Sub-optimal proton pump inhibitor dosing is prevalent in patients with poorly controlled gastro-oesophageal reflux disease


Dr N. T. Gunaratnam, Department of Internal Medicine, Division of Gastroenterology, St Joseph Mercy Hospital, Huron Gastro, 5300 Elliott Drive, Ypsilanti, MI 48197, USA.




Proton pump inhibitors are the most potent drug treatment for gastro-oesophageal reflux disease. Premeal dosing maximizes efficacy while sub-optimal dose timing may limit efficacy.


To determine the prevalence of sub-optimal proton pump inhibitor dosing in a community-based gastro-oesophageal reflux disease population.

Materials and methods

One hundred patients on proton pump inhibitors referred for persistent gastro-oesophageal reflux disease symptoms were questioned about their proton pump inhibitor dosing habits and classified as optimal or sub-optimal dosers. Optimal dosers took proton pump inhibitors with or up to 60 min before meals. Sub-optimal dosers took proton pump inhibitors >60 min before meals, after meals, as needed, or at bedtime.


Forty-six percent dosed optimally. Fifty-four percent dosed sub-optimally with 21 of 54 (39%) dosing >60 min before meals, 16 (30%) after meals, 15 (28%) at bedtime and two (4%) as needed. Only 6% of the subjects on once-daily proton pump inhibitor regimens and 33% of subjects taking proton pump inhibitors two- to three times daily dosed in a manner that maximized acid suppression (15–30 min before a meal).


In this study, 54% of patients dosed proton pump inhibitors sub-optimally and only 12% dosed in a manner that maximized acid suppression. As sub-optimal proton pump inhibitor dose timing can limit efficacy, patients with refractory symptoms should be asked about dose timing to avoid inappropriate and costly dose escalations.


Gastro-oesophageal reflux disease (GERD) is a common and costly clinical syndrome. Up to one-half of all American adults experience symptoms of heartburn at least once a month1–2 and 4–7% have daily reflux symptoms.3 A questionnaire-based study conducted in Minnesota revealed that the prevalence of either heartburn or acid regurgitation in the past year was 59/100 and 20/100 for at least weekly symptoms.4 Dyspepsia is associated with over 2 million out-patient doctor visits annually and almost 40% of gastroenterology referrals in the US.5–6 It has been estimated that in the US more than $2 billion are spent per year on over-the-counter antacids and histamine-2 receptor antagonists. In addition, it is estimated that $6 billion are spent on prescription histamine-2 receptor antagonists and proton pump inhibitors (PPIs).2, 3, 7 A recent evaluation of a rural northeastern managed care organization revealed the mean annual cost of treating GERD patients over nine study years was approximately $510/year and patients with GERD incurred approximately twofold higher medical care costs compared with those without GERD.8

Gastro-oesophageal reflux disease therapy options range from lifestyle modifications to pharmacologic treatment with motility agents, histamine-2 receptor antagonists and PPIs.9 In comparison studies, PPIs have consistently exhibited superior efficacy in the treatment of GERD symptoms and GERD-associated oesophageal erosions.9–13 To inhibit gastric acid secretion PPIs must accumulate in the acidic environment of actively secreting parietal cells, undergo conversion to a reactive species via an acid-catalysed reaction, and covalently inhibit ATPase molecules recruited to the luminal parietal cell surface.12, 14 As gastric acid secretion is maximally stimulated by food ingestion, maximal PPI efficacy should be achieved when they are taken before a meal.12–14 The concept of premeal PPI dosing has been validated in several intragastric pH probe studies.14–16 The objective of this study was to examine PPI dosing practices in GERD patients with poor symptom control. Our specific aim was to assess the prevalence of sub-optimal PPI dose timing in GERD patients with poor symptom control within a community-based gastroenterology practice.

Materials and methods


Patients referred by their primary care physicians to a 10-person community-based gastroenterology practice for the evaluation of medically refractory GERD symptoms were eligible subjects for this study. Medically refractory GERD symptoms were defined as heartburn and regurgitation symptoms, which were persistent at least weekly, despite PPI therapy for at least 1 month. Patients were excluded from the study if they complained of alarm symptoms like dysphagia, weight loss or manifested symptoms of gastrointestinal bleeding. Potential study subjects were identified by one nursing assistant (DPL), who recruited patients after assessing eligibility in an examination room prior to physician interaction. The study was approved by the St Joseph Mercy Hospital Institutional Review Board.

Conduct of the study

One nursing assistant (DPL) followed a scripted interview in English, which queried subjects about PPI brand used, dose and timing of dose relative to meals. The scripted interview was validated on five patients prior to the start of the study by comparing the results of the scripted interview to an independent physician assessment of the same study variables. In all cases, there was an agreement in all study variables between the nursing assistant and physician. The results of the interview were abstracted to a secondary database. The results of the tabulated data were independently verified by random assessment of 10-data sheets during the course of the study to ensure that there was agreement between the nursing assistant and physician on input of study variables.

