9.9 Musculoskeletal diseases*
Oxford Textbook of Public Health
Jennifer L. Kelsey and MaryFran Sowers
Magnitude of the problem
Some common problems in epidemiological studies of musculoskeletal conditions
Selected musculoskeletal disorders of adults
Osteoporosis and associated fractures
Low back pain
Selected musculoskeletal disorders of children
Slipped capital femoral epiphysis
Developmental (congenital) dislocation/dysplasia of the hip
Musculoskeletal diseases are among the most important public health problems. They are common, affect both sexes and all age groups, and are responsible for a substantial amount of impairment and disability. Musculoskeletal disorders range from minor aches and pains to chronic disabling conditions. Although they are occasionally fatal, their main effects are on quality of life and on economic productivity.
Musculoskeletal conditions have been less well studied than other major diseases. This limited attention has probably occurred because musculoskeletal diseases are not usually fatal, they are generally not dramatic because of their gradual onset, and some of the more important musculoskeletal diseases, such as osteoarthritis and osteoporosis, are often considered as inevitable consequences of ageing. In addition, most chronic musculoskeletal conditions do not come to medical attention until they are quite advanced, thus making case ascertainment more difficult in epidemiological studies.
Among the many musculoskeletal conditions that could be considered here, the most common and disabling in adults will be described (osteoarthritis, rheumatoid arthritis, osteoporosis and associated fractures, and back and neck pain), followed by some of the important diseases of adolescence, childhood, and infancy (scoliosis, slipped epiphysis, fractures, and developmental dislocation/dysplasia of the hip).
Magnitude of the problem
Musculoskeletal diseases are a major cause of disability worldwide and place a considerable social and economic burden on all societies (WHO 1992). In the United States, for instance, musculoskeletal problems are the most frequently reported impairment, affecting about 14 per cent of the population (Praemer et al. 1999). Table 1 shows that impairment of the back or spine is the most frequent, followed by the lower extremity or hip, and then the upper extremity or shoulder. The prevalence of musculoskeletal impairments increases with age. Among those aged 85 years and older, it is 20 per cent, with impairments of the back or spine and lower extremity or hip accounting for slightly more than half of musculoskeletal impairments. Among impairments, musculoskeletal disorders are the leading cause of days of restricted activity and of days in bed. In Ontario, Canada, musculoskeletal diseases rank first as a cause of long-term health problems and long-term disability, and second as a cause of days of restricted activity (Badley et al. 1994).
Table 1 Prevalence of musculoskeletal impairments, United States, 1995
A survey in the United Kingdom found that 17 per cent of adults report a long-standing musculoskeletal disorder (Bowling 1996). The most important areas of life affected by long-standing disorders of the musculoskeletal system were ability to get around, stand, walk, and go shopping (24 per cent), and to participate in social and leisure time activities (24 per cent). Ability to work was also a major issue (17 per cent), particularly for those of working age. Another study in England (Thompson et al. 1974) found that many of the elderly with musculoskeletal diseases live alone. Half of them had difficulties because of stairs, one-third said they would not be able attract attention even in the event of an emergency, 20 per cent were dependent on others for such everyday tasks as taking a bath, doing housework, and getting out of the house.
Acute injuries of the musculoskeletal system, including fractures, dislocations, sprains, and strains, are also common. According to the United States Health Interview Survey (Praemer et al. 1999), there are 14.5 musculoskeletal injuries per 100 people per year of sufficient severity to warrant medical care or at least one half day of restricted activity. Among acute conditions, musculoskeletal conditions rank second to respiratory conditions in the frequency with which they are reported.
Musculoskeletal conditions generate substantial health care utilization. About 11 per cent of all hospital admissions in short-stay hospitals in the United States are attributed to musculoskeletal conditions (Praemer et al. 1999). Fractures, arthritis, intervertebral disk disorders, and other back problems are responsible for the greatest number of hospital admissions. The population aged 65 years and older accounts for 44 per cent of hospital admissions for musculoskeletal conditions. Figure 1 shows that among visits to physicians in office-based practice, musculoskeletal conditions rank first, accounting for 17 per cent of all office visits. Fifteen per cent of visits to hospital outpatient departments and 26 per cent of visits to emergency departments are attributable to musculoskeletal conditions. In addition, a significant number of people in the United States use complementary medicine for musculoskeletal problems. In a national sample of adults it was found that 36 per cent of those with back problems, 18 per cent of those with arthritis, and 22 per cent of those with sprains or strains had used unconventional therapy in the past year (Eisenberg et al. 1993). In Ontario, Canada, musculoskeletal diseases rank first as a reason for consultations with a health professional and second as a cause of use of both prescription and non-prescription drugs (Badley et al. 1994).
Fig. 1 Estimated number of visits to physicians in office-based practice by category of first-listed diagnosis, United States 1995, National Ambulatory Medical Care Survey. (Source: Praemer et al. 1999.)
Musculoskeletal conditions are also a significant cause of workplace injuries and disability. Among the conditions for which worker disability allowances were granted in the United States in 1992, back disorders ranked sixth, osteoarthritis eighth, and rheumatoid arthritis sixteenth (Social Security Administration 1994, unpublished data).
The monetary cost of musculoskeletal diseases to society is substantial, especially because of the decreased productivity of those affected. The total cost of musculoskeletal diseases in the United States was estimated to be US$215 billion in 1995, with 41 per cent of this amount attributable to direct costs such as for hospital care, physicians and other health care providers, drugs, nursing home care, and administrative costs, and 59 per cent attributable to indirect costs from lost productivity because of morbidity and to a small extent mortality (Praemer et al. 1999). Felts and Yelin (1989) reported that costs of musculoskeletal diseases accounted for 1 per cent of the gross national product in the United States. Using somewhat different criteria, Coyte et al. (1998) estimated that musculoskeletal diseases cost 25.6 billion Canadian dollars in 1994, or 3.4 per cent of the gross domestic product. Seventy-one per cent of the costs were indirect. Badley (1995) noted that musculoskeletal diseases account for 32 per cent of all chronic disability costs in Canada.
In summary, musculoskeletal conditions have a substantial impact on quality of life in industrialized countries. In the United States, they rank highest among disease groups in the frequency with which they cause impairments and limitation of activity. They rank first in frequency of visits to physicians, and fourth in frequency of hospital admissions and frequency of surgical procedures within hospitals. The costs of musculoskeletal conditions are enormous, including both direct costs for medical care and indirect costs from lost productivity.
Moreover, the enormous societal burden from musculoskeletal diseases is not restricted to industrialized countries with high life expectancies. The limited data that are currently available indicate a large impact of musculoskeletal diseases on the work and personal lives of people in developing countries as well. As in industrialized countries, pain, disability, fatigue, depression, and loss of employment because of musculoskeletal diseases are major problems. For example, the prevalence of rheumatic complaints and of disability from rheumatic complaints in both urban and rural areas of Indonesia is similar to that in Australia (WHO 1992). Work-related problems, many of which could be prevented with proper ergonomic techniques, are particularly common in developing countries (Ahasan et al. 1999). In a survey in Thailand, about 50 per cent of female workers in five industries (garment, fertilizer, pharmaceutical, textile, and cigarettes) reported low back symptoms (Chavalitsakulchai and Shahnavaz 1993). Back problems and other musculoskeletal symptoms could often be attributed to modifiable situations, including manual handling of heavy materials, prolonged sitting and standing, awkward work positions, poor machine design and operation, monotonous and repetitive movements, poor work organization, and unsatisfactory work environments. Thus, opportunities already exist to reduce the burden of job-related musculoskeletal disorders in developing countries.
Some common problems in epidemiological studies of musculoskeletal conditions
Conducting epidemiological studies of musculoskeletal conditions, even in industrialized countries, presents several difficulties not experienced to the same extent in studies of other important chronic diseases such as cancers and coronary heart disease (Kelsey 1982). Firstly, because many musculoskeletal conditions are not fatal, are not necessarily seen in hospitals, have a gradual onset, and often do not even come to medical attention, ascertainment of representative cases for epidemiological study may be difficult. Consequently, many studies of musculoskeletal diseases are based on the experience of particular clinical practices or hospitals. However, cases representative of those occurring in the general population are needed in order to provide an accurate description of disease occurrence, aetiology, and progression in the community. Even if all cases coming to medical attention within a defined geographical area are included in a study, unless the disease under study almost invariably comes to medical attention (such as hip fractures), the factors that bring persons with a given condition to medical attention often cannot be separated from factors that may be related to disease aetiology. For instance, are psychological characteristics of persons seeking care for back pain related to the aetiology of back pain, or are they related to the care-seeking behaviour among those with back pain?
Sometimes a condition for which medical care is virtually always sought, and for which case ascertainment is therefore relatively easy, is used as a surrogate for a disease for which medical care is not necessarily sought. For instance, hip fracture, which is almost always seen in a hospital, has been used as a surrogate for osteoporosis. However, hip fracture depends not only on whether a person has low bone mass, but also on the bone architecture and on whether the person falls and how he or she falls. Thus the risk factors for osteoporosis and hip fracture will overlap, but they will also differ in some respects.
The diagnostic criteria used for a disease may differ from one study to another, making it difficult to compare results. For many of the arthritic disorders, diagnosis is not straightforward. For example, people with arthritis on radiograph may not have symptoms, while people with symptoms do not necessarily show radiographic changes, resulting in an ambiguous case definition of osteoarthritis. Expert committees have established and then periodically revised the criteria for several of the arthritic disorders. When the diagnostic criteria are uniformly applied, this has increased comparability of cases from one study to another. However, the criteria have not always been used uniformly, and since the criteria have changed over time, it is sometimes not possible to examine changes in disease frequency over time. People may say that they have low back pain, but how does a researcher or practitioner know that a person is really in pain if there are no radiographic or other signs? The majority of asymptomatic people have evidence of bulging or protruding intervertebral disks on MRI, so that even MRI may be of limited value without clinical evidence of pathology (Jensen et al. 1994). The person may report low back pain only in order to obtain worker’s compensation or the person may be truly affected.
Diseases that constitute one end of a continuous distribution, such as osteoporosis and scoliosis, present other problems. To define cases of these conditions, relatively arbitrary cut-off points aimed at classifying individuals as ‘diseased’ and ‘normal’ must be employed, but different investigators may chose different cut-off points, thus precluding comparisons from one study to another. In recent years, standard cut-off points have been recommended by committees, so that this situation should improve. Another issue is that it may be difficult to differentiate the aetiology of diseases such as osteoporosis, degenerative disk disease, and osteoarthritis from what is considered normal ageing.
Labels for diseases once considered a single entity may become irrelevant when subcategories of the disease are more clearly delineated. For instance, what was once called ‘juvenile rheumatoid arthritis’ is now known to be composed of several subtypes, each with its own aetiology. Much information may be lost when several distinct entities are considered as one disease, as exemplified by studies of low back pain that do not consider the contribution of many different disease processes, such as sprains and strains, intervertebral disk herniations, and osteoporosis.
Another concern relates to study design. Cross-sectional studies of the relation between neuromuscular abnormalities and scoliosis, for instance, are difficult to interpret because it is not known whether the neuromuscular abnormalities lead to curvature of the spine or whether curvature of the spine leads to neuromuscular defects. Cross-sectional studies of psychological symptoms and low back pain and of obesity and osteoarthritis present similar difficulties in interpretation. In such situations longitudinal studies, despite their great expense, are needed.
Problems in conducting epidemiological studies of musculoskeletal diseases are magnified considerably in developing countries. Because of shortages of money and human resources, most studies are hospital-based. However, in many developing countries most people do not seek care from a physician or in a hospital unless the disease is quite severe. Instead, inexperienced community nurses often provide care (WHO 1992; Dans et al. 1997). In a survey in the Philippines, for instance, it was found that most people with rheumatic complaints who sought treatment were seen by someone other than a physician. Only 23 per cent had seen a general practitioner, and 2 per cent saw a rheumatologist. Care from a physician is frequently not available, accessible, or affordable (Dans et al. 1997; Croft 1996). In fact, some countries have no rheumatologists at all (Adebajo 1990). Patriarchal societies may limit access to medical care for women, and relatively few women survive to old age. Consequently, both comparisons of disease frequency in men and women within such countries and comparisons to other countries may be difficult to interpret (Hameed et al. 1995; Farooqi and Gibson 1998). In addition, even when affected individuals are seen by qualified medical personnel, the diagnostic criteria vary considerably (Ferraz 1995).
If community surveys are undertaken in developing countries, many difficulties are likely to be encountered. Poor training of personnel, inadequate facilities, limited ability to reach populations for study, a low level of public support, political instability, and bureaucratic rigidity are often encumbrances to epidemiological studies (Ferraz 1995). The diagnostic tests used in such surveys must be simple and inexpensive. If the study is being conducted in a tropical region, reagents and tests used in other geographical areas may not work because of either the heat or the characteristics of the populations being studied. Tests for rheumatoid factor in West Africa are not useful because similar proportions of people with and without rheumatoid arthritis are positive (Adebajo 1990). Rheumatoid nodules are rare in southern Asians with rheumatoid arthritis, suggesting that diagnostic criteria may have to be modified (Hameed et al. 1995). In addition, pain, complaints, and manifestations of disability may vary cross-culturally (Croft 1996). Measures of disability often require modification (Adebajo 1990).
It should also be kept in mind that risk factors may be quite different in developing countries. It has been hypothesized, for instance, that osteoporosis may not be such an important cause of fracture in developing countries because high levels of physical activity make bones stronger (Scrimgeour 1992).
Finally, great variation in disease frequency and in risk factor prevalence occur among developing countries, and there is a need for information on specific diseases and risk factors in specific regions. In north Pakistan, for instance, osteoarthritis of the knee is an especially common complaint. Fibromyalgia, low back pain, and soft-tissue rheumatism are next most common, but mainly in the urban poor (Farooqi and Gibson 1989). Thus, not only should one avoid generalizing from one country to another, but frequently one cannot generalize from one part of the same country to another.
