J. Biomedical Science and Engineering, 2010, 3, 664-671 JBiSE
doi:10.4236/jbise.2010.37090 Published Online July 2010 (http://www.SciRP.org/journal/jbise/).
Published Online July 2010 in SciRes. http://www.scirp.org/journal/jbise
Analysis of spine loads in dentistry—impact of an altered
sitting position of the dentist
Max Wunderlich1, Thomas Eger2, Thomas Rüther1, Andreas Meyer-Falcke3, Dieter Leyk1,4
1German Sport University Cologne, Department of Physiology and Anatomy, Cologne, Germany;
2Department of Periodontology, German Armed Forces Central Hospital, Koblenz, Germany;
3Health Care Campus North Rhine-Westphalia, Universitätsstraße 136, Bochum, Germany;
4Department IV - Military Ergonomics and Exercise Physiology, German Armed Forces, Koblenz, Germany.
Received 25 March 2010; revised 6 April 2010; accepted 8 April 2010.
ABSTRACT
Neck and low back pain causes highest disability
rates in industrialized countries. Apart from blue-
collar occupations dentists and dental care personnel
are also strongly affected by work related back pain.
However, due to missing analysing systems spine ex-
posure could not be adequately estimated for sitting
and non load bearing tasks. Therefore, a new biome-
chanical measurement system was developed to ana-
lyze spine and trunk kinematics. To explore the po-
tential kinematics impact on the spine two days of
periodontology care have been examined. Moreover,
to asses the influence of an altered sitting position
data was collected in 9 (day 1) and 12 (day 2) o’clock
position. Data were collected and analyzed by the
new 3D-SpineMoveGuard. The systems analyze pos-
tures and movements of spine and trunk by ultra-
sound and inclinometer. Two days of routine non-
surgical periodontal treatment were recorded by a
frequency of 10 Hz. Moderate awkward postures
could be assessed for 90% of the treatment. Up to
85% of the working days were spent in isometric po-
sition. Movements were rare and of small amplitude.
There were no relevant differences between 9 and 12
o’clock position. The frequently reported back pain
in dentists is not only related to severe body postures
rather than to the high amount of isometric spine
loads. Moreover, an altered sitting position did not
reduce this physiological stress. Therefore, dentist
specific interventions concepts should pay more at-
tention to physical training during and off work. Er-
gonomic investments should be carefully evaluated
before implemented.
Keywords: Occupational Medicine; Periodontal Treat-
ment; Back Pain, Ergonomics; 3D-SpineMoveGuard
1. INTRODUCTION
About 80% of the adult population of industrialized
countries are affected by at least some episode of back
pain [1,2]. The highest prevalence rates can be esti-
mated for employees in the “mid-ages”. Moreover, the
major economic burden results from the high disability
rates and the persistent recurrence of the preliminary
cervical or lumbar pain [3,4]. As a large part of these
disabilities are caused by specific job demands preven-
tive strategies have to focus on the reduction of occu-
pational stressors to avoid chronic episodes of back
pain.
The most common causes for occupation related back
pain are biomechanics impacts. Beside the negative ef-
fects of load bearing tasks, there is evidence from epide-
miological studies that awkward trunk postures (bending,
twisting, isometrics, repetitive movements) are the most
reasonable cause for work related back pain [5,6]. Fur-
thermore, numerous of psycho-social co-factors could be
estimated to intensify the pain episode or to be involved
in the recurrence of the unspecific pain.
Back pain in dentistry
Regarding the existing tools to assess spine loads at
the workplace it appears that the tasks which dental
workers routinely perform are at low risk for ergonomi-
cally related disorders [7]. This rating is founded be-
cause of the probably good environmental conditions (no
cold or wet climate), excellent individual job motivation
and social background, no load bearing tasks, no ex-
treme body posture etc. However, low back and neck
pain is a severe and common occupation related injury in
dentistry. Survey data revealed a one year prevalence of
back and neck pain of about 70%-90% [8-10]. Moreover,
Wunderlich et al. [10] could estimate that about 80% of

M. Wunderlich et al. / J. Biomedical Science and Engineering 3 (2010) 664-671 665
the affected German military dentist population occupy
medical help to deal with the pain. These data indicate
that back pain is a major health problem for dental prac-
titioners and methods are needed objectifying the causes
and to identify appropriate intervention strategies [11].
