S38 Journal of Pain and Symptom Management Vol. 24 No.

1S July 2002

Proceedings from the Roundtable on “The Role of Coxibs in Successful Pain Management”

The Impact of Pain Management

on Quality of Life
Nathaniel Katz, MD
Pain Trials Center, Brigham and Women’s Hospital, Harvard Medical School, Boston, Massachusetts, USA

Abstract
Although its inclusion in medical research is relatively recent and its interpretation is often
variable, quality of life is increasingly being recognized as one of the most important
parameters to be measured in the evaluation of medical therapies, including those for pain
management. Pain, when it is not effectively treated and relieved, has a detrimental effect on
all aspects of quality of life. This negative impact has been found to span every age and every
type and source of pain in which it has been studied. Effective analgesic therapy has been
shown to improve quality of life by relieving pain. Opioid analgesics, cyclooxygenase (COX)-2
inhibitors (or coxibs), and several adjuvant analgesics for neuropathic pain have been
demonstrated to significantly improve quality-of-life scores in patients with pain. Coxibs
provide effective, well-tolerated analgesia without some of the issues faced with opioids—
benefits that should translate into improved quality of life. Recent studies have demonstrated
that the COX-2 inhibitor rofecoxib significantly improves quality of life in patients with
osteoarthritis and chronic, lower back pain. Quality-of-life measurements, especially symptom
distress scales, can also be used as sensitive means of differentiating one agent from another in
the same class. In future pharmacotherapeutic research, quality of life should be included as
an outcome domain as are the traditionally measured variables of efficacy and safety. In
particular, future studies of coxibs should include symptom distress scores as important
quality-of-life measurements, to identify meaningful differences between this new class of
analgesics and nonselective nonsteroidal anti-inflammatory drugs. J Pain Symptom
Manage 2002;24:S38–S47. © U.S. Cancer Pain Relief Committee, 2002.

Key Words
Analgesia, coxibs, COX-2, pain, quality of life, symptom distress

Introduction World Health Organization, individuals who

Pain is not only a highly noxious experience
per se, but it can also have an overwhelmingly
negative effect on nearly every other aspect of
life, including mood and capacity to function
in daily roles. According to a study by the
Address reprint requests to: Nathaniel Katz, MD, 212
Winding Brook Road, New Rochelle, NY 10804, USA.
Accepted for publication: March 3, 2002.
live with persistent pain are four times more
likely than those without pain to suffer from
depression or anxiety, and more than twice as
likely to have difficulty working.1
Pain is one of the most significant healthcare
crises in the United States. Nearly half of
Americans see a physician with a primary com-
plaint of pain each year,2 making pain the sin-
gle most frequent reason for physician consul-
tation in the United States.3 Even this fact
belies the true magnitude of the problem,

© U.S. Cancer Pain Relief Committee, 2002 0885-3924/02/$–see front matter

Published by Elsevier, New York, New York PII S0885-3924(02)00411-6
Vol. 24 No. 1S July 2002 Pain Management Impact on Quality of Life
since a substantial number of people with pain
do not consult a physician. In one of the larg-
est survey studies on the subject of pain, 18%
of respondents who rated their pain as severe
or unbearable had not visited any healthcare
professional, because they did not think that
anyone could relieve their suffering.4
The costs associated with pain are extremely
high, both to the healthcare system and to soci-
ety at large. Not only do individuals with pain
have a greater rate of utilization of the health-
care system, but their productivity is substan-
tially diminished. It has been estimated that
more than 4 billion workdays are lost to pain
annually. If one assumes a very conservative
median US income of $23,000, then pain costs
society $55 billion in lost productivity for full-
time workers alone.4
While these costs are enormous, one of the
greatest tolls exacted by pain is on quality of
life. Pain is widely accepted to be one of the
most important determinants of quality of life,
which can be defined as an individual’s ability
to perform a range of roles in society and to
reach an acceptable level of satisfaction from
functioning in those roles.5,6 However, quality-
of-life research is, relatively speaking, in its in-
fancy, and the effect of symptoms such as pain
on quality of life is just beginning to be under-
stood.7,8 Increasingly, however, quality of life is
coming to be accepted as one of the most im-
portant outcome domains to be measured in
the evaluation of any therapy or health-related
intervention.9 Quality of life is a more subtle
indicator than the typically measured variables
of efficacy and safety, but it is arguably more in-
dicative of treatment value and may be more
relevant to both patient satisfaction and will-
ingness to adhere to treatment.
