Sie können Operatoren mit Ihrer Suchanfrage kombinieren, um diese noch präziser einzugrenzen. Klicken Sie auf den Suchoperator, um eine Erklärung seiner Funktionsweise anzuzeigen.
Findet Dokumente, in denen beide Begriffe in beliebiger Reihenfolge innerhalb von maximal n Worten zueinander stehen. Empfehlung: Wählen Sie zwischen 15 und 30 als maximale Wortanzahl (z.B. NEAR(hybrid, antrieb, 20)).
Findet Dokumente, in denen der Begriff in Wortvarianten vorkommt, wobei diese VOR, HINTER oder VOR und HINTER dem Suchbegriff anschließen können (z.B., leichtbau*, *leichtbau, *leichtbau*).
Osteoarthritis (OA) is associated with joint pain and disability, with increasing prevalence and economic burden. This study explored non-steroidal anti-inflammatory drugs (NSAID) treatment patterns, gastrointestinal (GI) complications and burden of disease in patients with OA based on Nordic registry data, with emphasis on individuals aged 60 or younger.
Methods
This non-interventional observational study analysed registry data from Norway, Finland, and Sweden. Adults with primary OA diagnosis in specialty care were matched 1:1 by age and sex with individuals of the general population. The primary outcome was NSAID treatment patterns, assessed by dispensations and daily defined doses (DDD) 1 year before and after diagnosis. Secondary outcomes included prevalence of comorbidities, the incidence of GI complications, sick leave, and joint replacements.
Results
The study included 189,553 patients with OA from Norway, 341,548 from Sweden, and 218,253 from Finland, 34.1% to 40.4% aged ≤ 60. About half of patients had NSAIDs dispensed before and after diagnosis, (Norway 56.6% and 48.6%; Sweden 50.1% and 44.0%; Finland 58.8% and 58.2%), compared to about one-fifth of the matches (Norway 19.1% and 18.5%; Sweden 12.7% and 12.2%; Finland 22.5% and 21.7%). NSAID use increased with age until approximately 55 years, then declined. After diagnosis, fewer patients had NSAID dispensations, but mean DDDs per prescription were generally higher. Secondary analyses showed a higher comorbidity burden, higher rates of GI complications, more sick leave days, and more joint replacements in patients with OA than in matched individuals.
Conclusion
Across Nordic countries, patients with OA had more NSAID dispensations, and GI complications were more prevalent compared to matched individuals. Notably, this was already evident in younger patients under the age of 60, underscoring the need for comprehensive GI risk assessment for every patient with OA. The substantial burden of OA was also evidenced by considerable numbers of sick leave days and joint replacements.
Prior Presentation A poster based on this study was presented at the Austrian Pain Society (ÖSG) Annual Scientific Meeting, in Vienna, Austria, on May 22, 2025; and at the Scandinavian Rheumatology Congress in Malmö, Sweden, on September 5, 2025. As part of the presentation at the Scandinavian Rheumatology Congress, the corresponding abstract was also published in the Scandinavian Journal of Rheumatology (54(sup132): 60–304, 2025. https://doi.org/10.1080/03009742.2025.2533698).
Key Summary Points
Why carry out this study?
Non-steroidal anti-inflammatory drugs (NSAIDs) use is common in osteoarthritis (OA) management and recommended by international guidelines, but is associated with increased risk of gastrointestinal (GI) complications.
In addition to NSAID use and associated GI complications, the broader burden of OA includes comorbidities, sick leave, and the need for joint replacements. Although extensive research on OA has focused on vulnerable populations, particularly the elderly, epidemiological studies with emphasis on working-age individuals remain limited.
The study assessed NSAID use before and after OA diagnosis and evaluated comorbidities, GI complications, PPI use, sick leave, and joint replacements to better understand treatment risks, productivity loss, and disease progression in individuals with OA, with emphasis on younger patients, where risks may be underestimated.
What was learned from the study?
A higher proportion of patients with OA receive NSAID dispensations compared to matched individuals, particularly among younger patients (≤ 60 years). The proportion of patients treated with NSAIDs rises from age 18 to 50–55 but decreases thereafter, potentially as a result of transitioning to specialized treatments, surgical intervention, or concerns over adverse drug reactions.
Despite changes in treatment patterns, the risk of GI complications remains notably higher in patients with OA than in matched individuals across all age groups. This risk increases gradually with advancing age, without a sudden rise around 60–65 years, highlighting the need to consider GI risks even in younger patients.
Younger patients with OA experience a significant burden, including high rates of NSAID use, GI complications, sick leave, and the need for joint replacements. These findings underscore the importance of early intervention and comprehensive management strategies to address the long-term effects of OA, even in younger populations.
Introduction
Osteoarthritis (OA) is a long-term chronic disease characterized by the deterioration of cartilage in joints. It is often thought of as a degenerative condition, but it does not arise only because of gradual wear and tear. Instead, it should be viewed as an abnormal remodelling of the joint tissues, articular cartilage, and bone which is driven by many inflammatory mediators. Synovial inflammation is directly linked to clinical symptoms such as joint swelling, synovitis, and inflammatory pain [1‐3]. The most frequently affected joints are the knees, hips, hands, feet, and spine [4].
Anzeige
The burden of OA in the Nordic region is increasing. Between 1990 and 2015, the number of prevalent OA cases increased by 43% to 1,507,587, with contribution of OA to years lived with disability rising from 1.3% in 1990 to 1.6% in 2015 [5]. Moreover, the economic burden of OA in the Nordic countries is substantial, and the largest cost drivers are productivity loss and surgeries, underscoring the urgent need for effective treatment options [6]. Despite perceptions that OA only affects older adults, many people develop OA before 60 years of age and live decades with pain and disability [7]. The prevalence of OA in younger adults may rise further in the coming years as a result of increases in two key risk factors: sports-related injuries and obesity [8]. Although extensive research has been conducted on vulnerable populations, particularly the elderly, epidemiological studies specifically targeting working-age individuals with OA remain limited [9].
