Critical Review of Arthritis Self-Management
JULIE J. KEYSOR,
BRENDA M. DEVELLIS,
GORDON H. DEFRIESE,
ROBERT F. DEVELLIS,
JOANNE M. JORDAN,
THOMAS R. KONRAD,
ELIZABETH J. MUTRAN,
AND LEIGH F. CALLAHAN
Notions of self care and self management of common, but
serious, health conditions were once viewed with skepti-
cism by clinicians. However, over time, enlightened health
care professionals have come to embrace the idea of in-
formed, participative collaboration with patients as a key
strategy for enhancing the effectiveness of clinical diag-
nostic and therapeutic interventions (1). Although these
ideas are being absorbed into mainstream contemporary
medicine and health care, our understanding of the range
of self-initiated actions consumers use and the factors re-
lated to use of these actions is inadequate.
Sobel (2) suggests that self care is the “hidden” health
care system and that self care, rather than primary, sec-
ondary, or tertiary care, comprises the majority of health
care. Others suggest that self-management activities are
determinants of well being, potentially intervening be-
tween health status and health outcomes (3–5). If these
assertions are true, we must have a clear understanding of
the factors related to and outcomes associated with strat-
egy use so people with chronic conditions can be guided
in effective self-management practices.
The terms self management and self care, often used
interchangeably among lay and professional persons, are
broadly deﬁned as the activities people engage in to pro-
mote health and/or manage chronic conditions (6–9).
These activities are usually self initiated and often under-
taken with little to no supervision from health care pro-
fessionals. For this article we use both phrases (self care
and self management) to refer to the wide range of activi-
ties people use to promote personal health and to detect,
prevent, and treat common health problems.
The use of self-initiated actions to manage arthritis con-
ditions is well recognized in the scientiﬁc literature and
among health care providers. Applying the above deﬁni-
tion of self management to arthritis, a variety of strategies
could be used to manage the symptoms and consequences
of the disease, such as taking medication, exercising, wear-
ing splints or braces, taking herbs or supplements, or seek-
ing care from alternative care providers. Two review arti-
cles by Ernst (10,11) show that use of complementary and
alternative strategies is common but variable among peo-
ple with rheumatic disease, with most people using at
least 1 type of complementary therapy (e.g., special diets,
jewelry, vitamins, herbs, prayer, relaxation, or massage).
Little is known, however, about the broad range of strate-
gies that could be used to manage arthritis, the degree to
which people with arthritis use various strategies, and
factors related to use of speciﬁc strategies. The purpose of
this article is to address these gaps in the scientiﬁc litera-
The Medline database was used to identify studies that
met the following criteria: 1) study participants were from
a rheumatology clinical practice or were community-
dwelling adults who reported arthritis, rheumatoid arthri-
tis (RA), osteoarthritis, or chronic musculoskeletal joint
problems; 2) speciﬁc self-management strategy use or cat-
egories were assessed (e.g., heat, lotions, exercise, rest,
dietary practices, herbs and supplements, or alternative
care provider); and 3) frequency of use and/or sociodemo-
graphic or clinical disease-related correlates of strategy use
were reported. Sociodemographic factors included age,
sex, marital status, educational attainment, ethnicity, and
income or social status; disease-related factors included
functional status, pain, disease severity (e.g., joint involve-
ment or number of symptoms), disease duration, and co-
Supported by the National Institute on Arthritis and Mus-
culoskeletal Disease Grant 5-P60-AR30701-18, a National
Institute on Arthritis and Musculoskeletal Disease Predoc-
toral Grant, an Arthritis Foundation Dissertation Award,
University of North Carolina Center on Aging Pilot Grant,
and Postdoctoral Fellowship Award from NIDRR, US De-
partment of Education Grant #H133B990005.
Julie J. Keysor, PhD, PT: Boston University, Boston, Mas
Brenda M. DeVellis, PhD, Gordon H. DeFriese,
PhD, Robert F. DeVellis, PhD, Joanne M. Jordan, MD, MPH,
Thomas R. Konrad, PhD, Elizabeth J. Mutran, Leigh F. Cal-
lahan, PhD: University of North Carolina at Chapel Hill.
Address correspondence to Julie J. Keysor, PhD, PT, De-
partment of Physical Therapy, Sargent College of Health
and Rehabilitation Sciences, 635 Commonwealth Avenue,
Room 521, Boston University, Boston, MA 02215. E-mail:
Submitted for publication July 19, 2001; accepted in re-
vised form January 6, 2003.
Arthritis & Rheumatism (Arthritis Care & Research)
Vol. 49, No. 5, October 15, 2003, pp 724–731
© 2003, American College of Rheumatology
The search was limited to articles published in English
between 1980 and January 2002. The terms, “self-care” or
“self-management” were combined into 1 search strategy
and the terms “arthritis,” “rheumatoid arthritis,” “osteoar-
thritis,” or “musculoskeletal” were combined into another
search strategy. The 2 search strategies then were com-
bined to capture self-management practices among per-
sons with chronic arthritis and musculoskeletal problems.
