Pain. 1998 (Aug); 77 (2): 201–207 ~ FULL TEXT
Elisabeth I. Skargrena,, Birgitta E. O¨ berg
Department of Neuroscience and Locomotion,
Physiotherapy, Faculty of Health Sciences,
Linköping University, Sweden.
The inability to predict outcome in patients with low back/neck pain leads to inappropriate or unnecessary treatment. The aims of the study were to identify prognostic factors for disability at 1-year follow-up in patients with back pain visiting primary care, and to compare the effect of these in two treatment strategies--chiropractic and physiotherapy. Data were taken from a randomised trial on patients with back/neck pain visiting the general practitioner, in which patients were allocated to chiropractic and physiotherapy as primary management. Three hundred and twenty-three patients, aged 18-60 years, who had no contraindications to manipulation and who had not been treated within the previous month were included in the study. Multiple regression analysis was used to identify prognostic factors. Dependent variables were mean Oswestry score and mean change in Oswestry score at 12-month follow-up. The multiple regression analysis revealed five significant (P < 0.001-0.01) prognostic factors; duration of current episode, Oswestry score at entry, expectations of treatment, number of localisations, and well-being. Besides, the regression coefficients for the significant factors were compared between the two treatment strategies.
No significant difference in effect or regression coefficients for the prognostic factors were seen between the two treatment strategies. Twelve per cent of the patients had poor prognostic factors (duration > or = 1 month, more than one localisation, low expectations of treatment and low well-being) at entry. These patients had a mean Oswestry score above 20% at 1-year follow-up. Clinical decision models for the management of patients with back pain visiting primary care that consider prognostic factors need to be implemented and prospectively evaluated.
KEYWORDS: Chiropractic; Low back pain; Neck pain; Oswestry score; Physiotherapy; Predictors
From the FULL TEXT Article:
Back pain is a common problem in primary care and a
major cause of disability and loss of work. Physicians have
difficulty in choosing treatment forms and predicting which
of the patients will get better and which will go on to suffer
chronic back pain. One reason for the difficulties is the fact
that in most cases we cannot establish an organic cause
(Waddell, 1987; Frymoyer, 1988), so, mostly, back pain is
‘an illness looking for a disease’. Another reason is shown
by a number of systematic reviews of the quality of studies
concerning interventions for low back pain (Koes et al.,
1996; Tulder van et al., 1997) and neck pain (Hurwitz et
al., 1996; Kjellman et al., unpublished data). These showed
that the effectiveness of many interventions available for
treating back pain has not yet been demonstrated beyond
doubt. The quality of design, execution and reporting need
to be improved to establish evidence concerning effectiveness.
One important aspect of design is the study population.
A homogeneous study population enhance the possibility of
seeing similarities or differences better when comparing
treatments. A heterogeneous population can be regarded
as consisting of several more homogenous subgroups of
patients, where the effect varies among the subgroups. A
systematic review of randomised studies of spinal manipulation
for low back pain showed that 29 out of 36 studies did
not fulfil the criteria for homogeneity (0–1 points out of two
possible) (Koes et al., 1996). A similar systematic review of
chiropractic spinal manipulation showed that one out of
eight studies fulfilled the criteria for homogeneity (Assendelft
et al., 1996).
