Time since diagnosis, treatment pathways and current pain status: a retrospective assessment in a back pain population

Abstract

Objectives:

The purpose of this study is to report on long-term treatment patterns in a back pain population and to consider whether these may be considered as reflective of under-treatment of pain or poor pain control.

Methods:

Data are from a commissioned recontact SELECT survey of persons who had previously reported experiencing pain in the 2010 National Health and Wellness Survey. The analysis covers five countries: the UK, France, Spain, Germany and Italy. It is restricted to patients who reported moderate or severe pain in the last 3 months and who were judged to be experiencing back pain from the pain conditions reported. A total of 1018 subjects met the back pain criteria. Of these, 738 had their back pain initially diagnosed by a physician. It is this latter group that is the focus for the analysis. Subjects were further allocated to categories to assess potential pain chronicity and the intensity and self-limiting aspects of pain experience.

Results:

Irrespective of the pain classification used, 56.10% of subjects in this back pain population report their first physician diagnosis of pain as occurring at least 6 years ago. At the same time 36.81% report experiencing severe pain in the last 3 months and 63.69% moderate pain. Overall, 53.11% of respondents are assessed as experiencing probable chronic pain and 24.12% as possible chronic pain. Among those experiencing probable chronic pain, 61.48% were diagnosed with back pain more than 6 years ago. In addition, 43.22% of respondents found the impact of pain on daily activities to be severely limiting and 22.63% found it to be moderately limiting. The majority of subjects, whether they were diagnosed and treated initially by primary care or another physician specialty, apparently experience no change in their treatment regimen. The absence of treatment change is 69.03% in the case of those experiencing severe pain and 84.89% in the case of those with moderate pain (p < 0.05).

Limitations:

The NHWS is an internet survey, as a self-report there is no separate validation of pain experience reported; as subjects are asked to recall pain and treatment experience there is the possibility of recall error and bias.

Conclusions:

Two features of this analysis stand out: (i) time since first diagnosis of pain for those currently experiencing severe or moderate back pain and (ii) the ‘inertia’ that appears to characterize long-term treatment decisions. Irrespective of current pain status, 70% or more of those experiencing current severe or moderate pain report no change from their initial treating physician or no change in prescription once referred to a pain specialist. This occurs over an extended time frame. Whether this should be interpreted as a failure in pain management is a moot point. The fact remains, however, that despite time since first diagnosis subjects not only currently experience severe or moderate pain but a substantial proportion may be expected to continue.

Keywords::

• Diagnosis
• Opioids
• Pain duration
• Pain severity
• Treatment pathways

Introduction

A recurring theme in the chronic pain literature – notably in respect of back pain – is the inadequacy of pain management interventions. Even with a growing acceptance of the biopsychosocial model of pain management, pain management is seen as poorly conceived and infrequently applied1,2. This has been demonstrated in both in-patient care (to include emergency rooms and intensive care) as well as in general practice. Under-treatment is just as widespread in the US as it is in Europe, with the situation exacerbated in the US by the virtual absence of multidisciplinary pain management programs3. Indeed, the barriers to effective pain management, in particular barriers to the implementation of a multidisciplinary pain program, are well known4. These include health providers’ lack training or the knowledge to manage pain effectively, misconceptions regarding pain medication utilization, the lack of support by healthcare organizations for multidisciplinary pain programs, the reluctance of patients to report fully their pain experience and a lack of guidance from professional groups in the multidisciplinary management of chronic pain. The continuing emphasis in the literature on the under-treatment of pain is echoed in the belief that effective pain management is a fundamental human right2. This is echoed in the Declaration of Montreal which emphasizes, once again, the inadequacies of pain management5.

