To examine patterns of ambulatory care and hospital utilization for people with musculoskeletal disorders (MSDs), including arthritis and related conditions, bone and spinal conditions, trauma and related conditions, and unspecified MSDs.
To examine patterns of ambulatory care and hospital utilization for people with musculoskeletal disorders (MSDs), including arthritis and related conditions, bone and spinal conditions, trauma and related conditions, and unspecified MSDs.
Administrative data from the Ontario Health Insurance Plan database for ambulatory care physician visits, the National Ambulatory Care Reporting System database for day (outpatient) surgeries and emergency department visits, and the Discharge Abstract Database for hospital discharges were used to examine health care utilization for MSDs in fiscal year 2006–2007. Person visit rates (number of people with physician visits or hospital encounters per population) were calculated.
Overall, 22.3% of Ontario's population (2.8 million persons) saw a physician for an MSD in ambulatory settings. Person visit rates were highest for arthritis and related conditions (107.7 per 1,000 population), followed by trauma and related conditions (89.6 per 1,000 population), unspecified MSDs (71.0 per 1,000 population), and bone and spinal conditions (62.4 per 1,000 population). The majority of visits were to primary care physicians, with 83.2% of those with visits for all MSDs seeing a primary care physician at least once. Overall, 33.0% of people with a physician visit for an MSD saw a specialist, with orthopedic surgeons being the most commonly consulted type of specialist. In hospital settings, person visit rates for MSDs were highest in the emergency department, followed by day surgeries and inpatient hospitalizations.
The findings of our study highlight the magnitude of health care utilization for MSDs and the central role of primary care physicians in the management of these conditions.
Musculoskeletal disorders (MSDs) are among the most prevalent chronic conditions and a leading cause of global disability and pain (1–4). MSDs refer to a broad range of diseases and disorders of the musculoskeletal system such as arthritis, osteoporosis, spinal disorders, soft tissue disorders, and musculoskeletal trauma and injury (e.g., fractures). The prevalence of MSDs vary depending on the conditions included, with as many as 1 of 2 adults in the US reporting a musculoskeletal condition (when arthritis, chronic joint symptoms, and neck and back pain lasting >3 months are included) (4). In Canada, 36% of the population self-reported a chronic musculoskeletal condition based on data from the Canadian Community Health Survey (2006), which is more restricted in the breadth of musculoskeletal conditions available (Badley EM: unpublished observations). The economic burden of MSDs is substantial, representing ∼2.5% of the gross national product in the US (5). With the projected increase in MSDs due to the aging of the population (6–8), there will be concomitant increasing demands on the health care system. Despite this, the burden of MSDs on the health care system continues to receive inadequate attention (4, 9). Prior research of health care utilization has predominantly focused on specific types of MSDs (e.g., arthritis). However, most health care providers manage a range of MSDs in clinical practice. To date, there are few studies that have comprehensively examined health care utilization for a range of MSDs across service settings and health care providers.
Limited data are available from population surveys and administrative data on health care utilization for MSDs. Previous research using survey data found that people with MSDs were more likely to consult health care providers than people with other chronic conditions (1). Other Canadian research demonstrated that the majority of health care for MSDs was provided in ambulatory care (10, 11), and in the US, MSDs were among the top reasons for physician visits (12). In the 1998–1999 fiscal year, 24% of Canadians age >15 years made ≥1 physician visit for an MSD, excluding musculoskeletal trauma (1, 13). Similar results were found in the province of Ontario in fiscal year 2000–2001 (10, 11). While some studies have included trauma in prevalence estimates of MSDs (4, 14), few studies of health care utilization have included musculoskeletal trauma. Moreover, prior research has provided only a limited breakdown of physician visits by type of physician. It remains unclear as to what extent other types of physicians manage MSDs.