In a recently presented randomised study we compared
chiropractic and physiotherapy treatment in the management
of patients with low back and neck pain (Skargren
et al., 1997) Inclusion criteria were all patients between
18–60 years of age who visited a general practitioner in
primary care for low back or neck pain. The comparison
between the two groups showed similar results in effectiveness
and costs after treatment and at 6- and 12- month
follow-up. Our study population was heterogeneous in
cause, length of duration, etc. Therefore, subgroup analysis
was performed in more homogenous subgroups (Skargren
et al., 1998). The subgroupings were chosen according to
previous studies (Postacchini et al., 1988; Meade et al.,
1990) ‘similar problems during the previous 5 years’,
‘duration of current episode’ (£1 week, 1–4 weeks, ³1
month) and level of ‘Oswestry disability score at entry’
(,40%, ³40%). Our subgroup analyses showed that
patients with acute problems (duration of current episode
,1 week) benefited more from chiropractic than physiotherapy,
whereas patients with pain that had lasted for
³1 month gained more from physiotherapy (Skargren et al.,
1998). The study generated several questions about the
choice of study population in research and what prognostic
factors that are essential for the choice of treatment and the
prediction of outcome in clinical praxis. The latter is essential
knowledge also as a base for decisions of the future
structure and organisation of the management of patients
with back pain. Leboeuf-Yde et al. (1997) have also
pointed out the importance of subgrouping patients with
back pain to support the choice of treatment and scientific
evidence of treatment. To be able to form more homogenous
subgroups we have to know which are the potential
factors for the course of back pain. A number of studies on
prognostic factors for the course of low back pain have
been conducted in primary care (Pedersen, 1981; Roland
and Morris, 1983; Chavannes et al., 1986; Singer et al.,
1987; Lanier and Stockton, 1988; Von Korff et al., 1993;
Coste et al., 1994; Burton et al., 1995; Klenerman et al.,
1995; Cherkin et al., 1996; Hoogen van den et al., 1997),
but the results vary considerably.
The aims of this study were to identify prognostic factors
for disability at 1-year follow-up of patients with back pain
visiting primary care, and to compare the effect of these in
two treatment strategies – chiropractic and physiotherapy.
The data for the study to be described were taken from a
prospective clinical trial in which patients who visited a
general practitioner for low back and/or neck pain were
randomised to chiropractic or physiotherapy. Patients
were included from 10 primary care centres of different
size in a medium-sized Swedish county. Patients was followed
by questionnaires. Ninety-eight percent completed
the 12-month follow-up.
The study comprised patients between 18 and 60 years of
age who attended a general practitioner for low back or neck
problems. The main criteria for eligibility were that the
patients had not received active treatment for their problems
within the previous month and had no contraindication to
manipulation. Other criteria for eligibility were that they
had no other disease, were able to read and write in Swedish
and, finally, had problems relevant to chiropractic and physiotherapy
treatment. In total 323 patients participated in the
trial; 179 in the chiropractic group and 144 in the physiotherapy
group. A more careful description of reasons
for exclusion and withdrawal has been presented previously
(Skargren et al., 1997).
Baseline data were collected by questionnaire after randomisation.
Sociodemographic variables were; age, sex, life
style factors such as smoking, and exercise habits, job satisfaction,
similar problems the previous 5 years (no/yes),
duration of current episode (£1 week, 1–4 weeks, 1–3
months, ³3 months), and expectations of treatment on a
four-point scale (completely restored, quite improved, partial
relief, no expectations of being restored, or getting
Variables describing the severity of problems were different
aspects of perceived pain, function and general
Measures for pain were: pain intensity on a visual analogue
scale (VAS in mm) (Scott and Huskinsson, 1976), pain
frequency on a five-point scale and a pain drawing (Ransford
et al., 1976); from the latter the number of localisation’s
(neck, thoracic, lumbar) were estimated.
Measures for function were; sick-leave and the Oswestry
low back pain disability questionnaire. The result of the
Oswestry score is expressed on a scale ranging from 0%
(no pain or difficulties) to 100% (highest score for pain or
difficulty on all items) (Fairebank et al., 1980).
General health was followed with a six-point scale concerning
well-being, and on a VAS (in mm) with the endpoints:
0, best imaginable, and 100, worst imaginable
Localisation of the cause of treatment and degree of pain
radiation was reported by the general practitioner.
Each patient’s treatment was at the discretion of the individual chiropractor or physiotherapist. The mean number of
treatment sessions (±SD) during the treatment period was
lower in the chiropractic group 4.9 (2.0) than in the physiotherapy
group 6.4 (5.4); the difference (95% CI) was
-1.53 (-2.38; -0.68). The mean time for the treatment period
was somewhat longer, but not statistically significant, in
the physiotherapy group (mean ± SD 4.7 ± 4.7) than in the
chiropractic group (mean ± SD 4.1 ± 3.3). The majority of
the chiropractic patients received spinal manipulation. The
number of treatments and treatment forms varied more
between the individual patients in the physiotherapy group
(Skargren et al., 1997). None of the physiotherapists was a
specialist in manipulation.