Measuring Quality of Life: The Scales
and Beyond
Quality of life can be measured in a wide va-
riety of ways, and an array of instruments has
been developed to evaluate and attempt to
quantify it. Several questions need to be an-
swered to select the optimal instrument for any
given circumstance.10 In the present context, it is
assumed that we are referring to health-related
quality of life, which is more specific than general
quality of life.
S39
Which Is More Applicable, a Disease-Specific or
a Generic Instrument?
Specific instruments are designed to measure
quality of life in a particular disease state, such
as cancer or arthritis. Numerous specific instru-
ments are available in nearly every disease cate-
gory; for example, there are at least four instru-
ments that are specific to prostate cancer alone.
The disadvantage of specific instruments is that
their use makes it impossible to compare find-
ings across disease states. Generic instruments
are intended to measure quality of life in any
disease state and across disease states as well.
Their advantage is that they allow for groups of
patients with various conditions to be com-
pared with one another. Their disadvantage,
however, is that because they involve many dif-
ferent types of constructs, and are so general,
they are often not very effective at measuring
improvement in a specific disease state as a con-
sequence of an intervention. Thus, they may
not pick up subtle but important shifts in qual-
ity of life resulting from a given treatment.
The classic example of a generic quality of life
instrument is the Medical Outcomes Study Short
Form 36, or SF-36.10 The SF-36 is a 36-item survey
of general health status that was designed to
combine the comprehensiveness of much longer
surveys with the brevity of relatively coarse sin-
gle-item surveys. It can be self-completed, admin-
istered by computer, or conducted by a trained
interviewer in person or over the telephone.11
What Dimensions of Quality of Life Need
to Be Measured?
Quality of life is inherently a multidimen-
sional phenomenon, and most useful quality-
of-life instruments reflect this. There are do-
main-specific quality-of-life instruments, which
measure a single aspect of quality of life, such
as physical function or anxiety. However, mul-
tidomain instruments are generally preferred,
since an instrument that does not include several
dimensions will make it impossible to deter-
mine the nature of a score change.10 Although
some instruments have more domains,9 most
acceptable quality-of-life assessment strategies
address several or all of the following domains:
physical, psychological, social, somatic, and
spiritual.10 The SF-36 includes 8 domains: phys-
ical, role-physical, bodily pain, general health,
S40
vitality, social functioning, role-emotional, and
mental health.11
How Much Responder Burden Is Acceptable?
Responder burden refers to the amount of
effort that the patient must extend to complete
the evaluation. The number of instruments,
the number of questions in the instruments,
and the conceptual difficulty of the response
task must be considered. This is particularly
important when measuring quality of life for
patients in pain, because of the debilitating na-
ture of the condition. While some patients are
grateful for the caring and concern implied by
the effort to solicit their feelings about their
quality of life, others may be too incapacitated
to fully comply. Less-than-full compliance can
lead to inaccurate results.10
What Are the Administrative Issues That Need
to Be Considered?
Most quality of life evaluations require mea-
surement at a minimum of two intervals—the
baseline and then at a later point, typically af-
ter some form of treatment has been adminis-
tered. Therefore, arrangements need to be
made to ensure that the greatest possible num-
ber of patients complete follow-up evaluations.
In addition, decisions need to be made about
whether patients must complete their own eval-
uations (which may be difficult for patients
who suffer from severe pain) or whether proxy
completion will be permitted. It is generally
agreed that self-report data are preferred be-
cause they decrease the possibility that proxies
may unwittingly bias the results by allowing
their own feelings and opinions related to
quality of life to be reflected in their responses.
Instruments written at low-literacy levels or in
multiple languages can help decrease the need
for proxy involvement, as can allowing the op-
tion of audiotaped or computerized question-
naire completion, particularly for patients whose
movement is limited by severe pain.10
Has the Instrument Proven to Be Both Valid
And Reliable?
