Diabetes Spectrum
Volume 13 Number , 2000, Page 36
From Research to Practice/Diabetes and Quality of Life

Understanding and Assessing Diabetes-Specific Quality of Life

William H. Polonsky, PhD, CDE

In Brief

There is a growing interest in understanding and making good use of the concept of diabetes-specific quality of life. Unfortunately, the concept seems to have become more and more murky over the years, and there is significant confusion among clinicians and researchers about how it should be assessed. This paper puts forward a practical model of diabetes-specific quality of life and, using this model, reviews many of the self-report instruments that are currently used in this field. The paper concludes with recommendations for how diabetes-specific quality of life may be most effectively evaluated.

What is diabetes-specific quality of life (QoL), how important a concept is it, and how do we go about measuring it? To answer these questions, let us first consider the broad concept of health-related quality of life (HRQoL).

The medical community's interest in HRQoL, rather than the mere presence or absence of disease, has been growing at a phenomenal rate over the past few decades. As an illustration, note that the term quality of life was referred to in the medical literature approximately 40 times in the 8-year period between 1966 and 1974 versus more than 10,000 times in the 8-year period between 1986 and 1994.1 There can be little doubt that HRQoL is now a very hot topic—in diabetes as well as in most other areas of medical research.

The increasingly strong interest in evaluating HRQoL outcomes in clinical research is heartening. To a large degree, this interest reflects the growing recognition that HRQoL is, in the vast majority of cases, the single most important clinical and research outcome. The true impact of a successful medical intervention can be understood to largely reflect the degree to which said treatment has a positive influence on patients' immediate and/or future well-being. By emphasizing the assessment of HRQoL in medical research, health care providers are being subtly encouraged to give proper focus to patients' welfare. Consequently, medical research and care is slowly becoming more patient-focused, and there is a growing appreciation of the patient's perspective on health, disease, and medical treatments.

Unfortunately, clinical researchers in diabetes and other areas face a significant dilemma. Despite the growing pressure from funding agencies and/or government or institutional edicts to assess HRQoL, there is little agreement about how to do so. The concept of HRQoL remains ambiguous and poorly defined, and the number of different instruments that purport to assess the various aspects of this dimension is overwhelming.

In confusion and desperation, many researchers in diabetes seem to follow one of three rules in selecting appropriate instruments for their studies: a) use whatever everyone else seems to be using (thus, the ever-growing popularity of the Medical Outcomes Study SF-36 and its variants),2 b) assume that HRQoL is synonymous with psychosocial status and so use any available instrument that appears to assess some aspect of the patient's psyche (like depression), or c) choose any questionnaire, without worrying about the actual content, that has an appropriate name (i.e., includes the terms quality of life and diabetes in its title). With such an arbitrary approach towards the selection of appropriate instruments, it is not surprising that the hoped-for changes in HRQoL due to treatment intervention are often not found.

To overcome these problems, it is important to become more precise in defining HRQoL. In truth, the quality of one's life refers to the broadest range of human experience. In addition to the influence of health, it includes personal finances, job, housing, personal relationships, political and cultural climate, traffic, environmental considerations, and much more. HRQoL is primarily concerned with only one particular aspect of quality of life—how it may be affected by health and disease.

As a first step towards defining HRQoL, let us presume that there are two major types to consider, overall and disease-specific. Overall HRQoL refers to the patient's sense of his own health and well-being in the broad areas of physical, psychological, and social functioning. When evaluating overall HRQoL, objective health status is of secondary concern; it is the patient's personal perspective on his own well-being that is paramount. Overall HRQoL is understood to be a multidimensional construct, involving a variety of domains that can contribute independently to HRQoL. Researchers may make the mistake of evaluating overall HRQoL with a single questionnaire targeting a fairly limited range of subjective well-being—say, the patient's quality of sleep, level of depression, perceived functional limitations in daily life, or the burden of physical symptoms (e.g., chronic pain). However, these are only isolated components of the broader construct, and such questionnaires are not truly assessing HRQoL in a comprehensive fashion. The evaluation of overall HRQoL makes sense when the researcher is trying to compare different illnesses or different populations. As a means to measure within-subject changes in clinical trials (especially in diabetes-related studies), overall HRQoL scales are likely to be less sensitive to change than are disease-specific instruments.3,4

