Trends, utilisation and appropriateness of proton pump inhibitor prescriptions among recent initiators: a decade-long analysis of patients with a 1-year washout period (2013–2021)

Abstract

Objective

To describe temporal trends in initiation, dose intensity, utilisation and potentially inappropriate prescribing of proton pump inhibitors (PPIs) among recent initiators in an urban Chinese outpatient setting.

Design

Retrospective, annual identification of recent PPI initiators using a 1-year washout definition (repeated cross-sectional approach).

Setting

Outpatient prescriptions recorded in the hospital information system of Jinshan Hospital Affiliated to Fudan University, Shanghai, China, 2013–2021.

Participants

Adults aged ≥18 years who initiated PPI and had no PPI prescription in the preceding year.

Outcome measures

Primary and secondary outcome measures annual initiation rates (overall and stratified by sex, age group, dose intensity and indication), utilisation metrics (defined daily dose/1000 inhabitants/day, defined daily cost, drug utilisation index) and components of potentially inappropriate prescribing (indication-related, prophylaxis-related and incomplete diagnosis).

Results

Among 114 241 recent initiators (mean age 48.47 years; 51.43% female), initiation rates increased from 1.79% (95% CI 1.76% to 1.83%) in 2013 to 2.79% (2.74% to 2.83%) in 2021 (p<0.001), with the steepest rise in adults aged 18–34 years. High-dose initiation peaked in 2019 and then declined, while standard-dose initiation increased after 2019. Across 2013–2021, potentially inappropriate indication-related prescribing decreased (30.94% in 2013 to 10.36% in 2021), whereas prophylaxis-related prescribing remained common.

Conclusions

PPI initiation increased substantially over the past decade, particularly among younger adults. These patterns raise concerns about potential overuse and suboptimal prescribing. The results underscore the need for targeted prescription review and deprescribing-oriented stewardship, with particular attention to young adults, prophylactic indications and dose optimisation.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• Use of a large, hospital information system-based dataset with a 1-year washout definition to identify recent proton pump inhibitor initiators.

• Standardised drug utilisation metrics (defined daily dose/1000 inhabitants/day, defined daily cost, drug utilisation index) were applied using WHO definitions to enhance comparability across years.

• Appropriateness was assessed against explicit, guideline-based criteria covering indication, duration and prophylactic use.

• Records with incomplete key identifiers were excluded to reduce misclassification risk, but residual coding errors may remain.

• Single-centre outpatient data may limit generalisability and do not capture over-the-counter use or prescriptions outside the hospital catchment.

Introduction

Proton pump inhibitors (PPIs) represent a cornerstone in the management of acid-related gastrointestinal (GI) disorders.^{1 2} By inhibiting the H-K ATPase enzyme, PPIs effectively suppress gastric acid secretion, thereby providing therapeutic benefits in conditions such as Helicobacter pylori infection, peptic ulcer disease (PUD), GI bleeding, gastro-oesophageal reflux disease (GERD), and in the prevention of gastric ulcers induced by non-steroidal anti-inflammatory drugs (NSAIDs). Since their introduction in the late 1980s, PPIs—including omeprazole, esomeprazole, lansoprazole, rabeprazole and pantoprazole—have seen a dramatic increase in use, becoming among the most frequently prescribed medications globally.^{3 4} This burgeoning utilisation is highlighted by data from various countries; for instance, PPI usage in adult Americans surged from 3.9% in 1999 to 7.8% in 2012,⁵ while the UK saw prescriptions rise from 11.126 million in 2000 to 43.127 million in 2011.^{6 7} Moreover, the economic impact of PPIs is profound, with global spending exceeding US$25 billion in 2007–2008 and sales of esomeprazole alone topping US$5 billion in 2015.⁴

Recent literature has increasingly scrutinised the appropriateness of PPI prescriptions, noting a tendency towards both overuse and underuse in various settings.^{8 9} Concerns have been particularly pronounced regarding the duration of therapy and dose escalation without clear clinical indications.^{10 11} Furthermore, emerging evidence suggests potential adverse effects associated with long-term PPI use, including renal complications, nutritional deficiencies and increased susceptibility to infections.⁴ This underscores the necessity for healthcare providers to balance the undeniable benefits of PPIs with their risks, making the optimisation of their use a pressing clinical and public health priority.