Treatment guidelines in the Nordic countries are based on principles such as education about the disease, lifestyle changes, pain relief, optimization of physical function, preventing progression of harmful structural changes to cartilage, bone, and ligaments, and surgical procedures, when necessary [10‐14]. While lifestyle interventions can offer substantial benefits, additional pharmacological treatments are often necessary for managing OA. International guidelines recommend topical non-steroidal anti-inflammatory drugs (NSAIDs) as initial treatment option. If topical treatments are ineffective or unsuitable, oral non-selective NSAIDs such as ibuprofen or naproxen or selective COX-2 inhibitors like etoricoxib or celecoxib may be used. Regardless of the NSAID chosen, it is important to consider potential gastrointestinal (GI), renal, liver, and cardiovascular (CV) risks [15‐17].
Numerous studies support the efficacy of NSAIDs in OA [18‐20]. However, their use can be associated with GI complications such as GI ulcer development, which primarily affect the stomach and upper intestine. Their use may lead to significant morbidity as a result of bleeding, perforation, and obstruction, and even potentially death [21‐23]. To mitigate GI complications, gastroprotective agents such as proton pump inhibitors (PPIs) are recommended when using oral NSAIDs [15, 16], as symptoms of GI discomfort are not predictive of ulcer development [24].
Several factors increasing the risk of GI complications with NSAID use have been identified, with one of them being age above 60 years [25, 26]. However, GI risk in younger patients may be potentially underestimated, as comprehensive data on NSAID use and GI complications in this population is lacking. This study aimed to examine the treatment patterns of NSAIDs in patients with OA, with an emphasis on individuals aged 60 or younger, based on Nordic registry data. The primary objective was to assess NSAID use by dispensations and defined daily doses (DDD) 1 year before and after diagnosis. Secondary objectives were to describe comorbidities during 3-year pre-diagnosis period and evaluate GI complications and PPI use (to capture treatment risks), sick leave (reflecting productivity loss), and hip/knee replacements (indicating disease progression).
Anzeige
Methods
Study Design and Time Periods
This Nordic, non-interventional retrospective observational study used pseudonymized patient-level data from specialty care, excluding those diagnosed in primary care, derived from population-based registers in Norway, Sweden, and Finland (Supplementary Table 1). The overall study period spanned from January 1, 2008 to December 31, 2018, during which data were collected and analysed. The patient identification period, from January 1, 2011 to December 31, 2017, enabled the identification of patients and the setting of index dates (i.e. the first visit to specialty care with first OA diagnosis). A 3-year pre-index period was used to exclude those with a prior OA diagnosis and identify the Elixhauser comorbidity index. Additionally, a 1-year pre- and post-index period was used to capture rest of the outcomes. These timeframes allowed for a thorough examination of patients visiting specialty care (in- and outpatient hospital care) for the first time.
Study Population
The study population consisted of two cohorts: the OA cohort and a matched comparison cohort. The OA cohort included patients, aged 18 or older, with a first primary diagnosis of OA (ICD-10 codes M15–M19; see Supplementary Table 2) in specialty care from one of the three countries (Norway, Sweden, and Finland) between 2011 and 2017. Those with a diagnosis of OA anytime between 2008 and 2010 were excluded to ensure at least 3 years look-back period as part of the baseline data for all patients.
Each patient with OA was matched 1:1 on birth year and sex with a random individual from the general population. The matched individual was selected from national population registers and assigned the same index date as the patients with OA and was required to be at least 18 years old on the index date. This matching process was conducted without replacement, meaning that once a comparison was matched to a case, it could not be used again.
Baseline Characteristics
Patients’ age (continuous variable, measured in years) and sex (dichotomous variable) were recorded at the time of the index date.
Study Outcomes
Primary Outcome
The treatment patterns of NSAIDs in the OA cohort and in the matched comparison cohort and the number of NSAID dispensations (1 year before and after index, differentiated on the 5th ATC level) were assessed. The sum of DDD of NSAID dispensations (1 year before and after index, differentiated on the 5th ATC level) was also assessed.
Secondary Outcomes
Comorbidities During 3-Year Pre-diagnosis Period
The Elixhauser comorbidity index (continuous variable) was calculated for each patient [27], and individual comorbidities (dichotomous variables) included in the index were reported. The comorbidities were defined on the basis of ICD-10 codes (either as primary or secondary diagnoses) from specialty care during the 3-year pre-index period (see Supplementary Table 3 for ICD-10 codes).
Incidence of GI Complications
The number and proportion of patients experiencing at least one GI complication were reported for both the year before and the year after the index date in OA cohort and matched comparison cohort. The definition of GI complications was based on having a hospital visit or hospitalization with a primary diagnosis of one of the codes listed in Supplementary Table 4; these in turn are based on the definitions used in studies by Button et al. [28] and Park et al. [29]. The study by Button et al. was a record linkage study of hospital admissions for upper GI bleeding; the Park et al. study validated diagnostic codes for major clinical outcomes in a national health insurance database, including codes for GI bleeding. In addition to the codes from the two aforementioned studies, diagnoses for diverticulitis (K57) and ulcerative colitis (K51) were also used to define GI complications in this study.
For Norway, users of PPI (ATC code A02BC), including users of a combination of the NSAID naproxen and the PPI esomeprazole (ATC code M01AE52), were defined as a separate outcome. PPI data were not available in Finland and Sweden.
Sick Leave
The number of net sick leave days as registered in the social insurance registries over the 1-year period before and after index was recorded. Sick leave can be partial in the Nordic countries and net days for each sick leave episode were calculated by multiplying the number of ‘gross’ days by the level of compensation of the respective episode. For example, 30 days of 50% sick leave would result in 15 net sick leave days. However, the definitions of when sick leave starts to be recorded differ between countries: in Sweden, sick leave is only recorded in the sickness insurance registers if the sick leave has been lasting for at least 14 days (when the payment responsibility is shifted to the government). Similarly, in Norway, state-paid sick leave is registered only from day 17 and onwards, and in Finland from day 11 and onwards. This implies that short-term sick leave below these thresholds is not captured in the register data. However, the first day of sick leave was known for the episodes recorded in the data.