Unfortunately, the concept of disease-specific HRQoL has been, if possible, even more confused. Re-searchers have included a remarkable hodgepodge of variables under this rubric, including treatment satisfaction, disease-related self-efficacy, diabetes-related coping styles, and other feelings and beliefs about the disease. I would propose, however, a conceptual definition of disease-specific HRQoL based on subjective attributions of causality: as a subset of overall HRQoL, disease-specific HRQoL refers solely to patients' sense of how the disease in question is compromising their well-being in the three broad areas of physical, psychological, and social functioning. Within such a conceptual framework, disease-specific HRQoL includes two major categories of potential distress: intrinsic impairment (the disease, or some aspect of the disease, is perceived as directly burdensome or intrusive) and attributional impairment (the disease is perceived as being responsible for distress in one or more of the three broad areas of functioning).

Other important phenomena, such as coping style, disease-related self-efficacy, or most of the elements of treatment satisfaction, have no direct place in this model. They are important variables that will be associated with HRQoL (indeed, they may be powerful contributors to disease-specific HRQoL) and deserve to be carefully tracked in appropriate studies, but they cannot be considered to be synonymous with disease-specific HRQoL, nor are they truly elements of the construct. As mentioned above, disease-specific HRQoL measures may be of greater value than overall HRQoL instruments in most clinical trials because they are likely to be more sensitive to change. However, caution is indicated, since they may also be more sensitive to demand characteristics (change based on the assumption that an intervention should have produced change).

Turning our attention to the particular case of diabetes, we can now begin to explore the details of diabetes-specific HRQoL, a patient's perception of how diabetes affects his or her physical, psychological, and social functioning. In many ways, this construct is merely a formalized way of talking about the personal side of diabetes, the felt burden of living with the illness, how diabetes can get in the way of living a rich, full life.

And there is little doubt that diabetes can be a very tough illness with which to live. For many patients, the demands of self-care can be burdensome, frustrating, and overwhelming. The impact of long-term complications can be severe, leading to major changes in a patient's ability to function in daily life. The ongoing threat of complications can be worrisome and depressing. Social relationships may be severely affected. Adjustment to the disease is often accompanied by a variety of negative emotional responses, including anger, guilt, frustration, denial, and loneliness. The evaluation of diabetes-specific HRQoL involves the careful assessment of diabetes-specific impairment along the three major dimensions noted above.

How diabetes can compromise physical functioning
There are three major ways in which diabetes can negatively affect physical well-being. The most potent factor is the development of long-term complications. When patients suffer vision loss, kidney damage, significant heart disease, erection problems, peripheral neuropathy resulting in chronic pain, amputation, and/or difficulty walking, or any of a host of autonomic neuropathy problems (such as gastroparesis or loss of bladder function), there is likely to be a significant drop in perceived quality of life. The patient may now be unable or less able to work, to complete household tasks, or to enjoy pleasurable activities. The patient's ability to function independently may be impaired as well.

The second factor is short-term complications. Chronically elevated blood glucose levels may lead to increased fatigue, sleep problems, more frequent infections, and other associated problems. Tight glycemic control may lead to unwanted weight gain, more frequent hypoglycemia, and/or loss of hypoglycemic warning signs.

The third major factor concerns physical symptoms and lifestyle changes resulting from the demands of the diabetes regimen. Finally, when patients are forced (or believe that they are forced) to limit or curtail their activities in order to manage their diabetes effectively, quality of life is likely to be affected. Examples include declining late-night social engagements as a means to avoid dietary lapses or losing one's license to drive due to frequent and severe hypoglycemia. Unpleasant side effects due to prescribed medications may also affect perceived well-being—for instance, chronic gastrointestinal distress resulting from specific oral hypoglycemic agents or unsightly lypohypertrophy resulting from repeated insulin injections. To assess this dimension most effectively, evaluation might focus on the patient's perceived distress due to diabetes-specific symptoms as well as the perceived loss of physical function, interference with common role activities (including work, tasks at home, and social and recreational pastimes), and loss of independence due to diabetes.

