Skip Navigation



Rheumatology Advance Access published online on February 7, 2008
Rheumatology, doi:10.1093/rheumatology/kem375
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrowOA All Versions of this Article:
47/4/458    most recent
kem375v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Suh, D.-C.
Right arrow Articles by Mavros, P.
PubMed
Right arrow PubMed Citation
Right arrow Articles by Suh, D.-C.
Right arrow Articles by Mavros, P.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© 2008 The Author(s)
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/2.0/uk/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

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

D.-C. Suh1, E. H. Dipl rer pol2, H.-C. Shin1 and P. Mavros2

1School of Pharmacy, Rutgers University, Piscataway and 2Outcomes Research, Merck & Co., Inc., Whitehouse Station, NJ, USA.

Correspondence to: P. Mavros, Outcomes Research, Merck & Co., Inc., One Merck Drive, WS2E-76, Whitehouse Station, NJ 08889, USA. E-mail: panagiotis_mavros{at}merck.com.


   Abstract
Top
 Abstract
Introduction
 Methods
 Results
 Discussion
 Acknowledgements
 References
 
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 ≥40 yrs of age who received their first oral nsNSAID prescription between July and December 2002 and who had ≥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 ~50% of the days with nsNSAID therapy. After multivariate adjustment, age ≥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.

KEY WORDS: NSAIDs, PPIs, Retrospective observational study, Drug utilization


   Introduction
Top
 Abstract
 Introduction
Methods
 Results
 Discussion
 Acknowledgements
 References
 
NSAIDs are among the most frequently prescribed drugs world wide [1]. In March 2005, ~7 million patients in the UK received NSAID prescriptions [2]. Although these agents are frequently used for brief periods to relieve acute pain, they are also often prescribed to patients with chronic pain for long-term use. For example, the recent Pain in Europe survey found that 52% of chronic pain sufferers currently used prescription painkillers (most commonly NSAIDs, weak opioids or paracetamol) to treat chronic pain [3].

Non-selective NSAIDs (nsNSAIDs; henceforth, refers only to non-aspirin nsNSAIDs) inhibit both the COX-1 and COX-2 isoforms of the cyclo-oxygenase (COX) enzyme [4]. It is believed that the role of COX-1 in gastric protection accounts for the common side-effect of upper gastrointestinal (GI) symptoms among chronic nsNSAID users. The risk for serious upper GI events [perforations, ulcers and bleeds (PUB)] is four times higher in chronic users of nsNSAIDs than in non-users [5]. Chronic nsNSAID users are also at higher risk for less serious upper GI symptoms, such as dyspepsia and abdominal pain. A number of factors are known to increase the risk for GI adverse events among nsNSAID users, including a history of ulcers, advanced age, concurrent use of corticosteroids, anti-coagulants or aspirin and certain comorbidities [1].

Clinical practice guidelines in the UK [6], Europe [7] and North America [8, 9] recommend co-prescription of a gastroprotective agent (GPA) in high-GI-risk patients on chronic nsNSAID therapy. Acid suppressive proton pump inhibitors (PPIs) have been shown in randomized clinical trials to reduce symptoms of dyspepsia [10], risk for duodenal and gastric ulcers detected by endoscopy [11] and recurrence of symptomatic [12] and bleeding ulcers [13] among patients taking nsNSAIDs. Other GPAs are less frequently used due to lack of tolerability or efficacy: the prostaglandin analogue misoprostol effectively reduces ulcer complications but is associated with diarrhoea and dyspepsia, while anti-secretory therapy with histamine2-receptor antagonists (H2RAs) at standard doses reduces gastric but not duodenal ulcers [14]. In the UK, PPIs comprise ~60% of all anti-ulcer drug prescriptions [15].

Despite recommendations of clinical practice guidelines, inadequate co-prescribing of GPAs to high-risk nsNSAID users has been shown to persist in The Netherlands [16], the USA [17] and Canada [18, 19]. While prior studies demonstrated deficiencies in rates of GPA prescribing, only one published study has so far quantified adherence to recommended use of PPIs among nsNSAID users who received PPIs [20]. This Dutch study found that among patients with both nsNSAID and PPI prescriptions, PPI supply overlapped 82% of days of nsNSAID supply.