PPI dosing categories

The timing of dose was categorized in the following manner: (i) optimal dosers: (a) with or <15 min prior to meals, (b) 15–30 min prior to meals, (c) 30–60 min prior to meals; (ii) sub-optimal dosers: (a) >60 min prior to meals, (b) after meals, (c) as needed (p.r.n, one dose as symptoms occur), (d) at bedtime (q.h.s, one dose prior to sleeping). Based on available evidence and recommendations described in the literature, dosing 15–30 min prior to meals was selected as the interval that would maximize acid suppression.13–16 All PPI dosing behaviour of subjects was abstracted into the referenced categories prior to data analysis.

Data analysis

Frequencies and the associated percentages for each category of dosing characteristics were calculated. This pilot study was performed to determine baseline dosing patterns of patients. Future research will be based on results of these preliminary data. We arbitrarily decided to examine 100 consecutive patients fulfilling the study criteria.


One hundred consecutive patients were consented to participate in the study and no patients refused to participate. All subjects were insured by either private or government insurance. All patients spoke and understood English. The study sample consisted of 44 men and 56 women with mean ages of 50 years (s.d. = 10), and 55 years (s.d. = 14) for men and women, respectively. Of the 100 subjects, 69 were taking omeprazole; 21 lansoprazole; nine rabeprazole; and one pantoprazole.

Optimal dosers

Forty-six of the 100 subjects (46%) were categorized as optimal dosers; however, 26 subjects (57%) took their PPIs with meals or <15 min before meals. Only 12 subjects (26%) dosed at times which produced maximal acid suppression (15–30 min before meals). Eight subjects (17%) dosed between 30 and 60 min before meals.

Sub-optimal dosers

Fifty-four of the 100 subjects (54%) were categorized as sub-optimal dosers. Twenty-one of these (39%) took their PPIs >60 min before meals. Sixteen (30%) and 15 (28%) subjects dosed after meals and at bedtime (q.h.s.), respectively. Only two (4%) dosed p.r.n. (Figure 1).

Figure 1.

 The sub-optimal dosers. A dose timing breakdown for the 54 sub-optimal dosers.

Once daily dosers

Seventy-nine subjects took their PPIs once-daily (QD). Thirty-five (44%) were optimal dosers with 23 (29%) dosing with meals or <15 min before meals, five (6%) 15–30 min before meals (achieving maximal acid suppression), and seven (9%) between 30 and 60 min before meals. The remaining 44 (56%) of the QD patients were sub-optimal dosers with relatively equal percentages dosing >60 min before meals (17, 22%), after meals (13, 17%), and q.h.s. (12, 15%). Only two (3%) dosed p.r.n.

Multiple times a day dosers

Twenty-one subjects took their PPIs two- or three times daily (b.d./t.d.s.). Eleven (52%) were optimal dosers with seven (33%) dosing 15–30 min before meals (achieving maximal acid suppression), three (14%) with meals or <15 min before meals, and one (5%) between 30 and 60 min before meals. The remaining 10 (48%) of the b.d./t.d.s. patients were sub-optimal dosers. Six of them were taking one dose sub-optimally and were categorized based on the sub-optimal dose time. The remaining four took all doses sub-optimally and were categorized as follows: (i) morning with no food/q.h.s. (grouped in q.h.s.) (ii), morning with no food/after evening meal (after meals) (iii), after morning meal/afternoon with no food (after meals), and (iv) 2 h before morning meal/q.h.s. (>60 min before meals). Of the 21 b.d./t.d.s. dosers, the 10 sub-optimal dosers were relatively equally distributed among the >60 min before meals (4, 19%), after meals (3, 14%), and q.h.s. (3, 14%) categories. No b.d./t.d.s. subjects dosed p.r.n.


The prevalence of sub-optimal PPI dose timing in our community-based GERD population was high. Fifty-four percent of our subjects revealed that they took their PPI at times, which did not optimize acid suppression. Only 46% were categorized as optimal dosers and only 12 dosed in a manner, which maximized acid suppression (15–30 min before meals).

