Medial epicondylitis in occupational settings: prevalence, incidence and associated risk factors

Abstract

As medial epicondylitis has not been studied alone, we investigated its links between personal and occupational factors in repetitive work, and its course.

1757 workers were examined by an occupational health physician in 1993–94. 598 of them were re-examined three years later.

Prevalence was between 4 and 5%, with annual incidence estimated at 1.5%. Forceful work was a risk factor for medial epicondylitis (OR 1.95 CI [1.15–3.32]), but not exposure to repetitive work (OR 1.11, CI [0.59–2.10]). Workers with medial epicondylitis had a significantly higher prevalence of other work-related upper-limb musculoskeletal disorders (WRMD). Risk factors differed for medial and lateral epicondylitis. The prognosis for medial epicondylitis in this population was good with a three-year recovery rate at 81%.

Medial epicondylitis was clearly associated with forceful work and other upper-limb WRMD, and its prognosis was good.

Introduction

The frequency of work-related upper-limb disorders is high in the industrialized countries. In France, the number of claims for work-related musculoskeletal disorders has increased 350% from 1993 (3165 cases) through 1999 (11095).

Shoulder tendinitis, lateral epicondylitis, wrist tendinitis and carpal tunnel syndrome are the most common upper-limb disorders. Their occupational associations, especially those for lateral epicondylitis, have been the object of many studies ^1–6. No studies, however, have focused on medial epicondylitis in the workplace, probably because of its infrequency. It seems to be more common among athletes, including golfers ^6;7 (the source of “golfer’s elbow”), throwing athletes ⁸, baseball players ⁹ and tennis players ¹⁰. Most studies of epicondylitis have not distinguished medial from lateral disease ^11–15.

Medial epicondylitis arises from the inflammation of medial epicondylian muscles due to micro-tears ¹⁶ in any of the following muscles: the superficialis flexor digitorum, pronator teres, flexor radialis carpi, longus palmaris (fickle) and flexor ulnaris carpi ¹⁷. These muscles have in common their insertion on the medial epicondyle of the humerus and anterior position in the forearm. They (except for the palmaris longus, which is an abductor muscle) transmit pronation, flexion and adduction to the wrist and flexion to the elbow.

This study was based on the data from a national survey ^18–20, which focused on work-related upper-limb disorders ¹⁸. Previous articles presenting the results about carpal tunnel syndrome, wrist tendinitis and lateral epicondylitis have described the study ^19;20. The objective of the current study were to describe prevalence, incidence and associated risk factors for medial epicondylitis, which was assessed twice in two separate medical examinations (1993–1994 and 1996–1997). We looked at various risk factors ²¹ for its prevalence and incidence and examined the differences in the risk factors for medial and lateral epicondylitis.

Subjects and methods

Subjects

In 1993–1994, 1757 workers completed a self-administered questionnaire about their work conditions and upper-limb disorders. They also underwent a standardized physical examination by the occupational health physician responsible for medical surveillance in their company.

The workers were selected according to occupational criteria. They were required to be exposed to repetitive work as a worker in one of five activity sectors: (i) assembly-line manufacture of small electrical appliances, motor vehicle accessories, or ski accessories, packaging excluded; (ii) clothing or shoe industry, packaging excluded; (iii) food industry, packaging excluded (mainly meat industry); (iv) packaging (primarily in the food industry); (v) supermarkets (cashiers). A final control group (vi) was made up with workers from the same industries who were not exposed to repetitive work.

All subgroups included both women and men except for the supermarket cashiers who were exclusively women.

Eighteen of the 39 occupational health physicians who participated in the 1993–1994 evaluations were able to repeat the study 3 years later. The 700 workers in 18 different firms whom they had examined in 1993–1994 were thus the target population of the longitudinal study. In all, 598 workers completed the self-administered questionnaire and were re-examined by the same physician. Most had the same (or similar) occupational constraints as 3 years earlier, in the same company. Four no longer worked at the same firm; 134 (22.4%) said that their job or their work station had changed during the period in question.

Medical variables

Each worker underwent a standardized clinical examination, performed by the occupational physician at the beginning of the study and again at the end of follow-up. A list of criteria for the diagnoses recorded in the medical questionnaire was prepared for the survey. These guidelines covered 33 diagnoses at the beginning of the study and 35 at the time of the follow-up (the slight change between the two lists were limited to shoulder tendinitis).

