Low back pain is a major cause of morbidity in high-, middle- and low-income countries, yet to date it has been relatively under-prioritised and under-funded. One important reason may be the low ranking it has received relative to many other conditions included in the previous Global Burden of Disease studies, due in part to a lack of uniformity in how low back pain is defined and a paucity of suitable data. We present an overview of methods we have undertaken to ensure a more accurate estimate for low back pain in the Global Burden of Disease 2005 study. This will help clinicians to contextualise the new estimates and rankings when they become available at the end of 2010. It will also be helpful in planning further population-based epidemiological studies of low back pain to ensure their estimates can be included in the future Global Burden of Disease studies.
Low back pain is one of the most common health problems and affects 80–85% of people over their lifetime . In high-income countries, low back pain is the most frequent occupational problem, with an estimated 2–5% of people having chronic low back pain, many of whom are permanently debilitated as a result . It is also the most frequent activity-limiting complaint in the young and middle aged and the second leading cause of sick leave .
The impact from low back pain is multi-factorial, and includes pain and activity limitations, as well as broader impacts such as participation restrictions, carer burden, use of health-care resources and financial burden. The financial burden alone is enormous and includes the costs of medical care, indemnity payment, productivity loss, employee retraining, administrative expenses and litigation . For example, low back pain was found to be one of the most expensive diseases in Australia, with an estimated cost in 2001 of $9.17 billion .
Because low back pain is rarely fatal, unlike infectious and cardiovascular diseases or cancer, it is often seen as a trivial problem . With epidemiological evidence that severe low back pain increases into old age , and suggestions of an increasing prevalence in adolescence , it is imperative that low back pain be appropriately prioritised to ensure adequate allocation of health expenditure to address this growing public health problem, particularly with a view to developing preventive strategies. Demonstrating the impact from low back pain is the first step in this process.
Burden of disease studies
Burden of disease (BoD) estimates are one of the measures that governments and others may use when assessing health priorities, allocating resources and evaluating the potential costs and benefits of public health interventions. In BoD studies, diseases are ranked according to how much death and disability they cause. Since the introduction of BoD in the 1990s, largely non-fatal health problems such as mental disorders and sense organ disorders have been identified as global health priorities, whereas previously these would not have been given attention in mortality-based ranking lists.
Measurement of disability in BoD focuses on activity limitations rather than broader impacts such as participation, carer burden and economic impact. Therefore, to consider the full impact of low back pain, BoD estimates need to be supplemented with information on these other broader measures.
The World Bank commissioned the first Global Burden of Disease (GBD) study in the early 1990s (GBD 1990) , resulting in the assessment of disease burden for over 100 diseases and injuries . In 2000–2004, the World Health Organization (WHO) updated these data (GBD 2000–2004) . The GBD 2005 study (GBD 2005), funded in part by the Bill and Melinda Gates Foundation, is currently being undertaken through a collaboration between the University of Washington, Harvard University, WHO, the University of Queensland, Johns Hopkins University and a community of experts and leaders in epidemiology and other areas of public health research from around the world. New, more advanced methods are being applied to estimate the 2005 burden for more than 175 diseases and injuries. These methods will also be retrospectively applied to 1990 data to revise the original GBD estimates .
In past GBD studies, low back pain was ranked relatively low, due partly to a lack of available and suitable data for many countries . This is likely to have resulted in low back pain being under-prioritised by governments and others compared with many other conditions. GBD 2005 provides an opportunity to ensure low back pain is ranked more appropriately. In this article, we give an overview of the prevalence of low back pain, discuss BoD concepts, and describe and discuss the work undertaken to date in preparing to estimate the burden of low back pain for GBD 2005.
The prevalence of low back pain
The major indicators of morbidity at a population level are incidence (number of new cases of a disease in a given time period) and prevalence (number of individuals with existing disease at a given point in time) . Estimating the incidence of low back pain is problematic as the cumulative incidence of first-ever episodes of low back pain is already high by early adulthood and symptoms tend to recur over time . Therefore, most epidemiological studies on low back pain have estimated prevalence, either as point (i.e., at a specific time) or period estimates (e.g., 1 year and lifetime).