Selected musculoskeletal disorders of adults
Osteoarthritis is a condition affecting joints and their constituent parts, including muscle, bone, and cartilage. Increased bone stiffness, increased mineralization of bony tissue around the joint in the form of osteophytes, and loss of cartilage from the joint space are seen. This joint degradation results in pain, loss of mobility, disability, and, increasingly, the need for joint replacement to remediate the pain and loss of mobility. Indeed, osteoarthritis is second only to cardiovascular diseases in the frequency with which it produces severe chronic disability.
The peripheral joints most commonly affected in osteoarthritis are (in descending order of frequency) hands, feet, knees, and hips. Clinically, several subsets of disease are noted based on the pattern of the joint group involvement, such as isolated hip or knee involvement, hand plus knee involvement, or polyarticular involvement.
Estimates of osteoarthritis prevalence have generally used information from two sources: (a) an individual’s report of pain, decreased articular movement in the joint, and curtailment of physical functioning; (b) radiographs of the joint. Prevalence estimates depend on case definition (i.e. self-report of symptoms or radiographic appearance) as well as measurement approach (e.g. radiographs of the knees with or without bearing weight). The typical radiographic appearance of the joint includes narrowing of the joint space, subchondral sclerosis, and osteophyte formation, which are assumed to reflect the underlying processes of cartilage loss, bone change, and new bone formation (Kellgren and Lawrence 1963).
In the United States the prevalence of osteoarthritis has been estimated in two national studies: the National Health Examination Survey of 1960 to 62 (hands and feet) and the National Health and Nutrition Survey (NHANES-I) of 1971 to 75 (knees and hips). In NHANES-I there was radiographic evidence of osteoarthritis in more than 33 per cent of persons aged 25 to 74 years and more than 70 per cent of persons aged 55 to 74 years (Lawrence et al. 1989). NHANES-III has been recently completed. Data on radiographic osteoarthritis of the hands and knees will soon be available.
Many individuals with radiographic evidence of osteoarthritis are asymptomatic. In one American population, only 65 per cent of those older than 55 years of age with radiographic evidence of knee osteoarthritis reported knee pain, and 76 per cent of those with radiographic evidence of hand osteoarthritis reported hand pain (Carman 1989). In women aged 40 to 55 years, 9 per cent had knee osteoarthritis on radiograph and reported pain, 6 per cent had knee osteoarthritis on radiograph in the absence of reported pain, 31 per cent reported knee pain in the absence of radiographic osteoarthritis, and 54 per cent reported no pain and had no radiographic evidence of osteoarthritis in the knee (Lachance et al. 2001).
Despite differences in osteoarthritis definitions, several patterns of disease occurrence are evident (Lawrence et al. 1989; Lawrence et al. 1998). The prevalence of osteoarthritis increases with age, especially after 40 years of age. Almost everyone 65 years or older has at least one joint with a radiographically-defined feature of the disease. Individual joints show the same age-related rise in osteoarthritis prevalence, with the hip showing a later rise than the knee (Lawrence et al. 1998). Women are affected more often than men, especially for hand and knee osteoarthritis, and particularly after 50 years of age (Butler et al. 1988).
Caucasian populations in developed countries have similar prevalence rates (van Sasse et al. 1988), but hip osteoarthritis may be less common in Africa and Asia than in Western countries (Lawrence and Sebo 1980). Knee osteoarthritis shows less geographicalal variation than hip and hand osteoarthritis, possibly reflecting the importance of injury as a cause of knee disease (Lawrence and Sebo 1980).
The aetiology of osteoarthritis is complex and involves modifications in bone, joint, capsule, and muscle and the relation of these elements to each other (Fig. 2). Osteoarthritis appears to be a mechanically driven but chemically-mediated disease process, in which there is attempted (or aberrant) repair (Bleasel and Moskowitz 1995). An evolutionary theory has been put forward to help explain joint distribution (Hutton 1987) based on an understanding that it is a relatively recent evolutionary event that human joints provide pincer capability and full weight bearing on the legs. A strong genetic component has been proposed to occur in the biochemical constituents of bone and cartilage that places a person at increased risk. The biology and biochemistry of osteoarthritis is reviewed by Hamerman (1989).
Fig. 2 A schema depicting the aetiology of osteoarthritis.
Overweight persons have a higher than expected prevalence of knee osteoarthritis. This is true whether knee osteoarthritis is defined by symptom or radiograph and irrespective of whether the focus is on tibiofemoral (Anderson and Felson 1988; Felson et al. 1988) or patellofemoral disease (Manninen et al. 1996). The association between obesity and osteoarthritis of the knee is stronger for bilateral than unilateral disease and more marked in women than in men. Prospective data in women suggest that risk for knee osteoarthritis is increased by approximately 15 per cent for each additional kilogram per square metre of body mass index above 27 (Sowers et al. 2000). There is prospective evidence that weight loss reduces the risk of subsequent symptomatic knee disease (Felson et al. 1992).
The association with weight is weaker or absent for the hip in some studies (Kellgren 1961; Tepper and Hochberg 1993). Symptomatic hip osteoarthritis is more strongly associated with obesity than isolated radiographic disease (Roach et al. 1994).
Recent longitudinal studies show that increased weight precedes the occurrence of hand or knee osteoarthritis and is not merely a consequence of it (Carman et al. 1994; Manninen et al. 1996). For example, Manninen et al. found that body mass index was directly and strongly correlated with the risk of developing disabling knee osteoarthritis over a 10-year period. Furthermore, overweight persons with knee osteoarthritis are at higher risk of experiencing progressive disease than persons who are not overweight (Dougados et al. 1992).
Weight could act through several mechanisms to contribute to the development of osteoarthritis. Firstly, being overweight increases the amount of force across a weight-bearing joint (Schipplein and Andriacchi 1991). In addition, excess adipose tissue may produce abnormal levels of hormones or growth factors that affect cartilage or underlying bone so as to predispose to osteoarthritis development, although specific metabolic products have not been identified.
Evidence concerning sport and leisure activity as risk factors is conflicting. In part, this conflict may exist because, as a number of studies now indicate, brief applications of physiological loads, in the absence of traumatic twist or torque, stimulate matrix synthesis necessary to maintain cartilage, whereas extended application of loads can lead to cartilage deterioration. Studies of recreational and competitive athletes suggest that regular exercise, including jogging or moderate low-impact running, is not a detectable risk factor for hip and knee osteoarthritis in those with normal joints. However, these activities may increase risk in those with previous joint injuries or developmentally defective joints. Also, individuals who participate in certain other sports such as American football may have an increased risk (Vincelette et al. 1972). Prolonged decreased joint use and decreased loading generates changes that make cartilage more vulnerable to injury (Buckwalter et al. 1995).
Increased risks of both hip and knee osteoarthritis occur in men and women with jobs that stress the lower limbs. Studies have suggested that knee bending and heavy lifting are associated with radiographic knee osteoarthritis (Felson et al. 1991), and that heavy lifting among farmers confers enhanced risk of radiographic hip osteoarthritis (Croft et al. 1992). These risks are cumulative and there is evidence of increasing risk with greater number of years spent in such occupations. Hadler et al. (1978) showed that women who had worked for many years within the same cotton mill in the United States had patterns of osteoarthritis in the hand that could be directly related to specific tasks performed in the factory.
Persons with osteoarthritis of the knee or hip are more likely to have higher bone mass than others of their age (Sowers et al. 1996). Various hypotheses have been proposed to explain a negative association between osteoarthritis and osteoporosis, primarily centred around responses to mechanical forces. It has been hypothesized that stiff subchondral bone, as reflected in greater bone mass, increases cartilage damage with normal joint loading and leads to the development of osteoarthritis (Radin et al. 1972). Alternative explanations for the negative association between osteoarthritis and osteoporosis include genetic make-up and differences in response to changing ovarian hormone levels. Also, persons with osteoarthritis may have enhanced osteoblast production and/or a mineralization deposition defect as suggested by the presence of osteophytes and joint space narrowing in hip osteoarthritis (Sowers et al. 1999a).
Rare inherited diseases of cartilage have been recognized for many years. Also, the familial aggregation of Heberden’s nodes (bony protuberances on the margins of distal interphalangeal joints) was first reported many years ago by Stecher (1941). The genetics of the common forms of osteoarthritis are now of considerable interest, as recent studies have suggested that at least 50 per cent of the variability of osteoarthritis in the hands, knees, and hips is accounted for by genetic factors (Cicuttini and Spector 1996; Holderbaum et al. 1999). Studies have uncovered rare families with mutations in the collagen 2AI gene whose members express precocious osteoarthritis and varying degrees of chondrodysplasia. Collagen IX mutations have also been implicated. To date these mutations do not appear to influence common forms of osteoarthritis (Holderbaum et al. 1999). Current studies are underway to evaluate other candidate genes.
Based on current knowledge of risk factors, primary prevention is targeted at avoiding joint trauma, preventing obesity, and modifying occupationally-related joint stress through ergonomic approaches. Techniques for avoiding joint trauma include providing appropriate protection and padding for contact sports and modifying game rules to minimize the kind of contact that is associated with joint trauma (Buckwalter et al. 1995). Prompt diagnosis and treatment of earlier joint disease reduces the likelihood of osteoarthritis in these joints.
Some research suggests the importance of developing and maintaining adequate muscle strength as a mechanism to minimize joint stress. Ongoing research and intervention evaluations are aimed at identifying optimal conditioning exercise programmes to enhance musculoskeletal fitness without disease exacerbation of joint damage.
Weight reduction could potentially afford some protection against both the development and progression of osteoarthritis. Reduction of weight has been shown to reduce pain in symptomatic osteoarthritis of the knee (Felson et al. 1992).
Therapy can involve the use of analgesics and/or non-steroidal anti-inflammatory drugs to control symptoms of pain, combined with physical and occupational therapy to assure joint range of motion, muscle strength, and an individual’s ability to perform activities of daily living. Among those with severe pain and activity limitation, surgery, including joint replacement, is increasingly common. The role of social support is also recognized as important. The use of complementary therapies such as glucosamines and chondroitin sulphate is currently being evaluated in clinical trials.
Rheumatoid arthritis is a chronic inflammatory joint disorder characterized by proliferative synovitis leading to destruction of the articular cartilage and to bony erosions. These typically result in joint deformities. Although symptoms may vary, they typically include pain and stiffness of multiple joints (particularly the small joints of the hands and feet), soft-tissue swelling, increased temperature of affected joints, limitation in range of motion, weakness, fatigue, and loss of endurance.
Rheumatoid arthritis is usually not a fatal disorder, but generally results in substantial chronic disability. The disease is also associated with substantial comorbidity from respiratory and infectious diseases as well as gastrointestinal diseases (Kelsey and Hochberg 1988). Table 2 shows a revised classification for the diagnosis of rheumatoid arthritis published in 1987 by the American College of Rheumatology (ACR). Positive classification requires a positive finding for at least four out of the seven criteria (Arnett et al. 1988).
Table 2 The 1987 American Rheumatism Association Revised Criteria for the Classification of Rheumatoid Arthritisa
The prevalence of rheumatoid arthritis reported from the Rochester (Minnesota) Epidemiology Project (based on the 1987 ACR criteria) is 1 per cent (Gabriel et al. 1999), a prevalence consistent with the 0.8 per cent reported in several European populations (Silman and Hochberg 1993). The overall age- and sex-adjusted annual incidence in Rochester, Minnesota, residents aged 35 years and older from 1955 to 1985 was 75.3 per 100 000. There is evidence of a decline in rheumatoid arthritis incidence over the last few decades in both the Unied States and the United Kingdom (Linos et al. 1980; Hochberg 1990; Silman and Hochberg 1993; Gabriel et al. 1999). As incidence rates have been declining, prevalence has been rising, probably due in part to improved survivorship (Hochberg 1990).
Overall, the incidence is about twice as great in females as males. In Rochester, Minnesota, the annual incidence ranged from 17 per 100 000 in men aged 35 to 44 years to 111 per 100 000 in men aged 75 to 84 years. The incidence was 50 per 100 000 in women aged 35 to 44 years and 123 per 100 000 in women aged 75 to 84 years (Gabriel et al. 1999).
Rheumatoid arthritis prevalences are similar in African-Americans and white Americans, but higher prevalences of rheumatoid arthritis have been reported among several Native American tribes (Beasley et al. 1973; Harvey et al. 1983). Lower prevalences have been reported among some Asian populations (Beasley et al. 1983). Asians may also have more mild disease compared with Caucasians (Silman and Hochberg 1993; Lau et al. 1996). Reasons for this variation are not known.
The aetiology of rheumatoid arthritis is ill defined. Familial aggregation and a higher concordance rate in monozygotic than in dizygotic twins contribute to the characterization of rheumatoid arthritis as having a substantial heritable component. Genetic studies have focused primarily on autoimmune aspects of the disease, especially the role of the major histocompatibility locus. There is a strong association between rheumatoid arthritis and the class II major histocompatibility antigen HLA DR4, broadly observed across most racial/ethnic groups, and particularly in Caucasians, where the estimate of the relative risk exceeds 4.0 (Ollier and MacGregor 1995). Other work has identified an association with HLA DR1 and the shared epitope (Gregerson et al. 1987) between HLA DR1 and HLA DR4 (HLA DR1B1 gene on chromosome 6). The shared epitope is considered a marker for disease severity rather than susceptibility (Weyand et al. 1992). Current investigations are also exploring other genetic factors, such as HLA DR3/DR7 and an autoimmunity to type II collagen (Ollier and MacGregor 1995).
While the potential contribution of genetic susceptibility is widely recognized, there remain questions as to the initiating events that give rise to rheumatoid arthritis and its patterns of expression. Infectious agents that have been proposed and/or explored include the Epstein–Barr virus, parvovirus, rubella, and mycoplasma (Alarcón 1995), but no strong evidence exists for an aetiological role for these agents.
Based on early observations of symptom improvement during pregnancy, both hormonal and non-hormonal reproductive factors have been considered. It has been postulated that during pregnancy not only are there alterations in circulating hormone concentrations but that the placenta may also contribute glycoproteins with anti-inflammatory or immunosuppressive properties, particularly in those pregnancies where there are disparities in the mother and fetus HLA antigens (Alarcón 1995).