In this study, we employed a retrospective cohort analysis of 114,241 adults who initiated PPI therapy between 2013 and 2021. By dissecting the trends across different age groups, genders and diagnostic categories, this study illuminated the dynamics, appropriateness and PPI utilisation. We critically examined the evolution of prescribing practices in response to shifting clinical guidelines and patient demographics. Additionally, the economic aspects of PPI use, including cost-effectiveness and the impact of price variations on prescription trends, were analysed to provide insights into the broader implications of PPI therapy. Through this comprehensive decade-long analysis, the study aims to contribute to the optimisation of PPI prescribing practices, ensuring they align more closely with clinical guidelines and patient-specific therapeutic needs.

Methods

Study design and setting

This study employed a retrospective design, using longitudinal monitoring data to investigate the prescription of PPIs among outpatients at Jinshan Hospital, affiliated with Fudan University. Jinshan Hospital represents the sole third-tier healthcare facility within its catchment area, offering a unique dataset for this analysis.

Population and data source

Our study comprised individuals aged 18 years and above in outpatient service, who were identified as recent initiators of PPIs based on the Hospital Information System records. Recent initiators of PPIs refer to an individual who has initiated treatment with PPIs and has not had a prescription for PPIs for the prior 1 year to the start of the current therapy. Exclusion criteria included patients with incomplete records, such as a missing medical card number, or unknown date of birth or gender. Additionally, Ilaprazole was excluded from the analysis due to its limited outpatient use, being dispensed for the first time in 2020.

Prescription data

Prescription information extracted included the medical card number, patient’s gender and age, diagnosis, the generic name of the prescribed PPIs and the dosage. The PPIs considered in the study encompassed omeprazole (10 mg, 20 mg), pantoprazole (40 mg), rabeprazole (20 mg), lansoprazole (15 mg, 30 mg) and esomeprazole (20 mg).

The study categorised the participants into five distinct age brackets: 18–34, 35–59, 60–64, 65–79 and 80 years and older, following the demographic categorisations outlined in the Statistical Yearbook of Jinshan District. Demographic data from 2013 to 2021, including gender and age composition of Jinshan District, were sourced from its Statistical Yearbook. The International Classification of Diseases, ninth Revision, Clinical Modification was used for coding hospital diagnoses and procedures, ensuring standardisation and comparability of the data analysed.

Estimation of PPI initiation rate

The initiation rate of PPIs was determined by calculating the ratio of the annual number of new patients prescribed any PPI (numerator) to the total population within the hospital’s catchment area for each calendar year (denominator). In addition to overall rates, the initiation rates were further analysed by stratifying the data according to gender, age group, dose intensity of PPIs and the clinical indications for PPI use. This stratification enabled a nuanced understanding of PPI initiation patterns across different patient demographics and clinical needs.

For analytical clarity and consistency, the study delineated ‘standard’ and ‘high’ doses for the PPIs under consideration. The standard daily doses were established as follows: pantoprazole 40 mg, omeprazole 20 mg, esomeprazole 20 mg, lansoprazole 30 mg and rabeprazole 20 mg. High doses were classified as dosages exceeding these thresholds, specifically: pantoprazole >40 mg/day, omeprazole >20 mg/day, esomeprazole >20 mg/day, lansoprazole >30 mg/day and rabeprazole >20 mg/day.