Hip or Knee Joint Replacement
The number and proportion of individuals with hip/knee replacements 1 year before and after index were also assessed on the basis of NOMESCO surgical codes (NFB for hip replacements and NGB for knee replacements) [30].
Statistical Analyses
The data were analysed descriptively for all study outcomes and were presented as mean, standard deviation (SD) of the mean for continuous variables and as number and proportion of patients in each category for categorical and dichotomous variables.
For each of the objectives of interest, outcomes were reported for 1-year period before diagnosis and 1-year period after diagnosis, except the Elixhauser comorbidity index that was based on the 3-year pre-diagnosis period. Similarly, outcomes were reported for patients with OA and their matched individuals for comparison.
The analysis was stratified by age groups in 5-year bands, categorized by age at diagnosis: < 18, 18–25, 26–30, 31–35, 36–40, 41–45, 46–50, 51–55, 56–60, 61–65, 66–70, 71–75, 76–80, 81–85, 86–90, and 91+ years. Further analysis in this study focused on age groups between 18 and 90 years, except for sick leave days, where working-age population (between 18 and 65 years) was considered. Patients aged 60 and younger were considered the younger population, as they are typically deemed the economically active/productive subgroup, which remains underexplored in OA research.
Ethical statement
Ethical approval was obtained from the ethical review board in South-East Norway (28745) and the regional ethical review board in Stockholm (2018/1634-31/2) for Norway and Sweden, respectively. The use of Finnish data did not require ethical approval.
Anzeige
Results
Study Population
As a result of matching, the study included an equal number of patients with OA and matched individuals across Norway (n = 189,553), Sweden (n = 341,548), and Finland (n = 218,253) with the same proportion of men in the OA and matched comparison cohorts across all countries (42.8% [n = 81,217] in Norway; 44.3% [n = 151,151] in Sweden; 42.0% [n = 91,604] in Finland). The mean (SD) age of patients with OA was 62.5 (13.6) years in Norway, 64.5 (12.8) in Sweden, and 63.5 (13.3) years in Finland, with matched individuals having identical age by design (Table 1).
Table 1
Patient characteristics, by country and cohort
Characteristic
Norway
Sweden
Finland
OA cohort
Matched comparison cohort
OA cohort
Matched comparison cohort
OA cohort
Matched comparison cohort
Number of patients
189,553
189,553
341,548
341,548
218,253
218,253
Men
81,217 (42.8%)
81,217 (42.8%)
151,151 (44.3%)
151,151 (44.3%)
91,604 (42.0%)
91,604 (42.0%)
Age (years)
62.5 (13.6)
62.5 (13.6)
64.5 (12.8)
64.5 (12.8)
63.5 (13.3)
63.47 (13.3)
Elixhauser comorbidity index score
0.68 (1.16)
0.58 (1.14)
0.66 (1.17)
0.49 (1.07)
0.61 (1.11)
0.41 (0.94)
Data are presented as mean (SD)/N (%)
N number, OA patients with osteoarthritis, SD standard deviation
Overall, the age distribution of the OA cohort was skewed towards patients older than 60 years of age in all countries, but a notable portion of the cohort was younger. Younger patients (≤ 60 years) comprised 40.4% of the patients in Norway, 34.1% in Sweden, and 39.0% in Finland (Fig. 1).
Fig. 1
Age-wise distribution of patients with osteoarthritis
Across countries, about half of the patients with OA had NSAIDs dispensed both before and after diagnosis (Norway 56.6% before and 48.6% after diagnosis; Sweden 50.1% and 44.0%; Finland 58.8% and 58.2%), compared to about one-fifth of the matches (Norway 19.1% before and 18.5% after; Sweden 12.7% and 12.2%; Finland 22.5% and 21.7%). In general, the proportion of patients with OA with NSAID dispensations decreased after diagnosis across all age groups in Norway and Sweden and in patients with OA below the age of 60 in Finland. For example, in Norway, 43.0% of patients with OA aged 18–25 received NSAID dispensations before diagnosis, compared to 9.3% of matched individuals. After diagnosis, 34.0% of patients with OA in the same age group received NSAID dispensations, compared to 9.5% of matched individuals (Fig. 2). In the matched comparison cohort across all countries, the proportion of individuals with NSAID dispensations remained consistent before and after diagnosis.
Fig. 2
Proportion of individuals with an NSAID dispensation before or after diagnosis, by cohort (patients with osteoarthritis and matched comparisons) and age group (18–90 years). NSAID non-steroidal anti-inflammatory drugs, OA osteoarthritis
The pattern of NSAID dispensation for patients with OA varied by age, with a similar trend observed across countries. Even in the youngest age group between 18 and 25 years of age, a significant number of patients had NSAIDs dispensed. The proportion of patients with NSAID dispensations increased gradually with age up to the age of 55 years and gradually decreased again from above 60 years of age. Specifically, in Norway, 43.0% of patients aged 18–25 received NSAIDs before diagnosis, with the proportion increasing to 63.0% for those aged 51–55. Similarly, after diagnosis, the proportion of patients receiving NSAID dispensations increased from 34.0% to 55.0%. However, for those older than 55 years, the proportion decreased with age from 62.0% of patients aged 56–60 to 39.0% of patients aged 86–90. After diagnosis, 54.0% of patients aged 56–60 and 30.0% of patients aged 86–90 received NSAIDs (Fig. 2).
Not only the pattern of NSAID dispensation but also the mean DDD of NSAIDs varied by age group and ATC class among patients with OA. In the following, we focused on the most representative ATC classes: M01AB (including diclofenac), M01AE (including ibuprofen; naproxen), and M01AH (COX-2-selective inhibitors, including celecoxib) (Table 2).
Table 2
Most representative ATC classes for NSAIDs
ATC class
Examples of included NSAIDs
M01AB—Acetic acid derivatives and related substances
Diclofenac
M01AE—Propionic acid derivatives
Ibuprofen; naproxen
M01AH—COX-2-selective inhibitors (coxibs)
Celecoxib
NSAIDs non-steroidal anti-inflammatory drugs
ATC class M01AE consistently had the highest DDD across all age groups in Norway, Sweden, and Finland (Fig. 3). In most age groups, for this ATC, the DDDs were higher after diagnosis, except for older patients, where DDDs were similar (starting from > 75 years in Norway) or reduced (starting from > 75 years in Finland and > 60 in Sweden) after diagnosis.