Although previous studies show that utilization rates for MSDs are higher in ambulatory care, the highest expenditures for direct medical costs for MSDs result from hospitalization (13, 15). In a study of health care utilization in the US, ∼9% of hospital stays included a musculoskeletal procedure (16). Other studies have examined inpatient hospitalization rates, primarily with a focus on the rates of surgical procedures (13, 16). Few studies have examined emergency department visits and day (outpatient) surgeries have been largely neglected.
To address the need for a comprehensive examination of health care utilization for all MSDs, we undertook a study of MSD health care utilization across service settings and physician types within a defined population. By including all MSDs, we can examine the total demands on the health care system. This evidence can be used to support health system and policy development, including development of models of health care delivery and health human resources planning. The specific objective of this article is to examine patterns of ambulatory care and hospital utilization for people with MSDs in the province of Ontario, Canada.
In Canada's health care system, provincial health insurance plans provide universal coverage for medically necessary hospital and physician services with no copayments or other patient charges. Unlike other industrialized nations, Canada has no parallel private insurance system that pays for hospital care and physicians' services (17). Access to specialists is by referral from other physicians, usually a primary care physician. The majority of physicians operate on a fee-for-service basis; a claim is submitted to the provincial health insurance plan for each patient encounter.
Ambulatory visits refer to physician visits provided on an outpatient basis. Ambulatory physician visits were captured in the Ontario Health Insurance Plan (OHIP) database, using physician claims with fee codes indicating an ambulatory visit. A visit was defined as one claim per patient per diagnosis per service date. Linkage to Ontario's Registered Persons Database was used to derive age and sex for each person. Data on hospital services in Ontario were available from the Discharge Abstract Database (DAD) and the National Ambulatory Care Reporting System (NACRS) database, which contain demographic, administrative, and clinical data on hospital discharges and day surgeries/emergency department visits, respectively. Cross-sectional data from April 2006 to March 2007 were used. The data represent the population of Ontario (all age groups).
The classification scheme for diagnosis used in the OHIP database consists of a subset of 3-digit codes adapted from the International Classification of Diseases, Ninth Revision (ICD-9). The DAD and the NACRS use the full range of diagnostic codes from the International Classification of Diseases, Tenth Revision (ICD-10). In the OHIP database, there is only one diagnosis code recorded for each visit. The most responsible diagnosis (the diagnosis that was indicated as being the most responsible for the patient's visit or stay in the hospital) was used for the DAD and the NACRS. Diagnostic codes were grouped into condition groups and specific conditions based on clinical and research expertise, previous research (7), and volumes of visits within the constraints of the limited range of codes in the OHIP database and the compatibility of the ICD-9 and ICD-10 coding systems. The major condition groups are arthritis and related conditions (e.g., osteoarthritis [OA], rheumatoid arthritis [RA], and joint derangement), bone and spinal conditions (e.g., spine, bone), trauma and related conditions (e.g., fractures and dislocations, sprains and strains), and unspecified MSDs, which are ill-defined symptoms such as leg pain, joint pain, and undiagnosed conditions. Diagnostic codes for condition groupings and conditions are shown in Supplementary Appendix A (available in the online version of this article at http://www3.interscience.wiley.com/journal/77005015/home).
Physician specialty for each physician visit was defined using the Institute of Clinical Evaluative Sciences Physician Database (IPDB). The IPDB was linked to the OHIP database to identify physician specialty. Physician specialty was classified as primary care physicians and specialists. Specialists were further classified as medical or surgical specialists. Medical specialists include rheumatologists, internal medicine specialists (general internists), and physiatrists. In Canada, general internists are specialists who have hospital-based and referral practices, and do very little primary care. Some general internists further specialize in areas such as arthritis and may practice much like rheumatologists. Surgical specialists include orthopedic surgeons, neurosurgeons, and plastic surgeons. Data for specialists with extremely low volumes of visits for MSDs are not presented.