With regard to ergonomic improvements studies indi-
cated that a specific treatment position (“9 o’clock” vs.
“12 o’clock”) or a specific dentist’s stool directly affects
the impact on the spine and the dentist’s health [8,12].
However, up to date potential risk factors for musculos-
keletal disorders (MSD) in dentistry were only assessed
due to questionnaires and qualitative reports of observa-
tion studies [13]. As long as only a rough and almost
invalid analysis could be done it seems unlikely to ob-
jectify the real biomechanics impact on the spine and to
justify useful preventive or ergonomic interventions.
The aim of the present explorative study was to assess
information in detail on trunk posture and movements by
a newly developed ultrasound based mobile device:
3D-SpineMoveGuard (3-D SMG). Beside the objective
information on the working posture and spine kinematics
two different treatment positions (“9 o’clock” vs. “12
o’clock”) where compared to give insight in a potentially
less stressful working posture.
2. MATERIALS AND METHODS
The present study was proven and positively evaluated
by the ethics committee of the German Sport University
Cologne. Every subject voluntarily participated in the
study and gave their written informed consent.
2.1. The 3-Dimensional SpineMoveGuard
(3D-SMG)
Movements and postures of the trunk and spine were
captured with the 3D-SMG system. The 3D-SMG is a
further developed ultrasound based mobile measurement
device to assess three dimensional posture and move-
ment profiles of spine and trunk in real-time. Moreover,
to standardise data analysis and to reveal new kinematic
parameters the analysing software “JSpinal” was devel-
oped.
The basic component of the 3D-SMG is the ultra-
sound device “sonoSens® Monitor” (Friendly Sensors
AG, Jena, Germany). Details of the device are described
elsewhere [14,15]. In brief: Four pairs of ultrasound
transmitters and receivers (diameter 20 mm, height 4
mm) were fixated on the skin (Figure 1) with adhesive
tape. For each channel the skin distance between trans-
mitter and receiver is determined at a sampling rate of 10
Hz. The transmitters and receivers are cable connected to
a small data logger. Variations in inter transmitter skin
distance (shorter or longer running times of the ultra
Figure 1. Hardware application of the 3D-SpineMoveGuard.
Ultrasound transmitters/receivers and the sagittal inclino-
meter have their defined positions at the cervical (CS), tho-
racic (TS) and lumbar spine (LS).
sound signal) represent a change in spine posture within
the associated segment of the spine (Figure 1). The data
points of the 12 channel system represent a 3D model
(sagittal-, frontal- and horizontal plane) of the external
spine curvature and can be stored for up to 10 hours.
In addition to the changes in spine curvature knowl-
edge of the independent occurrence of trunk inclination
is important. Therefore, a one dimensional capacitive,
dielectric liquid-based inclinometer (NA4-70, Seika
Mikrosystemtechnik GmbH, Kempten, Germany) was
taped to the skin with the top edge at the middle of the
spinous process of the 2nd lumbar vertebra (Figure 1).
The inclinometer is cable-connected to a data logger.
The range of measurement was adjusted between +110°
and -30° (frontal/dorsal). Data was captured at a sam-
pling rate of 10 Hz.
2.2. The Analyzing Software “JSpinal”
The newly developed analyzing software JSpinal enables
the valid data management and the detection of move-
ments and postures. JSpinal analyses the segmental data
to
1) adjust the row data with the occupation specific
reference position (segmental spine length in upright
sitting position).
2) calculate relative (%) length indices to separately
detect sagittal (SLI), frontal (FLI) and horizontal (HLI)
movements or postures. The relative (%) spine curvature
enables the comparison of individuals with different
spine proportion.

M. Wunderlich et al. / J. Biomedical Science and Engineering 3 (2010) 664-671 666
3) automatically detect the direction of deviations
from the reference position (neutral zero): SLI forward
(flexion)/backward (extension), FLI lateral flexion
right/left and HLI torsion right/left
4) analyze the frequency distribution of spine and
trunk posture.
5) analyze the frequency and the amount of combined
spine postures: detecting the deviation from neutral zero
in at least two body planes independent of the direction
(e.g. flexion + torsion or torsion + lateral flexion etc.).
