A survey of physical therapy goals and interventions for patients with back and knee pain.

Key Words: Intervention, Pain, Physical therapy, Survey research.

Physical therapists are increasingly discussing the importance of studying the relationship between diagnoses and treatment.[1-3] Recently, it was shown that diagnostic categories may predict which interventions are used by physical therapists.[1] This finding concerns the relationship between diagnostic categories and interventions. Data on the nature of treatment are part of the report.[1] Rothstein[4] stated that although there is little doubt that outcome studies are needed, investigations are needed to determine what approaches are used in therapy practice. In this report, we attempt to provide such a description of practice for patients with complaints of pain. In order to do so we first formulated a framework to be used in the study of practice.

We believe that, in most cases, treatment starts with an emphasis on creating conditions for physiological recovery, that is, the reduction of impairments (eg, reduction of swelling). This, however, is an arguable point of view. In the course of a treatment, we contend that emphasis is placed on the reduction of disabilities, and ultimately the overall quality of life.[5-9] Consequently, throughout treatment a shift of emphasis can be expected. We studied this shift in emphasis with respect to two main aspects of treatment: (1) the goals and (2) the interventions.

We hypothesized that pain reduction and reduction of swelling should be important goals in the early stages of treatment because they are cardinal signs of the inflammatory stage.[10] Treatment goals that play a major role in the later stages of treatment may be more directly related to reduction of disabilities, increasing muscle $force production, and improving joint stability and function.[11-13]

In pain management, for example, thermal modalities (eg, ultrasound, shortwave diathermy)[10,14-16] and manual therapy[17] may be considered as preparation for exercise.[15,18-20] Application of thermal modalities is supposed to reduce pain and improve the extensibility of collagen.[10] Some therapists use manual therapy to restore function of the joints and the spine and to reduce pain.[1,6,17] We believe that only exercise can result in durable recovery, with minimal possibility of relapse.[7,9,21,22] In our view, attention during this second phase of patient care should focus on patient education (ie, instructions on home exercises and advice on living rules, such as regular walking) to optimize treatment.[23] We hypothesized that treatments often initially consist of the administration of thermal modalities and manual therapy. Massage is used by some therapists in this phase.[24,25] During the course of treatment, however, we believe that the emphasis will be on exercise,[15,18,20,26] which is accompanied by instruction and advice.

In developing a treatment framework, we distinguished subgroups on the basis of the time since the onset of pain. We believe that the treatment framework that we discussed (proceeding from passive modalities to exercise) seems to be appropriate for patients with subacute complaints (ie, complaints of 1-3 months) and not for patients with acute and chronic problems. For patients with long-existing complaints, different treatment goals and interventions usually are advised.[6,27-29] There may be less emphasis, for example, on pain reduction in persons with chronic conditions.[6] In persons with chronic conditions, the physiological basis for pain is usually less clear or not discernable at all.[30] In our view, therefore, treatment aimed at the physiologic basis of pain is less relevant and for persons with long-standing pain is sometimes even advised against.[27-29] We believe that the treatment framework described is also probably less relevant for patients with complaints of recent onset. In most of these patients, there are, in our opinion, no structural changes that could be determined by examination, and it is reasonable to expect less emphasis on the recovery of functional status in general. The two-phase treatment framework that we described seems to be appropriate for patients having their first episode of pain (not to be confused with complaints of recent onset) of the type for which they are seeking treatment. We believe that patients who have reinjured themselves (ie, patients treated previously for similar complaints) have more structural changes and therefore receive treatments that emphasize recovery of function more than physiologic recovery.[6]

The purpose of our study was to describe whether the treatment of patients with pain by Dutch physical therapists reflects the treatment framework that we formulated. Two groups of patients were studied: (1) patients with back pain and (2) patients with knee trauma, who were further divided into three groups related to the time since the trauma ([is less than] 1 month, 1-3 months, [is greater than] 3 months). These groups were selected because they are often seen by physical therapists in primary care settings. The groups were also selected because a broad range of nociceptive mechanisms may be causing their pain. In addition to investigating the treatment of these two groups of patients in general, we focused on two subgroups of patients within each group: (1) patients with subacute complaints and (2) patients who were having their first episode of pain of the type for which they were seeking treatment.

