Self-reported musculoskeletal pain predicts long-term increase in general health care use: A population-based cohort study with 20-year follow-up

Abstract

Aims: Musculoskeletal pain and disability is a modern epidemic and a major reason for seeking health care. The aim of this study is to determine absolute and relative rates of care seeking over 20 years for adults reporting musculoskeletal complaints. Methods: Interview data on musculoskeletal pain reported during the past two weeks from the Danish National Cohort Study were merged with data from the Danish National Health Insurance Registry and the National Patient Registry containing information on consultations in the Danish primary and secondary care sector. Absolute and relative rates for all seeking of care with general practitioners, physiotherapists, chiropractors, outpatient hospital contacts and hospital admissions are reported for persons reporting no musculoskeletal pain and for persons reporting pain in the neck, shoulder, wrist/hands, mid back, low back, hips, knees and ankles/feet. Results: Regardless of site, persons experiencing a musculoskeletal complaint had a statistically increased risk of consulting a general practitioner when compared with persons reporting no musculoskeletal complaint. For physiotherapists and chiropractors, only persons complaining of neck pain and back pain had an increased risk of seeking care. Regardless of pain site, except for shoulder pain, persons reporting musculoskeletal pain had a statistically significant increased risk of outpatient hospital consultations and hospital admissions. Few differences were found between pain sites in relation to any of the outcomes. Conclusions: Self-report of musculoskeletal pain reported within the past two weeks predicts a statistically significant long-term increase in general use of health care services in both the primary and the secondary health care sector.

Keywords

Back pain epidemiology health care knee pain neck pain musculoskeletal population studies shoulder pain

Background

Musculoskeletal pain is common across ages, genders and social groups [1] and reporting of musculoskeletal pain has been consistently increasing over the past decades [2]. The consequences of musculoskeletal pain are profound all over the world, because musculoskeletal disorders have evolved into leading causes of disability affecting over 1.7 billion people worldwide [3], representing a 45% increase in just two decades [4].

Naturally, the costs attributed to musculoskeletal disorders are enormous. In Germany, for instance, musculoskeletal disorders now account for over 11% of the entire cost of all illnesses [5]. Similarly, Danes suffering from long-term pain contact primary care physicians 70% more often than persons without pain, and have higher consumption of prescription medication [6]. Cote et al. found that 25% of Canadian adults who reported having had pain in the neck or back during the past month had seen a provider [7], and Lonnberg found that among Danes with back pain 30% had seen a primary care provider during the past year [8]. Patients with back and neck pain are also treated in secondary care settings, and the US witnessed an increase of 629% in the use of steroid injections, a 423% increase in opioid prescription, a 307% increase in magnetic resonance imaging and a 220% increase in spinal fusions between 1994 and 2004 [9]. The provision of total knee replacements was about doubled in the US Medicare population during the first 10 years of the 21st century [10] and more than three million procedures are projected in 2030 [11].

Because musculoskeletal pain is so prevalent and mostly intermittent in the general population, it is easy to regard such aches and pains as a normal part of life and focus attention on pain when it becomes more interfering, chronic and disabling. However, even self-reported pain in the muscles and joints may have consequences, such as an increased risk of seeking of health care, and these consequences may differ between pain sites or sub-groups in the population.

We accessed data from the Danish National Cohort Study [12] and started a series of analyses aiming to elucidate factors influencing long-term care seeking. The aims of this paper are, therefore, to 1) determine absolute and relative rates of care seeking over 20 years with general practitioners, physiotherapists and chiropractors as well as hospital outpatient consultations and hospital admissions for adults reporting musculoskeletal complaints within the past two weeks when compared with adults not reporting musculoskeletal complaints at baseline and 2) investigate whether some musculoskeletal pain sites were associated with higher rates of care seeking by determining relative rates for care seeking over 20 years for persons reporting primary pain in the neck, shoulders, hand/wrists, mid back, low back, hips, knees and ankles/feet during the past two weeks at baseline.

