Progression to insulin therapy among patients with type 2 diabetes treated with sitagliptin or sulphonylurea plus metformin dual therapy.

AIM: To assess time to insulin initiation among patients with type 2 diabetes mellitus (T2DM) treated with sitagliptin versus sulphonylurea as add-on to metformin. METHODS: This retrospective cohort study used GE Centricity electronic medical records and included patients aged ≥18 years with continuous medical records and an initial prescription of sitagliptin or sulphonylurea (index date) with metformin for ≥90 days during 2006-2013. Sitagliptin and sulphonylurea users were matched 1 : 1 using propensity score matching, and differences in insulin initiation were assessed using Kaplan-Meier curves and Cox regression. We used conditional logistic regression to examine the likelihood of insulin use 1-6 years after the index date for each year. RESULTS: Propensity score matching produced 3864 matched pairs. Kaplan-Meier analysis showed that sitagliptin users had a lower risk of insulin initiation compared with sulphonylurea users (p = 0.003), with 26.6% of sitagliptin users initiating insulin versus 34.1% of sulphonylurea users over 6 years. This finding remained significant after adjusting for baseline characteristics (hazard ratio 0.76, 95% confidence interval 0.65-0.90). Conditional logistic regression analyses confirmed that sitagliptin users were less likely to initiate insulin compared with sulphonylurea users [odds ratios for years 1-6: 0.77, 0.79, 0.81, 0.57, 0.29 and 0.75, respectively (p < 0.05 for years 4 and 5)]. CONCLUSIONS: In this real-world matched cohort study, patients with T2DM treated with sitagliptin had a significantly lower risk of insulin initiation compared with patients treated with sulphonylurea, both as add-on to metformin.

Impact on glycated haemoglobin of a biological response-based measure of medication adherence.

AIMS: The aim of this study was to examine the relationship between a specific glycated haemoglobin (HbA1c) measurement and a pharmaceutical dispensings-based measure of adherence calculated over the 90 days before each HbA1c measure among patients who have newly initiated metformin therapy. METHODS: We identified 3109 people with type 2 diabetes who initiated metformin as their first-ever antihyperglycaemic drug, analysing all 9918 HbA1c measurements that were taken over the next 2 years. We used an adaptation of the 'proportion of days covered' method for assessing medication adherence that corresponded to an ∼90-day interval preceding an HbA1c measurement, terming the adaptation the 'biological response-based proportion of days covered' (BRB-PDC). To account for multiple observations per patient, we analysed the association between HbA1c and BRB-PDC within the generalized estimating equation framework. Analyses were stratified by HbA1c level before metformin initiation using a threshold of 8% (64 mmol/mol). RESULTS: After multivariable adjustment using 0% adherence as the reference category, BRB-PDC in the range 50-79% was associated with HbA1c values lower by -0.113 [95% confidence interval (CI) -0.202, -0.025] among patients with pre-metformin HbA1c <8%, and by -0.247 (95% CI -0.390, -0.104) among those with HbA1c ≥8% at metformin initiation. Full adherence (≥80%) was associated with HbA1c values lower by -0.175% (95% CI -0.257, -0.093) and by -0.453% (95% CI -0.586, -0.320). CONCLUSIONS: Using this novel short-interval approach that more closely associates adherence with the expected biological response, the association between better adherence and HbA1c levels was considerably stronger than has been previously reported; however, the strength of the impact was dependent upon the HbA1c level before initiating metformin.

Post-fracture pharmacotherapy for women with osteoporotic fracture: analysis of a managed care population in the USA.

