Rheumatology 2005;44:521–528
Advance Access publication 22 February 2005
doi:10.1093/rheumatology/keh534
Hand use and patterns of joint involvement
in osteoarthritis. A comparison of female
dentists and teachers
S. Solovieva, T. Vehmas1, H. Riihimäki, K. Luoma2 and P. Leino-Arjas
Objectives. To investigate the effect of mechanical stress on finger osteoarthritis (OA) by comparing women from two
occupations with different hand load but the same socio-economic grade, and to investigate whether hand load may affect the
pattern of joint involvement in OA.
Methods. Radiographs of both hands of 295 dentists and 248 teachers were examined. Each interphalangeal (distal, proximal
and thumb interphalangeal) and the metacarpophalangeal joints were graded (0 ¼ no OA, 4 ¼ severe OA) separately by using
reference images. The co-involvement of different hand joints was analysed by logistic regression.
Results. The distal interphalangeal joints were the most frequently involved joints. The non-dominant hand was more frequently
affected by OA of grade 2 or more than the dominant hand. The prevalence of OA of grade 2 or more in any finger joint and
also in any distal interphalangeal joint was higher among the teachers compared with the dentists (59 vs 48%, P ¼ 0.020
and 58 vs 47%, P<0.010 respectively). Finger OA showed more clustering in the ring and little fingers and more row clustering
and symmetry in the teachers than in the dentists [age-adjusted odds ratio (OR) ¼ 1.57, 95% confidence interval (CI) 1.10–2.23,
OR ¼ 1.84, 95% CI 1.28–2.64, and OR ¼ 1.98, 95% CI 1.38–2.86 respectively]. The OR of more severe OA (grade 3 or more)
in the right-hand thumb and the index and middle fingers was significantly elevated among the dentists compared with the
teachers (OR 2.61, 95% CI 1.03–6.59).
Conclusion. Our findings indicate that finger OA in middle-aged women is highly prevalent and often polyarticular. Hand use
may have a protective effect on finger joint OA, whereas continuing joint overload may lead to joint impairment.
KEY WORDS: Hand, Osteoarthritis, Finger joint.
Osteoarthritis (OA) is the most common joint disorder in the
adult population, and hand joints are those most often affected
by it [1]. The prevalence of OA increases with age with evident
sex-specific differences [2]. Before the age of 50 yr, the prevalence
of OA in most joints is higher in men than in women, but hand
OA is more prevalent among women [2, 3].
Some studies have revealed certain patterns of disease clustering, such as rows (the same joint affected in several fingers) and
rays (multiple joints in the same finger), as well as symmetrical
associations among the different hand joints [4–7]. Egger et al. [5]
provided clear evidence for the polyarticular subset of hand
OA in women and pointed out that the three major patterns of
polyarticular involvement, in descending order of importance, are
symmetry, clustering by row and clustering by ray. Later, these
findings were verified in a longitudinal study of the members of the
Framingham Study cohort [6]. In a large British cohort study, the
pattern of hand joint involvement in clinical OA was found to be
almost identical between 53-yr-old men and women [7].
A high prevalence of OA in the right hand compared with the
left hand in right-handed persons was found in the New Haven
Survey of Joint Diseases [8]. Ever since this result and the ‘wear
and tear’ hypothesis proposed by Radin and colleagues [9], joint
degeneration has been linked to mechanical loading. According to
the hypothesis, wear comes from repeated stresses due to continued
loading of the joints. Job tasks that require intensive use (wear)
of the same muscles or motions for a long duration increase the
likelihood of both localized and general fatigue and in turn the
likelihood of injury (tear). Prolonged or repeated overuse of a
group of joints has been related to an increased frequency of OA in
some studies [10–13] although not all [14–16].
Epidemiological studies provide evidence linking occupational
physical activities with osteoarthritis of the lower extremity
weight-bearing hip and knee joints [17, 18], and highly repetitive
arm motions alone or in combination with other factors (such as
awkward posture, force, vibration and duration of the movements)
with hand/wrist disorders [19]. Although hand joints are the joints
most commonly affected by OA, there is little information on
the role of occupation in the aetiology of finger OA and data on
women are scarce.
Dentistry is one of the few occupations with an academic
background that involve extensive bimanual work. Dentists perform arm movements repeatedly, often rapidly and for extended
periods of time. The precision grip is used in the handling of
precision tools [20], some of which vibrate. The work postures tend
to be particularly awkward for the neck and upper limbs. Dentists
frequently assume static postures which require more than 50% of
the body’s muscles to contract to hold the body motionless while
resisting gravity [21]. Static postures are often combined with
repetitive movements inherent in the nature of tasks performed.
While dentists use the thumb, index and middle fingers heavily in
Department of Epidemiology and Biostatistics and 1Department of Occupational Medicine, Finnish Institute of Occupational Health, Helsinki and
2
Department of Radiology, Peijas Hospital, Helsinki University, Central Hospital, Vantaa, Finland.
Submitted 9 November 2004; revised version accepted 1 December 2004.