6) analyze the kinematic characteristic of the data:
duration, frequency and magnitude of movements as
well as duration, frequency and spine/trunk position of
isometric postures. Both, dynamic and isometric pa-
rameters were analysed within every single spine seg-
ment and body plane
7) rate the variability of the occupational demands
data can be chronologically analysed. Therefore the
time of the analysis will be divided in to 10 segments
(percentiles of the time) though the intensity/frequency
of spinal stressors can be compared between the seg-
ments.
2.3. Subject and Examination Procedure
The explorative assessment of spine loads in dentistry
was conducted by a single case study. Data collection
was done on the same days of two consecutive weeks. At
the first day data were assessed in the 9 o’clock position;
at the second date in the 12 o’clock position. To keep the
medical treatments comparable only 20 patients for pe-
riodontol maintenance were recruited for the two ex-
amination days. All patientes were treated with two dif-
ferent hand scalers, curettes, ultrasonic devices, rotating
and polishing instruments. Each treatment session lasted
45 minutes.
After the instrumentation of the measurement devices
maximal voluntary range of movements were assessed
within the three body planes in standardised positions
[14]. Furthermore, to identify the reference working po-
sition (neutral zero) the dentist set down on his working
stool with his hands relaxed on his thighs and an upright
trunk position with the head held according to the
Frankfurt horizontal. Each of the positions was kept for
about 30 seconds. Shortly thereafter the routinely treat-
ment of the patients began in the usual and necessary
duration and velocity.
2.4. Statistics
Data analyses were reduced to the time of treatment
procedure. Periods between the treatments and breaks
were excluded to reduce the bias due to individual be-
haviouristic. Because of the explorative character of the
study only descriptive data analyses were done by SPSS©
17.0. Data is shown as frequency and percentile (5th, 25th,
median, 75th, and 95th) distributions.
3. RESULTS
The maximal voluntary range of motion is shown in Ta-
ble 1. The magnitudes can be used to compare the devia-
tions from neutral zero (upright sitting) during the non
surgical periodontal treatment.
Figure 2 indicates the percentile values of spine and
trunk positions throughout the whole working day. The
different profiles of the 9 and 12 o’clock working posi-
tion indicate the necessary adoption of the posture by the
dentist. In detail:
Cervical spine (CS): sagittal deviations from neutral
zero are very small. The spectrum of lateral flexion and
torsion is about 3 times broader in both 9 and 12 o’clock
positions. In 9 o’clock position FLI and HLI values are
almost normal distributed around zero. During the 12 o’
clock manoeuvre the CS is in 75% of the treatment time
in a lateral left position and turned to the right.
Thoracic spine (TS): the range of movement is very
small. The TS were held in the almost identical position
in both 9 and 12 o’clock position. During the 12 o’clock
treatment FLI and HLI data ranges in a contra lateral
relation to the values of CS.
Lumbar spine (LS): the range of motion is moderate
and between the spectrum of CS and TS. In 9 o’clock
SLI and FLI indicate an almost upright sitting position.
HLI values show a continuous left rotation toward the
patient. During 12 o’clock treatment SLI is in a flexed
position during 75% of the treatments (the head of the
patient was in front of the abdomen of the dentist).
Moreover, LS is laterally flexed to the right (FLI) and
rotated to the left during the whole working day.
Trunk inclination (TI): the range of trunk motion in 9
o’clock indicating a broader spectrum of trunk positions
and is almost normal distributed. In 12 o’ clock TI is
negative in 75% of the treatments, indicating the tilt of
the pelvis.
Table 1. Maximum voluntary range of motion for each part of
the spine and within the three body planes.
Maximum voluntary range of
motion
upright sitting = 0%
Body planes
CS TS LS
Sagittal plane (Extension/Flexion) -25%/24% -8%/20% -10%/22%
Frontal plane (left/right) -18%/21% -16%/17%-15%/16%
Horizontal plane (left/right) -20%/21% -8%/10%-8%/12%
CS = cervical spine; TS = thoracic spine; LS = lumbar spine

M. Wunderlich et al. / J. Biomedical Science and Engineering 3 (2010) 664-671 667
Figure 2. Distributions of segmental spine positions of the dentist in 9 o’clock (blue/left symbol) and 12 o’clock treat-
ment positions. Deviations from upright sitting “0” can be determined for sagittal (SLI), frontal (FLI) and horizontal (HLI)
spine and trunk (sagittal inclination) postures. Box-Plots are showing 5th, 25th, 50th, 75th, and 95th percentile for cervical
(CS), thoracic (TS) and lumbar (LS) spine.
