To compare the outcomes of care provided by generalists with that provided by specialists, for patients with musculoskeletal and rheumatic conditions.

MEDLINE was searched from 1966 to April 1996; details of the search terms are provided. Additional published and unpublished references were identified by contacting relevant experts and authors of published abstracts, and by reviewing the reference lists of all articles. Only English language articles were considered.

Study designs of evaluations included in the review

Randomised controlled trials (RCTs), prospective cohorts, retrospective cohorts, and physician surveys were included.

Specific interventions included in the review

Treatment of lower-back pain, work-related injury, osteoarthritis, acute arthritis, and rheumatoid arthritis by both generalist and specialists.

Participants included in the review

Lower-back pain: hospital out-patient provider, chiropractor, primary care provider, orthopaedist, generalist, internist, osteopath, family physician and medical physician.

Work-related injury: medical physician and chiropractor.

Osteoarthritis: primary care physician and rheumatologist.

Acute arthritis: nonrheumatologist, rheumatologist and generalist.

Rheumatoid arthritis: intermittent or continuous rheumatological care.

Outcomes assessed in the review

Patient-centred outcomes (clinical outcomes or patient satisfaction with care), resource utilisation (duration of hospitalisation or cost of care) and appropriateness of the process of care.

How were decisions on the relevance of primary studies made?

The authors do not state how the papers were selected for the review, or how many of the authors performed the selection.

The studies were assessed using four main methodological criteria: practitioners (physician training, random assignment to provider type, comparison groups similar with regard to organisation and physical environment); patients (standard criteria for diagnosis, similar diagnoses between providers, similarity between patient groups, differences adjusted for in analyses); outcomes (validated measures, blinded assessment, evidence or consensus-based appropriateness criteria); and analysis (adequate study power). The authors do not state how the papers were assessed for validity, or how many of the authors performed the validity assessment.

The authors do not state how the data were extracted for the review, or how many of the authors performed the data extraction.

How were the studies combined?

A narrative synthesis was undertaken.

How were differences between studies investigated?

The methodological rigour of each of the studies are described in the review, highlighting methodological differences.

Sixteen trials were included (17 studies and 1 personal communication; for 2 trials, 2 studies reported the same trial).

Lower-back pain (over 9,000 patients): 1 RCT, 2 prospective cohort studies and 4 retrospective cohort studies.

Work-related injury: 1 retrospective cohort study (1,890 patients).

Osteoarthritis: 1 physician survey.

Acute arthritis: 3 retrospective cohort studies (136 patients) and 2 physician surveys.

Rheumatoid arthritis: 2 prospective cohort studies (282 patients).

All of the studies exhibit methodological limitations, in terms of study design, cohort selection, outcome definition and sample size calculations.

Lower-back pain: generalist care seems to be as effective and less expensive, but also less satisfying, to patients with lower-back pain. It is unclear whether these results generalise to other regional disorders, such as shoulder or knee pain. Clinical outcomes for lower-back pain seem to be similar across types of providers. Resource utilisation was higher in patients seen by chiropractors and orthopaedists, and satisfaction was highest in patients seen by chiropractors.

Work-related injury: the one study identified found that patients who saw chiropractors had less time off work than those who saw other practitioners.

Osteoarthritis: self-reported process of care revealed differences between rheumatologist and primary care physicians in the use of non-steroidal anti-inflammatory drugs and physical therapies.

Acute arthritis: no clear variation in clinical outcomes was identified, but rheumatologists may give more appropriate care and use less resources than nonrheumatologists or generalists.

Rheumatoid arthritis: specialists would appear to provide better outcomes than generalists. It is unclear whether this was true for other systemic rheumatic disorders.

Costs identified in individual primary studies are reported.

There appears to be a difference between generalist and specialists for a range of outcomes in various musculoskeletal and rheumatic conditions. Studies to date have important methodological limitations that need to be addressed in future research.

This is a clear and thorough review of the literature. However, the search only included English language articles, although it did attempt to identify unpublished research. The methods by which decisions were made on the relevance and quality of primary studies are not reported. It is unclear how generalisable the results of some of the included studies are to the organisational structure of the National Health Service.

American Academy of Orthopedic Surgeons; National Institutes of Health, grant number AR36308.

This additional published commentary may also be of interest. Smith RA. Specialists have better outcomes for rheumatoid disease but not for low-back pain. Evidence-Based Health Policy and Management, 1998;2:41.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.