Abstract
Background Based on a behavioral medicine perspective, modern recommendations for physical therapists treating patients with spinal pain include performing a trustworthy physical examination, conveying the message that back pain is benign, and stressing that activity is a key to recovery. However, little evidence is available on how patients perceive these biopsychosocial messages and how patients' perceptions of these messages relate to their recovery.
Objectives The aim of this study was to explore the relationships between perceptions of treatment delivery that are related to an evidence-based approach and psychological factors, treatment outcome, and treatment satisfaction.
Design A cohort study with 3 measurement points was conducted.
Methods Data on 281 participants were collected.
Results High catastrophizing and lower mood in the participants were correlated to “not perceiving the biopsychosocial message” measured at 6 weeks after treatment start. Participants who did not perceive the biopsychosocial message were at higher risk for disability and had lower treatment satisfaction 6 months after treatment start even when controlling for pretreatment pain intensity. “Not perceiving the biopsychosocial message” was not a mediator for treatment outcome and treatment satisfaction. Physical therapists' treatment orientations or attitudes were not related to the perception of the message by the patients.
Limitations There was no measure of actual practice behavior.
Conclusions Maladaptive cognitions and negative emotions appear to affect the way information provided during treatment is perceived by patients. The way information is perceived by patients influences treatment outcome and treatment satisfaction. Physical therapists are advised to check that patients with higher levels of catastrophizing and lower mood are correctly perceiving and interpreting a biopsychosocial message.
Most back pain episodes are mild and rarely disabling, with only a small proportion of individuals seeking care. The majority of patients with new episodes recover within a few weeks, but recurrences are common. Sixty to eighty percent of those initially consulting for back pain still have pain after a year.1 Recent systematic reviews of prospective studies2,3 conclude that psychological distress and associated behaviors can be potent risk factors for poor recovery.
Not surprisingly, evidence-based guidelines recommend that physical therapists assess these risk factors.4 All guidelines recommend a diagnostic triage where patients are classified as having (1) nonspecific back pain, (2) suspected or confirmed serious pathology, or (3) radicular syndrome. Most guidelines also advocate a comprehensive musculoskeletal and neurological examination. With nonspecific back pain, all 15 guidelines from North America, Europe, Australia, and New Zealand further recommend that physical therapists convey the evidence-based and behavioral medicine-based message to patients that they do not have a serious disease (“back pain is benign”) and should stay as active as possible and progressively increase their activity levels.4 However, little evidence is available on how these messages are perceived, which patient characteristics may influence perception, or how patients' perceptions of these messages may relate to their recovery and treatment satisfaction. In other words, the relationship between psychological risk factors and outcome could be mediated by patients' perceptions of the biopsychosocial message or could constitute an independent risk factor of treatment outcome.
Physical therapists can have different treatment orientations (or attitudes) that may influence which information they select to convey to patients. Two different treatment orientations can be extracted from the literature regarding nonspecific musculoskeletal pain: one with attitudes based on the biomedical model of disease and the other with attitudes based on the biopsychosocial model.5
In the treatment orientation based on the biomedical model of disease, diagnoses provide the basis for physical treatment of the illness. As pain is a signal of pathology or tissue damage, a physical therapist with a predominantly biomedical treatment orientation (toward, for example, chronic low back pain) will very likely adapt his or her treatment to the pain level of the patient (ie, use a pain-contingent treatment approach).6,7 Furthermore, treatment will primarily be aimed at finding the physical impairment that is the cause of the pain and consequently treating the impairment.8 Accordingly, the message or information to the patient will be focused on rest rather than activity or overlooking potential psychosocial risk factors.
In the treatment orientation based on the biopsychosocial model, the role of psychological and social factors in the development and maintenance of complaints is emphasized. This model stresses that pain does not have to be a sign of pathology or tissue damage but also is influenced by social and psychological factors. Because of these factors, disability due to pain can be maintained long after the initial pathology has healed. According to this model, it is not necessarily beneficial to adapt the treatment to the pain level of the patient. Rather, treatment should focus on an increase in activity according to a previously defined time frame.9 Physical therapists with a predominantly biopsychosocial treatment orientation generally hold a time-contingent treatment approach.8,10 Their information to patients will be more in accordance with evidence-based guidelines advising patients to stay as active as possible despite pain and informing the patients that back pain is benign.