Because suboptimal PPI utilization is expected to impair prophylaxis of nsNSAID-related gastropathy [21], it is important to determine the correspondence or overlap between consumption of nsNSAIDs and PPIs. The objectives of the study were to estimate concurrent usage of PPIs among chronic nsNSAID users in the UK and to determine factors associated with concurrent nsNSAID–PPI use.


   Methods
Top
 Abstract
 Introduction
 Methods
Results
 Discussion
 Acknowledgements
 References
 
Data source
This retrospective cohort study used data from the Intercontinental Marketing Services (IMS) Health UK MediPlus® database from 1 July 2001 to 31 December 2003. The MediPlus database, which includes longitudinal data on prescriptions, medical services (including tests and diagnostic procedures, hospital and specialist referrals) and medical diagnoses recorded by general practitioners for ~1.8 million active patients throughout the UK, has been validated and found to be accurate and reliable [22, 23], and has previously been used for studies of nsNSAID and GPA utilization [24]. To preserve patient and physician confidentiality, MediPlus records are coded anonymously. Our study was approved by the Independent Scientific and Ethics Advisory Committee.

Study design
Candidates for inclusion in the study cohort were identified based on a 6-month index period from 1 July 2002 to 31 December 2002. The date on which the physician wrote the first prescription for an oral nsNSAID during the index period was defined as the index date. Non-selective NSAIDs were classified according to the Anatomical Therapeutic Chemical (ATC) code M01A1, excluding meloxicam [which is considered a COX-2 inhibitor (coxib) in the UK] and multi-substance products such as benorilate and Arthrotec®. Patients were followed from the index date until the date of switching from an nsNSAID to a coxib or the end of 12 months of follow-up, whichever came first. Patients continued to be included in the study if they switched from one nsNSAID to another. A 12-month history period prior to the index date was reviewed to determine patients’ risk factors for GI events.

Patient inclusion criteria
Patients were included in the study cohort if they were prescribed any oral nsNSAID during the index period and had at least 60 days of nsNSAID supply during the follow-up period, thereby focusing the analysis on chronic nsNSAID users. Days of nsNSAID supply was determined based on the intended therapy duration of the prescribed medication(s), as recorded in MediPlus. All included patients were at least 40-yrs old at the index date, and had continuous active enrollment in the MediPlus data system throughout the study period.

Patients who received more than one type of nsNSAID at the index date were excluded to minimize confounding effects from multiple nsNSAIDs. Patients with no recorded information about the intended duration of their prescription were also excluded. Further, patients were excluded if they stopped taking nsNSAIDs and the date on which they stopped taking nsNSAIDs was unknown, since their days of medication supply could not be calculated. However, patients who switched from an nsNSAID to a coxib were included because the database allowed the nsNSAID duration to be calculated based on the date on which the coxib prescription was written.

The study population was divided into three cohorts based on PPI use during the history and follow-up periods:

  1. users with no PPI prescriptions during the follow-up period;
  2. continuing PPI users (PPIs were prescribed during both the history period and the follow-up period); and
  3. new PPI users (PPIs were not prescribed during the history period, but were added during the follow-up period).

GI risk factors
The following factors, which have been shown to be associated with an increased risk for nsNSAID-related gastropathy, were identified from the literature: advanced age, all-cause hospitalization (indicative of poor health status), history of GI events including episodes of PUB, comorbidities including history of cardiovascular (CV) disease and concurrent use of corticosteroids, aspirin or anticoagulants [1, 25–27]. Information on these risk factors was retrieved from the MediPlus database for the year prior to the index date. History of GI tests and prior use of GPAs served as indicators of prior GI events or elevated risk for these events. Although it may appear counterintuitive to include prior GPA use as a factor associated with higher GI risk, users of nsNSAIDs plus PPIs have been found to have a higher unadjusted rate of ulcers than users of nsNSAIDs alone due to channelling bias [28].