Methods

Design, study sample and data sources

The baseline for this prospective cohort study was the 1991 interview-based health survey conducted by the National Institute of Public Health in Denmark. The institute has been conducting nationally representative surveys of the adult Danish population aged 16 years or older since 1987 [12] and the 1991 survey focused on musculoskeletal complaints. Using the Danish Civil Registration System, 5986 persons chosen at random from the adult Danish population were invited to participate.

Follow-up was based on two national registries: the Danish Health Service Register (DHSR) and the National Danish Patient Register (DPR). DHSR started in 1990 and covers a range of primary care health services including consultations with general practitioners, physiotherapists and chiropractors [13]. DHSR is based on information on payments reimbursed by the public Danish National Health Insurance system. Not all consultations with physiotherapists and chiropractors are reimbursed by the public insurance system. DPR contains information on all outpatient hospital contacts since 1994 and hospital admissions since 1978 [14].

Baseline variables are based on the following questions in the 1991 survey: Have you during the past two weeks had trouble (pain or discomfort) in any of these locations: Neck, shoulders, upper back, elbows, low back, hands/wrist, hip, knee, feet/ankles, have not had any trouble. To aid the respondent in answering this question, the interviewer showed a diagram with the anatomical areas clearly shaded and labelled. Each person could respond ‘yes’ or ‘no’ to any of the complaints, i.e. more than one confirmatory answer was allowed. If ‘yes’ was answered to any of the above, the interviewer then asked: ‘In which area have you had the most trouble?’ This question had the same response options, and again the respondent was shown the diagram – however, the person had to choose only one area as being the primary pain area. Again, no distinction was made between minor and severe pain or discomfort. Finally, we created a variable indicating whether the person had reported pain in another location in addition to the primary pain site, i.e. reported pain in more than one anatomical site.

Follow-up variables include:

Visits to general practitioners defined as number of consultations per year between 1991 and 2011.

Visits to physical therapists defined as number of consultations per year between 2001 and 2011. Due to administrative data definitions, physiotherapy services provided before 2001 could not be defined on a consultation basis, and consequently data from 1991 to 2000 were not part of the present analysis.

Visits to chiropractors defined as number of reimbursed consultations per year between 1991 and 2011. From 1991 to 2003, up to five consultations per year were reimbursed and from 2004 to 2011 all consultations were reimbursed.

Outpatient hospital visits defined as number of contacts per year between 1991 and 2011.

Hospital admissions defined as number of admissions per year between 1991 and 2011.

All visits, without any attempt to single out visits possibly related to musculoskeletal pain, were included in the analyses.

Analysis

Each person was followed from the date of interview until the same date 20 years later, i.e. for a person interviewed on, for example, 15 April 1991 the first full follow-up year ended on 15 April 1992 and all follow-up ended on 15 April 2011. Data for persons who had died or emigrated were censored for the full year prior to that event.

For descriptive purposes, care-seeking rates are reported as rate per 10,000 person years (10,000 * number of contacts/risk time), where risk time is defined as the number of full years in which the participant was followed.

Because the number of health care contacts are correlated from one year to the next and clustered within the same individual, generalized mixed linear models based on the Poisson distribution were used to calculate relative risks with 95% confidence intervals. First, in order to determine whether persons reporting a primary musculoskeletal complaint had an elevated risk of seeking care over the follow-up period, absolute rates and relative risks are reported for all musculoskeletal primary pain areas separately in relation to the care-seeking outcomes, where relative risks are calculated with ‘no musculoskeletal trouble’ as the reference category. Then, in order to gauge whether risk of care seeking was greater for some pain sites than others and for persons with pain in more than one site, relative risks were calculated for each pain site in relation to all care-seeking outcomes with pain in the low back as the reference category. All analyses were adjusted for age and gender. Analyses were performed using the SAS version 9.3 software, specifically using PROC GENMOD and generalized estimation equations.

