Proton pump inhibitors (PPIs) are the most potent acid-suppressive drugs available. They are the treatment of choice for gastroesophageal reflux disease (GERD), peptic ulcer disease, Helicobacter pylori eradication and prevention of NSAID-associated mucosal injury [1]. Their efficacy, tolerability and low cost have made them some of the most widely used medicines worldwide. These same features have also encouraged prescribing well beyond the indications studied in clinical trials. For most indications, PPI therapy is meant to be time-limited. The American College of Gastroenterology (ACG) recommends that patients with GERD who respond to an initial course, and who have no erosive esophagitis or Barrett's esophagus, should be offered a trial of discontinuation. Those who need maintenance should receive the lowest effective dose [2]. Long-term therapy is appropriate in a defined group. This includes severe erosive esophagitis, Barrett's esophagus, peptic stricture, objectively confirmed GERD that relapses off therapy, hypersecretory states, and gastroprotection in patients whose bleeding risk remains high [3]. Duration alone therefore does not make a prescription inappropriate. What matters is whether a valid indication persists and whether the dose matches it. There is no single accepted definition of long-term use. A scoping review by Haastrup et al. found thresholds ranging from weeks to years and proposed 8 weeks as a practical point for reassessment [4]. Beyond this point, many prescriptions continue by inertia. Forgacs and Loganayagam drew attention to widespread overprescribing more than 15 years ago [5]. Heidelbaugh et al. showed that initiation without a recognised indication and failure to step down or stop after the indication resolves are the main drivers of overutilisation [6]. The consequences of unnecessary continuation are both clinical and economic. Observational studies have linked long-term PPI use with enteric infections, hypomagnesaemia, vitamin B12 deficiency, fractures and chronic kidney disease. Most of these associations may reflect confounding rather than causation [7]. Withdrawal after prolonged use can also cause rebound acid hypersecretion, which may lead patients and prescribers to restart treatment unnecessarily [8]. Conversely, stopping a PPI in a patient with complicated GERD or high bleeding risk can cause real harm. Decisions should therefore rest on a structured assessment of indication rather than on fear of adverse effects. This has led to formal deprescribing guidance. The 2022 American Gastroenterological Association (AGA) clinical practice update advises that every patient on long-term PPI therapy should have the indication documented and reviewed periodically. Patients without a definite chronic indication should be offered a deprescribing trial, and those on twice-daily therapy should be considered for step-down [9]. The evidence-based guideline by Farrell et al. supports dose reduction, on-demand use or discontinuation in suitable adults who have completed an adequate course and whose symptoms have resolved [10]. Indian studies have reported that 46–57% of hospital PPI prescriptions fall outside recognised indications [11,12]. However, most of these studies assessed short-term inpatient use and classified prescriptions only as appropriate or inappropriate. Few have looked specifically at patients already on long-term therapy, separated poorly documented cases from inappropriate ones, or estimated how many patients could be stepped down or stopped. The present study therefore aimed to determine the appropriateness of long-term PPI use against predefined evidence-based criteria. It also set out to identify patient and treatment factors associated with potentially inappropriate use and to quantify the scope for treatment optimisation.

