Evaluating the Real-World Use of Topical Diclofenac Sodium Gel 1% Using US Longitudinal Electronic Health Records Database: A study supporting OTC switch

Abstract

Introduction

Musculoskeletal conditions are a significant health challenge and second leading cause of disability worldwide. Diclofenac sodium topical gel 1% (DSG1%) provides effective relief of arthritis pain. While clinical studies show that DSG1% is safe and well-tolerated, long-term safety and tolerability in real-world settings are limited. This study aimed to profile users and prescribers of DSG1% and to evaluate its safety and tolerability over a screening period of 8.5 years with an average follow-up of nearly 1 year. The focus was on patients with risk factors and comorbidities, especially those taking concomitant medication in addition to topical nonsteroidal anti-inflammatory drugs (NSAIDs).

Methods

This retrospective, longitudinal cohort study used the US Department of Defense (DoD) electronic health records (EHR) database. The database included 521,593 individuals with ≥ 1 prescription for DSG1% for either indicated or non-indicated conditions with mean (standard deviation; SD) follow-up of 348.4 (562.4) days. The primary outcome measure assessed the incidence of predefined events of interest (EOIs), including gastrointestinal, hepatic, renal diseases, cardiovascular events, hypertension, skin reaction, misuse, abuse, and death (all-cause mortality).

Results

The average age of subjects was 56.7 years (SD = 18.1), with women comprising 60.4% of population. During study-period, 74.2% of subjects experienced no adverse EOIs after initiating treatment with DSG1%. Among the remaining 25.8%, average time to first EOI was 244.0 (SD = 368.6) days. Notably, the frequency of reported EOIs increased with age. Additionally, subjects with conditions such as rheumatoid arthritis, systemic lupus erythematosus, or diabetes had a higher incidence of cardiovascular EOIs.

Conclusions

The study results indicate that DSG demonstrated a favorable safety profile, particularly for patients with comorbidities and high-risk factors and when used with other medications. Despite an older population and high baseline risk factors (93%), only 26% DSG1% users experienced a predefined EOI, observed on average 244.0 (368.6) days from index date. These findings confirm the long-term tolerability of topical DSG1% for musculoskeletal disorders.

Supplementary Information

The online version contains supplementary material available at 10.1007/s40122-025-00723-9.

Key Summary Points

Why carry out this study?

Musculoskeletal (MSK) conditions, including low back pain (LBP) and osteoarthritis (OA), ranked as the second most prevalent cause of non-fatal disability, affecting 1.71 billion people globally in 2019.

Current treatment guidelines underscore the need for safe and effective pain management strategies in OA treatment. Topical nonsteroidal anti-inflammatory drugs (NSAIDs), like diclofenac sodium topical gel 1% (DSG1%), are frequently recommended as a first-line treatment due to their lower systemic exposure and improved safety profile.

Despite its widespread use, comprehensive long-term safety data for DSG1% in real-world applications remain scarce. Real-world evidence (RWE) is vital for evaluating safety in practical use.

This study investigated the safety and tolerability of DSG1% using the US Department of Defense (DoD) database over medium-to-long-term use, especially when used along with concomitant medications other than topical non-steroidal anti-inflammatory drugs (NSAIDs)

What was learned from this study?

There was no increase in events of interest (EOIs) observed during DSG1% therapy compared to baseline, which further reinforces the long-term tolerability of topical DSG1%

Utilizing a comprehensive database like the DOD in a real-world setting enhances data quality, increases generalizability, and supports well-informed regulatory decisions, including over-the-counter (OTC) switches.

Introduction

Musculoskeletal (MSK) conditions pose a substantial global health challenge, ranking as the second most common cause of disability worldwide in terms of years lived with disability. These conditions significantly impair mobility and dexterity, making them a notable contributor to reduced health-related quality of life (HRQoL) [1, 2]. Disability resulting from MSK diseases has increased by 24.6% from 2010 to 2021, primarily due to low back pain, neck pain and osteoarthritis (OA), which is the most prevalent form of arthritis [3].

Current international guidelines and definitions of pain emphasize that effective treatment should extend beyond merely alleviating pain intensity [4]. The emphasis must shift toward enhancing overall well-being and improving the patient's capacity to participate in daily activities [5]. Therefore, a multimodal and multidisciplinary approach that combines pharmacological, non-pharmacological, and interventional therapies is important to effectively manage MSK pain. For patients experiencing chronic MSK pain, the initial focus should be on non-pharmacological treatments, such as home exercises and rehabilitation programs. If these strategies do not yield adequate relief, non-steroidal anti-inflammatory drugs (NSAIDs) are typically recommended as the first-line pharmacological option [5–7]. NSAIDs are commonly used to manage pain and inflammation associated with MSK disorders along with other inflammatory disease. While oral NSAIDs offer systemic pain relief, they come with potential risks of gastrointestinal (GI), cardiovascular (CV), renal, and hepatic adverse events [8, 9]. Researchers are increasingly focusing on developing new scaffolds, like pyridazine and thiazole derivatives, to address the need for safer therapeutic options [10–15]. In this context, topical NSAIDs have emerged as a promising alternative, offering targeted pain relief with fewer systemic side effects. Topical NSAIDs offer localized action, reduced systemic exposure, and fewer drug interactions, making them a preferred treatment option for patients with MSK diseases who have comorbidities and are concurrently taking other medications [16–18]. Guidelines such as the American College of Physicians and the American Academy of Family Physicians recommend using topical NSAIDs as first-line therapy for acute pain from non–low back and MSK injuries. Additionally, clinical guidelines from the American College of Rheumatology and the Osteoarthritis Research Society International strongly recommend topical NSAIDs for knee osteoarthritis patients and conditionally recommend topical NSAIDs for patients with hand OA [4, 19].

