Assessment of Psychosocial Risk Factors of Chronicity—“Yellow Flags”



Assessment of Psychosocial Risk Factors of Chronicity—“Yellow Flags”


Craig Liebenson

Steven Yeomans

Richard Nicol






Introduction


Few will argue that the concept of acute low back pain (LBP) resolving in 4 to 6 weeks for 75% to 90% of sufferers is totally inaccurate.1,2,3,4 This unrealistic, overly optimistic picture, in the past, had led to false confidence and poor outcomes, because a passive management philosophy involving symptomatic approaches (bed rest and medication) or a nonmanagement approach of leaving it alone to let “nature run its course” simply did not pan out. There are two problems with this perspective. First, the view that most acute episodes resolve quickly and completely is disputed by a number of studies of primary care patients.5,6,7 Second, there has been a continuous growing body of evidence that it is much more cost-effective to prevent chronicity in those at risk for it rather than waiting to treat only those in whom it becomes fully apparent.

Von Korff et al demonstrated in a nonoccupational setting that after 1 month only 30% of neck pain and LBP patients achieved pain-free status, and after 1 year 50% still reported recurrent or persistent pain.7 Most recent studies show that the majority of acute episodes tend to improve rapidly, although not completely, and then run an intermittent chronic course with less severe “flare-ups.” The original episode frequently lasts for as long as 3 months—not 4 to 6 weeks—before it can even be said to have remitted.6,8 The “flare-ups,” which are predictable in the majority of cases 1 year later, are mild to moderately activity-limiting and painful and lead to general dissatisfaction with the symptoms.6,7,8 Thus, back problems typically run recurrent or chronic remitting courses with occasional acute self-limiting episodes.

Even though only a small percentage (7%) of individuals with acute LBP have chronic unremitting pain and disability, this group accounts for the majority of the costs.9,10 More specifically, 7.4% of patients account for approximately 75% of all the costs and 85% of the disability days.10,11 Thus, identifying potential risk factors for acute pain becoming chronic has become the “holy grail” of LBP research.12,13

The prevalence of acute LBP becoming chronic has remained similar as noted in more recent studies.14,15 In a 2012 meta-analysis that included 33 cohorts consisting of 11,166 participants, the difference between the acute LBP sufferers resolving and those becoming chronic was distinguishable after 6 weeks of tracking pain and disability levels, with the majority improving markedly in the first 6 weeks. In a review of a book chapter entitled “From Acute to Chronic Back Pain: Risk Factors, Mechanisms and Clinical Implications,” published in the British Journal of Anaesthesia, the following was reported: “A triumvirate can be defined as ‘any association of three in office or authority’: the political alliance of Gaius Julius Caesar, Marcus Licinius Crassus, and Gnaeus Pompeius Magnus being perhaps the most famous.”16 They go on to make the analogy of the biopsychosocial (BPS) model as being perhaps the most famous in pain medicine, stating that few would argue that these three components are inseparable when it comes to discussing the basis of modern pain management and its endorsement by the Faculty of Pain Medicine, the British Pain Society, and the International Association for the Study of Pain. Indeed, few would challenge this concept, because the evidence supporting the transition of acute to chronic pain has grown significantly since this model was first proposed in the mid- to late 1970s.17,18

In support of the BPS model, modifications were proposed 25 years later by Borrel-Carrio et al, who recommended that three clarifications be considered: “1) the relationship between mental and physical aspects of health is complex—subjective experience depends on but is not reducible to laws of physiology; 2) models of circular causality must be tempered by linear approximations when considering treatment options; and 3) promoting a more participatory clinician-patient relationship is in keeping with current Western cultural tendencies, but may not be universally accepted.”19 Further, they proposed the following “pillars” for the BPS-oriented clinical practice: “(1) self-awareness; (2) active cultivation of trust; (3) an emotional style characterized by empathic curiosity; (4) self-calibration as a way to reduce bias; (5) educating the emotions to assist with diagnosis and forming therapeutic relationships; (6) using informed intuition; and (7) communicating clinical evidence to foster dialogue, not just the mechanical application of protocol.”