Association between higher dose and indication for the use of PPI

To examine whether dose intensity varied by clinical indication, we conducted an indication-stratified descriptive analysis and multivariable regression. First, we summarised the proportion of high-dose initiation across indication categories using cross-tabulations (numbers and percentages). We then fitted multivariable logistic regression models with high-dose initiation (yes/no) as the dependent variable and indication category as the primary exposure. Age group and sex were included as covariates a priori, and calendar year (2013–2021) was additionally adjusted for to account for temporal changes in prescribing practice. Results are reported as adjusted ORs (aORs) with 95% CIs. All analyses were performed using complete-case records for the variables included in the model.

Mean prescribed daily doses and duration changes

In order to offer insights into trends in clinical management in the prescribing practices of PPIs from 2013 to 2021. The mean prescribed daily doses of the most commonly used PPIs were calculated to evaluate the patterns of PPI dosing over the study period. Treatment duration was categorised into four distinct intervals for analytical purposes: less than 30 days, 30–60 days, 60–90 days and more than 90 days. Furthermore, an examination of the differences in treatment duration between genders was conducted to identify any disparities in PPIs prescribing practices. This gender-based analysis contributed to a more comprehensive understanding of how PPI treatment durations might vary between male and female patients, potentially reflecting differing clinical approaches or patient characteristics.

Appropriateness of PPI prescriptions

The appropriateness of PPI prescriptions was assessed using the criteria outlined in the ‘Guiding Principles for Clinical Application of Proton Pump Inhibitors (2020 Edition)’ and ‘the Expert Consensus on Optimal Application of PPIs’ and other expert consensus of some provinces.¹²,¹⁷ This evaluation included the following components:

1. Approved indications: A prescription was considered appropriate if it adhered to approved indications, which include PUD, GERD, GI bleeding, H. pylori infection, Zollinger-Ellison syndrome and functional dyspepsia. Other conditions, such as chronic non-erosive gastritis, were not considered valid indications for PPI use.

2. Duration of therapy: The recommended duration of PPI therapy was 4–8 weeks for PUD, 4–8 weeks for GERD, 6–8 weeks for GI bleeding or 2 weeks for H. pylori infection.

3. Prophylactic use: The prophylactic use of PPIs was deemed appropriate only under specific conditions such as NSAIDs with aspirin, anticoagulants and steroids. Additionally, rabeprazole, lansoprazole and pantoprazole were not recommended for the prevention of acid-related diseases unless clearly indicated.

PPI utilisation

Next, we meticulously analysed the utilisation of PPIs employing three key metrics: the defined daily dose (DDD), the defined daily cost (DDC) and the drug utilisation index (DUI). The DDD for each PPI, as stipulated by the WHO’s Collaborating Centre for Drug Statistics Methodology,¹⁸ serves as the benchmark for the assumed average maintenance dose per day for a drug when used for its primary indication in adults. Utilisation rates were articulated in terms of DDD per 1000 inhabitants per day. The DDC was derived by dividing the total sales of PPIs (in Chinese Yuan, CNY) by the DDD value, thereby quantifying the average daily cost associated with PPI usage. The DUI, defined as the ratio of the DDD value to the actual days of PPI use, serves as a critical indicator of dosing appropriateness.

Patient and public involvement

Patients and the public were not involved in the design, conduct, reporting or dissemination plans of this retrospective study using deidentified archival data. We will disseminate the findings through peer-reviewed publication and feedback to hospital pharmacy and prescribing teams to inform local quality improvement.

Statistical analysis

Statistical analyses were executed using SAS software (V.9.2). Patient characteristics were detailed, with categorical data shown as frequencies and proportions, and continuous variables as means±SD. Records missing key variables required for classification were excluded. Precision was assessed using 95% CIs for PPI prescription prevalence and initiation rates. Trends in PPI prescriptions were examined with the Cochran-Armitage test, and differences across demographic groups were evaluated using Pearson’s χ² test and analysis of variance or Student’s t-test, as appropriate. A significance level of p<0.05 was set for all tests.