Results

Of the 5986 persons invited, 4817 (80.5%) participated at baseline. Twenty persons reported having had pain in at least one site during the past two weeks but did not report which one was primary, and another 20 persons died or emigrated during the first year after the interview and were therefore excluded from the analysis, leaving a final study sample of 4777 Danes aged 16 and older.

At baseline, 2836 (69.3%) reported having had no musculoskeletal complaint during the past two weeks. Using only persons reporting a primary pain site as the denominator, 375 (19.3%) reported neck pain, 222 (11.4%) shoulder pain, 43 (2.2%) pain in the elbow, 128 (6.6%) pain in the wrist and/or hands, 102 (5.2%) pain in the upper back, 579 (29.8%) pain in the low back, 111 (5.7%) pain in the hip, 243 (12.5%) knee pain and 138 (7.1%) pain in the ankles/feet. The distribution and patterns of pain have previously been reported in detail [15].

Absolute rates and relative risks for all-cause care seeking with general practitioners, physiotherapists and chiropractors are shown in Table I. On average, persons not reporting musculoskeletal pain saw their general practitioner 5.7 times per year over the 20-year period, whereas persons experiencing a musculoskeletal complaint had a statistically increased risk of consulting a general practitioner regardless of pain site, with persons reporting pain in the hip experiencing most consultations per year (9.2) on average. For consultations with physiotherapists and chiropractors, only persons complaining of neck pain and back pain had a statistically significantly increased risk of seeking care from those providers (Table I).

Table I.

Primary sector care seeking between 1991 and 2011 for 4777 Danes stratified by musculoskeletal (MSK) complaints with general practitioners (GP), physiotherapists (PT) and chiropractors (DC). Absolute rates (AR) per 10,000 person years and relative risk adjusted for age and gender (RR) with 95% confidence intervals.

	GP		PT		DC
	AR	RR	AR	RR	AR	RR
Neck	79,360	1.27 (1.16–1.40)	23,649	2.10 (1.32–3.34)	4327	1.48 (1.13–1.93)
Shoulder	85,859	1.36 (1.21–1.53)	20,526	1.56 (0.89–2.73)	4081	1.36 (0.91–2.05)
Hands	86,648	1.37 (1.18–1.69)	14,144	1.14 (0.63–2.07)	4503	1.49 (0.95–2.33)
Mid back	86,584	1.54 (1.27–1.86)	11,734	1.23 (0.75–2.03)	3720	1.25 (0.78–1.99)
Low back	81,514	1.33 (1.24–1.42)	15,188	1.47 (1.03–2.08)	4266	1.50 (1.16–1.94)
Hip	91,664	1.32 (1.09–1.60)	20,640	1.48 (0.58–3.81)	2627	0.99 (0.62–1.58)
Knee	74,187	1.24 (1.12–1.37)	10,522	1.25 (0.79–1.97)	2885	1.06 (0.63–1.78)
Feet	80,322	1.37 (1.20–1.57)	16,060	1.77 (0.86–3.63)	3377	1.21 (0.70–2.09)
No MSK	57,730	1	9962	1	2895	1

Absolute rates and relative risks for all-cause outpatient hospital consultations and hospital admissions are shown in Table II. Persons reporting musculoskeletal pain had a statistically increased risk of both outpatient hospital consultations and hospital admission regardless of primary pain site, except for persons reporting primary shoulder pain, which was only borderline significant as a risk factor for hospital admission (Table II).

Table II.

Secondary sector care seeking between 1991 and 2011 for 4777 Danes stratified by musculoskeletal complaints with ambulatory hospital visits and hospital admissions. Absolute rates (AR) per 10,000 person years and relative risk adjusted for age and gender (RR) with 95% confidence intervals.