In the context of spinal pain, although consistent evidence points to the effectiveness of management delivered within a biopsychosocial framework,4 clinicians' practice behaviors and beliefs in the management of spinal pain can be discordant with such a service delivery framework and with current evidence.11,12 Moreover, these non–evidence-based behaviors and beliefs result in poorer patient outcomes and management approaches that are less cost-effective.13–15 Thus, these therapists' characteristics may exert a direct influence on whether patients will actually receive evidence-based advice and, therefore, need to be controlled for when studying the relationships among patient characteristics, perceptions of a biopsychosocial message, and outcome.
Therefore, the primary aim of this study was to explore the relationships between patients' perceptions of a biopsychosocial treatment delivery and patient characteristics, treatment outcome, and treatment satisfaction in a patient population with spinal pain seeking physical therapy in primary care. Specifically, the aim was to investigate whether baseline levels of pain catastrophizing and depressive mood are related to perceptions of the important components of a biopsychosocial treatment delivery. The secondary aim was to investigate whether variations in patients' perceptions of a biopsychosocial message constitute an independent predictor or a mediator of treatment outcome and treatment satisfaction at 6-month follow-up while controlling for the possible influence of therapist treatment orientation and other confounders, such as pain intensity and functional disability.
Method
Overview
This study was a planned secondary analysis of a randomized controlled trial16 where we evaluated the result of an 8-day university training course for physical therapists, designed to educate physical therapists on the methods of integrating psychosocial factors into clinical practice on a patient level. In this trial, the physical therapists recruited consecutive patients with acute and subacute musculoskeletal pain in their clinical setting.
Setting and Participants
The study was carried out in primary care in several counties in Sweden. The inclusion criteria for patients enrolling in the study were musculoskeletal pain and age between 18 and 65 years. The exclusion criteria were sick leave for more than 3 months during the previous year as a result of the present musculoskeletal pain problem and the presence of “red flags,” such as widespread progressive motor weakness or gait disturbance, history of violent trauma, difficulty with micturition, loss of anal sphincter tone or fecal incontinence, and saddle anesthesia. All participants signed an informed consent form. Table 1 contains further descriptions of participant characteristics. Data on a total of 281 eligible patients were collected. The primary analysis of the original randomized controlled trial indicated no differences at 6 months for the primary outcome measures (pain, disability)16 between the patients recruited before the course and those recruited after the course. Therefore, participants were analyzed as a single cohort for this study.
Participant Characteristics at the Start of Physical Therapy
Participating physical therapists were Swedish licensed physical therapists. They both recruited and treated the patients in this study. Eighty-six percent of the participating clinicians were women (mean age=46 years, SD=6.8). The mean years of experience was 18.5 (SD=7.9). The vast majority were employed either at a health care center (48%) or in private practice (48%).
Outcomes and Follow-up
Participants.
All participants were asked to fill out questionnaires at 3 different occasions. First, participants completed an initial questionnaire at baseline (treatment start). Second, they completed a different questionnaire at 6 weeks after treatment start dealing with how they perceived treatment. Third, a follow-up was conducted at 6 months with the same questionnaire as at baseline. The questionnaire at baseline and at 6 months included demographic questions, a question about pain intensity during the previous week, the Quebec Back Pain Disability Scale,17 the Pain Catastrophizing Scale,18 and the Hospital Anxiety and Depression Scale.19 Descriptions of the different measures and measurement instruments at baseline, at 6 weeks after treatment start, and at 6-month follow-up are presented in the eTable. The physical therapists handed out the questionnaires, which the participants took home and posted directly to the researchers. This procedure ensured the physical therapists were blinded to the psychosocial scores at baseline.
The question about pain during the previous week was “How much pain have you had the last week?” and was taken from the Örebro Musculoskeletal Pain Screening Questionnaire.20 The question has a 0 (“not at all”) to 10 (“very much”) Likert scale.