Validity and reliability are crucial character-
istics of a useful quality-of-life questionnaire. A
valid instrument is one that actually measures
what it is intended to measure, whereas a reli-
able questionnaire is one that provides a repro-
ducible result. A questionnaire administered
Katz Vol. 24 No. 1S July 2002
on Day 1 and repeated a few days later should
produce a nearly identical score, provided that
no new variables have been introduced in the
interim.10
Newer Approaches to Quality-of-Life Assessment
New approaches to the field of quality-of-life
research add to the usefulness and interpret-
ability of quality-of-life questionnaires. The
Symptom Distress Inventory method, for exam-
ple, involves providing patients with a checklist
that allows them to indicate which disease-spe-
cific symptoms they have and how much dis-
tress each symptom produces. The magnitude
of symptom distress has been found to be
strongly correlated with traditional quality-of-
life assessment tools and may in some cases be
the most sensitive way to address health-related
quality of life.12 Furthermore, symptom distress
methods have been shown to be more sensitive
than traditional quality-of-life instruments in
differentiating the impact of various drugs on
quality of life.13 Thus, when two drugs have
equivalent efficacy but different side effect pro-
files (a common situation), these distinctions
in side effects, picked up most sensitively by
the symptom distress method, may underlie
important differences in quality of life for
those on the medications.
Utility methods enable the evaluation of treat-
ment-related factors that affect quality of life
(e.g., degree of pain relief or propensity to
cause a side effect such as nausea) in the con-
text of patient preferences. For example, pa-
tients may be asked to weigh the relative impor-
tance of various symptoms or other health-
related factors (e.g., cost of treatment, life pro-
longation). Different treatments are then com-
pared according to improvement in overall util-
ity, rather than using a simple unidimensional
outcome variable. Finally, calibration methods
allow changes in quality of life to be evaluated
comparatively against other stressful life events
(e.g., job loss), thereby providing a comparative
gauge of what magnitude of change on a quality
of life (or symptom distress) scale is significant.
The Impact of Uncontrolled Pain
on Quality of Life
Pain and quality of life are phenomena that
share several fundamental characteristics. Pain
Vol. 24 No. 1S July 2002 Pain Management Impact on Quality of Life S41

has been defined by the American Pain Society
as “an unpleasant sensory and emotional expe-
rience associated with actual or potential tissue
damage.” Similarly, the Joint Commission on
Accreditation of Healthcare Organizations
notes that pain is a common experience that
has adverse physiological and psychological ef-
fects when unrelieved.14 Hence, pain involves
cognitive, motivational, affective, behavioral,
and physical components. Quality of life, a con-
struct that incorporates all factors that impact
on an individual’s life, has a similar all-encom-
passing nature.5,15 Indeed, the World Health
Organization’s list of the domains and facets
that comprise quality of life confirms the all-
embracing nature of the concept (Table 1).9
Pain, when it is ongoing and uncontrolled,
has a detrimental, deteriorative effect on virtu-
ally every aspect of a patient’s life. It produces
anxiety and emotional distress; undermines
well-being; interferes with functional capacity;
and hinders the ability to fulfill family, social,
and vocational roles. With such broad-based ef-
fects, it is apparent that pain would have the ef-
fect of diminishing quality of life.
The deteriorative effect on quality of life is
universal; it spans every age and stage of life and
occurs regardless of the pain’s type or source.
For example, in a study of 49,971 elderly nursing
home residents with disorders of nearly every
kind, Won and colleagues found that more than
one in four (26.3%) experienced pain on a daily

Table 1
Domains and Facets of Quality of Life, as Defined by the World Health Organization9
Domain I: Physical
• Pain and discomfort.
• Energy and fatigue.
• Sexual activity.
• Sleep and rest.
• Sensory functions.
Domain II: Psychological
• Positive feelings.
• Thinking, learning, memory, and concentration (cognitions).
• Self-esteem.
• Body image and appearance.
• Negative feelings.
Domain III: Level of Independence
• Mobility.
• Activities of daily living.
• Dependence on medication or treatment.
• Dependence on nonmedicinal substances.
• Communication capacity.
• Working capacity.
Domain IV: Social Relationships
• Personal relationships.
• Practical social support.
• Activities as provider/supporter.
Domain V: Environmental Health
• Physical safety and security.
• Home environment.
• Work satisfaction.
• Financial resources.
• Health and social care; availability and quality (services).
• Opportunities for acquiring new information and skills.
• Participation in and opportunities for recreation and leisure activities.
• Physical environments.
• Transport.
Domain VI: Spirituality
• Spirituality, religion, and personal beliefs.
General Facet
• Overall perceptions of health and quality of life.
World Health Organization Quality of Life Group. 1995.
Adapted with permission from Ref. 9.