	Ambulatory hospital		Hospital admission
	AR	RR	AR	RR
Neck	5749	1.26 (1.15–1.38)	2204	1.18 (1.01–1.38)
Shoulder	5598	1.19 (1.05–1.36)	2210	1.18 (0.98–1.43)
Hands	7382	1.62 (1.34–1.96)	2531	1.36 (1.08–1.72)
Mid back	6443	1.54 (1.27–1.86)	3106	1.76 (1.24–2.48)
Low back	5680	1.27 (1.16–1.38)	2489	1.33 (1.16–1.51)
Hip	6263	1.31 (1.10–1.57)	2873	1.27 (1.04–1.56)
Knee	6190	1.46 (1.27–1.68)	2535	1.36 (1.11–1.66)
Feet	6623	1.52 (1.28–1.81)	2875	1.55 (1.24–1.95)
No MSK	4308	1	1771	1

Relative risks for all-cause care seeking in both primary and secondary care sector by pain site are reported in Table III. Using pain in the low back as reference, no differences were found between the pain sites in relation to any of the outcomes, except for an elevated risk for outpatient hospital consultations for persons reporting primary mid-back pain and wrist/hand pain and an increased risk for seeking care from a general practitioner for persons reporting pain in more than one site at baseline.

Table III.

Relative risk (adjusted for age and gender) for care seeking with general practitioner (GP), physiotherapist (PT), chiropractor (DC), ambulatory hospital visit (Amb) and hospital admission (Hosp) for 1961 Danes reporting a musculoskeletal complaint in 1991 using low back pain as the reference.

	GP	PT	DC	Amb	Hosp
Neck	0.95 (0.85–1.06)	1.44 (0.86–2.43)	0.98 (0.70–1.36)	1.00 (0.89–1.13)	0.90 (0.74–1.08)
Shoulder	1.02 (0.89–1.16)	1.08 (0.59–1.97)	0.93 (0.59–1.45)	0.95 (0.82–1.11)	0.90 (0.72–1.11)
Hands	1.03 (0.88–1.21)	0.81 (0.43–1.53)	0.98 (0.60–1.60)	1.29 (1.06–1.58)	1.03 (0.80–1.33)
Mid back	1.16 (0.96–1.41)	0.87 (0.50–1.50)	0.84 (0.50–1.40)	1.23 (1.00–1.50)	1.34 (0.93–1.93)
Hip	1.01 (0.83–1.24)	1.05 (0.40–2.79)	0.66 (0.40–1.10)	1.06 (0.88–1.28)	0.97 (0.78–1.22)
Knee	0.94 (0.84–1.06)	0.84 (0.51–1.38)	0.71 (0.40–1.25)	1.16 (0.99–1.35)	1.03 (0.83–1.29)
Feet	1.03 (0.90–1.19)	1.17 (0.56–2.45)	0.81 (0.46–1.44)	1.19 (0.99–1.43)	1.16 (0.91–1.49)
> 1 pain site	1.21 (1.13–1.31)	1.15 (0.82–1.60)	1.04 (0.76–1.41)	0.99 (0.81–1.22)	0.86 (0.64–1.16)
Low back	1	1	1	1	1

Discussion

In this 20-year follow-up in a population-based cohort of adult Danes, we found that self-report of two-week prevalence of any musculoskeletal pain in 1991 predicted a statistically significant increase in use of general health care services from general practitioners and from hospitals, as both outpatient visits and admissions. This was generally true regardless of pain site. Significantly increased risk of seeking care from a physiotherapist or chiropractor was found only for persons reporting a primary pain in the neck or low back, but estimates were similar across pain sites and lack of statistical significance probably reflects lower prevalence for some complaints. Thus, self-report of any musculoskeletal pain registered during a short random interval may be regarded as an indicator of poor general health and an indicator of future overall seeking of health care when compared with persons not reporting any pain. Curiously, the site of primary pain was not related to risk of care seeking, except for pain in the mid back and hand/wrist pain, which increased the risk for outpatient hospital consultations (Table III). We speculate that this is because mid-back pain may be perceived as a sign of cardiac problems and hand/wrist pain may be an annoying manifestation of arthritis, causing GPs to refer these patients to hospital specialists. We also found that persons reporting pain in more than one site had a statistically increased risk of seeking care from a general practitioner, likely reflecting both the general practitioner’s central role as gatekeeper in the Danish health care system and the fact that musculoskeletal pain is often part of a pattern of poor general health that is then captured by our outcome, which is general care seeking and not care seeking specific to musculoskeletal complaints.

