Drug utilization evaluation of nonprescription H₂-receptor antagonists and alginate-containing preparations for dyspepsia

Abstract

Aims

To evaluate the use, efficacy and adverse effects of nonprescription H₂-receptor antagonists and alginate-containing preparations obtained from community pharmacies.

Methods

Questionnaires were distributed to customers from 39 pharmacies in Scotland and Wales.

Results

Of 767 customers recruited, 608 (79.3%) returned an initial questionnaire and 472 (61.5%) customers a second questionnaire. The vast majority of respondents (424, 69.7%) had suffered their symptoms on three or more occasions and 369 (60.7%) had previously tried medicines to relieve their symptoms. Referrals to a doctor were less frequent than recommended in guidelines and few of those who were referred actually saw a doctor. Over a quarter of those returning the second questionnaire claimed to be taking more than one product simultaneously for symptom control. Eight customers who were taking prescribed ulcer-healing drugs obtained H₂-receptor antagonists. The majority of respondents (355/472, 75.2%) obtained some or complete symptom relief using the product obtained and 369/472 (78.2%) were completely satisfied with their product. H₂-receptor antagonists were more likely to produce complete relief of symptoms than alginate-containing preparations (P < 0.05). Only 14 respondents (3.0%) reported side-effects from the product used which were mostly gastro-intestinal.

Conclusions

The study demonstrated that drug utilization studies are feasible to carry out in a community pharmacy setting. While the results support published evidence of the efficacy and minimal toxicity of these products, they also highlight the possibility of H₂-receptor antagonists being used outwith their licenced indications.

Introduction

Investigations into purchasers' use of nonprescription medicines and their therapeutic outcomes are few in number. Few controlled, double-blind studies of nonprescription medicines have been published and this applies even to recently re-regulated products, such as H₂-receptor antagonists. It has been shown however, that both famotidine 10 mg and ranitidine 75 mg have a slower onset but more prolonged action than calcium carbonate tablets [1, 2] and that famotidine 10 mg has superior efficacy to alginate tablets in treating heartburn [3]. Ranitidine 75 mg and cimetidine 200 mg have been shown to be superior to placebo in relieving symptoms of gastro-oesophageal reflux disease [4], while ranitidine and famotidine have similar effects on nocturnal intragastric acidity [5]. A low incidence of adverse events has been shown for ranitidine [4, 6] and cimetidine [4].

A programme of product re-regulation in the UK has resulted in a substantial increase in self-medication [7, 8], and the availability of products without prescription may have contributed to changes in the prescribing behaviour of some general practitioners [9]. Recently re-regulated products often have the benefit of extensive experience with prescribed use, however, doses may differ for nonprescription use and indications may be restricted. While the availability of H₂-receptor antagonists without prescription may reduce overall health care costs by reducing the number of visits to general practitioners [10], concern has been expressed that nonprescription products for dyspepsia are likely to be used improperly [11–13].

Indigestion remedies constituted £63.7 m of nonprescription medicine purchases in 1995 [7], after the introduction of H₂-receptor antagonists, increasing to £69.2 m in 1997 [8]. However general practitioners view their prescribing of this class of drugs as having changed little as a result [11]. The need for information about how these products are used in the domiciliary setting is therefore of importance. Several methods have been used to study use of nonprescription medicines including interviews about products used in the past [14] or recently purchased [15], as well as health diaries [16] and postal questionnaires distributed by community pharmacists [13, 17, 18]. This paper describes a drug use evaluation of H₂-receptor antagonists and alginate-containing preparations for dyspepsia using the latter method as this has been shown to achieve comparable or higher response rates than other methods and is less intensive with regards to researchers' time than client interviews.

Methods

A pilot study was carried out in six pharmacies in Grampian from August to November 1996. As a result changes were made to the customer surveys and study design.

Pharmacists were informed of the study at continuing education meetings or via continuing education tutors, with the aim of recruiting 20 pharmacists from both Scotland and Wales. Once ethics committee approval had been granted, pharmacists were asked to recruit customers over 18 years of age to the study who requested a H₂-receptor antagonist or alginate-containing preparation, after this was deemed to be appropriate for sale. This main study was undertaken between June 1997 and February 1998. One pharmacy in Scotland dropped out shortly after the study began, due to pressure of work, therefore 20 pharmacies in Wales and 19 in Scotland took part in the study. Once written consent was obtained, customers were required to complete an initial questionnaire on their symptoms, other medication and use of dyspepsia treatment, which was provided by the recruiting pharmacist. Customers who returned the questionnaire to the researchers in a postage paid envelope were sent a second questionnaire after 2 weeks. One postal reminder was issued for each questionnaire. Pharmacists were asked to keep a written count of the number of customers who refused to participate and to ascertain the reasons why if possible. The information relating to prescribed medication and medical conditions provided by a small number of customers in their questionnaire responses was compared to that obtained from their general practice medical records. This sample comprised all participating customers who were registered with four different general practices in one region, to enable ease of accessibility.