Risk Factors of Chronicity

Assessment of spine patients has traditionally focused on finding the physical cause of the pain. Imaging techniques have figured prominently in this endeavor. Unfortunately, this has been an inefficient use of resources because of the poor specificity of this expensive
screening approach.20,21,22,23,24,25,26 Clinical scientists have summarized that the following measurable outcomes are representative criteria of patient recovery: pain, function (disability), well-being, work status, and satisfaction.13,27 According to Pinchus et al, the risk of long-term LBP-related activity limitations (disability) and work loss (participation) arises from four main sources that interact with each other (Table 7.1).28 Individual factors have been referred to as psychosocial “yellow flags.”29 “Yellow flags” are analogous to the concept of “red flags” in that they both influence the management and prognosis of the patient. Whereas “red flags” are indications for biomedical laboratory or imaging investigations and possibly specialist referral, “yellow flags” are indications for investigating the cognitive, affective, and behavioral aspects of LBP.








Table 7.1 Four Main Factors That Influence Chronic Disability









  • Individual



  • Treatment provider



  • Compensation or health care system



  • Workplace or home environment


From Pincus T, Burton AK, Vogel S, Field AP. A systematic review of psychological factors as predictors of chronicity/disability in prospective cohorts of low back pain. Spine (Phila Pa 1976). 2002;27(5):E109-E120.


Most yellow flags pertain to individual or work-related factors, yet the effect the treatment provider has on outcome is also important.30,31 Reis et al evaluated both the patients’ and clinicians’ perceptions of worry, coping, limitations, expectation of pain relief, and pain interference. When evaluated individually, both patients’ and clinicians’ perceptions were found to predict outcome at 2, 4, 8, and 12 months. Because many patient characteristics are stable and thus nonresponsive to change (such as premorbidity, high levels of depression, and catastrophizing), other risk factors that may be amenable to change such as patients’ or clinicians’ perceptions and expectations should receive greater attention.

The influence of perception on outcome is highlighted by Kalauokalani et al’s study of 135 patients with chronic LBP who were allocated randomly to receive either massage or acupuncture.32 Patient expectations regarding the potential helpfulness of each treatment correlated more than other variables with subsequent functional outcomes as assessed at 10 weeks using the modified Roland score.

Similarly, in a systematic review of the literature, 23 studies covering 18 different cohorts identified 16 psychosocial factors in three domains: social and socio-occupational, psychological and cognitive, and behavioral.33 They reported that depression, psychological distress, passive coping strategies, and fear-avoidance beliefs were sometimes found to be independently linked with poor outcome, whereas most social and socio-occupational factors were not. Because of biomedical confounding factors, they reported the predictive ability of a patient’s self-perceived general health at baseline was difficult to interpret. Again, the patient’s or health care provider’s initial perceived risk of persistent LBP was, reportedly, most consistently linked with the actual outcome.


Phase of Care

Because the majority of acute patients have a very good prognosis, overly aggressive early management is an inefficient use of limited health care resources. However, the same cannot be said for patients who are still symptomatic in the subacute phase. Thus, the subacute phase, beginning at the end of the first month, has now been recognized as a critical period when more aggressive management strategies can potentially have a large impact on preventing chronic pain and disability and thus reducing costs.34

Frank has presented the concept of the “number needed to treat” to determine the cutoff for when it would be more efficient and cost-effective to substitute more aggressive treatment with less aggressive approach. He states that it is possible to show that “the number (of individuals) needed to treat” to prevent a single case from passing into chronicity at 6 months off work declines swiftly over the first month and then remains rather stable.34

According to Frank, there are three distinct stages in terms of risk of an acute episode becoming chronic (Fig. 7.1).34 In the acute stage (first 4 weeks), the risk of chronicity is low. In the subacute stage (weeks 4-12), the risk is high “ipso facto,” and the survival curve suggests aggressive treatment will be cost-effective here. In the chronic stage (after 12 weeks), recovery halts.

