Spine (Phila Pa 1976) 2006 (Feb 15); 31 (4): 468–472 ~ FULL TEXT
Hestbaek L, Leboeuf-Yde C, Kyvik KO, Manniche C
Back Research Center, Backcenter Fynen,
Clinical Locomotion Science,
University of Southern Denmark,
STUDY DESIGN: Prospective study with 8–year follow-up.
OBJECTIVE: To describe the evolution of low back pain from adolescence into adulthood.
SUMMARY OF BACKGROUND DATA: High prevalence rates of low back pain among children and adolescents have been demonstrated in several studies, and it has been theorized that low back pain in childhood may have important consequences for future low back pain. It is important to understand the nature of such a link if effective preventive programs are to be established.
METHODS: Almost 10,000 Danish twins born between 1972 and 1982 were surveyed by means of postal questionnaires in 1994 and again in 2002. The questionnaires dealt with various aspects of general health, including the prevalence of low back pain, classified according to number of days affected (0, 1–7, 8–30, >30).
RESULTS: Low back pain in adolescence was found to be a significant risk factor for low back pain in adulthood with odds ratios as high as four. We also demonstrated a dose-response association: the more days with low back pain at baseline, the higher the risk of future low back pain. Twenty-six percent of those with low back pain for more than 30 days during the baseline year also had more than 30 days with low back pain during the follow-up year. This was true for only 9% of the rest of the sample.
CONCLUSIONS: Our study clearly demonstrates correlations between low back pain in childhood/adolescence and low back pain in adulthood. This should lead to a change in focus from the adult to the young population in relation to research, prevention, and treatment.
From the FULL TEXT Article:
Present evidence suggests that low back pain (LBP) is rarely
a self-limiting disorder but characterized by unpredictable
variations in pain status, with temporary, rather than permanent,
remissions. The few studies in which the nature of
such variations have been investigated [1–3] all confirm the
fluctuating pattern of LBP. These studies were based on
adult populations; thus, the knowledge of the natural
course of LBP appearing already in adolescence is limited.
During the past decades, there has been an increasing
awareness of LBP in children and adolescents,  an area
that previously was largely ignored. Several studies have
shown that LBP occurs to a large extent in the early years
of life. Combs and Caskey even showed “back pain with
no organic cause” to be the most frequent primary diagnosis
in a study of pediatric patients.  In the cohort used
for the present study, the cumulative incidence of LBP
was previously reported to increase considerably from
12 to 22 years with more than 50% of 20–year-olds having
experienced LBP. For subjects 23 to 41 years of age,
there was only a minor increase in prevalence.  Lifetime
prevalence rates above 50% have been reported for even
earlier ages in other populations. [7, 8]
Several cross-sectional studies of LBP in young people
have been performed. However, there exist only a few
longitudinal studies, and two of these have focused on
risk factors for LBP rather than on the course of the
disorder. [9, 10] To our knowledge, children and teenagers
have been followed into adulthood to study the association
between LBP in adolescence and LBP in adulthood
in three longitudinal studies. [11–13] Both Hellsing and
Bryngelsson11 and Darre et al  found that LBP at age
18, the time of military enlistment, significantly increased
the risk of LBP at age 40 and 30, respectively,
while Harreby et al found that 90% of schoolchildren
with LBP suffered from LBP 25 years later. 
Thus, there is a need to establish the extent and the
nature of the association between LBP in adolescence
and LBP in adulthood. It has been theorized that LBP in
childhood may have important consequences for chronic
LBP in adulthood.  This study aims to describe the evolution
of LBP as youngsters grow into adulthood.
The objective is to describe the change of LBP over an
8–year period as subjects change from youngsters to adults.
Besides the general description of change over time, the
specific question of whether persistent LBP in adolescence
predicts persistent LBP in adulthood will be addressed.
Materials and Methods
The Danish Twin Register is the most comprehensive
population-based twin register in the world, spanning a
period of more than 100 years. The twins of interest for this
study were born between 1972 and 1982. They were identified
through the Danish civil registration system and represent 95%
of the twins born during that period. The twins can be regarded
as representative of the general population since they have previously been shown to have the same mortality rate  and the
same prevalence of various diseases as the population at large,
e.g., insulin-dependent diabetes,  hand eczema,  asthma and
allergic rhinitis,  and LBP.  The database is described in detail
elsewhere.  Since the population can be regarded as representative
for the general population, the register provides a
valuable population-based study population for various purposes.