Results

Demographic characteristics of recent PPI initiators

Between 2013 and 2021, a cohort of 114, 241 adults aged 18 years or older was identified as recent PPI initiators, with a slight female predominance (51.43%) and an average age of 48.47 years (table 1). The majority of these recent initiators (51.93%) were in the 35 to 59 age group, followed by 22.34% in the 18–34 age group. Notably, individuals over 80 years of age comprised only 2.98% of the recent initiators. In terms of the specific PPIs dispensed at initiation, rabeprazole was the most frequently prescribed (47.36%), followed by omeprazole (28.76%) and pantoprazole (12.60%).

Table 1. Characteristics of recent PPI initiators (n=114 241) by a decade-long analysis from 2013 to 2021.

Characteristics	Value
Sex
Female	58 754 (51.43)
Age
Mean (SD)	48.47±15.65
18–34 years	26 230 (22.96)
35–59 years	59 325 (51.93)
60–64 years	9482 (8.30)
65–79 years	15 800 (13.83)
≥80 years	3404 (2.98)
PPIs at treatment initiation
Rabeprazole	54 105 (47.36)
Omeprazole	32 856 (28.76)
Pantoprazole	14 394 (12.60)
Dose intensity
High dose	57 303 (50.16)
Standard dose	46 051 (40.31)
Low doses	10 887 (9.53%)
Diagnosis
Helicobacter pylori infection	29 577 (25.89)
Gastrointestinal bleeding	685 (0.60)
GERD	5247 (4.59)
PUD	7894 (6.91)
Gastritis	19 613 (17.17)
Abdominal pain	3695 (3.23)
Anxiety/depression	1381 (1.21)
Kidney disease	982 (0.86)
Cerebrovascular disease	1577 (1.38)
Hypertension	2319 (2.03)
Diabetes	925 (0.81)
Coronary heart disease	982 (0.86)
PPI formulations
Oral	113 133 (99.03)
Injective	1108 (0.97)

GERD, gastro-oesophageal reflux disease; PPI, proton pump inhibitor; PUD, peptic ulcer disease.

The primary approved indications for PPI prescription included H. pylori infection (25.89%), PUD (6.91%), GERD (4.59%) and GI bleeding (0.60%). Additionally, a significant portion of the recent initiators was diagnosed with gastritis-related conditions (17.17%), while abdominal pain of unclear aetiology and anxiety were noted in 3.23% and 1.21% of the cases, respectively. Oral dosage formulations were overwhelmingly preferred, accounting for 99.03% of the prescriptions.

Trends in PPI initiation rate

The study’s analysis of PPI initiation rates from 2013 to 2021 revealed a significant upward trend. The annualised measure of PPI initiation rates increased by 55.87%, from 1.79% (95% CI 1.76% to 1.83%) in 2013 to 2.79% (95% CI 2.74% to 2.83%) in 2021 (p<0.001). A notable finding was the higher initiation rate among females compared with males throughout the study period (figure 1A) (p<0.001).

Figure 1. The recent PPI initiators from the hospital information system of Jinshan Hospital, Shanghai (2013–2021). (A) The initiation rates for recent PPI initiators categorised by gender; (B) The initiation rates for recent PPI initiators across various age groups; (C) The initiation rates for recent PPI initiators according to the clinical indication for PPI use; (D) The initiation rates for recent PPI initiators based on the prescribed intensity of PPI therapy. The error bars indicate the 95% CIs. GERD, gastro-oesophageal reflux disease; GI, gastrointestinal; HP, Helicobacter pylori; PPI, proton pump inhibitor; PUD, peptic ulcer disease.

