European Journal of Pain 2017 (May); 21 (5): 866–873 ~ FULL TEXT
Manogharan S, Kongsted A, Ferreira ML, Hancock MJ.
Faculty of Medicine and Health Science,
Sydney, NSW, Australia.
Low back pain (LBP) in older adults is poorly understood because the vast majority of the LBP research has focused on the working aged population. The aim of this study was to compare older adults consulting with chronic LBP to middle aged and young adults consulting with chronic LBP, in terms of their baseline characteristics, and pain and disability outcomes over 1 year.
METHODS: Data were systematically collected as part of routine care in a secondary care spine clinic. At initial presentation patients answered a self-report questionnaire and underwent a physical examination. Patients older than 65 were classified as older adults and compared to middle aged (45-65 years old) and younger adults (17-44 years old) for 10 baseline characteristics. Pain intensity and disability were collected at 6 and 12 month follow-ups and compared between age groups.
RESULTS: A total of 14,479 participants were included in the study. Of these 3,087 (21%) patients were older adults, 6,071 (42%) were middle aged and 5,321 (37%) were young adults. At presentation older adults were statistically different to the middle aged and younger adults for most characteristics measured (e.g. less intense back pain, more leg pain and more depression); however, the differences were small. The change in pain and disability over 12 months did not differ between age groups.
CONCLUSIONS: This study found small baseline differences in older people with chronic LBP compared to middle aged and younger adults. There were no associations between age groups and the clinical course.
SIGNIFICANCE: Small baseline differences exist in older people with chronic low back pain compared to middle aged and younger adults referred to secondary care for chronic low back pain. Older adults present with slightly less intense low back pain but slightly more intense leg pain. Changes in pain intensity and disability over a 12 month period were similar across all age groups.
From the FULL TEXT Article:
Chronic low back pain (CLBP) is responsible for a
substantial burden on the community, economy and
the health care systems (Walker et al., 2003; Murray
et al., 2013). Back pain is the most common health
condition forcing older Australians to retire involuntarily
and this problem reduces Australia’s GDP by
$3.2 billion per annum (Schofield et al., 2008).
According to the World Health Organisation low
back pain is one of the most disabling conditions
among the elderly (World Health Organization,
Although LBP in older adults is recognized as a
common and physically disabling condition it is
poorly understood because the vast majority of the
LBP research has focused on the working aged population
from 18 to 65 (Schiottz-Christensen et al.,
1999; Paeck et al., 2014). Patients aged over 65 are
often excluded from studies and only a small number
of studies have studied LBP in older adults
(Scheele et al., 2012; Paeck et al., 2014). The findings
in younger populations cannot necessarily be
generalized to older people with LBP (Scheele et al.,
2014). The United Nations predicts by 2050 people
aged 60 and older will account for 20% of the
world’s population and one-fifth of this population
will be above 80 years of age (Lutz et al., 2008).
With the increase in the ageing population and the
lack of studies currently available that focus on LBP
in older people, large cohort studies are needed to
gain a better understating of LBP in the older population.
Previous studies are mixed in their findings
regarding LBP in older adults and how it compares
to younger adults. Some studies suggest that older
people experience a higher prevalence of severe low
back pain and/or disabling episodes but report less
frequent benign or mild pain (Cayea et al., 2006;
Scheele et al., 2014). A cross-sectional populationbased
study revealed that those aged over 80 years
were twice as likely to be disabled by an episode of
LBP as those aged 25–40 (Macfarlane et al., 2012).
The international BACE (Back Complaints in the
Elders) consortium is recruiting a cohort of patients
>55 years of age presenting for care with a new episode
of LBP (Scheele et al., 2011). This cohort will
provide valuable data on LBP in older people; however,
it does not include younger adults to enable
direct comparison within the same study.
Studies that include both older and younger adults
with LBP are ideal to investigate potential differences
in low back pain between older and younger people.
