PRIMARY CARE & HEALTH SERVICES SECTION

Transcription

1 bs_bs_banner Pain Medicine 2013; 14: Wiley Periodicals, Inc. PRIMARY CARE & HEALTH SERVICES SECTION Original Research Article Prevalence, Causes, Severity, Impact, and Management of Chronic Pain in Australian General Practice Patients Joan V. Henderson, BAppSc(Hons), PhD (Medicine), Christopher M. Harrison, BPsych(Hons), MSocHlth, Helena C. Britt, BA, PhD, Clare F. Bayram, BAppSc(Hons), and Graeme C. Miller, MBBS, PhD, FRACGP Family Medicine Research Centre, Sydney School of Public Health, University of Sydney, Parramatta, New South Wales, Australia Reprint requests to: Joan V. Henderson, BAppSc(Hons), PhD (Medicine), Family Medicine Research Centre, University of Sydney, Level 7, Wentworth Street, Parramatta, NSW 2150, Australia. Tel: ; Fax: ; Disclosure/Conflict of interest: From 2008 to 2010, BEACH (Bettering the Evaluation And Care of Health) was funded by the Australian Government Department of Health and Ageing; Australian Government Department of Veterans Affairs; The Australian Institute of Health and Welfare; The National Prescribing Service Ltd; AstraZeneca Pty Ltd; Janssen-Cilag Pty Ltd; Merck Sharp and Dohme (Aust.) Pty Ltd; Novartis Pharmaceuticals Australia Pty. Ltd; Pfizer Australia Pty Ltd; Abbott Australasia; Sanofi-Aventis Australia Pty Ltd; Wyeth Australia Pty Ltd. The SAND (Supplementary Analysis of Nominated Data) substudy reported here was undertaken in collaboration with Janssen-Cilag Pty Ltd. The BEACH program is funded by a consortium of pharmaceutical manufacturers, Australian government agencies, and government-funded quality of care agencies such as the National Prescribing Service. Each group provides funding through a research agreement with the University of Sydney which provides complete research autonomy for the team conducting the BEACH program. Funders provide input into the research design development through an advisory board; however, the final decisions regarding research design, data collection and analysis, and reporting of findings remain with the principal investigators under the ethical supervision of the University. Each funding organization has the right to nominate two topics for the BEACH/SAND substudies. The organization works with the BEACH team to formulate the research questions for the SAND questionnaire. The final decision on the acceptability of the research questions rests with the BEACH team. All SAND questions are subject to approval by the Human Research Ethics Committee of the University. Once the SAND questions are approved, the processes of data collection, analysis, and reporting are conducted exclusively by the BEACH research team, including the development of published abstracts of the research. Any peer-reviewed articles developed from SAND substudies are authored or coauthored by the BEACH research team. The funding organization supporting the study has no editorial control over any aspect of the article including the presentation of results. No financial support for production of the article is accepted from the funding organizations. Abstract Objective. To determine the prevalence of chronic pain, its causes, severity, management, impact on sleep, mood and activity levels, and general practitioner (GP) and patient satisfaction with pain management. Design. A subset of 197 GPs and 5,793 patients from the BEACH program, a continuous, national crosssectional survey of Australian general practice. Results. The prevalence of chronic pain was 19.2% (95% confidence interval: ) (N = 1,113). The most commonly reported causal conditions were osteoarthritis (48.1%) and back problems (29.4%). For pain severity (using Von Korff s pain grades), 25.2% were at Grade I (lowest); 37.1% were at Grade II; 28.3% at Grade III; and 9.4% at Grade IV (highest). Medication was used for pain management by 86.1% of patients, and one third also used 1346

2 nonpharmacological managements. One third of patients were taking opioids, most commonly those at the highest pain severity grades. On Live Better with Pain Log scale, the impact of pain was similar across activity (mean = 4.0), sleep (mean = 4.8), and mood (mean = 4.8). On a scale of 1 (highest) to 5 (lowest), GPs satisfaction (mean = 2.5) was highly correlated (r = 0.7) with patients satisfaction (mean = 2.6) with pain management. Conclusions. Chronic pain impairs patient quality of life, and is a public health burden. This study provides a national overview of the prevalence, causes, severity, management and impact of chronic pain in Australian general practice patients, and the parity between GP and patient satisfaction with pain management. Key Words. Chronic Pain; Pain Management; Pain Severity; Impact of Pain Introduction Chronic pain results from many conditions. It contributes significantly to disease burden through expenditure for health and medical services and pharmaceuticals, and to high personal cost for individuals through physical suffering, loss of income, and social isolation [1]. Internationally, chronic pain is increasingly being recognized as a distinct entity, rather than merely a secondary consequence of other conditions [1 3]. At the 2010 Australian National Pain Summit, leading authorities in pain management supported this concept, calling for access to its treatment in the chronic disease model of care [4]. Prevalence is difficult to assess and estimates vary widely because chronic pain is associated with a multitude of conditions. Researching pain prevalence from clinical records is problematic because, even in practices using coding and classification systems, pain may be the reason for seeking treatment but is not identified as the problem being managed [5]. International prevalence estimates from population surveys over the past decade have ranged from 10.8% to 53.7% [6 13]. Study results often have limited comparability because methods vary significantly between studies in definitions of chronic pain [6,8,12,14 25], in data collection methods [6 8,10,12,14 17,21,26 30], or because they focus on: the body site affected by pain [14,31]; the condition causing pain [12,27,32,33]; the population subgroup experiencing pain [6,34 37]; or specific aspects of management [27,38,39]. Prevalence studies in Australia have similar variety, having been statefocused [15,18] or restricted to specific causes of pain [14,40]. The Australian Federal Government is responsible for national health care policy and initiatives, and for controlling and managing Medicare, Australia s publicly funded, universal health insurance system. The government is also responsible for the Pharmaceutical Benefits Scheme (PBS), which subsidizes costs of many medications. About 85% of Australians visit a GP at least once in any given year [41], and for most people, GPs are the first point of contact with the health system [42]. While the management of conditions causing chronic pain is often shared with specialists and other care providers, ongoing pain management usually occurs through primary care. A localized study of chronic pain patients in Sydney (Australia) found that people with pain were more likely to have consulted a GP than any other health practitioner or health service provider [43]. Pain clinics require a referral from a GP, and patients who attend these will return to the GP for ongoing pain management following discharge from the clinic [2]. Mantyselka et al. found that (long term) pain was the reason for 40% of visits in a primary care setting [30]. In this context, chronic pain management has potential to impact significantly on GP workload, and the complexity of its management is often increased by multimorbidity [44]. Guidelines generally focus on individual morbidities or on medication use, and research from secondary care populations cannot usually be extrapolated to primary care settings [2]. In a recent review of evidence available for guiding the management of chronic pain in primary care, Smith & Torrance concluded that management must be largely guided by consensus, experience, and judicious extrapolation from research in other contexts or conditions [2]. GPs have a variety of options for managing patients pain, but little information exists about whether these managements provide satisfactory pain relief from the GP and the patient perspective. The aims of this study are to determine the prevalence of chronic pain, its causes, severity, management techniques, its impact on sleep, mood and activity levels, and the satisfaction of GPs and patients with pain management, in a sample of patients attending general practice in Australia. Methods Chronic Pain in Australian General Practice This study was conducted through the BEACH (Bettering the Evaluation And Care of Health) program. BEACH is a continuous, national, cross-sectional survey of Australian general practice activity. The BEACH methods are described in detail elsewhere [45], but in summary, GPs from an ever-changing random sample of the (approximately) 18,000 currently practicing recognized GPs in Australia are invited to participate. Annually, 1,250 GPs are recruited and posted survey packs, from which approximately 1,000 completed surveys are returned. Each GP records details for 100 consecutive encounters with consenting, unidentified patients, on structured paper forms. Information is collected about what is managed for each patient at the single visit on the day the GP is participating [45]. 1347

3 Henderson et al. Because the survey is cross-sectional, patients may have health problems that were not managed at the recorded visit. To investigate these aspects of patient health, substudies are used whereby the GP records information additional to that recorded about the encounter. This method involves using the GP an as expert interviewer, and may provide more accurate information than patient self-report alone [46,47]. All questions are completed by the GP in discussion with the patient during the consultation, using the combined knowledge of both, and the patient s notes as an additional resource. These substudies of the BEACH program are known as SAND (Supplementary Analysis of Nominated Data). The SAND substudy questions are located in a separate area at the bottom of the encounter form. The pack of 100 forms is divided into three sections (1 40 forms, 2 30 forms) so that three different substudy topics can be surveyed. The SAND topics are run for a 5-week block and are sent to 125 GPs with the intention of receiving a completed sample of 3,000 patients (30 topic forms 100 GPs) per substudy [45]. The investigation of chronic pain was collected over two blocks with data being collected in April May 2008 and July August GPs were asked to record for each of 30 consecutive patients, the questions listed in Box 1. An instruction sheet at the beginning of the 30 question forms contained instructions and definitions for each question. Validated tools and definitions were employed for the study Chronic pain was defined as pain experienced every day for three months in the six months prior to this consultation [18]. The Chronic Pain Grades were those utilized by Von Korff et al. [48] (Box 2). The Chronic Pain Grades are four hierarchical classes that grade chronic pain as a function of pain intensity and pain-related disability, and were selected because they had been developed and evaluated by Von Korff et al. in a primary care setting, and were also used in other Australian studies, which would make results more comparable. As with all other SAND questions, the Pain Grade was recorded by the GP, resulting from discussion with the patient. An adaptation of the American Chronic Pain Association s Live Better with Pain Log [49] was used to score the patients pain level on occasions when pain was experienced, and to measure the impact of their pain on daily activity, sleep, and mood. For each of these elements, patients were asked to give a score between 1 and 10, 1 being the least impact and 10 the greatest. The Pain Log was chosen because of its simplicity (being pictorial and requiring minimal explanation to patients), for use in the time constraints of the GP patient consultation. For