Pharmacokinetic mechanisms that have been proposed to explain the blunted symptom control in refractory GERD patients include genetic variation in hepatic PPI metabolism, inter-individual variation in oral PPI bioavailability, and the selective inhibition of only actively-secreting ATPase molecules by PPIs.13 While the former two mechanisms may necessitate higher PPI doses in some patients, sub-optimal PPI dose timing in relation to meals may limit efficacy in others via the latter mechanism. Food ingestion stimulates gastric parietal cell ATPase molecules to secrete acid, allowing PPIs to inhibit acid secretion when they are taken before meals.12–16 It has been hypothesized that prebreakfast dosing maximizes PPI efficacy because ATPase molecules are able to accumulate in parietal cells during a prolonged overnight fast, making more ATPase molecules available for acid-catalysed conversion to their actively secreting form by food ingestion.12 The importance premeal PPI dosing has been explored in a number of intragastric pH probe studies. A study by Kuo and Castell revealed that omeprazole 20 mg b.d. (before breakfast and dinner) provided superior intragastric pH control compared with 40 mg q.d. before breakfast or dinner, despite the same cumulative daily dose of omeprazole in each group.14–15 Comparison of 24-h intragastric pH data in 19 subjects taking omeprazole 40 mg before breakfast, 40 mg before dinner, or 20 mg b.d. before breakfast and dinner revealed that the 20 mg b.d. regimen was superior in controlling intragastric pH.15 Hatlebakk et al.16 compared the gastric acidity at the end of two isolated 7-day periods in 21 patients taking omeprazole 20 mg or lansoprazole 30 mg daily. During one 7-day period, patients took the PPI 15 min before breakfast while during the other 7-day period they took their PPI in the morning with no food or drink (except water) until noon. Nine-hour pH monitoring on day 7 showed that the pH was acidic (pH <4) 17% of the time in the week, where the PPI was dosed before breakfast vs. 42% of the time in the without breakfast week (P < 0.01). The latter study, in conjunction with recommendations in the literature, led us to choose 15–30 min before meals as our optimal PPI dose timing category.12–13, 16 Given that there is no absolute consensus about the perfect time to dose PPIs, we chose to expand our definition of correct dosing to include meal-time to 1 h before meals. Even with this broadened definition of correct dosing, more than half of our subjects were sub-optimal dosers.

The optimal dosers in this study may have been taught the importance of premeal dosing or they may have learned to dose before meals because of better symptom control. The most common error in the sub-optimal dosers was taking the drug over 60 min before a meal (38%), followed by postmeal dosing (30%) and bedtime dosing (28%). It is possible that the high rate of incorrect dosing in our study population simply reflects the tendency of patients to take medications when it is most convenient. It is also possible that some were instructed to dose their PPIs at these times by health care providers who were not aware of the importance of premeal dosing. Patients who took the drug over 1 h before a meal frequently took the pill soon after awakening and did not eat breakfast, with lunch being their first meal of the day. Postmeal and bedtime dosing may be a learned response to postprandial or nocturnal GERD symptoms. It was interesting that there were only two subjects that were taking their PPIs on an as-needed basis, especially given that most over-the-counter reflux medications are marketed as p.r.n. medications. The concept of p.r.n. PPI dosing will likely mature over time as over-the-counter PPIs become widely utilized.

Although this is the first study to examine dosing habits in GERD patients with sub-optimal control, this study had limitations. Firstly, results from this study reflect those patients attending a private gastroenterology practice. It is unclear whether the dosing trends discovered in this study can be generalized to other populations. However, it is important to note that this is a community-based practice tied to a primary care medical center; therefore, we believe the results to be at least applicable to these types of patients. Secondly, definitions of optimal dose timing were based on literature recommendations rather than actual patient-oriented symptom control data as the availability of such data are limited. Thirdly, the actual clinical impact of sub-optimal PPI dose timing on GERD symptom control is unknown. No baseline GERD symptom data were collected to compare symptom severity in the optimal vs. sub-optimal dosers. It is possible, therefore, that similar symptom severity exists in the two groups. Finally, symptom severity was not followed after dose timing education to look for improvement. Further studies to assess refractory GERD symptom response to PPI dose timing optimization are being conducted to investigate optimal dose timing's impact on symptom control. Despite these limitations, this is the first study to examine dosing habits in GERD patients with sub-optimal control and provides evidence that dosing habits among these patients may play a significant role in their poor symptom control.

Education of healthcare professionals and pharmacists would help reduce the prevalence of sub-optimal PPI dosing. Healthcare professionals could educate their patients when the PPI is prescribed, while pharmacists could affix labels on pill bottles advising premeal dosing. If further studies revealed that GERD patients have better symptom control when they dose their PPIs optimally, patient education could lower per-patient PPI costs and improve overall quality of life in the GERD population. As billions of dollars are spent annually in the US on PPI prescriptions, the economic impact of reducing unnecessary PPI dose escalations could be profound. As a result, the evaluation of refractory GERD patients should include an exploration of their PPI dosing habits. If sub-optimal dosing is identified in a patient with poorly controlled GERD symptoms, a trial period of optimal dosing might be initiated before a modification of PPI dose or brand is initiated.


The authors received no grant support for this study.