The fact that they are twins is of no interest for the
purpose of this study. In 1994, when the sample was 12 to 22
years old, comprehensive questionnaires were sent to twins
who previously had agreed to participate in future studies
(96%). The questionnaires contained items relating to disease,
health, and health-related behavior. Similar questionnaires
were sent to the same population in 2002, when they were 20 to
30 years of age. The variable of interest for this paper was the
number of days with LBP during the past year (LBP-days). The
exact wording of the question was: “How many days have you
altogether had trouble with the lower part of your back during
the past year?” The possibilities for answering were: “0 days,”
“1–7 days,” “8–30 days,” “more than 30 days, but not daily,”
or “daily.” The group with daily pain was too small for reasonable
analyses; therefore, the two groups “more than 30
days, but not daily” and “daily” were combined into one
group: “more than 30 days.” Furthermore, the question,
“Have you ever had trouble with the lower part of your back?”
(LBP-ever), was used for validation-purposes.
The questions regarding LBP were modeled on the
Nordic Back Pain questionnaire,  which has been validated
previously.  The reliability of LBP questions from the 1994
omnibus has previously been considered satisfactory through
identification of logical errors.  Similarly, analyses of validity
were performed on the data from the 2002 omnibus by crosstabulating
LBP-days with LBP-ever. Finally, to get an indication
of the validity of lifetime recall in a young population, the
response to LBP-ever in 1994 was compared with the response
to the same question in 2002.
Responders and nonresponders at follow-up were compared with regard to age, gender, and LBP
status at baseline.
Persistent LBP (LBP-long) was defined as LBP for
more than 30 days during the previous year and LBP during the
past year (LBP-year) as LBP for more than zero days during the
previous year. The distributions between LBP categories in
1994 and in 2002 are presented in Tables 1 and 2 as proportions.
Odds ratios for having LBP-year and, specifically, for
having persistent LBP at follow-up were calculated for subgroups,
based on LBP status in 1994, by means of logistic
regression and adjusted for age and gender.
All analyses were done using STATA 8.0 statistical software
package. We were not testing any hypotheses but merely describing
the pattern of LBP in our population; thus, no significance
testing was performed. Instead, all results are presented with 95%
confidence intervals as they provide the key for referring from our
sample back to the general Danish population of that age. 
Description of Cohort
In 1994, the overall response rate was 84%, of which
98% answered the LBP questions, resulting in 9,569 participants.
Of those 9,569 individuals, 71% responded in
2002. Ninety-six percent of these answered the question
regarding number of days with LBP, leaving 6,540 individuals
for longitudinal analyses, i.e., 68% of the optimal
The mean age of this part of the cohort was 17.37
years (SD = 3.14 years; range, 12–22 years) in 1994 and
the male/female ratio was 43:57. The prevalence of LBP
in relation to age group and gender is shown in Table 1
for both 1994 and 2002.
Table 1 shows that the 1–year prevalence of LBP increases
for the youngest group and decreases for the
older groups, whereas the prevalence of persistent LBP
increases for all age groups, although this increase is not
significant for the oldest group.
Cross-tabulation of LBP-ever with LBP-days showed an
unacceptable response (“No” to “Have you ever had
trouble with the lower part of your back?” and more
than zero to “How many days have you altogether had
trouble with the lower part of your back during the past
year?”) in 1.4% of cases in 1994, and 0.3% of cases in
2002. We considered this to be an acceptable level of
invalid answers (logical errors). As for long-term recall,
35% of those reporting having had LBP in 1994 claimed
never to have experienced LBP in 2002. However, this
variable was only used for validation purposes in this
Comparisons between responders (n = 6,540) and nonresponders
(n = 3,029) at follow-up in 2002 demonstrate
similar age (responders: mean age = 17.00; SD
3.13; range, 12–22; nonresponders: mean age = 17.37
years; SD = 3.14 years; range, 12–22 years) and a similar
distribution of persistent LBP in 1994 (6%; range, 6%–7%; and 6%; range, 5%–7%, for responders and nonresponders,
respectively, P = 0.262). However, there is a
significant overrepresentation of females (57%; range,
55%–58%; vs. 39%; range, 37%–41%; P < 0.001) at
follow-up and likewise, subjects with LBP-year in 1994 are
overrepresented among responders in 2002 (35%; range,
34%–37%; vs. 29%; range, 28%–31%; P < 0.001).
Transition in LBP Status
The most stable group, as seen in Table 2, is the pain-free
group with 69% remaining pain-free from baseline to follow-
up. Among the subjects with persistent LBP at baseline,
there is a disproportionately large proportion of subjects
with persistent LBP in 2002 (26% vs. 9% for those with
none or nonpersistent LBP at baseline) and likewise a relatively
small proportion being pain free in 2002 (33% vs.