GI risk level was scored for each patient as the number of the following factors that were present during the 1-yr history period or at the index date (i.e. concurrent events):

  • Age 65 yrs or older.
  • History of all-cause hospitalization.
  • History of PUB.
  • History of diagnosis of dyspepsia/abdominal pain.
  • History of diagnosis of CV disease (acute myocardial infarction, ischaemic heart disease, congestive heart failure or cerebrovascular disease).
  • History of GI test (GI endoscopies, barium contrast tests, Helicobacter pylori tests and/or other GI monitoring).
  • History of GPA use (PPIs, H2RAs or misoprostol).
  • Concurrent use of anti-coagulants.
  • Concurrent use of oral corticosteroids.
  • Concurrent use of aspirin.

Data analysis
nsNSAID–PPI overlap
The total number of nsNSAID treatment days during the study period was calculated for each patient by totalling days of supply for all nsNSAID prescriptions, starting from the index date and ending on either the last day of nsNSAID supply or the date of a medication switch from an nsNSAID to a coxib, whichever came first. If total duration of medication supply during the 1-yr follow-up period exceeded 365 days, treatment days were truncated at 365.

The overlap in supply of PPIs and nsNSAIDs was calculated as a ratio for each patient by summing days of nsNSAID supply during which the patient had PPIs (inferred from the prescribing date and days of supply recorded for PPI prescriptions) and dividing this sum by the total number of nsNSAID treatment days as defined previously.

Because patients may not take medications as frequently as prescribed, and the MediPlus database does not provide direct measures of medication consumption, potential patient non-adherence was addressed in a sensitivity analysis by assuming that nsNSAIDs and PPIs were taken on 80% of prescribed days—a commonly cited threshold for acceptable adherence to chronic therapy [29]. Treatment days were recalculated as 1.25 (reciprocal of 80%) times the days of medication supply.

Factors associated with concurrent nsNSAID–PPI use
Logistic regressions were used to identify factors associated with PPI prescription, with and without adjustment for study covariates. Variables used in the regression models were gender and the GI risk factors listed in Table 1, including advanced age (≥ 65 yrs).


View this table:
[in this window]
[in a new window]

  		TABLE 1. Patient characteristics at the index date

 
Due to the fact that large proportions of nsNSAID users did not have any days with overlapping PPIs and nsNSAIDs, two-part models were used to identify factors associated with the degree of PPI overlap with nsNSAIDs [30, 31]. In the first part, logistic regressions were used to estimate the probability of any overlaps of PPIs with nsNSAID. In the second part, generalized linear models with log-link function and gamma distribution were used to determine factors associated with the overlapping ratio of PPI with NSAID among PPI users. The value of 10–10 was added to the overlapping ratios to address zero overlapping ratios when the models were tested with log-link function.

In the multivariate and generalized linear models, potential explanatory variables were assessed for colinearity using pairwise correlation analysis. Significance for all statistical tests was defined as P < 0.05. All statistical analyses were performed using SAS Version 8.02 (SAS Institute Inc., Cary, NC, USA).


   Results
Top
 Abstract
 Introduction
 Methods
 Results
Discussion
 Acknowledgements
 References
 
Patient baseline characteristics
Of 16 344 chronic NSAID users included in the study, 1586 (10%) received at least one PPI prescription during follow-up. Compared with patients not receiving PPI prescriptions during follow-up, the group receiving PPIs tended to be older, included more women, had more events and diagnoses indicative of higher GI risk during the prior year, and more frequently received a co-prescription at the index date for medications that increase the risk of nsNSAID-related gastropathy (Table 1). Patients who received PPIs during follow-up more frequently reported a history of dyspepsia/abdominal pain (20% vs 5%, respectively) and PPI use in the prior year (53% vs 1%) than patients not receiving PPIs. Compared with new users of PPIs during follow-up, the group of continuous PPI users, who had used PPIs during the prior year, tended to be younger, included more women, and had more prior events and diagnoses indicative of higher GI risk—notably, a much higher prevalence of prior dyspepsia/abdominal pain (30% vs 9%) and PUB (5% vs 1%).