Fig. 3
Mean defined daily dose of NSAID dispensations across NSAID users in OA cohort before and after diagnosis, by type of NSAID and age group (18–90 years). M01AB includes diclofenac, M01AE includes ibuprofen and naproxen, M01AH COX-2-selective inhibitors (coxibs), NSAID non-steroidal anti-inflammatory drugs, OA osteoarthritis
Trends for M01AB and M01AH were similar in Norway and Sweden, with M01AB consistently higher than M01AH. However, whereas DDDs for M01AB decreased after diagnosis in patients with OA > 50 years of age in both countries, M01AH showed higher DDDs after diagnosis across all age groups. Interestingly, in Finland, DDDs for M01AH class were consistently higher than M01AB, with slight difference in M01AH before and after diagnosis. DDDs for M01AB were reduced after diagnosis in patients with OA > 50 years of age, similarly to what was observed for Norway and Sweden. This analysis showed that despite a lower proportion of patients with OA with NSAID dispensations after diagnosis, overall, the mean defined daily dose per prescription was higher (Fig. 3).
Secondary Outcomes
Comorbidities During 3-Year Pre-diagnosis Period
Patients with OA had a consistently higher comorbidity burden than matched individuals, with higher Elixhauser Index scores across all countries (Norway 0.68 vs. 0.58; Sweden 0.66 vs. 0.49; Finland 0.61 vs. 0.41) (Table 1). Differences were notable for CV diseases and risk factors (hypertension, diabetes, obesity), rheumatoid arthritis, and depression. For example, uncomplicated hypertension affected 10.9–16.6% of patients with OA versus 7.7–11.8% in matches, uncomplicated diabetes 3.9–5.9% vs. 2.8–5.0%, obesity 2.0–2.3% vs. 0.7–0.8%, and depression 2.8–3.4% vs. 1.9–2.8% (Supplementary Table 5). Overall, the prevalence of listed comorbidities might be underestimated, as primary care diagnosis data were not captured.
GI Complications
Overall, across all countries, more patients with OA had GI complications as compared to matched individuals, both before and after diagnosis. In the year before diagnosis, 3442 patients with OA (1.8%) experienced at least one GI complication in Norway, compared to 2520 matches (1.3%); after diagnosis the numbers reached 3931 (2.1%) and 2777 (1.5%), respectively. A similar trend was observed in Sweden (before diagnosis 5967 [1.7%] vs. 4063 [1.2%]; after diagnosis 6933 [2.0%] vs. 4516 [1.3%]) and in Finland (before diagnosis 4023 [1.8%] vs. 2500 [1.1%]; after diagnosis 4641 [2.1%] vs. 2898 [1.3%]). When looking at trends by age groups, the difference between patients with OA and matched individuals emerged around the age of 40 years and was apparent in all following age groups. For individuals older than 40 years, the proportion of individuals with GI complications increased somewhat linearly with age, with a similar trend observed across countries in both patients (up to 4.0%) and matches (up to 2.6%). However, the slope of this linear trend was steeper for patients with OA compared to matched individuals, indicating higher increases of GI complications with age in patients with OA compared to matches, both before and after diagnosis (Fig. 4).
Fig. 4
Proportion of individuals with GI complications before and after diagnosis, by cohort (patients with osteoarthritis and matched comparisons) and age group (18–90 years). GI gastrointestinal, OA osteoarthritis
Data on PPI dispensation was only available for Norway. Overall, approximately twice the number of patients with OA had PPIs dispensed compared to matched individuals, both before (26.3% vs. 14.1%) and after (28.7% vs. 15.6%) diagnosis (Fig. 5). Proportion of PPI dispensation increased with age in patients with OA and matches. Despite the higher number of PPI dispensations among the group of patients with OA, a larger proportion experienced GI complications (before diagnosis 1.8% vs. 1.3%; after diagnosis 2.1% vs. 1.5%). Figure 5b focuses on patients with OA with and without NSAID dispensation. About twice the proportion of patients with NSAID dispensations also received PPI dispensations, both before (32.2% vs. 18.5%) and after (38.2% vs. 19.6%) diagnosis, compared to patients without NSAID dispensation. The proportion of GI complications was lower among patients with NSAID dispensations as compared to those without (before diagnosis 1.7% vs. 2.0%; after diagnosis 1.8% vs. 2.3%).
Fig. 5
a Proportion of individuals with PPI dispensations or GI complications, before or after diagnosis, by cohort (patients with osteoarthritis and matched comparisons). b Proportion of patients with osteoarthritis with PPI dispensations or GI complications, before or after diagnosis, by cohort (with or without NSAID dispensed). Data available for Norway only. GI gastrointestinal, NSAID non-steroidal anti-inflammatory drugs, OA osteoarthritis, PPI proton pump inhibitors
Working-age (18–65 years) patients with OA had considerably more mean sick leave days in all age groups compared to matched individuals in all countries. This difference was already apparent before diagnosis, with on average three to five times the number of sick leave days compared to matches (Norway 36.4 vs. 10.7; Sweden 23.7 vs. 7.3; Finland 28.1 vs. 5.6). This difference became even more prominent with four to seven times the number of sick leave days after diagnosis (Norway 45.3 vs. 10.6; Sweden 40.0 vs. 7.9; Finland 44.4 vs. 5.9) (Fig. 6).