Correspondence to: S. Solovieva, Department of Epidemiology and Biostatistics, Finnish Institute of Occupational Health, Topeliuksenkatu 41a A,
00250 Helsinki, Finland. E-mail: Svetlana.Solovieva@ttl.fi
521
Rheumatology Vol. 44 No. 4 ß British Society for Rheumatology 2005; all rights reserved
522
S. Solovieva et al.
precision gripping with their right hand, the ring and little fingers
remain static [22].
Teachers represent an occupational group with a comparable
academic background with the dentists but with a distinctively
different hand load. Although hand activities such as writing and
typing are common in a teacher’s work, they are usually of shorter
duration, require less force, and frequently alternate with nonhand activities. Furthermore, there is no exposure to vibration.
The aim of the study was to investigate the effect of mechanical
stress on finger OA by comparing women from an occupation
which places distinctive and continuous demands on the hands
(dentistry) with women from a different occupation in the same
socio-economic grade (teaching). Assuming that the pattern of
hand usage among dentists is characterized by stereotyped repetitive tasks for prolonged periods of time, we expected that the hands
of dentists will be significantly more often affected by OA than
the hands of teachers. We also investigated whether hand use
affects the pattern of joint involvement in OA. Based on the wear
and tear theory and on the assumption that the dominant hand is
subject to a greater daily stress than the non-dominant hand, we
hypothesized that the dominant hand will demonstrate a higher
prevalence of finger OA than the non-dominant hand.
Materials and methods
Study samples
Potential subjects were identified through the registers of the
Finnish Dental Association and the Finnish Teachers’ Trade
Union (comprising both the occupationally active and non-active).
In Finland 98% of the dentists and 95% of the teachers are trade
union members. In 2002, a questionnaire was sent to 436 female
dentists and 436 female teachers aged 45–63 yr randomly
selected from the registers, using the place of residence (Helsinki
or its neighbouring cities) as an inclusion criterion. Of those who
received the questionnaire, 295 (67.7%) dentists and 248 (56.9%)
teachers participated in a clinical examination between October
2002 and March 2003. Mean age of the dentists was 54 yr (S.D. 6,
range 45–63). Corresponding numbers for the teachers were 54 yr
(S.D. 4, range 45–61). Mean number of years in occupation was
26 (S.D. 7, range 11–40) for the dentists and 24 (S.D. 7, range 1–37)
for the teachers. Ninety-four per cent (n ¼ 277) of the dentists and
98.0% (n ¼ 243) of the teachers were occupationally active at the
time the study was conducted.
Participation in the study was voluntary and based on informed
consent. The Hospital District of Helsinki and Uusimaa Ethics
Committee for Research in Occupational Health and Safety
approved the study proposal.
Hand radiography and image analysis
Both hands of the participants were radiographed. Kodak X-ray
films were exposed with Siemens X-ray equipment (48 kV, 10 mAs,
focus film distance 115 cm). The analogue radiographs were
inspected on lighted view boxes and were first evaluated by an
experienced radiologist (T.V.) who was blinded to the occupation,
age and all health data of the subjects. The ratings were subsequently typed into a computer.
Each distal interphalangeal (DIP), proximal interphalangeal
(PIP), thumb interphalangeal (IP) and metacarpophalangeal
(MCP) joint of both hands was graded separately, and classified
for the presence of OA using the modified Kellgren and Lawrence
[23] system. The classification criteria were: grade 0 ¼ no OA
(normal finding); grade 1 ¼ doubtful OA (finding possibly slightly
abnormal); grade 2 ¼ mild OA (a single radiographic sign indicative of OA, slight to moderate lowering of the joint space, sometimes subluxation, minimal osteophytes, degeneration cysts or
slight marginal sclerosis, each of the latter signs without a clear
narrowing of joint space but little if any additional pathology)
(Fig. 1a); grade 3 ¼ moderate OA (considerable narrowing of joint
space with additional degenerative pathology as indicated in grade
2, no destruction of the joint) (Fig. 1b); grade 4 ¼ severe OA (joint
space destructed or poorly visible with various advanced degenerative changes) (Fig. 1c).
The reliability of the readings was estimated by measuring
intraobserver and interobserver agreement, using the weighted
Cohen’s coefficient with quadratic weights [24], for 46 randomly
chosen subjects. A second reading was independently performed
by T.V. and another experienced radiologist (K.L.). The values
can be interpreted as follows: less than 0.20, poor; 0.21–0.40, fair;
0.41–0.60; moderate; 0.61–0.80; good; 0.81–1.0, very good agreement. The interobserver agreement for OA ranged from 0.67 to
0.85 for DIP joints, from 0.39 to 0.61 for PIP joints and from 0.18
to 0.69 for MCP joints. The intraobserver agreement for OA
ranged from 0.73 to 0.88 for DIP joints, from 0.67 to 0.92 for PIP
joints and from 0.59 to 1.0 for MCP joints. The matrix of weighted
coefficients is shown in Appendix 1 (available as supplementary
data at Rheumatology Online). The readings of T.V. only were used
in the subsequent statistical analyses.