We began our study of the treatments administered by Dutch physical therapists with the following expectations:

1. When therapists first see patients, treatment goals are aimed at establishing conditions for physiologic recovery (pain reduction, reduction of swelling, and decreasing abnormal muscle activity).

2. In subsequent treatments, goals are aimed at reduction of disabilities, improvement of muscle force, and improvement of joint function and stability.

3. Massage, manual therapy, and thermal modalities are applied more often during initial treatments than during subsequent treatments.

4. Exercise and patient education (instruction and advice) are applied more often during initial treatments than during subsequent treatments.

Method

Physical Therapists

We conducted a survey in the Dutch primary health care system for a period of 4 years. Eighty-three physical therapists working in 32 private practices participated in this survey. In the Netherlands, about 60% of all physical therapists work in private practices providing primary health care. These practices have no direct access for patients: a physician's referral is needed.

Comparison of the data on the participating therapists with data on the general population of therapists working in primary care settings 31 showed that these groups were comparable in terms of gender (53% versus 52% male) and postgraduate training (84% versus 95%).32 Physical therapists younger than 35 years of age were somewhat overrepresented (60% in our sample versus 43% in the general population). Therapists having contracts with public health insurance companies were underrepresented (57% in our sample versus 69% in the general population). Practices consisting of three or fewer therapists were also overrepresented (97% in our sample versus 71% in the general population). We do not believe that these factors affected the results to a meaningful extent.

Assessment

A questionnaire was used to obtain information on every patient applying for treatment in the participating practices. This questionnaire consisted of three main sections. The first section focused on general patient characteristics and complaints and on the diagnosis (indication for referral) of the referring physician. The second section focused on the diagnostic findings of the physical therapist. These findings were described in terms of impairments and disabilities.(*) This classification is derived from the International Classification of Impairments, Disabilities, and Handicaps (ICIDH).[33] The impairments and disabilities that could be indicated in the questionnaire are listed in Tables 1 and 2. Information concerning the general patient characteristics, the medical diagnosis, and the diagnostic findings of the physical therapist were obtained at the first treatment session for each patient. Finally, the questionnaire included a section on treatment goals and interventions. In each treatment session, the therapist indicated the intervention that was applied and the treatment goal that was pursued with respect to the particular intervention. These treatment goals were derived from the diagnosed impairments and disabilities.[1] Table 3 shows the treatment goals and interventions that could be indicated. In our study, we did not examine whAt treatment goals and interventions were actually used. Instead, our data were derived from self-reports, and our study reflects perceptions of treatment and goal setting.

Table 1.
Frequency of Impairments

                                Percentage    With Percentage of
                                of Patients   Patients With
                                Back Pain     Knee Trauma
Impairments                     (n = 1,085)   (n = 416)

Pain                              100.0            100.0
  With rest                        65.7             31.5
  With movement                    82.8             94.0
  With change between rest
    and movement                   46.4             38.7

Restriction in joint range of
  motion                           90.7             74.5

Increased muscle activity          82.2             16.3

Decreased muscle activity           6.0             34.6

Diminished muscle force            31.3             74.5

Abnormal posture (kyphosis,
  lordosis, scoliosis)             63.3             20.0

Swelling                            4.4             61.5

Respiratory problems                2.5              0.7

Other impairments                  38.4             37.7
Table 2.
Frequency of Disabilities

                               Percentage    Percentage
                               of Patients   of Patients
                               With Back     With Knee
                               Pain          Trauma
                               (n = 1,085)   (n = 416)

Self-care
  Washing                        13.5           6.7
  Dressing                       19.7          10.1
  Using lavatory                  7.9           6.2
  Eating                          1.9           0.2

Physical control
  Sitting                        42.8          12.3
  Standing                       40.5          35.3
  Kneeling                       28.7          75.5
  Bending                        64.5          21.2
  Keeping balance                10.6          17.3

Mobility
  Getting in and out of bed      44.4          17.3
  Walking                        36.4          67.1
  Climbing stairs                31.8          81.0
  Cycling                        24.1          58.4
  Driving a car                  24.7          22.8

