SPINE Volume 25, Number 16, pp 2126 –2132
©2000, Lippincott Williams & Wilkins, Inc.
Do Functional Tests Predict Low Back Pain?
Esa-Pekka Takala, DMSc, and Eira Viikari-Juntura, DMSc
Study Design. A cohort of 307 nonsymptomatic work-
ers and another cohort of 123 workers with previous ep-
isodes of low back pain were followed up for 2 years. The
outcomes were measured by symptoms, medical consul-
tations, and sick leaves due to low back disorders.
Objectives. To study the predictive value of a set of
tests measuring the physical performance of the back in a
working population. The hypothesis was that subjects
with poor functional capacity are liable to back disorders.
Summary of Background Data. Reduced functional
performance has been associated with back pain. There
are few data to show whether reduced functional capacity
is a cause or a consequence of pain.
Methods. Mobility of the trunk in forward and side
bending, maximal isokinetic trunk extension, flexion and
lifting strength, and static endurance of back extension
were measured. Standing balance and foot reaction time
were recorded with a force plate. Clinical tests for the
provocation of back or leg pain were performed. Gender,
workload, age, and anthropometrics were managed as
potential confounders in the analysis.
Results. Marked overlapping was seen in the mea-
sures of the subjects with different outcomes. Among the
nonsymptomatic subjects, low performance in tests of
mobility and standing balance was associated with future
back disorders. Among workers with previous episodes
of back pain, low isokinetic extension strength, poor
standing balance, and positive clinical signs predicted
future pain.
Conclusions. Some associations were found between
the functional tests and future low back pain. The wide
variation in the results questions the value of the tests in
health examinations (e.g., in screening or surveillance of
low back disorders). [Key words: low back pain, muscu-
loskeletal physiology, occupational health, prospective
studies, range of motion] Spine 2000;25:2126 –2132
Tests measuring the functional performance of the low
back have been widely used among patients with low
back pain.16 Some tests have been advocated for the
screening or surveillance of low back disorders among
working populations.11,17,18 The results concerning the
benefits of these tests in screening have been contra-
dicted, however.4,7,12
The purpose of the study was to determine the predic-
tive value of results of some tests measuring physical
performance of the back in a working population. The
hypothesis was that subjects with poor functional capac-
ity risk development of low back disorders. The authors
From the Department of Physiology, Musculoskeletal Research Unit,
Finnish Institute of Occupational Health, Helsinki, Finland.
Supported by the Finnish Work Environment Fund.
Acknowledgment date: May 12, 1999.
First revision date: September 27, 1999.
Acceptance date: December 22, 1999.
Device status category: 10.
Conflict of interest category: 14.
2126
hypothesized further that among subjects with existing
disorders, low functional capacity predicts an adverse
outcome.
Methods
Study Design. A set of tests describing the functional capacity
of the musculoskeletal system was performed on two cohorts of
workers: 307 nonsymptomatic workers and 123 workers who
had had low back pain for at least in 30 days during the previ-
ous 12 months. The selection was based on the reported total
number of days with symptoms during the previous 12 months
and workload. The outcomes were the status of low back
symptoms after 1 and 2 years determined by questionnaire, and
medical consultations and sick leaves due to low back pain
during the follow-up of 2 years. In the nonsymptomatic cohort,
the workers who contracted pain during the follow-up were
compared with the consistently pain-free workers. In the co-
hort with previous episodes of back pain, the workers who
persistently reported pain in the follow-up were compared with
those who reported recovery of symptoms. In both cohorts,
patients who had had medical consultations or sick leave were
compared with those who had not. The predictors were the
measurements of function.
Subjects. The sample was recruited to provide representative
groups of symptomatic and nonsymptomatic subjects and even
distributions of potential confounding factors (workload, gen-
der, and age). The main selection criteria were duration of pain
in the low back and workload. The source population consisted
of 6848 workers who participated in March 1992 in a ques-
tionnaire survey in a forest industry enterprise. Postal surveys
were repeated in March of 1993, 1994, and 1995.
Based on the results of the surveys in 1992 and 1993, sub-
jects who fulfilled the study criteria were invited to participate
in laboratory tests between February and August 1993. Ex-
cluded from the study were subjects in whom the tests were
considered to involve a risk to health (e.g., presence of chronic
cardiopulmonary disease, infection, serious back pain, preg-
nancy) and subjects with disease that might have had an effect
on the test results (e.g., presence of neurologic disease, inflam-
matory joint disease). Subjects older than 54 years were
excluded to avoid withdrawal due to retirement during the
follow-up.
Cumulative number of days with pain in the low back dur-
ing the preceding 12 months, classified in five categories, was
determined by a modified version of the Nordic question-
naire.19 Three types of low back pain were considered: radiat-
ing pain or numbness below the knee, sudden attacks of pain,
and other kinds of low back pain. In the analysis the greatest
number of days with any of these symptoms during the preced-
ing 12 months represented the low back pain of the subject.
The subject was classified as nonsymptomatic if the total
number of days with low back pain during the preceding 12
months was fewer than 8 days and symptomatic if the number
of days with pain exceeded 30 days. The physical workload