Diagnosis of medial epicondylitis corresponded to the following criteria:

• medial epicondylian pain or pain at the medial side of forearm.

• medial epicondylian tenderness and pain on resisted pronation or elbow flexion.

The following three situations or levels were possible: (i) diagnosis proved during the medical examination, (ii) diagnosis proved before medical examination (for example, previous diagnosis by a specialist), and (iii) suspected diagnosis (not all the criteria were met in the medical examination or the diagnosis was based on the description of symptoms no longer present at the examination).

We defined disorders extensively and included proved, suspected and prior diagnoses. This last category in the second questionnaire included cases that had appeared during the three-year interval. Definitions were based exclusively on the results of the medical examination. The answers to the arm diagram in the self-administered questionnaire were not taken into account.

We used the same procedure for the other diagnoses examined as we studied the relation between medial epicondylitis and other upper-limb musculoskeletal disorders, including shoulder tendinitis, carpal tunnel syndrome and lateral epicondylitis.

Potential risk factors

The list contained personal and occupational variable that included the subject’s assessment of the presence of specific postural and biomechanical constraints and psychosocial stresses at work. It was based on results from other studies and included variables known or suspected to be risk factors for work-related upper-limb disorders, with a special emphasis on risk factors for lateral epicondylitis or epicondylitis in general ^4;5.

Personal and occupational variables

The personal and occupational variables included the following:

• Gender

• Age, in four categories (<30, 30–39, 40–49, >50 years)

• Occupational variables: activity sector (of the 6 possible) and number of years on the job (<1, 1–9, >10 years)

• Smoking (non-smoker, smoker, ex-smoker)

• Body mass index (BMI): overweight, defined as BMI > 27 kg/m2 for men and 26.5 kg/m2 for women

• Leisure activities (do you often spend time participating in sports, playing music, on do-it-yourself house projects?), in two categories: yes or no

Presence of psychosomatic or depressive problems according to a short version of the Langner’s questionaire ²²: “presence” was defined as a yes answer to at least one of the following questions: do you often have headaches; do you often have sleep disorders; are you often bothered by your heart beating hard; do you have personal worries that get you down physically; are you in low or very low spirits most of the time; are you bothered by nervousness; do you often feel that nothing ever turns out the way you want it to; do you sometimes wonder if anything is worthwhile anymore. Three categories were defined: no problem, some problems (score=1 or 2), problems (score =3 to 8).

Postures and biomechanical constraints

The job duties mentioned in the self-administered questionnaire included the following actions: “holding in position”, “turning and screwing”, forceful work (that is, involving any action performed with force), using tools forcefully, and holding a tool in position. Two categories were considered: whether the action was performed repetitively or not.

Psychosocial work factors

Job control (scored from 0 to 8) was calculated as the sum of the following yes or no items: no choice of the timing of breaks; unscheduled short breaks not possible; pace dictated by the machine, by colleagues, by other constraints; no control over the quantity of work; no control over the pace of work; lack of variety of work. A low level of job control was defined as a score of >=5.

A low level of satisfaction was defined as a negative answer about either satisfaction with the workstation or general job satisfaction.

For the comparison between risk factors for medial epicondylitis and risk factors for lateral epicondylitis the list of variables also included “pressing with a hand” since it was associated with the prevalence of lateral epicondylitis in this population ¹⁸.

“Observer effect”

Among the risk factors for prevalence of lateral epicondylitis in this study, an observer effect variable (occupational physician group A or B) was considered as a confounding factor ¹⁸; the physicians in group A had in common that their diagnoses of epicondylitis were nearly always accompanied by the worker’s report of elbow pain in the self-administered questionnaire. The small group of occupational physicians in group B clearly had a lower threshold for diagnosing epicondylitis. The observer effect was considered as a confounding factor because different physicians were in charge of surveillance of workers in different activity sectors, which could imply specific biomechanical constraints. Two confounding factors were defined here: group A or B for medial epicondylitis (group B contained 3 of the 39 occupational physicians), group A or B for medial or lateral epicondylitis (5 physicians in group B).