UNLABELLED: Pharmacologic therapy is recommended to reduce future fracture risk. We examined osteoporosis medications dispensed to older women after first fracture. Only 23 % received therapy during the first year post-fracture. Prior osteoporosis therapy, a prior osteoporosis diagnosis, and older age were good predictors of post-fracture osteoporosis therapy. INTRODUCTION: Pharmacologic therapy is recommended after osteoporotic fracture to reduce future fracture risk. The objective of this retrospective study was to examine osteoporosis therapy dispensed to women post-fracture. METHODS: We identified women ≥50 years old in a large administrative claims database from 2003 to mid-2012 who were continuously enrolled 2 years before (baseline) and 1 year after first osteoporotic fracture. Exclusions were Paget's disease or malignant neoplasm. Pre- and post-fracture osteoporosis therapies (oral and parenteral) were assessed overall and by fracture site. RESULTS: A total of 47,171 women of mean (SD) age of 63 (10) years were eligible; fractures included 8 % hip, 17 % vertebral, 73 % non-hip/non-vertebral, and 3 % multiple fracture sites. Only 18 % received osteoporosis therapy within 90 days and 23 % within 1 year post-fracture. Overall, 19 % of women had a prior osteoporosis diagnosis; 20 % had received osteoporosis therapy during baseline. Of 37,649 (80 %) women without baseline therapy, only 9 % initiated pharmacologic therapy within 1 year. The adjusted odds ratio (OR) of therapy within 1 year post-fracture was significantly greater for women who had received baseline osteoporosis therapy (versus none) and who had vertebral (OR 12.7, 95 % confidence interval (CI) 11.2-14.5), hip (15.2, 12.5-18.7), or non-hip/non-vertebral fracture (34.4, 31.7-37.3). Other significant predictors included pre-fracture osteoporosis diagnosis (1.6, 1.4-1.7) and older age (OR range, 1.3-1.7). Treatment adherence was significantly better among women with baseline osteoporosis diagnosis. CONCLUSIONS: The substantial post-fracture treatment gap represents an important unmet need for women with osteoporotic fractures. Fracture liaison or adherence programs could lead to improved post-fracture treatment rates.

A discrete-choice experiment of United Kingdom patients' willingness to risk adverse events for improved function and pain control in osteoarthritis.

OBJECTIVE: To assess patient preferences for treatment-related benefits and risks associated with the use of nonsteroidal anti-inflammatory drugs (NSAIDs) in the management of osteoarthritis (OA). DESIGN: Using a chronic-illness panel in the United Kingdom, patients 45 years or older with a self-reported diagnosis of OA were eligible to participate in the study. Patient preferences were assessed using a discrete-choice experiment that compared hypothetical treatment profiles of benefits and risks consistent with NSAID use. Benefit outcomes (ambulatory pain, resting pain, stiffness, and difficulty doing daily activities) were presented on a 0-to-100 mm scale. Risk outcomes (bleeding ulcer, stroke, and myocardial infarction [MI]) were expressed as probabilities over a fixed time period. Each patient answered 10 choice tasks comparing different treatment profiles. Preference weights were estimated using a random-parameters logit model. RESULTS: Final sample included 294 patients. Patients ranked reductions in ambulatory pain and difficulty doing daily activities (both: 6.32; 95% confidence interval [CI]: 5.0-7.6) as the most important benefit outcomes, followed by reductions in resting pain (2.80; 95% CI: 1.8-3.8) and stiffness (2.65; 95% CI: 0.9-4.4). Incremental changes (3%) in the risk of MI or stroke were assessed as the most important risk outcomes (10.00; 95% CI: 8.2-11.8; and 8.90; 95% CI: 7.3-10.5, respectively). CONCLUSION: Patients ranked ambulatory pain as a more important benefit than resting pain; likely due to its impact on ability to do daily activities. For a 25-mm reduction, patients were willing to accept four times the risk of MI in ambulatory pain vs resting pain.

Impact of concurrent macrovascular co-morbidities on healthcare utilization in patients with type 2 diabetes in Europe: a matched study.