Additional searches were performed for the terms “com-
plementary or alternative medicine,” “joint protection,” or
“adaptive equipment” in combination with the arthritis
and musculoskeletal terms speciﬁed above. The bibliogra-
phies of articles thus identiﬁed were scanned for further
relevant publications. Studies examining self-management
strategy use speciﬁcally for ﬁbromyalgia and systemic lu-
pus erythematosus were excluded because these condi-
tions usually do not manifest with chronic joint problems.
Study design, ascertainment and frequency of strategy
use, and sociodemographic and clinical disease-related
correlates of strategy use were assessed for each study that
met inclusion criteria. Frequency of strategy use was re-
ported as “recent use” (i.e., strategy used within the past 6
months) and “ever use” (i.e., use within the past year or in
previous years). Demographic and disease-related corre-
lates of strategy use were summarized by reporting signif-
icant associations (P ⱕ 0.05).
To further structure our review, frameworks described
by DeFriese et al (7) and Norburn et al (12) were adopted.
Seven broad categories representing different types of ar-
thritis self-management strategies were identiﬁed: 1) topi-
cal treatments (e.g., lotions, ointments, oils, heat, and
cold), 2) activity-related strategies (e.g., exercise, activity
modiﬁcation, assistive device use, joint protection behav-
iors, or environmental changes), 3) dietary practices (e.g.,
eating or avoiding special foods, taking supplements or
herbs, drinking or avoiding alcohol), 4) mind-body prac-
tices (e.g., relaxation, prayer, talking with others, positive
thinking, and meditation), 5) alternative and complemen-
tary practitioners (ACPs; e.g., chiropractors, acupunctur-
ists, or massage therapists), 6) home remedies (e.g., house-
hold oils such as WD-40, motor oil, snake venom, bee
sting, or special jewelry), and 7) conventional medical care
(e.g., prescribed and over-the-counter medications or use
of health care services).
Nineteen studies met inclusion criteria (Table 1), all were
cross-sectional (4,13–31). All but 1 study used self-report
surveys to ascertain strategy use; 1 study used self-report
diaries and surveys to ascertain strategy use (31).
Frequency of strategy use. Special dietary practices and
use of herbs and supplements were most frequently exam-
ined, with approximately 60% of the studies assessing
these strategies (see Table 2). Lotions, heat, exercise, rest,
relaxation, and special jewelry were examined in at least
40% of the studies; use of splints and joint protection
Table 1. Studies included in the review*
Study Sample description
Gray (13) Australian community-dwelling adults with self-reported arthritis (n ⫽ 76)
Kronenfeld and Wasner (14) Consecutive in-patient and out-patient rheumatology clinic patients, most with
RA or OA (n ⫽ 98)
Cassidy et al (15) Randomly selected patients with physician-diagnosed RA (n ⫽ 100)
Struthers et al (16) Patients with physician diagnosed RA (n ⫽ 199)
Coulton et al (17) Community-living elders with self-reported joint problems (n ⫽ 317)
Davis et al (18) Patients with physician diagnosed RA or OA and 3 months self-reported pain
(n ⫽ 82)
Cronan et al (19) Community-dwelling adults with self-reported OA (n ⫽ 382)
Hampson et al (20, 21) Patients with physician-diagnosed OA for at least 1 year (n ⫽ 61)
Boisset and Fitzcharles (22) Consecutive rheumatology outpatients (n ⫽ 235)
Arcury et al (23) Rural community-living adults with physician-diagnosed arthritis (n ⫽ 219)
Katz (4) Consecutive patients with physician-diagnosed RA from randomly selected
rheumatologic practices (n ⫽ 471)
Hammond (24) Patients with physician-diagnosed RA (n ⫽ 41)
Ramos-Remus et al (25) Consecutive patients from rheumatology clinical practices in Mexico (n ⫽ 250)
Rao et al (26) Consecutive patients from rheumatology practices (n ⫽ 232)
Anderson et al (27) Random sample of patients seen at least once during 1997 in rheumatology and
geriatric practices (n ⫽ 176)
Aceves-Avila et al (28) Consecutive patients from rheumatology practices in Mexico (n ⫽ 247)
Kaboli et al (29) Random sample of community-dwelling adults 65 years old or older with self-
report physician-diagnosed arthritis (n ⫽ 480)
McDonald-Miszczak et al (30) Community-dwelling adults 50 years or older who self report arthritis
randomly selected via random-digit dialing (n ⫽ 377)
Ramsey et al (31) Community-dwelling adults 55 years or older with physician-diagnosed OA
from a randomized clinical trial of water aerobics (n ⫽ 122)
*RA⫽ rheumatoid arthritis; OA ⫽ osteoarthritis.
Critical Review of Arthritis Self-Management Use 725
strategies, massage, prayer, coping, chiropractic care, and
any ACP were examined in 25% of the studies. The least
frequently examined strategies were activity modiﬁcation,
the application of cold or ice, acupuncture, some of the stress
and mind-body strategies, and some of the home remedies,
with less than 25% of the studies examining these strategies.