Table 1. Number of Subjects By Pain in the Lower Back in 1993, 1994, and 1995 (A), and Formation of Groups for
1 and 2 Years Follow-Up (B)
A
Pain in 1994
Pain in 1993 ⬍8 d 8–30 d ⬎30 d
⬍8 d 228 31 9 39
⬎30 d 16 29 65
Total 244 60 74
1 Year Follow-Up
“Soft” Criteria
“Consistently painfree” 228
“Contracted pain” 40 (⫽31 ⫹ 9)
“Recovered” 45 (⫽16 ⫹ 29)
“Persistent pain” 65
was classified according to job title as light (office work and
sedentary tasks) or moderately strenuous (cleaning, various
tasks at the paper machine, maintenance tasks). The nonsymp-
tomatic cohort included 49 men in light work and 176 in mod-
erately strenuous jobs and 49 women in light work and 33 in
moderately strenuous jobs. The symptomatic cohort comprised
9 men in light work and 68 in moderately strenuous jobs and
20 women in light work and 26 in moderately strenuous jobs.
Outcome.
Questionnaire Data. The outcome was based on the re-
sponses in 1994 and 1995 compared with those in 1993 (i.e.,
1-year and 2-year follow-up; Table 1). In retrospective ques-
tions, misclassification bias would most likely be between ad-
jacent classes of duration of pain. To reduce the effects of this
potential bias, two kinds of criteria for the changes in the fol-
low-up were formed
According to the “soft” criteria, any change from the base-
line was noted. Nonsymptomatic subjects in the beginning of
the study were classified as consistently pain free if they had
had pain during fewer than 8 days in the follow-up and as
having contracted pain if they had had pain during 8 or more
days. Similarly, symptomatic subjects in the beginning of the
study were classified as having persistent pain if in the fol-
low-up they had pain during more than 30 days and as recov-
ered if the number of days with pain was 30 or fewer.
According to “strict” criteria, only changes greater than a
move to the adjacent class were considered. The subjects with
no change during the follow-up (consistently pain-free and per-
sistent pain) were the same as in the soft classification. Non-
symptomatic subjects in the beginning of the study were clas-
sified as “contracted pain” if the total number of days with pain
was at least 30 in the follow-up. Subjects in the beginning of the
study were classified as recovered if the number of days was
fewer than 8.
Of the potential confounders, the distributions of sex,
workload, stature, and weight were similar in the nonsymp-
tomatic and symptomatic groups. The mean age of the symp-
tomatic subjects was approximately 3 years more than that of
the nonsymptomatic group.
Efficacy of Functional Tests • Takala and Viikari-Juntura 2127
Pain in 1995
Missing ⬍8 d 8–30 d ⬎30 d Missing
214 41 12 40
13 21 22 53 27
52 235 63 65 67
2 Year Follow-Up
“Strict” Criteria “Soft” Criteria “Strict” Criteria
228 214 214
9 53 (⫽41 ⫹ 12) 12
16 43 (⫽21 ⫹ 22) 21
65 53 53
Medical Files of the Occupational Health Service. All sick
leaves and medical consultations not leading to sick leave were
registered in the computerized medical files of the occupational
health service of the company with the diagnosis classified ac-
cording to the International Classification of Diseases.37 If a
medical physician outside the company prescribed a sick leave,
it was later registered in the files. The data on sick leaves and
medical consultations were collected in April 1995. The me-
dian follow-up time was 23.4 months. During this time, 93
workers had a low back disorder diagnosed (40 in the non-
symptomatic and 53 in the symptomatic baseline group), and
73 of them had been on sick leave due to a low back disorder
(34 in the nonsymptomatic and 39 in the symptomatic baseline
group).
Functional Tests. Mobility of the trunk in forward flexion
was measured with a tape measure6 and by determining the
flexion angles of the lumbar spine and pelvic tilt with two
goniometers (MIE, Medical Research, Ltd., London, UK).29
Mobility in side bending was measured by determining the
distance the subject could slide the fingertips down the lateral
aspect of the leg.1
Maximal isokinetic trunk extension and flexion strength
were measured by a dynamometer (Cybex 6000 TEF; Lumex,
Ronkonkoma, NY)23 with two speeds, 30 and 90 deg/sec. Iso-
kinetic maximal lifting force was measured with an apparatus
consisting of a horizontal handlebar connected with a nonelas-
tic rope through a hole in the supporting platform to a servo-
controlled electric engine below. The subject lifted the handle-
bar with both hands starting from the ankle level with bent
knees and terminating with erect body and straight arms above
the head. The subject repeated two lifts with two speeds, 0.5
and 1.0 m/sec. Maximal absolute force and maximal average
force from the ankle level to the hip level with both speeds were
computed. Static endurance of back extension was tested ac-
cording to the method of Biering-Sørensen.6
Coordination of the lower body was measured as standing
balance and foot reaction time with a force plate.34 Amplitude
and velocity of sway in two-footed and one-footed stance with
eyes open and closed were recorded. The time of achieving
2128 Spine • Volume 25 • Number 16 • 2000