To detect any significant differences between groups, chisquare
tests were used in comparing proportions of patients
and the Mann–Whitney U-test and t-test for groups, in comparing
means of pain intensity, general health and Oswestry
score. The Mann–Whitney U-test and t-test for groups gave
similar results, thus the t-test values are presented. Significance
level was P , 0.05 (two-sided). Linear multiple
regression was used to identify prognostic factors of importance.
Two dependent variables were used in the multiple
regression analysis, one representing the end level in Oswestry
score at 12-month follow-up and the other, the change in
Oswestry score between entry and 12-month follow-up.
Before the multiple regression, all background variables
and severity of problem variables at entry as well as the two
dependent variables, were entered in a bivariate correlation
analysis (Pearson’s product correlation coefficient) in order
to find the independent variables that significantly correlated
with the dependent variables (Tabachnick and Fidell,
1996). The threshold level P £ 0.10 was used. Pain radiation,
exercise habits and job satisfaction did not correlate
significantly with any of the dependent variables, so they
The multiple regression analysis was performed in two
steps with both of the dependent variables. Firstly, a stepwise
multiple regression was performed in order to find
the most important prognostic factors, the variables with a
significant regression coefficient. Two stepwise multiple
regression models were performed, the first including the
variable sick-leave. As sick-leave did not emerge as a factor
and excluded 15% of the population (those not at work), we
did a second multiple regression model without sick-leave.
The decision to exclude sick-leave was also based on the
fact that Oswestry score, which is a broad measure of
disability, strongly significantly correlated (r = 0.45) with
sick-leave (Tabachnick and Fidell, 1996).
Secondly, a separate multiple regression analysis with the
variables received in the stepwise multiple regression was
performed of chiropractic and physiotherapy. The regression
coefficients (B values) were compared using a covariance
analysis (Hassard, 1991).
Description of the study population
Background data and other characteristics of the two
groups confirming the study population did not differ significantly
initially, except for a slight difference in pain intensity
and general health (both measured with VAS). The
patients in the physiotherapy group estimated their pain
intensity and general health slightly to be worse than the
patients in the chiropractic group before treatment (Table 1).
Prognostic factors and comparison of regression coefficients
In total 14 variables were included in the stepwise multiple
regression analysis. The analysis showed that five of the
factors, i.e. duration of current episode, Oswestry score at
entry, number of localisations, expectations of treatment
and well-being, were significantly associated with Oswestry
score at the 12-month follow-up (Table 2). The remaining
nine factors: age, sex, smoking, similar problem previously,
neck or low back problems, pain intensity and frequency,
and general health (VAS) did not contribute significantly to
the prediction of outcome.
Separate multiple regression analysis of chiropractic and
physiotherapy showed no significant difference between
regression coefficients for the five significant prognostic
factors derived in the stepwise analysis (Table 3).
The same significant prognostic factors appeared when
the dependent variable change in Oswestry score from
entry to 12-month follow-up was used in the stepwise multiple
regression. The regression coefficients and significance
levels received were the same as when level of Oswestry at
12-month follow-up was used, with one exception. The
regression coefficient B (SE B) value for the Oswestry
score at entry was higher, -0.72 (0.05) compared to 0.28
(0.05). The adjusted R square for the model was 0.62. This
was the case for the multiple regression analysis with the
five significant prognostic factors as well. The regression
coefficients B (SE B) for Oswestry score at entry were
-0.71 (0.06) in the chiropractic group and -0.70 (0.07) in
the physiotherapy group. The adjusted R square was 0.66 for
chiropractic and 0.54 for physiotherapy.
Amount of patients with a poor prognosis
When the three prognostic factors, i.e. duration of current
episode, expectations of treatment, and number of localisations,
were applied in the study population, 14% of the
patients had poor prognosis (duration ³ 1 month, more
than one localisation, and less positive expectations of treatment,
while 13% had good prognosis (duration £ 1 week,
one localisation, and expected to be completely restored).
Those with poor prognosis estimated mean (SD) Oswestry
score significantly higher at 12-month follow-up 23 (14)%
than those with a good prognosis 7 (12)% the remainder,
who had a mixture of good and poor prognostic factors
scored 16 (14)%. The proportion of patients and mean
Oswestry score at 12-month follow-up was similar concerning
the chiropractic and the physiotherapy groups in the
If the prognostic factor ‘well-being’ (very good/less
good) was added, 12% of the patients had poor prognosis
and 4% good. Corresponding mean (SD) for Oswestry score
at the 12-month follow-up were 26 (14)% and 4 (8)%.