AIM: To examine and to quantify the impact of concurrent macrovascular co-morbidities (MVC) on healthcare resource utilization among patients with type 2 diabetes mellitus (T2DM) in Europe. METHODS: This is a matched cohort study based on the Real-Life Effectiveness and Care Patterns of Diabetes Management study, a multicentre, observational study with retrospective medical chart reviews of T2DM patients in Spain, France, UK, Norway, Finland, Germany and Poland. Included patients were aged > or =30 years at time of diagnosis of T2DM who added a sulfonylurea or a PPARgamma agonist to failing metformin monotherapy (index date) and had concurrent MVC (cases). A control cohort with T2DM but without concurrent MVC was identified using 1:1 propensity score matching. Logit models were used to identify the relationship between concurrent MVC and the likelihood of emergency room admission, receiving medical/surgical procedures, and hospitalization during the study period after controlling for baseline demographics, clinical information and baseline treatment. Negative binomial models were used to predict the number of office visits and length of hospital stay per year attributable to the concurrent MVC. RESULTS: Relative to controls, patients with MVC were significantly more likely to have emergency department admissions [odds ratio (OR) 2.69; 95% CI: 1.56-4.65], receiving medical/surgical procedures (OR 2.57; 95% CI: 1.56-4.21) and hospitalizations (OR 2.58; 95% CI: 1.64-4.07) after controlling for other predictors. Similarly, MVC were associated with 1.49 additional office visits per year (p = 0.036) and 0.32 days of hospital stay per year (p = 0.023). CONCLUSIONS: Within a seven-country European sample, this study showed that T2DM patients with MVC were more likely to use healthcare resources compared with T2DM patients without MVC.

Patient-reported outcomes in a survey of patients treated with oral antihyperglycaemic medications: associations with hypoglycaemia and weight gain.

AIM: To examine the association between medication side-effects (SEs) and patient-reported outcomes (PROs) among patients with type 2 diabetes treated with oral antihyperglycaemic agents (OAHAs). METHODS: A total of 1984 participants responded to an internet-based survey in the United States. Data were collected on hypoglycaemia 6 months and weight gain 12 months prior to the survey. Health-related quality of life (HRQoL) was measured using the EuroQol-5D (EQ-5D). Also administered were the Treatment Satisfaction Questionnaire for Medication v.1.4 (TSQM) and the Hypoglycaemia Fear Survey II (HFS). RESULTS: Symptoms of hypoglycaemia were reported by 62.9% of participants, and 36.9% reported weight gain. For those reporting hypoglycaemia, mean scores were lower for TSQM and EQ-5D and higher for HFS when compared with those with no symptoms (TSQM: 69.7 vs. 75.1; EQ-5D: 0.78 vs. 0.86; HFS: 17.5 vs. 6.2; all p < 0.0001). The same remained true when accounting for symptom severity, where severity was monotonically related with PRO scores (all p < 0.0001). Similarly, reported weight gain was associated with lower treatment satisfaction (69.0 vs. 73.3) and HRQoL (0.77 vs. 0.83), and increased fear of hypoglycaemia (15.7 vs. 11.8) (all p < 0.0001). In mixed linear regression analysis, the associations between medication SEs and PROs remained significant after adjusting for patient and disease characteristics. CONCLUSIONS: Among patients with type 2 diabetes treated with OAHAs, self-reported hypoglycaemia and weight gain were associated with decreased treatment satisfaction and HRQoL. In addition, the presence of these SEs was associated with increased fear of hypoglycaemia.

Glycaemic control among patients with type 2 diabetes mellitus in seven European countries: findings from the Real-Life Effectiveness and Care Patterns of Diabetes Management (RECAP-DM) study.