A role for hormones, including oestrogens, testosterone, and prolactin, has been suggested in both the initiation and the severity of rheumatoid arthritis. Early epidemiological studies reported that oral contraceptives were protective against rheumatoid arthritis (Wingrave 1978). A possible role of oral contraceptives was also suggested by the declining incidence rates among females, but not males, from the period 1960 to 1964 through to 1970 to 1974 (Linos et al. 1980). A meta-analysis of epidemiological studies has reported a protective effect (relative risk of ≅ 0.7), with the authors suggesting that the protection was more likely to be associated with progression rather than initiation of rheumatoid arthritis (Spector and Hochberg 1990). This latter suggestion however, does not account for the findings of an apparent shift in the age at onset to older ages. Thus, while a role for exogenous hormonal influence is intriguing, additional studies are needed to evaluate the use of exogenous hormones in the form of replacement therapy as well as contraception. Account must be taken of the selection factors that give rise to their use, as well as to the dose, constitutive hormones, and duration of use.
No viable screening or primary prevention measures are available for rheumatoid arthritis. Because of the excess mortality associated with respiratory and infectious diseases, public health interventions should include full utilization of vaccination programmes and the practice of hygienic techniques that limit the opportunity for exposure to infectious agents or transmission of infectious agents. Treatment of rheumatoid arthritis includes the use of analgesics and/or nonsteroidal anti-inflammatory drugs to control pain and stiffness, physical therapy to maintain and improve joint range of motion, and occupational therapy to maximize the individual’s ability to perform daily activities. Some persons with severe rheumatoid arthritis may be treated with immunosuppressive agents.
Osteoporosis and associated fractures
Osteoporosis has been defined at a Consensus Conference as a disease characterized by low bone mass and microarchitectual deterioration of bone tissue leading to enhanced bone fragility and a consequent increase in fracture risk (Anonymous 1997). Fractures of the hip, vertebrae, and distal radius are particularly common. The World Health Organization (WHO) defined osteoporosis as bone mineral density more than 2.5 standard deviations below the mean value of peak bone mass in normal young women (WHO 1994). It has been estimated that, by this definition, 70 per cent of women of age 80 years and older in the United States have osteoporosis in the hip, spine, or distal radius. Almost all women in this age group have bone mass that is below the mean for normal young women (Melton 1995).
Osteoporosis is much more common in females than males. In both females and males, its prevalence increases markedly with age, but the increase in prevalence occurs about 10 years earlier in females than males. Table 3 shows prevalence by age in females. A particularly rapid decrease in bone mass occurs in females in the years immediately following menopause, suggesting that loss of oestrogen is an important aetiological factor in women.
Table 3 Percentage of women in Rochester, Minnesota, with bone mineral measurements in the spine, hip, or mid radius more than 2.5 standard deviations below the mean for young normal women
Globally, hip fracture incidence rates are highest in white people in northern European and North American countries, slightly lower in Asians living in economically developed areas such as Hong Kong and the United States, still lower in Hispanics and black people in the United States and in South America, and lowest in less developed areas of Asia such as China and in Africa (Maggi et al. 1991; Schwartz et al. 1999). In the United States, hip fracture incidence rates are lowest in black people, highest in white people, and intermediate in Asians and Hispanic white people (Villa and Nelson 1996). Although bone mass is highest in black people, there is little difference in the bone mass of white people and Asian-Americans. Thus bone mass cannot be the only factor that explains the differing hip fracture incidence rates in these groups. Bone architecture, fall frequency, and manner of falling are also believed to be important.
Bone mass in later adulthood, when fracture risk is greatest, is a function of bone mass in young adulthood, when bone mass is at its peak, and rate of loss of bone mass after the peak is reached. Numerous family and twin studies have demonstrated a strong genetic component to bone mass in young adulthood. Estimates of the heritability of bone mass range from 45 to 84 per cent, depending on the skeletal site examined (Slemenda et al. 1991; Sowers et al. 1992). Numerous genes have been considered for their association with bone mass, including the genotypes for the vitamin D receptor, the sex hormone receptor genes, the matrix proteins (such as osteocalcin), and type I collagen genes (Morrison et al. 1992; Willing et al. 1998; Sowers et al. 1999b). However, the relative importance of any one gene, particularly that of specific vitamin D receptor genotypes, has been controversial (Eisman 1995; Peacock 1995). Moreover, causative mutations have yet to be identified in any genes, except in a few anecdotal cases (Spotila et al. 1994). Although results of studies are not consistent, it appears that such modifiable risk factors as weight, calcium intake, and physical activity also affect premenopausal bone mass but to a lesser extent than heredity (Bonjour and Rizzoli 1996). In some young people, anorexia nervosa and excessive athletic activity contribute to low bone mass. Premenopausal oophorectomy results in loss of bone mass if hormone replacement therapy or another appropriate pharmacological agent is not used.
Other factors known to increase the risk for low bone mass and osteoporotic fractures in postmenopausal women are prolonged immobility, prolonged corticosteroid use, a family history of an osteoporotic fracture, and cigarette smoking (Cumming et al. 1997). Established protective factors are hormone replacement therapy (in either the form of oestrogen alone or oestrogen with progestin), certain other pharmaceutical agents (discussed below), obesity, and, to a lesser extent, calcium supplementation. Figure 3 shows that hormone replacement therapy protects against bone loss for as long as it is used, but loss of bone mass continues when replacement hormone use ceases (Christiansen et al. 1981). Available evidence indicates that hormone replacement therapy also protects against osteoporotic fractures, but that recent use is again needed for this protection. Higher concentration of endogenous oestrogen around the time of menopause is also associated with a lower rate of bone loss (Cauley et al. 1996). Obese postmenopausal women have a reduced risk for low bone mass compared with thin women. On average, thin women have lower oestrogen production, lower concentration of circulating oestrogen, and less mechanical stress on their bones. In addition, fat padding around the hip provides some protection against fracture during a fall. Most randomized trials show a small protective effect of calcium supplementation that is not nearly so great as the protection from hormone replacement therapy (Cumming 1990; Heaney 1996). However, a trial of calcium and vitamin D supplementation among institutionalized elderly in France (Chapay et al. 1992) showed a 43 per cent decrease in risk for hip fracture. Older postmenopausal women whose dietary calcium intake is very low may be especially likely to benefit from supplemental calcium (Dawson-Hughes et al. 1990). Cigarette smoking increases the risk for osteoporosis, probably through a lowering of oestrogen concentration (Seeman 1996). Heredity plays a role in determining postmenopausal bone mass, but its contribution is not nearly so great as for premenopausal bone mass (Sambrook et al. 1996).
Fig. 3 Bone mineral content (BMC) as a function of time and treatment in 94 (study I) and 77 (study II) women soon after menopause. (Source: Christiansen et al. 1981.)
Most studies report that thiazide diuretics, which decrease urinary calcium excretion, are associated with increased bone mass and decreased hip fracture risk in adults (Cauley et al. 1996). Some studies, but not all, have shown an association between low levels of dietary calcium and osteoporosis. Among adults, the amount of dietary calcium consumed during old age may be especially significant (Heaney 1996). Sufficient vitamin D intake would be expected to be necessary for adequate calcium intake, but evidence to date is not consistent. Moderate physical activity probably affords a small amount of protection in adults, but this protective effect is probably lost if a person stops being physically active (Snow et al. 1996). Heavy alcohol consumption and caffeine consumption may also increase the risk for low bone mass and for fractures, but data are not consistent. The roles of other dietary constituents are even less certain. At present an area of active investigation is whether phyto-oestrogens protect against loss of bone mass.
In addition to low bone mass, certain architectural and geometric properties of bone affect fracture risk. One such geometric property is hip axis length, which is the distance from the greater trochanter to the inner pelvic rim. Long hip axis length is associated with a higher risk for hip fracture independent of bone mass (Faulkner et al. 1993). Differences in hip axis length may explain some of the variation in hip fracture incidence rates among countries and racial groups, since Asians and black people have shorter hip axis length than white people (Villa and Nelson 1996). An architectural property of bone that appears to affect hip fracture risk is the Singh index in the proximal femur (Singh et al. 1970). The Singh index ranges from grade VI (normal, all trabecular groups visible) to grade I (marked reduction of even the principal compressive trabeculae). Hip fracture patients have been found to have lower Singh grades, on average, than women of similar age without hip fracture (Peacock et al. 1995). It is likely that more geometric and architectural properties of bone that are associated with fracture risk will be identified in the future.
Most fractures of sites other than the spine depend on whether a person falls and how the person falls. Risk factors for falls have been found to be risk factors for hip fracture (Grisso et al. 1996). Number of previous falls and a recent increase in the number of falls are both predictive of hip fracture (Schwartz et al. 2000). Table 4 shows some host factors associated with falling. The roles of environmental hazards are less well documented (Grisso et al. 1996).
Table 4 Host risk factors for falls among the elderly for which the evidence is strong or moderate
How a person falls also affects the likelihood that a fracture will occur. Falling sideways or straight down and landing on the hip or leg, for instance, greatly increase the risk of a hip fracture, while breaking the fall with a hand decreases the risk of hip fracture (but increases risk of lower forearm fracture) (Hayes et al. 1993; Nevitt and Cummings 1993). Thus, not only will the likelihood of fracture depend on how a person falls, but the nature of the fall will also largely determine the skeletal site fractured. Frail women are more likely to fracture their hip or proximal humerus, while more healthy active women are more like to fracture their lower forearm (Kelsey et al. 1992; Nevitt and Cummings 1993). Falls from heights or on hard surfaces also increase the risk for fracture (Hayes et al. 1993; Nevitt and Cummings 1993).
Primary prevention includes trying to achieve high bone mass at young ages by having a diet appropriate in calcium and engaging in sufficient physical activity. Once loss of bone mass has begun, administration of oestrogens with or without progestin will limit further loss of bone mass, but the other benefits and risks of hormone replacement therapy need to be considered as well. Calcium supplementation can provide some protection. Moderate physical activity, such as brisk walking, is often recommended for older people to reduce loss of bone mass, but at most this probably has only a modest beneficial effect.
Two relatively new agents are available to reduce loss of bone mass. The bisphosphonate alendronate has been shown to reduce loss of bone mass in the hip and spine, although not to the same extent as hormone replacement therapy (Hosking et al. 1998). However, randomized trials in women have shown that it protects against fracture only in those who already have very low bone mass (Cummings et al. 1998). In addition, compliance is a problem because alendronate has to be taken in a rather strict manner to achieve maximal absorption and avoid unpleasant upper gastrointestinal effects (Ettinger et al. 1998). The selective oestrogen receptor modulator raloxifene has also been shown to reduce loss of bone mass in the hip and spine in women regardless of their baseline bone mass (Delmas et al. 1997; Ettinger et al. 1999) although, again, not to the same extent as hormone replacement therapy. However, in the one randomized trial with fracture as an endpoint, raloxifene protected against vertebral fractures but not against non-vertebral fractures, including hip fracture (Ettinger et al. 1999). Thus available evidence suggests that neither alendronate nor raloxifene will be useful in the primary prevention of osteoporosis, although both have utility as therapeutic agents.
Screening women in the perimenopausal and early postmenopausal years for high fracture risk by measuring their bone mass with either dual-energy X-ray absorptiometry or single- or dual-photon absorptiometry has achieved some popularity in recent years. However, there are many questions about the usefulness of such screening, such as who should be screened, whether multiple measurement over time are needed, what other information on risk should be obtained along with the measure of bone mass, and what therapy should be used in those with various degrees of low bone mass (Slemenda et al. 1996). Ultrasonography, usually of the heel bone, provides information about bone architecture and elasticity as well as bone mineral density, and has been found to predict fracture (Hans et al. 1996). Since ultrasound is less expensive and radiation-free compared with other methods of measuring bone mass, it is possible that it will be used more widely for screening in the future. At present, however, there is no consensus that any screening method should be used for healthy women in the general population.
Finally, reducing the frequency and impact of falls among those with osteoporosis is another potential means of reducing the risk of fracture. A randomized trial in a nursing home has shown that wearing protective hip pads substantially reduces the risk of hip fracture (Lauritzen et al. 1993), but compliance is a problem.
Low back pain
From 60 to 80 per cent of adults report having had low back pain at some time during their lives (Hult 1954; Walsh et al. 1992; Papageorgiou et al. 1995). In the United States, 4.6 per cent of workers lose at least one workday during the course of a year, and a total of 10 800 000 workdays are lost annually because of back pain (Guo et al. 1999). Fortunately, most episodes of low back pain are of limited duration. A Swedish study, for instance, found that 60 per cent of males in the general population had at some time had low back pain, 16 per cent had been incapacitated for periods ranging from 3 to 6 months, and 4 per cent had been incapacitated for more than 6 months (Hult 1954). The small proportion of cases that become chronic account for most of the cost associated with low back pain. Snook (1982) estimated that 25 per cent of the cases account for 90 per cent of the costs.
As noted above, the category of low back pain consists of a variety of entities with somewhat different aetiologies. However, almost all studies have considered low back pain as a whole, although a few have considered herniated lumbar disk specifically. When risk factors for low back pain as a whole and herniated lumbar disk differ, it will be so noted.
The best predictor of low back pain is a history of low back pain (Papageorgiou et al. 1996; Lagerström et al. 1998; van Poppel et al. 1998). First episodes of low back pain most frequently occur among persons in the age range 20 to 39 years, but the proportion of the population reporting low back pain (either old or new) is relatively uniform across the working years (Biering-Sorenson 1982; Guo et al. 1999). Whereas after about age 65 years the prevalence of low back pain decreases in men, prevalence increases in women, probably because of the increasing frequency of vertebral osteoporosis. If only cases seeking medical care or compensation are considered, low back pain is seen more frequently in males than females (Snook 1982; Shelerud 1998), but in surveys in the general population, males and females are affected with approximately equal frequency (Biering-Sorenson 1982; Walsh et al. 1992; Papageorgiou et al. 1995). Persons in low social classes are more likely to report low back pain than those in higher classes, probably because of their tendency to have jobs requiring heavy physical labour.