Potential difference in LBP in older people could
manifest as either differences in presentation and/or
differences in prognosis. A better understanding of
these potential differences between older and
younger people with low back pain is critical to
understanding how to optimize management of LBP
in older people and whether or not a different
approach is required compared to younger adults.
Therefore, the broad aim of the study is to compare
LBP in older adult patients with LBP in younger
adult patients to assess if important differences exist.
Specifically, the aims of the study are to
(1) compare the initial presentation of older adults and younger
adults with chronic LBP attending a secondary spine
care centre in Denmark and
(2) compare the prognosis over 1 year of older adults and younger adults
in this cohort.
This study is based on data collection from the SpineData
database (Kongsted et al., 2012, 2014; Kent
et al., 2015). It includes both a cross-sectional study
of patient presentation and a longitudinal study of
patient outcomes. The regional ethics committee of
Southern Denmark (Project ID S-200112000-29)
reviewed the protocol for the original study and stated
that it according to Danish law did not need ethical
approval due to the observational design.
Data were collected as part of routine clinical practice
in a secondary, non-surgical, outpatient Spine
Center in The Region of Southern Denmark (Kongsted
et al., 2012, 2014). The Spine Center performs
multidisciplinary, structured physical examinations
and treatment planning for patients with spinal pain
referred from general practitioners, chiropractors and
medical specialists. Patients referred to the centre
have received treatment in primary care without a
satisfactory outcome. Participants were asked to participate
in a follow-up questionnaire at 6 and 12-months. An electronic questionnaire was accessed
through a link sent by email or people could opt to
have a postal survey.
Data were collected in the Spine Center’s electronic
clinical registry named the SpineData database. Participants
answered a comprehensive self-reported
questionnaire on a touch screen in the waiting area
preceding their first consultation. The data were
entered directly into the SpineData database.
We used the following criteria to select patients from
the SpineData database for this study. All patients in
the database aged 17 or older who were seen at The
Spine Centre between January 1st 2012 and December
31st 2013 with LBP as their main complaint and
who gave informed consent for their data to be used
for scientific purposes were included in this study.
Only patients with follow-up data for either 6 or
12 months were included in the longitudinal study
The key variable of interest in this study was participant
age. Patients were categorized into three age
groups. Older age was defined as those >65 years of
age at the time they presented to the Spine centre.
Other participants were categorized into middle aged
(45–65) and younger (18–44) to serve as two comparison
groups for the older adults.
Outcomes for cross-sectional study (aim 1)
We investigated the differences between older adults
and the two groups of younger adults on the 10
baseline measures listed below. The evidence base
for the SpineData variables has been previously published
(Kent et al., 2015).
Outcomes for longitudinal study (aim 2)
Low back Pain Intensity and disability were measured
at 6 and 12-month follow-up using the same
methods as described for baseline measures.
Duration of present episode.
Duration was measured in months and categorized into three groups 0–3 months,
3–12 months and >12 months.
Previous LBP episodes.
Scored as yes or no in response to question ‘Have you had previous low back pain episodes?’
Intensity of LBP.
Average of ‘LBP now’, ‘worst in the last 14 days’ and ‘typical in the last 14 days’
on 0–10 Numerical Rating Scales (NRS).
Leg pain intensity.
Measured in a similar manner to intensity of LBP.
Proportion with leg pain below the buttock.
Using the body charts completed at baseline patients were classified as
having ‘No leg pain’ or ‘leg pain below the buttock’.
Self-reported disability was measured with the 23-item Roland Morris Disability Questionnaire (RMDQ)
converted to a score out of 100 (where 0% = no activity limitation and 100% = maximum activity limitation).
Patients needed to answer at least 17 questions for the score to be included and the %-score calculated based
on the number of items completed (Kent and
Two Primary Care Evaluation of Mental Disorders screening questions were used to determine
if depression was present.
(1) ‘During the past month have you often been bothered by feeling down, depressed or hopeless?’ and
(2) ‘During the past month, have you often been bothered by little interest or pleasure in doing things?’