At the same time the presence of widespread and chronic pain may be predictive of other conditions. Ohayon and Schatzberg point out that the presence of chronic pain may be predictive of depressive morbidity6. In a more recent study covering the UK, France, Spain, Germany and the UK, persons reporting chronic pain in the last 30 days also reported one or more of: sleep difficulties (63.65%), insomnia (37.27%), depression (25.21%), generalized anxiety disorder (14.21%) and social anxiety disorder (9.52%)7. Among those reporting severe pain the prevalence estimates were: sleep difficulties (64.50%), insomnia (47.42%), depression (35.3.38%), generalized anxiety disorder (18.69%) and social anxiety disorder (12.15%). Of particular note is the presence of emotional disorders and maladaptive cognitions such as depression which can interact with functional deficits and physical deconditioning to modulate symptom reporting and ongoing disability1. Hence the continuing emphasis in the pain assessment literature on the predictive role of risk factors such as psychiatric comorbidities (the so called ‘yellow flags’) and the introduction of psychosocial variables into instruments designed to assess the likelihood of continuing chronic pain8–12.

Back pain, in particular low back pain, is a common problem with over 70% of the population experiencing an episode at some time, but with a significant percentage experiencing recurrences and even continuous symptoms13. Even so, a recent assessment by Oostrom et al., reporting on the Doetinchem cohort study, emphasizes that the long-term course of chronic low back pain is relatively unknown14. Few studies go beyond 12 months in tracking the course of pain and virtually nothing is known about the long-term and dynamic nature of low back pain. Reporting on the results of a 10 year cohort study the authors found that over one in three reported long standing back pain and approximately 5% had persistent back pain over 10 years.

The research question considered here is whether or not treatment patterns, the duration of pain reported and the continuing experience of pain reflect under-treatment or poor pain control in the long-term management of chronic back pain. This study, in focusing on the long-term course of chronic back pain, builds upon previous studies that suggest that the long-term experience of pain is common and claims that patients often experience poor pain management and a failure by treating physicians to appreciate the complexities of pain experience.

Methods

Data are from a commissioned recontact SELECT survey of persons who had previously reported experiencing pain in the 2010 National Health and Wellness Survey. The NHWS is a syndicated, annual and biannual, internet-based, cross-sectional survey of the healthcare attitudes, behaviors, and characteristics of the adult population. It is undertaken in the US, UK, France, Spain, Germany, Italy, urban China, Japan, Brazil, and Russia (www.chsinternational.com). Since its inception in 1998, over 1 million survey responses across 165 conditions have been collected. The NHWS has been used to support a number of descriptive and multivariate studies to assess the burden of pain in Western Europe7,15–20.

The present SELECT analysis covers the UK, France, Spain, Germany and Italy. The analysis is restricted to those who reported moderate or severe pain in the last 3 months and who were judged to be experiencing back pain from the pain conditions reported. A random sample of those reporting pain experience in the 2010 NHWS was selected with a target number identified who reported pain in the last 3 months. A further filter included only those who were assessed as currently experiencing back pain. This was defined as one or more of the following conditions checked from a pre-defined list of pain conditions: upper back problems, lower back problems, neck problems or back/neck/spinal surgery or medical procedure. Respondents were then asked if their back pain was mild, moderate or severe: if mild then they were excluded from the survey.

The SELECT study captures the following aspects of treatment experience:

• Severity and characteristics of current pain experience

• Initial diagnosis and physician making diagnosis

• Time since initial diagnosis

• Pain quality

• Referral patterns

• Outcome of referrals

• Prescription changes

• Current treatment by a physician

together with responses to the Graded Chronic Pain Scale (GCPS) and the PHQ-9 depression instrument21,22.

A total of 1018 subjects met the back pain criteria. Of these, 738 reported that they had their back pain initially diagnosed by a physician. It is this latter group that is the focus for the present analysis.

Treatment pathways

The SELECT study allows us to identify five steps in treatment experience. These are:

• Initial diagnosis: type of physician who initially diagnosed back pain (general practitioner vs. other)

• Referral pattern: whether or not the respondent was referred to a pain specialist

• Therapy change: if referred to a pain specialist whether or not a change in prescription therapy occurred

• Influence of therapy on pain relief: if a change in therapy occurred whether this resulted in (i) an improvement; (ii) no change; or (iii) a worsening of symptoms

• Current pain status

The anchor is current pain status with treatment patterns separately identified for:

• Persons currently reporting severe pain initially diagnosed by a primary care physician

• Persons currently reporting severe pain initially diagnosed by a non-primary-care physician