OBJECTIVE: This study was undertaken to assess glycaemic control as well as changes in glycaemic control over time in patients with type 2 diabetes mellitus (T2DM) who added a sulphonylurea (SU) or thiazolidinedione (TZD) to their metformin monotherapy in typical treatment settings within seven European countries. METHODS: An observational, cross-sectional multicentre study with retrospective medical chart review was conducted in Finland, France, Germany, Norway, Poland, Spain and UK. T2DM patients who added a SU or a TZD to metformin monotherapy between January 2001 and January 2006 (i.e. index date) and who had > or = 1 haemoglobin A1C (HbA1C) measurement within 12 months before the visit date, which occurred from June 2006 to February 2007, were included in the study. Demographic and clinical data were collected from medical records. The main study outcome measure was the proportion of patients with adequate glycaemic control (defined according to the International Diabetes Federation as HbA1C < 6.5%) using the most recent HbA1C measurement before the visit date. In addition, patients were grouped based upon the interval from the index date to the most recent HbA1C measurement to evaluate goal attainment and treatment changes over time. FINDINGS: In this European cohort of 2023 T2DM patients on metformin and either an SU or a TZD (mean age = 60.4 years), 25.5% of patients had adequate glycaemic control. The average HbA1C level was 7.2% after a mean of 2.6 years of treatment with combination oral antihyperglycaemic agent (AHA) therapy. Among the patients (n = 227) with most recent HbA1C measurement within 1 year after first adding an SU or a TZD, 27% had adequate glycaemic control (HbA1C < 6.5%), with a mean (s.d.) HbA1C of 7.1% (1.0); 1.3% of these patients were using some type of insulin therapy. Among the patients (n = 176) with most recent HbA1C measurement occurring > or = 5 years after adding an SU or a TZD, 20% had adequate glycaemic control, with a mean (s.d.) HbA1C of 7.4% (1.17), and 29.6% of these patients were using insulin. Overall, patients with (vs. without) adequate glycaemic control had significantly (p < 0.05) lower HbA1C levels (7.6 vs. 8.2%) at the time of adding an SU or a TZD to ongoing metformin monotherapy, were less likely to report a history of macrovascular complications (20 vs. 26%) and were more often engaged in physical activity three to five times a week (29 vs. 23%). CONCLUSIONS: Approximately one quarter of European out-patients with T2DM had adequate glycaemic control after a mean of 2.6 years following initiation of combination AHA therapy. Overall glycaemic control modestly declined over time, even though more patients were being treated with insulin. These findings highlight the progressive nature of the disease and need for more effective disease management/therapeutic alternatives.

Hypoglycaemia in patients with type 2 diabetes treated with a combination of metformin and sulphonylurea therapy in France.

CONTEXT: Hypoglycaemia from antihyperglycaemic drugs may have a significant impact on patients' health-related quality of life. Combination use of metformin and a sulphonylurea has become increasingly common; yet, the impact of hypoglycaemia on quality of life in these patients is not well documented. OBJECTIVE: To examine patient-reported experience of hypoglycaemia, worry about hypoglycaemic symptoms and the impact of hypoglycaemia on patients' quality of life associated with use of sulphonylurea co-administered with metformin. DESIGN: This was an observational, cross-sectional, multi-centre study. SETTING: A total of 98 primary care centres in France during October to December 2005. PATIENTS: A total of 400 patients with type 2 diabetes, who were > or = 35 years old and who had been treated with metformin and a sulphonylurea for at least 6 months, completed questionnaires during their usual primary care office visit. MAIN OUTCOME MEASURES: Frequency and severity of hypoglycaemic symptoms in the past 6 months, the Worry subscale of the Hypoglycaemic Fear Survey-II (HFS-II) and the EuroQol-5 Dimensions (EQ-5D) questionnaire. RESULTS: A total of 136 (34%) patients reported experiencing hypoglycaemia, of whom 78 (58%) experienced mild, 40 (30%) experienced moderate and 16 (12%) experienced severe or very severe symptoms. Mean score on the HFS-II Worry scale was higher for patients who reported having hypoglycaemia than for those who did not (19.0 vs. 10.2; p < 0.0001) and increased with severity of hypoglycaemic symptoms. In linear regression analyses, more severe symptoms of hypoglycaemia were significantly associated with higher scores on the HFS-II Worry scale (p = 0.0162) among patients with hypoglycaemic symptoms. Summary scores on the EQ-5D were lower for patients who reported hypoglycaemia than for those who did not (p = 0.0001) and, in multivariate analysis, the experience of hypoglycaemia was negatively associated with the EQ-5D summary score (p < 0.0001). CONCLUSION: The occurrence and severity of hypoglycaemic symptoms were associated with increased patient worry about hypoglycaemia and lower health-related quality of life among type 2 diabetic patients being treated with both metformin and a sulphonylurea.