Table 2. Predictors That Differed Between the Contrasted Groups “Consistently Painfree” Vs. “Contracted Pain” in the
Univariate Analysis, Classification With “Soft” Criteria. Mean Values (SD in Parentheses) of Continuous Variables and
Number of Subjects for Categorical Items
Variables “Consistently Painfree” “Contracted Pain” P

Men
1 Year Follow-Up
N 165 28
Age, years 40.2 (7.9) 41.9 (7.2) NS
Pelvic movement in forward flexion,° 65 (13) 60 (13) 0.08
Static endurance time, s 131 (61) 151 (59) 0.09
Pain in side bending, hip flexion, SLR, sit-up, or passive extension 11/165 7/28 0.002
2 Year Follow-Up
N 154 38
Age, years 39.4 (7.9) 42.4 (7.8) 0.03
Maximal lifting force at 1.0 m/s, N 947 (186) 870 (186) 0.09
(n ⫽ 109) (n ⫽ 22)
Pain in side bending, hip flexion, SLR, sit-up, or passive extension 10/154 7/38 0.02

Women
1 Year Follow-Up
N 60 12
Age, years 42.3 (7.0) 48.4 (5.1) 0.006
Mean amplitude in the two feet standing balance with eyes closed, mm 6.4 (2.1) 7.8 (2.6) 0.02
Difference between the feet in step test: backward movement time, s 0.18 (0.18) 0.36 (0.12) 0.02*
2 Year Follow-Up
N 58 15
Age, years 43.2 (7.2) 44.4 (6.7) NS
Mean amplitude in the 2 feet standing balance with eyes closed, mm 6.3 (2.1) 8.2 (2.6) 0.004
*Tested with log transformation of the variable.
P ⫽ P-values in the ANCOVA with age, workload, and anthropometry as covariates, or in the ␹2 test for categorical variables.
NS: P ⬎ 0.10.
SLR ⫽ straight leg raising; s ⫽ second; ° ⫽ degree.

balance after a step movement (backward movement time) was
measured in a special test measuring balance and reaction in a
dynamic situation.34
A set of clinical tests was performed that produce results
with known predictive value: straight leg raising (SLR) for the
provocation of back or leg pain, ability to perform a sit-up,
resisted hip flexion, and passive extension of the back.22,35 Pain
in the low back during forward and side bending of the trunk
was recorded.
sistently pain-free subjects (Table 2). On the contrary,
the women with medical consultations had more mobil-
ity than those who had no consultation (Table 3). Sys-
tematic differences between the groups were seen in the
tests of balance and coordination but not in muscular
performance. The men who contracted pain in the 1 year
follow-up had even better results in the static endurance
test than those who did not contract symptoms.