This is borne out by a recent study of workers’ compensation claimants, in which it was found that the most robust predictors of future status (recurrence likelihood) were preadmission health care visits and earlier back-related claims.35


Psychosocial and Other Factors

Psychological variables have been demonstrated to account for 26% of self-reported pain and 36% of self-reported disability (Roland-Morris scale).36 Six separate review papers of varying methodological rigor all agreed that psychological characteristics such as coping strategies, self-efficacy beliefs, fear-avoidance behavior, and distress are examples of relevant factors than can be identified.28,37,38,39,40 Thus, the presence of psychosocial
“yellow flags” indicative of a decreased likelihood of recovery has been proposed as a technique for early identification and matched appropriate management of those with a poorer prognosis.29,41,42,43,44






Figure 7.1 Three-phase model of low back pain natural history. From Frank JW, Kerr MS, Brooker AS, et al. Disability resulting from occupational low back pain. Part 2: what do we know about secondary prevention? Spine. 1996;21:2918-2929.

Gatchel et al generated a statistical algorithm to identify acute patients at risk for chronic pain/disability.45 By including factors such as gender, self-reported pain, and disability scores; scores on Scale 3 (hysteria) of the Minnesota Multiphasic Personality Inventory (MMPI); and workers’ compensation and personal injury status, 90.7% of cases could be correctly classified as high or low risk for chronic pain/disability. The hysteria subscale of the MMPI had an odds ratio of 1.5 for predicting return to work. Thus, individuals with high hysteria scores on the MMPI are 1.5 times more likely to have chronic LBP than those without such scores. The large Boeing prospective trial also found this scale was predictive of future work-related injury.46 However, this is considered to be of minimal utility because it reflects personality, which is considered a trait measure that is not sensitive to change.28


Further validation of this model showed that a number of other factors also correlate with high risk.47 A less positive temperament identified with the Schedule for Nonadaptive and Adaptive Personality (SNAP), high workaholism (SNAP), an avoidant coping style, and an axis I (psychiatric) disorder were found to predict with 80.8% accuracy (80% sensitivity and 81.5% specificity) whether a person was at high or low risk for chronic LBP.47

Fransen et al also showed that early identification of risk factors can predict time off work because of back pain.48 This theoretically should lead to targeted interventions to those individuals at greatest risk for future disability. It was found that if workers, at the time they make their initial claim for workman’s compensation, report any of the following, the odds that they will still be receiving compensation 3 months later will be significantly increased:



  • Severe radiating lower limb pain


  • At least moderate physical disability (Oswestry)


  • Psychological distress


  • The need to lift for at least three-fourths of the day


  • A workplace unable to provide light duties on return to work

The authors concluded: “Importantly, these determinants each retained significant associations with
chronic occupational back pain, even when statistical adjustments were made for age, gender, and the other significant individual, psychosocial, or workplace risk factors.”48

Schultz et al found that cognitive factors were the most predictive of time off work for LBP over a 3-month period.49 Cognitive beliefs relating to perceptions of current health, physical status, and expectations of recovery were most relevant. Another very important predictor was sciatica. The overall correct prediction rate was 77.6%.

Thomas and colleagues performed a prospective study that followed 5,000 asymptomatic individuals for 18 months and correlated premorbid and clinical factors with development of chronic LBP.50 Ten percent of these individuals had LBP, with 34% of them reporting persistent, disabling LBP at 1 week, 3 months, and 12 months after onset. The premorbid features that correlated with persistent, disabling LBP were sex (female), age (increasing), high psychosocial distress, below-average self-rated health, low level of physical activity, history of LBP, and job dissatisfaction. Each of these had a 2- to 5-fold effect on the odds of being associated with persistent symptoms.