Person visit rates were defined as the number of persons with ≥1 ambulatory visit per 1,000 population for ambulatory care, and the number of persons with ≥1 hospital encounter (i.e., inpatient hospitalization, emergency department visit, or day surgery) per 100,000 population for hospital services. Person visit rates were calculated for each condition (e.g., OA) and major condition groupings (e.g., arthritis and related conditions) by age and sex. People with visits to physicians for >1 condition were counted in each condition group. Physician visit rates (the number of ambulatory visits per 1,000 population) were also calculated for all MSDs and the major condition groupings. All rates were calculated using census data for the Ontario population (2006).
In ambulatory care, the ratio of women to men making ≥1 visit to a physician and the mean number of visits per person were calculated for each condition. The percentages of individuals with ≥1 physician visit were examined by physician type. Individuals consulting different types of physicians or for multiple conditions were counted for each condition and physician type.
MSDs comprised a large percentage of all ambulatory physician visits in Ontario (27.3%) and a substantial proportion of emergency department visits (8.7% of all visits), day surgeries (4.2% of all day surgeries), and inpatient hospitalizations (5.0% of all hospitalizations) (Figure 1). Arthritis and related conditions was the most common condition group to receive ambulatory care, inpatient care, or day surgery. In contrast, the trauma and related conditions group was more common in the emergency department.
The majority of care for persons with MSDs was delivered in ambulatory settings. In 2006–2007, more than 2.8 million persons in Ontario made ambulatory visits to physicians for MSDs. This number represents 22.3% of the Ontario population (Table 1). The total number of visits to all physicians for MSD was 8.7 million, with a mean of 3.1 visits per person (a visit rate of 686.9 per 1,000 population). The person visit rate for trauma and related conditions was 89.6 per 1,000 population, representing 1.1 million people who made a total of 2.3 million visits (a visit rate of 178.1 per 1,000 population). For arthritis and related conditions, the person visit rate was 107.7 per 1,000 population, corresponding to 1.4 million people who made 2.9 million visits (a visit rate of 225.8 per 1,000 population). Analogous data for bone and spinal conditions were a person visit rate of 62.4 per 1,000 population, with 790,000 people making 1.8 million visits (a visit rate of 144.6 per 1,000 population). Finally, for unspecified MSD, the person visit rate was 71.0 per 1,000 population, with 898,000 people making 1.8 million visits (a visit rate of 138.4 per 1,000 population).
|Condition groups||Age groups, years||Sex||Women: men||No. visits per person, mean|
|Trauma and related conditions||89.6||53.5||85.8||110.9||109.1||91.8||87.3||1.1:1||2.0|
|Strains/sprains of spine||20.1||3.7||20.9||28.9||22.2||22.6||17.5||1.3:1||1.8|
|Arthritis and related conditions||107.7||21.9||67.3||167.0||242.5||122.3||92.7||1.4:1||2.1|
|Soft tissue disorders||9.7||1.9||8.1||15.7||13.8||11.8||7.5||1.6:1||1.6|
|Bone and spinal conditions||62.4||10.0||46.4||95.3||122.2||74.2||50.3||1.5:1||2.3|
Among trauma and related conditions, the most common reason for making at least one visit to a physician was “other sprains and strains.” Among arthritis and related conditions, the person visit rate was highest for OA and synovitis. There were also high person visit rates for spine and unspecified MSDs.
Person visit rates were higher in women than men. The ratio of women to men was 1.1:1 for visits for trauma and related conditions, 1.4:1 for arthritis and related conditions, 1.5:1 for bone and spinal conditions, and 1.3:1 for unspecified MSDs (Table 1). The highest ratio of visits by women to men was 3.8:1 for connective tissue diseases, followed by 2.3:1 for RA.