Five prognostic factors for disability at the 1-year followup
emerged from this study; duration of current episode,
Oswestry score, number of localisations, expectations of
treatment and well-being at entry. The same prognostic
variables emerged for level of disability and change of disability
at the 12-month follow-up. The result partly confirmed
what previous studies in the area have shown, but
also highlighted that the number of localisations, patients’
expectations of treatment and well-being were important
factors that need to be considered in clinical praxis and in
the prediction of outcome.
Our study population represents a substantial part of the
patients visiting general practitioners for back pain, which
was judged by the physician as needing physical treatment.
The fact that the data were taken from a randomised study in
which part of the patients were seen ineligible (Skargren et
al., 1997) might be a limitation. However, our study population
did not differ from those in most of the previous
studies on prognostic factors for outcome in patients with
back pain visiting primary care, according to such background
data as age, proportion of patients with previous
back problems and degree of patients at work (Roland and
Morris, 1983; Lanier and Stockton, 1988; Von Korff et al.,
1993; Coste et al., 1994; Burton et al., 1995; Cherkin et al.,
1996). In some previous studies the proportion of patients
with acute low back problems was similar to ours (Singer et
al., 1987; Von Korff et al., 1993), in others the proportion of
patients with acute low back problems was slightly higher
than in our study (Pedersen, 1981; Cherkin et al., 1996;
Hoogen van den et al., 1997). Two of the previous studies
only included patients with acute low back pain (Coste et
al., 1994; Klenerman et al., 1995).
The separate multiple regression analysis of chiropractic
and physiotherapy showed that the five prognostic factors
had similar effects in the two treatment strategies. The comparison
of regression coefficients between the two treatment
strategies did not show any significant difference, but a
slight tendency to difference in the factor ‘duration of current
episode’ strengthened the previous subgroup analysis.
The subgroup analysis showed that patients with acute (£1
week) problems benefited more from chiropractic or manipulation
whereas patients with more chronic problems (³1
month) benefited more from physiotherapy in the change in
Oswestry score (Skargren et al., 1998).
Disability at entry (also called activity discomfort in
some previous studies) and duration of current episode
have appeared beside other factors in several previous studies
of patients with back pain in primary care (Pedersen,
1981; Roland and Morris, 1983; Singer et al., 1987; Von
Korff et al., 1993; Coste et al., 1994; Burton et al., 1995;
Klenerman et al., 1995; Hoogen van den et al., 1997), suggesting
that they are important factors in this population and
need to be considered in clinical praxis. Previous problems
with back pain have been shown to be a prognostic factor for
outcome in disability in a number of previous studies in the
area (Pedersen, 1981; Singer et al., 1987; Deyo and Diehl,
1988; Coste et al., 1994; Burton et al., 1995), but this did not
emerge in this study. However, the variable ‘previous problems’
emerged only for patients with acute problems (Burton
et al., 1995) or was more specified in those studies than
in ours e.g. number of previous episodes (Singer et al., 1987;
Deyo and Diehl, 1988) previous chronic episodes (Coste et
al., 1994), duration of previous period more than 3 weeks
(Pedersen, 1981). Different findings suggest that more specified
questions of the frequency and length of previous
problems combined with definitions of subgroups might
be more important for outcome than simply if the patients
have had previous problems or not.
The number of localisations (pain from more than one
region) has not been shown previously to be a prognostic
factor for disability in back pain patients visiting primary
care. This might be explained by the fact that the variable
has not been inserted in multivariate analysis previously.