The episode-specific factors that correlated with the development of persistent disabling LBP were the presence of widespread pain, long duration of symptoms before consultation, leg pain, and significant restrictions in spinal movement. Widespread pain was the most highly correlating item with an odds ratio of 6.4. The other factors were associated with a 2- to 5-fold increased chance of poor outcome. Only 6% of patients with a poor outcome were missed if a minimum of three factors were used to identify risk!!!

Shaw et al showed that low back disability was related to the following problem-solving approaches: problem avoidance, lack of positive problem-solving orientation, and impulsive decision making.51



Fear-Avoidance Beliefs

One of the major goals of care is to reduce activity intolerances associated with pain.44 Thus, the cognitive association of activity with pain or anticipation of pain is an important psychological construct.52,53,54,55,56 In fact, the belief that an activity will be painful has been shown to be more predictive of physical performance than purely nociceptive factors.57,58 Anxious patients predict pain sooner during the performance of physical tasks such as range of motion (ROM) or straight leg raise (SLR) tests.52,53,59 Council et al60 documented substantial correlations between pain expectancies and self-rated physical disability with the performance of simple motor tasks.13

It is important to distinguish those factors that are associated with chronic pain from those that predict it. For instance, Ciccione showed that depression, somatization, and current pain ratings combined to explain 34% of the variance in work disability in a chronic group.52 However, these factors explained only 8% of the variance in an acute sample! More significant is the finding that pain expectancies accounted for 33% of the variance in acute subjects (P < 0.001).52 Fritz et al have also confirmed that initial fear-avoidance beliefs were significant predictors of subacute status at 4 weeks independent of pain intensity, physical impairment, disability, or therapy received.61,62 Thus, fear-avoidance beliefs such as pain expectancies begin in acute pain and precede other psychosocial problems that develop as acute pain becomes chronic.

Linton and colleagues found that fear-avoidance beliefs were even prospectively related to the development of acute pain and dysfunction in asymptomatic individuals.63 Those with scores above the median had twice the risk for acute LBP (odds ratio 2.4). Catastrophizing was also evaluated, but its predictive power was more limited (odds ratio 1.5).

Although numerous studies demonstrate the effectiveness of cognitive-behavioral strategies,54,64,65,66 simpler reactivation approaches may be all that is needed. Mannion reported that three different active care approaches, none of which consisted of psychological or cognitive-behavioral approaches, all improved psychological variables related to self-report of pain and disability.36

Abnormal illness behavior contributes to a slower or inadequate recovery.67,68 Patients who equate hurt with harm develop a disabling form of thinking. They develop fear-avoidance behavior that promotes deconditioning (see Fig. 1.9).44,69 It is important to identify the patient who is fearful and avoid encouraging them to take on a “sick role.” According to Troup70: “If fear of pain persists, unless it is specifically recognized and treated, it leads inexorably to pain-avoidance and thence to disuse.”



Cervical and Upper Quarter Risk Factors

Tenenbaum et al have shown that whiplash-associated disorders classification II patients with neuropsychological problems have a worse prognosis over a 3-year follow-up period.71 Confidence in one’s ability to work after 2 years is correlated with 3-year outcome (P < 0.0001) for neck pain caused by whiplash.72 Carroll et al have demonstrated that high levels of passive coping are associated with disabling cervical or lumbar spine pain.73 These patients have difficulty functioning with pain, are less likely to take responsibility for care, and have lower self-rated health.

Macfarlane et al performed a prospective study aimed at determining the relative contributions of psychological and work-related factors in the onset of forearm pain74; 1,953 individuals were followed up for 1 year, and 105 (8.3%) developed forearm pain. Increased risks for forearm pain were associated with a number of factors. Psychological distress had a relative risk (RR) of 2.4 (95% confidence interval 1.5-3.8). Multiple areas of pain had an RR of 1.7 (95% confidence interval 0.95-3.0). Repetitive movements of the arm had an RR of 4.1 (95% confidence interval 1.7-10), whereas that of the wrist was 3.4 (95% confidence interval 1.3-8.7). Dissatisfaction with a colleague or supervisor support had an RR of 4.7 (95% confidence interval 2.2-10).