All data were analysed using Epi-info version 6.04. Groups were compared statistically using chi-square (χ²) tests.

Results

Recruitment and response rate

Overall 391 customers were recruited by pharmacists in Scotland and 376 in Wales (767 in total). Only 20 pharmacies kept records of customers who refused to participate, noting a total of 362 refusals. A total of 546 customers were successfully recruited from these 20 pharmacies, thus the refusal rate was 40%. Reasons given for nonparticipation were no time (n = 113, 31.2%), no interest (n = 61, 16.9%), the medicine was for somebody else (94, 26.0%), language difficulty (n = 10, 2.8%), no reason given (n = 42, 11.6%) and other reasons given (n = 42, 11.6%). The range of recruitment varied considerably, from 0 to 91 over the 6 month study period. There was no significant difference in the recruitment rates between Scotland and Wales or in the rates of questionnaire return [19].

Of the 608 respondents to the first questionnaire (79.3% response rate), 354 were female and 252 male (two did not state); 133 were aged under 40 years, 333 were between 40 and 65 years and 138 were aged over 65 years (four did not state). Thirteen of the females were pregnant and none was breast-feeding. There were 472 respondents to the second questionnaire (77.6%).

Symptoms reported and comorbidity

Respondents to the first questionnaire provided information on their medical problems and prescribed therapy, the details of which are shown in Table 1. Many respondents cited factors which they felt aggravated their symptoms, the most common being rich, spicy or greasy food (144 respondents) and excess intake of alcohol (65), large meals (35) or coffee (11). Stress was cited as an aggravating factor by 85 respondents and smoking by 21. Some respondents (34) felt that their symptoms were aggravated or caused by their prescribed medicine; 13 of those taking nonsteroidal anti-inflammatory drugs, five taking aspirin and five taking oral corticosteroids.

Table 1.

Medical conditions and prescribed medicines reported by customers obtaining H₂-receptor antagonists or alginate-containing preparations without prescription.

Medical conditions	Number reporting	Prescribed medicines	Number claiming use
Upper GI	34	H2-receptor antagonists	19
		Proton pump inhibitors	25
		Antacids/alginates	20
		Misoprostol	4
		Lower GI	36
		Other GI drugs	18
Angina/IHD	17	Nitrates, calcium-channel blockers and potassium-channel activators	39
		Aspirin	22
Other cardiovascular	73	Other cardiovascular drugs	54
Respiratory	41	Bronchodilators and prophylactic therapy	26
Musculoskeletal	78	NSAIDs	38
Pregnancy	13	Alginate	1
		Iron preparations	4

The commonest symptoms described and their reported duration are shown in Table 2. Other conditions reported included cough (3), breathlessness (4), headache (3), hiatus hernia (4) and interstitial cystitis (1). The vast majority of customers (424, 69.7%) had suffered their symptoms on three occasions or more, only 61 (10.0%) experiencing them for the first time. A total of 235 (38.7%) respondents stated they had contacted a doctor previously about their symptoms and 369 (60.7%) had previously tried medicines to relieve their symptoms. The majority of these had used simple antacids, although H₂-receptor antagonists had been used by 68 respondents, which could have been obtained on prescription or purchased. Proton pump inhibitors had been taken by 22, prokinetic agents by two and one respondent had received a course of triple therapy for Helicobacter pylori eradication. The overall efficacy of these products was reported to be low, but there was a trend towards H₂-receptor antagonists producing the greatest symptom relief (Table 3).

Table 2.

Frequency of symptom presentation in customers obtaining H₂-receptor antagonists or alginate-containing preparations without prescription.

Symptom	Total describing symptom (n)	Obtained alginate (n)	Obtained H2-receptor antagonist (n)	Symptom present 7 - 28 days (n)	Symptom present over 1 month (n)
Feeling bloated/wind	259	160	103	20	15
Burning in throat/acid in mouth	298	190	109	22	31
Nausea	140	82	58	12	12
Stomach pain	233	131	104	15	24
Discomfort around chest area	227	149	78	15	22
Tight chest pain, spreading to arms or jaw	44	31	13	2	6

1. Three patients obtained combination products, two of these had stomach pain with wind, one had burning in mouth.

2. Customers who did not indicate the product obtained are excluded from this table.

Table 3.