• Persons currently reporting moderate pain initially diagnosed by a primary care physician

• Persons currently reporting moderate pain initially diagnosed by a non-primary-care physician

Current pain experience

As well as categorizing respondents reporting severe or moderate pain, The SELECT study identifies two complementary definitions of current pain experience. These are: (i) the application of a prognostic risk score (PRS) to assess the likelihood that the subject is experiencing possible or probable chronic pain11,12,23 and (ii) the application of the Graded Chronic Pain Scale (GCPS) to categorize the intensity and the self-limiting impact of pain experience21. These measures of pain involvement form the reference points for the profile of current pain experience.

The PRS is a validated measure of a respondent’s likelihood of experiencing chronic pain. It was developed to overcome a number of methodological issues in defining chronic pain in terms of subjects’ retrospective assessment of the time they have experienced pain. The PRS defines chronic pain in terms of probability that a subject is presently experiencing chronic pain and will continue to experience chronic pain. The categories are:

• Probable chronic pain: 80% chance or better of future clinically significant pain

• Possible chronic pain: 50% chance or better of future clinically significant pain

• Intermediate risk of chronic pain

• Low risk of chronic pain

The PRS combines items from the GCPS with the depression component of the Patient Health Questionnaire (PHQ-9). The PHQ-9 categories range from minimal level of depression severity (score 1–4) to severe depression (score 20–27)22. Items from the two scales are weighted to produce an overall PRS score.

The GCPS was developed to provide a basis for assessing the severity of chronic or recurrent pain for use in general population surveys and primary care settings21. The GCPS measures an underlying severity continuum defined by pain intensity and interference with daily activities. It embodies a hierarchical structure where lower levels of severity are differentiated by pain intensity and higher levels by interference with activities. The GCPS recognizes four categories of pain experience: Grade 1 – low intensity; Grade 2 – high intensity; Grade 3 – moderately limiting; and Grade 4 – severely limiting.

Current pain scenarios

In addition, the SELECT study allows respondents to describe the course of their pain. Respondents are asked to match pain experienced to one of four courses of pain scenarios (which are described as pain sequence pictograms in the survey questionnaire). The pictogram categories are:

• Persistent pain with slight fluctuations

• Persistent pain with pain attacks

• Pain attacks without pain between them

• Pain attacks with pain between them

Results

Current severity and frequency of pain

Reported severity and frequency of back pain experienced by those initially diagnosed by a physician are detailed in Table 1. Overall, 268 respondents reported experiencing severe pain (36.81%) and 470 moderate pain (63.69%). At the same time 70.52% of those with severe pain reported it on a daily basis; daily moderate pain is reported by only 39.57% of subjects (p < 0.05). Only 7.46% of those reporting severe pain experienced it once a week or less compared to 15.96% of those with moderate pain (p < 0.05).

Table 1. Back pain severity and frequency, respondents diagnosed by a physician.

Pain frequency	All respondents (%)	Severe pain (%)	Moderate pain (%)
Daily	375 (50.81)	189 (70.52)	186 (39.57)
4–6 time a week	119 (16.12)	31 (11.57)	88 (18.72)
2–3 times a week	149 (20.19)	28 (10.45)	121 (25.74)
Once a week or less	95 (12.87)	20 (7.46)	75 (15.96)
Total	738 (100.00)	268 (100.00)	470 (100.00)

Source: NHWS SELECT Pain Study 2010.

Time since diagnosis

Subjects were asked to indicate time since their back pain had been first diagnosed. The distribution of responses is given in Table 2 for those reporting an initial diagnosis by a physician. Overall, 56.10% of those with back pain reported an initial diagnosis at least 6 years ago; a figure that increases to 59.70% for those reporting severe back pain. The estimate is little different for those reporting moderate pain (54.04%). Irrespective of whether or not respondents currently experience severe or moderate pain, the time since first diagnosis is broadly similar with the majority reporting first diagnosis six or more years ago. The SELECT data do not allow us to assess whether or not pain experienced since first diagnosis has been continually experienced at its current severity level.