Both questions were measured on the 0–10 NRS, where 0 represented never and 10 represented always.
A mean was calculated, and depression was considered to be present if both scores were >6 out of 10.
These cut off scores have been validated in this setting relative to the Major Depression Inventory
(Kent et al., 2014).
Fear of movement.
Measured using 0–10 NRS (proportion with a total score on two screening questions
from the Fear Avoidance Belief Questionnaire equal to or above 14). This threshold has been validated
in this setting relative to the physical activity subscale of the Fear Avoidance Belief Questionnaire
(Kent et al., 2014).
Self-reported general health.
Assessed with the EuroQol Health Thermometer where participants
rate their own health at the present time. It was expressed on a scale of 0–100, where 0 was the
worst health state and 100 is the best health state
(Williams and Williams, 1990).
Cross-sectional part (aim 1)
Dichotomous baseline characteristics were described
as proportions (95% confidence intervals (CI)) and
continuous baseline characteristics were described as
means (95% CI) or medians (IQR) if continuous
variables were not normally distributed. Pairwise
comparisons between older age group and the two
younger age groups for each baseline characteristic
were tested using t tests for continuous normally
distributed variables, and a Mann–Whitney U-test
for variables with a non-normal distribution. Chi-
Square Tests were used to compare proportions.
Group differences were considered significantly different
at p < .05.
Longitudinal part (aim 2)
Longitudinal models using a Linear Mixed Model
were used to analyse if older patients had different
outcomes to the two younger adult groups in terms
of pain and disability at 6 and 12-months. These
models take into account that the repeated outcome
measures are correlated. The outcome measures
(pain and activity limitation) were treated as continuous
measures. The age variable was introduced in
the model as a 3-level categorical variable with old
age as the reference category. The time variable was
introduced into the model as a categorical number of
months. The primary analysis involved a simple
model with no other covariates. We also ran 2 secondary
analyses where we first entered gender and
duration of low back pain, and then also added
depression and fear of movement, to assess if these
explained any differences observed between different
age groups on pain and disability outcomes. Pairwise
comparisons of absolute scores on 6- and 12-months
outcomes between the older age group and the 2
younger age groups were performed in the same
way as the baseline comparisons. Analyses were performed
using STATA 14.0 (StataCorp, TX, USA)
4905 Lakeway Drive; College Station, Texas 77845
Between January 1st 2012 and December 31st 2013
14,567 people presented to the spine centre with
LBP as their main complaint and were registered in
the database. After excluding those aged 16 or
younger 14,479 patients were included in the study.
Of these 5321 (37%) were aged 17–44 (median 36,
IQR 29–41), 6071 (42%) were aged 45–65 (median
54, IQR 49–59) and 3087 (21%) patients were aged
>65 (median 72, IQR 68–76). There were slightly
more females (55%) and the mean age was 51.
Study aim 1: baseline differences
Patient baseline characteristics for all three age
groups are summarized in Table 1. The older aged
group was statistically different to the younger aged
group for almost all baseline measures (proportion
female, back pain intensity, leg pain intensity, previous
episodes, pain below buttock, general health,
disability, depressive symptoms and fear of movement).
The older age group was statistically different
to the middle aged group for the baseline measures
of proportion female, back pain intensity, leg pain
intensity, previous episodes, pain below buttock and
depressive symptoms. Older adults tended to have
slightly lower average back pain intensity, but
slightly higher average leg pain intensity and disability.
A slightly higher percentage of older adults had
pain extending below the buttock, and a smaller
proportion was fear avoidant or depressed as compared
to the young adult group. However, all differences
were small in size.