Estimated efficacy (customer self-reports) of nonprescription medicines in relieving dyspepsia symptoms.

	Simple antacid	Alginate-containing product	H2-receptor antagonist
Estimates of efficacy of medicines previously used	n = 208*	n = 63*	n = 68*
Complete or some symptom relief	67 (32.2%)	15 (23.8%)	30 (44.1%)
Initial relief but then symptoms return	95 (46.1%)	29 (46.0%)	25 (36.8%)
No symptom relief	26 (12.5%)	5 (7.9%)	3 (4.4%)
Estimates of efficacy of medicines obtained in the study	–	n = 297	n = 178
Complete or some symptom relief	–	210 (70.7%)	144 (80.9%)
Initial relief but then symptoms return	–	80 (26.9%)	32 (18.0%)
No symptom relief	–	7 (2.4%)	2 (1.1%)

^*Not all responders who indicated use of a product provided an estimate of efficacy.

Efficacy and toxicity of products obtained

The products purchased were not specified by 13 customers. Of the remainder, 355 (58.4%) obtained an alginate-containing preparation, including all 13 who were pregnant, and 240 (39.5%) a H₂-receptor antagonist. Most customers (523, 86.0%) asked for the product they obtained by name, the majority of whom (401, 66.0%) had used it previously. Advertising contributed to 87 (16.6%) requests, recommendations by friends or relatives to 119 (22.8%) and by doctors to 113 (21.6%).

A total of 162 customers claimed to have one or more reasons which should, according to guidelines [20], have resulted in referral to their general practitioner. Of these, 45 claimed to experience tight chest pain radiating to chest or jaw, 43 were aged over 40 years and presenting their symptoms for the first time and 103 stated they were taking prescribed ulcer-healing agents or ulcerogenic drugs. However only 48 of these were referred to a doctor. In addition, eight customers claiming in their responses to the questionnaire to be taking prescribed ulcer-healing drugs obtained H₂-receptor antagonists. A total of 175 customers were advised by pharmacy staff to contact a doctor, the large majority of whom (161, 92.0%) were advised to do so if symptoms worsened or were not relieved, that is, they received a conditional referral. Only 24 of the 137 who completed a follow-up questionnaire reported seeing a general practitioner and a further 2 were admitted to hospital for endoscopy. Fifteen of these 26 respondents were among those who fulfilled referral criteria.

Almost all customers who returned follow-up questionnaires had used the product obtained (470/472), 247 (52.6%) of whom claimed to have taken it until the pack was finished, 209 (44.5%) took it a few times and 17 (3.6%) took it once only. A high proportion of these 470 respondents claimed to have obtained complete (249, 53.0%) or some relief (105, 22.3%) from the product, with a further 112 (23.8%) obtaining initial relief, followed by return of symptoms. Only 9 obtained no relief from the product (Table 3). There was a significantly higher proportion of customers who were given H₂-receptor antagonists obtaining relief than those using alginate-containing preparations (χ² = 8.12, d.f. = 3, P < 0.05). Since most customers identified more than one symptom, many were also taking other products and all used the products obtained with varying frequencies, it was not possible to identify whether symptom relief was related to any of these factors individually.

A substantial number of respondents to the follow-up questionnaire claimed to have taken other medicines at the same time as those obtained from the pharmacy. The majority of these (67) were ulcer-healing agents being taken in most (58) cases on the advice of a doctor. A similar number of respondents (65) stated they had taken simple antacids concurrently with the product obtained from the pharmacy. The majority of respondents cited additional methods they were using to relieve symptoms, many of which were in line with the aggravating factors identified. The most frequently cited method was the avoidance of rich, spicy or greasy foods (309 respondents), eating smaller meals (160), weight reduction (118) and reduction of alcohol intake (112). Many fewer respondents cited attempts to reduce stress (7) than were aware of stress as an aggravating factor (85) and only two of the 21 respondents who had previously cited smoking as a factor were attempting to stop.

Most (439, 93.4%) would consider obtaining a further supply of the nonprescription medicine. Furthermore, 368 (78.3%) of the respondents who had used it were completely satisfied with the medicine overall and 96 (20.4%) were partly satisfied. Only 8 (1.7%) were not at all satisfied. Fourteen respondents claimed to have experienced symptoms which could have been side-effects of their dyspepsia treatment, nine of whom were taking alginate-containing preparations. The symptoms reported were diarrhoea, constipation, bloated feelings and flatulence from alginate preparations and dry mouth, altered bowel movements, diarrhoea and constipation from H₂-receptor antagonists.

