pump inhibitor deprescription: A rapid review

Proton pump inhibitors (PPI) are drugs that suppress gastric acid secretion. Its use, without support from scientific evidence, can contribute to polypharmacy, lead to drug interactions and, in the long term, cause serious adverse reactions. Studies advise physicians to deprescribe PPI. A rapid review of scientific evidence, also called a rapid systematic review, on the deprescribing of PPI was performed. Evidence searches were performed in the LILACS, Embase, PubMed and NICE evidence databases with the terms “omeprazole”, “proton pump inhibitors”, “deprescription”, “deprescribing”. At LILACS these descriptors were also used in Portuguese and Spanish. Of 118 studies identified, four systematic reviews were selected for analysis. Abrupt deprescribing was associated with an increased risk of symptom recurrence. Fear of symptom recurrence is one of the major barriers to patient-related deprescribing. Educational interventions directed at prescribers, pharmacists, and patients are effective strategies in the deprescribing of PPI. Deprescribing process showed to be feasible in different contexts, with different strategies. The process is most effective through actions with educational and guidance materials directed to health professionals and patients, and with the involvement or leadership of the pharmacist.


INTRODUCTION
Proton pump inhibitors (PPI) are drugs that suppress gastric acid secretion by inhibiting the enzyme H+/K+-ATPase, indicated for the treatment of gastric and duodenal ulcers, erosive esophagitis, eradication of H. pylori in combination with antibiotics, prophylaxis of ulcers associated with non-steroidal anti-inflammatory drugs and hypersecretory conditions such as Zollinger-Ellison syndrome. There is no evidence as to the benefit of using PPI for non-ulcer dyspepsia (Wallace, Sharkey, 2012).
For most acid secretion-related illnesses, the duration of PPI treatment varies from two to twelve weeks, but the efficacy, safety profile and tolerability of the drug stimulate long-term use without timely re-evaluation to determine the need for the drug maintenance (Boghossian et al., 2017).
Prolonged use is justified only in the treatment of complications of gastroesophageal reflux disease such as Barrett's esophagus, under hypersecretory conditions such as Zollinger-Ellison syndrome and in patients with erosive esophagitis (Wilsdon et al., 2017). However, according to Reimer et al. (2009) the prevalence of long-term treatment is increasing and up to 70% of patients with chronic acid suppression do not have an indication for PPI treatment. The use of PPI has increased over the past decade, with no new indications being added to their use (Haastrup et al., 2014), according to studies conducted in Denmark and the United Kingdom that reveal this increase after 1990 (Pottegård et al., 2016;Othman, Card, Crooks, 2016).
Patient safety is a relevant topic in the health policy agenda, and it is mandatory to consider it even before the effectiveness of medicines. Primary adverse effects associated with short-term use of PPI include headache, diarrhea, constipation, rash and nausea. Prolonged use may trigger drug interactions, such as reducing the antiplatelet effect of clopidogrel, as well as serious adverse effects such , as pneumonia, hypomagnesemia, vitamin B12 deficiency, C. difficile infection, bone fractures, polyp formation (Fohl, Regal, 2011;Ament, Dicola, James, 2012;Chubineh, Birk, 2012), chronic and acute kidney disease, iron deficiency anemia, dementia (Gomm et al., 2010;Schoenfeld, Grady, 2016;Wilsdon et al., 2017;Guedes et al., 2020).
Adverse effects may be confused with new diseases, leading to the prescription of other medications (Anthierens et al., 2010). The chronic use of PPI, as a consequence, contributes to the increase in unnecessary costs for health systems and polypharmacy (Hasstrup et al., 2014;Boghossian et al., 2017). The increase in prevalence of chronic diseases, multidisciplinary prescriptions and pharmacological choices for health intervention contribute to polypharmacy and expose the elderly population to prescription of potentially inappropriate drugs, with the risk of adverse reactions outweighing the clinical benefits (Gomes et al., 2019, Oliveira et al., 2012.
This practice is common in the elderly and may be beneficial for treating various diseases, but is associated with increased risks of drug interactions, adverse reactions, falls, iatrogenesis, hospitalizations and mortality (Hilmer, Gnjidic, 2009;Gnjidic et al., 2012;Dills et al., 2018;Machado et al., 2017;Motter et al., 2018e Santos et al., 2019. A population-based study, conducted in primary care in Brazil, observed a 47.4% prevalence of clinically important drug interactions in elderly patients (Obreli et al., 2012).