How diabetes can compromise psychological functioning
The demands of diabetes care can have a potent impact on mood, both short-term and long-term. Many patients may become chronically frustrated, discouraged, and/or enraged with a disease that often does not seem to respond to their best efforts. They may also feel hopeless or despondent about the possibility of avoiding long-term complications. It can be a difficult, emotional struggle to find a way to include diabetes in one's life and to confront the sense of mortality that diabetes may represent. This may be especially problematic at those specific time points in the natural history of the illness when diabetes suddenly seems quite "real"—such as at diagnosis, if and when insulin is first started, and when long-term complications begin to occur.

In addition, chronically elevated blood glucose levels may lead to persistent fatigue, which can exacerbate depressed mood. Similarly, frequent hypoglycemic episodes can be exhausting, debilitating, discouraging, and potentially quite frightening.

Facing a disease that is often difficult and confusing to manage, patients may feel a pervading sense of helplessness that detracts significantly from the overall sense of well-being. To assess this dimension, evaluation might focus on patients' perceived emotional distress due to diabetes-related symptoms, self-care, relevant problematic situations, and broader diabetes issues.

How diabetes can compromise social functioning
The mere presence of diabetes can affect the quantity and quality of a patient's relationships. As patients begin to institute changes in daily habits in order to manage diabetes most effectively, loved ones may begin to rebel—choosing not to participate in any necessary changes (e.g., "It's your disease, not mine."), or actively sabotaging any and all self-care efforts (e.g., "C'mon, can't you just forget about your diabetes for one night?").

Alternatively, friends or family members may begin to push for self-care changes even when the patient is unwilling to make them. As loved ones begin acting like "diabetes police," the opportunities for interpersonal conflict escalate.

In either case, it is easy to begin feeling alone with diabetes, feeling different and unsupported, and believing that no can understand what living with diabetes is really like. To assess this dimension, evaluation might focus on a patient's perceived emotional distress due to diabetes-related social situations.

At this time, there is no well-accepted measure that comprehensively evaluates the many aspects of diabetes-specific HRQoL as defined here. Thus, careful clinical researchers will be forced to compromise, having to select an instrument or instruments that are less than perfect for the task at hand. Still, this selection process should not be as arbitrary as it often seems. There are many scales available, and they vary greatly in the degree to which they assess relevant areas of diabetes-specific HRQoL. Through careful review of the published questionnaires, we can decide on the best of what is currently available and make a more educated choice.

This review will be limited to self-report instruments that focus on the perspective of adults with diabetes. In truth, there are so many scales that touch on one or another aspect of HRQoL that it would be impossible to review them all in a responsible manner. Instead, a representative sample of relevant instruments that are widely used, well known, or have been recently published will be reviewed.

While reliability and responsiveness to change are important properties to consider when deciding whether to use an instrument, special attention must be given to validity. Typically, authors document their scale's psychometric validity by comparing it to one of a number of selected "gold standards"—glycosylated hemoglobin levels, SF-36 scores, number of long-term complications, and the like. However, in an area of study such as diabetes-specific HRQoL, in which definitions remain so murky, there is no acceptable gold standard. Perceived HRQoL will not necessarily be closely tied to biomedical markers of diabetes, so it makes little sense to consider glycemic control or severity of complications as an appropriate gold standard. Thus, while I make mention where validity studies have been conducted, I also follow a more "low-tech" solution approach—a review of each scale's items and their correspondence to the multi-faceted definition proposed above.

The Diabetes Quality of Life Measure (DQOL) was a groundbreaking instrument originally developed for use in the Diabetes Control and Complications Trial (DCCT) in the early 1980s.5 Designed for patients with type 1 diabetes, it evaluates the relative burden of an intensive diabetes treatment regimen. There are 46 core items (10 additional items for adolescents) and four major dimensions: treatment satisfaction, treatment impact, worry about long-term complications, and worry about social/vocational issues. Patients respond to all items on a 5-point Likert scale. A score of 1 indicates no impact, no worries, or always satisfied. A score of 5 represents always affected, always worried, or never satisfied.