The distribution of patients by number of GI risk factors listed above is shown in Fig. 1. Compared with patients who did not receive PPIs during follow-up, a higher proportion of those who received PPIs had multiple GI risk factors. At least one GI risk factor was present in 83% of PPI users and 53% of non-users, with three or more risk factors in 27% of PPI users vs 7% of non-users. Among PPI users, a higher proportion of those who had received PPIs in the prior year had multiple GI risk factors compared with those receiving PPIs for the first time during follow-up. This is partly because prior GPA use was scored as a GI risk factor, which is reflected by none of the continuing PPI users having zero GI risk factors.


Figure 1
View larger version (23K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		FIG. 1. Distribution of GI risk factors among nsNSAID users categorized by PPI usage during follow-up. New PPI users: PPIs during the follow-up period but not the history period; continuous PPI users: PPIs during the history and follow-up periods; all PPI users: PPIs during the follow-up period; Non-users of PPI: no PPIs during the follow-up period.

 
nsNSAID and PPI utilization
Treatment days were truncated at 365 for 7.25% of the patients who refilled their medication more frequently than would be expected based on days of supply. Duration of nsNSAID supply was longer among patients who received PPI prescriptions during follow-up than among those who did not (median 168 vs 154 days, respectively). Mean duration of nsNSAID supply was longer than median duration (Table 2); the same pattern was observed for duration of PPI supply and duration of nsNSAID–PPI overlap.


View this table:
[in this window]
[in a new window]

  		TABLE 2. Patterns of nsNSAID and PPI utilization during follow-up

 
Among patients who received PPI prescriptions, the duration of PPI supply was lower than the duration of nsNSAID supply (median 112 vs 168 days, respectively; Table 2). The nsNSAID–PPI overlap ratios indicated that among patients who received PPI prescriptions, PPIs were used on ~50% of the days during which nsNSAIDs were used. This was true whether patients were assumed to consume their medication as prescribed (primary analysis) or only on 80% of days of NSAID and PPI therapy (sensitivity analysis to address non-adherence).

Factors associated with concurrent nsNSAID–PPI use
When considered separately in the univariate models, being female and all GI risk factors were significant predictors of both PPI use and any overlap in supply of nsNSAIDs and PPIs during follow-up (Table 3). GPA use in the prior year (which included prior PPI use) had the strongest association with PPI utilization during follow-up: the odds of any PPI use were 18 times higher and the odds of any nsNSAID–PPI overlap were 20 times higher among those with prior GPA use than those without. The next strongest predictor in the univariate analyses was a history of PUB in the prior year, which increased the odds of both PPI use and nsNSAID–PPI overlap 6-fold.


View this table:
[in this window]
[in a new window]

  		TABLE 3. Univariate analysis and adjusted logistic regression models of any PPI use (prescription) and of any concomitant nsNSAID–PPI use (NSAID–PPI overlap)

 
Pairwise correlation analyses did not indicate colinearity between independent variables; the largest r was 0.47 for history of CV disease and concurrent aspirin use at the index date. Therefore, all variables were retained in the final multivariate models.

After adjusting for all other variables in the multivariate model, those that remained significantly predictive of any PPI use during follow-up included advanced age (≥65 yrs), all-cause hospitalization in the prior year and diagnoses in the prior year of CV disease or dyspepsia/abdominal pain (Table 3). Presence of these factors was associated with 18–47% higher odds of PPI use. GPA use in the prior year had the largest effect in the multivariate model, being associated with 16 times higher odds of any PPI use during follow-up.

Variables in the multivariate model that remained significantly predictive of an increased likelihood of nsNSAID–PPI overlap during follow-up after adjusting for other variables included advanced age and all-cause hospitalization in the prior year. In particular, the odds ratio of nsNSAID–PPI overlap was 1.21 times higher among patients aged ≥65 yrs and 1.55 times higher among those reporting any hospitalizations during the prior year (Table 3). GPA use in the prior year was associated with 19 times higher odds of nsNSAID–PPI overlap. Co-prescriptions at the index date for anti-coagulants or oral corticosteroids increased the odds of nsNSAID–PPI overlap by 64 and 68%, respectively.

Among patients who used PPIs during the follow-up, the only independent variable in the generalized linear model that was significantly associated with the amount of overlap between days of nsNSAID supply and days of PPI supply was a history of GPA use during the prior year (Table 4). On average, patients receiving PPIs during follow-up who had received any GPA during the year prior to the index date had 2.46 times higher nsNSAID–PPI overlap ratio than those with no GPA use during the prior year.