Of the available topical NSAIDs, diclofenac sodium topical gel 1% (DSG1%) is an over-the-counter (OTC) NSAID indicated for the temporary relief of arthritis pain of the hand, wrist, elbow, foot, ankle, and knee [20]. It is a preferred option for targeted treatment, aligning with guidelines recommending topical NSAIDs as a first-line choice for OA pain management, especially when a few or small joints are affected [16, 21–25]. DSG1% became the first topical NSAID approved by the U.S. Food and Drug Administration as a prescription product in 2007 and as an OTC product in 2020. Topical NSAIDs, such as DSG1%, can provide local pain relief as effectively as oral NSAIDs, and also offer the benefit of comparatively limited systemic exposure [17], thus reducing the risk of adverse events. In clinical studies, DSG1% was shown to be safe and well-tolerated in patients with hand or knee OA for up to 8 weeks’ and 12 weeks’ duration, respectively [26]. Localized skin irritation or allergic reactions at the application site were the commonly reported AEs observed in the clinical studies [22–26]. Thus, to ensure comprehensive safety assessments over long-term use, it is crucial to monitor for all potential side effects. This includes vigilance for any systemic effects which may arise, as the overall safety profile of topical NSAIDs warrants thorough investigation beyond localized reactions. Additionally, evidence regarding the long-term safety and tolerability of DSG1% in real-world settings is still limited specifically in patients having comorbidities and taking concomitant medication.

A recent study has highlighted that real-world evidence (RWE), derived from a robust database, has been utilized to support the switch of drugs like desogestrel and ulipristal acetate from Rx-only to OTC availability primarily based on the intrinsic safety of the drug itself and the safety associated with consumer usage. RWE enhances our understanding of safety and efficacy by reflecting real-world performance [27]. Thus, a long-term RWE study is instrumental to evaluate the safety of DSG1%, particularly regarding NSAID-related adverse events in vulnerable patient populations, such as the elderly and individuals with comorbidities. This study involves evaluating the risk of localized side effects, such as skin reactions, as well as more serious events including GI, CV, hepatic, and renal complications.

Building on the safe and localized pain-relieving effects of DSG, our hypothesis proposes that DSG1% maintains its safety and tolerability profile during medium- to long-term use for patients with musculoskeletal diseases. This is particularly relevant for those who are taking other medications (excluding topical NSAIDs) for comorbid conditions and those with risk factors. Therefore, this retrospective study was conducted with the aim of profiling the users and prescribers of DSG1% within a large, diverse US-based cohort. The study examined treatment patterns, prescribing practices, and the safety and tolerability of DSG1% by measuring events of interest (EOIs) among all users as well as among subgroups experiencing selected outcomes. The screening period covered 8.5 years, with an average follow-up time of nearly 1 year. This research provides valuable insights for researchers, healthcare practitioners, and patients by improving our understanding of the safety profile DSG1% in a real-world setting. It provides vital evidence that can inform future investigations on NSAID-related adverse events. Moreover, it aids clinicians in making informed decisions by balancing efficacy and safety, ultimately leading to improved pain-management strategies. As a result, patients experience enhanced quality of life and are empowered to make informed choices about their care. Furthermore, it contributes data that strengthens the regulatory decision regarding the Rx-to-OTC switch of DSG1% in the United States.

Methods

Data Source

This retrospective, longitudinal cohort study was conducted using the US Department of Defense (DoD) Military Health System (MHS) electronic health records (EHR) database. The MHS database encompasses healthcare and administrative data for care delivered across the network of military hospitals and clinics, as well as care delivered by civilian providers throughout the US purchased through TRICARE, the insurance arm of the DoD. The database includes over 30 billion archived records, 5 billion online health records spanning more than 11 years, 20 years of ePrescribing, and comparable decades of prescription, inpatient, and outpatient data. Records of veterans who receive medical coverage through the Veterans Administration are not captured in the DoD database. [28]. The study period was from January 1, 2007, through December 31, 2016, and consisted of an 8.5-year screening period for patient identification from January 1, 2008, through June 30, 2016, along with a 365-day pre-index, baseline period and minimum follow-up of 180 days (Fig. 1).

Fig. 1.