Methods

Statistical Analysis Software (SAS) was used for the entire study.

The first portion of the analysis considered the data from the first questionnaire. Relations between medial epicondylitis and risk factors were studied by bivariate associations (chi-square and Fisher test). Gender, age, and the variables associated with medial epicondylitis at a P-level<0.20 were entered into a logistic model. A backward stepwise strategy was used, with gender and age kept in the model. When two risk factors were strongly associated, only one of them was retained in the logistic model.

We used polytomous logistic regression as the most appropriate analytic method for examining the difference in the risk factors for prevalence of medial and lateral epicondylitis. Three outcomes were compared: medial epicondylitis (n=49) lateral epicondylitis (n=183) and neither lateral nor medial epicondylitis (n=1506). The 19 subjects with both medial and lateral epicondylitis were not included in this model.

The first and the second questionnaires were used to estimate incidence. In the study of the predictive factors for medial epicondylitis, we considered only those risk factors measured at baseline.

Results

Description of the population (1993–1994)

Most subjects were women, except in the food industry. On average, assembly-line and clothing/shoe industry workers were the oldest, and supermarket cashiers the youngest. Clothing/shoe industry workers had the most years on the job, assembly-line workers the fewest. The highest proportion of smokers was found among cashiers and food industry workers. Body mass index did not differ between job sectors.

Forceful work was most frequent in the food industry (65% of the workers reported repetitive exposure to forceful work) and among cashiers (69%). On the other hand, those in the control group and those in the clothing/ shoe industry sector were less exposed to forceful work (respectively 44% and 25).

Results from the first questionnaire

The prevalence of medial epicondylitis was 3.8% (68/1757). Most cases were affected only on the right side (45.6%), compared with 19.1% on the left side and 35.3% bilaterally. The side affected by medial epicondylitis was not significantly associated with the worker’s handedness. Among the right-handed dominant cases, however, medial epicondylitis was unevenly distributed (47 on the right side, 14 on the left side), whereas the six cases among left-handed dominant workers were evenly distributed, 3 on the right side and 3 on the left.

Table 1 summarizes the relations between prevalence of medial epicondylitis and the potential risk factors.

Table 1.

Potential risk factors for prevalence of medial epicondylitis, 1993–1994, Odds ratio and 95%CI

	Number of subjects	Number of cases	Prevalence	OR [95% CI]
TOTAL	1757	68
Gender
Men	419	18	4.3	1
Women	1338	50	3.7	0.86 [0.50–1.50]
Age
<30 years	391	11	2.8	1
30–39 years	564	20	3.6	1.27 [0.60–2.68]
40–50 years	581	25	4.3	1.55 [0.76–3.19]
>50 years	221	12	5.4	1.98 [0.86–4.57]
Repetitive work
No	337	12	3.6	1
Yes	1420	56	3.9	1.11 [0.59–2.10]
Activity sector
Assembly line	479	25	5.2	1.49 [0.74–3.01]
Supermarket cashers	210	4	1.9	0.53 [0.17–1.65]
Clothing and shoes industry	264	7	2.7	0.74 [0.29–1.90]
Food industry	307	16	5.2	1.49 [0.69–3.20]
Packaging	160	4	2.5	0.69 [0.22–2.19]
Number of years on the job
<1	649	22	3.4	1
1 to 10	970	35	3.6	1.07 [0.62–1,84]
>10	138	11	8.0	2.47 [1.17–5.22]
Smoking
Non-smoker	580	15	2.6	1
Smoker	1042	48	4.6	1.82 [1.01–3.28]
Ex-smoker	135	5	3.7	1.45 [0.52–4.06]
BMI
Normal	1431	56	3.9	1
Overweight	200	7	3.5	0.89 [0.40–1.98]
Leisure activities
No	861	33	3.8	1
Yes	896	35	3.9	1.02 [0.63–1.66]
Presence of psychosomatic or depressive problem
No	944	31	3.3	1
Score 1 or 2	570	23	4.0	1.24 [0.72–2.15]
Score 3 to 8	243	14	5.8	1.80 [0.94–3.44]
Holding in position
No	599	24	4.0	1
Yes	1098	44	4.0	1.00 [0.60–1.66]
Turning and screwing
No	511	18	3.5	1
Yes	1156	50	4.3	1.24 [0.72–2.14]
Forceful work
No	861	22	2.6	1
Yes	896	46	5.1	2.06 [1.23–3.46]
Using tools forcefully
No	1469	53	3.6	1
Yes	288	15	5.2	1.47 [0.82–2.64]
Holding a tool in position
No	1289	43	3.3	1
Yes	468	25	5.3	1.64 [0.99–2.71]
Job control
Low level	1235	46	3.6	1
High level	454	22	4.6	1.30 [0.77–2.20]
	1281
Satisfaction at work
Yes	476	36	3.2	1
No	617	32	5.2	1.68 [1.03–2.73]