People who do heavy manual labour are at increased risk for low back pain (Hales and Bernard 1996). Lifting objects of 11.25 kg or more appears to be particularly detrimental. Specific activities that probably further increase the risk are frequent lifting of heavy objects while bending and twisting the body, holding heavy objects away from the body while lifting, and failing to bend the knees while lifting (Table 5) (Andersson 1981; Kelsey et al. 1984b). Nursing is one occupation with a particularly high risk for low back pain. Nurses who transfer patients between a bed and a chair, who manually reposition patients, who lift patients in and out of the bath with a hoist, and especially those who must lift in ‘save the patient’ situations are especially prone to low back pain (Lagerström et al. 1998).
Table 5 Estimated relative risk for prolapsed lumbar intervertebral disc associated with lifting more than 11.25 kg on job
Driving of motor vehicles either on or off the job and exposure to whole-body vibration increase the risk for low back pain (Pope et al. 1998 ). Evidence is inconsistent as to whether prolonged sitting or standing in one position confers an increased risk (Hales and Bernard 1996).
Cigarette smoking is associated with an increased risk, probably because of the pressure exerted by frequent coughing or the decreased diffusion of nutrients into the intervertebral disks resulting from smoking (Heliövaara et al. 1991). Some studies indicate that tallness is a risk factor for low back pain or specifically sciatica (Heliövaara et al. 1991), and that heavy body weight has no, or at most a slight, effect (Hales and Bernard 1996). A narrow spinal canal increases the risk for herniated lumbar disk (Heliövaara et al. 1986).
The role of psychological factors in the aetiology of low back pain has been difficult to determine because it is often difficult to separate whether the psychological factors preceded or followed the onset of low back pain. The psychological factor most consistently predictive of low back pain in longitudinal studies is job dissatisfaction (Papageorgiou et al. 1997; Krause et al. 1998). Other factors reported from longitudinal studies include a high frequency of job problems, psychological job demands, poor social relations at work, work pressure, lack of job control, and monotonous work (Hales and Bernard 1996).
Most low back pain improves without any specific treatment. Predictors of disability from low back pain reported in various studies include previous episodes of low back pain, long duration of pain, a history of past disability and hospital admissions, sciatica and use of analgesia in association with previous low back pain, an onset of pain attributable to trauma, lack of recognition and respect at work, low supervisory support, being unemployed, other disabilities, low educational level, psychological factors, heavy physical demands on the job, dissatisfaction with the job, whether insurance payments are being received, the perception of fault, whether a lawyer has been retained, and heavy body weight (Deyo and Diehl 1988; Cats-Baril and Frymoyer 1991; Hales and Bernard 1996; Wickstrom and Pentti 1998; Müller et al. 1999).
Various approaches have been used for the primary prevention of low back pain in the workplace. Low back radiographs and medical examinations have not proved useful as predictors of who will develop back pain on the job (Snook 1982). However, careful selection of workers for jobs involving heavy manual work by strength testing may be helpful (Keyserling et al. 1980). A randomized trial of an educational programme to reduce back injury in postal workers did not find that the programme reduced the occurrence of back injury, the cost per injury, the time off work per injury, or the rate of repeat injury after return to work (Daltroy et al. 1997). Similarly, available evidence suggests that better training in nurses does not result in a decrease in the frequency of low back problems (Lagerström et al. 1998). It has been found that training workers to bend their knees while lifting does not reduce the likelihood of low back injuries, in part because of poor compliance (Snook 1982). A better approach may be to redesign jobs to minimize bending and twisting motions while lifting and to reduce the amount of weight that must be lifted (Snook 1988). Redesigning jobs in these ways may also allow injured workers to return to work sooner. Other ways of reducing the frequency of back pain may include smoking cessation, improved physical fitness, moving around from time to time in situations requiring prolonged exposure to one position, vibration dampening, and motor vehicles with good lumbar support and positioning.
To reduce the likelihood of acute back pain progressing to chronic back pain, it is important for those affected to continue their normal activity to the extent that they are able (Malmivaara et al. 1995). Also important is return to work after a short period of rest (Nachemson 1983). Upon return to work, however, the worker should avoid activities that may exacerbate the problem, such as heavy lifting or staying in one position for long periods of time.
Back schools have been developed in an attempt to decrease pain and disability through self-involvement and self-reliance (Fish et al. 1983). Most back schools teach patients about spinal mechanics so that they can use their backs effectively without pain and damage, try to change attitudes through psychological approaches, and encourage appropriate exercise and physical fitness. There is some evidence to suggest that back schools are quite effective for people with recent onset of low back pain, but not for those with chronic back pain (Lankhurst et al. 1983).
Neck pain has been less well studied than low back pain, and, as with low back pain, a variety of conditions can result in neck pain. About 50 to 70 per cent of adults report having experienced neck pain at some time during their lives (Hult 1954; Côté et al. 1998a).
Some studies report that males and females are affected with approximately equal frequency, while others show a female excess. Chronic neck pain and disability from neck pain occur more frequently in females than males (Bovim et al. 1994; Côté et al. 1998a; Leclerc et al. 1999), but males appear to be affected more frequently with herniated cervical intervertebral disk (Kelsey et al. 1984a). Most neck pain is mild, but about 5 per cent results in disability (Hult 1954; Côté et al. 1998a). Neck pain occurs most frequently in young adulthood, and there is some evidence that its prevalence decreases with age (Côté 1998a; Leclerc et al. 1999).
Only a few studies to identify other risk factors for either neck pain in general or herniated cervical intervertebral disk in particular have been undertaken. Prolonged exposure to awkward postures may be associated with mild neck pain. For instance, frequent use of video display terminals with a fixed keyboard height that requires a bent neck can result in neck pain (Yu and Wong 1996). Other risk factors for either general neck pain or herniated cervical disk found in one or more studies include heavy lifting (Kelsey et al. 1984a; Magnusson et al. 1996; Krause et al. 1997), cigarette smoking (Kelsey et al. 1984a; Brage and Bjeredal 1996), driving motor vehicles (Kelsey et al. 1984a; Krause et al. 1997), exposure to whole-body vibration (Kelsey et al. 1984a; Magnusson et al. 1996), psychological distress (Leclerc et al. 1999), psychosomatic problems (Leclerc et al. 1999), headaches (Leclerc et al. 1999), and frequent diving from a board (Kelsey et al. 1984a). One study found that vehicles in which there were problems in adjusting the seat were particularly detrimental (Krause et al. 1997). Repetitive motions, forceful exertions, and constrained positions may increase risk (Hales and Bernard 1996).
Little research has been undertaken on predictors of disability from neck pain and on primary prevention. However, it would appear that reduction in the amount of heavy lifting, cigarette smoking, awkward positions (especially at video display terminals), prolonged driving in motor vehicles, and exposure to other sources of whole-body vibration should reduce the incidence of neck pain.
Selected musculoskeletal disorders of children
The Scoliosis Research Society defines scoliosis as a lateral spinal curve of greater than 10° (Skaggs and Bassett 1996). It is usually associated with rotation of the vertebrae. Only adolescent idiopathic scoliosis will be discussed here. About 2 to 3 per cent of adolescents develop curves of 10° of more before growth ceases, and 2 to 3 per 1000 children develop curves of 30° or more (Shands and Eisberg 1955; Morais et al. 1985). Those with large curvature usually develop spinal osteoarthritis in their adult years. Lung and heart complications may also occur. Additional progression of curves sometimes takes place in adults with scoliosis, thus worsening the situation.
The ages at which adolescent idiopathic scoliosis is most frequently diagnosed are 11 to 14 years in girls and 14 to 16 years in boys, the difference in ages reflecting the later onset of the adolescent growth spurt in males. The ratio of female to male cases among the more severe cases seen at surgery is about 5 to 1, but curves of less than 15° are seen with about equal frequency in females and males. Reports from surgical case series indicate that curves occur most frequently at the thoracic level, but school screening programmes that identify mild as well as severe cases have found curves to be most common at the thoracolumbar level (Brooks et al. 1975).
The risk for scoliosis in first-degree relatives of cases is about three to four times that in children from the general population (Wynne-Davies 1968). A recent prospective study in a cohort of prepubertal children in Finland free of scoliosis at baseline found several factors that predict the development of scoliosis, including the female sex, trunk asymmetry as indicated by large rip humps on the forward bend test (described below), the degree of thoracic kyphosis, and the degree of lumbar lordosis in boys. Standing height, sitting height, recent increase in sitting height, and early age at gain in sitting height appeared to be risk factors as well, although they did not reach statistical significance (Nissinen et al. 1993a,b). Other reports have also indicated that children who are skeletally more mature at the beginning but not at the end of puberty and children who are taller and thinner than average at the beginning but not at the end of adolescence are at high risk, suggesting that the scoliotic spine grows faster and earlier than the normal spine (Willner 1975; Nissinen et al. 1993b; Hazebroek-Kampschreur et al. 1994). Once girls reach menarche, their risk is considerably reduced (Hazebroek-Kampschreur et al. 1994). Other factors that may contribute to the development of scoliosis but for which the evidence is not conclusive are impaired visual and vestibular functioning (Nachemson and Sahlstrand 1977), defects in proprioceptive postural control (Keesen et al. 1992), asymmetric muscle activity (Henssge 1968; Alexander and Season 1978), leg-length discrepancies (Nachemson and Sahlstrand 1977), collagen disorders (Francis et al. 1977), and abnormalities of the elastic fibre system (Hadley-Miller et al. 1994).
Once a curve has developed, several risk factors for progression have been identified, including double curves as opposed to single curves, thoracic curves, larger curves, curves in females, the absence of a sacral tilt, leg-length inequality, early chronological age, and skeletal immaturity (Dickson et al. 1980; Lonstein 1988).
No means of primary prevention are known. Therefore secondary prevention by screening in schools has become widespread and is in fact required by law in many localities in the United States. The assumption of the screening programme is that if cases of scoliosis are detected early, they can be treated conservatively and surgery can be avoided.
The forward bend test has been used for screening for many years. In the forward bend test, the child’s back is examined while the child bends forward from the waist. In scoliosis, the rotation of vertebrae that is often associated with the lateral curvature results in prominent ribs on the concave side of the curve. Accordingly, in this test evidence of a ‘rib hump’ is considered positive for scoliosis. Nissinen et al. (1993b) suggest that hump size of 6 mm or greater is the optimal cutoff point for a positive test. In another screening test called moiré topography, a photograph of the back is taken and the degree of topographic asymmetry is measured. Moiré topography has high sensitivity, but because of the large number of false positives it is no longer recommended (Nissinen et al. 1993b). Recently, the scoliometer, an inclinometer used to measure axial trunk rotation during forward bending (Bunnell 1984), has been introduced as a screening test. Only limited evaluation of the scoliometer has been undertaken to date (Grossman et al. 1995; Côté et al. 1998b). Most school screening programmes use the traditional forward bend test.
School screening programmes identify relatively large numbers of children with possible curvature in their spine. Positive tests are followed by radiographic examination for more definitive diagnosis. Curves of more than 5° are generally monitored by subsequent radiographs every few months. If a curve progresses to 20 to 25° or more, treatment is usually started to prevent further progression. Methods of treatment include exercises, braces, external or internal muscle stimulators, and sometimes surgery.
Despite the widespread acceptance of screening for scoliosis, its effectiveness has been questioned (Williams 1988; United States Preventive Services Task Force 1993b; Goldberg et al. 1995). Many children screened as positive are not followed up for definitive diagnosis. Many false positives occur even with the forward bend test, resulting in an excess of radiograph examinations and considerable expense and anxiety. A report from a screening programme in Ireland (Goldberg et al. 1995) indicates that the positive predictive value of the forward bend test is only 8 per cent, i.e. of every 100 forward bend tests classified as positive, only eight develop clinically significant scoliosis. It is uncertain that school screening programmes have actually caused a reduction in the number of cases of severe curvature that require surgery. There is disagreement about the optimal ages for screening and about whether males should be screened at all. Criteria for referral for diagnosis and treatment need to be better specified, and improved training and evaluation of the nurses who do the screening is needed (Morais et al. 1985; Williams 1988). It would be helpful if predictors of progression were better defined. In view of all these uncertainties, the United States Preventive Services Task Force (1993a) did not recommend for or against routine screening for scoliosis in adolescents.
Slipped capital femoral epiphysis
In slipped capital femoral epiphysis, the head of the femur is displaced backward and downward off the diaphysis. The actual separation takes place through the layer of hypertrophied cartilage next to the zone of calcified cartilage of the epiphyseal plate. The usual symptoms are pain, stiffness, limp, and a limited range of motion of the hip joint. Slipped epiphysis usually occurs during the adolescent growth spurt, and does not occur once the epiphysis is fused to the shaft of the femur.
The annual incidence of slipped epiphysis has been reported to be 3.4 per 100 000 in the population under age 25 years in Connecticut, United States (Kelsey 1971), 0.7 per 100 000 in New Mexico (Kelsey et al. 1970), and to range from 2 to 13 per 100 000 in the age group 7 to 17 years in Gothenburg, Sweden, over a period of several years (Henrikson 1969). In Connecticut and Gothenburg, about 1 in 800 males and 1 in 2000 females will be diagnosed with a slipped epiphysis over the age range at risk (Kelsey 1971; Jerre et al. 1996).
Most cases occur between the ages of 10 and 17 years in males and between 8 and 15 years in females. The median age at diagnosis is 13 years for males and 11 years for females (Kelsey 1971; Jerre I>et al. 1996), the earlier age in females corresponding to their earlier onset of puberty. Males are affected more frequently than females, although male to female ratio appears to have decreased over time and varies from one geographical area to another (Hansson et al. 1987; Jerre et al. 1996; Loder 1996a). In males, the left hip is affected about twice as frequently as the right, whereas in females the left and right hips are affected with approximately equal frequency. In both sexes, about 20 to 25 per cent of those with slipped epiphysis have both hips affected (Sorenson 1968; Jerre et al. 1996; Loder 1996a). Black people are affected with slipped epiphysis two to three times more often than white people (Kelsey 1971). There are several reports that symptoms begin more frequently in spring and summer than in autumn or winter, at least in northern latitudes (Hansson et al. 1987; Loder 1996b).