Study aim 2: 6 and 12 month outcome differences
Either the 6 month or the 12 month follow-up was
completed by 48% (6995/14,479) and 46% (6705/
14,479) of participants for pain intensity (37% at
6 months and 34% at 12 months) and disability
(36% at 6 months and 33% at 12 months), respectively,
and included in the longitudinal analyses. The
proportion of patients not responding to any of the
follow-ups was largest in the youngest age group
and smallest in the older age group (younger aged
65% drop-out, middle aged 46%, older age 37%;
p < 0.01). Differences in baseline characteristics
between responders and drop-outs were small
although drop-outs were less likely to have had previous
episodes and more likely to have depressive
symptoms or fear of movement (Table 2). The differences
between responders and drop-out were not
statistically different across age groups apart from
women being less likely to drop-out in the youngest
age group and more likely to drop-out in the oldest
age group. Also, middle aged and older adults who
dropped out had on average longer episode durations
than the responders, whereas that was not the
case in the young group (Table 2).
Average pain intensity for the three different age
groups at each time point are provided in Table 3
and represented visually in Figure 1. Older adults had
significantly lower average back pain intensity than
both younger and middle aged participants; however,
the differences were small. The interaction
between time and age in the longitudinal linear
mixed model was non-significant when comparing
older participants to either middle aged (p = 0.967)
or younger participants (p = 0.254), suggesting that
age did not result in a difference in change in pain
intensity over time. The findings were very similar
in both secondary analyses where we first entered
gender and duration of the episode, and then also
added depression and fear of movement into the
Average disability (RMDQ) for the three different
age groups at each time point is provided in Table 3
and represented visually in Figure 2. Older participants
had significantly more disability than younger adults
at 6 months but not at 12 months. There were no
significant differences between disability in older
adults and middle aged adults at either 6 or
12 months. The interaction between time and age in
the longitudinal linear mixed model was non-significant
when comparing older participants to either
middle aged (p = 0.479) or younger participants
(p = 0.129), suggesting that age did not result in a
difference in change in disability over time. The
findings were very similar in both secondary analyses
where we first entered gender and duration of
the episode, and then also added depression and fear
of movement into the model.
Discussion and conclusions
We found baseline differences existed between older
adults with chronic LBP and young or middle aged
adults with chronic LBP. However, while many differences
were statistically significant, the size of the
differences in outcome measures was small and
mostly unimportant. Older adults did present with
slightly less intense LBP but slightly more intense leg
pain. Consistent with this finding the percent of
older participants describing pain extending below
the buttock was a little higher than in the middle
aged or younger adults.
At 6- and 12-month follow-up back pain intensity
was slightly lower in the older adults compared to
both middle aged and younger adults. Disability outcomes
at 6 and 12 months were marginally higher
in older adults when compared to younger adults,
but not different to middle aged adults. Changes in
pain intensity and disability over the 12 month period
were similar across all age groups.
Strengths and weaknesses of the study
Strengths of this study include the large sample of
consecutive patients presenting for care enabling
direct comparison between older adults and 2 comparison
age groups from the same clinical population.
The data were collected as part of routine care
therefore likely increasing the generalizability of the
findings. However, our sample included Danish people
with chronic low back pain presenting to a secondary
care spine clinic and it is unclear how well
our findings would generalize to primary care or
other more diverse geographical regions.
The key limitation of this study is the low percent of
participants that completed follow-up questionnaires.
We have previously shown that patients in this cohort
who completed follow-up were very similar at baseline
to those who did not (Kent et al., 2015). There is
no consensus on what age is considered older and it is
possible the results could be different if a different
threshold for age was used. We selected >65 as this
age has commonly used to exclude older adults and as
such this is the age group with low back pain for
whom little is known.
A further limitation of this
study is that some of the baseline characteristics we
investigated (e.g. depression) were collected using single
item screening questions rather than multi item
tools. These screening questions have been demonstrated
to represent full questionnaires well and are
an effective way to gather information when questions
on many aspects of health are asked in a clinical
setting (Kent et al., 2014). It is possible that there
were unmeasured confounders that may influence
the relationship between age and the study outcomes;
however, our study aimed to describe any differences
observed between age groups but did not aim to
explore causal pathways.