No association was found with repetitive work (OR=1.11). Four variables were associated with medial epicondylitis at a P-level of 0.20 or less and were considered, with gender and age, in the logistic model: number of years on the job, smoking, forceful work, and work satisfaction. “Forceful tool use” and “holding a tool in position” were not included because of their close relationship with “forceful work”.

The descending stepwise selection excluded smoking and satisfaction at work from the final model (at a P-level of 0.20 or less).

The final model, reported in Table 2, found no association between gender and medial epicondylitis, but found that the risk was associated with forceful work.

Table 2.

Results from the logistic model (1) for prevalence in 1993–1994

	OR [95% CI]	p=
Gender
Men	1	0.81
Women	1.07 [0.60–1.91]
Age
<30 years	1	0.72
30–39 years	1.32 [0.61–2.87]
40–50 years	1.43 [0.66–3.09]
>50 years	1.67 [0.68–4.08]
Number of years on the job
<1	1	0.17
1 to 10	1.10 [0.61–1.96]
>10	2.05 [0.93–4.51
Forceful work
No	1	0.01
Yes	1.95 [1.15–3.32]
Frequency	Cases =68	Total=1757

(1) Controlling for observer effect

Medial epicondylitis was strongly associated with other work-related musculoskeletal disorders (Table 3): 84% of those with medial epicondylitis suffered from another work-related musculoskeletal disorder, most often shoulder tendinitis, carpal tunnel syndrome or lateral epicondylitis. Ulnar nerve entrapment at the elbow was associated with medial epicondylitis, with a P-value on the borderline of significance (P=0.0518).

Table 3.

Medial epicondylitis in 1993–1994 and presence of other work-related musculoskeletal disorders (WRMD)

	At least one WRMD		Shoulder Pathology		Lateral epicondylitis		Carpal tunnel syndrome		Ulnar nerve entrapment at elbow
	number of cases	%	number of cases	%	number of cases	%	number of cases	%	number of cases	%
Medial epicondylitis
No (n=1689)	997	59.0	593	35.1	183	10.8	277	16.4	33	2.0
Yes (n=68)	57	83.8	37	54.4	19	27.9	20	29.4	4	5.9
Total	1054	60.0	630	35.9	202	11.5	297	16.9	37	2.1
Test (1)	p<0.001		p<0.005		p<0.001		p=0.005		p=0.052

(1) Chi-square tests, Fisher exact test for ulnar nerve entrapment at elbow

The risk factors considered in the polytomous logistic model (Table 4) were those in Table 2 and also “repetitive pressing with a hand”, psychosomatic or depressive symptoms (in two categories, score 0 to 2 versus score 3 to 8), and satisfaction at work, since these three factors were associated with the prevalence of lateral epicondylitis among the 1757 workers of the first questionnaire (18). The polytomous model indicated that lateral epicondylitis was associated with age (older), repetitive pressing with a hand and psychosomatic or depressive problems. “Forceful work” was the only risk factor significantly associated with medial epicondylitis. Nonetheless, the odds ratios of lateral and medial epicondylitis for this variable did not differ significantly.

Table 4.