Household and professional
   activities
  Shopping                       33.2          24.8
  Preparing meals                16.6          11.8
  Changing beds                  32.3          15.1
  Doing housework                33.4          20.0
  Caring for other household
   members                       11.7           8.9
  Using telephone                 1.2           0.5
  Standing for long periods      66.3          57.7
  Sitting for long periods       67.4          19.0
  Lifting                        72.7          33.4
  Maintaining a normal tempo
   during work                   62.5          51.2
  Stress resistance              19.1           5.8

Sports/hobbies
  Sports                         21.7          50.5
  Hobbies                        18.1          12.5
  Other activities               13.7           7.0

[TABULAR DATA 3 NOT REPRODUCIBLE IN ASCII]

Patients

Data from two patient groups were used in our study: (1) patients with back pain and (2) patients with knee trauma. These groups were selected using the medical diagnosis and the diagnostic findings of the physical therapist.

Patients with back pain were classified using International Classification of Primary Care (ICPC) codes L02 (back symptoms and complaints) and L03 (low back complaints without radiation of pain).[34] Patients with other diagnoses concerning a clear pathology, that is, syndromes of the cervical spine (ICPC code L83), osteoarthritis of the spine (ICPC code L84), acquired deformities of the spine (ICPC code L85), lumbar disk lesion or radiation of pain (ICPC code L86), and rheumatoid arthritis (ICPC code L88) were excluded. The group of patients with knee trauma were classified using ICPC codes L96 (acute injury of the meniscus or ligament of the knee) and L97 (chronic internal derangement of the knee). In addition, the physical therapist had to report one or more of the following conditions: pain with rest, pain with movement, and pain with change between rest and movement. Only patients who had therapy for 6 weeks or more (55% of the patients treated in private practices[32]) were included. These criteria resulted in a study sample of 1,085 patients with back pain and 416 patients with knee trauma.

Data Analysis

Physical therapy management of the patients was divided into three stages: less than 3 weeks, 3 to 6 weeks, and more than 6 weeks. This division is based on the idea that physical therapy management of patients changes with time. After 3 weeks, we believe that pain caused by uncomplicated injuries usually resolves.[35] After 6 to 8 weeks, we believe that the healing of connective tissue structures is nearly finished.[6]

Data on treatment goals and interventions were gathered during each treatment session. For each session, therapists described treatment goals and interventions. Data were then reduced to the level of the patients. For example, a patient was treated in 10 sessions in which "recovery of joint range of motion" was the only treatment goal. This treatment goal was pursued by means of massage (sessions 1-5) and by means of exercise (sessions 6-10). The frequency of "recovery of joint range of motion" as a treatment goal in this treatment is therefore 100%. The frequency of the two interventions is 50% each.

For each stage of treatment, the frequencies of treatment goals and interventions were computed. These proportions were used in the statistical analyses and are presented in Figures 1 through 4. Data on goals and interventions from the first session were omitted from these analyses because of the deviating character of this session (the first session focused mainly on the physical examination of the patient). During the first sessions, priority is usually given to the intake and physical examination, and we therefore did not believe that data on treatment would necessarily relate to the treatment framework.

[Figures 1-4 ILLUSTRATION OMITTED]

We also analyzed treatment for subgroups of patients. First, a selection was made of patients having their first episode of pain of the type for which they were seeking treatment. Thus, patients with previous treatments for similar complaints were excluded. Second, a selection was made based on the recency of onset of the complaints. Patients with complaints of 1 to 3 months (referred to as patients with subacute problems) were studied. This 3-month cutoff corresponds to the guidelines of the International Association of Pain,[30] which suggest setting the time at which pain can be called chronic at 3 months postonset. Evidence for this cutoff is given by Abbott et al[35] and Philips and Grant.[36]

Because data were not normally distributed, nonparametric analyses were performed. Friedman's two-way analysis of variance for related samples[37] was used to examine differences in the treatment goals and interventions among the three stages of treatment ([is less than] 3 weeks, 3-6 weeks, [is greater than] 6 weeks). The level of significance was set at .05. The data were analyzed using SPSS-X version 4.1 software.([dagger])

Results

Patients

General characteristics of the two groups of patients are shown in Table 4. The mean age of the patients with back pain was 43.5 years (SD = 16. 1, range = 10 - 84), and the mean age of the patients with knee trauma was 36.2 years (SD=17.6, range=9-78). Men and women were about equally represented in both groups (58.9% of the patients with back pain and 51.4% of the patients with knee trauma were female). Patients with back pain were referred by a specialist less frequently than patients with knee trauma were (4.3% versus 51.9%). In both groups, approximately 42% of the patients had complaints with a recent onset ([is less than] 1 month). Half of the patients with back pain and 29.3% of the patients with knee trauma, however, had a reinjury; that is, they had been treated previously for the same complaints.