S42
basis.16 A strong association was found between
daily pain and indices of poor quality of life: Pa-
tients who suffered from daily pain were more
likely to have impairment in activities of daily liv-
ing (odds ratio [OR]: 2.47), mood disorders
(OR: 1.66), and decreased involvement in activi-
ties (OR: 1.35). These associations persisted
even after the investigators adjusted for the po-
tentially confounding effects of age, gender,
race, cognitive status, and such debilitating con-
ditions as arthritis, stroke, congestive heart fail-
ure, and Parkinson’s disease.16
The younger end of the age spectrum is
equally vulnerable to the detrimental effects of
pain on quality of life. In a study of 128 adoles-
cents with chronic pain, Hunfeld and research-
ers found that quality of life decreased as in-
tensity and frequency of pain increased.17 The
domains of psychological functioning (including
feeling less at ease), physical status (including an
increase in incidence of other somatic com-
plaints), and functional status (defined as greater
impediments to leisure and daily activities) were
particularly affected. Notably, surveys of the pa-
tients’ mothers revealed that the adolescents’ pain
reduced their families’ quality of life as well.
The damaging effects of pain on quality of life
have been demonstrated for nearly every kind of
pain, including neuropathic pain, other chronic
nonmalignant pain such as that associated with
arthritis, and malignant pain.5,8,18–20 For exam-
ple, in a study of 150 patients with chronic pain,
including pain of neuropathic, somatic, psy-
chogenic, and visceral origins, Becker and col-
leagues found that scores on both the Psycholog-
ical General Well-Being (PGWB) scale and the
SF-36 were significantly reduced compared with
scores in the normal population (P  0.001).19
(The PGWB is a 22-item instrument designed to
measure subjective psychological well-being in
population-based studies. It includes six pa-
rameters: anxiety, depression, vitality, positive
well-being, self-control, and general health.19 All
eight SF-36 subscores, including bodily pain,
general health, mental health, physical function-
ing, role-emotional, role-physical, social function-
ing, and vitality were significantly reduced
compared with subscores for individuals without
pain (Figure 1).19 Furthermore, 40% of patients
with pain had scores on the Hospital Anxiety
and Depression scale that indicated the pres-
ence of a depressive disorder, whereas 50% had
scores indicating a comorbid anxiety disorder.
Katz Vol. 24 No. 1S July 2002
Fig. 1. Effect of chronic nonmalignant pain on qual-
ity of life, as indicated by SF-36 subscores, mean (SD)
(n 150).19 •  population norm values; PF  physi-
cal functioning; RP  role-physical; BP  bodily
pain; GH  general health; VT  vitality; SF  social
functioning; RE  role-emotional; MH  mental
health. *P  0.001. Adapted with permission from
Ref. 19.
The impact of malignant pain on quality of
life is similarly severe. Rummans and cowork-
ers studied the effect of pain on quality of life
in 117 patients with recurrent breast or gyneco-
logical cancer.5 The investigators found a sub-
stantial correlation between the presence of
pain and the physical and social dimensions of
quality of life. To their surprise, however, they
found a weaker correlation between pain and
the psychiatric and spiritual quality-of-life do-
mains. They attributed this aberrant finding to
the fact that the majority of these patients were
experiencing mild to moderate pain and none
were experiencing severe, incapacitating pain.5
The majority of studies have demonstrated
that there is a dose-response relationship be-
tween pain and quality of life: as one increases,
the other proportionately decreases.8,17,19,21 For
example, in their study of 216 adults with vari-
ous forms of cancer grouped by level of pain
severity, Wang and colleagues found that those
with moderate or severe pain had consistently
lower SF-36 scores than patients with no pain
or mild pain (Figure 2).8 All mean Mental and
Physical Component Summary scores declined
as pain severity increased (P  0.001 for both),
and this relationship was found to exist inde-
pendent of Eastern Cooperative Oncology
Group (ECOG) performance status.8
Cleeland and Ryan stated that it is more im-
portant to know the intensity of a patient’s
pain than to know merely whether or not pain
Vol. 24 No. 1S July 2002 Pain Management Impact on Quality of Life S43

dertreated and poorly controlled.8,22,23 The in-

adequacy of current efforts at pain control,
which is widely acknowledged by physicians, is
perhaps particularly striking in the field of on-
cology. In the aforementioned study of cancer
patients conducted by Wang and colleagues,
59% of patients who received treatment for
pain had a negative Pain Management Index,
meaning that their analgesic treatment did not
meet the minimum standards of the World
Health Organization guidelines.8
Similarly, in a study supported by ECOG, the
National Cancer Institute, the National Insti-
Fig. 2. Mean SF-36 subscale scores shown by pain se-
verity level.8 PF  physical functioning; RP  role tutes of Health, and the Department of Health
limitations due to physical problems; BP  bodily and Human Services, Cleeland and colleagues
pain; GH  general health perception; VT  vital- asked a group of 1308 outpatients with meta-
ity; SF  social functioning; RE  role limitations static cancer from 54 ECOG-affiliated locations
due to emotional problems; MH  mental health. to rate the severity of their cancer pain during
Adapted with permission from Ref. 8.