Design, setting and ethics. A hospital-based, observational, cross-sectional study was conducted over 12 months in the Department of Pharmacology and the participating outpatient and inpatient departments of Amaltas Institute of Medical Sciences, a tertiary care teaching hospital in India. The Institutional Ethics Committee approved the protocol, and all participants gave written informed consent. No PPI was altered for research purposes, and recommendations arising from the review remained the responsibility of the treating physician. Reporting follows the STROBE statement [13]. Participants. Consecutive adults (≥18 years) currently receiving any PPI for more than 8 weeks were enrolled [4]. Exclusions were short-course H. pylori eradication regimens, unverifiable PPI exposure, inability or refusal to consent, and participation in an intervention that determined PPI prescribing. The sample size was calculated as n = Z²pq/d², with p = 50%, Z = 1.96 and d = 5%. This gave 385, which was increased to 425 to allow for 10% incomplete records. Data collection. A structured case record form captured age, sex, clinical setting, diagnoses, PPI agent, dose, frequency, route and duration, and the initial and current indication. It also recorded previous upper gastrointestinal bleeding or ulcer, endoscopy findings, concurrent NSAIDs, aspirin, other antiplatelets, anticoagulants and corticosteroids, and any previous discontinuation attempt. Sources were prescriptions, case records and patient or caregiver interviews. Assessment of appropriateness. Each prescription was assessed against predefined criteria derived from the ACG GERD guideline, the AGA deprescribing update, the review of long-term indications by Kamboj et al., and the evidence-based deprescribing guideline [2,3,9,10]. • Category A – Appropriate: a continuing evidence-based indication at an appropriate intensity. Examples are severe erosive esophagitis, Barrett's esophagus, peptic stricture, objectively proven GERD needing maintenance, hypersecretory states, previous upper GI bleeding, or gastroprotection in patients at high bleeding risk (e.g., age ≥65 years, prior ulcer, or NSAID or antithrombotic therapy combined with corticosteroids or anticoagulants). • Category B – Potentially inappropriate: no continuing indication, for example uncomplicated dyspepsia or resolved symptoms after an adequate course. Also included were duration beyond what the indication requires, or intensity greater than needed (e.g., twice-daily dosing without a qualifying indication). • Category C – Indeterminate: documentation insufficient to judge appropriateness despite reasonable attempts at verification. Documentation was graded as complete, partial but sufficient, or insufficient. For every patient, a single proposed management category was recorded: continue, dose reduction or step-down, on-demand therapy, supervised discontinuation, or specialist review. Two investigators classified each prescription independently, and disagreements were resolved by discussion. Statistical analysis. Data were analysed with SPSS. Categorical data are presented as frequencies and percentages. The prevalence of potentially inappropriate use is reported with a Wald 95% confidence interval (CI), both for all prescriptions and for assessable prescriptions (Categories A + B). Associations between patient or treatment characteristics and appropriateness category were tested with Pearson's chi-square test. The chi-square test for linear trend (Cochran–Armitage) was used for ordered duration groups. Where expected cell counts were below 5, categories were merged (twice-daily and more than twice-daily dosing). Unadjusted odds ratios (OR) with 95% CI were calculated for potentially inappropriate versus appropriate prescriptions. A two-sided p-value <0.05 was considered significant.

All 425 enrolled long-term PPI users were included. Most were middle-aged or older (78.8% aged ≥40 years), 54.1% were male, and 61.2% were outpatients. Pantoprazole was the most commonly used agent (44.7%), and 42.4% had received PPIs for more than one year. Table 1. Appropriateness of long-term PPI prescriptions (N = 425) Appropriateness category n % (95% CI) Category A – Appropriate 180 42.4 (37.7–47.1) Category B – Potentially inappropriate 190 44.7 (40.0–49.4) Category C – Indeterminate 55 12.9 (9.7–16.1) Total 425 100.0 Potentially inappropriate prescriptions formed the largest category (44.7%), slightly exceeding appropriate ones (42.4%). Excluding the 55 indeterminate cases, 190 of 370 assessable prescriptions (51.4%; 95% CI 46.3–56.4) were potentially inappropriate. This means that about one in two long-term users whose records could be judged lacked a continuing evidence-based justification. Table 2. Quality