Education-based approach to addressing non-evidence-based practice in preventing NSAID-associated gastrointestinal complications

Table 1 A summary of the main questions (from a total of 22) assessing physicians’ knowledge of current evidence in the field of NSAID use and adverse effects

NSAID use is associated with adverse effects. Which of the following do you believe is not associated with NSAID use?

What is the expected annual incidence of upper GI complications in patients taking NSAIDs, as reported in the most recent large outcome studies?

The occurrence of dyspepsia in patients who take NSAIDs has been reported to be less than 25% (true or false)

NSAIDs may induce GI complications in the lower GI tract (true or false)

Which of the following factors do you believe is/are risk factors for GI complications in patients who take NSAIDs? (list)

Which of the following NSAIDs do you believe is more toxic to the GI tract? (list)

Concerning COX-2 selective inhibitors, for each of the following, indicate whether the statement is true or false:

They are not as effective as traditional NSAIDs in the treatment of OA or RA

The use of these compounds is associated with a 50% reduction in the risk of GI complications compared to NSAIDs

The concomitant use of low-dose aspirin reduces or eliminates the GI benefit of these compounds when compared to NSAIDs

The use of these compounds has been associated with an increased risk of CV events

In high-risk patients, the combination of NSAIDs plus a PPI is safer than a coxib alone

Concerning gastroprotective agents, indicate for each of the following statements whether they are true or false:

H2-RAs are effective in the prevention of gastric ulcers, duodenal ulcers, and GI complications

PPIs are effective in the prevention of gastric ulcers, duodenal ulcers, and GI complications

Misoprostol is effective in the prevention of gastric ulcers, duodenal ulcers, and GI complications

Which of the following agents has been proved to be effective in the treatment or prevention of NSAID-induced dyspepsia? (list)

NSAID: Nonsteroidal anti-inflammatory drug; GI: Gastrointestinal; CV: Cardiovascular; PPI: Proton pump inhibitor; H2-RAs: H2 receptor antagonists.

Table 2 Responses to the question, “Which of the following factors do you believe is/are risk factors for GI complications in patients who take NSAIDs”n (%)

	Rheumatologists	Orthopedic surgeons	Others	Total
History of peptic ulcer	115 (99.1)	275 (98.2)	21 (100.0)	411 (98.6)
History of complicated peptic ulcer	116 (100.0)	275 (98.2)	21 (100.0)	412 (98.8)
Age > 65 yr	114 (98.3)	229 (81.8)	18 (90.4)	361 (86.6)
Concomitant use of low-dose aspirin for CV prevention	114 (98.3)	228 (81.4)	19 (90.4)	361 (86.6)
Concomitant use of anticoagulants	112 (96.5)	247 (88.2)	20 (95.3)	379 (90.9)
Helicobacter pylori infection	103 (88.8)	257 (91.8)	19 (90.4)	379 (90.9)
Smoking	87 (75.00)	223 (79.6)	13 (61.7)	323 (77.5)
Dyspepsia history	73 (62.9)	250 (89.3)	19 (90.4)	342 (82.0)
Alcohol	105 (90.5)	257 (91.8)	20 (95.3)	382 (91.6)
High dose of NSAIDs	113 (97.4)	275 (98.2)	21 (100.0)	409 (98.1)

Table 3 Characteristics of patients included in the educational program of the study¹n (%)

Variable	Phase I (n = 1732)	Phase II (n = 1722)
Age (mean ± SD)	61.06 ± 13.37	60.81 ± 13.89
Female	1038 (60.4)	980 (57.6)
History of ulcer	238 (13.7)	307 (17.8)
History of ulcer bleeding	61 (3.5)	69 (4.0)
ASA use	167 (9.6)	168 (9.8)
CV history	203 (11.7)	205 (11.9)
Increased blood pressure	845 (48.8)	810 (47.0)
Anticoagulant use	126 (7.3)	120 (7.0)
Corticosteroid use	162 (9.3)	190 (11.0)
History of dyspepsia	782 (45.1)	766 (44.5)

¹No statistical differences were found between patients enrolled in the two phases.