Comparison to other studies
A recent publication from the Back pain Outcomes
Longitudinal Data (BOLD) cohort presented baseline
characteristics of over 5000 adults with low back
pain presenting for care (Jarvik et al., 2014). Despite
the different setting the baseline pain intensity was
relatively similar (5.0 on a 0–10 point scale) to our
secondary care cohort. As the BOLD cohort does not
include a younger comparison group it is not possible
to assess if the small differences we observed in
most baseline variables also occurred in this primary
The BACE cohort study examined the differences
in baseline characteristics between patients with low
back pain aged 55–74 years and those aged
≥75 years (Scheele et al., 2014). They found patients
aged >75 years reported more disability and more
psychological health problems. These findings are
somewhat different to our findings that disability
was no different between our middle aged group
and older group and also that presence of depression
was lowest in our older group. The differences might
be explained by the different age threshold used, the
different setting or the different duration of participants’
back pain (only approximately 25% of participants
in BACE had back pain lasting longer than
It is possible that low back pain in those
over 75 results in greater disability than in our group
who were over 65. The population-based MUSICIAN
study (Macfarlane et al., 2012) reported that while
prevalence of LBP peaked at 41–50 years of age the
rates of more severe disabling LBP increased with
age and were approximately twice as high in those
over 80 years of age compared to those 40 or
younger. This study also reported that older people
were less likely to have been referred to physiotherapy
or a specialist which could explain some differences
between our clinical cohort and this
population-based cohort. In our secondary care
cohort, the proportion of people aged over 65 was
21%, however, we do not know how this compares
with people presenting to primary care or whether
referral to Danish secondary care is triggered by
other factors in older compared to younger adults.
Meaning of the study
The small size of any differences identified between
our older group and younger adults suggest that
while older adults with LBP do appear to have some
differences to younger adults with LBP the differences
may not be that important. The slightly higher
prevalence and intensity of leg pain may represent a
greater degree of neurological involvement in older
adults due to degenerative stenosis. We found no
differences in change over time for either pain or
disability suggesting the prognosis of older adults
with chronic LBP is similar to that of younger adults.
Our study does not answer the question of whether
older people with low back pain respond differently
to specific interventions than younger adults with
low back pain, but the similar prognosis and minimal
differences in baseline do not support this
hypothesis. There is no strong evidence to suggest
older adults with LBP respond better to any particular
intervention than younger adults with LBP (Ferreira
et al., 2014). Despite this the MUSICIAN study
found older people were more likely to be opioid
users and less likely to be prescribed exercise than
younger adults (Macfarlane et al., 2012).
This study found statistically significant baseline differences
in older people with chronic LBP compared to
middle aged and younger adults referred to secondary
care with LBP. Older people generally had lower back
pain intensity, higher prevalence and severity of leg
pain, and lower levels of depression. However, given the
small size of these differences their clinical importance is
questionable. There were no differences in change in
pain or disability over a 12 month period between older
adults with LBP and younger adults with LBP.
All authors (1) made substantial contributions to conception
and design, or acquisition of data, or analysis and
interpretation of data; (2) drafted the article or revised it
critically for important intellectual content; (3) approved
the version submitted.
Cayea, D., Perera, S., Weiner, D.K. (2006).
Chronic low back pain in older adults: What physicians know, what they think they know,
and what they should be taught.
J Am Geriatr Soc 54, 1772–1777
Ferreira, M.L., Ferreira, P., Henschke, N., Kamper, S., Koes, B. (2014).
Age does not modify the effects of treatment on pain in patients with low back pain:
Secondary analyses of randomized clinical trials.
Eur J Pain 18, 932–938
Jarvik, J.G., Comstock, B.A., Heagerty, P.J., Turner, J.A., Sullivan, S.D., Shi, X. (2014)
Back pain in seniors: The Back pain Outcomes using Longitudinal Data (BOLD) cohort baseline data.
BMC Musculoskelet Disord 15, 134
Kent, P., Lauridsen, H.H. (2011).
Managing missing scores on the Roland Morris disability questionnaire.