In recent years, the need to reduce over prescription of drugs through an approach called deprescription has been discussed. The term "deprescription" was first mentioned in 2003 in the article"Deprescribing: Achieving Better Health Outcomes for Older People Through Reducing Medications". It is a process planned and supervised by a healthcare professional to reduce, replace or discontinue inappropriate medications to control polypharmacy (Woodward, 2003;Reeve et al., 2015). Scott et al. (2015, p. 827) define deprescription as "the systematic process of identifying and discontinuing drugs in cases where existing or potential harm outweighs existing or potential benefits (...)". Considers the same principles as starting a prescribed therapy, ie, it is a patient-centered process with shared decision making and monitoring of effects.
Planning this process involves recognizing polypharmacy and knowing the list of drugs used by the patient and their indications, identifying inappropriate drugs, evaluating each one and setting priorities for deprescribing, implementing the strategy, and monitoring withdrawal syndrome, rebound effect, recurrence of the disease and patient´s quality of life (Couteur et al., 2011;Reeve et al., 2013;Machado et al., 2017;Santos et al., 2019).
The beneficial consequences of deprescription include the cessation of adverse reactions and drug interactions. It includes also the minimization of future risks, reduced patient and health care costs, improved treatment adherence and patient´s quality of life, and a decreased medication associated errors (Couteur et al., 2011).

MATERIAL AND METHODS
A rapid review of the scientific literature on PPI was conducted, with emphasis on deprescription. The rapid review, also called the systematic rapid review, is a secondary study design that has been increasingly used to inform health policies, especially useful for managers and decision makers (Bortoli et al., 2017).
The search for scientific evidence was performed in the LILACS, Embase, PubMed and NICE evidence databases on July 7, 2019, without the use of filters. The terms extracted from the Descriptors in Health Sciences -DeHS and the Medical Subject Headings (MeSH) were used. In Embase, PubMed and Nice Evidence, the terms "omeprazole", "proton pump inhibitors", "deprescription" and "deprescribing" were used. The same search strategy was employed in LILACS, however, including also the descriptors in Portuguese and Spanish. The details of the search strategy are in Table I.
The article selection process was performed by the author and discussed with the co-author, starting with reading the titles, followed by reading the abstracts and later the full articles. Inclusion criteria were: Systematic reviews, published in English, Spanish and Portuguese. The selected systematic reviews were evaluated for methodological quality through the Assessment of Multiple Systematic Reviews -AMSTAR 2 instrument

Objective
To determine the effects associated with long-term deprescription of PPI therapy in adults compared with chronic daily use (28 days or more).
To determine the effectiveness of interventions to reduce inappropriate use of PPI in the elderly.
To identify barriers and enablers that may influence the patient's decision to discontinue medication use (ME).
To evaluate the result of deprescription in reducing the amount of ME and controlling chronic medical and mental conditions compared with standard treatment in the nonterminal adult population. (Shea et al., 2017), being applied by the author, followed by discussion with the co-author. The SR were classified as high (13-16/16), moderate (9-12/16), low (5-8/16) and critically low (0-4/16) methodological quality. The data were extracted from the SR by the author in an Excel spreadsheet, containing the following information: author/year, objective, quantity and study designs included, most recent search date, AMSTAR 2 score, intervention studied, participant characteristics, location and countries of achievement, outcomes, barriers to implementation, facilitators of implementation and knowledge gaps (Table II). ,
Randomized clinical trials (n = 58).   Pharmacist's intervention in the educative actions with the doctor and the patient, and in the specific recommendations in patient's treatment.

RESULTS
The searches allowed to identify 118 studies, of which six SR were considered eligible and four were selected, according to the selection process presented in Figure 1. The systematic reviews of Page et al., 2016 andMalhotra et al., 2018 were excluded because they did not contemplate the objectives of this study. Of the four SR included, one is of high methodological quality and the others of moderate quality.