Diagnostic Findings: Impairments and Disabilities

The frequency of impairments, according to the physical therapists, is provided in Table 1. Only patients who reported pain were included in the groups. In addition to pain, other frequently diagnosed impairments for the patients with back pain were restricted joint range of motion, increased muscle activity, and, to a lesser extent, impairments in posture. Frequently diagnosed impairments for patients with knee trauma, besides pain, were diminished muscle force, restricted joint range of motion, and swelling.

The frequency of disabilities is shown in Table 2. Frequently diagnosed disabilities for the patients with back pain were problems in lifting, sitting, and standing for long periods. For the patients with knee trauma, frequently diagnosed disabilities were problems with climbing stairs, kneeling, and walking.

Episode of Care and Number of Visits

According to our selection criteria, all episodes of care in this study lasted more than 6 weeks. After 9 weeks, treatment was finished for 45% of the patients with back pain and for 44% of the patients with knee symptoms. After 15 weeks, 17% of the patients with back symptoms and 21% of the patients with knee trauma were still being treated.

About half of the patients (48% of the patients with back pain and 41% of the patients with knee trauma) received treatment consisting of a maximum of 12 visits. An episode of care of more than 24 visits was registered for 12% of the patients with back pain and for 19% of the patients with knee trauma.

Treatment Goals

Treatment goals with a frequency of 1% or less were omitted from our analysis because we believe that they cannot be considered meaningful in the context of our study (eg, reduction of swelling for patients with back pain). These treatment goals are of no importance in the treatment of patients in these categories. Therefore, no interesting shifts in emphasis can be expected in these treatment goals.

In the questionnaire, two types of treatment goals could be indicated by the therapists: (1) those goals concerning the alleviation of or recovery from impairments and (2) those goals concerning the alleviation of or recovery from disabilities. In Figure 1, the treatment goals concerning the alleviation of or recovery from impairments are presented for the three stages of treatment for the patients with back pain.

For the patients with back pain, the choice of pain reduction as a treatment goal differed across the three stages of treatment ([chi square] = 26.2, df = 2, P = .0000). As expected, pain reduction was chosen more often early in the course of treatment. Contrary to our expectation, the choice for treatment goals aimed at the reduction of disabilities, improvement of muscle force, and improvement of joint function and stability did not differ across stages of treatment.

Figure 2 shows the treatment goals relating to the alleviation of or recovery from impairments across the three stages of the treatment for the patients with knee trauma. Treatment goals relating to reduction of swelling ([chi square] = 12.2, df = 2, P = .0022) and improvement of joint stability ([chi square] = 8.0, df = 2, P = .0175) differed across the stages of treatment. As expected, reduction of swelling was chosen as a treatment goal less often during the course of the treatment, whereas improvement of joint stability was chosen more often. In contrast to our expectations, no increase occurred in the choice of the treatment goal relating to improvement of muscle force and no decrease occurred in the choice of pain reduction as a treatment goal.

For both groups of patients, no differences were found for treatment goals relating to the alleviation of or recovery of disabilities across the three stages of treatment. This finding suggests that our expectation that treatment goals directed toward the reduction of disabilities would be chosen more frequently as treatment progresses was incorrect.

Interventions

Interventions with a frequency of 1% or less were omitted from our analysis because we believe that they cannot be considered meaningful in the context of our study (eg, heat therapy and cryotherapy for patients with knee trauma)

The treatment course of interventions for patients with back pain is shown in figure 3. The use of two interventions--exercise ([chi squared] = 18.3, df=2, P=.0001) and manual therapy ([chi squared] =27.1, df-2, P=.0000)--differed across the three stages of treatment. As expected, exercise was applied more often and manual therapy was used less often during the course of treatment. There were no other differences in the reported application of interventions.