most questions, categorical responses (tick boxes) were offered, to reduce the response burden (as recording occurred during the consultation). Free text responses to other chronic pain causal conditions were secondarily coded, and classified according to the International Classification of Primary Care, version 2 [50]. Medications were coded and classified according to an in-house classification called the Coding Atlas for Pharmaceutical Box 1 Does the patient suffer from chronic pain? Yes No if no end questions. (Chronic pain defined as pain experienced every day for 3 months in the 6 months prior to this consultation ). If yes from what condition? Cancer Osteoarthritis Other arthritis Back problem Other condition (please specify) ( Causal defined as the condition/s you identify as being the cause/s of the patient s chronic pain ) When the patient is in pain, how severe do you judge the pain to be? Grade I Grade II Grade III Grade IV If the pain is currently being managed, how? Medication (please specify) Name & Form Strength Dose Frequency Other management (please specify) NO management Satisfaction with pain management is: GP satisfaction level l l l l l Highly Highly satisfied dissatisfied Patient satisfaction level l l l l l Highly Highly satisfied dissatisfied Ask the patient to rate the impact of pain on the following functions when in pain: (see card) Pain level Activity Sleep Mood Box 2 Severity of chronic pain Chronic pain grades I = low disability low intensity II = low disability high intensity III = high disability moderately limiting IV = high disability severely limiting 1348

4 Substances (CAPS), which is mapped to the Anatomical Therapeutical Chemical (ATC) Classification [51]. Both classifications were used to analyze medications reported, as CAPS has more specificity for classifying some combination products than does the ATC. Statistical Methods A cluster sample design was used with the GP being the primary sampling unit, and the patient at the encounter the unit of analysis. Procedures in SAS Version [52] accounting for the cluster study design were used to determine robust 95% confidence intervals (CIs) for the resulting estimates and percentages. Significance of differences was judged by nonoverlapping CIs. Where responses were not provided, missing data were removed from the analyses of that variable. An unadjusted prevalence estimate was calculated as the number of patients with chronic pain as a proportion of the total sample of respondents. This estimate can be interpreted as the prevalence of chronic pain among patients who present to GPs at any given time. Because patients were sampled at the GP consultation, the likelihood of being sampled is dependent on visit frequency. Female and older patients have been shown to visit a GP more frequently than male or younger patients [53]. Therefore, the data were weighted by patient age and sex to reflect the age and sex distribution of patients that attended Percent GP patient encounters Chronic Pain SAND substudy: N = 5,793 MBS GP Consultation Services: N = million general practice at least once in the year [54], to provide an estimate of prevalence in the attending population [55]. Correlation between GP and patient satisfaction with pain management was tested using the Pearson correlation statistic. Ethics Approval The BEACH program and this substudy were approved by the Human Research Ethics Committee of the University of Sydney (Reference no ) and the Ethics Committee of the Australian Institute of Health and Welfare. Results Chronic Pain in Australian General Practice Completed surveys were returned by 197 of the 250 GPs, giving a GP response rate of 79% and a sample of 5,793 patients. Of the GP participants, the sex distribution was similar to that of the sample frame of all GPs practicing in Australia in (58% male cf. 62% male). In age distribution, the participants were underrepresented in the <35 year and 55 year age groups and overrepresented in the year age group. For patient participants, Figure 1 shows the age distribution of the patients in this SAND sample to be highly representative of the patient age distribution at the million general practice encounters claimed across Australia through Medicare in [54]. The sex distribution (40.9% male) also years years years years years 75+ years Patient age Figure 1 Age distribution of patients at general practitioner (GP) patient encounters in this study and at all GP Consultation Service Items claimed through Medicare in

5 Henderson et al. accurately reflects the proportion of male patients at encounters claimed through Medicare (42.5%) for the same period [54]. Prevalence Among the 5,793 respondents, the prevalence of chronic pain was 19.2% (95% CI: ) (N = 1,113). Prevalence increased with patient age to 29.9% (95% CI: ) in patients aged years, and 36.2% (95% CI: ) in patients aged 75 years and older (Figure 2). Prevalence did not differ among males (18.0; 95% CI: ) and females (20.0; 95% CI: ). After adjusting for visit frequency, the estimated prevalence of chronic pain in the population of patients attending general practice was 15.7% (95% CI: ). Response rates varied across questions, so for each question, the number of respondents was the denominator used. Causal Conditions Over 10% of chronic pain patients reported more than one causal condition, 1,087 patients reporting 1,210 conditions. Osteoarthritis was most common (48.1% of patients with chronic pain), followed by back problems (29.4%) (Figure 3). Of the other conditions nominated by 28.2% of patients with chronic pain, three quarters were either musculoskeletal (56.2%) or neurological (19.6%) in nature. The former included bursitis, tendonitis, fibromyalgia, osteoporosis, and the latter included migraine, peripheral neuropathy, and neuralgia of various sites. Percent The small number of patients with chronic pain caused by cancer (N = 25) suggests that results for this group should be interpreted with caution. Pain Severity One in ten patients (9.4%) was rated as Grade IV, high disability severely limiting. The