Apart from the postures within one body plane the
analyses further revealed the amount of “combined pos-
tures”. Figure 3 shows the frequency of combined spine
postures during periodontology care in 9 and 12 o’clock
position. The postures are grouped from “small” (0-2%),
“medium” (> 2-6%), “large” (> 6-10%) to “extreme” (>

M. Wunderlich et al. / J. Biomedical Science and Engineering 3 (2010) 664-671 668
Figure 3. Frequency of multi-directional combined postures of
the spine. Values are shown depending on magnitude (devia-
tion from upright sitting) of the combined postures for cervical
(CS), thoracic (TS) and lumbar (LS) spine.
10%). During 90% of the treatment time the combined
postures ranged between “small” and “medium” with
clear differences between the segments of the spine. Po-
sition related (9 vs. 12 o’clock) varieties could only be
estimated for the LS, where about 2/3 of the treatment
time in 12 o’clock position where spend in a “medium”
combined posture.
The further analysis of the kinematic work demand
revealed that in more than 75% of the treatment time no
spine movements occur. Small or medium movements of
the trunk could be estimated in about 25% of the work
time (Table 2). In addition, during 75% to 85% of the
periodontology care the dentist worked in an isometric
posture (Figure 4). As could be drawn from Figure 3
there exit some differences in static work posture be-
tween the segments of the spine but none between
treatment positions of the dentist (9 vs. 12 o’clock).
Moreover, Figure 5 shows the durations (s) of isometric
Figure 4. Frequency of static work posture during periodon-
tologic care. Values are shown for cervical (CS), thoracic (TS)
and lumbar (LS) spine as well as trunk inclination (TI).
Figure 5. Duration of isometric work postures within the cer-
vical (CS), thoracic (TS) and lumbar (LS) spine as well as for
trunk inclination (TI). Box-Plots are showing 5th, 25th, 50th, 75th,
and 95th percentile.

M. Wunderlich et al. / J. Biomedical Science and Engineering 3 (2010) 664-671 669
Table 2. Distribution of movement amplitudes of the cervical (CS), thoracic (TS) and lumbar (LS) spine. The amplitudes
represent relative (%) differences in segmental spine length per movement. Amplitudes of the trunk are expressed as degree
(°) changes per movement.
Distribution of movement amplitudes
Amplitude (%) CS Amplitude (%) TS Amplitude (%) LS Amplitude (°) Inclination
Percentile
9 o’clock 12 o’clock 9 o’clock12 o’clock 9 o’clock12 o’clock 9 o’clock 12 o’clock
5. ,00 ,00 ,00 ,00 ,00 ,00 ,00 ,00
25. ,00 ,00 ,07 ,07 ,02 ,03 ,00 ,00
50. ,22 ,20 ,16 ,16 ,14 ,15 3,00 3,00
75. ,44 ,44 ,54 ,48 ,40 ,34 5,00 5,00
95. 3,07 3,54 2,28 1,74 2,51 2,51 14,00 11,00
Rating of spine movements Rating of trunk inclination movements
0-<2 % = “no movement” 0°-< 3° = “no movement”
2-6 % = small to medium amplitude 3-20° = small to medium amplitude
> 6 % = large amplitude >20° = large amplitude
work postures. The median values could be estimated by
3 to 4 seconds for both spine and trunk. However, inter-
ruptions where even shorter (median duration of move-
ments ≈ 1 s) and as shown in Table 2 of small ampli-
tudes.
4. DISCUSSION
During periodontology care spine and trunk of the den-
tist where held within a small range of movement. The
deviation from neutral zero was very small in TS and
higher in CS and LS. The frequently postulated assump-
tion that dentists work in a sewer bended and twisted
posture [8,11] could not be confirmed. During 90% of
the non-surgical treatments the combined postures
ranged between “small” and “medium” deviation from
neutral zero. It turned out that the preliminary posture
specific risk factor could be the continuous phases of
isometric spine and trunk position. Moreover, the
changes of work postures were done very slowly and by
small movements. The revealed kinematics and postures
of spine and trunk differed only slightly between the
most popular treatment positions (9 o’clock vs. 12
o’clock) of the dentist.