Reliability of the data

In a small sample of respondents, medical records were examined to ascertain the completeness of questionnaire responses. Fifteen out of the 23 records examined confirmed the information given by the customer, while 8 records gave extra information about prescribed medication being taken concurrently. Two of these customers were prescribed lansoprazole, only one of whom had returned a follow-up questionnaire. It is therefore possible that the data underestimate the concurrent use of prescribed and nonprescribed medication. Incompleteness of the data was also illustrated by 6 respondents to the second questionnaire, who claimed having taken proton pump inhibitors concurrently with their purchased medicine, yet denied having contacted a general medical practitioner.

Discussion

This study has provided information about how nonprescription medicines were used to relieve minor symptoms in uncontrolled (that is ‘real life’) conditions, estimates of their efficacy in this regard and adverse events. Of 767 customers who agreed to participate, 608 (79%) returned the initial questionnaire. This is a high response rate, comparing favourably with those obtained in previous studies using a similar method [13, 17, 18, 21] and is important for the generalisability of the results.

The results suggest that in many cases referrals to a medical practitioner in line with guidelines [20] were not made, however, this may have been due to a lack of information provided by customers to pharmacy staff. Previous studies have shown that some customers do not want [22] or expect advice [23] from pharmacy staff and referral may depend, at least in part, on whether the product is being requested by name or whether the person was asking for advice. The significance of giving information to pharmacy staff may not be apparent to customers nor may the importance of following referral advice.

Of those who were referred, few actually stated following this advice and contacted a doctor. This emphasizes concerns about the possibility of self-treatment of symptoms which may be indicative of serious underlying conditions [12]. Duration of symptoms, their severity, efficacy of the product and whether a direct or conditional referral had been made could be additional factors which determine whether a customer would seek medical advice. In addition a substantial majority of customers had used these products previously, hence requested them by name and some were using other nonprescription products or prescribed ulcer-healing agents concurrently. A Finnish study using a similar methodology found that 75 per cent of responders to questionnaires used these products long-term and that some users who should have seen a doctor about symptoms had not done so [13].

These findings may suggest that more emphasis should be placed on referrals by pharmacy staff, but also indicate the need for complete information to be provided about concurrent medicines, both purchased and prescribed, to inform their advice. The data obtained from respondents to the questionnaires appear to underestimate the extent of concurrent use of medicines as identified by the discrepancy between self-reports of prescribed medication in questionnaires and information contained in medical notes. The information provided to pharmacy staff may be even less complete. Conversely, customers may have received advice to contact a doctor, but not reported this in their responses to questionnaires. Previous work has shown that recall of advice given relating to nonprescription medicines in pharmacies is poor [15].

The products obtained were in the main considered to be effective and were well tolerated, although customers reported greater symptom relief from H₂-receptor antagonists than from alginate-containing preparations. The number of customers describing complete symptom relief was higher than that found in a controlled study [4], which could be due to the use of additional medicines. Adverse effects were minimal and all were previously reported, reflecting previously published work [4, 6]. Product efficacy and safety were the two most important factors identified by pharmacists which could influence their recommendation of a product [24], therefore information about these factors in a real life setting is of value.

The finding that fewer customers using H₂-receptor antagonists experienced return of symptoms than those who had previously used simple antacids is in agreement with the results of published work from controlled studies [1, 2]. The results also suggest that H₂-receptor antagonists had a longer effect on symptom relief than alginate-containing preparations [3].

The study demonstrates the feasibility of obtaining data about product use and efficacy of nonprescription medicines outside controlled conditions. While any such study relies on users' estimates of efficacy, this is the factor most likely to determine future behaviour including use of the same or similar products and consultation with general medical practitioners. Both user estimates [17] and more standard methods of measuring symptom control [25] have been used to study hay fever remedies, demonstrating the efficacy of nonprescription products. Similar symptom severity between patient self-reports and clinical observers has also been found for colds [26].

While there are clearly benefits to be gained from the availability of nonprescription products, as demonstrated by their efficacy and minimal toxicity, this study shows that guidelines on referral to a general medical practitioner, which are based on licensed indications, may not be being followed on all occasions. The regular use of nonprescription products to relieve dyspepsia is also evident from the results, which could also be outwith licensed indications.