of documentation supporting long-term PPI prescriptions (N = 425) Documentation status n % Appropriateness assessment Complete clinical documentation 300 70.6 Assessable Partial but sufficient documentation 70 16.5 Assessable Insufficient documentation 55 12.9 Indeterminate Total 425 100.0 Complete documentation of indication and treatment history was available for 70.6% of prescriptions. In 12.9%, records were too sparse to judge whether continuation was justified, even after the patient interview. Together with the partially documented group, 29.4% of prescriptions lacked a complete record of why the PPI was being continued. Table 3. Appropriateness according to duration of PPI therapy (N = 425) Duration Appropriate n (%) Potentially inappropriate n (%) Indeterminate n (%) Total Inappropriate among assessable, % OR (95% CI)* >8 weeks–6 months 70 (53.8) 45 (34.6) 15 (11.5) 130 39.1 1.00 (reference) >6–12 months 55 (47.8) 45 (39.1) 15 (13.0) 115 45.0 1.27 (0.74–2.19) >1–2 years 35 (35.0) 55 (55.0) 10 (10.0) 100 61.1 2.44 (1.39–4.30) >2 years 20 (25.0) 45 (56.3) 15 (18.8) 80 69.2 3.50 (1.83–6.68) Total 180 (42.4) 190 (44.7) 55 (12.9) 425 51.4 Potentially inappropriate vs appropriate (indeterminate excluded). Row percentages. χ² (A vs B) = 20.24, df = 3, p <0.001; χ² for linear trend = 19.53, p <0.001. The share of appropriate prescriptions fell steadily with longer treatment, from 53.8% at up to 6 months to 25.0% beyond 2 years. Among assessable prescriptions, potentially inappropriate use rose from 39.1% to 69.2% (p for trend <0.001). Compared with patients treated for up to 6 months, the odds of inappropriate use were 2.4 times higher at 1–2 years and 3.5 times higher beyond 2 years. Taken together, patients treated for more than one year had 2.5 times the odds of patients treated for one year or less (OR 2.53, 95% CI 1.65–3.87). Indeterminate prescriptions were also most frequent beyond 2 years (18.8%), suggesting that the original reason for treatment is lost over time. Table 4. Appropriateness according to dosing frequency (N = 425) Dosing frequency Appropriate n (%) Potentially inappropriate n (%) Indeterminate n (%) Total n (%) Inappropriate among assessable, % Once daily 140 (48.3) 115 (39.7) 35 (12.1) 290 (68.2) 45.1 Twice daily 35 (30.4) 65 (56.5) 15 (13.0) 115 (27.1) 65.0 More than twice daily 5 (25.0) 10 (50.0) 5 (25.0) 20 (4.7) 66.7 Total 180 (42.4) 190 (44.7) 55 (12.9) 425 (100.0) 51.4 Once daily vs twice daily or more (A vs B): χ² = 12.05, df = 1, p <0.001; OR 2.28 (95% CI 1.45–3.60). More than half of twice-daily prescriptions (56.5%) were potentially inappropriate, compared with 39.7% of once-daily prescriptions. Among assessable prescriptions, the odds of inappropriate use were more than twice as high with multiple daily dosing (OR 2.28). This points to step-down to once-daily therapy as a major optimisation target. Table 5. Appropriateness according to demographic and clinical characteristics (N = 425) Characteristic Appropriate n (%) Potentially inappropriate n (%) Indeterminate n (%) Total OR (95% CI)* p-value† Age (years) 0.245 18–39 45 (50.0) 35 (38.9) 10 (11.1) 90 1.00 (reference) 40–59 75 (42.9) 80 (45.7) 20 (11.4) 175 1.37 (0.80–2.36) ≥60 60 (37.5) 75 (46.9) 25 (15.6) 160 1.61 (0.92–2.81) Sex 0.708 Female 85 (43.6) 85 (43.6) 25 (12.8) 195 1.00 (reference) Male 95 (41.3) 105 (45.7) 30 (13.0) 230 1.11 (0.73–1.66) Clinical setting 0.399 Inpatient 75 (45.5) 70 (42.4) 20 (12.1) 165 1.00 (reference) Outpatient 105 (40.4) 120 (46.2) 35 (13.5) 260 1.22 (0.81–1.86) Potentially inappropriate vs appropriate. †Chi-square test, Category A vs B. Each characteristic is a separate stratification of the same 425 patients. Potentially inappropriate use was somewhat more frequent with increasing age (38.9% at 18–39 years vs 46.9% at ≥60 years), in men, and in outpatients. None of these differences reached statistical significance. Inappropriate prescribing was therefore spread across all patient groups and was linked to treatment characteristics rather than patient demographics. Table 6. Proposed optimisation of long-term PPI therapy (N = 425) Proposed management n % Continue current treatment 210 49.4 Dose reduction / step-down 90 21.2 Supervised discontinuation 65 15.3 On-demand therapy 45 10.6 Further specialist review 15 3.5 Total 425 100.0 One category per patient. These were review recommendations; no treatment was changed during the study. About half of patients (49.4%) were judged suitable to continue their current regimen. The other 200 patients (47.1%) were candidates for some form of optimisation: dose reduction or step-down (21.2%), supervised discontinuation (15.3%) or on-demand use (10.6%). A further 3.5% were referred for specialist review before any change could be recommended.

