| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Research Reports |
JA Cleland, PT, PhD, OCS, FAAOMPT, is Associate Professor, Department of Physical Therapy, Franklin Pierce University, 5 Chenell Dr, Concord, NH 03301 (USA); Physical Therapist, Rehabilitation Services, Concord Hospital, Concord, New Hampshire; and Faculty, Regis University Manual Therapy Fellowship Program, Denver, Colorado
JM Fritz, PT, PhD, ATC, is Associate Professor, Department of Physical Therapy, University of Utah, Salt Lake City, Utah, and Clinical Outcomes Research Scientist, Intermountain Healthcare, Salt Lake City, Utah
GP Brennan, PT, PhD, is Director for Clinical Quality and Outcomes Research, Intermountain Healthcare
J Magel, PT, DSc, OCS, FAAOMPT, is Director, Intermountain Orthopedic and Spine Therapy, Intermountain Healthcare
Address all correspondence to Dr Cleland at: joshcleland{at}comcast.net
Submitted January 27, 2008;
Accepted August 18, 2008
 | Abstract |
|---|
Participants: The study participants were 19 physical therapists who attended a 2-day CE course focusing on the management of neck pain. All patients treated by the therapists in this study completed the Neck Disability Index (NDI) and a pain rating scale at the initial examination and at their final visit.
Methods: Therapists from 11 clinics were invited to attend a 2-day CE course on the management of neck pain. After the CE course, the therapists were randomly assigned to receive either ongoing education consisting of small group sessions and an educational outreach session or no further education. Clinical outcomes achieved by therapists who received ongoing education and therapists who did not were compared for both pretraining and posttraining periods. The effects of receiving ongoing education were examined by use of linear mixed-model analyses with time period and group as fixed factors; improvements in disability and pain as dependent variables; and age, sex, and the patient's initial NDI and pain rating scores as covariates.
Results: Patients treated by therapists who received ongoing education experienced significantly greater reductions in disability during the study period (pretraining to posttraining) than those treated by therapists who did not receive ongoing training (mean difference=4.2 points; 95% confidence interval [CI]=0.69, 7.7). Changes in pain did not differ for patients treated by the 2 groups of therapists during the study period (mean difference=0.47 point; 95% CI=–0.11, 1.0). Therapists in the ongoing education group also used fewer visits during the posttraining period (mean difference=1.5 visits; 95% CI=0.81, 2.3).
Discussion and Conclusion: The results of this study demonstrated that ongoing education for the management of neck pain was beneficial in reducing disability for patients with neck pain while reducing the number of physical therapy visits. However, changes in pain did not differ for patients treated by the 2 groups of therapists. Although it appears that a typical CE course does not improve the overall outcomes for patients treated by therapists attending that course, more research is needed to evaluate other educational strategies to determine the most clinically effective and cost-effective interventions.
| Introduction |
|---|
 TopAbstract Introduction MethodResultsDiscussionConclusionReferences |
|---|
One explanation offered for the slow transfer of new evidence to clinical practice is the insufficiency of existing continuing education (CE) methods.15 The primary purpose of CE is to enhance the transfer of new knowledge and change clinician behaviors to lead to overall improvements in clinical performance and patient outcomes.16 Physical therapists frequently attend traditional CE courses, consisting of a few days of intensive training at a location away from their typical clinical environments, to acquire new skills with the goal of improving the quality of care provided to their patients.17,18 Other CE methods that have been used include the dissemination of written educational materials, ongoing audit and feedback of clinical performance, and educational outreach visits in which clinicians receive training in their professional settings.19,20 All of these methods appear to improve overall clinician performance to various degrees; however, no data exist to suggest which CE interventions are most effective.19,21–23 Additionally, these results cannot be generalized because differences in professional roles, education, values, and status may affect the results of educational interventions.24
Although several educational strategies designed to enhance the transfer of evidence to clinical practice have been proposed,19 the impact of such strategies on patient outcomes has received little attention in the physical therapy literature. Recently, Brennan and colleagues25 found that a traditional 2-day CE course did not lead to improved outcomes for patients with neck pain treated by the attending therapists. The study did reveal that when therapists attended the CE course and then participated in a quality-improvement project involving ongoing audit and feedback, their patients’ clinical outcomes did improve. This result may indicate the need for a longitudinal approach involving feedback and follow-up with clinicians to affect clinical outcomes. The therapists in that study,25 however, were not randomly assigned to participate in the quality-improvement project but were selected by the researchers; this strategy could have biased the results.25
The purpose of this randomized clinical trial was to determine whether therapists who attended a traditional 2-day CE course and received ongoing education experienced better clinical outcomes for their patients with neck pain than therapists who only attended the 2-day CE course.