63% for those with none or nonpersistent LBP at baseline).
LBP in Adolescence as a Predictor for Persistent LBP in Adulthood
Odds ratios stratified for age indicate no difference between
ages. Thus, results are not presented stratified.
Likewise, there is no gender difference. To illustrate this,
odds ratios are presented both raw and adjusted for age
and gender in Table 3.
Table 3 shows a significant increase in the probability of
getting LBP-year as well as persistent LBP at follow-up in
case of LBP in adolescence. Thus, LBP in adolescence is a
significant risk factor for LBP in adulthood. Furthermore,
the risk increases with increasing duration of LBP
The pattern demonstrated in this study is similar to that
demonstrated in adults, [1–3] with one exception. There
are, despite a large number remaining in the group with
more than 30 days of LBP, 33% with persistent LBP at
baseline who are symptom-free at follow-up, compared
with only 9% found in an adult population.  This might
reflect a specific subgroup of LBP that is active in early
years but nonsymptomatic in adults. Previous populationbased
MRI studies of 13– and 40–year-olds showed that
endplate defects are associated with pain in children but not
in adults. [25, 26] We therefore speculate that this could represent
one etiologic explanation of the relatively large group
of children, who recover from persistent LBP.
In accordance with previous studies, [11–13] we also
found LBP at baseline to be a significant predictor for
LBP at follow-up. We further demonstrated a doseresponse
association between number of days with LBP
at baseline and the occurrence of LBP at follow-up, i.e.,
the more days with LBP at baseline, the higher the risk of
future LBP. To our knowledge, this has not been shown
Our cohort has two major strengths: its large sample
size and the fact that it covers the transition from childhood
into adulthood. It could be argued that the spread
in age of the cohort represents a problem, i.e., the development
from 12 to 20 years of age is significantly different
from that of 22 to 30. However, when results are
stratified for age, there is no difference between groups;
therefore, we do not consider this to devaluate our results.
The response rate of our study is not completely
satisfying, but the fact that 96% of responders answered
the LBP-questions diminishes the risk of disease-specific
bias. With regard to representativeness, the overrepresentation
of people with LBP-year at baseline must be
considered, when comparing the 1–year prevalence in
1994 with that from 2002. This could lead to an inflated
estimate for 2002 compared with the target population.
Fortunately, the equal distribution of persistent LBP
among responders and nonresponders lend credibility to
our results relating to this variable. Likewise, the transition
in LBP status is equal for males and females; thus,
the female overrepresentation is unlikely to influence our
Finally, we found long-term recall to be of limited
value. Researchers should take the large number of forgotten
episodes of LBP (at least 35%) into account in
future studies. These findings are in line with other studies,
where lifetime history of LBP has proven unreliable.
Burton et al reported that 60% of adolescents forgot
previous episodes of pain ; and in a study by Szpalski et
al,  50 of 51 children reporting LBP within the past 2
years also reported having had LBP the day before examination.
As it is highly unlikely that 98% of all LBP
cases during the previous 2 years involved the same day,
such a pattern of reporting must indicate severe recall bias.
Therefore, we feel confident that the problem is related to
long-term recall, rather than to the quality of the used questionnaire.
Therefore, the poor validity of the variable does
not affect the validity of our study, since the lifetime prevalence
variable was used only for validation purposes in our
Our results leave no doubt that LBP, and especially
persistent LBP, at an early age is a strong predictor of
persistent LBP later in life with odds ratios as high as 4,
and that the importance of early LBP as a risk factor is
independent of both age and gender. This indicates that
our findings are not specifically linked to adolescence but
can be applied to all incidence cases of LBP regardless of
age, bearing in mind that the majority of incidence cases
occur before the age of 20 years. In clinical practice, this
line of thought could lead to more attention to first-time
cases of LBP, regardless of severity, to prevent later chronicity.
It might be counterproductive to postpone treatment/
prevention until the problems become more severe
We have aimed for a better understanding of the importance
of LBP in adolescence. The age of our cohort and
the length of follow-up allow us to investigate the change
from adolescence into adulthood. Our study showed
strong correlations between LBP in adolescence and LBP
in adulthood, especially for persistent LBP, which demonstrated
a four times increase in risk.
A total of 35% of those reporting low back pain
in 1994 claimed in 2002 never to have experienced
low back pain.
Odds ratio for having low back pain in 2002 in
case of persistent low back pain in 1994 was 4.29
(95% confidence interval, 3.45–5.34).
A total of 26% of those with persistent low back
at baseline also had persistent low back pain at
follow-up; this was true for 9% of the rest.