Overall, 83.2% of people who visited any type of physician for MSDs saw a primary care physician at least once (Table 2). Approximately 78.0% of persons with visits for trauma and related conditions and arthritis and related conditions saw a primary care physician. The proportion was even higher for persons with visits for bone and spinal conditions (86.0%) and unspecified MSDs (88.9%). A higher proportion of people had visits to primary care physicians than to specialists with the exception of connective tissue disorders, ankylosing spondylitis, joint derangement, and fractures and dislocations. In total, 33.0% of people with MSDs saw a specialist at least once, with 16.2% seeing both a primary care physician and a specialist. Compared with the other condition groups, a higher proportion of persons with a visit for arthritis and related conditions saw a specialist (35.1%). Persons with RA, connective tissue disorders, and ankylosing spondylitis were more likely to consult specialists. Persons with unspecified MSDs were least likely to consult a specialist (13.9%). While ∼90% of persons with disorders of the spine saw a primary care physician, only 17.3% saw a specialist.
|Primary care||All specialists||Medical specialists||Surgical specialists|
|All medical specialists||Rheumatologist||General internist||Physiatrist||All surgical specialists||Orthopedic surgeon||Neurosurgeon||Plastic surgeon|
|Trauma and related conditions||78.2||28.9||4.9||0.1||0.7||2.2||24.5||20.5||0.2||2.8|
|Strains/sprains of spine||94.6||6.2||5.8||0.9||2.0||2.9||3.4||1.7||0.6||<0.1|
|Arthritis and related conditions||78.0||35.1||17.2||9.9||1.7||2.1||20.5||16.2||0.3||3.0|
|Soft tissue disorders||80.2||21.8||15.3||9.2||1.3||2.5||6.5||1.3||<0.1||4.1|
|Bone and spinal conditions||86.0||20.0||12.6||3.7||2.8||2.2||8.2||5.8||1.5||0.1|
When examining the types of specialists seen, a higher proportion of people with visits for MSDs saw surgical specialists (21.5%) than medical specialists (14.4%) (Table 2). This was true for trauma and related conditions and arthritis and related conditions overall. Conversely, a greater proportion of persons with bone and spinal conditions and unspecified MSD saw medical specialists, especially for bone disorders (likely related to the management of osteoporosis). Overall, orthopedic surgeons were the most commonly consulted specialist, with 58.1% of persons with physician visits for fractures and dislocations consulting an orthopedic surgeon, 66.8% for joint derangement, and 19.9% for OA. The proportion of people consulting other types of surgeons was relatively small. While only 2.6% of people with physician visits for all MSDs consulted a plastic surgeon, it was more common for people with fractures/dislocations (9.6%). Less than 1% of people with physician visits for all MSDs saw a neurosurgeon; visits were most common for ankylosing spondylitis (2.2%) and joint derangement (3.1%).
Overall, rheumatology was the most commonly consulted group of medical specialists (Table 2). For arthritis and related conditions, 17.2% of people with ambulatory visits went to medical specialists: 9.9% to rheumatologists followed by physiatrists (2.1%) and general internists (1.7%). Persons with inflammatory arthritis visits were more likely to see a medical specialist than a surgical specialist, including 81.1% of persons with visits for connective tissue disorders followed by ankylosing spondylitis (67.0%) and RA (53.4%). Rheumatologists were the most commonly consulted medical specialist for bone and spine conditions and unspecified MSDs, and physiatrists were more commonly seen for trauma and related conditions.
Overall, the person visit rates for all MSDs were higher in emergency departments (3,202.0 per 100,000 population) than day surgeries and inpatient hospitalizations (444.0 and 391.0 per 100,000 population, respectively) (Table 3). Person visit rates for day surgeries and inpatient hospitalizations were highest for persons with arthritis and related conditions and least for unspecified MSDs. OA had the highest rate for inpatient care, while joint derangements were highest in day surgery. In the emergency department, person visit rates for trauma and related conditions were highest at 1,214.0 per 100,000 population, with high rates for other sprains and strains and fractures and dislocations, while the rate for arthritis and related conditions was 1,077.8 per 100,000 population. Person visit rates for disorders of the spine and soft tissue disorders were also among the highest rates.
|Emergency department||Hospital inpatient||Day surgery|
|Trauma and related conditions||1,214.0||97.0||96.0|
|Strains/sprains of spine||64.0||1.0||1.0|
|Arthritis and related conditions||1,077.8||307.8||375.7|
|Soft tissue disorders||654.1||19.0||48.9|
|Bone and spinal conditions||837.4||106.3||100.8|
This study describes health care utilization for MSDs across health care settings and physician types. The findings demonstrate the magnitude of health care utilization by people with MSDs, particularly in ambulatory care, and underscores the importance of primary care management of MSDs. This study is one of the few to examine a comprehensive range of MSDs, including trauma, and to highlight the need for consideration of all MSDs when planning health services.