the preceding week, the degree of pain-related
functional impairment they experienced, and
the degree of relief provided by their analgesic
is present.21 “Many adults, including cancer pa- regimens.22 Of the group, 871 of 1308 (67%) re-
tients, function quite effectively with back- ported that they had experienced pain or taken
ground levels of pain which, for the most part, analgesics in the week preceding the study, and
are not attended to. As pain increases, how- 475 of 1306 (36%) said that their pain was se-
ever, it passes a threshold beyond which it can vere enough that it impaired their ability to
no longer be ignored. At this point, it becomes function. Of the 597 patients for whom com-
disruptive to many aspects of the person’s life.” plete information was available, 250 (42%) re-
According to their model, a progressively ceived inadequate analgesia. Factors associated
greater number of quality-of-life domains are with poor pain management included minority
impacted as pain becomes progressively worse race/ethnicity; greater discrepancy between pa-
(Table 2).21 tient and physician in judging degree of pain in-
The direct and unambiguous association terference with activity; and pain unrelated to
that exists between pain and quality of life cancer, older age, female sex, and better ECOG
would seem to highlight the importance of performance status (i.e., physician’s judgment
treating and effectively relieving pain. Unfortu- that the patient was relatively less ill).
nately, the evidence overwhelmingly demon- When queried, physicians admit that their
strates that despite the availability of effective efforts at pain management are largely inade-
analgesic pharmacotherapy, pain is often un- quate. In a survey of all ECOG-affiliated physi-
cians with pain management responsibility,
Von Roenn and coworkers found that only
Table 2 51% believed that pain control in their own
Activities/Quality-of-Life Domains Impaired by practice settings was good or very good; 31%
Increasing Pain Severity21 described it as fair, and 18% said that it was
Relate poor or very poor.23
Walk Walk
Sleep Sleep Sleep Sleep
Active Active Active Active
Mood Mood Mood Mood
Work Work Work Work Work Effective Pain Control: Its Salutary
Enjoy Enjoy Enjoy Enjoy Enjoy Enjoy Effect on Quality of Life
3 4 5 6 7 8
Worst pain rating If poorly controlled pain has a deteriorative
Note: Boldface indicates an additional dimension that is impaired at
effect on quality of life, then the implication is
the given level of pain severity. Adapted with permission from Ref. 21. that analgesics, by decreasing pain, will in-
S44
crease quality of life. Several recent studies
have demonstrated that this intuitive associa-
tion is true.7,24–26 For example, our group mea-
sured changes in Brief Pain Inventory scores in
332 patients with postherpetic neuralgia treated
with a 5% lidocaine patch for 28 days.26 We
found that treatment was associated with de-
creased pain-related interference with quality
of life in all domains examined (Figure 3).26
Rowbotham and colleagues had similar results
in their study of 229 patients with postherpetic
neuralgia who were randomized to receive
gabapentin or placebo for four weeks.25 At the
conclusion of the study, average daily pain scores
were reduced from 6.3 to 4.2 points in the
gabapentin-treated patients, compared with a
change from 6.5 to 6.0 points in the placebo
group (P  0.001). Simultaneously, SF-36 mea-
sures relating to physical functioning, role-
physical, bodily pain, vitality, and mental health
were all significantly better in the gabapentin
group than in the placebo group (P  0.01).
Gabapentin-treated patients also had signifi-
cantly greater improvements than patients in
the placebo group in Profile of Mood States as-
sessments of depression-dejection, anger-hos-
tility, fatigue-inertia, confusion-bewilderment,
and total mood disturbance (P  0.01).25
The link between new treatments for arthri-
tis and patient quality of life has also been eval-
uated. Ehrich and coworkers recently reported
Fig. 3. Effect of analgesic therapy with lidocaine patch 5% on quality-of-life indicators in patients with posther-
petic neuralgia.26 BPI  Brief Pain Inventory. Adapted with permission from Ref. 26.