As it has been frequently reported more than 90% of
the back pain is unspecific: No structural damage or in-
jury can be detected to explain the cause of the aches
[16]. Hence, the pain seems to be triggered by the active
and/or passive structures of the back and spine. In this
context it seems important to remember that the muscles
are much more than a contractile organ. The muscles are
rather a large sensory part of the body which registers a
broad spectrum of chemical and mechanical changes
[17,18].
With regard to the new quantitative findings of the
task analysis the static none neutral work postures can
lead to reduced muscle circulation. As a result metabo-
lites accumulate within the extracellular space. Beside
these perfusion and metabolic related volumes and con-
centration shifts mechanical pressure to the tissue could
also lead to acute or delayed exposure induced pain
[17,19]. Chemical or mechanical changes will be regis-
tered by free interstitial nerve endings (chemo and
mechano receptors) of slow afferent nerve fibres (group
III and IV). This will not only lead to an increase in
heart circulation and breathing but can also reveal re-
flectively pain phenomena and changes in muscle ten-
sion [17,19]. In addition to these acute reactions pro-
longed static none neutral posture can also cause struc-
tural damage of the spine. Additionally, none neutral
work posture of the dentist leads to a continuous eccen-
tric exposure of the intervertebral disc. Static and eccen-
tric pressures to the disc can lead to deficits in locale
nutrition, metabolic imbalance and can cause cumulative
trauma within the structure of the disc and the vertebral
end plates [20].
Former studies concluded that changes in treatment
positions of the dentist and/or ergonomic sitting stools
and/or arrangements of the medical practice will lead to
significant reductions in spine exposure [8,12]. However,
the detailed data of the explorative study did not reveal
substantial differences of exposure relevant body pos-
tures due to an altered treatment position of the dentist.

M. Wunderlich et al. / J. Biomedical Science and Engineering 3 (2010) 664-671 670
Moreover, the data suggests that dental care is per se
strongly associated with a certain, uncomfortable work
posture and spinal loads. Intervention concepts should
consider both ergonomics and individual alterations. To
be more precisely, specific physical intervention con-
cepts have to be adapted for dental care personnel to
cope with the physical stress of work postures. The ac-
tivity programs have also to be adapted to the daily rou-
tine of the personnel to interrupt and reduce the cumula-
tive isometric load to the spine [21,22].
The results of the present explorative study are of
course not representative for body postures during dental
care. Moreover, due to limited methodical capabilities of
former studies the parameters are not directly compara-
ble with the literature. But to the best of our knowledge
there is no comparable detailed data of kinematics pa-
rameter of trunk and spine during dental care. However,
the data documents exemplarily the task specific work
load characteristic, enables the objective evaluation of
interventions (sitting position) and leads to occupation
relevant intervention concepts.
5. CONCLUSIONS
The frequently reported back pain in dentist is not only
related to severe body postures rather than to the high
amount of isometric spine loads. Moreover, with regard
to an altered sitting position of the dentist only slightly
exposure relevant differences could be estimated.
Therefore, intervention concepts for dental care person-
nel should pay more attention to individual physical
training during and off work.
6. ACKNOWLEDGEMENTS
The analysis was funded by the Medical Service of German Armed
Forces (M/SAB1/4/A009) and supported by the Department of Perio-
dontology of the German Armed Forces Central Hospital Koblenz.
7. CONFLICT OF INTEREST
The authors declare that they have no conflict of interest.
REFERENCES
[1] Schneider, S., Schmitt, H., Zoller, S. and Schiltenwolf, M.
(2005) Workplace stress, lifestyle and social factors as
correlates of back pain: A representative study of the
German working population. International Archives of
Occupational and Environmental Health, 78(4), 253-269.
[2] Schneider, S. and Schiltenwolf, M. (2006) Occupations
associated with a high risk of self-reported back pain:
Representative outcomes of a back pain prevalence study
in the Federal Republic of Germany. European Spine
Journal, 15(6), 821-833.
[3] Kumar, S. (2001) Theories of musculoskeletal injury
causation. Ergonomics, 44(1), 17-47.
[4] Juniper, M., Le, T.K. and Mladsi, D. (2009) The epide-
miology, economic burden, and pharmacological treat-
ment of chronic low back pain in France, Germany, Italy,
Spain and the UK: A literature-based review. Expert
Opinion on Pharmacotherapy, 10(16), 2581-2592.