Illustration of the study design and timelines. MHS Military Health System

Patient Selection

A patient’s index date was defined as the date of the first DSG1% prescription fill within the 8.5-year screening period. Patients with ≥ 1 prescription for DSG1% were included in the study if they had ≥ 365 days of continuous eligibility in the MHS before the index date. Resumption of therapy with DSG1% was allowed; thus, patients could have multiple treatment episodes. Patients were excluded if they had any of the following: history of cancer, human immunodeficiency virus, major organ transplantation within the baseline period, or evidence of pregnancy among women within the 280-day period before or after the index date. Patient follow-up data were assessed from the day after the index date until the date of discontinuation, disenrollment from the MHS, death, or the end of the study follow-up period.

Baseline Variables

Baseline measurements included patient demographics, comorbidities, baseline risk factors, concomitant medications, and hospitalizations during the 12-month period prior to index. Comorbidities and risk factors were identified using diagnosis codes [International Statistical Classification of Diseases, Ninth Edition, Clinical Manifestation (ICD-9-CM)/International Statistical Classification of Diseases, Tenth Edition, Clinical Manifestation (ICD-10-CM)], procedure codes [ICD-9-CM/ICD-10-Procedure Coding System (PCS)], Current Procedural Terminology (CPT) codes, and medications as proxy.

Outcome Measure

The primary outcome measure was the occurrence of EOIs. The system organ classes associated with adverse reactions to NSAIDs were used to select the following eight prespecified EOIs to be included in the study: GI injury/disease, hepatic injury/disease, renal injury/disease, CV events/disease, incident or exacerbated hypertension (HTN), skin reaction outcomes, misuse and abuse, and all-cause mortality. As described in the above section, EOIs were identified in a similar manner using ICD-9/10-CM diagnosis codes and ICD-9-CM/ICD-10-PCS and CPT procedure codes, and medications as proxy.

The code sets used to identify the EOIs are detailed in Fig.2. The assessment of EOIs occurred from the day after the index date through the earliest occurrence of the following: last DSG1% treatment episode end date, disenrollment from the MHS, death, or the end of the study follow-up period. Additionally, EOIs were assessed across all follow-up, even for patients with gaps between therapy periods.

Fig. 2.

ICD-9-CM diagnoses in baseline records of all patients with DSG1% prescriptions by codebook chapter. DSG diclofenac sodium topical gel 1%, ICD-9-CM International Statistical Classification of Diseases, Ninth Edition, Clinical Manifestation

Compliance with Ethics Guidelines

Approval from an Institutional Review Board was obtained to access the DoD database. The study was conducted in accordance with the ethical principles laid down in the Declaration of Helsinki and complied with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement (Supplementary Material). All participants provided informed consent before study initiation.

Statistical Analysis

All patient characteristics were described using means and standard deviations (SD) for continuous variables and frequencies and percentages for categorical variables. Frequency distributions are also reported for selected continuous variables. All analyses were conducted using SAS version 9.3 (Cary, NC, USA).

Results

Cohort Characteristics

This large, retrospective cohort study consisting of 521,593 study-eligible patients is based on comprehensive EHR of patients receiving ≥ 1 fill of DSG1% between January 1, 2008, and June 30, 2016. The population (Table 1) had a mean age of 56.7 (18.1) years, was predominantly women (60.4%), and had a mean (SD) length of follow-up of 348.4 (562.4) days (Supplementary Table 1), which was primarily driven by the length of DSG1% exposure. This study represents a robust assessment of a population with OA and related conditions with well-documented demographics, medication use, and medical histories. The data showed 79.3% had baseline diagnoses of ≥ 1 MSK/connective tissue disorder. Of these, 32.5% of the study population had a baseline diagnosis of OA (Table 2; Fig. 2). OA specific to the knee, ankle, or foot was identified in17.1% of patients, and OA of the hand, wrist, or elbow was identified in 4.8% of patients (Table 2). Rheumatoid arthritis (RA) and ankylosing spondylitis (AS) were identified in 4.0% and 2.8% of the study population, respectively, with 64.7% of patients having no arthritis-indicated condition for DSG1% at baseline (Table 2). The other most common MSK/connective tissue disorder diagnoses included: arthropathies and related disorders (60.3% of patients); rheumatism, excluding the back (51.8% of patients); dorsopathies (41.3% of patients); and osteopathies, chondropathies, and acquired MSK deformities (24.8% of patients).

Table 1.