Age-stratified analysis revealed significant variations in initiation rates across different age groups. Except for the year 2020, all age groups exhibited a growing trend in PPI initiation, particularly notable in the 18–34 and 35–59 year age groups (figure 1B). The 18–34-year age group experienced the most rapid increase, from 1.79% (95% CI 1.71% to 1.87%) in 2013 to 4.43% (95% CI 4.28% to 4.58%) in 2021 (p<0.001). The group aged over 80 years consistently had the lowest initiation rates throughout the study duration. Moreover, the initiation rate for H. pylori infection saw a nearly fivefold increase, from 0.20% (95% CI 0.19% to 0.21%) in 2013 to 1.03% (95% CI 1.00% to 1.05%) in 2021 (p<0.001) (figure 1C).

The analysis also highlighted trends in the prescribed doses of PPIs. The initiation rate for high dose nearly tripled from 2013, reaching a peak in 2019 at 1.99% (95% CI 1.95% to 2.03%), before significantly declining through to 2021 (figure 1D). Conversely, the initiation rate for standard dose PPIs increased from 2019 to 2021, with a statistically significant difference observed between the initiation rates of high dose and standard dose PPIs.

Association between higher dose and indication for the use of PPI

High-dose initiation differed by indication category. In multivariable logistic regression adjusting for age group, sex and calendar year, high-dose initiation was more likely for H. pylori–related indications (aOR 3.50, 95% CI 3.42 to 3.59), PUD (1.63, 95% CI 1.57 to 1.69), oesophageal indications (1.37, 95% CI 1.31 to 1.43) and GI bleeding (1.48, 95% CI 1.28 to 1.72) compared with other indications. Older age groups were associated with lower odds of high-dose initiation, and males had slightly higher odds than females.

Mean prescribed daily doses and duration changes

The investigation into the prescribed daily dosages of Pantoprazole and Rabeprazole from 2013 to 2019 reveals a notable upward trajectory, followed by a gradual decline. This trend was found to be statistically significant. Particularly striking was the increase in the mean prescribed daily dose of pantoprazole in 2019, which surged by 150% in comparison to 2013, as detailed in table 2. Conversely, the dosing of omeprazole exhibited significant fluctuations throughout the period from 2013 to 2021.

Table 2. Mean prescribed daily doses of PPI prescriptions between 2013 and 2021.

Years	Omeprazole	P for trend	Pantoprazole	P for trend	Rabeprazole	P for trend
2013	30.45±11.22	<0.001	58.09±21.00	<0.001	30.19±10.09	<0.001
2014	28.81±11.52		58.35±20.46		31.50±11.05
2015	27.84±13.07		58.08±19.91		30.63±10.49
2016	27.62±11.22		58.79±20.34		31.48±10.04
2017	28.16±11.52		63.62±19.96		31.75±9.91
2018	26.73±11.19		72.53±15.59		33.85±9.23
2019	26.61±11.79		73.23±15.00		34.48±8.94
2020	27.29±11.34		57.52±19.83		30.37±10.18
2021	29.22±9.97		50.84±17.79		29.70±10.00

PPI, proton pump inhibitor.

A gender-based analysis of the recent initiators of PPIs indicated a disparity in the duration of use, with male users exhibiting a longer usage period than their female counterparts. Specifically, the average usage duration in 2021 was recorded at 23.99±23.19 days for females and 25.39±25.04 days for males. Notably, the years 2017 and 2018 were exceptions, where no statistical difference was observed in the average duration of PPI use between the genders (see online supplemental table S1).

Furthermore, an analysis of the duration of PPI usage among the recent initiators revealed that up to 80% used these medications for less than 30 days within the observed years, from 2013 to 2021. This duration exhibited fluctuations across the years, with these variations proving to be statistically significant. It was also observed that the proportion of recent initiators engaging in PPI use for more than 90 days was relatively low, with a peak at 3.10% in 2020, while remaining around 2.0% in other years, as illustrated in figure 2.

Figure 2. Duration changes of PPIs between 2013 and 2021 among the recent initiators. PPIs, proton pump inhibitors.