In this cohort of long-term PPI users, 44.7% of all prescriptions and 51.4% of assessable prescriptions were potentially inappropriate. A further 12.9% could not be judged because of poor documentation. Inappropriate use rose steeply with duration and was more common with twice-daily dosing, whereas age, sex and clinical setting were not significant factors. Almost half of patients could be considered for step-down, on-demand use or supervised withdrawal. Our estimate falls within the range reported from Indian hospitals. Juneja et al. found that 54% of inpatient PPI prescriptions in North India were appropriate, implying that about 46% were not [11]. Venkataraman et al. reported that 57% of inpatients received PPIs outside approved indications [12]. In Danish primary care, Reimer and Bytzer verified a valid indication in only 27% of long-term users [14]. Differences in criteria and setting limit direct comparison. Even so, the consistent finding across health systems is that around half of PPI use lacks a clear continuing justification. Our study adds to this by restricting the analysis to established long-term users and by separating undocumented from inappropriate prescriptions. This avoids labelling every poorly recorded case as inappropriate, which would overstate the problem. Duration was the strongest correlate of inappropriate use. Odds rose 3.5-fold beyond two years, and appropriate prescriptions fell from 54% to 25%. This pattern is consistent with the overutilisation pathway described by Heidelbaugh et al., in which a PPI started for a valid short-term reason is continued after that reason has resolved [6]. Haastrup et al. proposed 8 weeks as the point at which treatment should be reassessed [4]. In our data, failure to reassess after that point appeared to be the main driver of inappropriate use. The rise in indeterminate cases beyond two years points the same way: the original indication is often no longer recorded in long-standing prescriptions. Twice-daily or more frequent dosing more than doubled the odds of inappropriate use. Intensified therapy is justified in severe erosive esophagitis, refractory GERD and hypersecretory states [2]. However, the AGA update explicitly recommends considering step-down to once daily for patients on twice-daily therapy, even when they have a valid chronic indication [9]. The 21.2% of patients identified for dose reduction reflects this opportunity. The absence of a significant association with age, sex or setting suggests that inappropriate prescribing is a system-level problem rather than one confined to particular patient groups. Interventions should therefore target prescribing processes, such as mandatory indication fields, automatic review dates and pharmacist-led medication review, rather than specific demographics. Deprescribing must also be selective. Patients with complicated GERD, Barrett's esophagus or high bleeding risk should continue therapy. In the COMPASS trial, pantoprazole did not reduce clinically significant upper GI events in unselected patients on antithrombotic therapy, which underlines that gastroprotection should be guided by risk rather than drug exposure alone [15]. Patients should also be counselled about transient rebound symptoms after withdrawal [8,10]. This study has limitations. It was conducted at a single centre, and its cross-sectional design cannot establish causality. The associations reported are unadjusted, because multivariable regression requires patient-level modelling that was not undertaken here, so duration and dosing may confound each other. Appropriateness criteria involve clinical judgement, and recall may have affected duration estimates. The optimisation categories are recommendations; whether deprescribing was achieved and sustained was not assessed. Strengths include an a priori sample size, inclusion of both outpatient and inpatient settings, explicit guideline-based criteria, and transparent reporting of indeterminate cases.

About half of assessable long-term PPI prescriptions in this tertiary care hospital lacked a continuing evidence-based indication. A further one in eight could not be judged because of poor documentation. The likelihood of inappropriate use increased sharply with treatment duration beyond one year and with twice-daily dosing, but did not differ by age, sex or clinical setting. Nearly half of long-term users were candidates for dose reduction, on-demand therapy or supervised discontinuation. Recording the indication and a review date on every PPI prescription, structured reassessment after 8 weeks, and a pharmacist-supported PPI stewardship programme could safely reduce unnecessary exposure while protecting patients who need continued therapy. A prospective study of the outcomes of deprescribing is warranted.