| Method |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionConclusionReferences |
|---|
The CE course was presented by 2 physical therapists, each of whom was an Orthopaedic Certified Specialist and had fellowship status in the American Academy of Orthopaedic Manual Therapy. The CE course was delivered over 2 days (4 hours per day) and focused on the management of neck pain with a previously developed classification system.12 All attending therapists completed a questionnaire that included age, sex, number of years of practice, specialist certification status, and residency or fellowship training. Additionally, all therapists were asked to score their level of confidence in the management of neck pain on a 5-point Likert scale ranging from not confident at all (0) to very confident (5).
The CE course included both lectures (approximately 25%) and hands-on practical sessions (approximately 75%). The lecture portion included discussion of the current best evidence in support of the classification system for the management of neck pain.3–12 The practical sessions consisted of demonstration and practice of manual physical therapy techniques (thrust and nonthrust) directed at the cervical and thoracic spine, as well as therapeutic exercises targeting the deep neck flexor, lower trapezius, middle trapezius, and serratus anterior muscles. Details regarding the exact techniques demonstrated and practiced by the participants can be found elsewhere.26 We chose to focus on these intervention strategies because of the evidence supporting their effectiveness for certain subgroups of patients,3–12 as well as our previous research13,25 and the research of others2,27 suggesting that the use of these interventions is unjustifiably low. All participants received a course manual that contained lecture notes related to the evidence for the use of the various interventions, as well as figures with detailed descriptions of each manual therapy technique and exercise covered. The course concluded with a question-and-answer session.
Randomization
After the 2-day CE course, all participating therapists who had treated at least 10 patients with neck pain in the preceding year were randomly assigned to 1 of 2 groups. All therapists were informed that they were not required to participate and that, if they declined, their data would not be included in the analysis of the quality-improvement project. Therapists who had not treated at least 10 patients with neck pain in the preceding year were not included in the study because we wanted to ensure that participating therapists had at least moderate exposure to this patient population.
One group was randomly assigned to receive ongoing education in the use of the evidence-based interventions, and the second group received no further education organized by IHC beyond the CE course. Randomization was accomplished with a computer-generated random numbers table corresponding to numbers assigned to each participating therapist in alphabetical order of the last name; the individual performing the randomization was unaware of the random numbers table. Randomization to groups was performed after completion of the 2-day CE course to ensure that the instructors were unaware of the group assignments when providing feedback on skill performance during the course. Because therapists in both groups worked in the same clinics, therapists assigned to the ongoing education group were specifically asked not to discuss with others the further education that they received in an attempt to minimize contamination bias.
Educational Interventions
Only therapists in the group randomly assigned to receive ongoing education received training beyond the 2-day course. The additional training included two 1.5-hour educational meetings provided by the same clinicians who delivered the 2-day CE course. These meetings occurred 4 and 7 weeks after the completion of the 2-day course. The meetings entailed review of the classification system for the management of neck pain, skill demonstration, and practice of previously learned techniques with feedback from the instructors. Therapists also had the opportunity to ask questions regarding skills that they had learned and to discuss the management of specific cases.
In addition to the 2 educational meetings, all therapists randomly assigned to receive ongoing education participated in an outreach visit.19 The outreach visit included a 1-hour co-treatment of a patient with neck pain in the therapist's own clinical practice setting with the principal investigator of the study. The therapist identified the patient for whom the co-treatment would occur on the basis of patient availability and the consent of the individual patient. After the co-treatment, the therapist and the principal investigator discussed the clinical presentation of the patient, assignment to the appropriate classification, and the therapist's clinical decision-making process regarding management strategies. All therapists were encouraged to use interventions with evidence for effectiveness for the particular classification and to appropriately use techniques learned during the CE course.