This instrument remains the most widely used instrument for the assessment of diabetes-specific HRQoL, yet many problems are apparent. First, careful examination of the individual items points to significant content limitations in this measure. For example, many of the items pertain to overall HRQoL issues and are not diabetes-specific ("How satisfied are you with your social relationships and friendships?"; "How often do you feel good about yourself?"). Also, many of the items pertaining to treatment satisfaction do not readily fit into the restricted definition of diabetes-specific HRQoL as described above. In support of the scale's validity, lower scores on the DQOL have been found to be associated with more frequent and severe long-term complications and with glycemic control.4 However, given the problems with the scale's content, it is not surprising that some researchers have cast doubt on the sensitivity of the instrument as well as its validity as a measure of diabetes-specific HRQoL.6

The Diabetes-Specific Quality of Life Scale (DSQOLS) is a new instrument with 64 items, of which 10 focus on treatment goals, 10 on treatment satisfaction, and 44 on the perceived burden of diabetes.6 The treatment goals and treatment satisfaction questions do not meet the proposed diabetes-specific HRQoL criteria, but the third category of items, the perceived burden of diabetes, certainly does. These items fall into six dimensions: social relations, leisure time restrictions and flexibility, physical complaints, worries about the future, diet restriction, and daily hassles.

The scale was originally developed in Germany and has only been applied to type 1 diabetes populations. (An English language version has been published, but it has not yet been validated.) Good reliability and validity have been demonstrated, but there is nothing yet known about responsiveness to change.

This is a promising scale, but revision of items may be needed. For example, one item asks patients to what degree they agree with the statement "I am worried about my future health." Linear interpretation of this and similar items may be inappropriate. After all, a low to moderate amount of worrying may not be perceived as particularly bothersome to the individual (indeed, this may be the most realistic and proper response).

The Diabetes Quality of Life Clinical Trial Questionnaire—Revised (DQLCTQ-R) is a new, 57-item scale that comprises 8 dimensions: physical function, energy/fatigue, health distress, mental health, satisfaction, treatment satisfaction, treatment flexibility, and frequency of symptoms.7 Unlike many of the other scales described here, the authors have created a measure that attempts to assess overall HRQoL as well as diabetes-specific HRQoL. Toward this end, the majority of items are taken directly from the DQOL and from the SF-36. Results from the initial study suggest satisfactory validity and responsiveness to change. However, as a measure of diabetes-specific HRQoL, this instrument suffers from the same content limitations as the DQOL.

The Appraisal of Diabetes Scale (ADS) is a brief, 7-item scale that attempts to assess patients' feelings and attitudes about diabetes.8 Representative items include "How upsetting is having diabetes for you?" and "To what degree does your diabetes get in the way of developing your life goals?" All items require a response on a 5-point Likert scale.

To date, two studies have found a significant link between ADS scores and glycemic control (a more positive appraisal associated with a lower glycohemoglobin level).8,9 There is good support for the instrument's reliability and validity, but nothing is yet known about the scale's responsiveness to change.

A review of the items suggests that the scale does address one important aspect of diabetes-specific HRQoL, the impact of diabetes on psychological functioning, albeit quite briefly. However, it must be noted that a positive appraisal of diabetes is not to be equated with a realistic appraisal. Strongly positive responses to many of the questions, such as "How much control over your diabetes do you have?" may be indicative of a positive appraisal, but may also point to an inaccurate portrayal of the real world (and may lead to frustration and disappointment in the future).

The ATT-39 was one of the first instruments designed to assess psychosocial adjustment to diabetes.10 The 39 items comprise a range of attitudinal statements linked to patients' perception of diabetes and diabetes care. A representative item is "Most people would find it difficult to adjust to having diabetes." Patients indicate the degree to which they agree or disagree to each item along a 5-point Likert scale, with scores ranging from "disagree completely" to "agree completely." There are six subscales: diabetes stress, coping, guilt, alienation/cooperation, illness conviction, and ambiguity tolerance.