The Quebec Back Pain Disability Scale is a 20-item, self-administered scale designed to assess the level of functional disability in individuals with back pain. Each item on the scale represents an everyday-life activity. The items are rated on a 6-point Likert scale ranging from “not difficult at all” to “unable to do.” The Quebec Back Pain Disability Scale has been shown to have good reliability and validity.17 In a comparison of 5 commonly used questionnaires for assessing disability in people with low back pain, it was shown to be 1 of the 3 most reliable questionnaires and had sufficient width scale to reliably detect improvement or worsening in most individuals.21 The internal consistency in this study was α=.93.
The Pain Catastrophizing Scale is a 13-item, self-administered scale in which people are asked to reflect on past painful experiences and indicate the degree to which they experienced thoughts or feelings during pain on a 5-point scale, ranging from 0 (“not at all”) to 4 (“always”). The scale has been shown to have good reliability and validity in pain outpatient samples.22,23 The internal consistency in this study was α=.94.
The Hospital Anxiety and Depression Scale is a 14-item, self-administered scale with 2 subscales: 1 subscale of 7 items for anxiety and 1 subscale of 7 items for depression. It was designed to measure both anxiety and depression. It has a 4-point scale for each question, ranging from 0 (“not at all”) to 3 (“very often”). It has been shown to be a valid clinical indicator of the possibility of depression and clinical anxiety in a Swedish population.24 The internal consistency in this study was α=.88.
To measure how participants perceived treatment, we sent a questionnaire to all participants 6 weeks after treatment start. Details of this questionnaire are reported elsewhere.25 This questionnaire focused on the patients' perception of treatment delivered. For this study, we selected the following 3 questions to serve as a proxy of an evidence-based approach within a biopsychosocial framework: (1) “Did your physical therapist ensure you your complaint is not a sign of a serious disease?,” (2) “Did your physical therapist advise you to stay active despite your complaint?,” and (3) “Were you satisfied with the physical examination your physical therapist conducted?” The participants could answer “yes,” “no,” or “do not remember.” We dichotomized the answers into “perceived the biopsychosocial message” or “not perceived the biopsychosocial message.” We did this by defining “perceived the biopsychosocial message” as either “yes” on all 3 questions or “yes” on 2 of the 3 questions. Answers of “no” or “do not remember” on 2 or all 3 questions represented “not perceived the biopsychosocial message.” This dichotomization was done to increase power and facilitate interpretation.
For treatment results at 6 months, we combined 3 questions: (1) “How satisfied are you with your treatment result?,” (2) “How satisfied are you with the overall care?,” and (3) “Would you recommend the same type of treatment to a friend?” The participants could answer on a 10-point Likert scale, ranging from “not satisfied at all” to “very satisfied” on the first 2 questions and “no, definitely not” to “yes, definitely” on the last question. The scores from the 3 questions were combined into one score. This approach gives a composite score, ranging from 0 to 30, where a higher score indicates more treatment satisfaction.
Physical therapists.
We measured physical therapists' attitudes and beliefs by means of the Health Care Providers' Pain and Impairment Relationship Scale (HC-PAIRS)26 before the start of the course.
The HC-PAIRS is a questionnaire for assessing the attitudes and beliefs of health care providers in general in terms of functional expectations for patients with back pain.8,27 It consists of 15 statements that are rated on a 6-point Likert scale ranging from 0 (“totally disagree”) to 6 (“totally agree”). A high score on the HC-PAIRS reflects a belief in a strong relationship between pain and impairment, indicating an attitude that pain justifies disability and limitation of activities. Houben et al28 validated the HC-PAIRS and found that scores significantly correlated with all work and activity recommendations reported by therapists in the Netherlands, and a regression analysis showed the HC-PAIRS to be the strongest predictor of these recommendations. These findings indicate that HC-PAIRS scores can be used as an indicator of the recommendations presented to the patient.
The HC-PAIRS was translated into Swedish. An independent translator back translated the questionnaires to identify discrepancies. No discrepancies were found.