greatest proportion (37.1%) were judged to be at Grade II, 28.3% at Grade III, and 25.2% at Grade I, the lowest pain grade. (Figure 3). Pain Management Of the 1,074 patients for whom pain management was reported, 86.1% took at least one medication, two thirds of these without any other form of management. Approximately one in three (30.2%) took medication in conjunction with nonpharmacological pain managements. Smaller proportions of respondents were using only nonpharmacological managements (7.4%), or no management at all (6.5%) (Figure 4). Medication Management A total of 1,405 medications were reported for the 926 patients using medication. The most common was paracetamol, being taken by 42.8% of the 1,074 patients who responded to the management question. Opioids were taken by 34.0%, and 21.5% were taking nonsteroidal anti-inflammatory drugs (NSAIDs). A total of 418 opioids were recorded for the 365 patients who were taking them. Fifty-one patients were taking two, and two were taking three. The majority of years years years years years 75+ years Patient age preva- Figure 2 Age-specific lence of chronic pain. 1350

6 Chronic Pain in Australian General Practice Causal conditions* Osteoarthritis Total respondents with chronic pain - N = 1,113* 48.1 Back problems Other arthritis Figure 3 Causal conditions and severity of chronic pain. patients on multiple opioids took different strengths of the same product. The most frequently used were codeine (30+ mg) combinations (11.0%), followed by tramadol (8.9%) or oxycodone (7.7%). The NSAIDs most commonly used were meloxicam and celecoxib (Figure 4). Nonpharmacological Management More than one third (37.6%) of patients with chronic pain reported using nonpharmacological managements in combination with, or instead of, medication. For 398 patients, 480 managements other than medication were reported, the most common being physiotherapy, followed by therapeutic exercise, heat therapy, and exercise advice or education (Figure 4). Management and Severity Other conditions Pain Severity* Cancer Grade I Grade II Grade III Grade IV The relationship between management and pain severity is summarized in Table 1, presented as Severity Gradespecific rates. At least one medication, with or without other treatments, was predominant in the treatment options for all four Chronic Pain Grades. Medication alone was the most common form of management. Use of paracetamol tended to decrease as pain severity increased, while management with opioids increased significantly with increase in Pain Severity Grade. The use of NSAIDs was variable, but significantly higher in Severity Grades II and III than in Grade IV was apparent in the use of nonpharmacological treatments without medication. Patients were less likely to use no treatment as pain severity increased (Table 1). Management and Causal Conditions Percent The relationship between management and causal condition is summarized in Table 2, where condition-specific rates are presented. Because management was so similar for all types of arthritis, all arthritis was combined for this table. Also, managements for other conditions were not included because the conditions were dissimilar enough to render the reporting of them collectively as not meaningful. For all causes of chronic pain, patients were most commonly managed by medication. Arthritis was most commonly managed with paracetamol (more than half of arthritis cases), which was used at a significantly higher rate than either opioids (one in three) or NSAIDs (one in four). Nearly 40% of patients with arthritis used at least one nonpharmacological treatment. For patients with chronic pain caused by back problems, the rate of medication use alone (44.4%), and medication plus other nonpharmacological treatments (39.6%) did not significantly differ. Paracetamol and opioids were used at similar rates, both significantly higher than NSAID use. One in two back problems incorporated at least one nonpharmacological treatment in the management regimen, and 1 in 10 were managed with nonpharmacological treatments alone. The combination of medication and nonpharmacological treatment was significantly more likely in Pain Grades III and IV than in the lower Pain Grades, and the opposite For chronic pain caused by cancer, medication alone was the most common form of management. Opioid use rated highest (88.0%), with paracetamol being used in one in 1351

7 Henderson et al. Pain management* Total respondents with chronic pain - N = 1,113* Medication +/ other management 86.1 Medication only 56.0 Medication + other management 30.2 Other management only Medication Dextropropoxyphene comb.excl. psycholeptics Fentanyl - Other management No management Paracetamol Opioids Codeine 30+mg combinations Tramadol Oxycodone Buprenorphine Morphine NSAIDs Meloxicam Celecoxib Diclofenac sodium systemic Ibuprofen Glucosamine Other Medications medication Codeine <30mg combinations Amitriptyline Gabapentin Pregabalin Physiotherapy Therapeutic exercise Heat therapy Advide/education; exercise Referral Acupuncture Massage Surgical prodecure 1.8 Hydrotherapy 1.3 Support device Percent Figure 4 Management of chronic pain. 1352