A revised version, the ATT-19, has recently been completed and appears to comprise the central core of the ATT-39.11 It comprises a single large factor, which has been labeled "diabetes integration."

While both scales overlap to some degree with the concept of diabetes-specific HRQoL (especially the psychological functioning component), the focus of the scales is on diabetes-related health beliefs, a construct that is distinctly separate from HRQoL.

The Questionnaire on Stress in Patients with Diabetes—Revised (QSD-R) contains 45 items that describe common stressful situations for people with type 1 or type 2 diabetes.12 For each item, patients indicate whether the item applies to them and then rate on a 5-point scale the degree to which that particular item is a problem for them. (1 indicates only a slight problem; 5 indicates a very big problem.) There are eight dimensions: leisure time, work, relationship with partner, doctor-patient relationship, problems with hypoglycemia, treatment regimen, physical complaints, and worries about long-term complications.

While several of the items are not directly linked to diabetes (e.g., "I suffer from irritability."), the majority of items fit well within our definition of diabetes-specific HRQoL and tap all three of the major dimensions. Good reliability and validity have been demonstrated, and research indicates that each of the subscales is has been found to be significantly correlated with glycemic control (in the expected direction). Like the DSQOLS, this was originally a German scale. An English version does exist and can be obtained from the authors. However, it has not yet been validated.

The Type 2 Diabetes Symptom Checklist is a 34-item scale designed to evaluate the perceived burden of six categories of symptoms presumed to be associated with diabetes: hyperglycemic, hypoglycemic, cardiac, neuropathic, psychological, and vision-related.13 Patients respond on a 5-point Likert scale, with a 1 indicating that the symptom has not occurred or was not perceived as troublesome and a 5 indicating that the symptom was felt to be extremely troublesome.

The instrument appears to be valid, reliable, and responsive to change. Recent evidence suggests that higher scores are associated with poorer glycemic control, albeit weakly.14

There is no doubt that this instrument covers important sections of the physical functioning aspect of diabetes-specific HRQoL. However, one of the difficulties with this approach is that, for many of the symptoms listed, it may be difficult to determine whether the reported symptom distress is actually due to diabetes. Patients may report being extremely bothered by such symptoms such as "moodiness," "sleepiness or drowsiness," or "deteriorating vision," but this may have little to do with their diabetes. To be fair, however, it is hard to imagine a self-report scale that could accurately determine symptom burden solely due to diabetes.

The scale is of Dutch origin, but an English version has been developed and validated.

The Problem Areas in Diabetes Scale (PAID-1) is a 20-item scale that describes common problematic situations for people with type 1 or type 2 diabetes, each representing a unique area of diabetes-specific emotional distress.15,16 These range from anger ("feeling angry when you think about having diabetes and living with diabetes") and interpersonal distress ("feeling that your friends and family are not supportive of your diabetes efforts") to frustration with aspects of the diabetes regimen ("not having clear and concrete goals for your diabetes care"). On a 6-point Likert scale, patients rate the degree to which each item is currently problematic for them, from 1 ("no problem") to 6 ("serious problem").

As can be seen, the scale is similar in content and format to the QSD-R. Internal reliability is high, and there are no subscales. In addition, validity has been well-established,17 and recent intervention studies have shown the instrument to be responsive to change.18,19

To broaden the scope of the instrument, a new version has been recently developed, the PAID-2.20 The new scale is somewhat longer (28 items), but four diabetes-specific dimensions now are assessed: overall emotional distress, interpersonal distress, regimen-related distress, and physician-related distress. Reliability and validity look promising, and data are currently being collected to evaluate responsiveness to change.

The Audit of Diabetes-Dependent Quality of Life (ADDQoL) is a new, 15-item scale that makes use of a quite innovative approach towards evaluation.21 The instrument presents a comprehensive list of 13 life domains that diabetes might affect: employment/career opportunities, social life, family relationships, friendships, sex life, recreational opportunities, ease of travel, personal worries about the future, worries about the future for one's family and friends, motivation to achieve things, physical activities, potential loss of independence, and eating enjoyment.