Fig. 6
Number of sick days, by cohort (patients with osteoarthritis and matched comparisons) and age group (18–65 years), before and after diagnosis. OA, osteoarthritis
Overall, a higher proportion of patients from the OA cohorts had hip or knee joint replacements across all age groups as compared to the general population matches, especially after diagnosis (Norway 14.7% vs. 0.2%; Sweden 15.8% vs. 0.1%; Finland 16.8% vs. 0.1%). Notably, this trend was evident also in younger patients, with a higher proportion of patients below 60 years of age that underwent replacement surgeries in the year after diagnosis in contrast to matched individuals (Norway 6.7% vs. 0.0%; Sweden 6.6% vs. 0.0%; Finland 6.5% vs. 0.0%). Overall, across all countries, the proportion of patients with OA who had hip or knee joint replacement surgeries after diagnosis increased with age, peaking at the 75–80 age group, and decreasing thereafter (Fig. 7).
Fig. 7
Proportion of individuals with hip/knee replacements before and after diagnosis, by cohort (patients with osteoarthritis and matched comparisons) and age group (18–90 years). OA osteoarthritis
This study offers several important insights into the burden of OA in Norway, Sweden, and Finland, highlighting key patterns in NSAID usage, comorbidities, GI complications, work absence, and joint replacement surgery.
Patient Characteristics
Our study included a notable proportion of younger patients with OA; however, the majority of patients were older, consistent with the well-documented relationship between OA prevalence and increasing age [31, 32]. While aging is a key factor in the onset and progression of OA, the findings from our study highlight that OA impacts patients across a wide age spectrum, with 34.1–40.4% of patients being under 60 years of age. This is consistent with a recent descriptive analysis conducted in Canada, which reported that 55.4% of individuals with OA were younger than 65 years. The average age at diagnosis was 50 years, with 30.4% of individuals being diagnosed before the age of 45 [8]. These findings underscore that OA is not solely a condition affecting the elderly; rather, it is prevalent among younger and middle-aged adults as well.
Anzeige
Comorbidities
In our study, the most common comorbidities among patients with OA were cardiovascular diseases such as uncomplicated hypertension. This finding is consistent with a systematic review and meta-analysis of observational studies, which also identified hypertension as the most prevalent disease-specific comorbidity (50.0%) and CV diseases as a major system-wise comorbidity in patients with OA [33].
Obesity was also highly prevalent. In fact, obesity diagnoses were twice as common in the OA cohort compared with controls across all countries. Obesity may further contribute to the increased CV risk, as together with metabolic syndrome, they are common shared risk factors for both OA and CV conditions [33].
NSAID Treatment Patterns
Overall, NSAID dispensation rates were higher in the OA cohort compared to the matched comparison cohort, across all age groups. In the OA cohort, the proportion of patients with NSAID dispensations was lower after diagnosis compared to before (with Finland as a partial exception; see Fig. 2). This may be due to several reasons: (1) older patients undergoing surgical intervention (post-consultation) may no longer rely heavily on NSAIDs for pain management; (2) long-term risks associated with NSAID use such as GI and CV risks, which may prompt clinicians to explore alternative or tailored therapies as the disease progresses [34]. It is also possible that treatment approaches differ between general practitioners and specialists, with specialists potentially opting for more specific treatments after diagnosis.
Despite the overall observation that fewer patients were dispensed NSAIDs after diagnosis, the results for the defined daily dose reveal an interesting trend where, for certain NSAID categories, the DDDs were higher after diagnosis, especially in patients below 70 years of age. This may reflect a shift towards dispensing more potent doses or longer durations of NSAIDs for patients who continued to require pain management, potentially indicating a need for more aggressive treatment. In patients above 70 years of age, the differences in DDDs after diagnosis were less prominent, pointing to a different dispensing pattern or possibly a more conservative approach to NSAID use in older age groups.
Anzeige
These findings underscore the complexity of NSAID usage in OA treatment, which could have implications for patient safety, long-term management strategies, and the risk of adverse effects. Especially given the higher prevalence of cardiovascular comorbidities and risk factors in this patient group, careful consideration should be given to the choice of NSAID to be dispensed, as NSAIDs are a non-homogeneous drug class with notable differences in terms of adverse event risk for GI and CV complications [35, 36]. Among the NSAIDs categories, in terms of DDD, the ATC class M01AE was the most dispensed and DDDs overall were even higher after diagnosis, whereas DDDs overall were lower after diagnosis for the M01AB class. This is consistent with literature suggesting that ibuprofen and in particular naproxen are often preferred over diclofenac because of lower associated CV risk [37, 38]. Interestingly also DDDs for M01AH class, containing COX-2-selective anti-inflammatory drugs, were higher after diagnosis, potentially indicating a shift of treatment after diagnosis in specialty care. It is generally recommended to use the lowest effective dose of NSAID, with a flexible dosing regimen tailored to the intensity of pain and stiffness [39].
GI Complications
As a result of the high prevalence of OA, small differences in proportions of individuals with at least one GI complication between patients with OA and matched individuals translated to a considerable 922–1904 additional GI complications registered in patients with OA before diagnosis and 1154–2417 after, expressed as ranges across countries. These findings further highlight the substantial burden posed by the increased risk of GI complications in patients with OA compared to matched individuals. When looking at trends by age groups, there were consistently more individuals with GI complications in the OA cohorts compared to matched comparison cohorts at all ages, with an approximately linear increase observed from age 40 onwards, emphasizing the need to monitor GI risks not only in older patients but also in the younger population.
Proton Pump Inhibitor Dispensation
In Norway, PPIs were dispensed to more patients with OA receiving NSAIDs than those not receiving NSAIDs. This coincided with relatively lower rates of GI complications in NSAID-treated patients. The high rates of NSAID dispensations, combined with a higher comorbidity score in patients with OA compared to matched individuals, underscore the need for careful evaluation of GI and CV risks when prescribing NSAIDs. Clinicians should consider tailoring NSAID selection and co-prescribing PPIs to mitigate GI risks, with an emphasis on addressing these risks even in the younger population.
Sick Leave
Overall, patients with OA had considerably more sick days than matched individuals. This was evident even in the youngest age groups and before the index date, pointing to a high burden of disease already before diagnosis. After diagnosis, the mean number of sick days further increased. Scandinavian registry studies report approximately 2.8-fold higher sick leave days in patients with OA than matched controls (Finland) [40] and substantially high sick leave in patients before and after joint replacement (Sweden) [41]. In another Swedish study, Hubertsson et al. reported an almost twofold higher risk of sick leave in patients with OA compared with the general population [42]. This underscores the economic burden of OA, particularly in working-age individuals [6, 40, 41, 43, 44], as also evidenced by high number of patients undergoing hip or knee replacements, and the occurrence of GI complications.