Katz Vol. 24 No. 1S July 2002
the effect of a cyclooxygenase-2-selective inhib-
itor, rofecoxib, on health-related quality of life
in 672 patients with osteoarthritis of the knee
or hip.7 Patients were randomized to receive
once-daily placebo or rofecoxib at doses of 5,
12.5, 25, or 50 mg, and the SF-36 was adminis-
tered at baseline and at the conclusion of week
6 of treatment.7 All doses of rofecoxib were
significantly superior to placebo in relieving
arthritis pain. This improvement in arthritis
symptoms was found to correlate directly with
improvements in quality of life. Adjusted within-
group mean change scores demonstrated that
all doses of rofecoxib brought about significant
improvement on both the mental and physical
component summary scores (Figure 4),7 as well
as on all eight physical and mental health do-
mains of the SF-36. These improvements were
significantly greater (P  0.05) than those ob-
tained with placebo in all domains except gen-
eral health. A dose-response relationship was
noted, such that the mean changes in quality
of life for the 12.5-, 25-, and 50-mg groups were
of a larger magnitude than that for the 5-mg
group.7 The investigators hypothesized that the
improvement in overall emotional well-being
experienced by the rofecoxib-treated patients
was probably due to increased ability to per-
form and enjoy routine tasks and leisure activi-
ties as a result of relief of osteoarthritis signs
and symptoms.7
Vol. 24 No. 1S July 2002 Pain Management Impact on Quality of Life
Fig. 4. Mean change from baseline in SF-36 compo-
nent summary scores after Week 6 of treatment with
rofecoxib in patients with hip or knee osteoarthritis
(n  672).7 Error bars represent 95% confidence
intervals. PCS  physical component summary;
MCS  mental component summary. Adapted with
permission from Ref. 7.
Quality of Life as a Differentiator:
When Efficacy Is Similar, Can
Quality-of-Life Measures Be Used
to Show the Superiority of One
Medication Over Another?
Quality of life is clearly an important variable
to measure in and of itself. However, another
use for quality-of-life measurement is increas-
ingly being recognized. In any therapeutic
area, drugs within the same pharmacologic
class often have similar efficacy profiles. In
such cases, quality of life and other such indi-
cators have been used successfully to differenti-
ate one agent from another.
At least two studies have made quality-of-life
comparisons in the area of antihypertensive
therapy.6,13 In one, a study conducted by Testa
and researchers comparing quality of life in
379 patients being treated with either captopril
or enalapril, no differences were found be-
tween the two agents in either efficacy or ad-
verse effects.13 Nevertheless, captopril-treated
patients were found to have significantly better
quality-of-life scores than enalapril-treated pa-
tients.13 In the other study, a comparison of
verapamil and nifedipine, no difference in effi-
cacy between the two agents was reported.6
However, a significant distinction was noted in
Physical Symptom Distress Index scores (a
measure of the distress caused by drug-related
adverse effects) in favor of verapamil (P 
S45
0.002), which corresponded to a difference be-
tween the two groups in quality-of-life scores.
The variations in symptom distress scores
tended to predict adherence; there were more
discontinuations in the nifedipine group than
in the verapamil group. The investigators con-
cluded that measurement of symptom distress
is a sensitive technique for evaluating the effect
of antihypertensive therapy on quality of life.6
Typically, adverse events are captured by re-
cording symptoms spontaneously reported by
patients; this is an insensitive method com-
pared with prospectively capturing relevant
side effects and their magnitude.27
A similar relationship between improved ad-
verse effects and quality of life was reported in a
randomized crossover trial in which transdermal
fentanyl was compared with sustained-relief
morphine in patients with chronic noncancer-
related pain.28 In addition to more effective
pain relief, fentanyl-treated patients reported
significantly less trouble with side effects than
those receiving morphine (P  0.001). They
also had significantly higher SF-36 scores in
bodily pain, vitality, social functioning, and
mental health (P  0.005). These results sug-
gest that tolerability may be a critical marker
for quality of life.
Because NSAIDs, including cyclooxygenase-2
inhibitors (coxibs), provide analgesia by non-
opioid mechanisms, they may have an opioid-
sparing effect in patients treated with both
agents. Opioid sparing via use of coxibs can be
expected to decrease common quality-of-life-
impairing adverse effects associated with opio-
ids, including drowsiness, dizziness, constipation,
nausea, and tolerance.29 These studies support
the notion that drug therapies for pain can po-
tentially be differentiated in terms of overall
impact on quality of life, and that the most rel-
evant driver of quality of life in this setting may
be symptom distress due to medicinal side
effects.