If the subject had at least one finger joint with radiographic OA,
she was classified as having finger OA. Two cut-offs for radiographic finger OA were used: OA of grade 2–4 (mild OA) and OA
of grade 3–4 (moderately severe OA). A particular joint group (i.e.
DIP, PIP, thumb IP, MCP) was defined as having OA if at least
one joint of that joint group had radiographic OA. A particular
finger group (i.e. thumb, finger and middle fingers, little and ring
fingers) was defined as having OA if at least one joint of that finger
group had radiographic OA. Symmetrical OA was defined as
a subcategory of OA: OA in at least one symmetrical pair of the
joints—if the joint of one hand is affected, the same joint of the
opposite hand is also affected.
Statistical analyses
The prevalence of OA was calculated for each joint separately.
The differences in the prevalence of finger OA between dentists
and teachers were compared using Fisher’s exact probability test.
To assess the relationship between handedness and finger OA,
the prevalence ratio (prevalence of OA in the dominant hand/
prevalence of OA in the non-dominant hand) and 95% confidence
interval was calculated by the Mantel-Haenszel method.
To test whether OA is likely to affect multiple finger joints in
some individuals, the subjects were divided into three age groups:
45–49, 50–54, 55–63 yr, and the age-specific prevalence of OA
was obtained separately for each joint. The number of subjects
expected to have 0, 1, 2, 3 or 4þ joints with OA, assuming that
the presence of the disease in different hand joints in a subject
is independent of the presence in other joints, was calculated using
the Poisson distribution. The distribution parameter used to
generate the expected number of subjects was the average number
of joints per individual. The observed frequencies were compared
with the expected frequencies using the 2 test.
The interrelation of OA occurrence in different finger joints was
analysed by logistic regression. First, for each pair of joints, the
relationship was described by the conditional odds ratio (OR: the
relative odds of having OA in one joint if the other joint is also
affected by OA). Then, a single summary OR was obtained for
associations within the groups of joints using the Mantel-Haenszel
OR estimator. Estimates of OR were obtained for the clustering of
joint involvement by row (the same joint of several fingers) and ray
(several joints of the same finger involved) in the same hand and
symmetrical (same joint, same-finger, opposite hands) involvement
of the two hands, and were adjusted for age (year of birth).
To evaluate whether occupational hand load was associated
with the location of OA at a particular site (left vs right hand, in the
thumb, index and middle fingers, or in the little or ring fingers)
Hand use and patterns of joint involvement in OA
523
FIG. 1. (a) Mild OA (grade 2). See text for further definition. (b) Moderate OA. Considerable narrowing of joint space with additional
degenerative pathology. (c) Severe OA. Joint space destroyed/poorly visible. Various advanced degenerative changes.
a further analysis was performed among women with OA in
any finger joint: the odds of having a particular pattern of multiple joint involvement among dentists was estimated by logistic
regression analysis, using the teachers as a reference group. ORs
and their 95% confidence intervals (CIs) were adjusted for age
(year of birth).
Analyses were performed with the SAS statistical software
version 8.2 (SAS, Institute, Cary, NC, USA).
S. Solovieva et al.
524
Results
Prevalence of finger OA
Only 2.7% of the dentists and 4.8% of the teachers were lefthanded. The prevalence of OA in each joint group and each finger
is presented in Table 1. OA of grade 2 or more occurred most
frequently in the DIP joints and the joints of the little finger.
The MCP joints and the thumb joints were the least frequently
affected by OA. A similar tendency was observed for OA of grade 3
or more. The prevalence of OA of grade 2 or more was higher in
the non-dominant hand compared with the dominant (Table 1).
The difference between non-dominant and dominant hand
involvement reached statistical significance for the joints of the
ring and middle fingers [prevalence ratio (PR) ¼ 0.76, 95% CI
0.61–0.95 and PR ¼ 0.75, 95% CI 0.57–0.98 respectively]. The
prevalence of more severe OA (grade 3 or more) was slightly
higher, although not statistically significantly so, in the joints of the
dominant hand compared with the non-dominant hand.