Two of the included systematic reviews specifically address PPI (Boghossian et al., 2017;Wilsdon et al., 2017) to determine the effects (Boghossian et al., 2017) and effectiveness of interventions (Wilsdon et al., 2017) associated with deprescribing. The third analyzed barriers and facilitators that influence the patient in the decision to deprescribe (Reeve et al., 2013) and the fourth evaluated the result of deprescription in reducing the amount of medication and controlling medical conditions (Dills et al., 2018). Boghossian et al. (2017) analyzed the effects of two strategies (n=1758): on demand PPI deprescription in patients aged 48 to 57 years old with moderate gastroesophageal reflux disease and mild esophagitis, and abrupt deprescription in patients ≥ 65 years old with mild to moderate esophagitis compared to continuous use (28 days or more). In the on-demand deprescription, 16.3% of participants had return of gastrointestinal symptoms or inadequate relief versus 9.2% in continuous use (RR 1.71; 95% CI 1.31 to 2.21). Fifteen participants in the intervention group developed esophagitis compared to none in the control group. There was a reduction in use on average, of 3.79 tablets of PPI/week (95% CI -4.73 to -2.84). The use of PPI on demand caused greater dissatisfaction among participants compared to the control group, respectively 15.8% and 8.8% (RR 1.82; 95% CI 1.26 to 2.65). Abrupt deprescription was associated with an increased risk of symptom recurrence, with relapse in 69.6% of participants with a history of esophagitis compared with 20.4% of those with continuous PPI use (RR 3.41; 95% CI 1.91 to 6.09). Wilsdon et al. (2017) reported effective and targeted interventions to promote high-dose-reduced PPI deprescription through educational material (leaflets) prepared on the basis of scientific evidence directed to physicians, pharmacists and patients who were in different programs and periods in Australia. The number of low-dose prescriptions increased by 0.6% per month and after 20 months increased to 0.9% per month (p = 0.007). In one of the studies reviewed, these interventions were rated as useful or very useful by 81% of physicians, 95% of pharmacists and 72% of patients. Reeve et al. (2013) studied 1310 participants who were taking or recently discontinuing use of drugs, in order to identify barriers and facilitators that may influence the patient's decision to deprescribe. Two qualitative studies analyzed were conducted in the United Kingdom and cite as barriers to PPI deprescription: Belief in the benefit of the drug for the clinical condition, unwillingness to try alternatives, fear of the return of the clinical condition or the return of symptoms and poor experiences with previous deprescription. On the other hand, the fear of adverse effects, the possibility of restarting the use of the medication, the influence of the primary care physician and the cost of the medication were cited as facilitators of the PPI deprescription. Dills et al. (2018) included adult participants over 18 years old to evaluate the outcome of deprescribing in reducing the amount of medication and in controlling medical conditions. Pharmacist-led educational interventions on symptom management and prescription, directed at prescribers and patient-directed educational interventions on inappropriate drug use resulted in a reduction in PPI dose to maintenance dose in 50% of patients.

DISCUSSION
This review has limitations inherent in the design of a rapid review, such as fewer databases searched, selection processes and data extraction not performed independently, focusing on systematic reviews. On the other hand, this type of review has the advantage of providing timely answers to the demands of managers in the daily routine of health services.
Deprescription is a process that begins prior to the formal act of prescribing a change in conduct. For the deprescription process to be developed effectively and safely, barriers must be considered by both doctors and patients. Confidence in drug therapy for cure or remission of symptoms, limited time for consultation with the healthcare professional, fear of discontinuation of therapy initiated by another prescriber, market influences, lack of communication between prescribers, disagreement between professionals and patients regarding the strategy for deprescription, lack of knowledge in the management of deprescription are barriers experienced by prescribers. In addition, patient resistance to discontinuation or replacement of therapy for fear of symptom recurrence, reporting of unsuccessful experiences of others and pressure from family and community to continue drug use should be considered (Reeve et al., 2013;Boghossian et al., 2017;Wilsdon et al., 2017;Dills et al., 2018). Patient education about the risks and benefits of drug therapy, a structured process of drug withdrawal, monitoring and support facilitate deprescription (Dills et al., 2018). Boghossian et al. (2017) demonstrated that abrupt deprescription was associated with an increased risk of recurrence of gastric symptoms. In the case of PPI, deprescription may be performed with abrupt discontinuation, use on demand until relief of gastric symptoms, use of a lower dose or alternative therapy such as histamine-2 receptor antagonists (Thompson et al., 2018). Despite gaps in the scientific literature regarding agreement on the best strategy for deprescription, considering the clinical effects, the gradual on demand or dose-reduction process of PPI is more effective in controlling the recurrence of gastric symptoms compared to an abruptly withdrawal. (Katz, Gerson, Vela, 2013;Haastrup et al., 2014;Farrell et al., 2017). According to Reimer et al. (2009) abrupt withdrawal of PPI after 8 weeks of treatment may cause rebound acid hypersecretion in healthy adults. In a qualitative study, patients reported that they would use PPI at low or on demand doses (Grime, Pollock, 2002).