8 Chronic Pain in Australian General Practice Table 1 Severity-specific management of chronic pain Severity Grade I (N = 262) Grade II (N = 386) Grade III (N = 296) Grade IV (N = 98) At least one medication 74.4 ( ) 86.3 ( ) 94.3 ( ) 94.9 ( ) Paracetamol 48.5 ( ) 46.1 ( ) 37.2 ( ) 33.7 ( ) Opioids 10.7 ( ) 28.8 ( ) 51.0 ( ) 67.3 ( ) NSAIDs 17.6 ( ) 26.2 ( ) 23.0 ( ) 10.2 ( ) At least one nonpharmacological treatment 32.4 ( ) 34.5 ( ) 44.3 ( ) 41.8 ( ) Combination of treatments Medication only 53.4 ( ) 60.4 ( ) 53.4 ( ) 55.1 ( ) Medication + nonpharmacological treatment 21.0 ( ) 25.9 ( ) 40.9 ( ) 39.8 ( ) Nonpharmacological treatment only 11.5 ( ) 8.5 ( ) 3.4 ( ) 2.0 ( ) No treatment 14.1 ( ) 5.2 ( ) 2.4 ( ) 3.1 ( ) Missing data removed. NSAIDs = nonsteroidal anti-inflammatory drugs. four cases. NSAID use was very low (4.0%), as was the rate of nonpharmacological management, and nonpharmacological management without medication were not utilized at all for chronic pain caused by cancer (Table 2). Satisfaction with Pain Management The full range of options (from 1 highly satisfied to 5 highly dissatisfied ) was used by both GPs and patients. The mean score for patients was 2.6, and the mean score for GPs was 2.5. The median score for both was 2 (results not tabled). There was high correlation between the patient s satisfaction with their pain management and the GP s satisfaction with the patient s pain management (r = 0.7). Table 2 Condition-specific management of chronic pain Impact of Chronic Pain For the level of pain when in pain, and its impact on activity, sleep, and mood, the numbers of respondents and the mean and median scores for each variable are shown in Figure 5. The mean score for patient pain level when in pain was 5.7, and the median score was 6. Discussion In summary, almost 20% of patients attending a GP in this study experienced chronic pain, and more than a third of these had pain ranking in the two highest Pain Grades. The main cause of pain was osteoarthritis, particularly among older patients. This suggests that as the Australian Selected Causal Conditions Arthritis (N = 581) Back Problems (N = 313) Cancer (N = 25) At least one medication 90.9 ( ) 84.0 ( ) 96.0 ( ) Paracetamol 54.2 ( ) 37.4 ( ) 28.0 ( ) Opioids 31.8 ( ) 39.0 ( ) 88.0 ( ) NSAIDs 25.6 ( ) 19.8 ( ) 4.0 ( ) At least one nonpharmacological treatment 29.3 ( ) 50.5 ( ) 16.0 ( ) Combination of treatments Medication only 64.2 ( ) 44.4 ( ) 80.0 ( ) Medication + nonpharmacological treatment 26.7 ( ) 39.6 ( ) 16.0 ( ) Nonpharmacological treatment only 2.6 ( ) 10.9 ( ) 0.0 ( ) No treatment 6.5 ( ) 5.1 ( ) 4.0 ( ) Missing data removed. NSAIDs = nonsteroidal anti-inflammatory drugs. 1353

9 Henderson et al. (N = 1,057) Mean = 5.7 Median = 6 (N = 1,056) NSAIDs Mean = 4.0 Median = 4.6 (N = 1,053) Mean = 4.8 Median = 5.0 (N = 1,052) Mean = 4.8 Median = 5.0 Adapted from: Live Better with Pain Log; 2005 The American Chronic Pain Association Figure 5 Pain level and impact of chronic pain. population ages, the prevalence of osteoarthritis-induced chronic pain will increase. A variety of managements were employed, the majority involving medication. A third of patients were managing pain with opioids, and the highest opioid use was among patients at the highest Pain Severity Grades. Two in five chronic pain patients used nonpharmacological managements in conjunction with, or instead of, medication, and 6.5% of chronic pain patients reported having no management for their chronic pain. The patients and GPs in this study were closely aligned in their levels of satisfaction with pain management. The impact of pain was similar across activity, sleep, and mood. Prevalence The prevalence estimate of chronic pain from this national study (19.2%), even when adjusted for the population 1354

10 attending general practice (15.7%; 95% CI: ), was not significantly different to the estimates found in the previous two state-based studies undertaken in Australia by Blyth et al. (~18.5%) [18] and Currow et al. (17.9%; 95% CI: ) [15], or in a national New Zealand study (16.9%; 95% CI: ) [10]. All the above studies used the same definition of chronic pain as that used in this investigation, but sampled population groups rather than general practice patients. Our result may be a slight underestimate of population prevalence, as only 88% of Australians visited a GP at least once in [45], and just as some patients in this study were not using any management for their chronic pain, there may also be a small proportion of Australians who manage their chronic pain with over-the-counter medications or other nonpharmacological therapies, and do not consult a GP. Pain Severity The variety of methods used in other studies to assess pain severity confounds comparison with these results. In studies where Von Korff s Chronic Pain Grades [48] was also used to measure pain severity or intensity, there are differences in participant samples (e.g., Dobscha et al. s investigation was in a primary care setting, but among a sample of Veterans Affairs [VA] patients) [22] or studies are limited to investigating chronic pain by causal condition [56 59]. Our study used a different sampling method to that of Blyth et al. [43], and while the proportions of patients in each Chronic Pain Grade differed between the two studies, there was similarity in the finding that the majority of participants in both were categorized at the two lowest Chronic Pain Grades (62.3% and 74.0%, respectively). Von Korff et al. [48] found that an individual may experience differences in Chronic Pain Grades by pain site, and this may apply to patients in this study, given that a number of them experienced pain from more than one cause or at more than one body site. We asked for an overall estimate in cognisance of the time constraints of the consultation and that the GP will manage the patient rather than pain at an individual site. Causal Conditions For the majority of patients, chronic pain was caused by musculoskeletal conditions such as osteoarthritis and back problems, a finding similar to that of other national and international studies [9,16,30,43]. That one in ten patients reported more than one causal condition complicates management. Even where only one condition is present, management may still require different strategies depending on the number of body sites affected. For example, patients with arthritis commonly feel its effects in more than one body site [60], and different managements may be more effective in different areas of the body (e.g., walking may ease pain associated with hip joints but may not have much affect on shoulders). Management Chronic Pain in Australian General Practice Pain management is often further complicated by the presence of other comorbid conditions that need to be considered in overall management. Some comorbidities directly influence the prevalence of chronic pain. For example, obesity has a strong influence on both the cause and progression of