Anzeige
For younger patients, the higher number of sick days could reflect the challenges of managing OA symptoms while maintaining an active work life, possibly due to the onset of more severe symptoms or the need for adjustments in treatment. This pattern suggests that OA may begin to have a more pronounced impact on daily functioning and productivity in younger individuals, especially before the age of 60, when they are often in the midst of their careers. These findings emphasize the need for tailored treatment plans that consider not only the clinical management of OA but also its impact on daily life, particularly in younger patients who are still engaged in the workforce. Further investigation into the reasons behind these trends could provide valuable insights into improving work-related outcomes for individuals with OA.
Hip/Knee Replacements
Given that OA progresses over time, the increased incidence of joint replacements with age is not surprising, as joint deterioration tends to worsen over time, leading to surgical interventions. However, the timing of such surgeries is influenced not only by disease onset or progression but also by considerations of implant durability [45] and the risks associated with revision procedures. National arthroplasty registries have been instrumental in characterizing long-term outcomes in real-world practice. The Nordic Arthroplasty Register Association (NARA), which aggregates data from Sweden, Denmark, Norway, and Finland, reports that among total knee arthroplasties (TKAs) for OA performed between 2000 and 2017, the 15-year cumulative revision rate was 6.6% in Sweden, compared to higher rates in Denmark (9.6%), Norway (9.1%), and Finland (7.0%) [46]. These findings highlight that although revisions are relatively infrequent, they are clinically important and vary across healthcare systems.
Revision procedures are more difficult, particularly in older adults with potential bone deterioration; thus, joint replacements are often performed later in life when patients are less likely to require additional surgeries, allowing them to benefit from the prosthetic without the need for future revisions. This underscores the complexity of treatment decisions, where the timing of joint replacement is guided not just by disease progression but also by considerations of long-term outcomes and surgical risks. Interestingly, despite OA being more common in older individuals, a higher proportion of younger patients with OA underwent joint replacement surgeries compared to a matched comparison cohort, demonstrating the increased burden of OA also in younger populations.
Limitations
Data Source and Generalizability
This study relies on register-based data, which typically has high completeness due to mandatory reporting of many variables in national health care registers. While this minimizes missing data related to health care visits and dispensations, any unrecorded events would be undetectable.
The study includes patients diagnosed with OA in specialty care excluding primary care data, meaning the date of the first diagnosis may reflect the referral date. OA is often a clinical diagnosis initiated and managed in primary care, which could affect generalizability [47, 48]. Patients referred to specialist care could be those with more severe health problems that cannot be sufficiently managed in primary care.
The study period ends in 2018 and does not capture the most recent potential changes in prescription patterns and safety events over time. However, we do not anticipate significant changes in these patterns since the data were collected.
As the data are Nordic registry-based, findings may not be generalizable to other healthcare systems or populations with different demographic or prescribing patterns.
Confounding Factors and Statistical Approach
Patients and comparators were matched on age and sex, but residual confounding remains possible. In particular, we did not account for area-level factors such as socioeconomic deprivation, or individual-level factors such as BMI, smoking, and lifestyle, which may influence both treatment patterns and outcomes.
Given the exploratory nature of the study analysing the current situation for patients with OA in the Nordic countries only descriptive statistics are presented. The focus of the analyses is on identifying trends that can inform clinical judgment and may provide valuable insights without the need for hypothesis testing.
Clinical and Treatment Data Limitations
NSAID treatment dates are based on dispensation fills, not actual administration. Therefore, the results do not capture adherence and actual consumption. Additionally, while dispensed NSAIDs and PPIs were recorded, over the counter (OTC) use was not captured, potentially underestimating total NSAID and PPI consumption. OTC NSAID use is common in Norway [49] and Nordic utilization data confirm substantial non-prescription consumption across Norway, Sweden, and Finland [50]. PPI dispensations are not covered in the registers in Finland and Sweden; however, the similarity in NSAID dispensations and GI complications observed in Norway may suggest comparable dispensing behaviours in those countries.
Coverage of GI complications was limited to hospital-related visits, potentially underestimating the total number of GI events occurring in non-hospital settings, such as primary care. As the study was descriptive, no causal relationships are claimed. For instance, patients at lower risk for GI complications may be preferentially selected for NSAID treatment. Therefore, no conclusions on the safety of NSAIDs or the causes of GI events can be drawn. Future studies should apply causal methods and include primary care data to clarify these associations.
Sick Leave
The first day of sick leave is registered differently across the Nordic countries, which may result in an underestimation of work absences, particularly in case of short-term sick leave. This may underestimate the total burden of work loss, especially among working-age patients.
Conclusions
In this large Nordic registry study, a higher proportion of patients with OA, particularly those ≤ 60 years, receive NSAID dispensations compared to matched individuals, even before the first hospital visit with OA diagnosis. The risk of GI complications was notably higher in patients with OA than in matched individuals at all ages, including among the younger OA population, underscoring the importance of risk management. The additional burden of sick leave and joint replacements highlights the need for early intervention and comprehensive management strategies even in younger populations.
Anzeige
Acknowledgements
We thank the individuals whose data contributed to this research.
Medical Writing/Editorial Assistance
Professional medical writing and editorial assistance in the preparation of this article was provided by Kripi Syal, PhD, of Quantify Research (https://quantifyresearch.com/). Medical writing and editorial assistance service was funded by Grünenthal.
Declarations
Conflict of Interest
Tore K Kvien has received fees for speaking and/or consulting during last 2 years from AbbVie, Almirall, Astra Zeneca, Janssen and Sandoz, and received research funding to Diakonhjemmet Hospital from AbbVie, Galapagos, Novartis and Pfizer. Manfred Gross has received fees for speaking and/or consulting during last 2 years from Bristol-Myers Squibb, Janssen-Cilag, Luvos Heilerde, Merz Pharmaceuticals, Norgine, Pfizer, Reckitt Benckiser, Weber & Weber. Johan L Hansen is an employee of Quantify Research. Quantify Research, part of the Athagoras group, provides consulting and research services to the life science industry. Jan Jastorff, Robert Böttger, Rita Freitas, Elena Mascia are employees of Grünenthal GmbH or affiliates.