Hypoglycaemic symptoms, treatment satisfaction, adherence and their associations with glycaemic goal in patients with type 2 diabetes mellitus: findings from the Real-Life Effectiveness and Care Patterns of Diabetes Management (RECAP-DM) Study.

AIMS: This study was undertaken to evaluate (i) factors associated with patient-reported hypoglycaemia; (ii) association of patient-reported hypoglycaemic symptoms with treatment satisfaction and barriers to adherence and (iii) association between treatment satisfaction, adherence and glycaemic control among patients with type 2 diabetes who added a sulphonylurea or a thiazolidinedione to ongoing metformin. METHODS: This observational, cross-sectional, multicentre study was conducted in seven countries (Finland, France, Germany, Norway, Poland, Spain and UK) from June 2006 to February 2007. Patients with type 2 diabetes who added a sulphonylurea or a thiazolidinedione to ongoing metformin therapy on a date (index date) from January 2001 through January 2006 and who had at least one haemoglobin A1C (HbA1C) measurement in the 12-month period before the visit date were eligible. Questionnaires were used to ascertain patients' reports of hypoglycaemic symptoms, treatment satisfaction, and treatment adherence. The Treatment Satisfaction Questionnaire for Medication was used to measure patients' treatment satisfaction. An adherence and barriers questionnaire was used to measure patients' adherence to treatment. Glycaemic control was based on documented HbA1C measurements within the prior 12 months. RESULTS: The mean +/- s.d. age was 62.9 +/- 10.6 years, and the mean +/- s.d. duration of diabetes was 7.8 +/- 5.1 years. HbA1C in this population of patients who had failed metformin monotherapy and were treated with oral antihyperglycaemic agents was below the International Diabetes Federation goal of 6.5% in only 477 (27.9%) patients. Approximately 38% of patients reported hypoglycaemic symptoms during the past year. Hypoglycaemia was significantly more likely in patients with a history of macrovascular complications of diabetes (OR = 1.346; 95% CI = 1.050-1.725) and with no regular physical activity (OR = 1.295; 95% CI = 1.037-1.618). Patients reporting hypoglycaemia had significantly lower treatment satisfaction scores (71.6 +/- 17.6 vs. 76.3 +/- 16.8; p < 0.0001 for global satisfaction). Compared with their counterparts reporting no hypoglycaemic symptoms, patients with such symptoms were also significantly more likely to report barriers to adherence, including being unsure about instructions (37.0 vs. 30.5%; p = 0.0057). Patients at HbA1C goal had significantly higher treatment satisfaction and adherence compared with those who were not. CONCLUSIONS: Patients' reports of hypoglycaemic symptoms are common in European outpatients with type 2 diabetes and are associated with significantly lower treatment satisfaction and with barriers to adherence. In addition, being at HbA1C goal is associated with treatment satisfaction and adherence.

Co-prescribing of proton pump inhibitors among chronic users of NSAIDs in the UK.

OBJECTIVES: Co-prescribing of proton pump inhibitors (PPIs) with non-selective NSAIDs (nsNSAIDs) is recommended in patients at risk of gastrointestinal (GI) events. This study estimated usage of PPI co-therapy among chronic nsNSAID users and determined factors associated with concurrent nsNSAID-PPI use. METHODS: The retrospective study was based on the Intercontinental Marketing Services (IMS) Health UK MediPlus database and included subjects > or = 40 yrs of age who received their first oral nsNSAID prescription between July and December 2002 and who had > or = 60 days of nsNSAID supply during the following year. Days with nsNSAID-PPI overlap were calculated and logistic regression was used to identify factors associated with nsNSAID-PPI overlap. A generalized linear model was used to assess the degree of association of GI risk factors with the nsNSAID-PPI overlap ratio among PPI users. RESULTS: Of 16,344 patients included, 1586 received at least one PPI prescription. Among PPI users, PPIs were available on approximately 50% of the days with nsNSAID therapy. After multivariate adjustment, age > or = 65 yrs, history of any hospitalization and co-prescriptions for anti-coagulants or oral corticosteroids increased the odds of any nsNSAID-PPI overlap by 21-68%. Prior gastroprotective agent (GPA) use increased the odds of any PPI use during follow-up 16-fold and nsNSAID-PPI overlap 19-fold. Among PPI users, patients with prior use of any GPA had a 2.46 times higher nsNSAID-PPI overlap ratio. CONCLUSIONS: PPI utilization correlates poorly with nsNSAID use in the UK. GI safety of nsNSAID-PPI co-therapy observed in controlled trials may therefore not be achieved in clinical practice.