osteoarthritis [61,62]. Osteoarthritis was the leading cause of chronic pain for the patients in this study, and more than 60% of Australian adult general practice patients are either overweight (35.1%) or obese (26.7%), particularly patients aged years [63]. Prevalence of both overweight and obesity have increased significantly over the past decade [64]. Existing comorbidity could not be defined explicitly for the patients in this study, but other SAND substudies have shown high prevalence of multimorbidity among older age groups, with more than 50% of patients in the age groups most affected by chronic pain (65 years and over) having four or more morbidity domains on the Cumulative Illness Rating Scale [65]. The impact of this level of multimorbidity was highlighted by Dominick et al. who found accumulated comorbid load (defined as a count of other chronic conditions) to be independently associated with chronic pain [44,66]. Primary care physicians have reported being less than comfortable or confident in managing nonmalignant chronic pain, particularly with prescribed opioids [67 69]. The concerns most often cited are lack of training, risk of adverse events, substance misuse, and addiction [67 70]. Two recent studies reported a significant increase in prescribing of opioids in Australia, based on prescription data from the Australian PBS [71,72]. The PBS data include prescriptions from all clinicians (not just GPs) and lack any data about the conditions or indications for which the prescriptions were given. Yet, in both cases, and in an Editorial accompanying the latter [73], the authors inferred that this increase in opioid prescribing was attributable to GPs, for nonmalignant chronic pain. A more recent examination of BEACH data (which includes indication for prescription) showed that 43.9% of opioid prescribing by GPs was for chronic conditions (as defined by O Halloran et al., [74]) and 3.5% for malignant neoplasms leaving a significant proportion prescribed for conditions considered to be acute and most likely to be only for short-term use [75]. There is ongoing debate about the use of opioids in chronic pain management in Australia. Oxycodone prescribing by GPs has increased almost sevenfold over the past decade [75], with a correlating increase in deaths and hospitalizations attributed to overdose with prescription opioids [72,73]. Pain experts attribute the increased opioid prescribing to lack of other pain management alternatives available to GPs, particularly the lack of access to specialist support services such as pain clinics and other proven therapies such as cognitive behavioral therapies [76,77]. Semple & Hogg suggest that, in the absence of access to such services, and in light of a Medicare funding model 1355

11 Henderson et al. that discourages longer consultations required for managing complex conditions, time-poor GPs are likely to opt for therapy that is proven and readily available rather than allow patients to go untreated [76]. Our results show that one in five patients visiting a GP on any given day suffers from chronic pain, which must significantly impact on GP workload, yet they also show that these general practice patients are using a variety of managements for their chronic pain, not just relying on prescribed medication. Where opioids were taken, their use increased with pain severity so that the highest rate of usage was for the most severe pain. However, it is concerning that 10% of patients at Pain Grade 1 were taking opioids. Their pain grading may result from the opioid working well for them; however, patients were asked to rate their pain when in pain it may also mean that the opioid has been prescribed when other options, less associated with tolerance and dependency, might have achieved a satisfactory result. GPs often have minimal training in pain management [76], and while there are multiple guidelines available to GPs on management of chronic pain and rational use of opioids for chronic nonmalignant pain [78 81], these are produced by different bodies, differ in some recommendations, and are cumbersome up to 50 pages is not uncommon. An agreed, concise, easily accessible standardized approach by a national body might be of considerable benefit to ongoing clinician education. The use of NSAIDs by 21.5% of pain patients shows they are commonly prescribed for pain management. NSAID prescribing does not attract the same level of scrutiny as opioid prescribing, probably because of fewer concerns about risk of addiction and substance abuse. However, caution should also be applied to their long-term use. Traditional NSAIDS have potential significant gastrointestinal side effects, and while the more recent selective Cyclooxygenase-2 inhibitors are considered safer in terms of gastrointestinal risks, they also carry significant risk to the renal and cardiovascular systems [82 84]. Nonpharmacological treatment options were common for the patients in this study. Managements such as cognitive behavioral therapy [5,85], education, lifestyle modification, exercise [85], massage, physical therapy, and acupuncture [21] have proven beneficial to pain relief and improved activity levels. Artus et al. reported a high use of complementary and alternative medicine among primary care patients with chronic pain, either instead of, or in conjunction with, conventional treatments [27]. Patients often use these treatments without informing their GP, further complicating pain management through potential interactions with prescribed medications [86,87]. The capacity of these products to interact adversely is not commonly known to patients, and their availability exists without the stringent scrutiny and restrictions that apply to conventional pharmaceutical products in Australia. A small proportion of pain patients in this study (6.5%) reported having no pain management. We were unable to determine the reasons for this, but it may be that their pain was not being managed because they had not previously reported it to their GP. In the United States, Waktins et al. found that 22% of patients