Our study adds to the growing body of evidence indicating that a period with musculoskeletal pain is associated with long-term consequences, in this case care seeking. We have long known that an episode of, for instance, back pain or neck pain is strongly associated with future episodes of annoying pain and trouble [16,17]. Recently Ropponen et al., using data from a large Finnish twin cohort, found that just one-time reporting of musculoskeletal pain in the back, shoulders or neck that interfered with the ability to carry out normal duties predicted an increased risk of work disability and award of disability pension up to 23 years later for a range of musculoskeletal disorders, including back pain and osteoarthritis [18]. In addition, they found that an episode of interfering pain increased the risk for disability pension regardless of whether the episode was experienced only at baseline or persisted six years into the follow-up period [18].

Our results raise a number of questions that require further analyses in this and other databases. For example, we do not know whether the reason for care seeking was the musculoskeletal complaint or another symptom or morbidity. Clearly, musculoskeletal pain is associated with comorbidity in other body systems, and patients with musculoskeletal conditions report comorbidities in other body systems more frequently than persons without these conditions. For instance, patients with osteoarthritis of the hip or knee report the co-existence of heart disease (54%), eye, ear and throat problems (96%), urogenital (44%) and endocrine disorders (46%), exceeding the prevalence of these conditions in the general population [19]. Kadam et al. found that primary care patients with a diagnosis of osteoarthritis had a two-fold chance of also having gastritis, diaphragmatic hernia, heart and bowel disease when compared with matched primary care patients without osteoarthritis [20]. Older people reporting pain in the back and/or neck pain have a significantly greater chance of also suffering from heart disease, neurological disorders and gastric ulcers [21]. The combined effect is often greater than the effect of either the musculoskeletal complaint or the comorbidity alone [20], and significantly affects health-related quality of life [22], pain levels [22], functional status [23] and also care-seeking behaviour [24]. This body of information indicates that musculoskeletal pain – even of short duration – may be regarded as an indicator of poor general health and not just as a local pain phenomenon.

Another question, of course, is whether pain reporting and care seeking are a reflection of an underlying health condition or perhaps a reflection of a tendency in some persons to report pain and seek care while others have a more stoic attitude and ‘suffer in silence’. Indeed, psychological distress in back pain patients is associated with increased levels of care seeking and also with poorer outcome of care [25]. Self-efficacy and ability to cope thus may reflect both reporting of pain and seeking of care and could potentially be targets for intervention at the population or patient level. Finally, we do not know whether our results reflect an average tendency of persons reporting musculoskeletal pain to seek care or whether the higher health care use in the pain group is driven by a small group of high consumers.

We found that only persons reporting pain in the neck and back had an increased risk of seeking care from physiotherapists and chiropractors. In Denmark chiropractors are licensed to see patients with musculoskeletal complaints as a primary care provider, but over 80% of patients see chiropractors for back and neck pain [26], which is also true internationally [27]. Physiotherapists see patients with a broader range of functional limitations and musculoskeletal and neurological complaints; however, apparently neck and back pain patients greatly outnumber patients with other types of complaints. With newer international [28] and Danish [29] guidelines for the treatment of knee and hip osteoarthritis and shoulder pain [30] emphasizing non-surgical interventions and exercise, this pattern may change, and in the future general practitioners will likely refer more patients with these complaints to physiotherapists and chiropractors. Finally, the lack of statistical significance for pain sites other than the back or neck in this analysis could be due to the relatively low prevalence of some complaints.