EOI cohorts stratified by baseline characteristics

Characteristics	All	No EOI	Hepatic EOI	CV EOI	HTN EOI	GI EOI	Renal EOI	Skin reaction EOI	Misuse/abuse EOI	Death
All patients, n (%)	521,593 (100)	387,177 (74.2)	19,124 (3.7)	63,512 (12.1)	23,618 (4.5)	11,010 (2.1)	17,158 (3.3)	59,133 (11.3)	5399 (1.0)	2210 (0.4)
Age on index date, mean (SD), years	56.7 (18.1)	54.1 (18.3)	62.8 (13.2)	70.5 (12.1)	63.2 (13.8)	64.2 (14.3)	70.2 (12.2)	61.6 (15.5)	50.6 (16.6)	76.8 (10.4)
Age, %
< 18 (n = 5,444)a	1.0	95.1	0.4	0.3	0.1	0.2	0.2	3.9	0.2	0
18–24 (n = 21,808)a	4.2	93.3	0.4	0.8	0.4	0.4	0.2	4.0	1.2	0.0
25–34 (n = 48,787)a	9.4	90.2	0.9	1.1	1.3	0.7	0.4	5.5	1.8	0.0
35–44 (n = 60,743)a	11.6	85.3	2.0	2.5	2.9	1.2	0.7	8.1	1.4	0.0
45–54 (n = 83,006)a	15.9	78.9	3.8	5.3	4.4	1.7	1.4	10.9	1.3	0.1
55–64 (n = 99,884)a	19.1	72.5	5.0	10.4	5.7	2.3	2.9	13.0	1.1	0.2
65–74 (n = 105,116aa	20.2	66.6	5.1	17.9	6.1	3.1	4.9	14.5	0.8	0.4
75–84 (n = 72,213)a	13.8	61.2	4.3	26.7	5.7	3.1	7.1	13.8	0.5	1.1
≥ 85 (n = 24,592)a	4.7	55.7	3.0	34.1	5.3	2.7	8.6	12.6	0.3	3.0
Sex, %
Female (n = 315,273)a	60.4	71.8	4.4	13.2	4.9	2.5	3.5	13.0	1.1	0.4
Male (n = 206,317)a	39.6	78.0	2.6	10.6	4.0	1.6	2.9	8.7	1.0	0.4
Race, %
White (n = 149,959)a	28.8	81.9	2.5	6.4	3.4	1.3	1.6	8.3	1.1	0.1
Black (n = 48,105)a	9.2	81.3	2.1	4.9	4.2	1.4	1.7	9.0	0.9	0.1
Asian/Pacific Islander (n = 11,681)a	2.2	82.9	2.0	3.5	3.6	1.2	1.1	9.3	1.0	0.0
American Indian/Alaska Native (n = 2121)a	0.4	82.2	2.4	6.1	3.2	1.7	1.4	8.3	1.4	0.1
Other (n = 7942)a	1.5	81.4	3.2	4.5	3.6	1.5	1.4	9.2	1.1	0.1
Unknown (n = 301,785)a	57.9	68.7	4.6	16.8	5.2	2.7	4.5	13.4	1.0	0.7

CV cardiovascular, EOI event of interest, GI gastrointestinal, HTN hypertension

^aTotal number of people within each subgroup (age, sex, race)

Table 2.

DSG1%-indicated conditions, other arthritis conditions, risk factors, and comorbidities stratified by EOI cohort

Characteristics	All	Hepatic EOI	CV EOI	HTN EOI	GI EOI	Renal EOI	Skin reaction EOI	Misuse/abuse EOI	Death
All patients, n (%)	521,593 (100)	19,124 (3.7)	63,512 (12.1)	23,618 (4.5)	11,010 (2.1)	17,158 (3.3)	59,133 (11.3)	5,399 (1.0)	2,210 (0.4)
DSG indicated conditions, %
OA (n = 169,460)a	32.5	5.5	20.6	5.3	3.4	5.7	16.0	1.2	0.7
OA of hand/wrist/elbow (n = 25,193)a	4.8	6.2	20.1	5.5	3.9	5.2	17.2	1.0	0.5
OA of knee/ankle/foot (n = 89,440)a	17.1	5.6	20.4	5.5	3.4	5.9	16.4	1.2	0.6
Other arthritis conditions, %
RA (n = 21,084)a	4.0	8.0	22.0	6.2	4.3	6.8	19.1	1.8	0.8
AS (n = 14,384)a	2.8	6.6	18.3	5.3	3.7	5.4	17.3	3.2	0.5
No indicated condition at baseline, % (n = 337,357)a	64.7	2.6	7.9	4.1	1.4	2.0	8.8	0.9	0.3
Baseline risk factors, %b
≥ 1 risk factor (n = 484,458)a	92.9	3.9	13.0	4.5	2.2	3.5	11.9	1.1	0.4
Hepatic (n = 257,204)a	49.3	5.3	17.3	4.9	2.8	5.1	13.9	1.2	0.6
CV (n = 330,599)a	63.4	4.8	17.5	4.1	2.7	4.8	13.6	1.1	0.6
HTN (n = 272,385)a	52.2	5.0	18.1	4.5	2.8	5.2	13.7	1.0	0.6
GI (n = 409,774)a	78.6	4.1	13.1	4.6	2.4	3.6	12.2	1.2	0.4
Renal (n = 145,311)a	27.9	6.6	24.0	5.2	3.2	7.7	14.1	1.2	0.9
Skin (n = 145,028)a	27.8	5.0	15.0	4.2	2.9	4.1	16.5	1.3	0.5
Misuse/abuse (n = 114,697)a	22.0	4.3	12.9	3.5	2.6	3.5	11.9	2.8	0.5
Baseline comorbid conditions of interest, %
History of obesity (n = 115,692)a	22.2	3.8	11.4	3.9	2.1	3.5	11.5	1.2	0.3
Diabetes (n = 106,608)a	20.4	5.6	21.4	5.4	3.0	7.5	13.5	1.0	0.8
Smoking (n = 60,152)a	11.5	4.2	15.3	3.1	2.6	4.2	10.9	3.1	0.6
Alcohol use (n = 4956)a	1.0	6.2	11.4	2.4	2.6	3.4	8.7	5.0	0.7
SLE (n = 4223)*	0.8	9.0	20.4	5.9	4.4	8.7	21.5	2.4	0.5