Prevalence of finger OA and occupation
The prevalence of OA for each joint separately by hand laterality
and occupation is shown in Appendix 2 (available as supplementary data at Rheumatology Online). The prevalence of OA
increased with age in both occupations (Fig. 2). The prevalence of
OA of grade 2 or more in any finger joint as well as OA in any DIP
TABLE 1. Association between handedness (dominant or non-dominant) and prevalence of finger OA among female dentists and teachers (N ¼ 543)
Dominant hand
Joint group
OA of grade 2 or more
DIP
PIP
MCP
Thumb IP
Little finger
Ring finger
Middle finger
Index finger
Thumb
OA of grade 3 or more
DIP
PIP
MCP
Thumb IP
Little finger
Ring finger
Middle finger
Index finger
Thumb
Non-dominant hand
(n)
(%)
n
(%)
PR
95% CI
230
18
3
20
214
112
77
66
22
42.4
3.3
0.6
3.7
39.4
20.6
14.2
12.2
4.1
255
24
2
32
243
147
103
80
35
47.0
4.4
0.4
6.0
44.7
27.1
19.0
14.7
6.4
0.90
0.75
1.50
0.63
0.88
0.76
0.75
0.83
0.63
0.79–1.03
0.41–1.36
0.30–7.51
0.36–1.07
0.77–1.01
0.61–0.95
0.57–0.98
0.61–1.12
0.38–1.05
62
11
2
8
45
20
25
27
8
11.4
2.0
0.4
1.5
8.3
3.7
4.6
5.0
1.5
56
7
2
4
40
23
29
26
4
10.3
1.3
0.4
0.7
7.4
4.2
5.3
4.8
0.7
1.10
1.57
1.00
2.00
1.13
0.87
0.86
1.04
2.00
0.79–1.55
0.63–3.91
0.18–5.65
0.65–6.23
0.75–1.69
0.49–1.55
0.51–1.45
0.62–1.75
0.65–6.23
The prevalence ratio (PR; prevalence of OA in the dominant hand/prevalence of OA in the non-dominant hand) and 95% CI were calculated by
the Mantel-Haenszel method.
Dentists
100,0 %
Teachers
90,0 %
Prevalence of OA (%)
80,0 %
65 %
66 %
64 %
63 %
70,0 %
57 %
60,0 %
p=0.02
59 %
57 %
48 %
50,0 %
40 %
41 %
40 %
58 %
47 %
38 %
40,0 %
30 %
p=0.01
30 %
30,0 %
20,0 %
10 %
7%
10,0 %
6%
2 %3 %
2% 0%
4%
0,0 %
Any
Age-group:
DIP
45-49 years
PIP
Any
DIP
PIP
50-54 years
Any
DIP
55-63 years
PIP
Any
DIP
PIP
All
Hand joint type
FIG. 2. Prevalence of OA of grade 2 or more in any finger joint (Any), in any distal interphalangeal (DIP) and proximal
interphalangeal (PIP) joints among dentists and teachers by age categories. The numbers of females (dentist/teachers) were 93/47 in the
45–49 age group, 61/68 in the 50–54 age group, 74/108 in the 55–59 age group and 67/25 in the 60–63 age group.
Hand use and patterns of joint involvement in OA
joint was significantly higher among teachers than dentists (59
vs 48%, P ¼ 0.020 and 58 vs 47%, P ¼ 0.010, respectively; Fisher’s
exact probability test). There was no difference in the prevalence of
more severe OA (grade 3 or more) in any finger joint between the
dentists and teachers (13 and 17% respectively, P ¼ 0.19; Fisher’s
exact probability test).
within the ring and little fingers (OR ¼ 9.75, 95% CI ¼ 4.78–19.88)
than in the thumb, index and middle fingers (OR ¼ 1.50, 95%
CI ¼ 0.96–2.23). No symmetrical occurrence of moderately severe
OA was observed in either occupation (OR ¼ 0.83, 95% CI ¼
0.54–1.26 in dentists and OR ¼ 0.90, 95% CI ¼ 0.58–1.40 in
teachers).
The age-adjusted OR of having a particular pattern of OA
among the dentists compared with the teachers was calculated
(Table 4). Among the dentists with OA of grade 2 or more in any
finger joint, the OR of having it in the ring and little fingers of
the left hand was significantly lower than the corresponding
OR among the teachers with OA (OR 0.16, 95% CI 0.05–0.47).
However, the OR of more severe OA (grade 3 or more) in the
thumb, index and middle fingers of the right hand was significantly
elevated among the dentists with severe OA compared with the
teachers (OR 2.61, 95% CI 1.03–6.59).
Effect of hand use on the patterns of joint involvement
Table 2 shows the observed frequencies of the number of joints
involved per person and the corresponding expected frequencies
assuming that there was no clustering within the individual.
The clustering of hand joint involvement was statistically highly
significant (2 test, P ¼ 0.001).
There were distinct patterns of multiple joint involvement in
the hands (Table 3). In both occupations, the strongest association
was between the ring finger’s joint involvement and the little
finger’s joint involvement of the same hand (OR ¼ 16.90, 95% CI
10.24–27.80 for the dentists and OR ¼ 47.33, 95% CI 21.52–104.09
for the teachers). Furthermore, if one DIP or PIP joint was
involved the OR of any other DIP or PIP joint in that hand being
affected increased about 3-fold among the dentists and about
6-fold among the teachers. The clustering of finger joint involvement in OA was symmetrical. Clustering within the ring and little
fingers, clustering by row, and symmetrical joint involvement were
clearly more apparent among the teachers than the dentists (ageadjusted OR ¼ 1.57, 95% CI 1.10–2.23, OR ¼ 1.84, 95% CI
1.28–2.64, and OR ¼ 1.98, 95% CI 1.38–2.86, respectively).