Patients are asked to indicate, on a 7-point Likert scale, the degree to which each particular domain might be different "if I did not have diabetes." Potential scores range from 3, indicating that life in this domain would be a great deal better, to +3, indicating that life in this domain would be a great deal worse. Patients are then directed to rate how personally important each domain is to them along a 4-point scale, ranging from very important to not important at all. Using the latter responses, a patient's 13 scores can then be arithmetically weighted, such that the total HRQoL score is more strongly influenced by those domains that a patient has selected as being most important. (Two additional items concern estimates of overall HRQoL, comparing life with and without diabetes.)

Importantly, of the nine scales described here, the ADDQoL is the only one that attempts to evaluate diabetes-specific HRQoL in a comprehensive manner from an attributional perspective (how diabetes may be perceived as interfering with well-being). In contrast, the majority of the other scales are working from an intrinsic perspective (how the different aspects of diabetes may be perceived as burdensome). Initial reliability and validity for the ADDQoL look good, but there are as yet no data available on responsiveness to change.

The major point of concern about this questionnaire is that the logic of the stem question is quite different from all other available instruments. Rather than asking about the degree to which problems associated with diabetes are occurring, this scale asks patients to imagine how life might be different without diabetes. This seems a more complex cognitive task, one step removed from direct questions about diabetes-specific HRQoL. And it is not yet clear that such an approach will actually allow re-searchers to accurately determine how diabetes is affecting these broad areas. However, one of the great advantages of this approach is that it allows respondents to indicate how diabetes may be having a positive effect in certain domains. In sum, this is a fascinating and promising approach towards diabetes-specific HRQoL that requires further study. At this time, additional revision and refinement of the initial instrument is apparently underway.

While these nine instruments cover the broad array of diabetes-specific HRQoL assessment tools, there are many others. These range from scales that cover a small though important portion of the concept, such as the Hypoglycemic Fear Survey,22 to more extensive measures, such as the Diabetes Care Profile23 and the HRQoL scale developed by Testa and Simonson.24,25 These latter instruments are designed to evaluate a range of factors beyond diabetes-specific HRQoL, but each includes subsets of questions that are directly relevant to diabetes-specific HRQoL. Other promising questionnaires appear to be in an early phase of development, such as the Diabetes-3926 and the Diabetes Health Profile,27 and may turn out to be quite valuable as research progresses.

Given the plethora of available instruments, what should savvy clinical researchers do? Of course, the actual selection of measures should follow from the research questions that are being investigated, but general recommendations are possible. Studies concerned with HRQoL should generally include instruments that evaluate overall HRQoL as well as diabetes-specific HRQoL. In this issue, the paper by Snoek (Quality of Life: A Closer Look at Measuring Patients' Well-Being p. 24) covers the assessment of overall HRQoL and the selection of appropriate instruments in that domain. Regarding diabetes-specific HRQoL, most of the measures described above provide a valuable means for the evaluating at least a portion of the larger construct. Unfortunately, none of the instruments manages to comprehensively assess all three major domains, neither from an intrinsic perspective nor from an attributional perspective. There-fore, to make appropriate recommendations, we need to consider evaluation within each of the three domains.

Assessing the impact of diabetes on physical functioning
Many of the instruments described here include items or subscales concerning the physical functioning domain (including the DQOL, DSQOLS, DQLCTQ-R, QSD-R, and ADDQoL), but the Type 2 Diabetes Symptom Checklist is the only scale that appears to evaluate this dimension in a broad, comprehensive manner. The perceived burden of diabetes-specific symptoms is the central aspect of this domain, and this is captured relatively well by this measure (though, as mentioned earlier, there are some limitations to this instrument). Unfortunately, I am not aware of any instrument that examines the other major aspect of the physical functioning domain, impairment in common role activities due to diabetes, in a thorough fashion.