Statistical Analysis. The contracted pain group was com-
pared with the consistently pain-free group and the persistent
pain group with the recovered group. Differences between the
groups were studied by analysis of covariance (ANCOVA)
with sex, age, workload, and anthropometrics as covariates.
Categorical variables were investigated by ␹2 test.
The variables that had shown the largest differences be-
tween the groups in the first analysis (P ⬍ 0.10) were included
in logistic regression models. Only the variables statistically
significant at P ⬍ 0.05 were included in the final model. The
time until the first medical consultation or sick leave after the
measurements was studied by survival analysis with Cox’s pro-
portional hazards regression models.
Results
There were several differences among the contrasted
groups. With application of strict criteria, the differences
were generally of the same direction as those with appli-
cation of soft criteria.
Nonsymptomatic Subjects
Among the men, those who contracted symptoms during
follow-up had generally less mobility than did the con-
Logistic Regression. In the 1 year follow-up with appli-
cation of soft criteria, age and clinical findings were the
only statistically significant predictors of future low back
pain in the men. In the women, age and difference be-
tween the feet in backward movement time in the dy-
namic step test were included in the final model. With
application of strict criteria, the final model included age,
workload, spinal movement, and difference between feet
in the step test. In the 2-year follow-up, trends similar to
those of the first year were seen but the associations were
weaker.
Lumbar spine–pelvic movement ratio and clinical
findings were associated with medical consultations
among the men. Among the women only one variable
describing balance (difference in sway velocity between
feet during one-footed stance) showed constant statisti-
cal significance. The lumbar spine–pelvic movement ra-
tio and the step test (difference between the feet in back-
ward movement time) were associated with future sick
leaves in the men, but no combination of the predictors
was statistically significant among the women.
 Efficacy of Functional Tests • Takala and Viikari-Juntura 2129

Table 3. Predictors That Differed in the Univariate Analysis When Medical Consultations Were Used as Outcome.
Nonsymptomatic Subjects in the Beginning of the Study. Mean Values (SD in Parentheses) of Continuous Variables
and Number of Subjects for Categorical Items
Medical Consultation

Variables No Yes P

Men
N 194 29
Age, years 39.6 (7.8) 42.9 (8.1) 0.04
Movement of the lumbar spine in forward flexion,° 49 (9) 46 (9) 0.08
Lumbar spine/pelvic movement ratio 0.81 (0.25) 0.70 (0.17) 0.03
Difference in mean amplitude between feet in one foot stance, mm 0.12 (0.09) 0.15 (0.12) 0.05
Pain in side bending, hip flexion, SLR, sit-up, or passive extension 16/194 6/29 0.04

Women
N 70 11
Age, years 43.4 (7.1) 46.0 (5.9) NS
Pelvic movement in forward flexion,° 73 (16) 82 (11) 0.04
Total movement in forward flexion,° 119 (18) 122 (11) 0.09
Difference in sway velocity between feet in one foot stance, mm/s 23 (21) 38 (22) 0.02
P ⫽ P-values in the ANCOVA with age, workload, and anthropometry as covariates, or in the ␹2 test for categorical variables.
NS: P ⬎ 0.10.
SLR ⫽ straight leg raising; ° ⫽ degree.

The models classified correctly 95% to 100% of the
consistently healthy subjects but only 0% to 33% of the
subjects who contracted pain, had medical consulta-
tions, or took sick leaves.
Survival Analysis. Among the men lower lumbar spine–
pelvic movement ratio, greater difference in the ampli-
tude between feet in the balance test, and lower power in
maximal flexion test with 90-deg/sec speed predicted
shorter time until the first consultation after the func-
tional tests (P ⫽ 0.001). Among the women a greater
difference in balance sway velocity between feet pre-
dicted a shorter time until the first consultation (P ⫽
0.025).
Symptomatic Subjects
Mobility was not systematically associated with the per-
sistence of symptoms (Tables 4 and 5). Men with persis-
tent pain had better results in the static endurance test,
and the women were stronger in the trunk flexion test

Table 4. Predictors That Differed Between the Contrasted Groups “Persistent Pain” Vs. “Recovered” in the Univariate
Analysis, Classification with “Soft” Criteria. Mean Values (SD in Parentheses) of Continuous Variables and Number of
Subjects for Categorical Items
Variables “Recovered” “Persistent Pain” P