Table 2. Time since back pain first diagnosed, respondents diagnosed by a physician.

Time since first diagnosed	All subjects	Subjects reporting severe pain	Subjects reporting moderate pain
Under 3 years	183 (24.80)	60 (22.39)	123 (26.17)
3 to 5.9 years	141 (19.11)	48 (17.91)	93 (19.79)
6 years and over	414 (56.10)	160 (59.70)	254 (54.04)

Source: NHWS SELECT Pain Study 2010.

Prognostic risk score

For those assessed by the PRS as probably experiencing chronic pain (Table 3), 61.48% reported that it was at least 6 years since their pain was first diagnosed, with 79.59% reporting 3 years or more (with a corresponding estimate of 70.79% for those with possible chronic pain). Overall, 392 respondents (53.11%) are assessed as experiencing probable chronic pain and a further 178 (24.12%) as experiencing possible chronic pain (p < 0.05). Expressing these estimates somewhat differently, 241 out of 414 (58.21%) respondents who were diagnosed six or more years ago were assessed as probably experiencing chronic pain with a further 89 (21.50%) assessed as experiencing possible chronic pain.

Table 3. Time since back pain first diagnosed, subjects by prognostic risk score, respondents diagnosed by a physician.

Time since first diagnosed	Low risk	Intermediate risk	Possible chronic pain	Probable chronic pain
Under 3 years	10 (27.78)	41 (31.06)	52 (29.21)	80 (20.41)
3 to 5.9 years	5 (13.89)	28 (21.21)	37 (20.79)	71 (18.11)
6 years and over	21 (58.33)	63 (47.73)	89 (50.00)	241 (61.48)
Total	36 (100.00)	132 (100.00)	178 (100.00)	392 (100.00)

Source: NHWS SELECT Pain Study 2010.

Graded Chronic Pain Scale

A similar pattern emerges in the case of the GCPS (Table 4). The proportion of subjects reporting at least 6 years since their first diagnosis for pain increases with the severity of pain experienced with 61.13% for those with Grade 4 and 55.69% for Grade 3 pain. Overall, 319 (43.22%) respondents found their pain to be severely limiting and a further 167 (22.63%; p < 0.05) respondents found their pain to be moderately limiting. Of those who were diagnosed six or more years ago, 195 out of 414 (47.10%) find their current pain experience severely limiting and a further 22.46% moderately limiting.

Table 4. Time since back pain first diagnosed, subjects by graded chronic pain scale, respondents diagnosed by a physician.

Time since first diagnosed	Grade 1: low intensity	Grade 2: high intensity	Grade 3: moderately limiting	Grade 4: severely limiting
Under 3 years	32 (35.96)	47 (28.83)	41 (24.55)	63 (19.75)
3 to 5.9 years	20 (22.47)	27 (16.56)	33 (19.76)	61 (19.12)
6 years and over	37 (41.57)	89 (54.60)	93 (55.69)	195 (61.13)
Total	89 (100.00)	163 (100.00)	167 (100.00)	319 (100.00)

Source: NHWS SELECT Pain Study 2010.

Concordance in pain experience measures

Among those respondents whose back pain was initially diagnosed by a physician, 65.67% of those reporting severe pain were categorized by the GCPS as experiencing severely limiting pain with 30.43% of those with moderate pain falling in the same category (p < 0.05) (Table 5). Overall, 43.22% reported severely limiting pain (compared to 36.15% of all respondents) and 22.63% moderately limiting pain (p < 0.05).

Table 5. Graded chronic pain scale and prognostic risk score by back pain severity, respondents diagnosed by a physician.

Graded Chronic Pain Scale	All respondents (%)	Severe pain (%)	Moderate pain (%)
Grade 1: low intensity	89 (12.06)	6 (2.24)	83 (17.66)
Grade 2: high intensity	163 (22.09)	34 (12.69)	129 (27.45)
Grade 3: moderately limiting	167 (22.63)	52 (19.40)	115 (24.47)
Grade 4: severely limiting	319 (43.22)	176 (65.67)	143 (30.43)
Prognostic risk score
Low risk	36 (4.88)	4 (1.49)	32 (6.81)
Intermediate risk	132 (17.89)	13 (4.85)	119 (25.32)
Possible chronic pain (50%)	178 (24.12)	41 (15.30)	137 (29.15)
Probable chronic pain (80%)	392 (53.12)	210 (78.36)	182 (38.72)

Source: NHWS SELECT Pain Study 2010.