with chronic pain had not reported it to their physician, despite many experiencing moderate to severe pain [88]. In any event, the 6.5% reported here is a very small proportion compared with studies in other countries: 36% in New Zealand [10] and a similar proportion (one third) in a study of 15 European countries and Israel [21]. Impact of Pain Pain management will ultimately be tailored to the individual patient [89] and be influenced by the intensity of pain and its impact on their lives. For the patients in this study, the mean score of 5.7 and median score of 6 implies a considerable degree of pain on occasions when pain is experienced. The impact of pain on patients activity, the patterns and quality of their sleep, and on their mood scored in the mid range (median of on scale of 1 10). The capacity for pain to adversely influence an individual s quality of life has been well documented [21,29,35,43,90,91], and while we were able to include only measures of activity, sleep, and mood in this study, it is likely that the participants also experienced impact on employment, social relationships, and education, as reported by others [9,15,30,35,37,92]. Satisfaction with Pain Management Despite the demand on GPs time, the multidimensional nature of both the causes and effects of chronic pain, and the complexity of pain management (in consideration of probably multimorbidity), the majority of patients and GPs in this study were reasonably satisfied with pain management overall. There was a high correlation coefficient, and similarity between their mean and median satisfaction levels. From both perspectives, the full range of responses on the scale were recorded (from 1 highly satisfied to 5 highly dissatisfied ) but the 17.1% of GPs and 20.4% of patients who scored 4 or 5 on the satisfaction scale shows that pain management could be improved for a considerable number of patients. Mitchinson et al. found that 74% of primary care physicians were at least somewhat satisfied with the quality of care they provided to patients with chronic pain at VA medical centers, yet these clinicians still thought there was room for improvement and that further education and training, and better access to specialist care would improve pain management for their patients [34]. Mitchinson s study did not report patient satisfaction. Conversely, the pan-european study by Breivik et al. reported patient satisfaction rather than that of the clinicians 60% responded that their pain was adequately controlled, but the significant remaining proportion were not satisfied with the overall effectiveness of treatment for their chronic pain [21]. 1356

12 Satisfaction with pain management and the previously reported low proportion of patients with unmanaged pain compared with other international studies [10,21,88] probably reflects levels of access to care in Australia not always available to patients in countries with different health care payment models. Many practices still bulkbill (i.e., the patient is charged the standard Medicare rebate for the consultation, and it is directly billed to the Government such that the patient pays nothing out-ofpocket for the visit), and the PBS provides access to medications that would be prohibitive in countries where these costs are wholly born by the individual. Limitations and Strengths There were several limitations to this study. The GPs were underrepresented in the <35 year and 55 year age groups and overrepresented in the year age group. Other research has shown that the management of chronic problems increases with GP age, as does their prescribing rate [93]. Because the survey was patientfocused, and the distribution of patients was so highly representative of all GP patient encounters in Australia, we believe these data give an accurate reflection of the issues regarding chronic pain in ambulatory care patients. The Live Better with Pain Log has been validated elsewhere [94,95] and is used by patients as part of their daily assessment of pain impact, as recommended by the American Chronic Pain Society [49]. However, the section used in our study was applied retrospectively, which may have introduced a degree of recall bias because patients may give a different response to how pain affected them previously, if asked on a day when pain is less. As a further limitation, while we had a measure of mood (through the Pain Log), we had no opportunity to determine the coexistence of individual psychological problems with the causal conditions for chronic pain. Chronic pain has been found to be positively associated with the risk of psychological problems such as anxiety or depression [7,10,21,96]. The visual analog pain scale allows for a 1-item response only. There are tools which offer more specificity in gauging the particular areas where satisfaction is higher than others, for example, where efficacy of a medication is high but overall satisfaction may be reduced because of a side effect. The cost of larger, more specific tools, is the burden on the GP s time as we were asking for this research to be undertaken during the consultation, and historically, the more complex the substudies, the higher the rate of withdrawal. This type of scale allowed us to investigate the overall satisfaction with pain management from both the GP and patient perspectives, in consideration of the time constraints of the consultation. While the 1-item response is limited in specificity, it does provide this overall perspective. More importantly, it has shown that where satisfaction was poor to average, and pain management could be improved, the agreement between patient and GP is quite high (r = 0.7) suggesting good communication between clinician and patient. The strengths of the study lie in the large number and highly representative nature of the patient respondents, and the capacity through this method to utilize the GP as an interviewer to collect information with the patient s input during the consultation. This allows the patient to contribute knowledge not necessarily known to the GP, and for the GP to