Table 2. Frequency of self-management strategies used among people with arthritis
Ever use, % (reference; speciﬁc
strategy) Recent use, % (reference, speciﬁc strategy)
Heat (applied or not speciﬁed) 49 (4); 66 (43); 70 (18); 71 (23) 38 (17); 40 (20); 47 (23)
Hot shower, bath, spa 4 (29); 63 (4); 71 (18) 29 (19); 34 (17)
Cold 22 (18) 4 (17)
Lotions, oils, and ointments 21 (26); 26 (25); 36 (23; lotions); 38 (29);
66 (25); 81 (14,23; ointments)
11 (26); 18 (23; lotions); 30 (19); 52 (23;
Exercise 56 (24); 74 (4); 75 (23); 78 (18) 33 (19); 63 (23); 65 (20; range of motion);
68 (30); 74 (17); 92 (20; gentle activity)
Rest 62 (23); 71 (43); 89 (4); 90 (18) 29 (30); 33 (19); 59 (23); 60 (17); 65 (20)
Modify activity 22 (24); 62 (4) 15 (17)
Splints/braces, joint protection 29 (18); 38 (4); 44 (24) 3 (17); 5 (20; splints); 35 (20; joint
Special diets (avoid or eat special
7 (29; special diet); 13 (22); 20 (26;
special diet); 25 (26; vinegar
drinks); 28 (4; avoid foods); 33 (16;
special diet); 34 (18; avoid foods);
46 (15; cod liver oil); 49 (23; special
foods); 54 (14; speciﬁc diets); 60
(15; dietary modiﬁcations)
2 (17; special foods); 5 (17; avoid foods);
12 (26); 26 (23); 32 (30)
Supplements or herbs 4 (25; shark cartilage); 4 (29; herbs); 16
(26; minerals or megavitamins); 18
(25; garlic); 22 (26; supplements); 25
(14); 29 (26; herbs); 38 (4;
supplements, vitamins, and special
foods); 44 (25; vitamins); 74 (25;
1 (27; shark cartilage); 4 (17); 5 (27; grape
seed); 5 (27; cranberry or ginkgo); 5 (27;
glucosamine); 9 (26; minerals and
megavitamins); 11 (26; supplements); 12
(26; herbs); 18 (17; herbs); 27 (19;
vitamins); 31 (20); 47 (27; vitamins); 51
Alcohol 15 (23; whiskey); 1 (17); 6 (30; reduce
5 (23; whiskey)
Stress and mind-body
Prayer/spiritual 5 (26); 37 (23; church services); 38 (15;
religious services); 39 (22; prayer,
meditation, relaxation); 42 (29); 92
2 (17); 2 (26); 31 (23; religious services); 44
(19); 92 (23; prayer)
Relaxation or meditation 5 (29); 10 (29); 21 (4); 32 (43); 38 (18) 1 (17); 33 (19); 40 (21)
Diverting attention 3 (23; special trips); 12 (14; special
trips); 72 (4); 79 (18)
1 (23; special trips); 10 (17)
Social support 60 (18); 66 (4) 28 (30)
Stress control 31 (4); 44 (18) 43 (30)
Support or self-help groups 6 (4; support group); 8 (4; self help
group); 15 (18)
Alternative and complementary
ACP* 13 (22); 25 (13); 28 (29); 28 (15); 36 (14)
Chiropractor 20 (25); 26 (29); 31 (26) 7 (26)
Acupuncture 3 (29); 7 (26); 7 (15); 11 (16); 14 (25) 1 (26)
Massage 17 (25); 26 (4); 46 (18) 9 (17); 1 (19); 35 (21)
Homeopathist ⬍1 (29); 1 (15)
Jewelry ⬍1 (22); 5 (29); 12 (25); 28 (23); 29 (26);
37 (15); 38 (14); 38 (16)
4 (23); 6 (26); 10 (17)
Household motor oils 5 (25); 17 (23; WD-40); 18 (23;
3 (23; WD-40); 7 (23; turpentine); 9 (17)
Snake venom/bee sting 3 (25); 7 (23) 2 (23)
* ACP ⫽ Alternative and complementary practitioners.
726 Keysor et al
Frequencies of use between and within the “recent”
versus “ever” timeframes were varied. In most instances,
people were less likely to have used a strategy when strat-
egy use was assessed within the past 6 months than when
assessed within the past year or longer. This conclusion is
consistent with those determined by others (23,26), that
people with arthritis may try various strategies but are less
likely to continue them.