AS ankylosing spondylitis, CV cardiovascular, DSG1% diclofenac sodium topical gel 1%, EOI event of interest, GI gastrointestinal, HTN hypertension, ICD-9-CM International Statistical Classification of Diseases, Ninth Edition, Clinical Manifestation, ICD-10-CM International Statistical Classification of Diseases, Tenth Edition, Clinical Manifestation, OA osteoarthritis, RA rheumatoid arthritis, PCS Procedure Codes, SLE systemic lupus erythematosus

^aTotal number of people within each subgroup (DSG-indicated conditions, baseline risk factors, baseline comorbid conditions of interest)

^bRisk factors were identified through ICD-9-CM/ICD-10-CM diagnosis, ICD-9-CM/ICD-10-PCS, Current Procedural Terminology codes, and medications (as proxy), and are correlated to events of specific interest. Among patients with an EOI, the proportion having risk factors for any of these conditions ranged between 92 and 99%

Among the patients, 57.8% received only one DSG1% prescription, 17.4% received two prescriptions, and 24.7% received three or more prescriptions. On average, patients had 2.5 (SD 3.6) prescription fills per patient (Table 3). The mean days of DSG1% exposure were 94 days (Table 3).

Table 3.

Prescriptions of DSG1% dispensed and exposure time

Treatment patterns	All	Hepatic EOI	CV EOI	HTN EOI	GI EOI	Renal EOI	Skin reaction EOI	Misuse/abuse EOI	Death
All patients, n (%)	521,593 (100)	19,124 (3.7)	63,512 (12.1)	23,618 (4.5)	11,010 (2.1)	17,158 (3.3)	59,133 (11.3)	5399 (1.0)	2210 (0.4)
Number of DSG1% prescriptions dispensed, mean (SD)a	2.5 (3.6)	5.6 (6.2)	5.1 (6.1)	5.3 (6.2)	6.1 (6.7)	6.1 (6.8)	5.2 (5.9)	5.3 (6.6)	6.2 (7.9)
Total person-years of DSG exposureb	134,310.7	182,430.2	163,764.4	178,667.2	186,835.9	184,095.0	165,716.6	188,863.2	191,152.5
Number of therapy periods, mean (SD)c	1.9 (1.8)	3.7 (2.9)	3.3 (2.7)	3.5 (2.8)	3.8 (3.0)	3.7 (2.9)	3.5 (2.8)	3.4 (2.8)	2.8 (2.6)

CV cardiovascular, DSG 1% diclofenac sodium topical gel 1%, EOI event of interest, GI gastrointestinal, HTN hypertension

^aOf patients with an EOI, the average number of DSG1% prescriptions filled ranged from 5.1 to 6.2 per EOI cohort. Patients filling 1, 2, or ≥ 3 prescriptions were 301,441, 91,065, and 129,087s, respectively

^bTotal person-years of DSG exposure is defined as the sum of days in all DSG therapy episodes that occurred during patient follow-up periods/365. From this formula, mean days of DSG exposure per person is 94 days

^c63.1%, 18.4%, and 18.5% of the overall population had 1, 2, or ≥ 3 therapy periods, respectively. Median number of therapy periods was 1 with a maximum number of 31 therapy periods

Subgroup Analysis: EOIs Rate Prevalence

During the study period, 74.2% of the population reported no EOIs after initiation of treatment with DSG1%. Among the remaining 25.8%, mean (SD) time from index date to first event was 244.0 (368.6) days (Supplementary table 2). The mean (SD) follow-up time for patients with no EOI was 187.8 (338.7) days. In contrast, mean (SD) follow-up times were longer in each of the EOI cohorts, ranging from 551.3 (649.2) days for patients who died followed by 878.5 (814.7) in CV EOI, 984.5 (835.7) in HTN EOI, 1,015.8 (865.2) in misuse/abuse EOI, 1024.0 (825.1) in skin reaction EOI, 1032.2 (839.2) in renal EOI, 1119.0 (845.0) in GI EOI, and 1153.5 (861.2) days for patients with hepatic events (Supplementary Table 1). Among all cohorts, CV was the most common (reported by 12.1% of patients) EOI, followed by skin reaction (11.3%), HTN (4.5%), hepatic (3.7%), GI (2.1%), and misuse/abuse (1%).