The relation between reported severe and moderate pain and the PRS is also detailed in Table 5. Of those with self-reported severe pain 78.36% were allocated to the probable chronic pain category and 15.30% to possible chronic pain (a total of 93.66%). In the moderate pain category 58.15% were assigned to the possible/probable chronic pain categories (p < 0.05).

Course of pain

Amongst those who reported severe pain 34.70% describe persistent pain with pain attacks and 31.34% as pain attacks with pain between them (corresponding estimates for the moderate pain group are 27.23% [p < 0.05] and 15.96% respectively [p < 0.05]) (Table 6). Combining the two persistent categories with those who reported pain attacks with pain between them yields an overall estimate of 78.18% of respondents experiencing continual pain. Persons reporting severe pain have a substantially higher percentage reporting continuous pain than those reporting moderate pain (86.19% vs. 73.62%; p < 0.05).

Table 6. Course of pain by back pain severity, respondents diagnosed by a physician.

Course of pain by pain severity	All respondents (%)	Severe pain (%)	Moderate Pain (%)
Persistent pain with slight fluctuations	197 (26.69)	54 (20.15)	143 (30.43)
Persistent pain with pain attacks	221 (29.95)	93 (34.70)	128 (27.23)
Pain attacks without pain between them	161 (21.82)	37 (13.81)	124 (26.38)
Pain attacks with pain between them	159 (21.54)	84 (31.34)	75 (15.96)

Source: NHWS SELECT Pain Study 2010.

Depression and pain

Given the role of the depression component of the Patient Health Questionnaire (PHQ-9) as an element of the PRS, together with the emphasis in the pain literature on the prevalence of depression in chronic pain populations, it is worth considering the allocation of respondents to PHQ-9 categories. Categories range from minimal level of depression severity (score 1–4) to severe depression (score 20–27). The distribution of respondents by PHQ-9 score is given in Table 7. Overall, 13.14% of respondents were assessed as having moderate to severe depression and 10.70% as having severe depression. The assigned distribution was, however, markedly different between those reporting severe as opposed to moderate pain. Among those reporting severe pain 37.31% reported moderately severe or severe depression compared to only 16.17% of those with moderate pain (p < 0.05).

Table 7. PHQ-9 depression status, respondents diagnosed by a physician.

PHQ-9 category	All respondents (%)	Severe pain (%)	Moderate pain (%)
Minimal	183 (24.80)	32 (11.94)	151 (32.13)
Mild	213 (28.86)	65 (24.25)	148 (31.49)
Moderate	166 (22.49)	71 (26.49)	95 (20.21)
Moderately severe	97 (13.14)	48 (17.91)	49 (10.43)
Severe	79 (10.70)	52 (19.40)	27 (5.74)

Source: NHWS SELECT Pain Study 2010.

Treatment patterns since initial pain diagnosis

A summary of treatment patterns reported is given in Table 8. Overall, fewer persons experiencing severe pain reported being initially diagnosed by a primary care physician compared to those experiencing moderate pain (46.64 vs. 59.79; p < 0.05) while a higher percentage of those currently experiencing severe pain report being referred to a pain specialist or another physician (60.07% vs. 42.34%; p < 0.05). At the same time, a higher percentage of those currently experiencing severe pain reported being referred and switched to opioids compared to those with moderate pain (27.99% vs. 9.15%; p < 0.05) although there is no significant difference between those who were referred yet reported no change in treatment by current severe or moderate pain status (29.10% vs. 27.23%); p > 0.05).

Table 8. Treatment experience by current back pain severity and frequency, persons diagnosed by a physician.