Therapists
Thirty-eight therapists from 11 different IHC physical therapy clinics were initially invited to attend the CE course. A total of 30 therapists elected to attend and participate in the CE course. Of these, 1 therapist declined to participate in the quality-improvement project, 1 therapist was switching his place of employment, and 9 therapists had treated fewer than 10 patients with neck pain in the preceding year and therefore were not included in the study. Therapists who were not included were younger (P=.03) and had less confidence (P=.02) in treating patients with neck pain (Fig. 1). The remaining 19 therapists (mean age=41.0 years, SD=7.6) were randomly assigned to the group receiving ongoing education (n=10; mean age=38.5 years, SD=8.0) or to the control group (n=9; mean age=43.8 years, SD=6.5), which received no further educational interventions. Four therapists (2 in each group) had obtained Orthopaedic Certified Specialist status. None of the therapists had completed a residency or fellowship training.
|
Clinical Data Collection
Clinical data were collected from the clinical outcomes database maintained by the Rehabilitation Agency of IHC. Beginning in 2002, all outpatient physical therapy clinics in the Rehabilitation Agency began tracking clinical outcomes for all patients receiving physical therapy. In the clinical outcomes database, each new patient is entered by use of a Web-based application. At each visit, patients provide a condition-specific disability outcome score and a numeric pain rating (0–10),28 and these values are entered into the database. The Neck Disability Index (NDI)29 is the condition-specific disability scale used for patients with a chief complaint of neck pain. Also included are the patient's age, sex, symptom duration, and date of surgery (if applicable). The number of physical therapy visits, the duration of physical therapy services, typical costs billed for physical therapy, and insurance provider also can be obtained from the database.
The sample of patients for this study was drawn from the clinical outcomes database for therapists who had treated at least 10 patients with neck pain in the preceding year and who attended the CE course. Data for all patients undergoing an episode of care for the treatment of neck pain between June 1, 2005, and June 15, 2006, were retrospectively extracted from the database to examine the pretraining clinical outcomes achieved by the therapists. Data for all patients undergoing an episode of care between August 1, 2006 (completion of educational intervention), and August 1, 2007, were extracted to examine the posttraining clinical outcomes achieved by the therapists. Patients who received treatment between June 15, 2006, and August 1, 2006, were not included in the analysis because this was the time frame during which the educational intervention occurred. Figure 2 outlines the timeline for study procedures. Patients receiving treatment for neck pain were identified as those for whom the NDI was used as the condition-specific disability scale. Data were excluded for patients who did not complete at least 2 physical therapy visits, who had an initial NDI score of less than 20%, or who had a surgery date recorded.
|
10% was categorized as achieving the MCIC.
The NPRS was used to capture a patient's perceived level of pain. Patients were asked to indicate the intensity of their current pain using an 11-point scale ranging from 0 (no pain) to 10 (worst pain imaginable).28 A change in the NPRS score (NPRS change score) was calculated for each patient by subtracting the final NPRS score recorded for the patient from the initial NPRS score. The MCIC for the NPRS has been reported to be 2 points.36 Therefore, all patients with an NPRS change score of 2 points were categorized as achieving the MCIC.
Data Analysis
Baseline patient variables were compared between patients treated in the pretraining period and those treated in the posttraining period, as well as between patients receiving treatment from a therapist in the ongoing education group and those receiving treatment from a therapist in the control group. Baseline comparisons were made with independent t tests for continuous data and with chi-square tests of independence for categorical data.