Association between health-related quality of life and clinical efficacy endpoints in rheumatoid arthritis patients after four weeks treatment with anti-inflammatory agents.

BACKGROUND: Improvement of health-related quality of life (QoL) is increasingly recognized as a maj or treatment goal for patients with rheumatoid arthritis (RA). There are several measures of general health status and of physical functioning for assessing treatment effects on QoL in patients with RA, however, the relationship between QoL outcomes and conventional clinical efficacy endpoints is not completely understood. OBJECTIVE: To describe the association between changes in QoL and changes in other efficacy measures, among patients with RA after four weeks of treatment with etoricoxib, naproxen or placebo, and to explore differences in the association of changes in efficacy and changes in QoL parameters across treatment groups. METHODS: The study used data from 1684 patients with RA enrolled in two identical clinical trials (one US and one multinational). Patients were randomized to placebo, etoricoxib 90 mg once daily, or naproxen 500 mg twice daily in a 2 : 2: 1 allocation ratio. Primary efficacy endpoints were tender joint count, swollen joint count, patient global assessment of disease activity (100 mm VAS), and investigator global assessment of disease activity (0 - 4 Likert scale). QoL assessments were based on the Health Assessment Questionnaire (HAQ) and the Medical Outcomes Survey Short Form 36 (SF-36). Mean differences between baseline and week four were calculated for each parameter studied. Linear regression analysis was performed to assess the association between changes in clinical efficacy and changes in QoL parameters, adjusted for covariates. RESULTS: The degree of association between changes in tender or swollen joint counts and changes in QoL variables was low, explaining less than 10% of the variability for most QoL variables, except bodily pain (SF-36). In contrast, changes in patient global assessment of disease activity explained 33% of the variability in the overall HAQ score, and in the physical component score (SF-36; adjusted regression models). Values for investigator global assessment of disease activity were below those for patient global assessment but above joint count measures. Results were similar between the etoricoxib, naproxen and placebo groups in the degree of association between changes in efficacy and QoL variables. CONCLUSION: Currently used efficacy endpoints are less than ideal predictors of change in QoL. There is no evidence from this study that the association between changes in CE endpoints and QoL was different across treatments. Our results highlight the need to assess both conventional efficacy measures and QoL in clinical trials of RA treatments.

Initiation of antihypertensive therapy among new users of cyclooxygenase-2-selective and nonselective NSAIDs.