Future Directions in Quality-of-Life Research on
Analgesics: Beyond Efficacy
The pain-relief efficacy of the coxibs is ap-
proximately equivalent to that of the nonselec-
tive nonsteroidal anti-inflammatory agents
(NSAIDs). However, the lower risk of gastrointes-
tinal (GI) adverse effects associated with the cox-
ibs compared with traditional NSAIDs is an im-
S46
portant quality-of-life consideration. Both
rofecoxib and celecoxib have been shown, in
very large randomized clinical trials (n  8076
and 8059, respectively) to result in a significantly
lower rate of GI ulcers, blood loss, intolerabil-
ity, and other GI events relative to conven-
tional NSAIDs.29,30 During nine months of fol-
low-up, rofecoxib was found to be associated
with 2.1 confirmed GI events per 100 patient-
years compared with 4.5 events per 100 pa-
tient-years with naproxen (P  0.001).30 Cele-
coxib was associated with an annualized inci-
dence rate of upper GI ulcer complications of
0.76% compared with a rate of 1.45% for ibupro-
fen or diclofenac (P  0.09).29 Rather than de-
pend on surrogate markers, however, future
research should compare coxibs and nonselec-
tive NSAIDs with regard to effect on quality of
life, focusing on symptom distress measure-
ments. It is critical to examine whether overall
quality of life is improved with coxibs com-
pared with nonselective NSAIDs, based on im-
proved tolerability. Similar comparisons of var-
ious opioids or modes of opioid administration
should focus on quality of life.
Conclusion
Uncontrolled pain has a universal and pro-
foundly negative effect on quality of life; no
age group is spared, and no type of pain is ex-
cepted. There is ample evidence, however, that
effective analgesia improves quality of life—a
finding that makes inadequate efforts at pain
control unacceptable. Opioid analgesics have
been shown to improve quality of life for pa-
tients with chronic pain, as have nonnarcotic
agents, including nonselective NSAIDs and
coxibs. In addition to providing a realistic indi-
cator of how a given treatment will affect pa-
tients’ lives, quality-of-life indicators can also
be used to differentiate two agents of the same
pharmacologic class. Future research in the
area of analgesic pharmacotherapy should in-
clude quality of life as a key variable. Analgesic
agents should be compared both within and
between classes, incorporating the use of symp-
tom distress scales, which may be the most sen-
sitive way of discriminating among analgesics
in effects on quality of life. The meaningful-
ness of these differences can be addressed di-
rectly by methods that calibrate symptom dis-
Katz Vol. 24 No. 1S July 2002
tress to stressful real-life events or by using
utility-based methods.
References
1. Gureje O, Von Korff M, Simon GE, et al. Persis-
tent pain and well-being: A World Health Organiza-
tion study in primary care. JAMA 1998;280:147–151.
2. MayoClinic.com. Managing pain: attitude, med-
ication and therapy are keys to control. Mayo Clinic
Web Site. June 21, 2001. Available at: http://www.
mayoclinic.com/invoke.cfm?idHQ01055. Accessed
September 19, 2001.
3. Abbott FV, Fraser MI. Use and abuse of over-
the-counter analgesic agents. J Psychiatry Neurosci
1998;23:13–34.
4. Sternbach RA. Survey of pain in the United States:
The Nuprin Pain Report. Clin J Pain 1986;2:49–53.
5. Rummans TA, Frost M, Suman VJ, et al. Quality of
life and pain in patients with recurrent breast and gy-
necologic cancer. Psychosomatics 1998;39:437–445.
6. Anderson RB, Hollenberg NK, Williams GH.
Physical Symptoms Distress Index: a sensitive tool to
evaluate the impact of pharmacological agents on
quality of life. Arch Intern Med 1999;159:693–700.
7. Ehrich EW, Bolognese JA, Watson DJ et al. Ef-
fect of rofecoxib on measures of health-related qual-
ity of life in patients with osteoarthritis. Am J Manag
Care 2001;7:609–616.
8. Wang XS, Cleeland CS, Mendoza TR, et al. The
effects of pain severity on health-related quality of
life: a study of Chinese cancer patients. Cancer
1999;86:1848–1855.
9. Skevington SM. Investigating the relationship
between pain and discomfort and quality of life, us-
ing the WHOQOL. Pain 1998;76:395–406.
10. Esper P, Redman BG. Supportive care, pain
management, and quality of life in advanced pros-
tate cancer. Urol Clin North Am 1999;26:375–389.