The effects of physical therapists receiving ongoing education were examined by comparing the clinical outcomes for patients treated by therapists receiving ongoing education with the clinical outcomes for patients treated by therapists in the control group. Linear mixed-model analyses were used to account for the possibility that patient outcome scores would not be independent but might be correlated on the basis of the physical therapist performing the treatment and the clinic within which the treatment occurred. A hierarchical approach considered patients nested within therapists, who were nested within clinics. The variables of therapist and clinic were modeled as random effects. A variance components covariance structure was used for the random effects. Time period (pretraining or posttraining), group (ongoing education or control), and the interaction between time period and group were modeled as fixed factors, and NDI change scores served as the dependent variable. Age, sex, and the patient's initial NDI and NPRS scores were modeled as fixed-effect covariates. We examined the significance of the random effects in the model to determine whether the variance attributable to the clustering of patients within therapists and within clinics contributed to the model. We examined the coefficients for each of the fixed effects in the model, including the covariates, to determine which effects contributed to the model. The interaction between time period and group was examined to determine whether the NDI change scores for patients treated by therapists in the ongoing education group would differ from those for patients treated by therapists in the control group in the posttraining period but not in the pretraining period. A significant interaction term was analyzed further with planned pair-wise comparisons of the estimated marginal cell means.
We planned to compare the NDI change scores for patients treated by therapists in the ongoing education and control groups during the pretraining and posttraining periods and to compare the NDI change scores for patients during the pretraining and posttraining periods within each group. A separate, hierarchical, mixed-model analysis with the same methods as those described above was performed with NPRS change scores as the dependent variable.
We also used chi-square tests to compare therapists in the ongoing education and control groups with regard to the proportions of patients achieving the MCIC for the NDI and NPRS. Mann-Whitney U tests were used to compare the 2 groups of therapists with regard to the duration of physical therapy and charges. Data analysis was performed with the SPSS version 13.0 statistical software package.*
| Results |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionConclusionReferences |
|---|
|
|
|
Effectiveness of Ongoing Education
The linear mixed-model analysis examining NDI change scores did not reveal a significant effect for the variance attributable to therapists (variance estimate=1.76, P=.48) or clinics (variance estimate=0.15, P=.95). The residual variance estimate was 175.97; therefore, therapists and clinics explained very small percentages of the total variance in NDI change scores (0.99% and 0.084%, respectively). Examining the fixed-effect coefficients for the covariates in the model revealed that age (P=.62) and sex (P=.51) did not significantly contribute to the model. Coefficients for the baseline NDI score (P<.001) and baseline NPRS score (P=.004) were significant, indicating that these variables contributed to the prediction of patients’ NDI change scores. The fixed effect of time period was not significant (mean difference between adjusted mean NDI change scores in pretraining and posttraining periods=–0.74; 95% CI=–2.49, 1.01; P=.41). The fixed effect of group also was not significant (mean difference between adjusted mean NDI change scores for therapists in ongoing education and control groups=1.59; 95% CI=–0.82, 4.0; P=.18). The interaction between time period and group was significant (P=.019).
These results indicated that time period and therapists’ training were not significant predictors of NDI change scores. The combination of time period and training was a significant predictor, such that therapists in the ongoing education group achieved greater improvements in NDI change scores for their patients than therapists in the control group (mean difference=4.2; 95% CI=0.69, 7.7; P=.019).
Pair-wise comparisons of NDI change scores for various combinations of time period (pretraining or posttraining) and group (ongoing education or control) revealed no difference in NDI change scores between the ongoing education group and the control group during the pretraining period (P=.73). During the posttraining period, patients treated by therapists in the ongoing education group had higher NDI change scores than patients treated by therapists in the control group (adjusted mean difference=3.7; 95% CI=0.84, 6.5; P=.013) (Tab. 3).
|
These results indicated that time period and therapists’ training and the interaction between these factors were not significant predictors of NPRS change scores. Pair-wise comparisons did not reveal significant differences in the pretraining or posttraining period on the basis of therapists’ training (Tab. 3).
| Discussion |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionConclusionReferences |
|---|
The findings of the present study are similar to those of our previous study25 showing that educational interventions including strategies for follow-up outreach visits to therapists were more effective for improving clinical outcomes than traditional educational interventions involving short courses of training without any follow-up strategies. These findings support the conclusions of other authors38 that the traditional strategy for CE, emphasizing short-term, intensive courses with no follow-up or individualized outreach, is ineffective for improving patient outcomes. However, it should be recognized that educational interventions including small-group sessions and individualized outreach are likely to be more costly than the traditional strategy.38 We did not examine the cost-effectiveness of the ongoing educational intervention used in the present study. The costs associated with care provided by a therapist who received ongoing education during the posttraining period were significantly lower than those for care provided by a therapist in the control group (median=$599.76 and $735.83, respectively). We expect that these cost savings would exceed the costs associated with the ongoing educational intervention; however, this hypothesis has not been tested.