Heat, rest, and exercise were used by the majority of
participants in the studies, although frequency of use and
ascertainment of strategy varied. In Katz’s study (4), 49%
of the respondents applied heat to parts of their body and
63% used a heated pool, tub, or shower. Arcury et al (23)
reported 71% of the participants used heat (unspeciﬁed)
and Davis et al (18) found that 70% of participants applied
heat to painful areas. Ascertainment of exercise also dif-
fered among the studies; prevalence of use estimates were
varied. In the studies reporting the highest rates of exer-
cise, participants were asked about their use of “exercise
as tolerated” (18), “exercise” (23), and “gentle activity”
(20,21). Cronan et al (19), reporting the lowest frequency of
exercise, asked participants about use of “exercise or
swimming not prescribed.”
Findings related to use of rest and lotions and ointments
were similar to heat and exercise: frequency estimates
varied and so did ascertainment. In the study that reported
the highest use of topical applications, participants were
from a southern community and were speciﬁcally asked
about the use of “ointments-liniments” (23). Lower esti-
mates were found when use of lotions, creams, or oils was
assessed (23,25,26). The highest rates of the use of rest as
a self-management strategy were ascertained by questions
relating to “rest,” “resting,” or “used rest” (4,18,23); the
lowest estimates of rest were reported when the use of rest
was assessed in a more restrictive manner, such as “bed
rest” (19) or “resting more” (30).
Use of activity modiﬁcation practices and joint protec-
tion strategies was examined in only 5 studies; results
indicated that these strategies were not commonly used.
One study reported that 35% of the participants used joint
protection principles within the past 6 months (20), but
other studies reported that fewer than 15% used joint
protection activities or braces or splints (17,20).
Dietary practices were examined frequently and as-
sessed in many different ways (see Table 2). Generally,
about one-third of the participants used these strategies.
Some studies examined the use of speciﬁc herbs or sup-
plements, such as shark cartilage or glucosamine (25,27),
garlic (25), grape seed (27), or ginkgo (27), although most
assessed the general use of herbs or supplements
(4,14,17,19,25–27,29). In addition, 1 assessment included
special foods with supplements and herbs (4).
Spiritual and relaxation practices (including prayer and
meditation) were fairly common, although estimates
ranged from 5% (26,29) to as high as 92% (23). Spiritual
practices included attending church or religious services
and engaging in prayer. In 2 studies (15,23), approximately
40% of the participants reported using church services as
a self-management practice; whereas in one study, 92% of
the participants used prayer (23).
ACP use was usually assessed within the past year or
longer; only a few studies assessed current use of ACP.
Although estimates varied, approximately 25% of the par-
ticipants used some type of ACP. Chiropractors and mas-
sage therapists were the ACPs most frequently seen; acu-
puncturists and homeopathists were least frequently seen.
Of the home remedies, special jewelry was the most
frequently used strategy; however, in 3 studies, fewer than
10% of the participants used special jewelry for their
arthritis within the past 6 months. Studies showed that
less than 20% of the participants had tried household oils,
such as WD-40, turpentine, and other motor oils. Estimates
of use within the past month were lower: 3% used WD-40
and 7% used turpentine (23).
The following strategies were reported only in 1 study
and were, therefore, not listed in Table 2: assistive devices
(24), environmental changes (30), positive thinking (23),
natural healing (26), spiritual healing (26), hypnotism (29),
energy healer (29), reﬂexology (25), dimethyl sulfoxide
(17), quit or decrease tobacco use (30), mothballs (23),
special clothing (23), orthopedic shoes (17), biofeedback
(18), and transcutaneous electrical nerve stimulation (18).
In summary, much of the literature focused on use of
topical treatments, diet, alternative care strategies, exer-
cise, and rest. Use of activity modiﬁcation practices and
joint protection strategies was not examined often, despite
the potential beneﬁt and long-term importance of these
strategies on health outcomes of people with chronic joint
problems. Furthermore, people do not seem to be using
these strategies to a great degree.
Correlates of strategy use. In Table 3, the sociodemo-
graphic and disease-related correlates of strategy use are
summarized. Because few studies reported these correla-
tions, only patterns and trends are reported. Functional
status was the only disease-related factor examined often
enough to identify patterns and trends. Correlations be-
tween disease severity, disease duration, comorbidity, and
pain with self-management strategies were not reported
because the relations were not examined frequently
enough to identify any patterns or trends.
Dietary practices, alternative care, or home remedies,
generally, were not related to sociodemographic factors,
although there were a few exceptions. Katz (4) reported
that people with RA who had more educational attainment
were more likely to avoid certain foods and use massage
compared with people with fewer years of education, even
after adjusting for age, sex, race, marital status, income,
comorbidity, disease duration, arthritis, number of painful
joints, presence of severe fatigue, and functional status.
Most studies examining these relationships, however,
found no differences in strategy use. This discrepancy may
be due to variability in the way educational attainment
was operationalized. Katz categorized education into 6
groups: 0 – 8 years of education, 9 –11 years, 12 years,
13–15 years, 16 years, or 17 or more years of education, but
other researchers used educational attainment as a contin-
uous variable. Sex differences were found with the use of
alcohol: men were more likely to use whiskey than women
(23). Ethnic differences were also found: African Ameri-
cans were more likely to use turpentine than whites (23).