11. Ware JE Jr. The SF-36 health survey. In: Spilker
B, ed. Quality of Life and Pharmacoeconomics. 2nd
ed. Philadelphia, Pa: Lippincott-Raven Publishers;
1996:337–345.
12. Hollenberg NK, Williams GH, Anderson R.
Medical therapy, symptoms, and the distress they
cause: relation to quality of life in patients with an-
gina pectoris and/or hypertension. Arch Intern
Med 2000;160:1477–1483.
13. Testa MA, Anderson RB, Nackley JF, et al. Qual-
ity of life and antihypertensive therapy in men: a
comparison of captopril with enalapril. N Engl J
Med 1993;328:907–913.
14. Joint Commission on Accreditation of Healthcare
Organizations. Assessment of persons with pain.
Pain Assessment and Management: An Organiza-
tional Approach. Oakbrook Terrace, IL: Joint Com-
Vol. 24 No. 1S July 2002 Pain Management Impact on Quality of Life
mission on Accreditation of Healthcare Organiza-
tions; 2000:13–25.
15. Torrance GW. Utility approach to measuring
health-related quality of life. J Chron Dis 1987;40:
593–600.
16. Won A, Lapane K, Gambassi G, et al. Correlates
and management of nonmalignant pain in the nurs-
ing home. J Am Geriatr Soc 1999;47:936–942.
17. Hunfeld JA, Perquin CW, Duivenvoorden HJ, et
al. Chronic pain and its impact on quality of life in
adolescents and their families. J Pediatr Psychol
2001;26:145–153.
18. Haythornthwaite JA, Benrud-Larson LM. Psy-
chological aspects of neuropathic pain. Clin J Pain
2000;16:S101–S105.
19. Becker N, Thomsen AB, Olsen AK, et al. Pain
epidemiology and health related quality of life in
chronic non-malignant pain patients referred to a
Danish multidisciplinary pain center. Pain 1997;73:
393–400.
20. Hill CL, Parsons J, Taylor A, et al. Health related
quality of life in a population sample with arthritis. J
Rheumatol 1999;26:2029–2035.
21. Cleeland CS, Ryan KM. Pain assessment: global
use of the Brief Pain Inventory. Ann Acad Med
1994;23:129–138.
22. Cleeland CS, Gonin R, Hatfield AK, et al. Pain
and its treatment in outpatients with metastatic can-
cer. N Engl J Med 1994;330:592–596.
23. Von Roenn JH, Cleeland CS, Gonin R, et al.
Physician attitudes and practice in cancer pain man-
agement: a survey from the Eastern Cooperative On-
cology Group. Ann Intern Med 1993;119:121–126.
S47
24. McCarberg BH, Barkin RL. Long-acting opioids
for chronic pain: pharmacotherapeutic opportuni-
ties to enhance compliance, quality of life, and anal-
gesia. Am J Ther 2001;8:181–186.
25. Rowbotham M, Harden N, Stacey B, et al. Gaba-
pentin for the treatment of postherpetic neuralgia:
a randomized controlled trial. JAMA 1998;280:
1837–1842.
26. Katz N, Davis MW, Dworkin R. Topical lidocaine
patch produces a significant improvement in mean
pain scores in treated PHN patients: results of a mul-
ticenter open-label trial. In: Posters of the 20th An-
nual Scientific Meeting of the American Pain Soci-
ety; Phoenix, AZ; April 19–22, 2001; Poster 741.
27. Anderson RB, Testa MA. Symptom distress
checklists as a component of quality of life measure-
ment: comparing prompted reports by patient and
physician with concurrent adverse event reports via
the physician. Drug Inf J 1994;28:89–114.
28. Allan L, Milligan K. Randomized, crossover and
open-label trials demonstrate the efficacy of trans-
dermal fentanyl (Duragesic®) for the treatment of
chronic non-cancer pain. European League Against
Rheumatism; Prague, Czech Republic; June 13–16,
2001. Abstract SAT0127.
29. Silverstein FE, Faich G, Goldstein JL, et al. Gas-
trointestinal toxicity with celecoxib vs nonsteroidal
anti-inflammatory drugs for osteoarthritis and rheu-
matoid arthritis—the CLASS study: A randomized
controlled trial. JAMA 2000;284:1247–1255.
30. Bombardier C, Laine L, Reicin A, et al. Compar-
ison of upper gastrointestinal toxicity of rofecoxib
and naproxen in patients with rheumatoid arthritis.
N Engl J Med 2000;343:1520–1528.