The outcomes of CE interventions may cover numerous domains, including therapists’ behavioral changes or knowledge retention; satisfaction of the patient, therapist, or both; costs; and clinical outcomes.39 We did not capture the specific interventions delivered by the therapists in the present study and, therefore, cannot make inferences regarding the effectiveness of our CE interventions for facilitating behavioral changes in the participating therapists. It is unclear whether the therapists receiving the ongoing education actually treated patients in a manner more consistent with current best evidence or whether the use of manual therapy or specific strengthening interventions was increased. Previous research40 showed that changing clinicians’ behaviors or enhancing knowledge does not guarantee improved clinical outcomes. For example, a recent randomized trial investigated the effectiveness of an active intervention strategy for implementing evidence-based guidelines for the physical therapy management of low back pain.41 The results demonstrated that although the implementation strategy was effective in improving therapists’ adherence to treatment recommendations,41 there were no improvements in clinical outcomes or reductions in direct medical costs.42 This research illustrates that improved clinical outcomes cannot be inferred from seemingly positive changes in clinical behaviors. We agree with other authors43 who have chosen to prioritize clinical outcomes as the desired effect of any educational intervention strategy. From the results of the present study, we can only infer that the ongoing educational program led to improved clinical outcomes. The reasons explaining these changes are an important topic for further research.
In the present study, several additional interventions were provided for therapists in the ongoing education group; these included small- group follow-up sessions and individualized feedback on performance. The design of the study could not isolate the effectiveness of the additional interventions independently; therefore, it is not possible to determine which components may have been responsible for the improved outcomes. A recent systematic review19 reported that quality-improvement studies investigating the effects of multifaceted intervention strategies generally result in improved care, but it is difficult to determine which components are critical to achieving the improvements. We believe that our ongoing educational intervention may have been successful because it was longitudinal in scope, with both individual and group follow-up sessions, and it was pragmatic for participating physical therapists because it occurred in their practice settings and provided feedback on patients they were treating. These factors have been reported by Greenhalgh and colleagues44 to be potential contributors to the success of an educational intervention strategy. However, further research is needed to investigate the optimal number and type of educational interventions needed to optimize clinical outcomes.
There are limitations to the present study that should be considered. We had a finite amount of descriptive data on the therapists and patients included in the present study. There may be some patient or therapist characteristics that were not collected but may have affected the outcomes of this study. Psychosocial issues, including depression and fear-avoidance beliefs, can affect the prognosis of patients with neck pain45,46 but were not captured in the present study. We were unable, therefore, to adjust for these factors. Although therapists in both groups were similar in terms of experience and training levels, other factors may affect outcomes. For example, if the classification approach to the management of neck pain was within the therapists’ clinical values, then adherence to such an approach might be more likely.47 Additionally, some physical therapists randomly assigned to different groups worked in the same clinics, a factor that could lead to contamination bias. We also collected pretraining data retrospectively, a factor that could introduce potential bias, because therapists during that time period were not aware that they would be participating in a clinical trial. However, this bias would have been exhibited by both therapists receiving and therapists not receiving ongoing education and, therefore, would not have had a direct effect on differences between the groups at any time period.
| Conclusion |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionConclusionReferences |
|---|
| Footnotes |
|---|
The study was approved by the Institutional Review Board of Intermountain Healthcare.
* SPSS Inc, 233 S Wacker Dr, Chicago, IL 60606. 
| References |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionConclusionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  C. M. Williams, C. G. Maher, M. J. Hancock, J. H. McAuley, A. J. McLachlan, H. Britt, S. Fahridin, C. Harrison, and J. Latimer Low Back Pain and Best Practice Care: A Survey of General Practice Physicians Arch Intern Med, February 8, 2010; 170(3): 271 - 277. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Jull Invited Commentary Physical Therapy, January 1, 2009; 89(1): 48 - 50. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
(SD) years. Confidence is reported as 