However, some related studies have been presented, suggesting
that it is important if the pain is localised to more
than one region. Murphy and Cornish, 1984 showed that
acute low back pain patients who became chronic reported
pain over a wider area of the body according to a pain
drawing, and Buskila et al., 1997 showed that the development
of fibromyalgia syndrome was more frequent following
cervical spine injury than following lower extremity
injury. Another recent study showed that the number of
pain sites was a prognostic factor for worse long-term outcome
for work disability in patients with musculoskeletal
soft tissue injury (Crook and Moldovsky, 1996). Finally,
Leboeuf-Yde et al., 1997 showed a difference in association
between potential risk indicators (age, sex, marital status,
attitude to a healthy life style, self-reported physical activity
at work, and smoking) between subgroups of patients with
low back pain with and without pain in the neck or upper
Expectations of treatment have not previously been
shown to be a prognostic factor for level or change in disability
due to back pain. Like the number of localisations
this variable has not previously been used in multivariate
analysis of back pain patients in primary care. However,
other closely related factors such as attitudes towards and
beliefs about pain in chronic back pain patients (Strong et
al., 1990), fear-avoidance beliefs (Waddell et al., 1993) and
fear-avoidance behaviour (Klenerman et al., 1995) have
been shown to be prognostic factors for outcome in disability.
According to the literature, expectation is regarded as
part of satisfaction and fear-avoidance behaviour (Linder-
Pelz, 1982; Klenerman et al., 1995; Thompson and Sun˜ol,
1995). A couple of papers presented recently have show an
increasing interest in the importance of patients’ expectations
on outcome in back pain (Deyo et al., 1994; Borkan
and Cherkin, 1996).
The last factor that emerged in the stepwise analysis was
well-being. The factor might be regarded as closely related
to psychosocial factors like anxiety, depression, coping
strategy and always feeling sick (part of SIP), which have
previously been shown to be factors of importance for disability
(Deyo and Diehl, 1988; Lanier and Stockton, 1988;
Burton et al., 1995).
The variables age, sex, smoking, exercise habits, similar
problems in the previous 5 years, pain radiation, pain intensity
and pain frequency and general health (VAS) did not
emerge as significant prognostic factors. This does not mean
that they are of no importance for the prognosis of low back
pain, but merely that other factors were more important in
this study population. Pain radiation, for instance, has been
shown to be a prognostic factor in a number of previous
studies of back pain patients in primary care (Pedersen,
1981; Chavannes et al., 1986; Lanier and Stockton, 1988;
Cherkin et al., 1996), but not in this study. This might partly
be explained by the low proportion of patients in the study
population who had pain radiating distally.
Two of the five prognostic factors received in this study
have been found in previous studies. The prognostic factors
received in patients with back pain in primary care vary
considerably between different studies. Restrictions of
study population (acute, subacute or a mixture of problems,
high number of drop outs (Burton et al., 1995; Klenerman et
al., 1995; Hoogen van den et al., 1997), small sample (Pedersen,
1981; Lanier and Stockton, 1988; Coste et al., 1994),
long time to data collection (Von Korff et al., 1993; Klenerman
et al., 1995), and the length of follow-up period (Klenerman
et al., 1995; Cherkin et al., 1996) are factors in the
study design and performance that influence what factors
will emerge. Other factors of importance for which prognostic
factors will emerge are what independent variables
were inserted in the multivariate analysis and the choice of
dependent variable. Many of the differences in received
prognostic factors between studies within the area can be
explained by the differences in preconsumption.
Even though the identified factors do not explain the
whole outcome, 12% of the patients in our study were identified
as having four of the poor prognostic factors before
treatment, e.g. duration of current episode ³1 month, more
than one localisation of pain, low well-being and did not
expect that they would be completely restored by the treatment.
The mean Oswestry score never reached below 20%
for these patients, which means that they still had moderate
disability according to the definition by Fairebank et al.,
1980. It might be that if we use subgroupings according to
prognostic factors, a proportion of patients do not reach a
clinically acceptable level of disability. These patients with
poor prognosis can easily be identified in clinical praxis. It is
important to identify these patients early, and it might be
important to consider a supplementary treatment model not
included in this study. Previous studies have shown that
cognitive-behavioural models added to traditional physiotherapy
are more effective than solely traditional treatment
in patients with back pain (Fordyce et al., 1986;
Nicholas et al., 1992; Linton, 1994; Turner, 1996). Others
have shown that multimodal approaches are effective in
chronic back pain (Pfingsten et al., 1997).