A large proportion of children with slipped epiphysis are very overweight. In fact, about half are at or above the 95th percentile for their age (Morscher 1968; Kelsey et al. 1972). Children with slipped epiphysis tend to have undergone slower skeletal maturation than average for their age (Sorenson 1968). They tend to be tall for their age at the time of diagnosis, but at maturity their heights are almost normal for their chronological age (Hansson et al. 1987). Some familial aggregation of cases has been noted (Rennie 1967).
Most of the established risk factors for slipped epiphysis are related either to a weakening of the epiphyseal plate or to an increase in the shearing stress on the plate. The epiphyseal plate is weaker during periods of rapid growth, such as during the adolescent growth spurt. The growth spurt in males is of greater magnitude and of longer duration than that in females, and the male growth spurt is more likely to have periods of acceleration and deceleration (Acheson 1966). The excess of onset of symptoms in spring and summer may be attributable to more rapid growth in spring and summer (Morscher 1968).
Animal experiments indicate that a deficit of sex hormones relative to growth hormone brings about a widening of the epiphyseal plate and a reduction in the shearing force needed to displace the epiphysis. Oestrogens protect against slipped epiphysis, whereas androgens are protective only in large doses after prolonged exposure, providing another reason for the higher incidence in males than females. Slowly maturing children are exposed for a longer period to high levels of growth hormone relative to sex hormones than are children who mature faster. Tall children also would have a longer exposure to growth hormone relative to sex hormones (Morscher 1968).
During adolescence the epiphyseal plate changes from a horizontal to an oblique plane, so that it becomes more vulnerable to stress from superincumbent weight. Children who are overweight will put more stress on their plate than lighter children. Children with the unusual combination of being overweight and are slow maturers are probably at particularly high risk.
Prevention of adolescent obesity is the only known means of reducing the risk for slipped epiphysis. No screening tests are available, but slipped epiphysis should be considered in adolescents who have a limp and hip or knee pain or restriction of motion in the hip. The contralateral hip in children with slipped epiphysis in one hip needs to be carefully monitored, especially if the first slipped epiphysis occurred at an early age (Loder 1996a). Early diagnosis is important, as slight displacement treated early by pinning the hip has a favorable prognosis, whereas cases that are diagnosed late and that have severe displacement usually have early onset of osteoarthritis of the hip and permanent disability.
Figure 4 indicates that there are two age groups at particularly high risk for fractures: the elderly and children. Fractures in the elderly were considered in the section on osteoporosis. Here, epidemiological and preventive aspects of fractures in children are described.
Fig. 4 Average annual rate of persons having fractures, by age and sex, United States, 1992 to 1994, National Health Interview Survey. (Source: Praemer et al. 1999.)
In the United States, about 1 in 36 persons less than 18 years of age fractures a bone each year (Holbrook et al. 1984; Praemer et al. 1999). In Wales, 64 per cent of boys and 39 per cent of girls can expect to fracture a bone by 15 years of age (Lyons et al. 1999). Most fractures heal quickly in children, and the younger the age, the more rapid the healing. However, if the growth plate is involved in the fracture, growth in that bone may be adversely affected. Other rare complications, as in other age groups, are infection, delayed union, nonunion, avascular necrosis, and malunion.
Among children, incidence rates of fracture increase with age until about ages 11 to 14 years in females and until about ages 20 to 24 years in males, and males are at higher risk of fracture throughout childhood (Landin 1983; Worlock and Stower 1986; Cheng et al. 1999; Lyons et al. 1999). In one study (Cheng et al. 1999), the ratio of male to female fracture cases rose from 1.4 to 1 in infancy to 4.9 to 1 in adolescence. Almost all forms of trauma associated with fractures are more common in boys than girls (Landin 1983).
The bones reported to be most frequently fractured in children have varied somewhat from one series to another depending on the sources of cases. The United States Health Interview Survey indicates that the sites most frequently affected are the hand, radius and ulna (especially the distal radius and ulna), carpals, skull and face, clavicle, foot, and humerus (Holbrook et al. 1984). Children who have one fracture are at increased risk of having an additional fracture (Landin 1983).
Accidents while playing, sports accidents, and traffic accidents accounted for over half of childhood fractures in a large series from Sweden (Landin 1983). In a series of childhood fracture cases in Wales, sports and leisure-time activities accounted for 36 per cent of fractures, assaults for 3.5 per cent, and road traffic accidents for 1.4 per cent (Lyons et al. 1999). Among fractures of specific sites (Landin 1983), fractures of the phalanges of the hands most commonly resulted from contact sports as well as from skating, playing, and fighting. Distal forearm fractures were most often caused by a fall on an outstretched hand; when environmental factors were involved, ball games, bicycle accidents, playground accidents, and skateboard accidents were most common. Carpal and metacarpal fractures tended to result from fighting, falls, bicycle accidents, ball games, skiing, and skating. Fractures of the clavicle were most often the result of falls, ball games, and contact sports. Among foot fractures, the ankle was most frequently involved. Ankle fractures most often resulted from the foot being caught in a bicycle wheel and from falls, but ball games, skateboards, roller skates, mopeds, motorcycle, skiing, skating, cycling, and playing also were responsible for some ankle fractures.
Accident prevention is the key to reducing the number of fractures in children, including decreasing the number of sports and recreational injuries, falls, bicycle, motorcycle, and automobile accidents, non-accidental child injuries, and other childhood traumas (Garraway et al. 1979; Landin 1983; Lyons et al. 1999).
Developmental (congenital) dislocation/dysplasia of the hip
In developmental dislocation/dysplasia of the hip, the head of the femur is displaced completely or partially out of the acetabulum. Because many dislocations tend to occur immediately after birth or occasionally later during the first year of life, the term ‘developmental dislocation/dysplasia of the hip’ is now frequently used instead of the older term ‘congenital dislocation of the hip’ (Mooney and Eman 1995). The diagnosis is made shortly after birth in about 80 per cent of cases, while about 20 per cent of the time the diagnosis is made later, especially when the child starts to walk. Available evidence indicates that most of the late-diagnosed cases in fact do develop some time after birth and are not merely cases that were missed close to the time of birth (Mooney and Emans 1995).
Developmental dislocation/dysplasia of the hip includes the following: (a) dislocated hips, which are hips dislocated in a resting position, with the dislocation present at birth; (b) located but unstable or dislocatable hips, which are hips that rest in a located position but are unstable or dislocatable on clinical examination and provocative manoeuvres; (c) dysplastic hips, in which the acetabulum is shallow or dysplastic (Mooney and Emans 1995).
Many hips diagnosed as unstable or dislocatable immediately after birth stabilize during the first few days of life. Hips may become stable so quickly that estimates of prevalence based on examination after the third day of life are less than half of the prevalence estimates at birth (Sharrard 1993). Thus, prevalence estimates are highly dependent on the conditions under which the estimates were made.
These reservations not withstanding, there appears to be considerable variation in the frequency of developmental dislocation/dysplasia of the hip between geographical areas and between racial/ethnic groups. Using data collected before screening for developmental dislocation/dysplasia of the hip at birth became widespread, in most North American and Western European countries and in Australia, New Zealand, and Israel, prevalence rates of around 1 per 1000 to 10 per 1000 births were found. In the Navajo, Apache, and Cree-Ojibwa of North America, the Lapps, and in Hungary, northern Italy, Brittany, and the Faroe Islands, rates from 10 per 1000 to 100 per 1000 births were reported. Conversely, developmental dislocation/dysplasia of the hip was rare in black people in South Africa, the West Indies, and Uganda, and among Chinese living in Hong Kong (reviewed in Kelsey 1982). It must be kept in mind, however, that the nature of the neonatal examination of the infant, the experience of the examiner, the timing of the examination, and the criteria for developmental dislocation/dysplasia of the hip could all affect these figures.
Females are affected about four to six times more frequently than males, possibly because female infants are more susceptible to capsular and ligamentous laxity from the hormones that cause maternal pelvic relaxation during labour (Gunther et al. 1993). In the United States, white people are affected more often than black people. In most geographical areas, an excess of cases is diagnosed in children born in late autumn and winter relative to summer (Robinson 1968). First-born children are at higher risk than later children. The ligaments and other tissues in and around the maternal uterus have already been stretched during previous pregnancies, thus allowing more fetal movement in pregnancies after the first (Gunther et al. 1993). In unilateral cases, the left hip is affected more frequently than the right, probably because the left hip is positioned against the maternal spine more often than the right hip (Mooney and Emans 1995).
Familial aggregation of developmental dislocation/dysplasia of the hip occurs, and both hereditary and environmental factors are believed to contribute to the familial excess (Record and Edwards 1958; Gunther et al. 1993). For a family with one affected parent and one affected child, the risk of an unstable hip in a subsequent child is about 36 per cent (Wynne-Davies 1970). Infants with developmental dislocation/dysplasia of the hip are considerably more likely to have been born by breech delivery than other infants (Robinson 1968; Cyvin 1977; Gunther et al. 1993). Possible reasons for the association with breech delivery are that breech position in utero limits hip motion and elongates the ligament of the hip joint capsule by persistent upward pressure of the greater trochanter (Jones 1965). Children born by caesarean section are also at elevated risk (Gunther et al. 1993). Infants with developmental dislocation/dysplasia of the hip tend to have had longer gestations than other infants (Cyvin 1977).
No means of primary prevention of developmental dislocation/dysplasia of the hip are known. However, early detection and treatment of hips with developmental dislocation/dysplasia of the hip is highly important, since without prompt treatment the affected leg may be shorter, the child may limp, surgery may be required, and osteoarthritis of the hip is likely to occur in young adulthood. Accordingly, screening by examining newborn infants for developmental dislocation/dysplasia of the hip is now accepted as a routine procedure.
Two screening tests have generally been used: the Ortolani test and the Barlow test. In the Ortolani test, the hip is placed in flexion and is gently adducted and then abducted. The Ortolani test is considered positive if a palpable jerk and audible click are heard as the head of the femur returns to the acetabulum. Some physicians consider just an audible click to be a positive test. The Barlow test involves exerting gentle downward pressure over the lesser trochanter with the hip in flexion and adduction; an unstable hip will shift from the acetabulum and a sensation similar to the Ortolani sign is produced. When the leg is allowed to abduct, the hip is reduced.
Because many hips noted to be unstable at birth soon become stable spontaneously, these tests are often repeated at around 3 weeks. Infants who are positive by either the Ortolani or Barlow test are generally treated with braces, splints, or harness for 2 to 4 months. Routine checks of the hips of these infants should be performed until they are walking well. If diagnosis is delayed until after the neonatal period, surgery is usually required and the prognosis is poorer.
Although screening for developmental dislocation/dysplasia of the hip by the Ortolani or Barlow test is routine in many localities, the effectiveness of these tests has been questioned (Leck 1986; Knox et al. 1987). It appears that the frequency of developmental dislocation/dysplasia of the hip requiring prolonged treatment is no lower now than before screening became widespread. Both the sensitivity and specificity of the screening tests are poor. One study (Knox et al. 1987) reported that only one-third of true cases were detected, and that the ratio of false positives to true positives was 10 to 1. No consensus has emerged regarding indications for treatment, the timing of treatment, or the type of splint to be used. It is possible that the screening procedures themselves may bring about hip dislocation (Moore 1989; Jones 1991). These screening tests require experienced examiners (Macnicol 1990), but inexperienced examiners are often used. A better understanding of which hips will spontaneously stabilize would allow more informed decisions to be made about which hips need immediate treatment. There is disagreement about the significance of a soft audible or palpable click without evidence of abnormal movement between the femoral head and acetabulum, and hence there is disagreement about how such infants should be monitored (Fulton and Barer 1984). More studies addressing these issues are needed.
Recently, ultrasound, in which a defined image of the bony and cartilaginous hip can be examined, has become widely available for screening for developmental dislocation/dysplasia of the hip. However, its use as a routine screening test has not been found to be cost-effective, and even its use in high-risk infants is controversial (Hernandez et al. 1994; Rosendahl et al. 1994; Geitung et al. 1996). Some of the problems are that ultrasound is expensive, developmental dislocation/dysplasia of the hip is relatively infrequent, many hips classified as positive on ultrasound develop normally, and cases that develop after the neonatal period will not be detected. It is unclear how an infant with a normal clinical examination but with abnormal ultrasonogram should be treated. Better training is needed to improve the scans and their interpretation (Rosendahl et al. 1995). At present, the American Academy of Pediatrics does not have a policy on developmental dislocation/dysplasia of the hip screening.
The impact of musculoskeletal diseases, though substantial now, will probably become much greater over the next several decades. Musculoskeletal diseases are common at all ages, but their greatest impact is on the elderly, among whom they are often associated with severely compromised quality of life and high costs for medical care. In industrialized countries, the ageing of the population is well recognized. In developing countries, the number of people aged 65 years or older is expected to increase by 200 to 400 per cent between 1990 and 2025. Thus, musculoskeletal diseases will account for a significant portion of health care costs in countries throughout the world in the twenty-first century. Identifying feasible prevention strategies and affordable methods of treatment for such ubiquitous diseases of the elderly as osteoarthritis and osteoporosis is thus a high global public health priority. Although some progress has been made in identifying methods of prevention and treatment for osteoporosis through lifestyle changes and pharmaceutical agents, osteoporosis remains a serious problem for many older people, and still better methods need to be developed. Even less is known about means of preventing the development and progression of osteoarthritis, and much more work in this area is clearly needed.
Among people in their working years, osteoarthritis and neck and low back pain are major causes of disability. Some potentially modifiable risk factors have been well established, including injury to joints, cumulative trauma to joints, heavy lifting, and exposure to vibration in the back and neck. However, to date, preventive programmes have done little to reduce the frequency and impact of osteoarthritis and neck and low back pain. For rheumatoid arthritis, another relatively common disease of adults, almost nothing is known about means of primary prevention except for possible risk reduction from the use of oral contraceptives.
For the major musculoskeletal disorders of adults, including osteoarthritis, osteoporosis, low back and neck pain, and rheumatoid arthritis, much more needs to be learned about how to limit the pain and disability among those who are affected. Once people develop osteoarthritis, osteoporosis, rheumatoid arthritis, and, in some instances, back or neck pain, they are likely to be affected for the rest of their lives. Thus, helping people learn how best to cope with these conditions is essential.