Critical Review of Arthritis Self-Management Use 727
Table 3. Associations of demographic and disease-related factors with self-management strategy use*
strategies Dietary practices
Stress and mind-body
practices Alternative care providers
Age Heated pool (Y; 18) Bracing (Y; 18) Eat or avoid foods Stress control (Y; 18) Massage (Y; 18) Turpentine (23)
Apply heat (18,23) Exercise (18,23) (14,18,23) Support (18) Alternative care providers (14,29,31) WD-40 (23)
Apply cold (18) Rest (18,23) Vitamins (14) Distraction (18) Jewelry (14,23)
Lotion (14,23) Whiskey (23) Relaxation (18)
Positive thinking (23)
Sex Heat (M; 23) Rest (F; 23) Whiskey (M; 23) Prayer (F; 23) Alternative care providers (14,29,31) Turpentine (23)
Lotions (F; 23) Eat or avoid foods (14,23) Positive thinking (23) WD-40 (23)
Lotions (14) Vitamins (14) Trips (14) Jewelry (14,23)
Race Heat (W; 23) Rest (W; 23) Eat or avoid foods (14,23) Prayer (AA; 23) Alternative care providers (14) Turpentine (23)
Lotions (AA; 23) Rest (4,17) Vitamins (14) Positive thinking (W; 23) WD-40 (23)
Lotions (14) Whiskey (23) Trips (14) Jewelry (14,23)
Education Apply heat (4) Exercise (4) Eat or avoid foods (4) Support (4) Massage (4) Jewelry (14)
Heated pool (4) Rest (4) Eat or avoid foods (14) Stress control (4) Alternative care providers (14,22,29,31)
Heat (22) Joint protection (4) Vitamins (14) Relaxation (4)
Lotions (14) Change routine (4) Supplements (4)
Socioeconomic Heat (Lower; 22) Eat or avoid foods (14) General stress (22) Alternative care providers (14,22,29,31) Jewelry (14)
status Lotions (14) Vitamins (14,22) Trips (14)
Functional Heat (23) Rest (23) Eat or avoid foods (23) General stress (14,23) Alternative care providers (14,31) WD-40 (23)
status Lotions (14,23) Whiskey (23) Jewelry (23)
Eat or avoid foods (14) Turpentine (23)
Vitamins (14) Jewelry (14)
* Items in bold are statistically signiﬁcant. Numbers in parentheses indicate study references. Y ⫽ youth; M ⫽ male; F ⫽ female; W ⫽ white; AA ⫽ African American.
728 Keysor et al
Regarding physical functioning, ﬁndings were inconsis-
tent. One study showed that people with worse physical
functioning were more likely to eat special foods, drink
whiskey, use WD-40, or wear special jewelry (23); how-
ever, another study found no relationship between func-
tional status and use of similar dietary strategies or home
Regarding topical treatments, age and functional status,
generally, were not related to use of heat or lotions. In 1
study, however, which dichotomized people as age 50
years of age or younger and over 50, the younger aged
adults were more likely to use a heated pool than the older
adults. Findings for sex, race, education, and income were
mixed. Men, whites, and those with more income and
more years of educational attainment were more likely to
use heat (4,22,23); women and African Americans were
more likely to use lotions (23). On the other hand, 2 stud-
ies showed no differences in use of topical treatments by
sex, race, education, or income (14,22).
Activity-related strategies were examined in only a few
of the studies, thus, our interpretations are limited. Gen-
erally, age and education were not related to exercise, rest,
or activity modiﬁcation practices, although Davis et al (18)
found that people who were younger were more likely to
use braces or splints. Findings with sex and race were
inconsistent. Women and people who were white were
more likely to rest than were men or African Americans
(23). On the other hand, no association was found in 2
other studies (4,17).
Stress and mind-body practices were not related to age,
socioeconomic status, or functional status, except in 1
study (18) that showed younger adults were more likely to
engage in stress control practices. However, there were
some associations between stress and mind-body practices
and sex, race, and educational attainment, though only a
few studies examined these relations. Katz (4) found that
people with more educational attainment were more likely
to attend support groups, use stress control strategies, and
use relaxation strategies than were people with fewer years
of education. Women and African Americans were more
likely to use prayer and church services compared with
men and people who were white; and whites were more
likely to use positive-thinking strategies (23).
In sum, we found no conclusive evidence relating socio-
demographic and disease-related factors to self-manage-
ment strategy use. Some positive associations were found
between sex, race, and functional status and use of topical
treatments, activity-related strategies, stress and mind-
body practices, and folklore strategies. Age, education, and
income were generally not related to strategy use. These
ﬁndings, at least in part, are supported by studies that
show sociodemographic factors contribute only a small
amount of variance to self-management strategy use
The primary aims of this study were 1) to identify the
range of strategies that people with arthritis or chronic
joint problems used to manage their condition, and 2) to
examine the sociodemographic and disease-related factors
associated with use of these self-management strategies.