Men
1 Year Follow-Up
N 28 40
Age, years 40.7 (6.7) 45.0 (6.0) 0.008
Maximal side bending, mean of left and right, mm 183 (46) 189 (41) 0.05
Static endurance time, s 94 (61) 108 (66) 0.08
Difference in sway velocity between feet in one foot stance, mm/s 32 (29) 44 (27) 0.05
2 Year Follow-Up
N 29 31
Age, years 40.7 (7.1) 46.5 (4.5) 0.0004
Mean amplitude in the two feet standing balance with eyes open, mm 4.5 (0.8) 4.1 (1.2) 0.07

Women
1 Year Follow-Up
N 16 25
Age, years 42.7 (6.8) 45.3 (6.4) NS
Total work of maximal flexion 30°/s, J 141 (32) 156 (21) 0.05
Average power in maximal flexion 30°/s, W 46 (11) 51 (7) 0.03
Mean amplitude in the two feet standing balance with eyes closed, mm 6.8 (1.6) 8.2 (2.4) 0.06
Sway velocity in the two feet standing balance with eyes closed, mm/s 23 (6) 27 (9) 0.10
2 Year Follow-Up
N 14 22
Age, years 43.2 (6.0) 45.9 (6.7) NS
Extension/flexion force ratio with 30°/s 0.80 (0.19) 0.90 (0.26) 0.08
P ⫽ P-values in the ANCOVA with age, workload, and anthropometry as covariates, or in the ␹2 test for categorical variables.
NS: P ⬎ 0.10; s ⫽ second; J ⫽ joole; W ⫽ watt.
2130 Spine • Volume 25 • Number 16 • 2000

Table 5. Predictors That Differed in the Univariate Analysis When Medical Consultations were Used as Outcome.
Symptomatic Subjects in the Beginning of the Study. Mean Values (SD in Parentheses) of Continuous Variables and
Number of Subjects for Categorical Items
Medical Consultation

Variables No Yes P

Men
N 38 39
Age, years 41.8 (6.6) 43.7 (6.7) NS
Extension/flexion force ratio with 30°/s 1.07 (0.23) 0.91 (0.19) 0.01
Extension/flexion total work ratio with 30°/s 1.01 (0.20) 0.91 (0.19) 0.04
Extension/flexion average power ratio with 30°/s 1.00 (0.22) 0.89 (0.20) 0.04
Difference in mean amplitude between feet in one foot stance, mm 0.09 (0.07) 0.15 (0.12) 0.01
Pain in side bending, hip flexion, SLR, sit-up, or passive extension 15/38 28/39 0.004

Women
N 31 14
Age, years 43.7 (7.1) 44.9 (5.2) NS
Maximal flexion force, 90°/s, Nm 140 (26) 119 (43) 0.05
Pain in resisted hip flexion 4/31 6/14 0.03
Pain in SLR 3/31 5/14 0.03
Pain in resisted hip flexion or in SLR 5/31 7/14 0.02
P ⫽ P-values in the ANCOVA with age, workload, and anthropometry as covariates, or in the ␹2 test for categorical variables.
NS: P ⬎ 0.10.
SLR ⫽ straight leg raising.