View this table:
[in this window]
[in a new window]

  		TABLE 4. Generalized linear model examining the degree of association between risk factors and nsNSAID–PPI overlap among PPI users (n = 1586) during follow-up

 

   Discussion
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
Acknowledgements
 References
 
Given the high prevalence of nsNSAID use, particularly among elderly and other high-risk populations, preventing nsNSAID-related gastropathy continues to represent an important public health objective. This study found that chronic nsNSAID users with GI risk factors were significantly more likely to receive PPI prescriptions than those at lower GI risk, indicating that UK GPs consider patients’ GI risk level to some extent when prescribing nsNSAIDs. However, 53% of chronic nsNSAID users who did not receive any PPIs during follow-up had one or more GI risk factors; 12% had two GI risk factors and 7% had three or more. It is possible that some patients received GPAs other than PPI, which were not assessed in the follow-up of this study. Nevertheless, this would not represent optimal care since H2RAs are less effective in the prevention of GI events than PPIs and misoprostol is associated with more side-effects than PPIs [32].

Similar patterns of incomplete adherence to clinical practice guidelines that recommend gastroprotection for at-risk nsNSAID recipients [7–9] have been documented elsewhere. Although these studies do not always distinguish between PPIs and other GPAs, the results indicate lower overall utilization of GPAs. A study contacted in The Netherlands found that only 7% of new nsNSAID users received a GPA [16]. Although patients at higher risk of GI complications were more likely to receive GPA, 87 and 81% of the patients with one and two risk factors respectively were not managed by any GI preventive strategy. Two studies in Canada report rates of nsNSAID plus concomitant GPA use that range from 10% to 45% of nsNSAID users [18, 19]. Similarly, a study in USA found that only 33% of nsNSAID users received any GPA, with less than 5% receiving concomitant PPI therapy [17]. Concomitant use of GPA therapy was higher among patients at increased risk of GI complications; ~46% of the nsNSAID users with at least two risk factors were prescribed GPA therapy.

Among chronic nsNSAID users who did receive PPIs during follow-up, PPI supply overlapped only half the days of nsNSAID supply. The generalized linear model indicated that patients who had used a GPA in the prior year had more complete nsNSAID–PPI overlap during follow-up, but no other GI risk factors significantly predicted the degree of nsNSAID–PPI overlap. Although reasons for PPI prescribing could not directly be assessed in this study, this finding suggests that GPs may have prescribed PPIs to many chronic nsNSAID users on an ‘as-needed’ basis to treat symptoms such as dyspepsia instead of for the prevention of nsNSAID-associated gastropathy. However, GI prevention is not required only in patients experiencing GI symptoms such as dyspepsia, as these symptoms are not a predictor of serious GI complications [33].

Similarly to this study, Sturkenboom et al. [20] analysed overlapping nsNSAID-GPA utilization using records from the Integrated Primary Care Information database in The Netherlands from January 1996 to April 2002. For patients who received PPI co-prescriptions within 2 days of an nsNSAID prescription, 82% of days of nsNSAID supply were overlapped by PPI supply. For patients who received H2RA co-prescriptions, the nsNSAID-H2RA overlap ratio was 0.73. The higher nsNSAID–PPI overlap ratio in the Dutch study than in this one may in part be due to the exclusion of patients who were prescribed GPAs for reasons other than prevention of nsNSAID-related upper GI toxicity. Sturkenboom et al. [20] categorized patients as adherent if days of GPA supply overlapped at least 75% of days of nsNSAID supply. By this definition, 31% of PPI users and 44% of H2RA users were found to be non-adherent.