1. Strand DS, Kim D, Peura DA. 25 years of proton pump inhibitors: a comprehensive review. Gut Liver. 2017;11(1):27-37. doi:10.5009/gnl15502 2. Katz PO, Dunbar KB, Schnoll-Sussman FH, Greer KB, Yadlapati R, Spechler SJ. ACG clinical guideline for the diagnosis and management of gastroesophageal reflux disease. Am J Gastroenterol. 2022;117(1):27-56. doi:10.14309/ajg.0000000000001538 3. Kamboj AK, Patel DA, Yadlapati R. Long-term proton pump inhibitor use: review of indications and special considerations. Clin Gastroenterol Hepatol. 2024;22(7):1373-6. doi:10.1016/j.cgh.2024.01.042 4. Haastrup PF, Thompson W, S√∏ndergaard J, Jarb√∏l DE. When does proton pump inhibitor treatment become long term? A scoping review. BMJ Open Gastroenterol. 2021;8(1):e000563. doi:10.1136/bmjgast-2020-000563 5. Forgacs I, Loganayagam A. Overprescribing proton pump inhibitors. BMJ. 2008;336(7634):2-3. doi:10.1136/bmj.39406.449456.BE 6. Heidelbaugh JJ, Kim AH, Chang R, Walker PC. Overutilization of proton-pump inhibitors: what the clinician needs to know. Therap Adv Gastroenterol. 2012;5(4):219-32. doi:10.1177/1756283X12437358 7. Haastrup PF, Thompson W, S√∏ndergaard J, Jarb√∏l DE. Side effects of long-term proton pump inhibitor use: a review. Basic Clin Pharmacol Toxicol. 2018;123(2):114-21. doi:10.1111/bcpt.13023 8. Reimer C, S√∏ndergaard B, Hilsted L, Bytzer P. Proton-pump inhibitor therapy induces acid-related symptoms in healthy volunteers after withdrawal of therapy. Gastroenterology. 2009;137(1):80-7.e1. doi:10.1053/j.gastro.2009.03.058 9. Targownik LE, Fisher DA, Saini SD. AGA clinical practice update on de-prescribing of proton pump inhibitors: expert review. Gastroenterology. 2022;162(4):1334-42. doi:10.1053/j.gastro.2021.12.247 10. Farrell B, Pottie K, Thompson W, Boghossian T, Pizzola L, Rashid FJ, et al. Deprescribing proton pump inhibitors: evidence-based clinical practice guideline. Can Fam Physician. 2017;63(5):354-64. 11. Juneja S, Rana P, Manoj SV, Kalia R, Singh RP. Appropriateness of proton pump inhibitor use in hospitalized patients: a cross-sectional study in a tertiary care hospital in North India. J Pharm Care. 2023;11(2). doi:10.18502/jpc.v11i2.13360 12. Venkataraman R, Rashid M, Shrestha H. Inappropriate medication use and cost comparison analysis of proton pump inhibitors: evidence from an Indian tertiary care facility. Curr Drug Saf. 2020;15(2):147-55. doi:10.2174/1574886315666200311120151 13. von Elm E, Altman DG, Egger M, Pocock SJ, G√∏tzsche PC, Vandenbroucke JP. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet. 2007;370(9596):1453-7. doi:10.1016/S0140-6736(07)61602-X 14. Reimer C, Bytzer P. Clinical trial: long-term use of proton pump inhibitors in primary care patients - a cross sectional analysis of 901 patients. Aliment Pharmacol Ther. 2009;30(7):725-32. doi:10.1111/j.1365-2036.2009.04092.x 15. Moayyedi P, Eikelboom JW, Bosch J, Connolly SJ, Dyal L, Shestakovska O, et al. Pantoprazole to prevent gastroduodenal events in patients receiving rivaroxaban and/or aspirin in a randomized, double-blind, placebo-controlled trial. Gastroenterology. 2019;157(2):403-12.e5. doi:10.1053/j.gastro.2019.04.041