Treatment experience	All respondents n = 738 (%)	Severe pain n = 268 (%)	Moderate pain n = 470 (%)
Initial diagnosis
Primary care	406 (55.01)	125 (46.64)	281 (59.79)
Other	332 (44.99)	143 (53.36)	189 (40.21)
Referred
Yes	360 (48.78)	161 (60.07)	199 (42.34)
No	378 (51.22)	107 (39.93)	271 (57.66)
Therapy switching on referral
Yes: to opioids	118 (15.99)	75 (27.99)	43 (9.15)
Yes: to non-opioids	19 (2.57)	2 (0.75)	17 (3.62)
Yes: to no medication	17 (2.30)	6 (2.24)	11 (2.34)
No change in medication	206 (27.91)	78 (29.10)	128 (27.23)
Outcome of therapy switching
Improvement	93 (12.60)	49 (18.28)	44 (9.36)
No change (not sure)	17 (2.30)	8 (2.99)	9 (1.91)
Worse	44 (5.96)	26 (9.70)	18 (3.83)
Not switched or no change in treatment	584 (79.13)	185 (69.03)	399 (84.89)
Current treatment status: no treatment	146 (19.78)	19 (7.09)	127 (27.02)

Source: NHWS SELECT Pain Study 2010.

A higher percentage of those with current severe pain who were referred and who changed therapy reported an improvement compared to those reporting moderate pain (18.28% vs. 9.36%; p < 0.05) while a substantially higher percentage of those currently experiencing moderate pain report not receiving treatment compared to those with moderate pain (7.09% vs. 27.02%; p < 0.05). Overall, 79.13% of respondents reported either no referral or a referral with no change in therapy; among those reporting severe pain the figure is 69.03% and for those reporting moderate pain 84.89% (p < 0.05).

The impact of therapy switching on outcomes is detailed in Table 9. Overall, the majority of respondents who were referred indicated that there had been no change in therapy (57.22%) while persons currently experiencing severe pain had had a greater chance of medication switching than those with moderate pain (51.57% vs. 35.68%; p < 0.05). Taking all those who were referred, the most common switch had been to opioids (32.78%) with the likelihood of a switch to opioids for those with severe pain over twice that for moderate pain (46.58% vs. 21.61%; p < 0.05). Where an opioid switch had occurred the majority reported an improvement in pain experience (61.86%) while the difference between those reporting severe and moderate pain was not significantly (5% level) different.

Table 9. Impact of therapy switching by current back pain severity and frequency, persons diagnosed by a physician.

Treatment experience	All respondentsn = 360 (%)	Severe painn = 161 (%)	Moderate painn = 199 (%)
Therapy switching on referral
Yes: to opioids	118 (32.78)	75 (46.58)	43 (21.61)
Yes: to non-opioids	19 (5.28)	2 (1.24)	17 (8.54)
Yes: to no medication	17 (4.72)	6 (3.73)	11 (5.53)
No change in medication	206 (57.22)	78 (48.45)	128 (64.32)
Outcome of referral
Therapy change: improvement	93 (25.83)	49 (30.43)	44 (22.11)
Therapy change: worse	44 (12.22)	26 (16.15)	18 (9.05)
No change in therapy or unsure	223 (61.94)	86 (53.42)	137 (68.84)
Opioids: outcome of therapy switching	n = 118	n = 75	N = 43
Improvement	73 (61.86)	44 (58.67)	29 (67.44)
No change	12 (10.17)	7 (9.33)	5 (11.63)
Worse	33 (27.97)	24 (32.00)	9 (20.93)

Source: NHWS SELECT Pain Study 2010.

Discussion

In the general population estimates of the long-term experience of back pain are mixed. In follow-up studies, estimates of persistent back pain range from 30% to 40%24–26. However, as Oostrom et al. point out, these rates, which are substantially higher than reported in the Doetinchem cohort study, may reflect differences in definition and the number of pain conditions covered14. The present study, however, differs from previous ones in asking a randomly selected cohort of persons currently assessed as experiencing chronic back pain in the previous 3 months to report on (i) when their back pain was first diagnosed (and by whom) and (ii) whether or not they had ever been referred for treatment (and the consequent change in prescription treatment regimen and whether that improved their situation).