Figure 4 shows the application of interventions for patients with knee trauma. As expected, experience was applied more often during the course of treatment ([chi squared] = 15.5, df = 2, P = .0004). Consistent with expectations, application of ultrasound decreased during the course of treatment ([chi squared] = 6.2, df = 2, P = .0445). The application of other interventions was not in accordance with our expectations.

Treatment of Patients With Subacute Complaints or Having Their Episode of Pain

Table 1 gives an overview of the results for subgroups of patients: patients with subacute complaints (1-3 months) and patients having their first episode of pain. Differences in the use of treatment goals and interventions during the course of treatment are shown. For the purpose of comparison, the results for the patients in general are also shown (these results correspond to the results illustrated in Figs. 1-4).

The treatment framework does not appear to apply to patients with subacute complaints or to patients having their first episode of pain. The treatment of patients with subacute complaints showed even fewer changes that were in accordance with the treatment framework than did the treatment of patients with back pain and patients with knee trauma in general.

There was hardly any difference between the total groups of patients and the two subgroups. Mean differences in treatment goals or interventions hardly varied among the groups.

Discussion

We provided an overview of the treatment of two groups of patients based on therapists' reports. These groups were (1) patients with back pain and complaints and (2) patients with knee trauma. A framework was formulated in which physical therapy is considered to be a process consisting of two phases: (1) a phase in which conditions are created for physiological recovery and (2) a phase directed toward the reduction of disabilities.

The results suggest some use of the first part of the framework (ie, creation of conditions for physiological recovery). The results also showed that the central intervention was the application of exercise, which was consistent with second part of the framework (ie, reduction of disabilities). Specifically, with respect to treatment goals established in the first phase, the therapists' reports showed that the emphasis on pain reduction decreased for the patients with back pain and the emphasis on reduction of swelling decreased for the patients with knee trauma during the course of treatment. Furthermore, with respect to interventions, decreases in the application of manual therapy for patients with back pain and in the use of ultrasound for patients with knee trauma were shown over time.

In the second phase, the expected emphasis on the treatment goals of improvement of muscle force, improvement of joint function, improvement of joint stability, and reduction of disabilities was hardly observed as treatment progressed. The only result regarding treatment goals that corresponded to the framework was the use of improvement of joint stability as a treatment goal for patients with knee trauma. Our expectation of an emphasis on exercise as treatment progressed was reflected in an increase in the application of exercise for both patient groups. This finding is in accordance with the notion that exercise is an important intervention for a durable recovery.[7,9,21,22]

We expected the framework to apply to the treatment of two subgroups. The first subgroup consisted of patients with subacute complaints. Their treatment characteristics, however, did not show a better fit with the treatment framework. Compared with the patients with back pain and the patients with knee trauma in general, this subgroup was treated with less consistency with e treatment framework. A possible explanation for this finding is the smaller number of patients in these subgroup, resulting in reduced power of the tests. The mean scores for this subgroup, however, showed no indications of a more pronounced application of the framework compared with the mean scores for the patients in general. The second subgroup consisted of patients having an initial episode of pain. As for the patients with subacute complaints, we expected that the treatment framework would apply to this group in particular. The evidence, however, did not support this expectation.

The generalizability of our results to other countries can be questioned because comparable treatment processes may not be expected outside the Netherlands. Comparisons of US and Dutch physical therapy practice on two measures--episodes of care and average number of therapy visits--show similar results with regard to at least one of the largest patient groups (ie, patients with low back pain). A recent analysis of data of the American Physical Therapy Association's Survey of Physical Therapy Outpatient Practice[38] on patients with low back pain suggests a moderate difference in the average episode of care (5.3 weeks in the United States versus 7 weeks in the Netherlands[39]), whereas the number of visits per episode of care shows a small difference (approximately 11 in the United States versus approximately 10 in the Netherlands for patients with similar complaints[39]). These figures suggest that, to a certain extent, some conditions are comparable and therefore that the present findings may not be unique to the Netherlands. Many other differences, however, cannot be accounted for, and comparisons can only be done with caution.

Conclusion

Our results show that a framework based on two phases of patient management is only partially reflected in Dutch physical therapy practice. The use of such a framework can be considered a means for understanding physical therapy practice, and this can be an important first step in studying the outcome of treatment.[1,4]

Acknowledgments

We express our gratitude to all physical therapists who participated in the study and to Ms Rinie de Groot, who played a vital role in the collection of data. We also thank Dr Marij Roebroeck for her helpful comments on an earlier draft of the manuscript.