communicate and clarify aspects of the patient s health state or management a benefit to the quality of care as well as to the quality of the data. The method also enables the determination of the true breadth of chronic pain in patients attending general practice, because questions are focused on pain as the primary aspect, and the conditions associated with pain are secondary rather than the usual condition-based manner of recording in patient s medical records. Conclusion Chronic Pain in Australian General Practice The collective consequences of chronic pain on patient quality of life and capacity to work etc, has enormous impact on public health. Investigation is challenging because chronic pain is rarely recorded independent of its causal condition. This is the first comprehensive study in Australia to scope the problem of chronic pain nationally, and to demonstrate how chronic pain is managed in general practice the clinicians at the coal face of patient care. We have provided valid, national data from a representative sample of general practice patients, as a baseline from which educators and policy makers may develop strategies to improve service provision and quality of care for chronic pain patients. This evidence shows that GPs are already incorporating a multidisciplinary method of patient care, and not relying on prescription medications alone. Broadening a multidisciplinary team approach, with greater access to pain clinics, specialists and allied health professionals, and with standardized guidelines to reaffirm judicious prescribing, will support GPs in their clinical practice and enhance pain management for patients. Acknowledgments We thank the GP participants for their generosity, and all members of the BEACH research team. The SAND substudy reported here was undertaken in collaboration with Janssen-Cilag Pty Ltd. The Live Better with Pain Log was used with permission of the American Chronic Pain Association. References 1 Siddall PJ, Cousins MJ. Persistent pain as a disease entity: Implications for clinical management. Anesth Analg 2004;99(2): Smith BH, Torrance N. Management of chronic pain in primary care. Curr Opin Support Palliat Care 2011; 5(2):

13 Henderson et al. 3 Tracey I, Bushnell MC. How neuroimaging studies have challenged us to rethink: Is chronic pain a disease? J Pain 2009;10(11): Australian Pain Management Association. National pain summit. A national healthcare policy initiative Available at: (accessed May 7, 2013). 5 McBeth J, Prescott G, Scotland G, et al. Cognitive behavior therapy, exercise, or both for treating chronic widespread pain. Arch Intern Med 2012;172(1): Ng KF, Tsui SL, Chan WS. Prevalence of common chronic pain in Hong Kong adults. Clin J Pain 2002;18(5): Ohayon MM, Schatzberg AF. Using chronic pain to predict depressive morbidity in the general population. Arch Gen Psychiatry 2003;60(1): Boulanger A, Clark AJ, Squire P, Cui E, Horbay GL. Chronic pain in Canada: Have we improved our management of chronic noncancer pain? Pain Res Manag 2007;12(1): Catala E, Reig E, Artes M, et al. Prevalence of pain in the Spanish population: Telephone survey in 5000 homes. Eur J Pain 2002;6(2): Dominick C, Blyth F, Nicholas M. Patterns of chronic pain in the New Zealand population. N Z Med J 2011;124(1337): Rustoen T, Wahl AK, Hanestad BR, et al. Prevalence and characteristics of chronic pain in the general Norwegian population. Eur J Pain 2004;8(6): Bouhassira D, Lanteri-Minet M, Attal N, Laurent B, Touboul C. Prevalence of chronic pain with neuropathic characteristics in the general population. Pain 2008;136(3): Gerdle B, Bjork J, Henriksson C, Bengtsson A. Prevalence of current and chronic pain and their influences upon work and healthcare-seeking: A population study. J Rheumatol 2004;31(7): Walker BF, Muller R, Grant WD. Low back pain in Australian adults: Prevalence and associated disability. J Manipulative Physiol Ther 2004;27(4): Currow DC, Agar M, Plummer JL, Blyth FM, Abernethy AP. Chronic pain in South Australia Population levels that interfere extremely with activities of daily living. Aust N Z J Public Health 2010;34(3): Hasselstrom J, Liu-Palmgren J, Rasjo-Wraak G. Prevalence of pain in general practice. Eur J Pain 2002;6(5): Clark JD. Chronic pain prevalence and analgesic prescribing in a general medical population. J Pain Symptom Manage 2002;23(2): Blyth FM, March LM, Brnabic AJ, et al. Chronic pain in Australia: A prevalence study. Pain 2001;89(2 3): Moulin DE, Clark AJ, Speechley M, Morley-Forster PK. Chronic pain in Canada prevalence, treatment, impact and the role of opioid analgesia. Pain Res Manag 2002;7(4): Kerssens JJ, Verhaak PF, Bartelds AI, Sorbi MJ, Bensing JM. Unexplained severe chronic pain in general practice. Eur J Pain 2002;6(3): Breivik H, Collett B, Ventafridda V, Cohen R, Gallacher D. Survey of chronic pain in Europe: Prevalence, impact on daily life, and treatment. Eur J Pain 2006;10(4): Dobscha SK, Corson K, Perrin NA, et al. Collaborative care for chronic pain in primary care: A cluster randomized trial. JAMA 2009;301(12): Mantyselka PT, Turunen JH, Ahonen RS, Kumpusalo EA. Chronic pain and poor self-rated health. JAMA 2003;290(18): Reid MC, Engles-Horton LL, Weber MB, et al. Use of opioid medications for chronic noncancer pain syndromes in primary care. J Gen Intern Med 2002; 17(3): McDermott ME, Smith BH, Elliott AM, et al. The use of medication for chronic pain in primary care, and the potential for intervention by a practice-based pharmacist. Fam Pract 2006;23(1): Verhaak PF, Kerssens JJ, Dekker J, Sorbi MJ, Bensing JM. Prevalence of chronic benign pain disorder among adults: A review of the literature. Pain 1998;77(3): Artus M, Croft P, Lewis M. The use of CAM and conventional treatments among primary care consulters with chronic musculoskeletal pain. BMC Fam Pract 2007;8: Gerdle B, Bjork J, Coster L, et al. Prevalence of widespread pain and associations with work status: A population study. BMC Musculoskelet Disord 2008; 9: McCarberg BH, Nicholson BD, Todd KH, Palmer T, Penles L. The impact of pain on quality of life and the 1358