The factors ‘duration of current episode’ and ‘Oswestry
score at entry’ that emerged strengthen previous results and
the factors ‘number of localisations, expectations of treatment’
and ‘well-being’ add new factors. Clinical decision
models for managing patients with back pain visiting primary
care that consider prognostic factors need to be implemented
and prospectively evaluated.
This research was supported by the County Council of
stergo¨tland and Va°rdalstiftelsen. The authors thank Per
Carlsson, Mikkel Gade and Annika Rosenbaum, Hans
Tropp and UllaMaija O¨ dman for offering their expert
knowledge in the planning and implementation of the
study and John Carstensen for statistical advice.
Assendelft, P., Koes, B., van der Heijden, G. and Bouter, L.,
The effectiveness of chiropractic for treatment of low back pain: an update and attempt at statistical pooling,
J. Manipulative Physiol. Ther., 19 (1996) 499–507.
Borkan, J. and Cherkin, D.,
An agenda for primary care research on low back pain,
Spine, 21 (1996) 2880–2884.
Burton, K., Tillotson, M., Main, C. and Hollis, S.,
Psychosocial predictors of outcome in acute and subchronic low back trouble,
Spine, 20 (1995) 722–728.
Buskila, D., Neumann, L., Vaisberg, G., Alkalay, D. and Wolfe, F.,
Increased rates of fibromyalgia following cervical spine injury. A controlled study of 161 cases of traumatic injury,
Arthritis Rheum., 40 (1997) 446–452.
Chavannes, A.W., Gubbels, J., Post, D., Rutten, G. and Thomas, S.,
Acute low back pain: patients´ perceptions of pain four weeks after initial diagnosis and treatment in general practice,
J. R. Coll. Gen. Pract., 36 (1986) 271–273.
Cherkin, D., Deyo, R., Street, J. and Barlow, W.,
Predicting poor outcomes for back pain seen in primary care using patients’ own criteria,
Spine, 21 (1996) 2900–2907.
Coste, J., Delecoeuillerie, G., Cohen de Lara, A. and Paolaggi, J.,
Clinical course and prognostic factors in acute low back pain: an inception cohort study in primary care practice,
Br. Med. J., 308 (1994) 577–580.
Crook, J. and Moldovsky, H.,
The clinical course of musculoskeletal pain in empirically derived groupings of injured workers,
Pain, 67 (1996) 427–433.
Deyo, R. and Diehl, A.,
Psychosocial predictors of disability in patients with low back pain,
J. Rheumatol., 15 (1988) 1557–1564.
Deyo, R.A., Andersson, G., Bombardier, C., Cherkin, D.C., Keller, R.B., Lee, C.K., Liang, M.H.
Outcome measures for studying patients with low back pain,
Spine, 19 (1994) 2032s–2036s.
Fairebank, J., Couper, J., Davies, J. and O’Brien, J.,
The Oswestry low back pain disability questionnaire,
Physiotherapy, 66 (1980) 271–273.
Fordyce, W., Brockway, A., Bergman, J. and Spengler, D.,
Acute back pain: a control-group comparison of behavioral vs traditional management methods,
J. Behav. Med., 9 (1986) 127–140.
Back pain and sciatica,
N. Engl. J. Med., 318 (1988) 291–300.
Moshby-Year Book, St. Louis, MO, 1991.
Hoogen van den, H., Koes, B.W., Devill, W., van Eijk, J.T.M. and Bouter, L.M.,
The prognosis of low back pain in general practice,
Spine, 22 (1997) 1515–1521.
Hurwitz EL, Aker PO, Adams AH, Meeker WC, Shekelle PG.
Manipulation and Mobilization of the Cervical Spine: A Systematic Review of the Literature
Spine (Phila Pa 1976) 1996 (Aug 1); 21 (15): 1746–1760
Klenerman, L., Slade, P.D., Stanley, I.M., Pennie, B., Reilly, J.P.,
The prediction of chronicity in patients with an acute attack of low back pain in a general practice setting,
Spine, 20 (1995) 478–484.
Koes, B., Assendelft, W., van der Heijden, G. and Bouter, L.,
Spinal manipulation for low back pain: an updated systematic review of randomized clinical trials,
Spine, 21 (1996) 2860–2873.