Ethical Approval
Ethical approval was obtained from the ethical review board in South-East Norway (28745) and the regional ethical review board in Stockholm (2018/1634-31/2) for Norway and Sweden, respectively. The use of Finnish data did not require ethical approval.
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License, which permits any non-commercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc/4.0/.
Arden N, Blanco FJ, Bruyère O, et al. Atlas Of Osteoarthritis, 2 ed. London: Springer Healthcare; 2018.
2.
Berenbaum F. Osteoarthritis as an inflammatory disease (osteoarthritis is not osteoarthrosis!). Osteoarth Cartil. 2013;21(1):16–21.CrossRef
3.
Sellam J, Berenbaum F. The role of synovitis in pathophysiology and clinical symptoms of osteoarthritis. Nat Rev Rheumatol. 2010;6(11):625–35.CrossRefPubMed
Kiadaliri AA, Lohmander LS, Moradi-Lakeh M, Petersson IF, Englund M. High and rising burden of hip and knee osteoarthritis in the Nordic region, 1990–2015. Acta Orthop. 2018;89(2):177–83. https://doi.org/10.1080/17453674.2017.1404791.CrossRefPubMed
6.
Hallberg S, Rolfson O, Karppinen J, et al. Economic burden of osteoarthritis—multi-country estimates of direct and indirect costs from the BISCUITS study. Scand J Pain. 2023;23(4):694–704. https://doi.org/10.1515/sjpain-2023-0015.CrossRefPubMed
7.
Driban JB, Harkey MS, Liu S-H, Salzler M, McAlindon TE. Osteoarthritis and aging: young adults with osteoarthritis. Curr Epidemiol Rep. 2020;7:9–15.CrossRef
8.
Wilfong JM, Badley EM, Perruccio AV. Old before their time? The impact of osteoarthritis on younger adults. Arthritis Care Res (Hoboken). 2024;76(10):1400–8. https://doi.org/10.1002/acr.25374.CrossRefPubMed
9.
Hawker GA, King LK. The burden of osteoarthritis in older adults. Clin Geriatr Med. 2022;38(2):181–92.CrossRefPubMed
Foreningen for utgivelse av Norsk legemiddelhåndbok [The Association for the publication of the Norwegian Drug Handbook]. Artrose [Osteoarthrosis] [updated 2016-06-10]. Available from: http://legemiddelhandboka.no/Terapi/21860/?ids=21861#i21861. Accessed 24 Sep 2024.
Bannuru RR, Osani MC, Vaysbrot EE, et al. OARSI guidelines for the non-surgical management of knee, hip, and polyarticular osteoarthritis. Osteoarthritis Cartilage. 2019;27(11):1578–89. https://doi.org/10.1016/j.joca.2019.06.011.CrossRefPubMed
17.
Chaplin S. NICE on the diagnosis and management of osteoarthritis. Prescriber. 2023;34(2):15–6.CrossRef
da Costa BR, Pereira TV, Saadat P, et al. Effectiveness and safety of non-steroidal anti-inflammatory drugs and opioid treatment for knee and hip osteoarthritis: network meta-analysis. BMJ. 2021;375:n2321. https://doi.org/10.1136/bmj.n2321.CrossRefPubMedPubMedCentral
20.
da Costa BR, Reichenbach S, Keller N, et al. Effectiveness of non-steroidal anti-inflammatory drugs for the treatment of pain in knee and hip osteoarthritis: a network meta-analysis. Lancet. 2017;390(10090):e21–33. https://doi.org/10.1016/s0140-6736(17)31744-0.CrossRefPubMed
21.
Sostres C, Gargallo CJ, Lanas A. Nonsteroidal anti-inflammatory drugs and upper and lower gastrointestinal mucosal damage. Arthritis Res Ther. 2013;15:1–8.CrossRef
22.
Straube S, Tramèr MR, Moore RA, Derry S, McQuay HJ. Mortality with upper gastrointestinal bleeding and perforation: effects of time and NSAID use. BMC Gastroenterol. 2009;9:1–7.CrossRef
23.
Lau JY, Sung J, Hill C, Henderson C, Howden CW, Metz DC. Systematic review of the epidemiology of complicated peptic ulcer disease: incidence, recurrence, risk factors and mortality. Digestion. 2011;84(2):102–13.CrossRefPubMed
24.
van de Laar MA, Schöfl R, Prevoo M, Jastorff J. Predictive value of gastrointestinal symptoms and patient risk factors for NSAID-associated gastrointestinal ulcers defined by endoscopy? Insights from a pooled analysis of two naproxen clinical trials. PLoS One. 2023;18(4):e0284358.CrossRefPubMedPubMedCentral
Anglin R, Yuan Y, Moayyedi P, Tse F, Armstrong D, Leontiadis GI. Risk of upper gastrointestinal bleeding with selective serotonin reuptake inhibitors with or without concurrent nonsteroidal anti-inflammatory use: a systematic review and meta-analysis. Am J Gastroenterol. 2014;109(6):811–9.CrossRefPubMed
27.
Elixhauser A, Steiner C, Harris DR, Coffey RM. Comorbidity measures for use with administrative data. Med Care. 1998;36(1):8–27.CrossRefPubMed
28.
Button L, Roberts S, Evans P, et al. Hospitalized incidence and case fatality for upper gastrointestinal bleeding from 1999 to 2007: a record linkage study. Aliment Pharmacol Ther. 2011;33(1):64–76.CrossRefPubMed
29.
Park J, Kwon S, Choi E-K, et al. Validation of diagnostic codes of major clinical outcomes in a National Health Insurance database. Int J Arrhythmia. 2019;20(1):1–7.CrossRef
30.