Across the entire population, the frequency of any reported EOI increased with advancing age. In the oldest subgroup (i.e., those aged ≥ 85 years), 55.7% experienced no EOIs, whereas 95.1% of those in the youngest age group (i.e., aged < 18 years) experienced no EOI (Fig 3). The CV, renal, and death EOI cohorts had average ages of ≥ 70 years (Table 1). There were no major differences in the frequencies of an EOI when comparing males versus females or by race (Table 1). The EOI cohort associated with misuse or abuse was the youngest age group, with a mean (SD) age of 50.6 (16.6) years.

Fig. 3.

Incidence of “No EOI” during the study period by age group. EOI event of interest

Among patients with OA, RA, and/or AS, a greater proportion fell into the EOI subgroup compared to those without a baseline arthritis-indicated condition (Table 2). Despite a high level of baseline risk (92.9%), 74.2% of patients did not experience an EOI during the follow-up period (Tables 1, 2). Among the baseline risk factors, GI risk factors were most prevalent (78.6%), followed by CV (63.4%), HTN (52.2%), and hepatic risk factors (49.3%) (Table 2). Conversely, EOI rates for GI (2.1%), hepatic (3.7%), and HTN (4.5%) events were lower, while CV (12.1%) EOI rates were higher (Table 2).

Baseline comorbid conditions were also reported (Table 2), with obesity (22.2%), diabetes (20.4%), and smoking (11.5%) being the most prevalent. Notably, patients with systemic lupus erythematosus (SLE) tend to experience higher rates of EOIs compared to other comorbidities excluding misuse-related EOIs or death. However, the association between the diabetes–CV EOI is an exception to this trend. Specifically, while overall population has 12.1% incidence of CV EOI, 21.4% of patients with diabetes experienced a CV EOI.

As expected, individuals with an EOI exhibited greater utilization of healthcare resources (Supplementary Table 3). The mean (SD) number of outpatient visits for the entire population was 12.3 (25.6), with average outpatient visits ranging from 37.0 (46.6) to 55.4 (61.6) among the EOI cohorts (excluding deaths). Similarly, the average (SD) number of emergency department visits was 0.9 (3.5) for the entire population, varying from 3.2 (7.2) to 7.4 (15.7) among the EOI cohorts. This trend may be attributed to several factors: patients experiencing an EOI tend to be older, have longer follow-up times, and exhibit a higher prevalence of comorbidities and baseline risk factors compared to the overall population.

Concomitant Prescription Fills in EOI Subgroups

In the entire study population, 43.0% of patients received at least one prescription for an oral NSAID (Table 4). Additionally, nearly 3% of patients had concomitant prescription fills for topical NSAIDs other than DSG1%. The majority of patients (70.8%) in the misuse/abuse EOI cohort were concurrently taking oral NSAIDs followed by 65.4% in the hepatic EOI cohort, 65.3% in the skin reaction EOI cohort, 64.4% in the GI and HTN EOI cohort, 55.4% in the renal EOI cohort, 51.9% in the CV EOI cohort, and 33.5% in the death EOI cohort. It’s important to note that this study focused on NSAID prescription therapies and did not include OTC products. Therefore, the total number of people using any NSAID may be higher than reflected in this study. Besides NSAIDs, other concomitant medications dispensed during the follow-up period are presented in Table 5.

Table 4.

Concomitant prescription NSAIDs dispensed to patients with DSG1% prescriptions during the follow-up period

	All	Hepatic EOI	CV EOI	HTN EOI	GI EOI	Renal EOI	Skin reaction EOI	Misuse/abuse EOI	Death EOI
All patients, n (%)	521,593 (100)	19,124 (3.7)	63,512 (12.1)	23,618 (4.5)	11,010 (2.1)	17,158 (3.3)	59,133 (11.3)	5399 (1.0)	2210 (0.4)
≥ 1 concomitant topical NSAID, n (%)	13,426 (2.6)	1856 (9.7)	5178 (8.2)	1867 (7.9)	1114 (10.1)	1,700 (9.9)	4,975 (8.4)	443 (8.2)	122 (5.5)
NSAIDs (nontopical)a, n (%)	224,104 (43.0)	12,506 (65.4)	32,969 (51.9)	15,208 (64.4)	7094 (64.4)	9,511 (55.4)	38,643 (65.3)	3,822 (70.8)	740 (33.5)

CV cardiovascular, EOI event of interest, GI gastrointestinal, HTN hypertension, NSAID nonsteroidal anti-inflammatory drug

^aNontopical NSAIDs with therapeutic class code 280,804

Table 5.