Evaluating appropriateness of PPI prescriptions

From 2013 to 2021, a total of 194 815 new user prescriptions for PPIs were analysed, revealing that approximately 50% were deemed potentially inappropriate. Among these, 36 878 prescriptions (18.93%) were issued for non-erosive gastritis, chronic gastritis and gastritis, which are not suitable indications for PPI use. Additionally, 5844 prescriptions (3.00%) were for symptoms such as abdominal pain, nausea and vomiting, which are also potentially inappropriate indications (see online supplemental table S2).

Regarding prophylactic use, 63 782 prescriptions (32.74%) included potentially inappropriate prophylactic use (rabeprazole, pantoprazole or lansoprazole, which are not recommended for the prevention of acid-related diseases, or no prophylactic use indications under the evaluated guidelines).

A comparative analysis between 2013 and 2021 showed a significant decrease in the rate of potentially inappropriate indications, from 30.94% in 2013 to 10.36% in 2021(p<0.001). Similarly, the rate of potentially inappropriate prophylactic use declined from 34.30% in 2013 to 27.16% in 2021(p<0.001), indicating a noteworthy trend towards more appropriate PPI prescription practices over time.

PPI utilisation

The overall utilisation of PPIs exhibited a notable increase, from 74.05 DDD/1000/day in 2015 to 100.88 DDD/1000/day in 2020, before witnessing a decline to 93.09 DDD/1000/day in 2021, as depicted in figure 3A. Analysis by gender revealed that male recent initiators had a higher DDD/1000/day than their female counterparts throughout the study period. Age-wise, PPIs utilisation showed fluctuations from 2013 to 2015 but increased steadily from 2015 to 2019, followed by a significant decrease in 2021, as shown in figure 3B. The over 60 years age group consistently demonstrated higher utilisation rates, whereas the 18–34-year age group had the lowest rates across all age groups during the entire study duration (figure 3B).

Figure 3. PPIs utilisation in recent PPI initiators from 2013 to 2021. (A) DDD/1000/day of all PPIs. (B) DDD/1000/day stratified by groups. (C) DUI of overall PPIs. (D) DUI of stratified by different PPIs. (E) DDC of overall PPIs. (F) DDC of stratified by different PPIs. DDC, defined daily cost; DDD, defined daily dose; DUI, drug utilisation index; PPIs, proton pump inhibitors.

The DUI values observed an upward trajectory from 1.11 in 2013 to 1.45 in 2019, subsequently dropping to 1.39 in 2021, indicating fluctuations in the appropriateness of dosing over time (figure 3C). Stratification by specific PPIs revealed that rabeprazole and pantoprazole exhibited increases in DUI values from 2013 to 2019, with a noticeable decline between 2020 and 2021, and values exceeding 1.50 between 2017 and 2019, suggesting periods of higher than recommended dosing (figure 3D).

The study also assessed the economic aspect of PPI use, identifying trends in price reductions over time. Significant price reductions were observed, with oral omeprazole experiencing the most substantial decrease at 96.49%, followed by oral pantoprazole at 40.94%. Among the various PPIs, oral rabeprazole had the highest average DDC value, highlighting differences in cost-effectiveness among the PPI options (figure 3E,F).

Discussion

This comprehensive study leverages extensive historical data from our hospital’s records to provide an in-depth analysis of PPI utilisation among new patients. It stands out as a significant contribution to the existing literature on PPI use, highlighting trends in initiation rates, dose intensity, duration and overall utilisation from 2013 to 2021. Our findings reveal a general increase in PPI initiation rates up to 2019, followed by a decrease in 2020. Notably, the 18–34 age group exhibited the highest and fastest-growing initiation rate, suggesting a demographic shift towards younger adults in the prevalence of acid-related illnesses. This shift underscores emerging socioeconomic challenges that warrant governmental attention, especially in Shanghai’s unique context of urbanisation and lifestyle changes.