Among infants, children, and adolescents, early detection and prompt treatment would appear to offer the best means of reducing the impact of developmental dislocation of the hip, scoliosis, and slipped epiphysis, but many questions have been raised about the effectiveness of existing screening programmes for developmental dislocation of the hip and scoliosis. Prevention of adolescent obesity and of childhood accidents should reduce the frequency of slipped epiphysis and fractures, respectively, if it were possible to modify the lifestyles of children in these respects.
Finally, musculoskeletal disorders are often not given high priority in developing countries, despite the large amount of disability that they cause and their high monetary cost to the community. On the basis of current knowledge, ergonomic improvements and better training in the workplace could be instituted now in many regions. In addition, more information on the most important musculoskeletal problems and risk factors in specific geographical areas is needed if limited funds are to be used wisely. To accomplish this, special surveys must be undertaken. Clinics will need to be established, professional staff and patients educated, and community participation encouraged. Collaboration between epidemiologists in developing and industrialized countries will often be needed to make such surveys successful (Adebajo 1990).
In summary, much remains to be learned about the epidemiology and prevention of this important group of diseases. Better methods of primary, secondary, and tertiary prevention would improve the quality of life of people worldwide and would contribute substantially to increased economic productivity and a reduction in health care costs.
*A small portion of the material presented in this chapter has been adapted from Kelsey (1998).
Acheson, R.M. (1966). Maturation of the skeleton. In Human development (ed. F. Faulkner), pp. 465–500. Saunders, Philadelphia.
Adebajo, A.O. (1990). Rheumatology in the Third World. Annals of the Rheumatic Diseases, 49, 813–16.
Ahasan, M.R., Mohiuddin, G., Vayrynen, S., Ironkannas, H., and Quddus, R. (1999). Work-related problems in metal handling tasks in Bangladesh: obstacles to the development of safety and health measures. Ergonomics, 42, 385–96.
Alarcón, G.S. (1995). Epidemiology of rheumatoid arthritis. Rheumatic Diseases Clinics of North America, 21, 589–604.
Alexander, M.A. and Season, E.H. (1978). Idiopathic scoliosis: an electromyographic study. Archives of Physical Medicine and Rehabilitation, 59, 314–15.
Anderson, J.J. and Felson, D.T. (1988). Factors associated with osteoarthritis of the knee in the first National Health and Nutrition Examination Survey (HANES I). American Journal of Epidemiology, 128, 179–89.
Andersson, G.B.J. (1981). Epidemiologic aspects of low-back pain in industry. Spine, 6, 53–60.
Anonymous (1997). Who are candidates for prevention and treatment for osteoporosis? Osteoporosis International, 7, 1–6.
Arnett, F.C., Edworthy, S.M., Bloch, D.A., et al. (1988). The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis and Rheumatism, 31, 315–24.
Badley, E.M. (1995). The economic burden of musculoskeletal disorders in Canada is similar to that for cancer, and may be higher. Journal of Rheumatology, 22, 204–6.
Badley, E.M., Rasooly, I., and Webster, G.K. (1994). Relative importance of musculoskeletal disorders as a cause of chronic health problems, disability, and health care utilization: findings from the 1990 Ontario Health Survey. Journal of Rheumatology, 21, 505–14.
Beasley, R.P., Wilkens, R.F., and Bennett, P.H. (1973). High prevalence of rheumatoid arthritis in Yakima Indians. Arthritis and Rheumatism, 16, 443–7.
Beasley, R.P., Bennett, P.H., and Lin, C.C. (1983). Low prevalence of rheumatoid arthritis in Chinese: Prevalence survey in a rural community. Journal of Rheumatology, 10 (Supplement), 11–15.
Biering-Sorenson, F. (1982). Low back trouble in a general population of 30-, 40-, 50-, and 60-year old men and women. Study design, representativeness, and basic results. Danish Medical Bulletin, 29, 289–99.
Bleasel, J.F. and Moskowitz, R.W. (1995). Osteoarthritis: etiology and specific therapy revisited. In Osteoarthritic disorders (ed. K. Kuettner and V. Goldberg), pp. 25–34. AAOS, Rosemont, IL.
Bonjour, J-P. and Rizzoli, R. (1996). Bone acquisition in adolescence. In Osteoporosis (ed. R. Marcus, D. Feldman, and J. Kelsey), pp. 465–76. Academic Press, San Diego, CA.
Bovim, G., Schrader, H., and Sand, T. (1994). Neck pain in the general population. Spine, 19, 1307–9.
Bowling, A. (1996). The effects of illness on quality of life: findings from a survey of households in Great Britain. Journal of Epidemiology and Community Health, 50, 149–55.
Brage, S. and Bjerkedal, T. (1996). Musculoskeletal pain and smoking in Norway. Journal of Epidemiology and Community Health, 50, 166–9.
Brooks, H. L., Azen, S. D., Gerberg, E. Brooks, R., and Chen, L. (1975). Scoliosis: a prospective epidemiological study. Journal of Bone and Joint Surgery, American Volume, 57A, 968–72.
Buckwalter, J.A., Lane, N.E., and Gordon, S.L. (1995). Exercise as a cause of osteoarthritis. In Osteoarthritic disorders (ed. K. Kuettner and V. Goldberg), pp. 405–17. AAOS, Rosemont, IL.
Bunnell, W.P. (1984). An objective criterion for scoliosis screening. Spine, 18, 1572–80.
Butler, W.J., Hawthorne, V.M., Mikkelsen, W.M., Carman, W.J., Bouthillier, D.L., and Lamphiear, D.E. (1988). Prevalence of radiologically defined osteoarthritis in the finger and wrist joints of adult residents of Tecumseh, Michigan, 1962–65. Journal of Clinical Epidemiology, 41, 467–73.
Carman, W. (1989). Factors associated with pain and osteoarthritis in the Tecumseh Community Health Study. Seminars in Arthritis and Rheumatism, 18, 1–4.
Carman, W.J., Sowers, M.F., Hawthorne, V.M., and Weissfeld, L.A. (1994). Obesity as a risk factor for osteoarthritis of the hand and wrist. A prospective study. American Journal of Epidemiology, 139, 119–29.
Cats-Baril, W.L. and Frymoyer, J.W. (1991). Identifying patients at risk of becoming disabled because of low-back pain. The Vermont Rehabilitation Engineering Center predictive model. Spine, 16, 605–7.
Cauley, J.A., Salamone, L.M., and Lucas, F.L. (1996). Postmenopausal endogenous and exogenous hormones, degree of obesity, thiazide diuretics, and risk of osteoporosis. In Osteoporosis (ed. R. Marcus, D. Feldman, and J. Kelsey), pp. 551–76. Academic Press, San Diego, CA.
Chapay, M.C., Arlot, M.E., Duboeuf, F., et al. (1992). Vitamin D3 and calcium to prevent hip fractures in elderly women. New England Journal of Medicine, 327, 1637–42.
Chavalitsakulchai, P. and Shahnavaz, H. (1993). Musculoskeletal disorders of female workers and ergonomics problems in five different industries of a developing country. Journal of Human Ergology (Tokyo), 22, 29–43.
Cheng, J.C.Y., Ng, B.K.W., Ying, S.Y., and Lam, P.K.W. (1999). A 10-year study of the changes in the pattern and treatment of 6,493 fractures. Journal of Pediatric Orthopaedics, 19, 344–50.
Christiansen C., Christensen, M.S., and Transbol, I. (1981). Bone mass in postmenopausal women after withdrawal of oestrogen/gestagen replacement therapy. Lancet, i, 459–61.
Cicuttini, F.M. and Spector, T.D. (1996). Genetics of osteoarthritis. Annals of the Rheumatic Diseases, 55, 665–7.
Côté, P., Cassidy, J.D., and Carroll, L. (1998a). The Saskatchewan Health and Back Pain Survey. The prevalence of neck pain and related disability in Saskatchewan adults. Spine, 23, 1689–98.
Côté, P., Kreitz, B.G., Cassidy, J.D., Dzus, A.K., and Martel, J. (1998b). A study of the diagnostic accuracy and reliability of the scoliometer and Adam’s forward bend test. Spine, 23, 796–803.
Coyte, P.C., Asche, C.V., Croxford, R., and Chan, B. (1998). The economic cost of musculoskeletal disorders in Canada. Arthritis Care and Research, 11, 315–25.
Croft, P. (1996). The occurrence of osteoarthritis outside Europe. Annals of the Rheumatic Diseases, 55, 661–4.
Croft, P., Coggon, D., Cruddas, M., et al. (1992). Osteoarthritis of the hip. An occupational disease in farmers. British Medical Journal, 304, 1269–72.
Cumming, R.G. (1990). Calcium intake and bone mass: a quantitative review of the evidence. Calcified Tissue International, 47, 194–201.
Cumming, R.G., Nevitt, M.C., and Cummings, S.R. (1997). Epidemiology of hip fracture. Epidemiologic Reviews, 19, 244–57.
Cummings, S.R., Black, D.M., Thompson, D.E., et al. (1998). Effect of alendronate on risk of fracture in women with low bone density but without vertebral fractures: results from the Fracture Intervention Trial. Journal of the American Medical Association, 280, 2077–82.
Cyvin, K.B. (1977). Congenital dislocation of the hip joint. Acta Paediatrica Scandinavica, 263, 1–67.
Daltroy L.H., Iversen, M.D, Larson, M.G., et al. (1997). A controlled trial of an educational program to prevent low back injuries. New England Journal of Medicine, 337, 322–8.
Dans, L.F., Tankeh-Torres, S., Amante, C.M., and Penserga, E.G. (1997). The prevalence of rheumatic diseases in a Filipino urban population: a WHO-ILAR COPCORD Study. Journal of Rheumatology, 24, 1814–19.
Dawson-Hughes, B., Dallal, G.E., Krall, E.A., Sadowski, L., Sahyoun, N., and Tannenbaum, S. (1990). A controlled trial of the effect of calcium supplementation on bone density of postmenopausal women. New England Journal of Medicine, 323, 878–83.
Delmas, P.D., Bjarason, N.H., Mitlak, B.H., et al. (1997). Effects of raloxifene on bone mineral density, serum cholesterol concentrations, and uterine endometrium in postmenopausal women. New England Journal of Medicine, 337, 1641–7.
Deyo, R.A. and Diehl, A.K. (1988). Psychosocial predictors of disability in patients with low back pain. Journal of Rheumatology, 15, 1557–64.
Dickson, R.A., Stamper, P., Sharp, A-M., and Harker, P. (1980). School screening for scoliosis: cohort study of clinical course. British Medical Journal, ii, 265–7.
Dougados, M., Gueguen, A., Nguyen, M., et al. (1992). Longitudinal radiologic evaluation of osteoarthritis of the knee. Journal of Rheumatology, 19, 78–83.
Eisenberg, D.M., Kessler, R.C., Foster, C., Norlock, F.E., Calkins, D.R., and Delbanco, T.L. (1993). Unconventional medicine in the United States. New England Journal of Medicine, 328, 246–52.
Eisman, J.A. (1995). Vitamin D receptor gene alleles and osteoporosis: an affirmative view. Journal of Bone and Mineral Research, 10, 1289–93.
Ettinger, B., Pressman, A., Schein, J. Chan, J., Silver, P., and Connolly, N. (1998). Alendronate use among 812 women: prevalence of gastrointestinal complaints, noncompliance with patient instructions, and discontinuation. Journal of Managed Care Pharmacy, 4, 488–92.
Ettinger, B., Black, D.M., Mitlak, B.H., et al. (1999). Reduction of vertebral fracture risk in postmenopausal women with osteoporosis treated with raloxifene. Results from a 3-year randomized clinical trial. Journal of the American Medical Association, 282, 637–45.
Farooqi, A. and Gibson, T. (1998). Prevalence of major rheumatic disorders in the adult population of north Pakistan. British Journal of Rheumatology, 37, 491–5.
Faulkner, K.G., Cummings, S.R., Black, D., Palmero, L., Glüer C.-C, and Genant, H.K. (1993). Simple measurement of femoral geometry predicts hip fracture: the Study of Osteoporotic Fractures. Journal of Bone and Mineral Research, 8, 1211–17.
Felson, D.T., Anderson, J.J., Naimark, A., Walker, A.M., Swift, M., and Meenan, R.F. (1988). Obesity and knee osteoarthritis: the Framingham Study. Annals of Internal Medicine, 109, 18–24.
Felson, D.T., Hannan, M.R., Maimark, A., et al. (1991). Occupational physical demands, knee bending and knee osteoarthritis. Results from the Framingham Study. Journal of Rheumatology, 18, 1587–92.
Felson, D.T., Zhang, Y., Anthony, J.M., Nalmark, A., and Anderson, J.J. (1992). Weight loss reduces the risk of symptomatic knee osteoarthritis in women: the Framingham Study. Annals of Internal Medicine, 116, 535–9.
Felts, W. and Yelin, E. (1989). The economic impact of the rheumatic diseases in the United States. Journal of Rheumatology, 16, 867–84.
Ferraz, M.B. (1995). Tropical rheumatology. Epidemiology and community studies: Latin America. Baillières Clinical Rheumatology, 9, 1–9.
Fish, J.R., DiMonte, P., and Courington, S.M. (1983). Back schools: past, present, and future. Clinical Orthopaedics and Related Research, 179, 18–23.
Francis, M.J., Smith, R., and Sanderson, M.C. (1977). Collagen abnormalities in idiopathic adolescent scoliosis. Calcified Tissue Research, 22 (Supplement), 381–4.
Fulton, M.J. and Barer, M.L. (1984). Screening for congenital dislocation of the hip: an economic appraisal. Canadian Medical Association Journal, 130, 1149–56.
Gabriel, S.E., Crowson, C.S., and O’Fallon, W.M. (1999). The epidemiology of rheumatoid arthritis in Rochester, Minnesota, 1955–1985. Arthritis and Rheumatism, 42, 415–20.
Garraway, W.M., Stauffer, R.N., Kurland, L.T., and O’Fallon, W.M. (1979). Limb fractures in a defined population: frequency and distribution. Mayo Clinic Proceedings, 54, 701–7.