than those who recovered. Low extension–flexion ratio
was associated with medical consultations and sick
leaves. The recovered workers had generally better mea-
sures of balance and coordination than the workers with
persistent symptoms. These associations were even
stronger when medical consultations and sick leaves
were used as outcomes. Clinical tests provoking low
back pain were associated with the persistence of symp-
toms, medical consultations, and sick leaves in both
genders.
Logistic regression. Age was the only variable statisti-
cally significantly associated with the persistence of
symptoms. Clinical findings showed associations with
medical consultations: pain in side bending, resisted hip
flexion, SLR, sit-up, or passive extension of back among
the men; pain in resisted hip flexion or SLR among the
women. Low extension–flexion force ratio was the only
predictor of sick leaves among the men and pain in re-
sisted hip flexion among the women.
Survival Analysis. A shorter time until the first medical
consultation was predicted by the lower extension–
flexion force ratio (P ⫽ 0.002) among the men and by
pain in side bending (P ⫽ 0.001), in resisted hip flexion
(P ⫽ 0.003), or in SLR (P ⫽ 0.005) among the women.
Discussion
In the occupational health service, the focus of actions is
on prevention of diseases. Measures for screening and
surveillance of diseases should have predictive value.
The current study was designed to test the value of
results of functional performance tests in the predic-
tion of low back pain among nonsymptomatic work-
ers and among those who had symptoms but were
working despite them.
Patients with low back pain often have reduced mo-
bility of the spine. In this study of nonpatient population,
limited mobility of the spine was associated with future
back problems among men who had been nonsymptom-
atic in the beginning of the study. In contrast, nonsymp-
tomatic women with future medical consultations had
more mobile backs in comparison with those who did
not seek medical advice. These controversial findings and
the wide variation in the measures limit the predictive
value of the measurements of back mobility and their use
for screening purposes.5
Isometric lifting strength did not predict back injury
claims in the prospective Boeing study,3 but isometric
lifting capacity in relation to the lifting requirements of
the job has shown an association with back injuries.11,21
Isokinetic strength measures have been recommended,
although their validity has been questioned.13,15,27,30 –32
In the current study, low isokinetic extension–flexion
strength ratio was associated with the outcomes but only
in the previously symptomatic subjects. This finding
should be interpreted with caution. In extension, the
back muscles work against the weight of the upper body,
and in flexion the gravity erroneously increases the mea-
sured output of muscle force. The manufacturer of the
dynamometer (Lumex) does not provide information
about the principle that governs the calculation of out-
put, and therefore the authors could not make biome-
chanical corrections according to the the weight of the
upper body. The results do not support the use of isoki-
netic measurements as a screening tool in physically light
or moderately strenuous work. Neither did short endur-
ance time in the static back extension test predict future
low back pain, although such associations have been
reported in two previous studies.6,25
 Efficacy of Functional Tests • Takala and Viikari-Juntura 2131
Patients with low back pain have been reported to
have low performance in balance tests.2,10,24,26 The as-
sociation between the measurements of balance and co-
ordination and future low back symptoms in a working
population is a new finding. Sudden loss of postural bal-
ance is common in daily work, and it causes corrective
muscular reactions. It can be hypothesized that the cor-
rective pendulous movements of subjects with poor mus-
cular control have a larger deviation from neutral joint
postures than those of subjects with good control. These
larger deviations result in greater lever arms and greater
instantaneous load moments on the lower back. Some of
these mechanical impulses could be strong enough to
cause microscopic traumas within the tissues. Therefore,
the frequency of potentially injurious accidents and even
losses of balance may be greater among subjects with
poor muscular control than among those with good con-
trol. Another explanation for our finding is that absence
of coordination may make people unwilling or less capa-
ble of training their musculoskeletal system.
The clinical tests for the provocation of pain had the
greatest power to predict persistent pain. Although these
findings were scanty among the originally nonsymptom-
atic subjects, positive associations were seen with future
pain. The current results support the view that these sim-