In general, our findings are consistent with results from other studies. Using physician claims data from Ontario, Canada (2000 and 2001), the person visit rate for MSDs, excluding trauma, was 239.1 per 1,000 population (11). The comparable figure in our study, restricting data to the same age groups, was 251.0 per 1,000 population. The higher rate in our study reflects the addition of musculoskeletal trauma, but is less than might be expected. This may be explained by the fact that people could have visits for more than one diagnosis (the sum of visits for each grouping is higher than the overall total). The person visit rate for arthritis and related conditions was similar, although slightly less (126.2 per 1,000 population) than in 2000–2001 (137.1 per 1,000 population). It is possible that physician visit rates for some conditions are decreasing over time, as related to availability of physicians (18–21) or changes in treatments. A study from The Netherlands showed that rofecoxib withdrawal resulted in a large proportion of patients who discontinued analgesic treatment altogether regardless of the original therapy (22). It is unclear as to what extent changes in the availability of treatments and concerns about medication side effects will affect a patient's decision to seek health care.
Studies from other countries have also shown comparable results. In the US, the number of ambulatory physician visits for arthritis and rheumatic conditions was 234.0 per 1,000 population (23), compared with 225.6 per 1,000 population in our study. In the UK, the annual rate of persons (age ≥15 years) consulting general practices for MSDs was ∼200 persons per 1,000 population, with variation depending on the data source (24). Our findings were similar when limiting our data to these age groups and primary care physician visits. The UK health care system is similar to the Canadian system with universal access to medical services and referral for access to specialists.
In comparison with other chronic diseases, people with MSDs are among the highest users of ambulatory care. In previous research in Ontario, MSDs (excluding trauma) had the second highest person visit rate next to respiratory conditions. In the US, musculoskeletal symptoms are also a leading cause of visits to physicians (25).
Person visit rates were highest in ambulatory care, with 22% of the population seeking care for MSDs. The majority of care was provided by primary care physicians. In a US study, more than 50% of office visits for arthritis and related conditions were to primary care physicians (7). A higher proportion of Canadians may seek consultation with primary care physicians compared with jurisdictions with direct access to specialists. While management of the majority of MSDs in primary care is likely appropriate, studies from Canada, the US, and the UK have reported deficiencies in the primary care management of MSDs, particularly arthritis, including lack of confidence in musculoskeletal examination (26–28), and suboptimum referrals to specialists (29–32). Recent findings from Canada showed that medical students continue to have limited hours of musculoskeletal physical examination training (32). Our findings reinforce the need for education about MSDs in medical schools and for continuing education for practicing physicians. Given the shortages of health human resources, including primary care physicians and specialists in countries such as Canada and the US (18–21, 33–35), there is a need for wider application of initiatives to enhance primary care management of MSDs. New models of care for MSDs, using extended roles for health professionals such as physical therapists and nurse practitioners, have emerged to enhance access and quality of care (36–38).
Orthopedic surgeons were the most commonly consulted specialty, which is similar to findings of other Canadian studies (11). While we anticipated that neurosurgeons might be consulted for disorders of the spine, the proportion was small. Although equivalent data are not available, in the US neurosurgeons performed a higher proportion of spinal surgeries than orthopedic surgeons (39).