(*) Although diagnosis in physical therapy may encompass more than a description of the patient's health status in terms of impairments and disabilities, we used impairments and disabilities as the diagnosis in this study.

([dagger]) SPSS Inc, 444 N Michigan Ave, Chicago, IL 60611.

References

[1] Dekker J, van Baar ME, Curfs EC, Kersens JJ. Diagnosis and treatment in physical therapy: an investigation of their relationship. Phys Ther. 1993; 73:568-580.

[2] Rose SJ. Physical therapy diagnosis: role and function. Phys Ther. 1989;69:535-537.

[3] Sahrmann SA. Diagnosis by the physical therapist--a prerequisite for treatment: a special communication. Phys Ther. 1988;68:1703-1706.

[4] Rothstein JM. Editor's note: When thoughtfulness dies. Phys Ther. 1996;76:342-343.

[5] Delitto A, Erhard RE, Bowling RW. A treatment-based classification approach to low back pain syndrome: identifying and staging patients for conservative treatment. Phys Ther. 1995;75:470-485.

[6] DeRosa CP, Porterfield JA. A physical therapy model for the treatment of low back pain. Phys Ther. 1992;72:261-269.

[7] Davis JN. Rehabilitation of sports injuries: a practical approach. In: Bernhardt DB, A Sports Physical Therapy. New York, NY: Churchill Livingstone Inc; 1986;10:155-172. Clinics in Physical Therapy.

[8] Sikorski JM. A rationalized approach to physiotherapy for low-back pain. Spine. 1985:10;571-579.

[9] Aufdemkampe G, Meijer OG, Winkel D, Witmaar GC. Manuele pijnbenaderingen. In: Matti H, Menges LJ, Spierdijk J, eds. Pijninformatorium. Alphen aan den Rijn, the Netherlands: Samson Stafleu; 1985:1-25.

[10] Reed B, Zarro V. Inflammation and repair and the use of thermal agents. In: Michlovitz SL, ed. Thermal Agents in Rehabilitation. 2nd ed. Philadelphia, Pa: FA Davis Co; 1990:1-16.

[11] Hellsing AL, Linton SJ, Kalvemark M. A prospective study of patients with acute back and neck pain in Sweden. Phys Ther. 1994;74: 116-128.

[12] Schlapbach P. Exercise in low-back pain. In: Schlapbach P, Gerber NJ, eds. Physiotherapy: Controlled Trials and Facts--Rheumatology. Basel, Switzerland: S Karger AG, Medical and Scientific Publishers; 1991: 34-46.

[13] Kergerreis S. The construction and implementation of functional progressions as a component of athletic rehabilitation. J Orthop Sports Phys Ther. 1983;5:14-19.

[14] Gam AN, Johannsen F. Ultrasound therapy in musculoskeletal disorders: a meta-analysis. Pain. 1995;63:85-91.

[15] Aufdemkampe G, Meijer OG, Winkel D. Fysiothechniek aan de extremiteiten. In: Matti H, Menges LJ, Spierdijkj, eds. Pijninformatorium. Alphen aan den Rijn, the Netherlands: Samson Stafleu; 1993:1-41.

[16] Pages IH. Elektrotherapie: hochfrequenzbereich. Physiotherapy. 1992;83:51-55.

[17] Koes BW, Assendelft WJJ, van der Heijden GJMG, et al. Spinal manipulation and mobilisation for back and neck pain: a blinded review. BMJ. 1991;303:1298-1303.

[18] Koes BW, Bouter LM, Beckerman H, et al. Physiotherapy exercises and back pain: a blinded review. BMJ. 1991;302:1572-1576.

[19] Falconer J, Hayes KW, Chang RW. Therapeutic ultrasound in the treatment of musculoskeletal conditions. Arthritis Care Research. 1990; 3:85-91.

[20] Wells P, Lessard E. Movement education and limitation of movement. In: Wall PD, Melzack R, eds. Textbook of Pain. Edinburgh, Scotland: Churchill Livingstone; 1989:952-963.