Ascertainment of strategy use was remarkably varied
among the studies, with no 2 studies examining arthritis
self-management strategy use in the same manner. The
range of self-initiated strategies used to manage arthritis
symptoms and consequences varied and included conven-
tional and unconventional activities. No conclusive evi-
dence relating sociodemographic and disease-related fac-
tors to self-management strategy use was found.
These ﬁndings may be explained by 2 factors: 1) a lack of
clearly deﬁned and consistent conceptual framework, and
2) methodologic limitations in studies. None of the studies
used the same conceptual framework to identify and de-
ﬁne strategies or categories of use; only 2 used a theoretical
framework to guide the examination of correlates of strat-
egy use (17,30). A clear conceptual framework for arthritis
self management—i.e., identify and deﬁne relevant cate-
gories or domains—would enable researchers to more
easily compare study ﬁndings and assess arthritis self-
management behaviors. Without a clear, conceptual frame-
work, assessment of the broad range of self-management
strategy use is unstructured, potentially misleading, and
difﬁcult to duplicate. The conceptual framework of De-
Friese and colleagues (7,34) and the 1 proposed in this
article could be useful in identifying the broad range of
arthritis self-management strategies and deﬁning catego-
In addition to using conceptual frameworks to assess
self-management strategy use, conceptual frameworks
could guide research examining the determinants of self-
management behaviors. The Behavioral Model of Health
Service Utilization (35–37) and the Health Belief Model
(38) were used in 2 studies of this review to guide the
examination of arthritis self-management strategy use. Us-
ing Andersen’s model as a framework, Coulton et al (17)
showed that pain and number of chronic conditions (iden-
tiﬁed as need factors in Andersen’s model) were the stron-
gest correlates of self-management strategy use, after ad-
justing for demographic factors. Similar ﬁndings were
reported in 2 other studies of self-management practices
among elderly persons (32,33), suggesting that disease-
related factors explained a larger part of the variance in
strategy use than sociodemographic factors.
Likewise, using the Health Belief Model as a framework,
McDonald-Miszczak et al (30) found that among 377 com-
munity-dwelling adults 50 years of age or older, perceived
seriousness and self-rated health status were associated
with self-care behaviors after adjusting for sociodemo-
graphic factors, disease-related factors, and health care
utilization. Peoples’ perceptions, therefore, may inﬂuence
the strategies they use to manage their conditions. The
self-regulation model (SRM) of Leventhal et al (39 – 41)
provides such a framework for understanding how people
interpret health threats and how these interpretations in-
ﬂuence self-management activities. In the SRM, cognitive
representations of and emotional responses to disease
states are viewed as proximal determinants of the actions
people take to enhance their health and to prevent, treat,
and rehabilitate from illness.
Despite a substantial amount of research on illness rep-
Critical Review of Arthritis Self-Management Use 729
resentations among people with arthritis, few studies have
examined the relation of these factors to the use of speciﬁc
strategies people use to cope with arthritis conditions.
Using this framework, if people with arthritis interpret
pain and fatigue as problematic, they may take actions to
decrease these symptoms by avoiding activity, resting, or
taking herbs or supplements; however, if someone inter-
prets these symptoms as a consequence of lack of exercise,
he or she might start or increase exercise activity.
Methodologic differences also may explain the inconsis-
tencies found in the articles. Some of the inconsistencies
could be due to how self-management strategy use was
ascertained, whether confounding factors were included
in modeling strategies, and differences in sample charac-
teristics. As shown in the Results section, ascertainment of
strategy use varied markedly. In some instances when
strategies were very general, such as “exercised” or “rest-
ed,” estimates of use were high. On the other hand, when
very speciﬁc strategies were assessed, such as “bed rest” or
“cod liver oil,” use estimates were lower. Differences in
ascertainment of strategy use may explain a substantial
amount of the variance in prevalence estimates. In addi-
tion, because many of the strategies were assessed glo-
bally, exercise for example, we do not know whether peo-
ple using these strategies engaged in the appropriate type
of exercise, safely performed the exercise (e.g., minimizing
abnormal joint forces), or exercised long enough to have
any health beneﬁt. A clear conceptual framework of arthri-
tis self-management categories or domains should enable
researchers to develop better assessment instruments.
Some of the discrepancies in the correlates of strategy
use may be due to differences in analytic approaches. For
example, several differences were found between whites
and African Americans in relation to use of medications,
rest, spiritual practices, and household oils in a large rep-
resentative sample from a rural southern community; how-
ever, socioeconomic factors were not adjusted in these
models, which may in part explain the ethnic differences.
In the study reported by Coulton et al (17), several ethnic
differences were found in univariate models; however,
when models were adjusted for socioeconomic factors and
physical disability, the associations between race and self-
management strategy use were no longer signiﬁcant.