Actual patient consumption of medications cannot be assessed in retrospective prescription database analyses like the present study and that of Sturkenboom et al. [20]. These studies measure prescriber adherence to recommended clinical practice rather than patient adherence to prescribed dosing. Consequently, findings are likely to overestimate nsNSAID–GPA overlap because patient adherence to prescribed daily GPA therapy (as well as NSAID therapy) is known to be poor. In one UK study, 84% of the patients had incomplete adherence to long-term PPI prescriptions [34]. Among PPI users with incomplete adherence to daily therapy, the explanation most often cited (by 80% of respondents) was that treatment was taken only when symptoms were a problem [34]. Achieving satisfactory patient adherence to daily PPI use is likely to be difficult because of several factors known to reduce patient adherence to long-term therapy [29]: polypharmacy (separate PPI and nsNSAID pills), the frequent absence of symptoms before a serious GI complication and the resulting lack of perceived effect of PPIs. Programs that may contribute to improvements in the observed inadequacy of practice include physician education on the nsNSAID GI-related risks and the benefits of nsNSAID–PPI concomitant use, prospective drug utilization review where pharmacists and physicians could collaborate in ascertaining that nsNSAIDs and PPI are prescribed to patients at high risk of GI complications, and patient education of the associated risks of non-adherence to PPI use.

This study has several strengths, most notably the analysis of real-world prescribing patterns using a large, validated, population-based database. Restricting analysis to chronic nsNSAID users reduced confounding factors. Furthermore, prescribing practice was likely more uniform than during the 1996–2002 period examined by Sturkenboom et al. [20], which encompassed years before and after coxibs became available and which saw a large increase in the use of gastroprotective strategies by practitioners in The Netherlands [16]. Follow-up in the present study was from 1 July 2002 to 31 December 2003, well after the introduction of coxibs but before the withdrawal of rofecoxib and valdecoxib and heightening of concern about CV side-effects of coxibs and nsNSAIDs.

This study also has limitations, many of which are common to retrospective database analyses, including lack of randomization. Although analyses controlled for known GI risk factors, results could be influenced by other variables not recorded in the MediPlus database. This study provides an incomplete picture of PPI utilization in nsNSAID users because the endpoints were written prescriptions—it was not possible to verify whether patients filled these prescriptions, or to monitor actual medication consumption or adherence to prescribed dosing. Consequently, it was not possible to verify that both nsNSAIDs and PPIs were actually taken on days scored for nsNSAID–PPI overlap. To simulate patient non-adherence, the effect on nsNSAID–PPI overlap ratio of 20% non-adherence to prescribed dosing frequency was tested in a sensitivity analysis. Overlap ratios were similar in the primary and sensitivity analyses, differing by a mean of 0.02 and a median of 0.04. It remains unknown, however, whether this reflects the true level of adherence in this cohort.

In conclusion, although chronic nsNSAID users at high risk for GI events were more likely to receive a PPI prescription than those with low GI risk, PPI supply overlapped only 51% of mean days of nsNSAID supply. These results support the conclusion of Sturkenboom et al. [20] that even when GPAs are prescribed to nsNSAID users for gastroprotection, their utilization pattern correlates poorly with nsNSAID use. These patterns of suboptimal PPI utilization may increase the risk for GI adverse events among chronic nsNSAID users. Assessments of the GI safety of nsNSAID–PPI co-therapy that are based on clinical trial results [35] may not be valid given the utilization patterns seen in actual clinical practice.

Formula


   Acknowledgements
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 Acknowledgements
References
 
This study was made possible by a grant from Merck & Co., Inc. This article was prepared with the assistance of BioMedCom Consultants Inc., Montreal, Canada.

Disclosure statement: E.H.D.rp. was an employee of Merck & Co. at the time of submission of this article. P.M. is an employee of Merck & Co. All other authors have no conflicts of interest.


   References
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 Acknowledgements
 References
 