Geitung, J.T., Rosendahl, K., and Sudmann, E. (1996). Cost-effectiveness of ultrasonographic screening for congenital hip dysplasia in newborns. Skeletal Radiology, 25, 251–4.
Goldberg, C.J., Dowling, F.E., Fogarty, E.E., and Moore, D.P. (1995). School scoliosis screening and the United States Preventive Services Task Force. Spine, 20, 1368–74.
Gregerson, P.K., Silver, J., and Winchester, R.J. (1987). The shared epitope hypothesis. An approach to understanding the molecular genetics of susceptibility to rheumatoid arthritis. Arthritis and Rheumatism, 30, 1205–13.
Grisso, J.A., Capezuti, E., and Schwartz, A.V. (1996). Falls as risk factors for fractures. In Osteoporosis (ed. R. Marcus, D. Feldman, and J. Kelsey), pp. 599–611. Academic Press, San Diego, CA.
Grossman, T.W., Mazur, J.M., and Cummings, R.J. (1995). An evaluation of the Adams forward bend test and the scoliometer in a scoliosis school screening setting. Journal of Pediatric Orthopaedics, 15, 535–8.
Gunther, A., Smith, S.J., Maynard, P.V., Beaver, M.W., and Chilvers, C.E.D. (1993). A case-control study of congenital hip dislocation. Public Health, 107, 9–18.
Guo, H.-R., Tanaka, S., Halperin, W.E., and Cameron, L.L. (1999). Back pain prevalence in US industry and estimates of lost workdays. American Journal of Public Health, 89, 1029–35.
Hadler, N.M., Gillings, D.B., Imbus, H.R., et al. (1978). Hand structure and function in an industrial setting. Arthritis and Rheumatism, 21, 210–20.
Hadley-Miller, N., Mims, B., and Milewicz, D.M. (1994). The potential role of the elastic fiber system in adolescent idiopathic scoliosis. Journal of Bone and Joint Surgery, American Volume, 76A, 1193–206.
Hales, T.R. and Bernard, B.P. (1996). Epidemiology of work-related musculoskeletal disorders. Orthopedic Clinics of North America, 27, 679–709.
Hameed, K., Gibson, T., Kadir, M., Sultana, S., Fatima, Z., and Syed, A. (1995). The prevalence of rheumatoid arthritis in affluent and poor urban communities of Pakistan. British Journal of Rheumatology, 34, 252–6.
Hamerman, D. (1989). The biology of osteoarthritis. New England Journal of Medicine, 320, 1322–30.
Hans, D., Dargent-Molina, P., Schott, A.M., et al. (1996). Ultrasonographic heel measurements to predict hip fracture in elderly women: the EPIDOS prospective study. Lancet, 348, 511–14.
Hansson, L.I., Hagglund, G., and Ordeberg, G. (1987). Slipped capital femoral epiphysis in southern Sweden, 1910–1982. Acta Orthopaedica Scandinavica, 226, 1–67.
Harvey, J., Lotze, M., Arnett, F.C., et al. (1983). Rheumatoid arthritis in a Chippewa Band: II. Field study with clinical, serologic and HLA-D correlations. Journal of Rheumatology, 10, 28–32.
Hayes, W.C., Myers, E.R., Morris, J.N., Gerhart, T.N., Yett, H.S., and Lipsitz, L.A. (1993). Impact near the hip dominates fracture risk in elderly nursing home residents who fall. Calcified Tissue International, 52, 192–8.
Hazebroek-Kampschreur, A.A.J.M., Hofman, D., Van Dijk, A.P., and Van Linge, B. (1994). Determinants of trunk abnormalities in adolescence. International Journal of Epidemiology, 23, 1242–7.
Heaney, R.P. (1996). Nutrition and risk for osteoporosis. In Osteoporosis (ed. R. Marcus, D. Feldman, and J. Kelsey), pp. 483–509. Academic Press, San Diego, CA.
Heliövaara, M., Vanharanta, H., Korpi, J., and Troup, J.D.G. (1986). Herniated lumbar disc syndrome and vertebral canals. Spine, 11, 433–5.
Heliövaara, M., Mäkelä, M., Knekt, P., Ollimpivaara, and Aromaa, A. (1991). Determinants of sciatica and low-back pain. Spine, 16, 608–14.
Henrikson, B. (1969). The incidence of slipped capital femoral epiphysis. Acta Orthopaedica Scandinavica, 40, 365–72.
Henssge, J. (1968). Are the signs of denervation of the muscles of the spine primary or secondary findings in cases of scoliosis. Journal of Bone and Joint Surgery, British Volume, 50B, 882.
Hernandez, R.J., Cornell, R.G., and Hensinger, R.N. (1994). Ultrasound diagnosis of neonatal congenital dislocation of the hip. A decision analysis assessment. Journal of Bone and Joint Surgery, British Volume, 76B, 539–43.
Hochberg, M.C. (1990). Contrasting trends in the incidence and prevalence of rheumatoid arthritis in England and Wales, 1970–1982. Seminars in Arthritis and Rheumatism, 19, 294–302.
Holbrook, T.L., Grazier, K.L., Kelsey, J.L., and Stauffer, R.N. (1984). The frequency of occurrence, impact and cost of musculoskeletal conditions in the United States. AAOS, Rosemont, IL.
Holderbaum, D., Haqqi, T.M., and Moskowitz, R.W. (1999). Genetics and osteoarthritis. Arthritis and Rheumatism, 42, 397–405.
Hosking, E., Chilvers, C.E.D., Christiansen, C., et al. (1998). Prevention of bone loss with alendronate in postmenopausal women under 60 years of age. New England Journal of Medicine, 338, 485–92.
Hult, L. (1954). Cervical, dorsal, and lumbar spine syndromes. Acta Orthopaedica Scandinavica, 17 (Supplement), 1–102.
Hutton, C. (1987). Generalised osteoarthritis: an evolutionary problem? Lancet, i, 1463–5.
Jensen, M.C., Brant-Zawadzi, M.N., Obuchowski, N., Modic, M.T., Malkasian, D., and Ross, J.S. (1994). Magnetic resonance imaging of the lumbar spine in people without back pain. New England Journal of Medicine, 331, 69–73.
Jerre, R., Karlsson, J., and Henrikson, B. (1996). The incidence of physiolysis of the hip. A population-based study of 175 patients. Acta Orthopaedica Scandinavica, 67, 53–6.
Jones, D.H. (1965). The early diagnosis of congenital dislocation of the hip joint. British Journal of Clinical Practice, 19, 443–9.
Jones, D.A. (1991). Neonatal hip stability and the Barlow test. A study in stillborn babies. Journal of Bone and Joint Surgery, British Volume, 73B, 216–18.
Keessen, W., Crowe, A., and Hearn, M. (1992). Proprioceptive accuracy in idiopathic scoliosis. Spine, 17, 149–55.
Kellgren, J.H. (1961). Osteoarthrosis in patients and populations. British Medical Journal, 7, 1–6.
Kellgren, J.H. and Lawrence, J.S. (1963). Epidemiology of chronic rheumatism. F.A. Davis, Philadelphia, PA.
Kelsey, J.L. (1971). Incidence and distribution of slipped capital femoral epiphysis in Connecticut. Journal of Chronic Diseases, 23, 567–87.
Kelsey, J.L. (1982). Epidemiology of musculoskeletal disorders. Oxford University Press, New York.
Kelsey, J.L. (1998). Musculoskeletal diseases. In Maxcy–Rosenau–Last public health and preventive medicine (ed. R.B. Wallace) (14th edn), pp. 1005–17. Appleton and Lange, Stamford, CT.
Kelsey, J.L. and Hochberg, M.C. (1988). Epidemiology of chronic musculoskeletal disorders. Annual Reviews of Public Health, 9, 379–401.
Kelsey, J.L., Keggi, K.J., and Southwick, W.O. (1970). The incidence and distribution of slipped capital femoral epiphysis in Connecticut and Southwestern United States. Journal of Bone and Joint Surgery, American Volume, 52A, 1203–16.
Kelsey, J.L., Acheson, R.M., and Keggi, K.J. (1972). The body builds of patients with slipped capital femoral epiphysis. American Journal of Diseases of Children, 124, 276–81.
Kelsey, J.L., Githens, P.B., Walter, S.D., et al. (1984a). An epidemiologic study of acute prolapsed cervical intervertebral disc. Journal of Bone and Joint Surgery, American Volume, 66A, 907–14.
Kelsey, J.L., Githens, P.B., White A.A., III, et al. (1984b). An epidemiological study of lifting and twisting on the job and risk for acute prolapsed lumbar intervertebral disc. Journal of Orthopaedic Research, 2, 61–6.
Kelsey, J.L., Browner, W.S., Seeley, D.G., Nevitt, M.C., and Cummings, S.R. (1992). Risk factors for fracture of the distal forearm and proximal humerus. American Journal of Epidemiology, 135, 473–89.
Keyserling W.M., Herrin, G.D., and Chaffin, D.B. (1980). Establishing an industrial strength testing program. American Industrial Hygiene Association Journal, 41, 730–6.
Knox, E.G., Armstrong, E.H., and Lancashire, R.J. (1987). Effectiveness of screening for congenital dislocation of the hip. Journal of Epidemiology and Community Health, 41, 283–9.
Krause, N., Ragland, D.R., Greiner, B.A., Fisher, J.M., Holman, B.L., and Selvin, S. (1997). Physical workload and ergonomic factors associated with prevalence of back and neck pain in urban transit operators. Spine, 22, 2117–26.
Krause, N., Ragland, D.R., Fisher, J.M., and Syme, L. (1998). Psychosocial job factors, physical workload, and incidence of work-related spinal injury: a 5-year prospective study of urban transit operators. Spine, 23, 2507–16.
Lachance, L., Sowers, M.F., Jamadar, D., Jannausch, M., Hochberg, M.C., and Crutchfield, M. (2001). The experience of pain and emergent osteoarthritis of the knee. In press.
Lagerström M., Hansson, T., and Hagberg, M. (1998). Work-related low-back problems in nursing. Scandinavian Journal of Work, Environment and Health, 24, 449–64.
Landin, L.A. (1983). Fracture patterns in children. Analysis of 8,682 fractures with special reference to incidence, etiology and secular changes in a Swedish urban population 1950–1979. Acta Orthopaedica Scandinavica, 202 (Supplement), 1–109.
Lankhurst, G.J., Van de Stadt, R.J., Vogelaar, T.W., Van de Korst, J.K., and Prevo, A.J.H. (1983). The effect of the Swedish back school on chronic idiopathic low back pain. Scandinavian Journal of Rehabilitation Medicine, 15, 141–5.
Lau, E.M.C., Symmons, D.P.M., and Croft, P. (1996). The epidemiology of hip osteoarthritis and rheumatoid arthritis in the Orient. Clinical Orthopaedics and Related Research, 323, 81–90.
Lauritzen, J.B., Petersen, M.M., and Lund, B. (1993). Effect of external hip protectors on hip fracture. Lancet, 341, 11–13.
Lawrence J.S. and Sebo, M. (1980). The geography of osteoarthritis. In The aetiopathogenesis of osteoarthrosis (ed. G. Nuki), pp. 155–83. Pitman, Kent.
Lawrence, R.C., Hochberg, M.C., Kelsey, J.L., et al. (1989). Estimates of the prevalence of selected arthritic and musculoskeletal diseases in the United States. Journal of Rheumatology, 16, 427–41.
Lawrence, R.C., Helmick, C.G., Arnett, F.C., et al. (1998). Estimates of the prevalence of arthritis and selected musculoskeletal disorders in the United States. Arthritis and Rheumatism, 41, 778–99.
Leck, I. (1986). An epidemiological assessment of neonatal screening for dislocation of the hip. Journal of the Royal College of Physicians of London, 20, 56–62.
Leclerc, A., Niedhammer, I., Landre, M.-F., Ozguler, A., Etore, P., and Pietri-Taleb, F. (1999). One-year predictive factors for various aspects of neck disorders. Spine, 24, 1455–62.
Linos, A., Worthington, J.W., O’Fallon, W.M., and Kurland, L.T. (1980). The epidemiology of rheumatoid arthritis in Rochester, Minnesota: a study of incidence, prevalence and mortality. American Journal of Epidemiology, 111, 87–98.
Loder, R.T. (1996a). The demographics of slipped capital femoral epiphysis. An international multicenter study. Clinical Orthopaedics and Related Research, 322, 8–27.
Loder, R.T. (1996b). A worldwide study of the seasonal variation of slipped capital femoral epiphysis. Clinical Orthopaedics and Related Research, 322, 28–36.
Lonstein, J.R. (1988). Natural history and school screening for scoliosis. Orthopedic Clinics of North American, 19, 227–37.
Lyons, R.A., Delahunty, A.M., Kraus, D., et al. (1999). Children’s fractures: a population based study. Injury Prevention, 5, 129–32.
Macnicol, M.F. (1990). Results of a 25-year screening programme for neonatal hip instability. Journal of Bone and Joint Surgery, British Volume, 72B, 1057–60.
Maggi, S., Kelsey, J.L., Litvak, J., and Heyse, S.P. (1991). Incidence of hip fractures in the elderly: a cross-national analysis. Osteoporosis International, 1, 232–41.
Magnusson, M.L., Pope, M.H., Wilder, D.G., and Areskoug, B. (1996). Are occupational drivers at an increased risk for developing musculoskeletal disorders? Spine, 21, 710–17.
Malmivaara, A., Häkkinen, U., Aro, T., et al. (1995). The treatment of acute low back pain—bed rest, exercises, or ordinary activity? New England Journal of Medicine, 332, 351–5.
Manninen, P., Riihimaki, H., Heliövaara, M., and Mäkelä, P. (1996). Overweight, gender and knee osteoarthritis. International Journal of Obesity, 20, 595–7.
Melton, L.J., III (1995). How many women have osteoporosis now? Journal of Bone and Mineral Research, 10, 175–7.
Mooney, J.F. and Emans, J.B. (1995). Developmental dislocation of the hip: a clinical overview. Pediatrics in Review, 16, 299–303.
Moore, F.H. (1989). Examining infants’ hips – can it do harm? Journal of Bone and Joint Surgery, British Volume, 71B, 4–5.