The patterns of joint involvement for moderately severe (grade 3
or more) OA were different between dentists and teachers. Among
the dentists, OA (grade 3 or more) was more frequently clustered
within the thumb, index and middle fingers (OR ¼ 1.80, 95% CI ¼
1.16–2.80) than within the ring and little fingers (OR ¼ 1.47, 95%
CI ¼ 0.95–2.27). The pattern among the teachers was quite the
opposite, with moderately severe OA more frequently clustered
Discussion
This study investigated the association of extensive hand use with
finger OA and characterized the patterns of hand joint involvement
among middle-aged women from two occupations with different
hand loads. Our findings provide evidence for a high prevalence of
radiographically defined finger OA in women aged 45–63 yr and
for that of a polyarticular subset of finger OA. The joints most
commonly affected were the same in both occupations, i.e. the DIP
joints, while the MCP joints were less often involved. Clustering
within the little and ring fingers, row clustering and symmetry were
the most pronounced features in the patterns of joint involvement.
Clustering within the thumb, index and middle fingers was more
commonly observed in the dentists.
Effect of hand load on the pattern of joint involvement
We examined the pattern of joint involvement by using mild and
moderate/severe cut-offs for radiographic finger OA. The clustering of affected joint groups within a hand was similar to that
observed in previous studies [5, 6], with a stronger relationship
between multiple involvement of the same joints in several fingers
than with involvement of joints of any given ray. A high degree
of symmetrical hand joint involvement was also confirmed.
Furthermore, clustering within the little and ring fingers was
revealed.
The remarkably high rate of OA co-occurrence in the joints of
the little and ring fingers can not be easily explained by the impact
of joint loading only. Possible explanations for this result include
an inherited predisposition to OA. The tendency towards symmetrical joint involvement also suggests the importance of genetic [25]
and systemic factors [26] in the aetiology of OA. It is possible that
there are differences between the two occupations in the presence
of risk factors or protective factors. The socio-economic status
of the dentists and the teachers was similar, although the average
income of dentists tends to be higher.
TABLE 2. Observed and expected numbers of hand joints affected by OA
in 543 women (295 dentists and 248 teachers)
OA (grade 2 or more)
Number
of joints
0
1
2
3
4þ
2
Degrees
of freedom
P value
Observed
254
59
70
28
132
186.5
4
Expected
OA (grade 3 or more)
Observed
298
179
48
10
8
0.001
Expected
461
31
18
10
23
58.8
4
525
457
79
7
0
0
0.001
TABLE 3. Presence of OA of grade 2 or more in a particular finger joint according the OA status of other finger joints
Dentists (n ¼ 295)
Pattern of joint involvement
Same row, same hand
Same ray, same hand
Same joint, opposite hand
Thumb or index and middle fingers, same hand
Ring and little fingers, same hand
Teachers (n ¼ 248)
n
OR
95% CI
n
OR
95% CI
104
13
94
70
135
2.97
0.10
1.96
0.96
16.90
2.27–3.89
0.07–0.15
1.53–2.50
0.76–1.20
10.24–27.80
125
9
118
75
142
6.25
0.07
4.37
1.07
47.33
4.48–8.71
0.04–0.11
3.25–5.86
0.85–1.34
21.52–104.09
The conditional OR (the relative odds of having OA in one joint if the other joint is also affected by OA) was calculated for each pair of joints using
logistic regression. A single summary OR was obtained for associations within the groups of joints using the Mantel-Haenszel OR estimator. ORs
shown in the table are age-adjusted.
S. Solovieva et al.
526
TABLE 4. ORs of finger OA patterns as a function of occupation among those with OA in any (1) finger joint
OA of grade 2 or more present (n ¼ 288)
OA of grade 3 or more present (n ¼ 82)
OR
95% CI
OR
95% CI
0.10
0.76
0.02–0.44
0.42–1.39
0.76
1.15
0.28–2.09
0.37–3.52
0.38
0.16
0.64
0.10–1.47
0.05–0.47
0.37–1.13
0.20
0.33
0.49
0.06–0.69
0.13–0.84
0.20–1.25
Thumb, index, and middle fingers
Overall
0.81
Left hand
0.88
Right hand
0.62
0.50–1.31
0.54–1.43
0.37–1.02
1.52
1.15
2.61
0.59–3.90
0.47–2.84
1.03–6.59
Any finger
Left hand
Right hand
Ring and little fingers
Overall
Left hand
Right hand
The table shows age-adjusted ORs and their 95% CIs for dentists with teachers as the reference group.
A number of differences in the pattern of joint involvement
between the two occupations were discovered. In the teachers,
there was a tendency to develop multiple joint involvement in the
little and ring fingers, in the same row of fingers (same joint but
different fingers), and symmetrically. The dentists were more likely
to develop severe OA in the joints of the right hand’s thumb and
the index and middle fingers compared with the teachers. These
fingers are the ones used by dentists in holding instruments accompanied by precise and repetitive movements. These fingers seem
to be subject to overload rather than limited mechanical stress.