Strengths of this study are interview-based data from a large population-based cohort with good response rate. In addition, we identified the primary musculoskeletal complaint at baseline, which ensured that different individual pain sites could be investigated together and contrasted; thus, we deliberately chose not to investigate the effect of multiple-site pain in this paper. The follow-up using comprehensive and complete national registries enabled a comprehensive look at care seeking in both the primary and the secondary care sector over a very long period. Weaknesses include the lack of diagnosis coding in Danish primary care, resulting in inability to identify consultations caused by musculoskeletal complaints. This is particularly annoying in relation to general practitioner consultations, because their scope of practice is not limited to musculoskeletal complaints, and consultations could be due to any number of health complaints. On the other hand, the consistent finding that persons with any musculoskeletal complaint were at increased risk of care seeking with general practitioners when compared with persons without a complaint indicates that musculoskeletal pain may indeed be an indicator of poor general health and not only an isolated biomechanical or overuse injury. The fact that we only had follow-up data for physiotherapy consultations for the last 10 years may have resulted in insufficient power to detect risk for physiotherapy consultations for most pain complaints. Likewise, the number of chiropractic consultations might be underestimated from 1991 to 2004, since during that period only the first five consultations per year were registered. It should also be acknowledged that patients sometimes pay out of pocket for visits to chiropractors and physical therapists, or have these visits reimbursed by workplace insurance or similar. In neither case would the visits be registered within the public insurance system. Taken together, our data therefore provide a conservative estimate for visits to chiropractors and physical therapists, and our analyses for these professional groups may have suffered from low power. Furthermore, results of studies dealing with prevalence of conditions or symptoms during a relatively short time window – such as our baseline assessment – typically include a disproportional number of longer-lasting conditions or symptoms, simply because the chances of capturing these are greater compared with shorter-lasting conditions or symptoms. Thus, our baseline two-week prevalence likely includes a fair proportion of longer-lasting symptoms. On the other hand, longer time windows greatly increase the risk of memory bias, so the short time window probably ensures valid self-report of pain.

We deliberately did not elaborate on the fact that musculoskeletal complaints often occur together in the same person because our aim was to examine the effect of site-specific pain on future care seeking. Indeed, analyses of such pain patterns in relation to a number of outcomes, such as seeking of health care and/or social effects, will be the subject of subsequent analyses and papers.

Conclusions

Self-report of musculoskeletal pain in the population within the past two weeks predicts long-term increase in general use of health care services in both the primary and the secondary health care sector.

Footnotes

Conflict of interest

The authors declare that there is no conflict of interest.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

References

Woolf

Erwin

March

. The need to address the burden of musculoskeletal conditions. Best Pract Res Clin Rheumatol 2012;26:183–224.

Busija

Bridgett

Williams

. Osteoarthritis. Best Pract Res Clin Rheumatol 2010;24:757–68.

Vos

Flaxman

Naghavi

. Years lived with disability (YLDs) for 1160 sequelae of 289 diseases and injuries 1990–2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet 2012;380:2163–96.

Lim

Vos

Flaxman

. A comparative risk assessment of burden of disease and injury attributable to 67 risk factors and risk factor clusters in 21 regions, 1990–2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet 2013;380:2224–60.

Office of German Federal Heath Monitoring. Cost of illness accounts 2011. http://www.gbe-bund.de (accessed 1 April 2013).

Eriksen

Sjogren

Ekholm

. Health care utilisation among individuals reporting long-term pain: an epidemiological study based on Danish National Health Surveys. Eur J Pain 2004;8:517–23.

Cote

Cassidy

Carroll

. The treatment of neck and low back pain: who seeks care? who goes where? Med Care 2001;39:956–67.

Lonnberg

. [The management of back problems among the population. I. Contact patterns and therapeutic routines]. Ugeskr Laeger 1997;159:2207–14.