Concomitant medications by therapeutic class dispensed during the follow-up period^a

Characteristics	All	Hepatic EOI	CV EOI	HTN EOI	GI EOI	Renal EOI	Skin reaction EOI	Misuse/abuse EOI	Death
All patients, n (%)	521,593 (100)	19,124 (3.7)	63,512 (12.1)	23,618 (4.5)	11,010 (2.1)	17,158 (3.3)	59,133 (11.3)	5,399 (1.0)	2,210 (0.4)
Therapeutic class, %
Opiate agonists	42.0	81.5	72.8	70.2	81.1	81.5	71.7	84.6	69.4
Proton-pump inhibitors	30.5	68.9	58.6	54.2	82.4	65.1	55.8	55.8	51.1
HMG-COA reductase inhibitors	30.0	51.9	59.4	53.3	52.9	63.0	48.0	38.8	42.4
Antidepressants	27.6	54.0	47.3	45.4	55.6	52.7	46.9	69.1	45.7
Centrally acting skeletal muscle relaxants	22.1	47.2	34.1	39.7	46.6	37.9	42.2	61.1	20.3
Adrenals	20.5	53.8	44.4	44.2	54.1	50.7	51.9	51.2	37.5
Beta-adrenergic blocking agents	20.1	41.5	56.0	44.9	42.7	57.6	33.7	33.0	45.9
Corticosteroids (EENT)	20.0	49.5	40.1	43.1	48.7	44.5	48.4	42.5	25.9

CV cardiovascular, EENT eyes, ears, nose, throat, EOI event of interest, GI gastrointestinal, HMG-CoA 3-hydroxy-3-methylglutaryl coenzyme A, HTN hypertension

^aMedications shown are those dispensed to ≥ 20% of the overall study population

Primary care providers were the most common prescribing provider type of DSG1% (Supplementary Table 4). In this cohort, 61.0% of all DSG1% prescriptions were prescribed by those in family practice, internal medicine, and physician assistants (26.7%, 21.3%, and 13.0%, respectively). An additional 13.3% of DSG1% prescriptions were prescribed by the following prescribing provider specialties: nurse practitioners, general practice, and pain medicine. This constituted 74.3% of the top 10 prescribing provider types (Supplementary Table 4). Interestingly, rheumatologists were not on the list despite the high presence of MSK disease and OA among the study population.

Discussion

This study represents a robust assessment of a population with MSK/connective tissue disorders with well-documented demographics, medication use, and medical histories. The study profiles more than 500,000 DSG1% users in the US DoD MHS diagnosed with DSG-indicated and non-indicated MSK disorders. In this study of DSG1% users identified across an 8.5-year time period averaged 2.5 prescription fills and nearly 1 year of continuous follow-up, only 25.8% of patients experienced an EOI. More notably, EOIs occurred on average 244.0 days after the last DSG1% prescription fill. Interestingly, 92.9% of the population had elevated baseline risk factors for ≥ 1 EOI. These findings suggest no increase in risk of EOIs from baseline (i.e., underlying conditions at the time of DSG1%) associated with DSG1% therapy. Such EOIs could also be inherited from the recruited participant population, which is a crucial factor with respect to real-world scenarios. These findings are further supported by the age-associated subgroup analysis. The age distribution of DSG users with EOIs skewed more toward older individuals. Notably, a higher proportion of the older age group (> 85 years) experience more EOIs compared to the younger age group. The observed trend aligns with prior research on the frequency of these events, stratified by age group. The data highlight specific risk categories and indicate that the incidence of EOIs tends to rise with age due to various comorbidities [29–31].

Furthermore, a higher incidence of CV EOIs was observed in the DSG users with conditions such as RA compared to other DSG-indicated and non-indicated MSK conditions. These data are consistent with a meta-analysis of 24 studies reporting a 50% increased risk of CV death in patients with RA. The study highlighted that the traditional CV risk factors and persistent chronic inflammation, an intrinsic component of RA lead to accelerated atherosclerosis—a major cause of CV disease and increased CV mortality [32]. This suggests that the elevated risk might be primarily due to underlying health conditions rather than DSG usage itself.

In our study, we found that GI baseline risk factors were the most prevalent (78.6%) followed by CV, hepatic, and HTN. Conversely, the rates of EOI for GI (2.1%), hepatic (3.7%), and HTN (4.5%) were lower. However, CV EOI rates were higher (12.1%). A study by Peniston et al. also supports this finding, demonstrating that the percentage of patients experiencing adverse events was similar between those with and without HTN, type 2 diabetes mellitus, or cerebrovascular/cardiovascular disease [33]. These data suggest that baseline risk factors do not significantly affect the use of topical diclofenac in patients with high baseline risk, as it does not substantially increase the likelihood of adverse events in the population. The slightly higher prevalence of CV EOI may be associated with other factors, such as age, especially considering that the majority of the EOIs group comprises older individuals.

In addition to assessing arthritic conditions and risk factors, the study also investigated comorbidities and their association with EOI. Notably, a larger proportion of patients with SLE were linked to CV EOI, except for the diabetes–CV EOI connection. This finding aligns with existing literature that highlights elevated CV risk in both SLE and diabetic patients. Inflammation in these individuals is increasingly recognized as a central factor in atherosclerosis pathogenesis and a significant risk factor for vascular disease [34, 35].