Phase I: Before physicians attended the evidence-based seminar; Phase II: After the seminar; ASA: Aspirin.

Table 4 Prescription of NSAIDs to patients in each of the two study phases of the educational program n (%)

Drug therapy	Phase I		Phase II
	Before visit	After visit	Before visit	After visit
No NSAID therapy	718 (41.45)	162 (9.35)	653 (37.92)	190 (11.03)
NSAID therapy	1014 (58.55)	1570 (90.65)	1069 (62.08)	1532 (88.97)
Aceclofenac	146 (8.43)	248 (14.32)b	148 (8.59)	202 (11.73)b
Celecoxib	45 (2.60)	100 (5.77)b	35 (2.03)	116 (6.74)b
Diclofenac	229 (13.22)	271 (15.65)	238 (13.82)	270 (15.68)
Etoricoxib	16 (0.92)	46 (2.66)b	18 (1.05)	79 (4.58)b
Ibuprofen	281 (16.22)	432 (24.94)b	297 (17.25)	406 (23.58)b
Indomethacin	63 (3.64)	62 (3.58)	73 (4.24)	75 (4.36)
Ketorolac	15 (0.87)	25 (1.44)	28 (1.63)	31 (1.80)
Meloxicam	71 (4.10)	234 (13.51)b	101 (5.87)	215 (12.49)b
Piroxicam	74 (4.27)	75 (4.33)	64 (3.72)	64 (3.72)
Other NSAIDs includes aproxen)	19 (1.10)	22 (1.27)	16 (0.93)	28 (1.63)
Analgesics
Paracetamol	137 (7.91)	120 (6.93)	136 (7.90)	122 (7.08)
Metamizol	35 (2.02)	28 (1.62)	53 (3.08)	26 (1.51)
Total	1732 (100)		1722 (100)

^bP < 0.001 vs before the visit.

Table 5 Risk factors (RFs) of patients reported by doctors in the educational program according to either a non-restrictive or a restrictive definition¹n (%)

Number of RFs	Non-restrictive		Restrictive
	Phase I2	Phase II3	Phase I2	Phase II3
0	347 (20.03)	352 (20.44)	961 (55.48)	891 (51.74)
1	660 (38.11)	598 (34.73)	558 (32.22)	573 (33.28)
2	517 (29.85)	536 (31.13)	176 (10.16)	213 (12.37)
> 2	208 (12.01)	236 (13.70)	37 (2.14)	45 (2.61)
Total	1732 (100)	1722 (100)	1732 (100)	1722 (100)

¹A non-restrictive definition of risk factors for NSAID-related complications included age > 60 years, history of dyspepsia, history of either complicated or non-complicated ulcer, concomitant therapy with NSAIDs and low-dose aspirin, or anticoagulants or corticosteroids. A restrictive definition of risk factors included age ≥ 70 years, history of complicated or non-complicated ulcer, concomitant therapy with NSAIDs and low-dose aspirin, or anticoagulants or corticosteroids;

²In Phase I, the specialists received an anonymous questionnaire regarding data and prescriptions for their last five consecutive patients;

³In Phase II, the process was repeated 4-5 mo later after specialists had attended an evidence-based seminar that reviewed current evidence on NSAID-related issues, with a focus on GI prevention strategies in NSAID users.

Table 6 Proportion of patients on NSAID therapy that received concomitant therapy with a PPI or misoprostol after the medical visit, according to the number of RFs n (%)

Number of RFs	Non-restrictive		Restrictive
	Phase I	Phase II	Phase I	Phase II
0	268/347 (77.2)	283/352 (80.4)	782/961 (81.4)	728/891 (81.7)
1	536/660 (81.2)	499/598 (83.4)	471/558 (84.4)	504/573 (87.9)
2	453/517 (87.6)	456/536 (85.1)	151/176 (85.8)	168/213 (78.9)
> 2	175/208 (84.1)	201/236 (85.2)	28/37 (75.7)	39/45 (86.7)