chest.
40. Jokic R, Klimaszewski A, Crossley M, Sridhar G, Fitzpatrick M. Positional treatment vs continuous positive airway pressure in patients with positional obstructive sleep apnea. 1999;115:771-781.
Otolaryngol Head Neck Surg.
41. Piccirillo JF, Gates GA, White DL, Schectman KB. Obstructive sleep apnea treatment outcomes pilot study. 1998;118:833-844.
Sleep.
42. Johns MW. A new method for measuring daytime sleepiness: the Epworth Sleepiness Scale. 1991;14:540-545.
Electroencephalogr Clin Neurophysiol.
43. Mitler MM, Gujavarty KS, Browman CP. Maintenance of Wakefulness Test: a polysomnographic technique for evaluating treatment in patients with excessive somnolence. 1982;53:658-661.
Sleep.
44. Carskadon MA, Dement W, Mitler MM, Roth T, Westbrook PR, Keenan S. Guidelines for the Multiple Sleep Latency Test (MSLT): a standard measure of sleepiness. 1986;9:519-524.
Sleep.
45. Weaver TE, Laizner AM, Evans LK, et al. An instrument to measure functional status outcomes for disorders of excessive sleepiness. 1997;20:835-843.
Am J Roentgenol.
46. Dodds WJ. The pathogenesis of gastroesophageal reflux disease. 1988;151:49-56.
Oral Surg Oral Med Oral Pathol Oral Radiol Endod
47. Silva MAGS, Damante JH, Stipp AC, Tolentino MM, Carlotto PR, Fleury RN. Gastroesophageal reflux disease: new oral findings. . 2001;91:301-310.
Dig Dis Sci.
48. Svedlund J, Sjodin I, Dotevall G. GSRS: a clinical rating scale for gastrointestinal symptoms in patients with irritable bowel syndrome and peptic ulcer disease. 1988;33:129-134.
Aliment Pharmacol Ther.
49. Moayyedi P, Duffett S, Braunholtz D, et al. The Leeds Dyspepsia Questionnaire: a valid tool for measuring the presence and severity of dyspepsia. 1998;12:1257-1262.
J Clin Epidemiol.
50. Garratt AM. Developing a condition-specific measure of health for patients with dyspepsia and ulcer-related symptoms. 1996;49:565-571.
Eur J Gastroenterol Hepatol.
51. El-Omar EM, Banerjee S, Wirz A, McColl KE. The Glasgow Dyspepsia Severity Score: a tool for the global measurement of dyspepsia. 1996;8:967-971.
Dig Dis.
52. Moyer CA, Fendrick AM. Measuring health-related quality of life in patients with upper gastrointestinal disease. 1998;16:315-324.
Br J Surg.
53. Eypasch E, Williams JI, Wood-Dauhinee S, et al. Gastrointestinal Quality of Life Index: development, validation, and application of a new instrument. 1995;82:216-222.
J Am Coll Surg.
54. Velanovich V, Vallance SR, Gusz JR, Tapia FV, Harkabus MA. Quality of life scale for gastroesophageal reflux disease. 1996;183:217-224.
Ann Intern Med.
55. Guyatt GH, Feeny DH, Patrick DL. Measuring health-related quality of life. 1993;118:622-629.
Am J Gastroenterol.
56. Irvine EJ, Zhou Q, Thompson AK, Canadian Crohn's Relapse Prevention Trial Investigators. The Short Inflammatory Bowel Disease Questionnaire: a quality of life instrument for community physicians managing inflammatory bowel disease. 1996;91:1571-1578.
Gastroenterology.
57. Guyatt G, Mitchell A, Irvine JE, et al. A new measure of health status for clinical trials in inflammatory bowel disease. 1989;96:804-810.
Eur J Surg Suppl.
58. Chassany O, Bergmann JF. Quality of life in irritable bowel syndrome: effect of therapy. 1998;583:81-86.
Eur J Surg Suppl.
59. Wiklund IK, Junghard O, Grace E, et al. Quality of life in reflux and dyspepsia patients: psychometric documentation of a new disease-specific questionnaire (QOLRAD). 1998;583:41-49.
Aliment Pharmacol Ther.
60. Talley NJ, Haque M, Wyeth JW, et al. Development of a new dyspepsia impact scale: the Nepean Dyspepsia Index. 1999;13:225-235.
J Allergy Clin Immunol.
61. Flemons WW, Tsai W. Quality of life consequences of sleep-disordered breathing. 1997;99:S750-S756.
Dis Colon Rectum.
62. Maunder RG, Cohen Z, McLeod RS, Greenberg GR. Effect of intervention in inflammatory bowel disease on health-related quality of life: a critical review. 1995;38:1147-1161.
Sleep.
63. Flemons WW. Measuring health related quality of life in sleep apnea. 2000;23(suppl 4):S109-S114.
Scand J Gastroenterol Suppl.
64. Irvine EJ. Measuring quality of life: a review. 1996;31:5-7.