Furthermore, because all studies in this review were
cross-sectional, we are unable to ascertain whether such
factors as functional status or pain inﬂuenced strategy
choice and use. In a longitudinal study on self-manage-
ment among community-dwelling elders, elderly persons
who reported consistent joint pain and stiffness (arthritis-
related symptoms) at 4 time points over 27 months were
more likely to use self-management strategies than were
persons who reported inconsistent symptoms (joint pain
or stiffness 3 or fewer times over a 27-month interval) (42).
Symptom recurrence over time, therefore, seemed to be
related to self-management practices, potentially support-
ing the SRM approach of Leventhal et al to illness inter-
pretation and self-management behaviors.
There are limitations to our approach to this article.
First, it was difﬁcult to classify some ﬁndings using our
classiﬁcation scheme, and as a result we may have ob-
scured some ﬁndings. Although we tried to establish clear
and reliable guidelines for interpreting the data, other
researchers may not agree with our classiﬁcation frame-
work. Nonetheless, a wide range of strategies was used,
many inconsistencies were present, ascertainment of strat-
egy use varied, and few studies reported speciﬁc relations
between sociodemographic and disease-related factors and
use of our self-management categories. Second, we found
only 19 articles meeting our criteria. The ﬁndings are,
therefore, limited to trends and patterns and should be
interpreted with caution. There is great interest among
health professionals regarding the strategies people use to
manage arthritis; however, those who have studied these
behaviors among persons with arthritis have approached
these issues in an ad hoc manner, using highly idiosyn-
cratic measures resulting in primarily descriptive informa-
tion and many inconsistent ﬁndings. We do not know
whether use of various self-management strategies is in-
ﬂuenced by episodes of acute pain or discomfort; func-
tional loss; cultural factors; or intentional, preventive, pre-
cautionary measures. Lastly, we know very little about the
levels of knowledge, skills, or experience with these self-
management strategies that are required to practice them
with any likelihood of realizing a therapeutic effect.
We thank Dr. Shannon Currey for her contributions to the
North Carolina Musculoskeletal Health Project and Dr.
Alan Jette for his comments on this article.
1. Von Korff M, Gruman J, Schaefer J, Curry SJ, Wagner EH.
Collaborative management of chronic illness. Ann Intern Med
2. Sobel DS. Rethinking medicine: improving health outcomes
with cost-effective psychosocial interventions. Psychosom
Med 1995;57:234 –44.
3. Dean K. Lay care in illness. Soc Sci Med 1986;22:275– 84.
4. Katz PP. Education and self-care activities among persons
with rheumatoid arthritis. Soc Sci Med 1998;46:1057–66.
5. Callahan LF, Pincus T. Education, self-care, and outcomes of
rheumatic diseases: further challenges to the “biomedical
model” paradigm. Arthritis Care Res 1997;10:283– 8.
6. Haug MR, Akiyama H, Tryban G, Sonoda K, Wykle M. Self-
care: Japan and the US compared. Soc Sci Med 1991;33:1011–
7. DeFriese GH, Konrad TR, Woomert A, Norburn JEK, Bernard
S. Self-care and quality of life in old age. In: Abeles RP, Gift
HC, Ory MG, editors. Aging and quality of life. New York:
Springer; 1994. p. 99–117.
8. Ory MG, DeFriese GH, Duncker AP. Introduction: the nature,
extent, and modiﬁability of self-care behaviors in later life. In:
Ory MG, DeFriese GH, editors. Self-care in later life. New
York: Springer; 1998. p. xv–xxvi.
9. Holman HR, Lorig KR. Patient education: essential to good
health care for patients with chronic arthritis. Arthritis
10. Ernst E. Usage of complementary therapies in rheumatology:
a systematic review. Clin Rheumatol 1998;17:301–5.
11. Ernst E. Complementary and alternative medicine in rheuma-
tology. Baillieres Best Pract Res Clin Rheumatol 2000;14:731–
12. Norburn JE, Bernard SL, Konrad TR, Norburn JE, Bernard SL,
Konrad TR, Woomert A, DeFriese GH, Kalsbeek WD, et al.
Self-care and assistance from others in coping with functional
730 Keysor et al
status limitations among a national sample of older adults. J
Gerontol B Psychol Sci Soc Sci 1995;50:S101–9.
13. Gray D. The treatment strategies of arthritis sufferers. Soc Sci
14. Kronenfeld JJ, Wasner C. The use of unorthodox therapies and
marginal practitioners. Soc Sci Med 1982;16:1119 –25.
15. Cassidy M, Jacobs A, Bresnihan B. The use of unproven rem-
edies for rheumatoid arthritis in Ireland. Ir Med J 1983;76:
16. Struthers GR, Scott DL, Scott DGI. The use of ‘alternative
treatments’ by patients with rheumatoid arthritis. Rheumatol
17. Coulton CJ, Milligan S, Chow J, Haug M. Ethnicity, self-care,
and use of medical care among the elderly with joint symp-
toms. Arthritis Care Res 1990;3:19 –28.
18. Davis GC, Cortez C, Rubin BR. Pain management in the older
adult with rheumatoid arthritis or osteoarthritis. Arthritis
Care Res 1990;3:127–31.