  1. Laine L. Approaches to nonsteroidal anti-inflammatory drug use in the high-risk patient. Gastroenterology (2001) 120:594–606.[CrossRef][Medline]
  2. IMS Health Disease Analyzer-Mediplus as assessed by Merck Sharp & Dohme Limited. (2005) Mar.
  3. Fricker J. for Mundipharma International Ltd. Pain in Europe - a report. On Pain in Europe website. http://www.painineurope.com/user_site/index.cfm?item_id=4405413. (updated 5 October 2004; 6 January 2006, date last accessed).
  4. MacDonald TM. Epidemiology and pharmacoeconomic implications of non-steroidal anti-inflammatory drug-associated gastrointestinal toxicity. Rheumatology (2000) 39(Suppl 2):13–20.[Abstract]
  5. Hernández-Díaz S, Garcia Rodríguez LA. Association between nonsteroidal anti-inflammatory drugs and upper gastrointestinal tract bleeding/perforation: an overview of epidemiologic studies published in the 1990s. Arch Intern Med (2000) 160:2093–9.[Abstract/Free Full Text]
  6. National Institute for Clinical Excellence, National Health Service. Management of dyspepsia in adults in primary care. Clinical Guideline 17. On NICE website. http://www.nice.org.uk/pdf/CG017NICEguideline.pdf. (updated 2004 August; 14 June 2005, date last accessed).
  7. Jordan KM, Arden NK, Doherty M, et al. EULAR Recommendations 2003: an evidence based approach to the management of knee osteoarthritis: report of a task force of the Standing Committee for International Clinical Studies Including Therapeutic Trials (ESCISIT). Ann Rheum Dis (2003) 62:1145–55.[Abstract/Free Full Text]
  8. American College of Rheumatology Subcommittee on Rheumatoid Arthritis Guidelines. Guidelines for the management of rheumatoid arthritis: 2002 Update. Arthritis Rheum (2002) 46:328–46.[CrossRef][ISI][Medline]
  9. Tannenbaum H, Peloso PM, Russell AS, Marlow B. An evidence-based approach to prescribing NSAIDs in the treatment of osteoarthritis and rheumatoid arthritis: the second Canadian consensus conference. Can J Clin Pharmacol (2000) 7(Suppl A):4A–16A.[Medline]
  10. Hawkey C, Talley NJ, Yeomans ND, et al. Improvements with esomeprazole in patients with upper gastrointestinal symptoms taking non-steroidal antiinflammatory drugs, including selective COX-2 inhibitors. Am J Gastroenterol (2005) 100:1028–36.[CrossRef][Medline]
  11. Graham DY, Agrawal NM, Campbell DR, et al. Ulcer prevention in long-term users of nonsteroidal anti-inflammatory drugs: results of a double-blind, randomized, multicenter, active- and placebo-controlled study of misoprostol vs lansoprazole. Arch Intern Med (2002) 162:169–75.[Abstract/Free Full Text]
  12. Lai KC, Lam SK, Chu KM, et al. Lansoprazole reduces ulcer relapse after eradication of Helicobacter pylori in nonsteroidal anti-inflammatory drug users–a randomized trial. Aliment Pharmacol Ther (2003) 18:829–36.[CrossRef][Medline]
  13. Chan FK, Chung SC, Suen BY, et al. Preventing recurrent upper gastrointestinal bleeding in patients with Helicobacter pylori infection who are taking low-dose aspirin or naproxen. N Engl J Med (2001) 344:967–73.[Abstract/Free Full Text]
  14. Laine L. The role of proton pump inhibitors in NSAID-associated gastropathy and upper gastrointestinal symptoms. Rev Gastroenterol Disord (2003) 3(Suppl 4):S30–9.
  15. Higham J, Kang JY, Majeed A. Recent trends in admissions and mortality due to peptic ulcer in England: increasing frequency of haemorrhage among older subjects. Gut (2002) 50:460–4.[Abstract/Free Full Text]
  16. Sturkenboom MC, Burke TA, Dieleman JP, Tangelder MJ, Lee F, Goldstein JL. Underutilization of preventive strategies in patients receiving NSAIDs. Rheumatology (2003) 42(Suppl 3):iii23–31.[Abstract]
  17. Smalley W, Stein CM, Arbogast PG, Eisen G, Ray WA, Griffin M. Underutilization of gastroprotective measures in patients receiving nonsteroidal antiinflammatory drugs. Arthritis Rheum (2002) 46:2195–200.[CrossRef][ISI][Medline]