Medial versus lateral epicondylitis in 1993–1994, comparison between risk factors. Results from the polytomous logistic regression(1)

	Lateral epicondylitis	Medial epicondylitis
Variables	OR [95% CI]	OR [95% CI]
Gender
Men	1	1
Women	0.99[0.67–1.46]	1.07[0.54–2.11]
Age
<30 years	1	1
30–39 years	1.43[0.82–2.48]	1.05[0.43–2.54]
40–50 years	1.97[1.16–3.35]	1.10[0.45–2.67]
>50 years	4.05[2.26–7.24]	1.52[0.54–4.32]
Number of years on the job
<1	1	1
1 to 10	1.10[0.77–1.56]	0.74[0.38–1.42]
>10	1.00[0.53–1.88]	1.08[0.39–3.02]
Forceful work
No	1	1
Yes	1.22[0.88–1.70]	1.91[1.03–3.55]
Pressing with the hand
No	1	1
Yes	1.55[1.12–2.15]	1.16[0.64–2.08]
Satisfaction at work
No	1	1
Yes	1.44[0.83–2.49]	1.41[0.53–3.70]
Presence of psychosomatic or depressive problems
Score 0 to 2	1	1
Score 3 to 8	1.63[1.08–2.46]	1.41[0.66–2.99]
Number of cases	183	49

(1) Controlling for observer effect

Results from the second questionnaire

In 1996–1997, the prevalence of medial epicondylitis was 5.2% (31/598). The incidence in three years, defined as the number of cases at the second questionnaire among those free from medial epicondylitis at the first questionnaire, was 4.5 (25 incident cases), which corresponds to an estimated annual incidence rate of 1.5%.

Of the workers with medial epicondylitis in 1993–1994, 81% had recovered by 1996–1997 (25/31). The frequency of change in job or workstation did not differ between those who had recovered and the rest of the sample.

We examined whether the risk factors associated with prevalence (listed in Table 1) were also related to incidence and found no significant associations. Incidence was similar among those exposed and not exposed to forceful work at baseline (4% and 5% respectively, p= 0.68).

Subjects with shoulder tendinitis, carpal tunnel syndrome, lateral epicondylitis or ulnar nerve entrapment at the elbow in 1993–94 developed medial epicondylitis significantly more often than the rest of the study population (relative risk 2.54; 95% CI [1.01–6.00]).

Discussion

This is the first study of medial epicondylitis in the workplace to take into account different kinds of risk factors and to use both cross-sectional and longitudinal approaches. Most epidemiological studies of occupational disease have not distinguished medial from lateral epicondylitis. While two cross-sectional surveys (Luopajarvi et al.²³, Ritz ²⁴) studied work-related epicondylitis and estimated the prevalence of medial epicondylitis, they did not study its risk factors. Several studies have focused on athletes ^6–10 and on prognosis, especially after surgery ^25–29.

The diagnostic criteria for medial epicondylitis are slightly different between studies ^23;30–34. The criteria in our study is close to that given in recent recommendations ³⁵.

The prevalence rate we found was close to those reported in the literature; Luopajarvi et al. ²³ found a prevalence rate of 3.2% in assembly-line workers exposed to repetitive tasks; Ritz ²⁴ reported a prevalence of 8.2% among gas- and water-work employees with high and moderate exposure to elbow-straining tasks. Andersen ³⁶ found no cases among 424 sewing machine operators, but this result was based on self-administered questionnaires in a survey focusing on neck and shoulder disorders.

The rate of recovery from medial epicondylitis was very high in our study (81% in three years), and recovery was not associated with a change in working conditions. Several sports physicians ^25–29 report that medial epicondylitis is cured in a few weeks with conservative treatment, and that surgery (proposed in resistant cases) also yields good results. These results suggest that the prognosis is good with an optimal treatment.

Medial epicondylitis was associated with other work-related upper-limb disorders, in particular, shoulder tendinitis, lateral epicondylitis, and carpal tunnel syndrome. The association with ulnar nerve entrapment is consistent with the literature ^25;37.

Although medial and lateral epicondylitis were associated, the polytomous logistic model suggests that they have different risk factors. Anatomical and physiological reasons make this highly plausible: medial and lateral epicondylian muscles are considered to be agonist-antagonists ³⁸. In specific movements, such as finger flexion, they work in synergy. They are also antagonists in most movements, including: turn and screw, wrist flexion or extension. Major utilization and a relative weakness of these muscles, relative to the medial ones, could explain the higher incidence of lateral epicondylitis. They should thus be studied separately, as two different disorders.