Data Analysis
Statistical analysis was carried out using SPSS version 15.0 software (SPSS Inc, Chicago, Illinois). The data were first summarized using descriptive statistics. Pearson product moment correlation coefficients were performed between independent and dependent variables in preparation of regression analyses. Specifically, a goal was to investigate whether pain catastrophizing and depressive mood were related to perceptions of the biopsychosocial message. The role of physical therapists' treatment orientations or attitudes as a possible confounder was investigated in relation to the independent and dependent variables. As it turned out to be unrelated in all cases, these measures were left out of further multivariate analyses. To investigate whether variations in perceptions of the biopsychosocial message constitute an independent predictor, or rather a mediator, of the relationship between patient characteristics and treatment outcome and treatment satisfaction, multiple linear regression was conducted with treatment satisfaction 6 months after treatment and disability 6 months after treatment as dependent variables. The eFigure provides a visual display of the hypothesized models that were tested for plausibility.
Separate analyses were performed for catastrophizing and depression, as it was hypothesized that perceiving the biopsychosocial message could mediate the influence of both variables on outcome.29,30 We controlled for baseline pain and disability. We investigated whether the prerequisites of mediation effects were met using the traditional recommendations by Baron and Kenny31 (eFigure). In case of indications of indirect effects, the significance would thereafter be assessed with a bootstrapping method (n=1,000 bootstrap resamples).32
Role of the Funding Source
This study received funding from the Swedish Research Council for Health, Working Life and Welfare.
Results
Two hundred eighty-one participants filled in both the initial questionnaire at baseline and the questionnaire at 6 weeks after treatment start dealing with how they perceived the biopsychosocial message and at 6-month follow-up. The results showed that 87% (n=244) of the participants indicated that they perceived the biopsychosocial message; 13% (n=37) indicated they did not perceive the biopsychosocial message. Thirty-eight physical therapists contributed with patients (range=1–26, median=6). Twenty-two of the 38 contributing physical therapists had patients who did not perceive the message.
Correlations
Table 2 displays simple correlations among potential predictor, mediator, and outcome variables. These results showed that pretreatment pain, pretreatment disability, and physical therapists' treatment orientation (measured with the HC-PAIRS) were not significantly related to “not perceiving the biopsychosocial message.” However, baseline levels of catastrophizing and depression were significantly related to whether the patient was “not perceiving the biopsychosocial message.” Moreover, pretreatment catastrophizing and depression and “not perceiving the biopsychosocial message” were prospectively related to treatment satisfaction and disability at 6-month follow-up. Treatment satisfaction was negatively correlated to “not perceiving the biopsychosocial message.” Although pain and disability were not related to perceiving the message in the univariate analysis, they were related to outcome and, therefore, were controlled for in the multivariate analysis.
Means (SD) and Cross-Sectional Correlations Among Pretreatment Pain, Pretreatment Disability, Catastrophizing, Depression, “Not Perceiving the Message,” Treatment Satisfaction at 6-Month Follow-up, Disability at 6-Month Follow-up, and HC-PAIRSa
Do Variations in Perceptions of the Biopsychosocial Message Constitute an Independent Risk Factor or a Mediator of Treatment Outcome and Treatment Satisfaction?
To investigate whether perceptions of the biopsychosocial message could be seen to constitute an independent risk factor for treatment outcome and treatment satisfaction or whether they mediate the relationship between catastrophizing or depression and outcome, multiple hierarchical regression analyses were performed to investigate whether the traditional prerequisites of mediation effects were met. As shown in Tables 3, 4, 5, and 6, the effects of catastrophizing and depression on the outcome variables were intact even when introducing “perceiving the biopsychosocial message” to the model. However, “perceiving the biopsychosocial message” did add more explained variance. Thus, these analyses indicate that “not perceiving the biopsychosocial message” has a small, unique predictive value but did not mediate the relationship between catastrophizing respiratory depression and disability at follow-up or treatment satisfaction at 6 months.
