Purpose: Exercise and patient education are considered first-line treatment for patients with knee and hip osteoarthritis (OA) and mild analgesics are recommended when first-line treatments are insufficient to reduce symptoms, while opioids are not recommended in the management of OA at all. However, studies have indicated that a large proportion of patients are not offered first-line treatment, but likely use analgesics instead, including opioids, for their joint pain.
The aims of this study were 1) to investigate the prevalence of analgesic use (i.e., dispensed prescriptions of analgesics) among patients with knee or hip OA initiating an exercise therapy and patient education program in primary care, and 2) to explore differences in patient characteristics among patients using different types of analgesics.
Methods: We used data from the Good Life with OA in Denmark (GLA:D®) registry (data collected from January 2013 to December 2018) linked with the Danish National Prescription Registry and other national health registers. The GLA:D® registry includes all patients who have participated in a standardised primary care program in Denmark. GLA:D® consists of 2 group-based patient education, and 12 supervised neuromuscular exercise sessions (1 hour) delivered by physiotherapists over an 8-week period.
We included patients ≥40 years with register data coverage the 365 days preceding the start of the intervention (i.e., index date). Participants with a diagnosis of cancer (ICD-10 code C00-97, excluding C44) or substance abuse (ICD-10 code F11) within the 365 days preceding the index date were excluded.
Based on data from the Danish National Prescription Registry covering 90 days preceding the index date, we included all dispensed prescriptions of paracetamol (Anatomical Therapeutic Chemical [ATC] code N02BE01), non-steroidal anti-inflammatory drugs (NSAIDs; M01A, excluding M01AX), and opioids (N02A, R05DA04, and N02BA75). Participants with at least one dispensed analgesic prescription within this 90-day period were classified into 1) paracetamol (only) users, 2) NSAIDs (only) users, 3) NSAIDs and paracetamol users, 4) opioids (only) users, 5) opioids and paracetamol users, 6) opioids and NSAIDs users, and 7) opioids, NSAIDs, and paracetamol users, while participants with no dispensed analgesic prescription within the time period were classified as no users.
The following baseline data from patients participating in GLA:D® was used: age, sex, level of education, body mass index (BMI), index joint (i.e., knee or hip), average pain intensity in most affected joint (i.e., within last month; visual analogue scale), symptom duration, presence of comorbidities, and number of pain sites (i.e., sum of body regions marked by the participant on a mannequin indicating the experience of pain within the last 24 hours; maximum number of pain sites was 56). Age and sex were derived from a national person identification number. Height and weight, used to calculate BMI, were collected by the clinician at baseline. Number of comorbidities was derived from self-reported data about the presence of hypertension, heart disease, stomach ulcer or gastrointestinal disease, respiratory disease, diabetes, kidney- or liver disease, anaemia, cancer, depression, rheumatoid arthritis, neurological disease, and other medical disease (i.e., maximum number of comorbidities was 12).
Differences in characteristics between analgesic use groups were investigated using ANOVA and Pearson’s Chi-squared test for continuous and categorical variables, respectively (p-value <0.05 was considered statistically significant).
Results: In total, 37,640 patients out of 38,745 eligible patients fulfilled our inclusion criteria, and 46% of these had at least one dispensed prescription of analgesics within the 90 days preceding the index date. Among analgesic users, paracetamol (only) was most common (15.4%), followed by opioids alone or in combination with paracetamol and/or NSAIDs (11.1%), NSAIDs and paracetamol (11.0%) and NSAIDs (only) (8.0%) (Table 1).
About 75% were women in most analgesic use groups, except for the no use- and NSAIDs (only) group, in which about 70% were women. In general, a larger proportion of study participants with no analgesic use reported a medium-cycle higher education or a long cycle higher education (or higher) compared to the other analgesic use groups. Finally, participants not using analgesics had a slightly lower BMI compared with all other analgesic use groups.
The study participants who used opioids differed from the other analgesic use groups on several characteristics. For example, opioid users, in particular opioids (only) and opioids in combination with paracetamol, were generally older than the other groups. Furthermore, opioid users in particular (and analgesic users in general), reported a higher average pain intensity compared to those not using analgesics, and the longest symptom duration was found among participants using opioids and paracetamol. In addition, study participants using opioids (alone or in combination with paracetamol and/or NSAIDs) reported a higher number of comorbidities than most other groups. Similarly, the highest number of pain sites was found among participants using opioids, only or in combination with other analgesics (and among paracetamol [only] users). For example, 9.6% of opioids (only) users reported 10 or more pain sites compared to 7.1%, 5.3%, and 4.7% of paracetamol (only) users, NSAIDs (only) users, and no users, respectively.