19. Cronan TA, Kaplan RM, Kozin F. Factors affecting unpre-
scribed remedy use among people with self-reported arthritis.
Arthritis Care Res 1993;6:149–55.
20. Hampson SE, Glasgow RE, Zeiss AM, Birskovich SF, Foster L,
Lines A. Self-management of osteoarthritis. Arthritis Care Res
21. Hampson SE, Glasgow RE, Zeiss AM. Personal models of
osteoarthritis and their relation to self-management activities
and quality of life. J Behav Med 1994;17:143–58.
22. Boisset M, Fitzcharles MA. Alternative medicine use by rheu-
matology patients in a universal health care setting. J Rheu-
matol 1994;21:148 –52.
23. Arcury TA, Bernard SL, Jordan JM, Cook HL. Gender and
ethnic differences in alternative and conventional arthritis
remedy use among community-dwelling rural adults with
arthritis. Arthritis Care Res 1996;9:384 –90.
24. Hammond A. The use of self-management strategies by peo-
ple with rheumatoid arthritis. Clin Rehabil 1998;12:81–7.
25. Ramos-Remus C, Gamez-Nava JI, Gonzalez-Lopez L, Ramos-
Remus C, Gamez-Nava JI, Gonzalez-Lopez L, Skeith KJ, Perla-
Navarro AV, Galvan-Villegas F, et al. Use of alternative ther-
apies by patients with rheumatic disease in Guadalajara,
Mexico: prevalence, beliefs, and expectations. Arthritis Care
Res 1998;11:411– 8.
26. Rao JK, Mihaliak K, Kroenke K, Bradley J, Tierney WM, Wein-
berger M. Use of complementary therapies for arthritis among
patients of rheumatologists. Ann Intern Med 1999;131:409 –
27. Anderson DL, Shane-McWhorter L, Crouch BI, Andersen SJ.
Prevalence and patterns of alternative medication use in a
university hospital outpatient clinic serving rheumatology
and geriatric patients. Pharmacotherapy 2000;20:958 –66.
28. Aceves-Avila FJ, Medina F, Fraga A. Herbal therapies in
rheumatology: the persistence of ancient medical practices.
Clin Exp Rheumatol 2001;19:177–83.
29. Kaboli PJ, Doebbeling BN, Saag KG, Rosenthal GE. Use of
complementary and alternative medicine by older patients
with arthritis: a population-based study. Arthritis Rheum (Ar-
thritis Care Res) 2001;45:398–403.
30. McDonald-Miszczak L, Wister AV, Gutman GM. Self-care
among older adults: an analysis of the objective and subjec-
tive illness contexts. J Aging Health 2001;13:120–45.
31. Ramsey SD, Spencer AC, Topolski TD, Belza B, Patrick DL.
Use of alternative therapies by older adults with osteoarthri-
tis. Arthritis Rheum (Arthritis Care Res) 2001;45:222–7.
32. De Klerk M, Huijsman R, McDonnell J. The use of technical
aids by elderly persons in The Netherlands: an application of
the Andersen and Newman model. Gerontologist 1997;37:
33. Kart CS, Engler CA. Predisposition to self-health care: who
does what for themselves and why? J Gerontol 1994;49:
34. DeFriese GH, Woomert A. Informal and formal health care
systems serving older persons. In: Ory MG, Abeles RP, Lip-
man PD, editors. Aging, health, and behavior. Newbury Park
(CA): Sage Publications; 1992. p. 57– 82.
35. Andersen RM. Behavioral model of families’ use of health
services. Chicago: University of Chicago; 1968.
36. Andersen R, Newman JF. Societal and individual determi-
nants of medical care utilization in the United States. Milbank
Mem Fund Quart (Health Society) 1973;51:95–124.
37. Andersen RM. Revisiting the behavioral model and access to
medical care: does it matter? J Health Soc Behav 1995;36:1–
38. Hochbaum GM. Public participation in medical screening
programs: a sociopsychological study. Washington (DC): US
Government Printing Ofﬁce; 1958.
39. Leventhal H, Diefenbach M, Leventhal EA. Illness cognition:
using common sense to understand treatment adherence and
affect cognition. Cognitive Therap Res 1992;16:143– 63.
40. Leventhal EA, Crouch M. Are there differences in perceptions
of illness across the lifespan? In: Petrie KJ, Weinman JA,
editors. Perceptions of health and illness. Amsterdam: Har-
wood Academic Publishers; 1997. p. 77–102.
41. Leventhal EA, Leventhal H, Robitaille C. Enhancing self-care
research: exploring the theoretical underpinnings of self-care.
In: Ory MG, DeFriese GH, editors. Self-care in later life. New
York: Springer; 1998. p. 118 –41.
42. Musil CM, Morris DL, Haug MR, Warner CB, Whelan AT.
Recurrent symptoms: well-being and management. Soc Sci
Med 2001;52:1729 –40.
Critical Review of Arthritis Self-Management Use 731