Multiple Regression Analysis to Explore Mediation of Receiving the Biopsychosocial Message in the Relationship Between Pretest Catastrophizing and Disability at 6-Month Follow-upa
Multiple Regression Analysis to Explore Mediation of Receiving the Biopsychosocial Message in the Relationship Between Pretest Depression and Disability at 6-Month Follow-upa
Multiple Regression Analysis to Explore Mediation of Receiving the Biopsychosocial Message in the Relationship Between Pretest Catastrophizing and Treatment Satisfaction at 6-Month Follow-upa
Multiple Regression Analysis to Explore Mediation of Receiving the Biopsychosocial Message in the Relationship Between Pretest Depression and Treatment Satisfaction at 6-Month Follow-upa
Discussion
This study sought to examine whether patient characteristics such as pain catastrophizing and depressive mood are related to perceptions of “satisfaction with physical examination,” “back pain is benign message,” and “stay active message.” Furthermore, the aim was to investigate whether variations in perceptions of this biopsychosocial message constitute a risk factor or is a mediator for treatment outcome and treatment satisfaction. The aim was also to study if physical therapists' treatment orientations influence patient perceptions of the biopsychosocial message. The study showed that a vast majority of the participants (87%) perceived the message and only a small minority (13%) did not. It also showed that baseline levels of catastrophizing and depression were significantly correlated with whether the patient was “not perceiving the biopsychosocial message,” and “not perceiving the biopsychosocial message” also was significantly related to treatment satisfaction and disability at 6-month follow-up. The results also indicate that participants who were less satisfied with their treatment results after 6 months had not perceived the message that back pain is benign, that they should stay as active as possible, and that they should progressively increase their activity level. They also were less satisfied with the physical examination their physical therapist conducted, even when controlled for pretreatment pain intensity. Furthermore, participants who did not perceive this biopsychosocial message were at risk of higher disability 6 months after the start of treatment. “Not perceiving the biopsychosocial message” was not a mediator for treatment outcome and treatment satisfaction. The results also showed that the physical therapists' treatment orientation was not related to whether the participants perceived the biopsychosocial message. There are several possible explanations for these results, and they can be explained from a patient's perspective or from a therapist's perspective.
First, when we look at the fact that pain catastrophizing and depression were correlated to perceiving the message from a patient's perspective, we have to remember that patients with a high level of catastrophizing tend to have fear and avoidance beliefs, cognitions in terms of harm, and negative emotions such as fear of injury.33 They often display distress, and taken together, this often leads to their own model of a serious disease.33–35 The message of back pain being benign and to stay active may not be in concordance with their own view of their back problems. Moreover, these patients often display selective attention, with their main interest being attention to pain and pain-related information.36 Distress, maladaptive cognitions and beliefs, negative emotions, and selective attention combined may simply lead to them not “hearing” the message that is contradictory to their own beliefs.
Another explanation would be the influence of recall bias. We attempted to minimize the influence of recall bias by providing the alternative “do not remember” as opposed to just “yes” and “no.” There is, however, no way of knowing the extent of it. We have no reason to assume that recall bias in this study was disproportional to other studies.
From a therapist's perspective, we cannot rule out that a number of participants with high catastrophizing and low mood may not have been given the message that back pain is benign and to stay active. As mentioned previously, these patients often are distressed and have a behavioral coping strategy aimed at eliciting emotional or tangible support from others (ie, health care providers). This strategy positively reinforces pain and illness behaviors and undermines successful adaptation to pain.37 The physical therapists might have reacted defensively on this coping strategy by not conveying the message and thereby reinforcing the patient's own model of a serious disease. Yet, the fact that physical therapists' treatment orientation did not differ between the participants perceiving the biopsychosocial message and those not perceiving the message indicates that patient characteristics might be more important for perceiving the message than physical therapist characteristics.