Hill et al recently reported that the most important factors related to persistent neck pain were age,75,76,77 concomitant LBP, and regular cycling.75 Age was, by far, the most significant factor. Both age over 40 and concomitant LBP were also found to be accurate predictors by Hoving et al.76 Other authors have also reported that concomitant LBP was a significant prognostic factor for chronic neck pain.78,79,80,81








Table 7.2 Risk Factors for Prolonged Cervical and Upper Quarter Pain

















1 month




  • Upper extremity comorbidity (RR 1.58)



  • Pain severity (RR 1.45)



  • Ergonomic risk factor (RR 1.07)



  • Low job support (RR 1.03)



  • Catastrophizing pain coping style (RR 1.54)


3 months




  • Pain severity (RR 10.46)



  • Job stress (RR 1.20)



  • Catastrophizing pain coping style (RR 1.98)


12 months




  • Number of pain treatment episodes (RR 1.77)



  • Past recommendations for surgery (RR 6.43)



  • Catastrophizing pain coping style (RR 1.87)


RR, relative risk.


From Feuerstein M, Huang GD, Haufler AJ, Miller JK. Development of a screen for predicting clinical outcomes in patients with work-related upper extremity disorders. J Occup Environ Med. 2000;42(7):749-761.









Table 7.3 The Sensitivity and Specificity of Cervical and Upper Quarter Pain Predictors





















Duration (months)


Sensitivity (%)


Specificity (%)


1


77.4


71.8


3


80.6


82.4


12


80.6


83.3


From Feuerstein M, Huang GD, Haufler AJ, Miller JK. Development of a screen for predicting clinical outcomes in patients with work-related upper extremity disorders. J Occup Environ Med. 2000;42(7):749-761.


Feuerstein82 followed acute (<6 weeks from onset) cervical and upper quarter pain patients for up to 1 year to ascertain what factors were predictive of 1-, 3-, and 12-month outcomes. The findings are summarized in Tables 7.2 and 7.3.


Assessment

Linton reviewed psychological risk factors in back and neck pain with the objective to summarize current knowledge concerning the role psychological factors play in the cause and development of back and neck pain.41 In doing so, 913 potentially relevant articles were located, with 37 studies consisting of only those with prospective designs to ensure quality. The review procedure resulted in the reporting of the main predictor variables and the outcome criteria. If a statistically significant relation was determined, a plus (+) or minus (−) was used to indicate a positive or negative association, respectively. If no statistically significant relationship was found, a zero (0) was used. The conclusions include a grading system similar to that used for meta-analysis review and guidelines preparation.83 These grades include the following:

Level A: Evidence is supported by two or more good-quality prospective studies

Level B: Evidence is supported by at least one good-quality prospective study

Level C: Inconclusive data exist

Level D: No studies were found to meet the criteria utilized

Table 7.4 offers a summary of the conclusions drawn from this review of prospective studies.

What follows is a list of those specific risk factors—called “yellow flags”—for acute LBP becoming chronic. These have been identified primarily from an assortment of prospective longitudinal studies. Few cross-sectional studies were used as sources for the “yellow flags.” They are divided into those related to
symptoms, examination, and psychosocial, functional, and work-related factors (Table 7.5). “Yellow flags” are primarily subjective and have a significant psychosocial predominance. Whereas “red flags” such as cauda equina syndrome, cancer, fracture, and infection require urgent attention, further testing, and possibly specialist referral, “yellow flags” only require a shift in the focus of care. These risk factors have been reported to predict future chronic pain or disability in the United States,5,84,85,86 New Zealand,29,41,43 and England.43,87

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Apr 17, 2020 | Posted by in PHYSICAL MEDICINE & REHABILITATION | Comments Off on Assessment of Psychosocial Risk Factors of Chronicity—“Yellow Flags”

Full access? Get Clinical Tree

Get Clinical Tree app for offline access