Hand use and finger OA
Our data did not support the hypothesis of increased radiographic
OA in the dominant hand compared with the non-dominant hand.
This is in conflict with the findings of the New Haven Survey of
Joint Diseases, in which a statistically significant increase in the
prevalence and severity of OA in the dominant hand was observed
[8]. No radiographic or clinical differences were found between
the right and left hands in a study of 134 consecutive community
subjects with a detailed description of the hand use history [14].
In agreement with the Mini-Finland study [16], the prevalence of
OA was higher in the joints of the non-dominant hand than in the
dominant hand. However, Cvijetı́c et al. [27, 28] and Egger et al. [5]
reported the reverse association.
In a Virginia textile mill, female workers whose jobs required a
repeated pinch grip had a much higher rate of OA in DIP joints
than other female workers [10]. The results of the present study
showed that the prevalence of OA in any finger joint or in any DIP
joint of the dentists was lower compared with those of the teachers.
Lehto et al. [14] did not find any differences in the prevalence of
OA between female dentists and female controls randomly selected
from the general population in a study of radiographic arthritis of
the hands in 136 dentists.
The current findings regarding higher prevalence of mild
degenerative changes in the little finger compared with other
fingers, in the non-dominant compared with the dominant hand
and in the teachers compared with the dentists indicate that wear
alone may be insufficient to explain the prevalence and location
of OA. This raises the question whether the non-dominant hand
and the little finger, which are engaged in somewhat more static
activities compared with the dominant hand and other fingers,
could be subject to some specific stress. It seems more plausible,
however, that persistent and intensive hand use increases the
strength of muscles and ligaments and through this or other
mechanisms protects against wear and tear injury. The results of
the study by Rogers et al. [29] showed that moderate/high joint
stress was associated with a reduced risk of hip/knee OA among
women.
The prevalence rates of radiographically-defined osteoarthritis
in this study were higher than those reported in several epidemiological studies [4, 6, 30, 31] and similar to those found in the
Zoetermeer survey [32] and the Mini-Finland study [16]. Our
findings on the prevalence of OA in different joint groups are
consistent with the results from other studies [5, 16, 31]. However,
some studies found a higher prevalence of OA in MCP than in PIP
joints [4, 32]. The forces generated by diverse hand activities
produce different loads on the finger joints. The MCP joints are
known to be especially affected by heavy manual labour [33].
Schmid et al. [34] found that the distribution of OA in the hands
with overproportional involvement of the MCP joints was associated with lifelong farming. Using biomechanical analysis of loads
in the finger joints, Radin et al. [9] found that forces were greatest
across the DIP joints and that fine pinch grip resulted in greater
mechanical stress than power grip. Therefore, the preferential
involvement of the DIP rather than the MCP joints in hand OA
among dentists and teachers in the present study was expected.
Methodological considerations
Some of the discrepancies of our findings with the previously
published reports can be due to differences in the classification
criteria. Most studies to date have used the system of grading
radiographic severity (0–4) developed by Kellgren and Lawrence
[23], though there has been some variation in the description of
these grades [35]. The Kellgren and Lawrence criteria perhaps put
too much emphasis on osteophytes, assuming that joint space
narrowing occurs after osteophyte formation [36]. A number of
alternative grading systems have been developed to overcome these
deficiencies [37–39]. A recent study by Neame et al. [12] showed
that osteophytes were more marked on the right side for all DIP
and PIP joints, whereas there were no right–left differences in joint
space width. The classification system used in the present study
gave less weight to the osteophytes, especially in the definition of
mild OA (of grade 2 or more).
The reliability of the OA classification criteria is critical for
the credibility of results. While there was substantial variation
(especially for the less frequently affected joint groups) in the
readings between the two radiologists, intraobserver agreement
was generally good. The lower level of interobserver agreement was
not surprising. Despite training and the use of reference images,
each reader graded the radiographs according to his or her
own inherent standard about what constituted a positive finding.
The high intraobserver agreement suggests that the classification criteria applied here are highly reproducible. Because all
Hand use and patterns of joint involvement in OA
radiographs were evaluated by one observer who was blind to the
subjects’ occupational status, the high intraobserver repeatability
implies that the comparison between teachers and dentists was
unbiased.
The assumption made in the last set of the logistic regression
analyses led to reduction in the number of subjects to be analysed.
Thus, the estimation of the odds of OA location at a particular site,
especially in the thumb, index and middle fingers, may not be
precise.