Although evidence would suggest that for a minority of the population the experience of back pain has been long term, the surprising feature of the present study is the relatively high proportion of those currently reporting back pain who indicated that they were diagnosed by a physician more than 6 years previously. For both severe and moderate pain groups, 56.10% of respondents reported that more than 6 years had elapsed since the initial diagnosis of back pain with a further 19.11% reporting it was between 3 and 6 years ago. At the same time, while there is no significant difference between time since first diagnosis for those reporting severe as opposed to moderate pain, a substantially higher percentage of those reporting severe pain indicate that it is experienced daily compared to those reporting moderate pain (70.52% vs. 39.57%; p < 0.05). While the PRS captures pain experience in the past 6 months, with emphasis on the past 3 months, it says nothing as to the time since first diagnosis of back pain. Given this, it is of considerable interest to assess the extent to which the PRS is associated with time since first diagnosis as a potential indicator not just of future risk but of previous chronic pain experience. The results presented suggest that the likelihood of probable chronic pain increases with time since first diagnosis with 61.48% of those assessed reporting 6 years or more since first diagnosis – this group comprises 32.66% of all respondents. This compares to only 50.00% of those possibly experiencing chronic pain (p < 0.05), noting that this group comprises only 12.06% of all respondents. Overall, four out of five respondents who were diagnosed six or more years ago are assessed as experiencing probable or possible severe or moderate chronic pain. The experience of possible or probable chronic pain would appear, on this analysis, to be endemic within the long-term pain population – a group that accounts for over 50% of the current back pain population. Given that the GCPS is a component of the PRS, care should be exercised in interpreting the results. Even so, as detailed, 43.22% of respondents find their pain experience over the last 6 months to be severely limiting with 26.42% of respondents diagnosed more than 6 years ago in the same category. Importantly, the majority of these respondents fall in the severe pain category. Of those reporting severe pain (with the majority reporting severe daily pain) 176 (65.67%) are assessed as experiencing severely limiting pain – which is in contrast to a much lower figure of 30.43% (p < 0.05) among those with moderate pain. As would be expected, a similar pattern is found for the PRS but in this case a much higher percentage of those with severe pain (78.36%) assessed as probable chronic pain candidates. Among those with moderate pain the figure is only 38.72% (p < 0.05). Indeed, virtually every respondent reporting severe pain is assigned to either the possible or probable chronic pain categories. The importance of this assignment to pain chronicity categories is also of interest given the course of pain described by respondents. Those in the severe pain category describe their pain as either persistent or as pain attacks with pain between them (86.19%) as opposed to those reporting moderate pain (73.62; p < 0.05).

From a pain management perspective, the question that arises is, given the number of years since first diagnosis of pain, with opportunities presented for effective interventions, why such a high percentage of respondents currently report (i) severe daily pain; (ii) the presence of probable chronic pain; and (iii) severely limiting pain? This does not mean that respondents initially diagnosed more than 6 years ago might not have experienced a reduction or remission in their pain symptoms in the intervening period or that other patients initially diagnosed at the same time might not have been managed effectively to resolve symptoms. However, the fact remains that a substantial proportion of a randomly selected population currently experiencing back pain appear not to have resolved their pain symptoms. Indeed, taking the biopsychosocial model of pain, we find that major depressive disorder, as assessed by the PHQ-9 instrument is also highly prevalent in this population – notably the prevalence of moderately severe and severe depression in those reporting severe pain. Again, a condition that appears unresolved and one which might have been exacerbated by a failure to resolve pain symptoms.

Although only exploratory, the treatment patterns and outcomes reported by respondents may suggest some answers. Overall, just over 50% of those initially diagnosed were diagnosed by a primary care physician. Among those currently reporting severe pain 53.36% were diagnosed by a non-primary-care physician compared to 59.79% of those with moderate pain (p < 0.05). When patterns of referral are considered the results indicate that while 42.34% of those with current moderate pain were referred to a pain specialist or other specialist for treatment some 60.07% of those with current severe pain were referred. At the same time, 48.35% of those initially diagnosed by a primary care physician were referred compared to 39.95% of those diagnosed by a non-primary-care physician. Unless it is argued that these rates of referral are too low, the patterns are consistent for those reporting current pain severity having a greater likelihood of referral.