Patients with high catastrophizing and those with a lower mood were to a greater extent not satisfied with the physical examination performed by the physical therapists. This finding also can be understood from both a patient's and a physical therapist's point of view. From a patient's view, one can assume that the explanation of the back pain given by the physical therapist after the examination was not in line with what they expected to hear. Moreover, a message that back pain is benign and to stay active is controversial when the patient's beliefs and cognitions are oriented toward a more serious back problem. The explanation and message were not what they expected, and a natural conclusion for the patients would be that the physical therapist did not examine them properly. A less plausible explanation would be that the physical therapists performed a different physical examination on patients with high catastrophizing than on other patients. There is no reason to believe this is the case.
The fact that not perceiving the biopsychosocial message is an independent risk factor for lesser function 6 months after treatment start may not be so surprising. Because these patients did not perceive the biopsychosocial message, the patients' own maladaptive beliefs and cognitions were not challenged. It is likely that they persisted in the coping strategy of avoiding activity, leading to greater disability.
The effect on treatment satisfaction can be twofold. First, a recent review on treatment satisfaction with physical therapy38 showed that aspects of communication that patients value in physical therapy are the ability to provide a helpful explanation about the patient's condition, the ability to give prognostic information, and the ability to explain the patient's role in the treatment process. However, when the information and explanation patients receive are contradictory to the patients' own beliefs, it is not surprisingly that this results in lower treatment satisfaction. Second, the review showed, unexpectedly, that the actual treatment outcome was infrequently and inconsistently a determinant of satisfaction with physical therapy care. It is believed that patient satisfaction with physical therapy care is determined more by interactions with the therapist and the process of care than by the outcome of treatment.38 There is evidence that the relationship between overall patient satisfaction with care and satisfaction with clinical outcome is weak.38 In our study, patients with high catastrophizing either received a message from the physical therapist that was opposite to their beliefs, or the physical therapist chose not to convey the message, thereby reinforcing the patients' maladaptive coping. Either way will probably not enhance patient satisfaction.
The fact that only a small minority of participants (13%) did not perceive the message should be seen as positive. This percentage is in line with the percentage of patients who end up in chronicity and is probably yet another marker for that. Yet, this is an important minority who should be recognized in clinical practice.
To our knowledge, no earlier studies have investigated the relationship among patient characteristics, perception of information, and treatment outcome. Although the correlations are rather low, the results of this study seem to contribute with another piece of the puzzle, explaining how patient characteristics influence the course of events during care, leading to treatment outcome. The clinical implication of the results of this study is that physical therapists cannot rely on the assumption that patients with high catastrophizing levels or lower mood hear or remember a biopsychosocial message or information when it is conveyed to them. Interpretation of a message or information is dependent on existing beliefs, cognitions, and emotional state. To make sure the message is heard and interpreted in the way it is meant, the message or information should be very clear and unmistakable. After this, the therapist should ask the patient to repeat the message in his or her own words. This approach would give an indication of whether the message was perceived and in which way it was interpreted by the patient. Misinterpretation of a message contradictory to one's beliefs is probably not uncommon.
The strength of this study is the large number of patients, which makes the results robust. However, there are some limitations as well. One weakness of this study is that we do not have any data on actual practice behavior and, therefore, do not know which message or care was given. As mentioned earlier, it might be that therapists changed their behavior or care according to how they perceived the characteristics of the patient in front of them.
In summary, this study shows the physical therapists' treatment orientation was not related to whether the patients perceived the biopsychosocial message. Yet, maladaptive cognitions and negative emotions such as patients' baseline levels of catastrophizing and mood are related to patients' perceptions of the message delivered by their physical therapists. The study also shows that the way information is perceived by patients influences treatment outcome and treatment satisfaction, and not perceiving the message is an independent risk factor for greater disability and lower treatment satisfaction at 6 months into treatment, even when investigators have controlled for pretreatment pain intensity. Physical therapists are advised to check that patients with higher levels of catastrophizing or lower mood perceive and interpret a biopsychosocial message in the way it is intended.
Footnotes
Both authors provided concept/idea/research design and data analysis. Dr Overmeer provided data collection and project management.
This study received funding from the Swedish Research Council for Health, Working Life and Welfare.
The Regional Ethical Committee at Uppsala University approved the study.
- Received December 7, 2014.
- Accepted July 5, 2015.
- © 2016 American Physical Therapy Association