BACKGROUND: The comparative effects of cyclooxygenase-2- (COX-2) selective inhibitors and nonselective, nonsteroidal anti-inflammatory drugs (NSAIDs) on blood pressure are debated. Clinicians have been concerned about the need for antihypertensive treatment following therapy with these agents. OBJECTIVE: To compare initiation of antihypertensive treatment among new users of the COX-2-selective inhibitor rofecoxib and of nonselective NSAIDs in clinical practice. METHODS: Retrospective cohort study using the MediPlus (UK) database that covers 1.8 million patients throughout the UK. Patients included were at least 50 years of age, had at least 1 prescription for either diclofenac, ibuprofen, naproxen or rofecoxib (drugs of interest, DOIs), and had no prescription for any NSAID, COX-2 inhibitor, or antihypertensive treatment during the 6 months prior to their first/index prescription date. A subset of patients, classified as chronic and persistent new users, had at least 3 prescriptions of the index prescription DOI and did not switch to another DOI during the 6-month follow-up period. Logistic regression analysis, adjusted for potential predictors, was used to assess initiation of new antihypertensive treatment. RESULTS: 18,737 suitable patients were identified (diclofenac 7,861, ibuprofen 8,423, naproxen 1,556 and rofecoxib 897). Those using rofecoxib were older and more likely to be female than those using NSAIDs. During the 6 months following the index prescription, 7.0% of all new users and 11.5% of chronic and persistent new users initiated antihypertensive treatment. After adjusting for potential predictors there were no statistically significant differences in the risk of initiating antihypertensive treatment between new or chronic and persistent new users of rofecoxib, diclofenac, ibuprofen and naproxen (p > 0.05). CONCLUSION: The results of this study did not indicate any significant differences in the initiation of antihypertensive therapy among patients who were prescribed rofecoxib and NSAIDs, even after multiple prescriptions.

Use of nonsteroidal anti-inflammatory drugs and gastroprotective agents before the advent of cyclooxygenase-2-selective inhibitors: analysis of a large United States claims database.

BACKGROUND: Previous studies have shown that 20% to 40% of patients requiring nonsteroidal anti-inflammatory drugs (NSAIDs) are concomitantly prescribed gastroprotective agents (GPAs) such as proton pump inhibitors (PPIs) and H2-receptor antagonists. OBJECTIVE: The purpose of this study was to examine NSAID prescription patterns and the concurrent use of GPAs in a large national sample of patients who were prescribed NSAIDs for the first time. METHODS: Patterns of NSAID use, particularly chronic NSAID use, and of concomitant use of GPAs were examined using a large US-based prescription database. Patients with at least 1 NSAID prescription dispensed between May 1 and August 31, 1998, were identified. Persons with any NSAID prescription within 4 months prior to the first (index) prescription were excluded. The remaining patients were defined as new NSAID users and then classified as chronic users (> or = 30 days of supply of NSAIDs during the 120 days of follow-up after the first NSAID prescription) or acute users (<30 days of NSAID supply during the 120 days of follow-up). Concomitant GPA use was defined as receipt of any GPA prescription between the fill date of NSAID prescription and 125% of days of supply. NSAIDs included diclofenac/misoprostol (in a fixed combination), diclofenac, naproxen, nabumetone, ibuprofen, and "other" (comprising several less frequently prescribed agents). Patients were classified as users of a particular NSAID based on the first NSAID prescription they received. GPAs included PPIs, H2-receptor antagonists, and misoprostol. RESULTS: A total of 3,028,808 new NSAID users were identified. Chronic NSAID users (47.8% of the sample) were older than acute users. The percentage of new chronic users aged > or = 65 years for each of the NSAIDs was 41.2% for diclofenac/ misoprostol, 33.0% for nabumetone, 30.8% for diclofenac, 20.4% for naproxen, and 20.3% for ibuprofen. The percentage of women was higher among patients treated with diclofenac/misoprostol than among patients treated with all other NSAIDs (P < 0.001). During the 120 days of follow-up, the percentages of NSAID users with concomitant GPA use were 22.7% for diclofenac/misoprostol, 16.3% for diclofenac, 11.5% for naproxen, 18.0% for nabumetone, 12.3% for ibuprofen, and 14.8% for other NSAIDs. Based on days of supply, the rates of concomitant GPA use were 31.1%, 23.6%, 17.6%, 27.3%, 18.8%, and 22.5% for diclofenac/misoprostol, diclofenac, naproxen, nabumetone, ibuprofen, and other NSAID users, respectively. Among those who were taking GPAs before the NSAID index prescription date, -89% continued GPA therapy. CONCLUSIONS: Approximately 22% of the days of NSAID supply were covered by GPAs. Prior GPA use was the strongest predictor of subsequent concomitant GPA/ NSAID use. Differences in GPA use were observed among patients using different NSAIDs.