Our results indicate that medial, like lateral, epicondylitis is related to work in certain conditions. According to Bernard in 1997 ⁵, both types of epicondylitis are associated with work that causes physical strain. French authors ^{39, 40} also consider that working conditions play a role in medial epicondylitis, and this disorder is on the official list of occupational diseases in France ⁴¹. This approach is far from universal, however ^{42, 25}.

We found a relation between medial epicondylitis and forceful work. Some studies have not found any such relation between strenuous work and epicondylitis, without differentiating lateral from medial epicondylitis ^{1, 11, 30}.

Other studies have found associations with forceful work. As early as 1979, Luopajarvi et al.²³ studied the role of work in epicondylitis (medial and lateral) in a cross-sectional study comparing 152 female assembly-line workers with 133 female shop assistants: the risk was higher, albeit not statistically significant, among the assembly-line workers. Medial and lateral epicondylitis were distinguished, but the five cases of medial epicondylitis made specific analyses impossible.

Most recent studies conclude that epicondylitis is related to forceful work. In a 1991 cohort study in the food industry, Kurppa et al. ⁴³ concluded that epicondylitis was associated with strenuous work, with a relative risk estimate of 6.7 (CI: 3.3–13.9). In a cross-sectional study of 230 pork processing plant workers, Moore et al. ⁴⁴ showed that those with “hazardous jobs”, a definition that included force (but also posture, compression, vibration and cold), had more epicondylitis than workers in “safe jobs” (OR=5.5 CI 1.5–6.2). Ritz in 1995 ²⁴ performed a cross-sectional study among 290 male workers from the Hamburg public gas- and water-works and observed an increasing risk of epicondylitis for 10 years of exposure to a high level of elbow-straining work (OR =1.9 CI 1.2–3.1). The 1997 National Institute of Occupational Safety and Health (NIOSH) report ⁵ concluded that there is evidence of a relation between forceful work and epicondylitis.

We found no association between medial epicondylitis and repetitive work. This is in accordance with the NIOSH review ⁵, which concluded that there is insufficient evidence for support of an association between epicondylitis and repetitive work.

The results were limited by the cross-sectional design of the study: the workers who suffered from medial epicondylitis may tend to describe their work as strenuous. To examine the plausibility of such a recall bias, we looked at the job descriptions in the questionnaires of the cases classified as exposed to forceful work. This group included a variety of jobs, all of which probably required the use of strength at work.

Possible selection bias may have occurred due to two reasons: first, more occupational physicians from the firms with higher prevalence of upper-limb disorders participated in the follow-up component of the study; in addition, 102 workers were lost to follow-up. The consequence of the first selection effect, dealing with the participation of the physicians, was to increase the prevalence rate. However, it is not expected to have consequences on the conclusions regarding the role of the risk factors. The fact that a part of the workers were lost to follow up has probably minimal effects on the results, since the 102 workers lost to follow-up did not differ very much from the rest of the sample. The reasons for the limited health selection in this study have been discussed previously ¹⁹.

We chose to take an “observer effect” into account as a possible confounder. However, we verified that the result for forceful work did not depend on the presence of this factor: a model without an “observer effect” yielded nearly the same OR for forceful work (2.10; CI [1,24–3,54]).

We adjusted for gender, rather than performing separate analyses, because of the small number of cases. Some biomechanical factors that may be risk factors for medial epicondylitis, such as unscrewing, wrist flexion, and strenuous movements, could not be studied because they were not included in the questionnaire.

None of the biomechanical and psychosocial factors assessed at baseline was predictive of medial epicondylitis. This may be due to a lack of power, or to the relatively long time (3 years) between the two questionnaires, especially if short-term effects are expected. The predictive factor found here was the presence of other work–related upper-limb disorders (shoulder tendinitis, lateral epicondylitis, carpal tunnel syndrome and ulnar nerve entrapment at the elbow). The underlying mechanism may be that the first disorder induces unusual movements that in turn cause epicondylitis. A better understanding of the complex relations between several upper-limb disorders and strenuous work could be achieved by studies that follow exposed workers and record the occurrence of upper-limb disorders accurately.

In conclusion, medial epicondylitis, a disorder with high recovery rate, is frequently associated with other upper limb WRMD. The medical surveillance of workers performing repetitive work, especially those exposed to forceful movements, should include early identification of medial epicondylitis.