ple clinical tests are useful in the examination of subjects
with back pain in primary health care.4,35,36
One reason for the contradictory results in different
studies may be the differences in outcomes. Some of the
studies have used simple dichotomized questions, such as
the prevalence of back pain during the past 12
months6,25 or claims of back injuries.3,5,7 Two indepen-
dent sources of information and several classifications
were used in the current study to determine outcomes.
Although the current study had a prospective design, the
questionnaires used dealt with retrospective health out-
comes. To reduce misclassification due to the recall bias,
the authors used two levels of outcome criteria. The med-
ical records, however, were prospective outcomes.
The outcome based on questionnaires resulted in a
classification with a broad spectrum of possible patho-
logic conditions causing back pain. The current study
population was too small to analyze associations be-
tween functional performance and outcomes formed
separately for the three different types of symptoms of
the low back (radiating pain, sudden attacks of pain,
other kinds of pain). In addition, the manner of asking
for the cumulative number of days with symptoms has
limitations. For instance, the prognosis of a single 8-day
bout of pain may be different from that of eight short-
duration intervals of pain during the year. Medical con-
sultations and sick leaves are more relevant, however,
than subjective symptoms from the practical point of
view, because they measure the costs of low back pain.
Psychosocial factors have been reported as predictors
of low back disorders in several studies.8,20,28,33 Several
psychological items were also assessed in the current
study population.14 These items did not show enough
statistical significance to be included into the final mul-
tivariate models when low back disorders were out-
comes.
The most probable explanation for the current results
is that so many additional intervening factors affect the
course of low back pain during a long follow-up period
that single measurements have little significance in an
individual outcome. The onset and course of low back
pain is a dynamic process with numerous interactions
between the external and individual factors over time.9
In conclusion, the results did not contradict the au-
thors’ hypothesis that persons with low functional ca-
pacity are liable to low back disorders and that those
with existing disorders have an adverse outcome if their
functional capacity is poor. The associations were weak,
however, and some findings of muscle strength were even
inverse to the hypothesis. The power to identify individ-
uals with future low back pain by results of the tests
studied was poor in a nonpatient population.
Key Points
● Predictive value of tests measuring the physical
performance of the back in a working population
was studied.
● Low performance in mobility, standing balance,
extension strength, and positive clinical signs pre-
dicted future pain.
● The wide variation in the results questions the
value of the tests in health examinations.
Acknowledgments
The authors thank Marja-Riitta Suikki, Aija Moilanen,
Sirpa Rauas, and Dr. Sirpa Lusa-Moser for help in per-
forming the measurements.
References
1. Alaranta H, Hurri H, Heliövaara M, Soukka A, Harju R. Flexibility of the
spine: Normative values of goniometric and tape measurements. Scand J Rehabil
Med 1994;26:147–54.
2. Alaranta H, Moffroid M, Elmqvist L-G, Held J, Pope M, Renstrom P. Pos-
tural control of adults with musculoskeletal impairment. Crit Rev Phys Rehabil
Med 1994;6:337–70.
3. Battié MC, Bigos SJ, Fisher LD, Hansson TH, Jones ME, Wortley MD.
Isometric lifting strength as a predictor of industrial back pain reports. Spine
1989;14:851– 6.
4. Battié MC, Bigos SJ, Fisher LD, et al. Anthropometric and clinical measures
as predictors of back pain complaints in industry: A prospective study. J Spinal
Disord 1990;3:195–204.
5. Battié MC, Bigos SJ, Fisher LD, et al. The role of spinal flexibility in back
pain complaints within industry: A prospective study. Spine 1990;15:768 –73.
6. Biering-Sørensen F. Physical measurements as risk indicators for low-back
trouble over a one-year period. Spine 1984;9:106 –19.
7. Bigos SJ, Battié MC, Fisher LD, Hansson TH, Spengler DM, Nachemson AL.
A prospective evaluation of preemployment screening methods for acute indus-
trial back pain. Spine 1992;17:922– 6.
8. Bigos SJ, Battié MC, Spengler DM, et al. A prospective study of work per-
ceptions and psychosocial factors affecting the report of back injury. Spine 1991;
16:1– 6.
9. Burdorf A, Rossignol M, Fathallah FA, Snook SH, Herrick RF. Challenges in
assessing risk factors in epidemiologic studies on back disorders. Am J Ind Med
1997;32:142–52.
10. Byl NN, Sinnott PL. Variations in balance and body sway in middle-aged
2132 Spine • Volume 25 • Number 16 • 2000