It is very important to take these findings into account, as the fear of recurrence of gastric symptoms, associated with an increased risk of abrupt withdrawal, is one of the main barriers to deprescription, in addition to the belief in the benefit of the drug, unwillingness to try alternatives, bad experiences with previous deprescriptions processes and costs. Patients consider the use of PPI for clinical treatment necessary, value the control of gastric symptoms and the quality of life provided by their use and point this class of drugs as the most effective for this purpose (Spijker-Huiges, Winters, Meyboom-De Jong, 2006;Farrell et al., 2017;Thompson et al., 2018). In the study by Spijker-Huiges, Winters, Meyboom-De Jong (2006) 68% of patients reported that they would not accept the return of any symptoms after deprescription.
Systematic reviews by Wilsdon et al. (2017) and Dills et al. (2018) showed that information directed to physicians, pharmacists and patients through educational actions involving teaching materials and explanatory content on the promotion of rational use of medicines, as well as guides and algorithms, guide the conduct in the deprescription process. Based on the awareness of health professionals about prescribing and symptom management, the deprescription process can be relied on through the use of tools for guidance (Walsh et al., 2016;Farrell et al., 2017).
In partnership with groups from other countries, Brazilian investigators performed the translation and cultural adaptation of deprescribing algorithms developed by the Canadian Deprescribing Network (Caden) for various drugs, including PPI (Sbrafh, 2020).
According to Thompson et al. (2018) physicians are also afraid of the return of adverse effects in the face of deprescribing, so a strategy to guide deprescribing should include the identification, evaluation and prioritization of drugs in relation to the potential for risk, in a shared way between doctors and pharmacists. The time limitation on primary care physicians imposed by the routine of the service, however, implies the lack of reevaluation of continuous use medications (Thompson et al., 2018).
In the midst of a process that involves technical knowledge, established care routines and patients' anxieties, the experiences and expectations should be considered and discussed as a component for the shared development of the best strategy for deprescribing. The adverse effects of long-term PPI use worry patients in inverse proportion to the degree of satisfaction with symptom control and the costs incurred to maintain treatment (Chey, Mody, Izat, 2010). According to studies (Farrell et al., 2017;Thompson et al., 2018), patients agree to discuss over prescription, are willing to decrease PPI use, and information exchange is important in this process. The patient is interested in understanding what is, how effective is, what actions and options are considered in view of the different outcomes, especially the occurrence of symptom recurrence and the possibility of resumption of PPI treatment. In the study by Smeets et al. (2009), patients considered the clarification of their involvement, the reasons for the deprescription, and the possibility of symptom recurrence to be of greater importance in the deprescription process.
In this sense, the inclusion of the pharmacist in health teams and their involvement in actions related to the promotion of rational use of medicines, educational actions to provide patient education and review of the list of medicines used, becomes increasingly relevant. This includes also monitoring symptoms in a shared and complementary manner to the physician (Farrell et al., 2017); and reducing indiscriminate drug use and health system costs (Bundeff, Zaiken, 2013).
The studies included in this review were conducted in Europe, the United States and the Middle East and show that deprescription is feasible in different contexts with different strategies. The findings of the systematic reviews indicate that the process is most effective through actions with educational and guiding materials directed to health professionals and patients, with the involvement or leadership of the pharmacist. There were no studies conducted in Brazil on PPI deprescription, however, at the care level, the factors implicated in greater effectiveness and the actors involved are generally common to health systems, yet adaptations may be necessary to adapt to the local reality.