The results of referral are, however, more interesting. A substantial proportion of all respondents reported being referred but with no change in prescriptions (57.22%), a figure which is substantially higher for those currently reporting moderate as opposed to severe pain (64.32% vs. 48.45%; p < 0.05). Even so, such a high percentage reporting a referral with no change in treatment seems unusual. At the same time, opioids are the treatment of choice where a referral results in a therapy change. Among those with current severe pain who are referred 46.58% are switched to opioids compared to only 21.61% currently reporting moderate pain. For those reporting severe pain the referral outcome is, therefore, either no change in therapy or a switch to opioids. For those currently reporting severe pain, virtually no one is switched to a non-opioid or to a no therapy option. As far as the outcome of therapy switching is concerned there is no significant difference between the two pain groups with just over two out of three respondents signaling an improvement in their pain status (severe pain 65.33%; moderate pain 70.97%). Finally, if the outcome of opioid switching is considered there is no significant difference between those with severe pain (58.67%) and those with moderate pain (67.44%) who claim an improvement in outcome.

The one characteristic that stands out in this summary of treatment pathways for those currently experiencing moderate or severe pain is what may be described as treatment ‘inertia’. The majority of subjects, whether they were diagnosed and treated initially by primary care or another physician specialty, apparently experience no change in either their choice of physician (i.e., are not referred to a pain specialist) or, if referred, experience no change in their prescription regimen. For those with severe pain, 69.03% are in this category and the figure is 84.89% for those with moderate pain. To this might be added a lack of improvement when therapy switching occurs, but this appears secondary to the absence of therapy change.

Whether this ‘inertia’ reflects decisions by the treating physician or a decision by the patient the question remains open as to whether it should be interpreted as indicative of poor pain management. It is certainly consistent with the views of professional bodies such as the IASP that the burden of pain is both under-appreciated and under-treated. The fact remains, however, that given time since first diagnosis a substantial proportion of patients, in fact the majority of those currently experiencing pain who were diagnosed six or more years ago, not only continue to experience severe or moderate pain but may be expected to continue to experience this as a possible or probable chronic condition.

Limitations

There are a number of limitations on the present study which need to be noted. First, as the NHWS (and SELECT follow-up) are internet-based there can be no separate validation of the reported pain and treatment experience reported. Second, as the SELECT study asks respondents to recall pain and treatment experience over a number of years there is the possibility of recall error and bias. Third, the analysis covers five countries and, as such, includes respondents who may be subject to different standards in access to and procedures in pain treatment and management. Future analyses utilizing a multivariate framework could accommodate country-specific effects through country-specific dummy variables. Weighting is not possible in the absence of national estimates of patients meeting the back pain definition. Even so, attempting to standardize for country effects begs the question of whether the degree of homogeneity in treatment patterns within countries is greater than that between countries. Fourth, the study is focused on pharmacological experience. It does not attempt to cover non-pharmacological pain interventions.

Conclusions

Two features of this analysis stand out: (i) time since first diagnosis of pain for those currently experiencing severe or moderate back pain and (ii) the ‘inertia’ that appears to characterize long-term treatment decisions. Irrespective of current pain status, 70% or more of those experiencing current severe or moderate pain report no change in their treatment regimen since first diagnosis by a physician, a diagnosis that may have occurred more than 6 years ago. Whether this should be interpreted as a failure in pain management is a moot point. The fact remains, however, that despite time since first diagnosis subjects not only continue to experience severe or moderate pain but a substantial proportion are possibly or probably experiencing chronic pain, a situation that appears to be unresolved for the majority who were first diagnosed with back pain six or more years ago.

Transparency

Declaration of funding

Funding for this paper was received from Grünenthal GmbH (Aachen, Germany).

Declaration of financial/other relationships

P.C.L. has disclosed that he is a consultant to Kantar Health, a company that undertook the analysis for and received funding from Grünenthal GmbH to conduct this study. H.L. is an employee of Grünenthal.

Acknowledgements

No assistance in the preparation of this paper is to be declared.