Our analysis demonstrates a notable global increase in PPI utilisation, aligning with the findings of Rosenberg et al, which highlighted annual initiation rates of PPI treatment fluctuating between 2.4% and 3.1% of the adult population from 2000 to 2015.¹⁹ This trend of escalating PPI use is corroborated by international reports from Australia, Korea, France, the UK and the USA, all indicating a rise in PPI prescriptions.²²⁰,²² A striking example from the USA shows PPI usage doubling from 3.9% in 1999 to 7.8% in 2012,^{5 21} underscoring the widespread and growing reliance on these medications across healthcare systems. However, the year 2020 marked a temporary decline in PPI initiation rates, likely due to COVID-19-related public health measures, including school closures, bans on large gatherings and recommendations for social distancing and staying at home, imposed by local governments.

Particularly noteworthy is the surge in PPI initiation among the 18–34-year age group, which rose from an initiation rate of 1.79% (95% CI 1.71% to 1.87%) in 2013 to 4.06% (95% CI 3.92% to 4.20%) in 2019, as evidenced by a significant test for trend in proportions (p<0.001). This age group also accounted for the highest initiation rate between 2015 and 2019, while individuals over the age of 80 consistently exhibited the lowest initiation rates throughout the study period. Additionally, the proportion of recent PPI initiators aged between 18 and 34 years was 22.55%, a figure that surpasses those reported in other countries. For comparison, Rotman and Bishop found that only 3.0% of American PPI outpatient study participants were aged between 18 and 29 in the years 2002–2009,²³ and in the UK, only 8.4% of recent PPI initiators were aged 18–30 years from 1990 to 2014.²⁴ This pattern may be driven by multiple factors, such as younger populations often facing lifestyle-related risks such as occupational stress, irregular diet and obesity. At the same time, increased health awareness and the growing availability of gastroscopy and H. pylori testing have contributed to the prescription of PPIs for non-specific symptoms like dyspepsia. This observation carries important clinical implications, for example, initiating PPI therapy in this age group should be more cautious, with clear indications required. For patients already on treatment, regular evaluation should be conducted to guide dose reduction or discontinuation when appropriate, avoiding unnecessary long-term use at higher doses.

The study also identified a substantial increase in the initiation rate of high-dose PPIs, nearly doubling from 2013 to 2019, contrasting with the standard dose’s more stable usage patterns. Our findings indicate an increasing reliance on high-dose PPI prescriptions, mirroring a trend observed in Manitoba, Canada, where the prevalence of high-intensity PPI use escalated from 9.7% in 1997 to 14.2% in 2004.²⁴ Reports suggest that high-dose PPI usage could reach up to 30% in specific populations,²⁵ despite the minimal benefits associated with such dosages for the majority of indications. This trend towards high-dose PPI utilisation, despite the lack of evidence supporting its superiority over standard doses for preventing GI complications, raises concerns about potential overuse and associated risks, such as hypomagnesaemia and osteoporotic fractures.²⁶,²⁹ Typically, high-dose PPIs are reserved for cases where standard doses fail to manage conditions like persistent GERD. While some clinicians advocate for high doses to potentially avert PUD or other GI complications, there is no conclusive evidence that high doses offer superior protection against PUD in patients at risk for GI complications, including those using NSAIDs, antiplatelet agents, anticoagulants or with a history of gastrointestinal bleeding.

Our analysis demonstrates a significant decrease in the DDC, underscoring the effectiveness of healthcare policies like the zero-markup drug policy in reducing patients’ financial burdens. Despite this progress, high DDC values for rabeprazole point to the need for more cost-effective prescribing practices.³⁰ In contrast, omeprazole’s low DDC and initial high utilisation from 2013 to 2015 suggest a price advantage, although its use declined subsequently. Zhang et al analysed the optimal application of different PPIs and pointed out that oral omeprazole possesses the most widely indicated use.¹² However, drug selection is often shaped by local formulary policies, pricing, reimbursement, availability and perceived tolerability. The utilisation was much higher in over 60-year age groups in our study. This indicated that the frequency of PPIs used by the elderly was high. The DUI data showed that oral lansoprazole, rabeprazole and pantoprazole were more than 1.50 in 2017–2019. It indicated that these PPIs were used in excessive doses in patients.