adults: Subjects with healthy backs compared with subjects with low-back dys-
function. Spine 1991;16:325–30.
11. Chaffin DB, Herrin GD, Keyserling WM. Preemployment strength testing:
An updated position. J Occup Med 1978;20:403– 8.
12. Dempsey PG, Burdorf A, Webster BS. The influence of personal variables on
work-related low-back disorders and implications for future research. J Occup
Environ Med 1997;39:748 –59.
13. Dueker JA, Ritchie SM, Knox TJ, Rose SJ. Isokinetic trunk testing and
employment. J Occup Med 1994;36:42– 8.
14. Estlander A-M, Takala E-P, Viikari-Juntura E. Do psychological factors
predict changes in musculoskeletal pain? A prospective, two-year follow-up
study of a working population. J Occup Environ Med 1998;40:445–53.
15. Gibbons L, Videman T, Battié M. Isokinetic and psychophysical lifting
strength, static back muscle endurance, and magnetic resonance imaging of the
paraspinal muscles as predictors of low back pain in men. Scand J Rehabil Med
1997;29:187–91.
16. Keeley J. Quantification of function. In: Mayer TG, Mooney V, Gatchel RJ,
eds. Contemporary Conservative Care for Painful Spinal Disorders. Philadelphia:
Lea & Febiger; 1991:290 –307.
17. Keyserling WM, Herrin GD, Chaffin DB. Isometric strength testing as a
means of controlling medical incidents on strenuous jobs. J Occup Med 1980;
22:332– 6.
18. Keyserling WM, Herrin GD, Chaffin DB, Armstrong TJ, Foss ML. Establish-
ing an industrial strength testing program. Am Ind Hyg Assoc J 1980;41:730 – 6.
19. Kuorinka I, Jonsson B, Kilbom Å, et al. Standardised Nordic questionnaires
for the analysis of musculoskeletal symptoms. Appl Ergonom 1987;18:233–7.
20. Leino PI, Hänninen V. Psychosocial factors at work in relation to back and
limb disorders. Scand J Work Environ Health 1995;21:134 – 42.
21. Liles DH, Deivanayagam S, Ayoub MM, Mahajan P. A job severity index for
the evaluation and control of lifting injury. Hum Factors 1984;26:683–93.
22. Lloyd DCEF, Troup JDG. Recurrent back pain and its prediction. J Soc
Occup Med 1983;33:66 –74.
23. Lumex. Cybex 6000 Testing and Rehabilitation System. User’s guide.
Ronkonkoma, NY: Lumex, Inc; 1992.
24. Luoto S, Aalto H, Taimela S, Hurri H, Pyykkö I, Alaranta H. One-footed and
externally disturbed two-footed postural control in patients with chronic low
back pain and healthy control subjects: A controlled study with follow-up. Spine
1998;23:2081–9.
25. Luoto S, Heliövaara M, Hurri H, Alaranta H. Static back endurance and the
risk of low-back pain. Clin Biomech 1995;10:323– 4.
26. Luoto S, Taimela S, Hurri H, Aalto H, Pyykko I, Alaranta H. Psychomotor
speed and postural control in chronic low back pain patients: A controlled fol-
low-up study. Spine 1996;21:2621–7.
27. Madsen OR. Trunk extensor and flexor strength measured by the Cybex
6000 dynamometer: Assessment of short-term and long-term reproducibility of
several strength variables. Spine 1996;21:2770 – 6.
28. Mannion AF, Dolan P, Adams MA. Psychological questionnaires: Do
“abnormal” scores precede or follow first-time low back pain? Spine 1996;21:
2603–11.
29. Mayer TG, Tencer AF, Kristoferson S, Mooney V. Use of noninvasive tech-
niques for quantification of spinal range-of-motion in normal subjects and
chronic low-back dysfunction patients. Spine 1984;9:588 –95.
30. Mostardi RA, Noe DA, Kovacik MW, Porterfield JA. Isokinetic lifting
strength and occupational injury A prospective study. Spine 1992;17:189 –93.
31. Newton M, Thow M, Somerville D, Henderson I, Waddell G. Trunk
strength testing with iso-machines. Part 2: Experimental evaluation of the Cybex
II back testing system in normal subjects and patients with chronic low back pain.
Spine 1993;18:812–24.
32. Newton M, Waddell G. Trunk strength testing with iso-machines. Part 1:
Review of a decade of scientific evidence. Spine 1993;18:801–11.
33. Papageorgiou AC, Macfarlane GJ, Thomas E, Croft PR, Jayson MIV, Silman
AJ. Psychosocial factors in the workplace: Do they predict new episodes of low
back pain? Evidence from the South Manchester Back Pain Study. Spine 1997;
22:1137– 42.
34. Takala E-P, Korhonen I, Viikari-Juntura E. Postural sway and stepping re-
sponse among a working population: Reproducibility, long-term stability, and
associations with symptoms of the low back. Clin Biomech 1997;12:429 –37.
35. Troup JDG, Martin JW, Lloyd DCEF. Back pain in industry: A prospective
survey. Spine 1981;6:61–9.
36. Viikari-Juntura E, Takala E-P, Riihimäki H, Malmivaara A, Martikainen R,
Jäppinen P. Standardized physical examination protocol for low back disorders:
Feasibility of use and validity of symptoms and signs. J Clin Epidemiol 1998;51:
245–55.
37. World Health Organization. Manual of the International Statistical Classi-
fication of Diseases, Injuries, and Causes of Death. 9th ed. Geneva: WHO, 1977.
Address reprint requests to
Esa-Pekka Takala
Finnish Institute of Occupational Health
Topeliuksenkatu 41 a A
FIN-00250 Helsinki, Finland
E-mail: esa-pekka.takala@occuphealth.fi