Lanier, D.C. and Stockton, P.,
Clinical predictors of outcome of acute episodes of low back pain,
J. Fam. Pract., 27 (1988) 483–489.
Leboeuf-Yde, C., Lauritsen, J. and Lauritzen, T.,
Why has the research for causes of low back pain largely been inconclusive?,
Spine, 22 (1997) 877–881.
Toward a theory of patient satisfaction, Soc.
Sci. Med., 16 (1982) 577–582.
Chronic back pain: integrating psychological and physical therapy – an overview,
Behav. Med., 20 (1994) 101–104.
Meade TW, Dyer S, Browne W, et al.
Low Back Pain of Mechanical Origin: Randomised Comparison of Chiropractic
and Hospital Outpatient Treatment
British Medical Journal 1990 (Jun 2); 300 (6737): 1431–1437
Murphy, K.A. and Cornish, R.D.,
Prediction of chronicity in acute low back pain,
Arch. Phys. Med. Rehabil., 65 (1984) 334–337.
Nicholas, M., Wilson, P. and Goyen, J.,
Comparison of cognitive-behavioral group treatment and an alternative non-psychological treatment
for chronic low back pain,
Pain, 29 (1992) 225–238.
Prognostic indicators in low back pain, J. R. Coll.
Gen. Pract., 31 (1981) 209–216.
Pfingsten, M., Hildebrandt, J., Leibing, E., Franz, C. and Saur, P.,
Effectiveness of a multimodal treatment program for chronic low-back pain,
Pain, 73 (1997) 77–85.
Postacchini, F., Facchini, M. and Palieri, P.,
Efficacy of various forms of conservative treatment in low back pain: a comparative study,
Neuro-Orthopedics, 6 (1988) 28–35.
Ransford, A., Cairns, D. and Mooney, V.,
The pain drawing as an aid to the psychological evaluation of patients with low-back pain,
Spine, 1 (1976) 127–134.
Roland, M. and Morris, R.,
A study of the natural history of low-back pain. Part II: development of guidelines for trials
of treatment in primary care,
Spine, 8 (1983) 145–150.
Scott, J. and Huskinsson, E.,
Graphic representation of pain,
Pain, 2 (1976) 175–184.
Singer, J., Gilbert, J., Hutton, T. and Taylor, D.,
Predicting outcome in acute low-back pain,
Can. Fam. Physician, 33 (1987) 655–659.
Skargren, E., Carlsson, P. andO¨ berg, B.,
One-year follow-up of a cost and effectiveness comparison of chiropractic and physiotherapy as
primary treatment for back pain: subgroup analysis, recurrence and additional health care utilization,
Spine, (1998) in press.
Skargren, E.,O¨ berg, B., Carlsson, P. and Gade, M.,
Cost and effectiveness analysis of chiropractic and physiotherapy treatment for low-back and neck pain:
Spine, 22 (1997) 2167–2177.
Strong, J., Ashton, R., Cramond, T. and Chant, D.,
Pain intensity, attitude and function in back pain patients,
Aust. Occup. Ther. J., 37 (1990)
Tabachnick, B. and Fidell, L.S.,
Using Multivariate Statistics,
Harper Collins College publishers, New York, 1996.
Thompson, A. and Sunol, R.,
Expectations as determinants of patient satisfaction: concepts, theory and evidence,
Int. J. Quality Health Care, 7 (1995) 127–141.
Tulder van, M.W., Koes, B.W. and Bouter, L.M.,
Conservative treatment of acute and chronic low back pain: a systematic review of randomized controlled
trials of the most common interventions,
Spine, 22 (1997) 2128–2156.
Educational and behavioral interventions for back pain in primary care,
Spine, 21 (1996) 2851–2859.
A new clinical model for the treatment of low back pain,
Spine, 12 (1987) 632–644.
Waddell, G., Newton, M., Henderson, I., Somerville, D. and Mian, C.,
A fear-avoidance beliefs questionnaire (FABQ) and the role of fear-avoidance beliefs in
chronic low back pain and disability,
Pain, 52 (1993) 157–168.
Von Korff, M., Deyo, R., Cherkin, D. and Barlow, W.,
Back pain in primary care: outcomes at 1 year,
Spine, 18 (1993) 855–862.
Return to LOW BACK PAIN
Return to CLINICAL PREDICTION RULE