Despite numerous guidelines and studies emphasising the importance of PPIs use, the prevalence of potentially inappropriate PPIs prescriptions remains high in clinical practice. Giannini et al³¹ reported that approximately 40% of outpatients received PPIs potentially inappropriately, while Thomas et al³² found an even higher rate of 68.8% among patients. Similarly, a study in US nursing facilities revealed that approximately one-quarter of residents prescribed PPIs lacked a valid medical reason.³³ Our study corroborates these findings, with 50% of prescriptions deemed potentially inappropriate. Specifically, 18.93% of PPIs prescriptions were for non-erosive gastritis, an indication not recommended by guidelines. This underscores the necessity of adhering strictly to guideline-recommended indications for PPIs use. However, it is important to note that functional dyspepsia, a frequent indication for PPI use, accounted for only 0.05% of prescriptions in this study. This discrepancy likely arises because many prescriptions for dyspeptic symptoms that would meet the criteria for functional dyspepsia were coded under non-specific symptom codes or alternative diagnostic labels like gastritis. This highlights the need for caution when interpreting indication data from large datasets, as coding practices often prioritise symptoms or definitive organic diagnoses over functional syndrome labels. Improving the accuracy of diagnostic coding and adhering to guideline-recommended indications could further enhance the rational use of PPIs.

Another significant proportion of potentially inappropriate use involves the prophylactic use of PPIs. Guidelines indicate that rabeprazole, lansoprazole and pantoprazole should not be used for the prevention of NSAID-related ulcers. Other PPIs can be used for prophylactic NSAID-related ulcers unless specific conditions are met, such as the concurrent use of NSAIDs with aspirin or steroids. The overuse of PPIs likely arises from prescription errors, highlighting the need for targeted education for healthcare providers regarding correct PPI indications to prevent adverse reactions and unnecessary treatments.

To address this critical issue, our hospital has implemented several initiatives, including deploying pre-review software for prescribing guidance, conducting community-based health education and providing regular training for health workers to scrutinise prescriptions. These interventions have resulted in a noticeable decline in potentially inappropriate PPI use from 2013 to 2021. This trend indicates that heightened awareness and rigorous adherence to guidelines have led to more rational PPI use. However, there remains potential for further optimisation. Continued efforts to educate healthcare providers and implement stringent prescription review processes are essential to minimise potentially inappropriate PPI use and enhance patient care.

While this study provides valuable insights into PPI utilisation patterns within a third-tier hospital setting, it acknowledges limitations, including the exclusion of community hospital data and the inability to capture over-the-counter PPI use post-initial hospital visit. second, this is a very large study (>100 k patients) enrolled over 9 years. However, this was conducted in a single centre. And this still represents a limitation of the study that should be acknowledged. Finally, the observational nature of the study suggests a need for randomised clinical trials to establish causality. Future research should adopt a longitudinal cohort design, employing more rigorous standards for initiation and follow-up to deepen our understanding of PPI therapy’s efficacy and safety.

Conclusions

PPI initiation increased markedly from 2013 to 2021, particularly among younger adults, and a substantial proportion of initiators received high-dose therapy. Although potentially inappropriate prescribing, especially indication-related inappropriateness, declined over time, persistent high-dose use and prophylactic prescribing raise concerns about possible overuse and suboptimal prescribing. These findings support implementing targeted prescription audits coupled with prescriber education focused on indication verification at initiation and dose optimisation or step-down therapy, with particular attention to young adults and prophylactic use. Embedding these measures within guideline-concordant workflows and decision-support systems may help promote more rational PPI use and minimise avoidable harms.

The funders had no role in study design, data collection, analysis, decision to publish or manuscript preparation.

Data availability statement

Data are available on reasonable request.