Comparing the prevalence of low back pain between populations and over time is challenging due to the previous lack of a uniform case definition, marked methodological heterogeneity across studies and difficulties in obtaining true population estimates . For example, in 2000, a systematic review of population-based prevalence studies on low back pain found only 30 studies that met methodological quality criteria . Of these, only 21 estimated pain specifically for the low back region, and only 10 reported estimates across most adult age groups. There were also very few studies that reported the prevalence of low back pain in low- and middle-income countries . However, since 2000, an increasing body of high-quality evidence suggests that prevalence of low back pain is also high in these countries . Figs 1–3 show point, 1-year and lifetime prevalence estimates from methodologically acceptable studies in the above review and a review of African studies .
While the high global prevalence of low back pain is well established, prevalence rates alone do not adequately illustrate its burden. BoD methods most commonly use Disability-Adjusted Life Years (DALYs), a summary measure that quantifies loss of healthy years of life due to premature mortality and morbidity against an ideal that everyone in the population lives into old age free of disease. Calculation of DALYs is described in more detail below . In addition to the occurrence of low back pain, the DALY also considers other measures, such as episode duration and disability.
The prevalence of low back pain
The major indicators of morbidity at a population level are incidence (number of new cases of a disease in a given time period) and prevalence (number of individuals with existing disease at a given point in time) . Estimating the incidence of low back pain is problematic as the cumulative incidence of first-ever episodes of low back pain is already high by early adulthood and symptoms tend to recur over time . Therefore, most epidemiological studies on low back pain have estimated prevalence, either as point (i.e., at a specific time) or period estimates (e.g., 1 year and lifetime).
Comparing the prevalence of low back pain between populations and over time is challenging due to the previous lack of a uniform case definition, marked methodological heterogeneity across studies and difficulties in obtaining true population estimates . For example, in 2000, a systematic review of population-based prevalence studies on low back pain found only 30 studies that met methodological quality criteria . Of these, only 21 estimated pain specifically for the low back region, and only 10 reported estimates across most adult age groups. There were also very few studies that reported the prevalence of low back pain in low- and middle-income countries . However, since 2000, an increasing body of high-quality evidence suggests that prevalence of low back pain is also high in these countries . Figs 1–3 show point, 1-year and lifetime prevalence estimates from methodologically acceptable studies in the above review and a review of African studies .
While the high global prevalence of low back pain is well established, prevalence rates alone do not adequately illustrate its burden. BoD methods most commonly use Disability-Adjusted Life Years (DALYs), a summary measure that quantifies loss of healthy years of life due to premature mortality and morbidity against an ideal that everyone in the population lives into old age free of disease. Calculation of DALYs is described in more detail below . In addition to the occurrence of low back pain, the DALY also considers other measures, such as episode duration and disability.
Previous estimates of the global burden of low back pain (Level A)
For the original GBD study (GBD 1990), no estimates were made for low back pain. Instead, all musculoskeletal conditions apart from osteoarthritis and rheumatoid arthritis were grouped in the category ‘ other musculoskeletal conditions’ . For the GBD 2000–2004 updates, separate estimates were made for three health states of low back pain :
- i.
Acute episode of low back pain resulting in moderate or greater limitations to mobility and usual activities;
- ii.
Episode of intervertebral disc displacement or herniation; and
- iii.
Chronic intervertebral disc disorder.
The global burden of low back pain in 2004 was estimated to be 2.5 million DALYs, representing 0.09% of the overall global disease burden. The burden from low back pain was greatest in adolescence to middle age for both sexes ( Fig. 4 ), and in low- and lower-middle-income countries than in the upper-middle-income and high-income countries.
However, the approach taken had a number of limitations. Firstly, intervertebral disc pathology was a defining factor for two of the health states, yet the presence of intervertebral disc pathology requires radiological imaging, and most population-based studies do not have the resources to perform these investigations. More importantly, the presence of intervertebral disc pathology correlates poorly with clinical symptoms and therefore is unlikely to be a good indicator of functional disability . In addition, low back pain with mild activity limitations, which is common and therefore potentially has a substantial global impact, was not included in the estimates . There were also limitations due to methodological heterogeneity between low back pain prevalence studies and a paucity of suitable data.