[5] Harkness, E.F., Macfarlane, G.J., Nahit, E.S., Silman, A.J.
and McBeth, J. (2003) Risk factors for new onset low
back pain amongst cohorts of newly employed workers.
Rheumatology, 42(8), 959-968.
[6] Hoogendoorn, W.E., Bongers, P.M., Vet, H.C.W., de
Douwes, M., Koes, B.W., Miedema, M.C. Ariëns, G.A.
and Bouter, L.M. (2000) Flexion and rotation of the trunk
and lifting at work are risk factors for low back pain:
Results of a prospective cohort study. Spine, 25(23),
3087-3092.
[7] Guay, A.H. (1998) Commentry: Ergonomically related
disorders in dental practice. Journal of the American
Dental Association, 129(2), 184-186.
[8] Thornton, L.J., Barr, A.E., Stuart-Buttle, C., Gaughan,
J.P., Wilson, E.R., Jackson, A.D., Wyszynski, C.T. and
Smarkola, C. (2008) Perceived musculoskeletal symp-
toms among dental students in the clinic work environ-
ment. Ergonomics, 51(4), 573-586.
[9] Wassan, A.L., Almas, K.A. and Shethri, S.E. (2001)
Rückenund Nackenbeschwerden bei Zahnärzten und
zahnmedizinischem Personal. Journal of Contemporary
Dental Practice, 2(3), 17-30.
[10] Wunderlich, M., Rüther, T., Eßfeld, D., Rohde, U., Eger,
T., Leyk, D. (2009) Rückenschmerzen und Wirbels-
äulenbelastungen – Befragung des zahnmedizinischen
Personals in der Bundeswehr. Wehrmed Mschr, 53(1),
230-234.
[11] Leggat, P.A., Kedjarune, U. and Smith, D.R. (2007) Oc-
cupational health problems in modern dentistry: A review.
Industrial Health, 45(5), 611-621.
[12] Morita Europe GmbH, J. (2008) Dr. Beach ergonomic
concept. Warum sich intuitive Arbeitswege lohnen und
Rückenschmerzen nicht sein müssen. DDZ, 117(4),
188-189.
[13] Marklin, R.W. and Cherney, K. (2005) Working postures
of dentists and dental hygienists. CDA Journal, 33(2),
133-136.
[14] Baum, K., Hoy, S. and Essfeld, D. (1997) Continuous
monitoring of spine geometry: A new approach to study
back pain in space. International Journal of Sports
Medicine, 18(Suppl 4), 331-333.
[15] Friedrich M (2002) Measuring lumbar sagittal posture in
sewgae workers using an ultrasonic device. Journal of
Musculoskelet Research, 6(3-4), 135-145.
[16] Waddell, G. and Burton, A.K. (2001) Occupational health
guidelines for the management of low back pain at work:
evidence review. Occupational Medicine, 51(2), 124-
135.
[17] Sjøgaard, G., Lundberg, U. and Kadefors, R. (2000) The
role of muscle activity and mental load in the develop-
ment of pain and degenerative processes at the muscle
cell level during computer work. European Journal of
Applied Physiology, 83(2-3), 99-105.
[18] Leyk, D. (2009) The preventive and therapeutic roles of
regular physical activity. Deutsches Aerzteblatt interna-
tional, 106(44), 713-714.
[19] Baum, K., Selle, K., Leyk, D. and Essfeld, D. (1995)

M. Wunderlich et al. / J. Biomedical Science and Engineering 3 (2010) 664-671 671
Comparison of blood pressure and heart rate responses to
isometric exercise and passive muscle stretch in humans.
European Journal of Applied Physiology, 70(3), 240-245.
[20] Davis, K.G. and Marras, W.S. (2000) The effects of mo-
tion on trunk biomechanics. Clinical Biomechanics,
15(10), 703-717.
[21] Finsen, L., Christensen, H. and Bakke, M. (1998) Mus-
culoskeletal disorders among dentists and variation in
dental work. Applied Ergonomics, 29(2), 119-125.
[22] Ratzon, N.Z., Yaros, T., Mizlik, A. and Kanner, T. (2000)
Musculoskeletal symptoms among dentists in relation to
work posture. Work, 15(4), 153-158.