Morais, T., Bernier, M., and Turcotte, F. (1985). Age- and sex-specific prevalence of scoliosis and the value of school screening programs. American Journal of Public Health, 75, 1377–80.
Morrison, N.A., Yeoman, R., Kelly, P.J., and Eisman, J.A. (1992). Contribution of trans-acting factor alleles to normal physiological variability: vitamin D receptor gene polymorphisms and circulating osteocalcin. Proceedings of the National Academy of Sciences, 89, 6665–9.
Morscher, E. (1968). Strength and morphology of growth cartilage under hormonal influence of puberty. Reconstructive Surgery and Traumatology, 10, 3–104.
Müller, C.F., Monrad, T., Biering-Sorensen, F., Darre, E., Deis, A., and Kryger, P. (1999). The influence of previous low back trouble, general health, and working conditions on future sick-listing because of low back trouble. Spine, 24, 1562–70.
Nachemson, A. (1983). Work for all: those with low back pain as well. Clinical Orthopaedics and Related Research, 179, 77–85.
Nachemson, A. and Sahlshand, T. (1977). Etiologic factors in adolescent idiopathic scoliosis. Spine, 2, 176–84.
Nevitt, M.C. and Cummings, S.R. (1993). Type of fall and risk of hip and wrist fractures: the Study of Osteoporotic Fractures. Journal of the American Geriatrics Society, 41, 1226–34.
Nissinen, M., Heliövaara, M., Seitsamo, J., and Poussa, M. (1993a). Trunk asymmetry, posture, growth, and risk of scoliosis. A three-year follow-up of Finnish prepubertal school children. Spine, 18, 8–13.
Nissinen, M., Heliövaara, M., Ylikoski, M., and Poussa, M. (1993b). Trunk asymmetry and screening for scoliosis: a longitudinal cohort study of pubertal schoolchildren. Acta Paediatrica Scandinavica, 82, 77–82.
Ollier, W.E.R. and MacGregor, A. (1995). Genetic epidemiology of rheumatoid disease. British Medical Bulletin, 51, 267–85.
Papageorgiou A.C., Croft, P.R., Ferry, S., Jayson, M.I.V., and Silman, A.J. (1995). Estimating the prevalence of low back pain in the general population. Evidence from the South Manchester Back Pain Survey. Spine, 20, 1889–94.
Papageorgiou, A.C., Croft, P.R., Thomas, E., Ferry, S., Jayson, M.I.V., and Silman, A.J. (1996). Influence of previous pain experience on the episode incidence of low back pain: results from the South Manchester Back Pain Study. Pain, 66, 181–5.
Papageorgiou, A.C., Macfarlane, G.J., Thomas, E., Croft, P.R., Jayson, M.I.V., and Silman, A.J. (1997). Psychosocial factors in the workplace—do they predict new episodes of low back pain? Evidence from the South Manchester Back Pain Study. Spine, 22, 1137–42.
Peacock, M. (1995). Vitamin D receptor gene alleles and osteoporosis: a contrasting view. Journal of Bone and Mineral Research, 10, 1294–7.
Peacock, M., Turner, C.H., Liu, G. Manatunga, A.K., Timmerman, L., and Johnson, C.C., Jr (1995). Better discrimination of hip fracture using bone density, geometry and architecture. Osteoporosis International, 5, 167–73.
Pope, M.H., Magnusson, M., and Wilder, D.G. (1998). Low back pain and whole body vibration. Clinical Orthopaedics and Related Research, 354, 241–8.
Praemer, A., Furner, S., and Rice, D.P. (1999). Musculoskeletal conditions in the United States. AAOS, Rosemont, IL.
Radin, E.L., Paul, I.L., and Rose, R.M. (1972). Role of mechanical factors in pathogenesis of primary osteoarthritis. Lancet, i, 519–22.
Record, R.C. and Edwards, J.H. (1958). Environmental influences related to the aetiology of congenital dislocation of the hip. British Journal of Preventive and Social Medicine, 12, 8–22.
Rennie, A.M. (1967). Familial slipped capital femoral epiphysis. Journal of Bone and Joint Surgery, British Volume, 49B, 535–9.
Roach, K.E., Persky, V., Miles, T., and Budiman-Mak, E. (1994). Biomechanical aspects of occupation and osteoarthritis of the hip. A case-control study. Journal of Rheumatology, 21, 2334–40.
Robinson, G.W. (1968). Birth characteristics of children with congenital dislocation of the hip. American Journal of Epidemiology, 87, 275–84.
Rosendahl, K., Markestad, T., and Lie, R.T. (1994). Ultrasound screening for developmental dysplasia of the hip in the neonate: the effect on treatment rate and prevalence of late cases. Pediatrics, 94, 47–52.
Rosendahl, K., Aslaksen, A., Lie, R.T., and Markestad. T. (1995). Reliability of ultrasound in the early diagnosis of developmental dysplasia of the hip. Pediatric Radiology, 25, 219–24.
Sambrook, P.N., Kelly, P.J., White, C.P., Morrison, N.A., and Eisman, J.A. (1996). Genetic determinants of bone mass. In Osteoporosis (ed. R. Marcus, D. Feldman, and J. Kelsey), pp. 477–82. Academic Press, San Diego, CA.
Schipplein, O.D. and Andriacchi, T.P. (1991). Interaction between active and passive knee stabilizers during level walking. Journal of Orthopaedic Research, 9, 113–19.
Schwartz, A.V. (1999). Epidemiologic aspects of osteoporotic fractures and falls. Ph.D. Dissertation, Stanford University.
Schwartz, A.V., Kelsey, J.L., Maggi, S., et al. (1999). International variation in the incidence of hip fractures: cross-national project on osteoporosis for the World Health Organization Program for Research on Aging. Osteoporosis International, 9, 242–53.
Scrimgeour, E.M. (1992). Prevention of fracture of the neck of the femur: evidence from developing countries of the relative unimportance of osteoporosis. Australia and New Zealand Journal of Medicine, 22, 85–6.
Seeman, E. (1996). The effects of tobacco and alcohol use on bone. In Osteoporosis (ed. R. Marcus, D. Feldman, and J. Kelsey), pp. 577–97. Academic Press, San Diego, CA.
Shands, A.R. and Eisberg, H.B. (1955). The incidence of scoliosis in the state of Delaware. Journal of Bone and Joint Surgery, American Volume, 37A, 1243–9.
Sharrard, W.J.W. (1993). Paediatric orthopaedics and fractures, Vol. 1. Blackwell Scientific, Oxford.
Shelerud, R. (1998). Epidemiology of occupational low back pain. Occupational Medicine, 13, 1–22.
Silman, A.J. and Hochberg, M.C. (1993). Epidemiology of the rheumatic diseases. Oxford University Press.
Singh, Y.M., Nagrath, A.R., and Maini, P.S. (1970). Changes in trabecular pattern of the upper end of the femur as an index of osteoporosis. Journal of Bone and Joint Surgery, American Volume, 52A, 457–67.
Skaggs, D.L. and Bassett, G.S. (1996). Adolescent idiopathic scoliosis: an update. American Family Physician, 53, 2327–34.
Slemenda, C.W., Christian, J.C., Williams, C.J., Norton, J.A., and Johnston, C.C., Jr (1991). Genetic determinants of bone mass in adult women: a reevaluation of the twin model and potential importance of gene interaction on heritability estimates. Journal of Bone and Mineral Research, 6, 561–7.
Slemenda, C.W., Johnston, C.C., and Hiu, S.L. (1996). Assessing fracture risk. In Osteoporosis (ed. R. Marcus, D. Feldman, and J. Kelsey), pp. 623–33. Academic Press, San Diego, CA.
Snook, S.H. (1982). Low back pain in industry. In American Academy of Orthopaedic Surgeons Symposium on Low Back Pain (ed. A.A. White III and S.L. Gordon), pp. 23–38. C.V. Mosby, St Louis, MO.
Snook, S.H. (1988). Approaches to the control of back pain in industry: job design, job placement, and education/training. Occupational Medicine, 3, 45–59.
Snow, C.M., Shaw, J.M., and Matkin, C.C. (1996). Physical activity and risk for osteoporosis. In Osteoporosis (ed. R. Marcus, D. Feldman, and J. Kelsey), pp. 511–28. Academic Press, San Diego, CA.
Sorenson, K.H. (1968). Slipped upper femoral epiphysis. Acta Orthopaedica Scandinavica, 39, 499–517.
Sowers, M.F. (1996). Premenopausal reproductive and hormonal characteristics and the risk for osteoporosis. In Osteoporosis (ed. R. Marcus, D. Feldman, and J. Kelsey), pp. 529–49. Academic Press, San Diego, CA.
Sowers, M.F., Boehnke, M., Jannausch, M.L., Crutchfield, M., Corton, G., and Burns, T.L. (1992). Familiality and partitioning of variability of femoral neck bone mineral density of women of childbearing age. Calcified Tissue International, 50, 110–14.
Sowers, M.F., Hochberg, M., Crabbe, J.P., Muhich, A., Crutchfield, M., and Updike, S. (1996). Association of bone mineral density and sex hormone levels with osteoarthritis of the hand and knee in premenopausal women. American Journal of Epidemiology, 143, 38–47.
Sowers, M.F., Lachance, L., Jamadar, D., Hochberg, M., Crutchfield, M., and Jannausch, M.L. (1999a). The associations of bone mineral density and bone turnover markers with osteoarthritis of the hand and knee in pre- and peri-menopausal women. Arthritis and Rheumatism, 42, 483–8.
Sowers, M.F., Willing, M., Burns, T., et al. (1999b). Genetic markers, bone mineral density, and serum osteocalcin levels. Journal of Bone and Mineral Research, 14, 1411–19.
Sowers, M.F., Lachance, L., Hochberg, M., and Jamadar, D. (2000). Prevalence of radiographically defined osteoarthritis of the hand and knee in a population of pre- and perimenopausal women. Osteoarthritis and Cartilage, 8, 69–77.
Spector, T.D. and Hochberg, M.C. (1990). The protective effect of oral contraceptives on the development of rheumatoid arthritis. An overview of analytic epidemiologic studies with a meta-analysis. Journal of Clinical Epidemiology, 43, 1221–30.
Spotila, L.D., Colige, A., Sereda, L., et al. (1994). Mutation analysis of coding sequences for type I procollagen in individuals with low bone density. Journal of Bone and Mineral Research, 9, 923–32.
Stecher, R.M. (1941). Heberden’s nodes. Heredity in hypertrophic arthritis of finger joints. American Journal of Medical Sciences, 201, 801–9.
Tepper, S. and Hochberg, M.C. (1993). Factors associated with hip osteoarthritis. Data from the First National Health and Nutrition Examination Survey (NHANES-I). American Journal of Epidemiology, 137, 1081–8.
Thompson, M., Anderson, M., and Wood, P.N.H. (1974). Locomotor disability—a study of need in an urban community (abstract). British Journal of Preventive and Social Medicine, 28, 70–1.
US Preventive Services Task Force (1993a). Screening for adolescent idiopathic scoliosis (policy statement). Journal of the American Medical Association, 269, 2664–6.
US Preventive Services Task Force (1993b). Screening for adolescent idiopathic scoliosis. Journal of the American Medical Association, 269, 2667–72.
van Poppel, M.N., Koes, B.W., Deville, W., Smid, T., and Bouter, L.M. (1998). Risk factors for back pain incidence in industry: a prospective study. Pain, 77, 81–6.
van Sasse, J.L.C.M., Vandenbroucke, J.P., van Romunde, L.K.J., and Valkenburg, H.A. (1988). Osteoarthritis and obesity in the general population. A relationship calling for an explanation. Journal of Rheumatology, 15, 1152–8.
Villa, M.L. and Nelson, L. (1996). Race, ethnicity, and osteoporosis. In Osteoporosis (ed. R. Marcus, D. Feldman, and J. Kelsey), pp. 435–47. Academic Press, San Diego, CA.
Vincelette, R., Laurin, C.A., and Levesque, H.P. (1972). The footballer’s ankle and foot. Canadian Medical Association Journal, 107, 872–4.
Walsh, K., Cruddas, M., and Coggon, D. (1992). Low back pain in eight areas of Britain. Journal of Epidemiology and Community Health, 46, 227–30.
Weyand, C., Hicock, K.C., Conn, D.L., and Goronzy, J.J. (1992). The influence of HLA-DR B1 genes on disease severity in rheumatoid arthritis. Annals of Internal Medicine, 117, 802–6.
WHO (1992). Rheumatic diseases. WHO Technical Report Series 816. WHO, Geneva.
WHO (1994). Assessment of fracture risk and its application to screening for postmenopausal women. WHO Technical Report Series 843. WHO, Geneva.
Wickstrom, G.J. and Pentti, J. (1998). Occupational factors affecting sick leave attributed to low-back pain. Scandinavian Journal of Work, Environment and Health, 24, 145–52.
Williams, J.L. (1988). Criteria for screening. Are the effects predictable? Spine, 13, 1178–86.
Willing, M., Sowers, M.F., Aron, D., et al. (1998). Bone mineral density and it change in white women. Estrogen and vitamin D receptor genotypes and their interaction. Journal of Bone and Mineral Research, 13, 695–705.
Willner, S. (1975). A study of height, weight, and menarche in girls with idiopathic structural scoliosis. Acta Orthopaedica Scandinavica, 46, 71–83.
Wingrave, S.J. (1978). Reduction in incidence of rheumatoid arthritis associated with oral contraceptives. Lancet, 1, 569–71.
Worlock, P. and Stower, M. (1986). Fracture patterns in Nottingham children. Journal of Pediatric Orthopedics, 6, 656–60.
Wynne-Davies, R. (1968). Familial (idiopathic) scoliosis. A family survey. Journal of Bone and Joint Surgery, British Volume, 50B, 24–30.
Wynne-Davies, R. (1970). Acetabular dysplasia and familial joint laxity: two etiological factors in congenital dislocation of the hip. Journal of Bone and Joint Surgery, British Volume, 52B, 704–16.
Yu, I.T.S. and Wong, T.W. (1996). Musculoskeletal problems among VDU workers in a Hong Kong bank. Occupational Medicine, 46, 275–80.