GBD 2005 methods for determining the global burden of low back pain
The remainder of this article details the steps we are taking to derive estimates of the global burden of low back pain for GBD 2005, including our rationale for the selected case definition, health states and lay descriptions ( Fig. 5 ). We also briefly describe the methods for assigning disability weights to each of the health states across all diseases in the GBD project, and the parameters that will be used to estimate the disease burden.
Case definition
For the purpose of the GBD 2005, the case definition of low back pain is required to align with definitions used in epidemiological datasets to maximise data available for burden estimates. We performed an abbreviated literature search for case definitions used in previous studies. We searched Medline, EMBASE and CINAHL databases using the following search string: ( back pain OR lumbar pain OR back ache OR backache OR lumbago ) AND ( prevalence OR incidence OR cross-sectional OR epidemiology ). We limited our search to studies from 1980 to the present and had no language limits. We found 50 articles that reported results of population-based studies on the prevalence of low back pain and could provide useful input into development of the case definition .
The case definition of low back pain varied widely between studies, particularly with respect to the area considered the low back, the recall period and minimum duration of the episode . Most studies did not provide a precise description of pain location but among those that did, the area between the inferior margin of the twelfth rib and inferior gluteal folds was the most commonly used. The most common recall periods were 1 year, at the current time and over the person’s lifetime. Episode duration was less often used in case definitions, but when it was, duration of at least 1 day was most common .
Based on these findings, and in line with a recent international consensus of a case definition for low back pain for future prevalence studies , we agreed on the following case definition for the current GBD study: Activity-limiting low back pain ( +/ − pain referred into one or both lower limbs) that lasts for at least 1 day. The ‘low back’ is defined as the area on the posterior aspect of the body from the lower margin of the twelfth ribs to the lower gluteal folds .
Health states
To develop a description of discrete health states that would characterise different levels of severity and take into account the variation in functional loss associated with different low back pain episodes, the same studies were reviewed. The final health states needed to be consistent with the natural history of low back pain , which is known to be extremely variable . The purpose of developing discrete health states representing different levels of severity was to enable the development of disability weights that would adjust for differences in severity of low back pain in the final burden estimates.
Most commonly, individuals have multiple episodes of low back pain separated by asymptomatic periods or periods of lesser severity . We found episodes could be short in duration, or longer and more disabling, extending into a chronic pain syndrome after the expected normal healing period has elapsed . We also found that episodes associated with referred pain into one or both lower limbs were usually associated with more significant loss of function .
We found sufficient epidemiological data to enable the estimation of the prevalence of low back pain according symptom duration: acute/sub-acute (less than 3 months), and chronic (3 months or more) . This is consistent with a recent international consensus exercise, which recommended the following categories for future epidemiologic studies: less than 3 months; 3 months or more, but less than 7 months; 7 months or more, but less than 3 years; and 3 years or more .
We developed four health states that were consistent with the epidemiological literature and best represented the natural history and functional loss associated with low back pain (see Fig. 6 ):
- •
Acute/sub-acute low back pain without leg pain: An episode of activity-limiting low back pain (with no pain referred into either lower limb) that lasts for at least 1 day and resolves either temporarily or permanently in less than 3 months.
- •
Acute/sub-acute low back pain with leg pain: An episode of activity-limiting low back pain (with pain referred into one or both lower limbs) that lasts for at least 1 day and resolves either temporarily or permanently in less than 3 months.
- •
Chronic low back pain without leg pain: An episode of activity-limiting low back pain (with no pain referred into either lower limb) that lasts for 3 months or more.
- •
Chronic low back pain with leg pain: An episode of activity-limiting low back pain (with pain referred into one or both lower limbs) that lasts for 3 months or more.
Related posts:
Exercise therapy for chronic nonspecific low-back pain
Treatment-based subgroups of low back pain: A guide to appraisal of research studies and a summary of current evidence
How important are back pain beliefs and expectations for satisfactory recovery from back pain?
Back pain and work
Lumbar Spinal Stenosis
Contemporary low back pain research – and implications for practice
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