[21] Twomey LT. A rationale for the treatment of back pain and joint pain by manual therapy. Phys Ther. 1992;72:885-892.

[22] Nachemson AL. Exercise, fitness, and back pain. In: Bouchard C, Shepard RJ, Stephens T, et al, eds. Exercise, Fitness, and Health: A Consensus of Current Knowledge. Champaign, Ill: Human Kinetics Inc; 1990:533-540.

[23] Sluijs EM, van der Zee J, Kok GJ. Differences between physical therapists in attention paid to patient education. Physiotherapy Theory and Practice. 1992;9:103-117.

[24] Haldeman S. Manipulation and massage for the relief of pain. In: Wall PD, Melzack R, eds. Textbook of Pain. Edinburgh, Scotland: Churchill Livingstone; 1989:942-951.

[25] Stratford PW, Levy DR, Gauldie S, et al. The evaluation of phonophoresis and friction massage as treatments for extensor carpi radialis tendinitis: a randomized controlled trial. Physiotherapy Canada. 1989;41:93-99.

[26] Falconer J, Hayes KW, Chang RW. Effect of ultrasound on mobility in osteoarthritis of the knee. Arthritis Care Research. 1992;5:29-35.

[27] Waddell G. A new clinical model for the treatment of low back pain. Spine. 1987;12:632-644.

[28] Cardenas DD, Egan JJ. Management of chronic pain. In: Kottke FJ, Lehmann JF, eds. Krusen's Handbook of Physical Medicine and Rehabilitation. Philadelphia, Pa: WB Saunders Co; 1990:1162-1168.

[29] McCulloch J. Approaches to dealing with musculoskeletal pain. In: Echternach JL, ed. Pain. New York, NY. Churchill Livingstone Inc; 1987;12:125-140. Clinics in Physical Therapy.

[30] International Association for the Study of Pain, Subcommittee on Taxonomy. Classification of chronic pain, description of chronic pain syndromes, and definitions of pain terms. Pain. 1986;44:s1-s226.

[31] Pool JB, Hingstman L. Cijfers uit de Registratie van Beroepen in de Eerstelijnsgezondheidszorg 1991: Fysiotherapeuten. Utrecht, the Netherlands: NIVEL; 1993.

[32] van der Valk RWA, Dekker J, Boschman M. Basisgegevens Extramurale Fsiotherapie 1989-1992: Gegevens uit het Project "Beleidsgericht Evaluatie: Effectonderzoek Extramuralee Fysiotherapie (BEEF)." Utrecht, the Netherlands: NIVEL; 1994.

[33] International Classification of Impairments, Disabilities, and Handicaps. Geneva, Switzerland; World Health Organization; 1980.

[34] Lamberts H, Wood M. International Classification of Prima7y Care. Oxford, England: Oxford University Press; 1987.

[35] Abbott FV, Gray-Donald K, Sewitch MJ, et al. The prevalence of pain in hospitalized patients and resolution over six months. Pain. 1992;50: 15-28.

[36] Philips HC, Grant L. The evolution of chronic back pain problems: a longitudinal study. Behav Res Ther. 1991;29:435-441.

[37] Siegel S, Castellan NJ. Nonparametric Statistics for the Behavioral Sciences. London, England: McGraw-Hill; 1988.

[38] Jette AM, Smith K, Haley SM, Davis KD. Physical therapy episodes of care for patients with low back pain. Phys Ther. 1994;74:101-115.

[39] Zuijderduin W, Dekker J, Abrahamse H. Determinanten van de omvang van behandeling in de extramurale fysiotherapie. Tidschrift voor Sociale Gezondheidszorg. 1995;75:274-281.

ME van Baar, is Research Associate, Netherlands Institute of Primary Health Care, PO Box 1568, 3500 BN, Utrecht, the Netherlands. Address all correspondence to Ms van Baar.

J Dekker, PhD, is Professor of Allied Health Care and Head of the Research Department, Netherlands Institute of Primary Health Care.

W Bosveld, PhD, is Senior Research Associate, Netherlands Institute of Primary Health Care.

This study was supported by a grant from the Dutch Ministry of Health.

The research was presented at the 12th International Congress of the World Confederation for Physical Therapy; June 25-30, 1995; Washington, DC.