Representativeness of the sample
There was some difference in the study participation rate between
the two occupational groups, especially among the participants
aged 45–49 and 60–63 yr. No teachers belonging to the age group
of 62–63 yr took part in the study. Although the official age for oldage pension in Finland is 65 yr, the actual average retirement age
is 59 yr [40]. Workplace stress is a significant reason for reduced
hours or part-time retirement, particularly for workers in their 60s
[41]. Perhaps it is the heavy workload of the teachers, characterized
by long hours devoted to teaching-related activities, and relatively
low reward through monetary compensation that tend to push
the teachers towards early retirement. The registers of the Finnish
Dental Association and the Finnish Teachers’ Trade Union,
through which the potential subjects were identified, include both
occupationally active and non-active persons. Thus, the largest
difference in participation rate, observed among the older age
group, may be due to a difference in willingness to participate in
a study on workload and health. Among the younger age group,
where the difference was smaller (63% for the dentists and 58% for
the teachers), those women who were occupationally active, had
experienced symptoms and/or had a family history of arthritis
were possibly more willing to take part in the study. However, the
most noticeable difference in the prevalence of OA between the
occupations was observed among those aged 50–54 yr, in whom
the participation rates were similar (62% for the dentists and 60%
for the teachers). Thus, the difference in participation rate is not
likely to account for the difference in the prevalence of OA between
the occupations.
Conclusions
The present findings suggest that moderate hand use does not have
a detrimental effect on finger OA but may even be protective,
whereas continuous joint overload may lead to joint impairment.
Furthermore, the excessive prevalence of OA found among
the teachers, and in the non-dominant hand joints of both
occupational groups, as well as the noticeable symmetry of joint
involvement seem to suggest the importance of non-mechanical
aetiological factors.
Rheumatology
Key messages
Clustering by finger groups and the row
and symmetrical patterns were the most
pronounced patterns.
Hand use may have a protective effect on
finger OA.
Continuous joint overload may lead to
joint impairment.
Acknowledgements
This work was financially supported by the Finnish Work
Environment Fund.
There is no conflict of interest.
527
Supplementary data
Supplementary data are available at Rheumatology Online.
References
1. Felson DT. Epidemiology of hip and knee osteoarthritis. Epidemiol
Rev 1988;10:1–28.
2. Felson DT, Lawrence RC, Dieppe PA et al. Osteoarthritis: New
insights. Part 1: the disease and its risk factors. Ann Intern Med
2000;133:635–46.
3. Jones G, Cooley HM, Stankovich JM. A cross sectional study of
the association between sex, smoking, and other lifestyle factors and
osteoarthritis of the hand. J Rheumatol 2002;29:1719–24.
4. Butler WJ, Hawthorne VM, Mikkelsen WM et al. Prevalence of
radiologically defined osteoarthritis in the finger and wrist joints of
adult residents of Tecumseh, Michigan, 1962–65. J Clin Epidemiol
1988;41:467–73.
5. Egger P, Cooper C, Hart DJ, Doyle DV, Coggon D, Spector TD.
Patterns of joint involvement in osteoarthritis of the hand: the
Chingford Study. J Rheumatol 1995;22:1509–13.
6. Chaisson CE, Zhang Y, McAlindon TE et al. Radiographic hand
osteoarthritis: incidence, pattern, and influence of pre-existing disease
in a population based sample. J Rheumatol 1997;24:1337–43.
7. Poole J, Sayer AA, Hardy R, Wadsworth M, Kuh D, Cooper C.
Patterns of interphalangeal hand joint involvement of osteoarthritis
among men and women: a British cohort study. Arthritis Rheum
2003;48:3371–6.
8. Acheson RM, Chan YK, Clemett AR. New Haven survey of joint
diseases. XII. Distribution and symptoms of osteoarthrosis in the
hands with reference to handedness. Ann Rheum Dis 1970;29:275–86.
9. Radin EL, Parker HG, Paul IL. Pattern of degenerative
arthritis. Preferential involvement of distal finger-joints. Lancet
1971;1:377–9.
10. Hadler NM, Gillings DB, Imbus HR et al. Hand structure and
function in an industrial setting. Arthritis Rheum 1978;21:210–20.
11. Elsner G, Nieenhaus A, Beck W. Arthroses of the finger joints
and thumb saddle joint and occupationally related factors.
Gesundheitswesen 1995;57:786–91.
12. Neame R, Zhang W, Deighton C, Doherty M, Doherty S, Lanyon P,
Wright G. Distribution of radiographic osteoarthritis between the
right and left hands, hips, and knees. Arthritis Rheum 2004;
50:1487–94.
13. Hunter DJ, Zhang Y, Nevitt MC et al. Chopstick arthropathy: the
Beijing Osteoarthritis Study. Arthritis Rheum 2004;50:1495–500.
14. Lehto TU, Rönnemaa TE, Aalto TV, Helenius HYM.
Roentgenological arthrosis of the hand in dentists with reference
to manual function. Community Dent Oral Epidemiol 1990;
18:37–41.
15. Lane NE, Bloch DA, Jones HH, Simpson U, Fries JF. Osteoarthritis
in the hand: a comparison of handedness and hand use. J Rheumatol
1989;16:637–42.
16. Haara MM, Manninen P, Kroger H et al. Osteoarthritis of finger
joints in Finns aged 30 or over: prevalence, determinants, and
association with mortality. Ann Rheum Dis 2003;62:151–8.