Assessing the impact of diabetes on psychological functioning
All of the instruments include items or subscales that attempt to assess one or more of the aspects of this domain (though some, such as the DQOL and the Type 2 Diabetes Symptom Check-list, do not ask specifically whether the felt impairment is due to diabetes in particular). The perceived burden of diabetes-related psychological distress is covered most carefully and comprehensively by the QSD-R, the DSQOLS, and the two PAID measures.

However, there is no validated English translation yet available for the QSD-R or DSQOLS. In addition, only the PAID-1 has been shown to be responsive to change. When validated English translations of the two German instruments are available, especially the QSD-R, they should be strongly considered for diabetes-related HRQoL in this domain, especially the QSD-R.

Assessing the impact of diabetes on social functioning
Many of the instruments include items or subscales concerning this domain (including the DQOL, DSQOLS, QSD-R, the two PAID measures, and ADDQoL), but all suffer from certain weaknesses. The ADDQoL addresses this domain fairly thoroughly from an attributional perspective, but the stem question (asking patients to estimate how HRQoL would be different if they did not have diabetes) is of some concern. From an intrinsic perspective, the perceived burden of diabetes-related social distress is covered fairly well by the QSD-R, the DSQOLS, and the PAID-2 (each has a separate subscale evaluating this domain). As mentioned earlier, neither the QSD-R nor DSQOLS is currently available in validated English translations, and the PAID-2 has not yet been published.

Proper assessment of diabetes-specific HRQoL, therefore, necessitates different instruments depending on the particular domain or domains that are of interest. None of the available scales manages to adequately cover all of the important facets of diabetes-related HRQoL as defined here. Of course, as current instruments are refined and tested more thoroughly (especially the ADDQoL) and new, more comprehensive instruments are published, these recommendations are likely to change.

This article has taken an initial step towards defining diabetes-specific HRQoL and discriminating this construct from other important psychosocial factors, such as treatment satisfaction. The importance of evaluating diabetes-specific HRQoL in clinical research has been explored, and many of the major instruments for assessing the construct have been described and critiqued. Based on this review, preliminary recommendations were made for selecting the best instruments to be used—based primarily on the scale content and its degree of fit with the proposed definition of diabetes-specific HRQoL.

Many are likely to disagree with the proposed definition and with the selection of recommended instruments. Such comments will be warmly welcomed. Indeed, one of the major purposes of this paper has been to move our professional conversation forward, to engender conversation among colleagues as to the nature of diabetes-specific HRQoL and its assessment, and to further refine our concepts and tools.

1Albrecht GL: Subjective health assessment. In Measuring Health and Medical Outcomes. Jenkinson C, Ed. London, UCL Press, 1994, p. 7-26.

2Ware JH, Sherbourne CD: The MOS 36-item Short Form Health survey (SF-36) I: conceptual framework and item selection. Med Care 30: 473-83, 1992.

3Anderson RM, Fitzgerald JT, Wisom K, Davis WK, Hiss RG: A comparison of global versus disease-specific quality-of-life measures in patients with NIDDM. Diabetes Care 20:299-305, 1997.

4Jacobson AM, de Groot M, Samson JA: The evaluation of two measures of quality of life in patients with type I and type II diabetes. Diabetes Care 17:267-74, 1994.

5Jacobson AM, The DCCT Research Group: The diabetes quality of life measure. In Handbook of Psychology and Diabetes. Bradley C, Ed. Chur, Switzerland, Harwood Academic Publishers, 1994, p. 65-87.

6Bott U, Muhlhauser I, Overmann H, Berger M: Validation of a diabetes-specific quality-of-life scale for patients with type 1 diabetes. Diabetes Care 21:757-69, 1998.

7Shen W, Kotsanos JG, Huster WJ, Mathias SD, Andrejasich CM, Patrick DL: Development and validation of the Diabetes Quality of Life Clinical Trial Questionnaire. Med Care 37:AS45-66, 1999.

8Carey MP, Jorgensen RS, Weinstock RS, Sprafkin RP, Lantinga LJ, Carnrike CL Jr, Baker MT, Meisler AW: Reliability and validity of the appraisal of diabetes scale. J Behav Med 14:43-51, 1991.