DSG offers a safer profile when used alongside other medications compared to oral NSAIDs. This safety advantage might be attributed to DSG’s lower systemic absorption [16]. Our study corroborates this finding, as we observed a notably higher presence of oral NSAIDs among patients with various EOIs, particularly those with misuse/abuse, hepatic issues, and skin reactions. By incorporating DSG into the treatment regimen, reliance on oral NSAIDs can be reduced, particularly in individuals with CV and GI comorbidities who often use multiple medications. DSG also mitigates the risk of drug–drug interactions, thereby enhancing safety. This approach aligns with studies recommending the minimization of oral NSAIDs doses to mitigate drug interactions and adverse effects in patients concurrently taking other medications [36, 37].

The study’s key strength is its use of detailed data from the US DoD MHS database, which covers an integrated health system. The cohort consisted of over 500,000 DSG1% users who were profiled in a real-world setting, providing patient characteristics and prescription fill patterns. This large cohort of patients was followed for nearly 1 year. While there are limited studies assessing the use of topical NSAIDs with long-term follow-up, even fewer studies have profiled topical NSAID users in a real-world setting [16, 18]. Approximately one-third of U.S. military veterans live with arthritis, with OA rates being twice as high as in the general population for those aged 40 and older [38]. The MHS database, which includes information on active and retired military personnel and their dependents, benefits from profiling users of DSG at a 1% prevalence rate. Although the active-duty population may be healthier, the demographics of DSG1% users align with those of arthritis patients in the general U.S. population.

The retrospective data analysis faced limitations due to the available data in the pharmacy dispensing records. The data included retail and mail-order dispensing of prescription drugs but did not account for OTC products. Consequently, the actual number of individuals using any NSAID may be greater than this study reflects. It is important to note that prescription dispensing alone does not always confirm adherence to labeled drug use.

Conclusion

This long-term RWE study effectively bridges the gap between controlled clinical trials and the diverse landscape of real-world medical settings, providing valuable insight into post-market authorization safety and effectiveness. Notably, there was no observed increase in EOIs from baseline during therapy with DSG1%. Moreover, long intervals before the occurrence of EOIs can provide more comprehensive safety data, contributing to a better understanding of the long-term effects of DSG1% therapy. These findings indicate that DSG is a viable treatment option with a favorable safety profile for managing OA pain, frequently used alongside concomitant medications and in the presence of comorbidities and elevated risk factors.

These results further support the long-term tolerability of topical DSG1% in patients with MSK disorders and supplement the clinical trial findings on the safety of DSG. Utilizing a robust database like the DOD in real-world settings enhances data quality, increases generalizability, and facilitates longitudinal analysis, thereby contributing to informed regulatory decisions such as OTC switch.

DSG’s OTC status offers further benefit by allowing direct access for patients without requiring a physician’s prescription. This accessibility benefits both elderly patients and working-age adults by enabling prompt treatment of OA pain, improving quality of life, preventing bad posture, and ultimately enhancing HRQoL.

Supplementary Information

Below is the link to the electronic supplementary material.

Author Contributions

Nicholas M. Sicignano, Frédérique Bariguian Revel, Richard Petruschke, Francis P. Barbone, Karin Nicholson, and Jess D. Edison were involved in conceptualization, design, data collection, analysis, manuscript drafting, critical revision, and final approval.

Funding

This study and the Rapid Service fee was funded by Haleon (formerly GSK Consumer Healthcare). The study sponsor participated in the study design, data collection, data review, data analysis and writing of the report.

Data Availability

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Conflict of interest

Frédérique Bariguian Revel, Richard Petruschke, Karin Nicholson and Francis Barbone are employees of Haleon, formerly GSK Consumer Healthcare. Nicholas M. Sicignano has no conflict of interests to declare. The sponsor, Haleon (Formerly GSK Consumer Healthcare), Warren, NJ, USA, participated in the study design, analysis, and interpretation. The sponsor also provided funding for medical writing support and participated in the decision to submit the manuscript for publication. The opinions and assertions expressed herein are those of the author(s) and do not reflect the official policy or position of the Uniformed Services University of the Health Sciences or the Department of Defense. Neither Dr. Jess Edison nor any of his family members have a financial interest in any commercial product, service, or organization providing financial support for this research. References to non-Federal entities or products do not constitute or imply a Department of Defense or Uniformed Services University of the Health Sciences endorsement. This research protocol was reviewed and approved by the Naval Medical Center Portsmouth (IRB) in accordance with all applicable Federal regulations governing the protection of animals in research. This work was prepared by a military or civilian employee of the US Government as part of the individual’s official duties and therefore is in the public domain and does not possess copyright protection (public domain information may be freely distributed and copied; however, as a courtesy it is requested that the Uniformed Services University and the author be given an appropriate acknowledgement). Research data were derived with oversight from Naval Medical Center, Portsmouth, VA, USA, IRB, NHRS identifier; number NMCP.2017.0096.

Ethical Approval

Approval from an Institutional Review Board (IRB) was obtained to access the DoD database. The study was conducted in accordance with the ethical principles laid down in the Declaration of Helsinki. All participants provided informed consent before study initiation.