17. Lievense A, Bierma-Zeinstra S, Verhagen A, Verhaar J, Koes B.
Influence of work on the development of osteoarthritis of the hip:
a systematic review. J Rheumatol 2001;28:2520–8.
18. Maetzel A, Makela M, Hawker G, Bombardier C. Osteoarthritis of
the hip and knee and mechanical occupational exposure—a systematic
overview of the evidence. J Rheumatol 1997;24:1599–607.
19. Bernard BP. Musculoskeletal disorders and workplace factors.
NIOSH Report 97–141. Cincinatti (OH): National Institute for
Occupational Safety and Health, 1997.
20. Mathiowetz V, Kashman N, Volland G, Weber K, Dowe M, Rogers S.
Grip and pinch strength: normative data for adults. Arch Phys Med
Rehabil 1985;66:69–74.
528
S. Solovieva et al.
21. Valachi B, Valachi K. Mechanisms leading to musculoskeletal disorders in dentistry. J Am Dent Assoc 2003;134:1344–50.
22. Fish DR, Morris-Allen DM. Musculoskeletal disorders in dentists.
N Y State Dent J 1998;64:44–8.
23. Kellgren JH, Lawrence JS. Radiologic assessment of osteoarthritis.
Ann Rheum Dis 1957;16:494–502.
24. Cohen J. Weighted kappa. Nominal scale agreement with provision
for scaled disagreement or partial credit. Psychol Bull 1968;70:213–20.
25. Spector TD, Cicuttini F, Baker J, Loughlin J, Hart D. Genetic
influences on osteoarthritis in women: a twin study. BMJ 1996;312:
940–3.
26. Spector TD, Campion GD. Generalised osteoarthritis: a hormonally
mediated disease. Ann Rheum Dis 1989;48:523–7.
27. Cvijetı́c S, Dekanic D, Kurtagic N, Roic G. Radiologically defined
osteoarthrosis in the finger joints of adult residents of Zagreb. Arh
Hig Rada Toksikol 1994;45:219–29.
28. Cvijetı́c S, Campbell L, Cooper C, Kirwan J, Potocki K.
Radiographic osteoarthritis in the elderly population of Zagreb:
distribution, correlates, and the pattern of joint involvement. Croat
Med J 2000;41:58–63.
29. Rogers LQ, Macera CA, Hootman JM, Ainsworth BE, Blairi SN.
The association between joint stress from physical activity and selfreported osteoarthritis: an analysis of the Cooper Clinic data.
Osteoarthritis Cartilage 2002;10:617–22.
30. Heliövaara M, Mäkelä M, Sievers K et al. Musculoskeletal diseases
in Finland [In Finnish with English summary]. Publication AL:35.
Helsinki: National Insurance Institution, 1993.
31. Sowers M, Lachance L, Hochberg M, Jamadar D. Radiographically
defined osteoarthritis of the hand and knee in young and middle-aged
African American and Caucasian women. Osteoarthritis Cartilage
2000;8:69–77.
32. van Saase JL, van Romunde LK, Cats A, Vandenbroucke JP,
Valkenburg HA. Epidemiology of osteoarthritis: Zoetermeer survey.
Comparison of radiological osteoarthritis in a Dutch population with
that in 10 other populations. Ann Rheum Dis 1989;48:271–80.
33. Williams WV, Cope R, Gaunt WD et al. Metacarpophalangeal
arthropathy associated with manual labor (Missouri metacarpal
syndrome). Clinical radiographic, and pathologic characteristics of an
unusual degeneration process. Arthritis Rheum 1987;30:1362–71.
34. Schmid L, Dreier D, Muff B, Allgayer B, Schlumpf U. Lifelong heavy
agricultural work and development of arthrosis of the hand—a case
study. Z Rheumatol 1999;58:345–50.
35. Lawrence JS. Rheumatism in population. London: William
Heinemann Medical Books, 1977.
36. Hart DJ, Spector TD. Definition and epidemiology of osteoarthritis
of the hand: a review. Osteoarthritis Cartilage 2000;8(Suppl. A):
S2–7.
37. Altman R, Fries JF, Bloch. Radiographic assessment of progression
of osteoarthritis. Arthritis Rheum 1987;31:1214–25.
38. Kallman DA, Wigley FM, Scott WW, Hochberg MC, Torbin JD.
New radiographic grading scales for osteoarthritis of the hand.
Arthritis Rheum 1989;32:1584–91.
39. Kessler S, Dieppe P, Fuchs J, Sturmer T, Gunther KP. Assessing
the prevalence of hand osteoarthritis in epidemiological studies.
The reliability of radiological hand scale. Ann Rheum Dis 2000;
59:289–92.
40. National Programme of Ageing Workers. Implementation report by
an expert group. Older workers in the labour market and outside.
Report No. 259. Helsinki: Ministry of Labour, 2000.
41. McNair S, Flynn M, Owen L, Humphreys C, Woodfield S. Changing
work in later life: a study of job transitions. Guildford: University of
Surrey, 2004.