9Trief PM, Grant W, Elbert K, Weinstock RS: Family environment, glycemic control, and the psychosocial adaptation of adults with diabetes. Diabetes Care 21:241-45, 1998.

10Dunn SM, Smartt H, Beeney L, Turtle J: Measurement of emotional adjustment in diabetic patients: validity and reliability of ATT39. Diabetes Care 9:480-89, 1986.

11Welch G, Dunn SM, Beeney LJ: The ATT39: a measure of psychological adjustment to diabetes. In Handbook of Psychology and Diabetes. Bradley C, Ed. Chur, Switzerland, Harwood Academic Publishers, 1994, p. 223-46.

12Herschbach P, Duran G, Waadt S, Zettler A, Amm C, Marten-Mittag B: Psychometric properties of the Questionnaire on Stress in Patients with Diabetes—Revised (QSD-R). Health Psychol 16:171-74, 1997.

13Grootenhuis PA, Snoek FJ, Heine RJ, Bouter LM: Development of a type 2 diabetes symptom checklist: a measure of symptom severity. Diabetic Med 11:253-61, 1994.

14Van der Does FE, De Neeling JN, Snoek FJ, Kostense PJ, Grootenhuis PA, Bouter LM, Heine RJ: Symptoms and well-being in relation to glycemic control in type II diabetes. Diabetes Care 19:204-10, 1996.

15Polonsky WH, Anderson BJ, Lohrer PA, Welch G, Jacobson AM, Schwartz C: Assessment of diabetes-specific distress. Diabetes Care 18:754-60, 1995.

16Polonsky WH, Welch G: Listening to our patients' concerns: understanding and addressing diabetes-specific emotional distress. Diabetes Spectrum 9:8-11, 1996.

17Welch G, Jacobson AM, Polonsky WH: The Problem Areas in Diabetes Scale: an evaluation of its clinical utility. Diabetes Care 20:760-66, 1995.

18Snoek FJ, Van der Ven NC, Heine RJ: Improving glycemic control and well-being: a cognitive behavioral based group training for poorly controlled type I diabetes patients (Abstract). Diabetes 48 (Suppl 1):A9, 1999.

19Weinger K, Kinsley BT, Bajaj M, Levy C, Simonson DC, Jacobson AM: Quality of life predictors of glycemic control during intensive treatment in type I diabetes. Ann Behav Med 20:S118, 1998.

20Polonsky WH, Dudl RJ, Peterson M, Steffian G, Lees J, Hokai H: The PAID-2: a newly revised and expanded version of the Problem Areas in Diabetes Scale. In preparation.

21Bradley C, Todd C, Gorton T, Symonds E, Martin A, Plowright R: The development of an individualized questionnaire measure of perceived impact of diabetes on quality of life: the ADDQoL. Qual Life Res 8:79-91, 1999.

22Cox D, Irvine A, Gonder-Frederick L, Nowacek G, Butterfield J: Fear of hypoglycemia: quantification, validation, and utilization. Diabetes Care 10:617-21, 1987.

23Fitzgerald JT, Davis WK, Connell CM, Hess GE, Funnell MM, Hiss RG: Development and validation of the Diabetes Care Profile. Eval Health Professions 19:208-30, 1996.

24Testa MA, Simonson DC: Measuring quality of life in hypertensive patients with diabetes. Postgrad Med J 64:50-58, 1988.

25Testa MA, Simonson DC, Turner RR: Valuing quality of life and improvements in glycemic control in people with type 2 diabetes. Diabetes Care 21:C44-52, 1998.

26Boyer JG, Earp JA: The development of an instrument for assessing the quality of life of people with diabetes. Diabetes-39. Med Care 35: 440-53, 1997.

27Meadows K, Steen N, McColl E, Eccles M, Shiels C, Hewison J, Hutchinson A: The Diabetes Health Profile (DHP): a new instrument for assessing the psychosocial profile of insulin requiring patients—development and psychometric evaluation. Qual Life Res 5:242-54, 1996.

Return to Issue Contents

Copyright 2000 American Diabetes Association

Last updated: 3/00
For Technical Issues contact