A total of 69% of those pain free at baseline were
also pain free at follow-up; this was true for 46%
of the rest.
Cedrashi S, Robert J, Goerg D, et al.
Is chronic non-specific low back pain chronic? Definitions of a problem and problems of a definition.
Br J Gen Pract 1999;49:358–62.
Hestbaek L, Leboeuf-Yde C, Engberg M, Lauritzen T, Bruun NH, Manniche C.
The Course of Low Back Pain in a General Population.
Results From a 5-year Prospective Study
J Manipulative Physiol Ther. 2003 (May); 26 (4): 213–219
Waxman R, Tennent A, Helliwell P.
A prospective follow-up study of low back pain in the community.
Balague´ F, Dudler J, Nordin M.
Low-back pain in children.
Lancet 2003; 361:1403–4.
Combs JA, Caskey PM.
Back pain in children and adolescents: a retrospective review of 648 patients.
South Med J 1997;90:789–92.
Leboeuf-Yde C, Kyvik KO.
At What Age Does Low Back Pain Become a Common Problem?
A Study of 29,424 Individuals Aged 12-41 Years
Spine (Phila Pa 1976). 1998 (Jan 15); 23 (2): 228–234
Kovacs FM, Gestoso M, Gil del Real MT, et al.
Risk factors for non-specific low back pain in schoolchildren and their parents: a population based study.
Sheir-Neiss GI, Kruse RW, Rahman T, et al.
The association of backpack use and back pain in adolescents.
Feldman DE, Shrier I, Rossignol M, et al.
Risk factors for the development of low back pain in adolescence.
Am J Epidemiol 2001;154:30–6.
Szpalski M, Gunzburg R, Balague´ F, et al.
A 2-year prospective longitudinal study on low back pain in primary school children.
Eur Spine J 2002;11: 459–64.
Hellsing AL, Bryngelsson IL.
Predictors of musculoskeletal pain in men: a twenty-year follow-up study from examination at enlistment.
Spine 2000; 25:3080–6.
Darre EM, Biering-Sørensen F, Deis A, et al.
Rygbesvær under aftjening af værnepligt– betydning for senere rygproblemer.
[Back problems during military service: significance for later back problems].
Ugeskr Laeger 1999;161: 1926–30.
Harreby M, Kjer J, Hesselsøe G, et al.
Epidemiological aspects and risk factors for low back pain in 38-year-old men and women: a 25-year prospective cohort-study of 640 Danish school children.
Eur Spine J 1996;5:312–8.
Watson KD, Papageorgiou AC, Jones GT, et al.
Low back pain in schoolchildren: occurrence and characteristics.
Christensen K, Vaupel J, Holm NV, et al.
Twin mortality after age 6: fetal origins hypothesis versus twin method.
Kyvik KO, Green A, Beck-Nielsen H.
Concordance rates of insulin dependent diabetes mellitus: a population-based study of young Danish twins.
Bryld LE, Agner T, Kyvik KO, et al.
Hand eczema in twins: a questionnaire investigation.
Br J Dermatol 200;142:298–305.
Genetic and environmental influence on asthma: A population-based study of Danish twins, PhD thesis.
Odense, Denmark, University of Southern Denmark, 1999.
The natural course of low back pain and early identification of high-risk populations, PhD thesis.
2003, Odense, Denmark, University of Southern Denmark, 2003.
Kyvik KO, Christensen K, Skytthe A, et al.
The Danish Twin Register.
Dan Med Bull 1996;43:467–70.
Kuorinka I, Jonsson B, Kilbom A, et al.
Standardized Nordic questionnaires for the analysis of musculoskeletal symptoms.
Appl Ergon 1987; 18:233–7.
Biering-Soerensen F, Hilden J.
Reproducibility of the history of back trouble.
Hestbæk L, Leboeuf-Yde C, Kyvik KO, et al.
Comorbidity with low back pain: a cross-sectional population-based survey of 12–22 year olds.
Swinscow TDV, Campbell MJ.
Statistics at Square One.
BMJ Books, London, England; 2002:41.
Kjaer P, Leboeuf-Yde C, Sorensen JS, et al.
An epidemiologic study of MRI and low back pain in 13-year-old children.
Kjaer P, Leboeuf-Yde C, Korsholm L, et al.
Magnetic resonance imaging and low back pain in adults: a diagnostic imaging study of 40-year-old men and women.
Burton AK, Clarke RD, McClune TD, et al.
The natural history of low-back pain in adolescents.
Return to the PEDIATRICS Section
Return to the LOW BACK PAIN Section