  18. Rahme E, Marentette MA, Kong SX, LeLorier J. Use of NSAIDs, COX-2 inhibitors, and acetaminophen and associated coprescriptions of gastroprotective agents in an elderly population. Arthritis Rheum (2002) 47:595–602.[CrossRef][ISI][Medline]
  19. Sebaldt RJ, Petrie A, Goldsmith CH, Marentette MA. Appropriateness of NSAID and coxib prescribing for patients with osteoarthritis by primary care physicians in Ontario: results from the CANOAR study. Am J Manag Care (2004) 10:742–50.[Medline]
  20. Sturkenboom MC, Burke TA, Tangelder MJ, Dieleman JP, Walton S, Goldstein JL. Adherence to proton pump inhibitors or H2-receptor antagonists during the use of non-steroidal anti-inflammatory drugs. Aliment Pharmacol Ther (2003) 18:1137–47.[CrossRef][Medline]
  21. Goldstein JL. Challenges in managing NSAID-associated gastrointestinal tract injury. Digestion (2004) 69(Suppl 1):25–33.[CrossRef][Medline]
  22. Lawrenson R, Williams T, Farmer R. Clinical information for research; the use of general practice databases. J Pub Health Med (1999) 21:299–304.[Abstract/Free Full Text]
  23. Lawrenson RA, Coles G, Walton K, Farmer RDT. Characteristics of practices contributing to the MediPlus database and the implications for its use in epidemiological research. J Inform Prim Care (1998) 1:14–8.
  24. Langman M, Kahler KH, Kong SX, et al. Drug switching patterns among patients taking non-steroidal anti-inflammatory drugs: a retrospective cohort study of a general practitioners database in the United Kingdom. Pharmacoepidemiol Drug Saf (2001) 10:517–24.[CrossRef][Medline]
  25. Hunt RH, Barkun AN, Baron D, et al. Recommendations for the appropriate use of anti-inflammatory drugs in the era of the coxibs: defining the role of gastroprotective agents. Can J Gastroenterol (2002) 16:231–40.[ISI][Medline]
  26. Wolfe MM, Lichtenstein DR, Singh G. Gastrointestinal toxicity of nonsteroidal antiinflammatory drugs. N Engl J Med (1999) 340:1888–99.[Free Full Text]
  27. Griffin MR. Epidemiology of nonsteroidal anti-inflammatory drug-associated gastrointestinal injury. Am J Med (1998) 104:23S–9S.[Medline]
  28. Wolfe F, Anderson J, Burke TA, Arguelles LM, Pettitt D. Gastroprotective therapy and risk of gastrointestinal ulcers: risk reduction by COX-2 therapy. J Rheumatol (2002) 29:467–73.[ISI][Medline]
  29. World Health Organization. Adherence to long-term therapies: evidence for action (2003) Geneva: Switzerland.
  30. Duan N, Manning WG, Morris CN, Newhouse JP. A comparison of alternative models for the demand for medical care. J Bus Econ Statist (1983) 1:115–26.[CrossRef]
  31. Manning WG, Mullahy J. Estimating log models: to transform or not to transform? J Health Econ (2001) 20:461–94.[CrossRef][ISI][Medline]
  32. Rostom A, Dube C, Wells G, et al. Prevention of NSAID-induced gastroduodenal ulcers. Cochrane Database Syst Rev (2004) 4(CD002296):1–111.
  33. Singh G, Ramey DR, Morfeld D, Shi H, Hatoum HT, Fries JF. Gastrointestinal tract complications of nonsteroidal anti-inflammatory drug treatment in rheumatoid arthritis. A prospective observational cohort study. Arch Intern Med (1996) 156:1530–6.[Abstract]
  34. Hungin AP, Rubin G, O’Flanagan H. Factors influencing compliance in long-term proton pump inhibitor therapy in general practice. Br J Gen Pract (1999) 49:463–4.[Medline]
  35. Hooper L, Brown TJ, Elliott R, Payne K, Roberts C, Symmons D. The effectiveness of five strategies for the prevention of gastrointestinal toxicity induced by non-steroidal anti-inflammatory drugs: systematic review. Br Med J (2004) 329:948–57.[Abstract/Free Full Text]
Submitted 5 May 2006; revised version accepted 12 December 2007.
Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrowOA All Versions of this Article:
47/4/458    most recent
kem375v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Suh, D.-C.
Right arrow Articles by Mavros, P.
PubMed
Right arrow PubMed Citation
Right arrow Articles by Suh, D.-C.
Right arrow Articles by Mavros, P.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?