Spine 36, 1878–1884
Kent, P., Mirkhil, S., Keating, J., Buchbinder, R., Manniche, C., Albert, H.B. (2014).
The concurrent validity of brief screening questions for anxiety, depression, social isolation,
catastrophization, and fear of movement in people with low back pain.
Clin J Pain 30, 479–489
B., Manniche, C. (2015).
SpineData–a Danish clinical registry of people with chronic back pain.
Clin Epidemiol 7, 369–380
Kongsted A, Kent P, Albert H, Jensen TS, Manniche C.
Patients with Low Back Pain Differ From Those Who Also Have Leg Pain or Signs of Nerve Root
Involvement - A Cross-sectional Study
BMC Musculoskelet Disord. 2012 (Nov 28); 13: 236
Kongsted, A., Vach, W., Axø, M., Bech, R.N., Hestbaek, L., Kongsted, A., Vach, W., Axø, M. (2014).
Expectation of Recovery from Low Back Pain: A Longitudinal Cohort Study Investigating
Patient Characteristics Related to Expectations and the Association Between
Expectations and 3-month Outcome
Spine (Phila Pa 1976). 2014 (Jan 1); 39 (1): 81–90
Lutz, W., Sanderson, W., Scherbov, S. (2008).
The coming acceleration of global population ageing.
Nature 451, 716–719
Macfarlane, G.J., Beasley, M., Jones, E.A., Prescott, G.J., Docking, R., Keeley, P. (2012).
The prevalence and management of low back pain across adulthood:
Results from a population-based cross-sectional study (the MUSICIAN study).
Pain 153, 27–32
Murray CJ, Vos T, Lozano R, et al.
Disability-adjusted life years (DALYs) for 291 diseases and injuries in 21 regions, 1990-2010:
a systematic analysis for the Global Burden of Disease Study 2010
Lancet 2013 Dec 15;380(9859):2197–223.
Paeck, T., Ferreira, M.L., Sun, C., Lin, C.W.C., Tiedemann, A., Maher, C.G. (2014).
Are older adults missing from low back pain clinical trials? A systematic review and meta-analysis.
Arthritis Care Res 66, 1220–1226
Scheele, J., Luijsterburg, P.A., Ferreira, M.L., Maher, C.G., Pereira, L., Peul, W.C., van Tulder, M.W. (2011).
Back complaints in the elders (BACE); design of cohort studies in primary care: An international consortium.
BMC Musculoskelet Disord 12, 193
Scheele, J., Luijsterburg, P., Bierma-Zeinstra, S., Koes, B. (2012).
Course of back complaints in older adults: A systematic literature review.
Eur J Phys Rehabil Med 48, 379–386
Scheele, J., Enthoven, W., Bierma-Zeinstra, S., Peul, W., Tulder, M., Bohnen, A. (2014).
Characteristics of older patients with back pain in general practice:
BACE cohort study.
Eur J Pain 18, 279–287
Schiottz-Christensen, B., Nielsen, G.L., Hansen, V.K., Schodt, T., Sorensen, H.T. (1999).
Long-term prognosis of acute low back pain in patients seen in general practice:
A 1-year prospective follow-up study.
Fam Pract 16, 223–232
Schofield, D.J., Shrestha, R.N., Passey, M.E., Earnest, A., Fletcher, S.L., Schofield, D.J. (2008).
Chronic disease and labour force participation among older Australians.
Med J Aust 189, 447–450
Walker, B.F., Muller, R., Grant, W.D. (2003).
Low back pain in Australian adults: The economic burden.
Asia Pac J Public Health 15, 79–87
Williams, A., Williams, A. (1990).
EuroQol – A new facility for the measurement of health-related quality of life.
Health Policy 16, 199–208
World Health Organization (2009).
Global Health Risks: Mortality and Burden of Disease Attributable to Selected Major Risks
(Geneva: WHO Press).
Return to SENIOR CARE
Return to LOW BACK PAIN