Rheumatologists were the most commonly consulted medical specialist, followed by general internists. Approximately 50% of people with physician visits for RA saw a medical specialist. Guidelines for the management of RA recommend early referral to specialists (40). In Canada, patients with RA were more likely to receive disease-modifying antirheumatic drugs, which have been shown to slow disease progression, if they saw a rheumatologist or a general internist (41). An American study suggested that persons whose RA is managed primarily by a rheumatologist had a small advantage over those whose main physician was not a rheumatologist (42). A higher percentage of people with RA in this study consulted a specialist than in previous Canadian studies (11, 43). However, these findings suggest that access to specialist care may remain less than recommended. Other research in Ontario has demonstrated geographic variation in access to rheumatologists (21). Compared with previous Ontario studies (10, 11), the proportion of visits to general internists for arthritis has decreased. This may be due to more general internists being coded under subspecialties in administrative databases. The types of medical specialists presented account for only a portion of people with visits to medical specialists (Table 2). Other specialties on their own accounted for a small proportion of people.
While utilization rates for hospital services were substantially lower than ambulatory care rates, they represent significant health care costs (5, 15). Person visit rates in the emergency department were the highest among hospital services with the highest rates for trauma and related conditions. There may be longer-term health system implications of musculoskeletal trauma, as injuries are risk factors for development of other MSDs (e.g., knee injuries are a known risk factor for OA) (44). Perhaps more surprising was the relatively high person visit rates for spinal conditions. Given the high prevalence of back pain (4), this supports the need for adequate primary care management for MSDs. Unspecified MSDs (ill-defined diagnoses) also contributed to the high rates of emergency department visits. While the emergency department is not considered an ideal setting for managing chronic conditions such as arthritis, our study found that >3% of people with emergency department visits had a most responsible diagnosis of arthritis and related conditions at their visit, which is similar to other research in Ontario (45).
Person visit rates for inpatients and day surgeries were highest for arthritis and related conditions. Other Canadian research has shown the most common surgery for arthritis is total joint replacement (46). The person visit rate was also high for fractures and dislocations. Hip fractures are the most common site of fracture needing surgery, account for the highest proportion of medical costs, and a high proportion of these fractures are attributed to osteoporosis (46–48).
Strengths of this study are the comprehensive study population (with universal health coverage), the inclusive definition of MSDs, and a broad range of physician specialties. The study also has limitations. There may be misclassification of diagnostic codes in administrative databases, including discrepancies between clinical diagnoses and administrative records (49). Estimates using OHIP data may not capture patients seeing physicians remunerated under alternate payment plans, unless there is shadow billing (i.e., billings made so that a record of service is available). While the proportion of non-fee-for-service is rising (50), most physicians are required to submit shadow billings and were included in this analysis. Only one diagnosis can be recorded for each physician visit, and it is possible that a visit including a consultation for MSD may be coded under another diagnosis, underestimating consultations. Similarly, inpatient admissions may be underestimated if there were significant comorbidity or complications associated with the admission that was coded as the most responsible diagnosis. In grouping diagnostic codes into condition groups, we were restricted by the limited range of codes available in the OHIP coding classification system, and how the individual condition codes (e.g., synovitis) were described. The extensive use of “other arthritis” and unspecified MSD codes, and the lack of comparability between the ICD-9–based OHIP codes and the ICD-10 codes for MSD were also challenging. These data sources do not include other nonphysician health services (e.g., physiotherapy), which are important to the management of MSD. Finally, these data are limited to one province. While it may be difficult to generalize findings to other settings, the findings are consistent with studies from the US and the UK, suggesting overall patterns may be similar and applicable to other health care systems. Future work should examine the relationship between utilization and other factors such as availability of health human resources. Rates for ambulatory physician visits in Ontario for arthritis and related conditions are less than other Canadian provinces (11). The reason for this rate discrepancy is unclear, but it may be related to the restricted range of codes available in OHIP. National rates for all MSDs are not available.
In summary, our findings demonstrate that care for MSDs places a significant burden on the health care system. As the prevalence of MSDs increases, there will be an escalating demand for health services. The health care system will need to be adequately prepared to ensure those affected have access to the health care they require.
All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be submitted for publication. Ms MacKay had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study conception and design. MacKay, Canizares, Davis, Badley.
Analysis and interpretation of data. MacKay, Canizares, Davis, Badley.