Deyo

Mirza

Turner

. Overtreating chronic back pain: time to back off? J Am Board Fam Med 2009;22:62–8.

10.

Cram

Kates

. Total knee arthroplasty volume, utilization, and outcomes among Medicare beneficiaries, 1991–2010. JAMA 2012;308:1227–36.

11.

Bitton

. The economic burden of osteoarthritis. Am J Manag Care 2009;15:S230–5.

12.

Ekholm

Hesse

Davidsen

. The study design and characteristics of the Danish national health interview surveys. Scand J Public Health 2009;37:758–65.

13.

Andersen

Olivarius Nde

Krasnik

. The Danish National Health Service Register. Scand J Public Health 2011;39:34–7.

14.

Lynge

Sandegaard

Rebolj

. The Danish National Patient Register. Scand J Public Health 2011;39:30–3.

15.

Hartvigsen

Davidsen

Hestbaek

. Patterns of musculoskeletal pain in the population: a latent class analysis using a nationally representative interviewer-based survey of 4817 Danes. Eur J Pain 2013;17:452–60.

16.

Henschke

Maher

Refshauge

. Prognosis in patients with recent onset low back pain in Australian primary care: inception cohort study. BMJ 2008;337:a171.

17.

Carroll

Hogg-Johnson

van der Velde

. Course and prognostic factors for neck pain in the general population: results of the Bone and Joint Decade 2000–2010 Task Force on Neck Pain and Its Associated Disorders. Spine 2008;33:S75–82.

18.

Ropponen

Svedberg

Kalso

. A prospective twin cohort study of disability pensions due to musculoskeletal diagnoses in relation to stability and change in pain. Pain 2013;154:1966–72.

19.

van Dijk

Veenhof

Schellevis

. Comorbidity, limitations in activities and pain in patients with osteoarthritis of the hip or knee. BMC Musculoskelet Disord 2008;9:95.

20.

Kadam

Jordan

Croft

. Clinical comorbidity in patients with osteoarthritis: a case-control study of general practice consulters in England and Wales. Ann Rheum Dis 2004;63:408–14.

21.

Hartvigsen

Christensen

Frederiksen

. Back and neck pain exhibit many common features in old age: a population-based study of 4,486 Danish twins 70–102 years of age. Spine 2004;29:576–80.

22.

Tuominen

Blom

Hirvonen

. The effect of co-morbidities on health-related quality of life in patients placed on the waiting list for total joint replacement. Health Qual Life Outcomes 2007;5:16.

23.

Dekker

van Dijk

Veenhof

. Risk factors for functional decline in osteoarthritis of the hip or knee. Curr Opin Rheumatol 2009;21:520–4.

24.

Hurwitz

Morgenstern

. The effect of comorbidity on care seeking for back problems in the United States. Ann Epidemiol 1999;9:262–70.

25.

Pincus

McCracken

. Psychological factors and treatment opportunities in low back pain. Best Pract Res Clin Rheumatol 2013;27:625–35.

26.

Hartvigsen

Sorensen

Graesborg

. Chiropractic patients in Denmark: a short description of basic characteristics. J Manipulative Physiol Ther 2002;25:162–7.

27.

Coulter

Hurwitz

Adams

. Patients using chiropractors in North America: who are they, and why are they in chiropractic care? Spine 2002;27:291–6.

28.

Zhang

Moskowitz

Nuki

. OARSI recommendations for the management of hip and knee osteoarthritis, Part II: OARSI evidence-based, expert consensus guidelines. Osteoarthritis Cartilage 2008;16:137–62.

29.

Danish Health and Medicines Authority. Guidelines for Osteoarthritis of the Knee. Copenhagen: Danish Health and Medicines Authority, 2007.

30.

American Academy of Orthopaedic Surgeons Clinical Practice Guideline on the Diagnosis and Treatment of Osteochondritis Dissecans . American Academy of Orthopedic Surgeons, 2010.