The Nordic Medico-Statistical Committee (NOMESCO). NOMESCO classification of surgical procedures (NCSP), Version 1.14. 2009.
31.
Felson DT, Naimark A, Anderson J, Kazis L, Castelli W, Meenan RF. The prevalence of knee osteoarthritis in the elderly. The Framingham osteoarthritis study. Arthritis Rheum. 1987;30(8):914–8. https://doi.org/10.1002/art.1780300811.CrossRefPubMed
32.
Felson DT, Zhang Y, Hannan MT, et al. The incidence and natural history of knee osteoarthritis in the elderly. The Framingham osteoarthritis study. Arthritis Rheum. 1995;38(10):1500–5. https://doi.org/10.1002/art.1780381017.CrossRefPubMed
33.
Swain S, Sarmanova A, Coupland C, Doherty M, Zhang W. Comorbidities in osteoarthritis: a systematic review and meta-analysis of observational studies. Arthritis Care Res (Hoboken). 2020;72(7):991–1000. https://doi.org/10.1002/acr.24008.CrossRefPubMed
34.
Sabzwari SR, Qidwai W, Bhanji S. Polypharmacy in elderly: a cautious trail to tread. J Pak Med Assoc. 2013;63(5):624–7.PubMed
35.
Pelletier JP, Martel-Pelletier J, Rannou F, Cooper C. Efficacy and safety of oral NSAIDs and analgesics in the management of osteoarthritis: evidence from real-life setting trials and surveys. Semin Arthritis Rheum. 2016;45(4 Suppl):S22-27. https://doi.org/10.1016/j.semarthrit.2015.11.009.CrossRefPubMed
Schmidt M, Lamberts M, Olsen AM, et al. Cardiovascular safety of non-aspirin non-steroidal anti-inflammatory drugs: review and position paper by the working group for Cardiovascular Pharmacotherapy of the European Society of Cardiology. Eur Heart J. 2016;37(13):1015–23. https://doi.org/10.1093/eurheartj/ehv505.CrossRefPubMed
38.
Bhala N, Emberson J, Merhi A, et al. Vascular and upper gastrointestinal effects of non-steroidal anti-inflammatory drugs: meta-analyses of individual participant data from randomised trials. Lancet. 2013;382(9894):769–79. https://doi.org/10.1016/s0140-6736(13)60900-9.CrossRefPubMed
39.
Kvien TK, Brørs O, Staff PH, Rognstad S, Nordby J. Improved cost-effectiveness ratio with a patient self-adjusted naproxen dosing regimen in osteoarthritis treatment. Scand J Rheumatol. 1991;20(4):280–7. https://doi.org/10.3109/03009749109096801.CrossRefPubMed
Stigmar K, Dahlberg LE, Zhou C, Jacobson Lidgren H, Petersson IF, Englund M. Sick leave in Sweden before and after total joint replacement in hip and knee osteoarthritis patients. Acta Orthop. 2017;88(2):152–7. https://doi.org/10.1080/17453674.2016.1269051.CrossRefPubMed
42.
Hubertsson J, Petersson IF, Thorstensson CA, Englund M. Risk of sick leave and disability pension in working-age women and men with knee osteoarthritis. Ann Rheum Dis. 2013;72(3):401–5.CrossRefPubMed
Sicras-Mainar A, Tornero-Tornero JC, Vargas-Negrín F, Lizarraga I, Sicras-Navarro A, Rejas-Gutierrez J. Sick leave and costs in active workers with chronic osteoarthritis pain in Spain: outcomes of the OPIOIDS real world study. Open Access Rheumatol. 2022;14:25–38. https://doi.org/10.2147/oarrr.S346746.CrossRefPubMedPubMedCentral
45.
Pakarinen O, Ponkilainen V, Varnum C, et al. Choice of bearings influences the implant survival of total hip arthroplasty in patients who have osteoarthritis aged 55 years or older: results of 158,044 patients from the Nordic Arthroplasty Register Association from 2005 to 2017. J Arthroplasty. 2025;40(2):437-442.e432. https://doi.org/10.1016/j.arth.2024.08.015.CrossRefPubMed
46.
Irmola T, Ponkilainen V, Mäkelä KT, et al. Impact of Nordic Arthroplasty Register Association (NARA) collaboration on demographics, methods and revision rates in knee arthroplasty: a register-based study from NARA 2000–2017. Acta Orthop. 2022;93:866–873. https://doi.org/10.2340/17453674.2022.5256.
Dale O, Borchgrevink PC, Fredheim OMS, Mahic M, Romundstad P, Skurtveit S. Prevalence of use of non-prescription analgesics in the Norwegian HUNT3 population: impact of gender, age, exercise and prescription of opioids. BMC Public Health. 2015;15(1):461. https://doi.org/10.1186/s12889-015-1774-6.CrossRefPubMedPubMedCentral
50.
Kristensen KB, Karlstad Ø, Martikainen JE, et al. Nonaspirin nonsteroidal antiinflammatory drug use in the Nordic countries from a cardiovascular risk perspective, 2000–2016: a drug utilization study. Pharmacother J Hum Pharmacol Drug Therapy. 2019;39(2):150–160. https://doi.org/10.1002/phar.2217.
Eine gepoolte Analyse von elf prospektiven Studien liefert eine umfassende Bestandsaufnahme zu Zusammenhängen von verschiedenen Kostformen und dem Krebsrisiko. Neben bekannten Vorteilen von eher pflanzenbasierten Kostformen fanden sich überraschenderweise auch Hinweise auf einzelne ungünstige Effekte.
Hohe Wochendosierungen Erythropoese-stimulierender Substanzen erhöhen möglicherweise das Krebsrisiko von dialysepflichtigen Personen: In einer Analyse aus Korea traten neue Tumorerkrankungen bei hohen Dosierungen um etwa ein Viertel häufiger auf.
Relevante Koronarstenosen vor einer Transkatheter-Aortenklappen-Intervention (TAVI) interventionell statt rein medikamentös zu behandeln, bietet Registerdaten zufolge einen überschaubaren klinischen Nutzen zum Preis eines erhöhten Blutungsrisikos.
