Recent research has shown that classifying patients with low back pain into treatment subgroups results in better improvements than treating all patients with low back pain the same. However, physical therapists may use different types of information to determine how to classify their patients. One method uses patient characteristics that have been shown by research to predict good results from a certain type of treatment. Another method uses specific impairments that the physical therapist identifies in a clinical exam to determine which treatment to provide. It is not currently known if one of these methods is better than the other. The purpose of this study is to determine if research-based classification or impairment-based treatment is more effective for treating patients with low back pain that has lasted less than 90 days in terms of improvements in pain and disability. The results of this study may help reduce the high financial cost associated with low back pain.

PURPOSE: The purpose of this study is to describe a multimodal impairment-based clinical model for the treatment of low back pain utilizing the concept of Regional Interdependence (Wainner et al 2007) and to determine the relative effectiveness of this model versus classification-based therapy using the Treatment-Based Classification System for acute and subacute low back pain (Delitto et al 1995, Fritz et al 2006, Fritz et al 2007) in terms of improvements in pain and disability at 4 weeks and 6 months.

BACKGROUND: Recent studies have shown that subclassifying patients with low back pain into treatment groups results in better outcomes than treating all patients with low back pain as homogenous groups. (Fritz et al 2003) However, no studies have directly compared the efficacy of impairment-based versus classification-based physical therapy for patients with acute and subacute low back pain.

HYPOTHESIS: We suspect that both methods of treatment will be effective, but we are uncertain if one will be more effective than the other. Therefore, our null hypothesis will be that both methods will be equally effective.

METHODS:

1. Patients referred to physical therapy at Hendricks Regional Health in Plainfield, IN will be screened for eligibility. Eligible participants will be offered the opportunity to voluntarily participate in the research study. Those patients who are both eligible and interested will receive an informed consent form approved by the Hendricks Regional Health Institutional Review Board.
2. All subjects will complete a standard intake form and self-report measures. The principal investigator will obtain a subjective history from the patient and perform a standardized baseline physical examination.
3. After the baseline examination, subjects will be randomized in blocks of 4, 6, or 8 using a concealed, computer-generated randomization sequence accessed electronically through a Microsoft Excel spreadsheet using the VLOOKUP function to return the group assignment when the subject number is entered.
4. All subjects in both groups will be treated by the principal investigator two times per week for two weeks followed by once per week for two weeks (total of 6 visits) regardless of treatment group assignment. All subjects will receive standardized education as well as additional individualized education as deemed appropriate by the treating clinician. Subjects will be given a home exercise program consisting of the same exercises they performed in physical therapy and asked to perform their exercises twice per day.
5. Patients in the impairment-based group will be treated using a multi-modal treatment approach utilizing manual therapy of the thoracolumbosacral spine and hip as well as motor control exercises. Patients in the classification-based group will be categorized into subgroups according to the previously described Treatment-Based Classification (TBC) Algorithm and treated accordingly.
6. After completion of the 6 visits, we intend to discharge most patients to a home exercise program. However, patients who require more or fewer visits than this will still be included in the intention-to-treat analysis. Patients will be encouraged to perform their home exercises at least 3 times per week until the 6 month follow up.
7. Primary and secondary outcome measures will be collected at 4 weeks and 6 months from baseline.

DATA ANALYSIS:

Study data will be analyzed using SPSS Statistical Software (IBM, Armonk, NY).

1. Baseline characteristics will be examined for between group differences.
2. Primary outcomes will be analyzed using a repeated measures analysis of co-variance (ANCOVA) to examine a group x time interaction for pain and disability at baseline, 4 weeks, and 6 months controlling for baseline covariates of Fear Avoidance Behavior Questionnaire physical activity subscale score (FABQpa), duration of symptoms, and prior history of low back pain. Median and range values will be calculated for Global Rating of Change at 4 weeks and 6 month. Primary outcome measures will also be dichotomized and combined in order to determine overall treatment success for the 4 week and 6 month follow ups.
3. Intention-to-treat analysis using multiple imputation to account for missing values will be used in all analyses. A secondary per-protocol analysis will be performed for patients who completed exactly 6 visits, performed their home exercise program at least 6 times per week, and completed all outcomes at all time points.

Inclusion Criteria:

• primary complaint of low back pain with or without leg symptoms
• duration of symptoms 24 hours to 90 days
• modified Oswestry score of 20 or greater
• able to read and speak English.

Exclusion Criteria:

• current pregnancy
• spinal steroid injections within the past month
• spinal surgery in the past 6 months
• current worker's compensation or motor vehicle accident claim
• presence of any red flag symptoms indicating possible serious pathology or non-musculoskeletal pain