Conclusions: Use of prescribed analgesics, including opioids, is common among patients with knee or hip OA initiating an exercise therapy and patient education program. Patients with knee or hip OA using opioids seems to be older and report a higher average pain intensity, longer symptom duration, higher number of comorbidities, and higher number of pain sites compared to patients not using analgesics, or patients using paracetamol and/or NSAIDs. Our findings indicate that a large proportion of patients with knee or hip OA who receive analgesics (second-line treatment) may experience a limited pain-relieving effect, and, therefore, seek exercise therapy and patient education (first-line treatment).
Table 1Characteristics of knee or hip osteoarthritis patients stratified by analgesic use group
| Characteristics | Analgesic use groups N = 37640 (100.0%) mean (SD) / n (%) | P-value | |||||||
|---|---|---|---|---|---|---|---|---|---|
| No use n = 20524 (54.5%) | Paracetamol (only) n = 5781 (15.4%) | NSAIDs (only) n = 2998 (8.0%) | NSAIDs and paracetamol n = 4140 (11.0%) | Opioids (only) n = 1118 (3.0%) | Opioids and paracetamol n = 1679 (4.4%) | Opioids and NSAIDs n = 334 (0.9%) | Opioids, NSAIDs, and paracetamol n = 1066 (2.8%) | ||
| Age, yrs. | 65.3 (9.4) | 68.7 (9.1) | 64.4 (9.3) | 64.5 (9.3) | 66.7 (9.5) | 68.8 (9.3) | 64.1 (9.2) | 64.3 (9.6) | <0.001 |
| Sex, women (vs. men) | 14339 (69.9) | 4335 (75.0) | 2027 (67.6) | 3073 (74.2) | 827 (74.0) | 1278 (76.1) | 244 (73.1) | 806 (75.6) | <0.001 |
| BMI | 27.7 (5.0) | 28.9 (5.3) | 28.8 (5.3) | 29.7 (5.7) | 28.8 (5.3) | 29.6 (5.8) | 29.9 (5.8) | 30.3 (6.1) | <0.001 |
| Index joint, knee (vs. hip) | 15392 (75.0) | 4104 (71.0) | 2378 (79.4) | 3065 (74.1) | 863 (77.3) | 1177 (70.1) | 246 (73.7) | 803 (75.4) | <0.001 |
| Average pain intensity last month, mm VAS | 42.9 (21.3) | 51.4 (20.6) | 49.5 (21.8) | 54.1 (20.8) | 54.8 (21.8) | 59.1 (21.1) | 58.7 (21.8) | 59.3 (23.1) | <0.001 |
| Symptom duration, months | 41.9 (65.0) | 40.0 (61.7) | 36.0 (60.2) | 36.5 (60.4) | 41.9 (65.6) | 44.3 (71.2) | 36.1 (54.5) | 35.5 (58.1) | <0.001 |
| Number of comorbidities | |||||||||
| 0 | 7075 (41.7) | 1346 (28.1) | 1025 (41.7) | 1336 (38.9) | 219 (25.1) | 245 (18.1) | 69 (26.3) | 278 (32.1) | <0.001 |
| 1-2 | 8695 (51.2) | 2869 (59.9) | 1280 (52.1) | 1854 (54.0) | 503 (57.6) | 787 (58.3) | 156 (59.5) | 485 (56.1) | <0.001 |
| ≥3 | 1215 (7.2) | 576 (12.0) | 152 (6.2) | 243 (7.1) | 152 (17.4) | 319 (23.6) | 37 (14.1) | 102 (11.8) | <0.001 |
SD, standard deviation; NSAIDs, non-steroidal inflammatory drugs; yrs., years; BMI, body mass index; VAS, visual analogue scale; P-value from ANOVA (continuous variables) and Pearson’s Chi-squared test (categorical variables). Number of comorbidities was derived from self-reported data about the presence of hypertension, heart disease, stomach ulcer or gastrointestinal disease, respiratory disease, diabetes, kidney- or liver disease, anaemia, cancer, depression, rheumatoid arthritis, neurological disease, and other medical disease (i.e., maximum number of comorbidities were 12). Results for level of education, and number of pain sites are not shown due to space limitations. Most variables had few missing values. However, level of education had 4946 (13.1%) missing values, symptom duration had 4745 (12.6%) missing values due to technical problems during data collection, number of comorbidities had 6622 (17.6%) missing values due to late introduction of the questions in the questionnaire, and number of pain sites had 8045 (21.4%) missing values partly because the question/pain mannequin was first introduced in April 2014, and partly because of a technical problem during data collection.
Article info
Publication history
Presentation Number: 508
Identification
Copyright
© 2021 Published by Elsevier Inc.
User license
Elsevier user license | How you can reuse 
Elsevier's open access license policy
Elsevier user license
Permitted
For non-commercial purposes:
- Read, print & download
- Text & data mine
- Translate the article
Not Permitted
- Reuse portions or extracts from the article in other works
- Redistribute or republish the final article
- Sell or re-use for commercial purposes
Elsevier's open access license policy
