Hemodialysis (HD) patients take more pills per day on average than any other chronically ill patient population. On average, an HD patient takes 19 medications per day, of which 70% may not be appropriate. The reason the medications may not be appropriate is that HD patients are rarely included in clinical trials for new medications and therefore the efficacy and safety data that exists for the general population may not actually apply to them. Tools to guide the re-assessment and discontinuation (deprescribing) of these specific medications that lack evidence for efficacy and safety in HD patients are needed. These tools will help reduce the amount of medications being taken and the potential negative consequences of taking so many medications (e.g. adverse drug reactions, drug interactions, non-adherence, increased risk of cognitive impairment, impaired balance and falls, and increased risk of morbidity, hospitalization, and mortality).

Nine medications that are often inappropriately prescribed to HD patients have been identified by the investigators. These medications are: Alpha-1 Blockers, Anticonvulsants, Benzodiazepines & Z-Drugs, Loop Diuretics, Prokinetic Agents, Proton Pump Inhibitors, Quinine, Urate Lowering Agents, and Statins. The investigators developed and validated tools to help medical teams in outpatient HD units with identifying and stopping these medications in their patients. The next step will be to perform a study where test these tools are tested in practice at multiple HD centers across Canada. This initiative should decrease the average number of medications per patient and inappropriate medication use in the HD units where these tools are used. The overall objective of this study is to improve current clinical practice by optimizing medication use and prescribing patterns in the HD units across Canada.

Background: Hemodialysis (HD) patients are rarely included in clinical trials, thus medication efficacy and safety data specific to this population is lacking. Toxicity from medications inadequately removed by dialysis is also a risk for them. HD patients take an average of 19 pills daily, with 70% potentially inappropriate.

This polypharmacy increases their risk of adverse events, drug-drug interactions, non-adherence, cognitive impairment, impaired balance and falls, morbidity, hospitalization and mortality. Using provincial databases, the investigators identified 9 medication classes with uncertain indications and/or safety in HD patients: Alpha-1 Blockers, Anticonvulsants, Benzodiazepines & Z-medications, Loop Diuretics, Prokinetic Agents, Proton Pump Inhibitors, Quinine, Urate Lowering Agents and Statins. The investigators developed and validated tools for deprescribing, safety monitoring and patient education for each of these medications. The next step will be to perform an implementation study evaluating these deprescribing tools at multiple HD units across Canada.

The investigators hypothesize that implementation of these deprescribing tools will decrease polypharmacy and improve safety and patient satisfaction in these HD units. Specific Aims are to determine:

1. Effectiveness of the deprescribing algorithms for decreasing polypharmacy (i.e. % of successful deprescribing of at least 1 of the medication classes at 6 month post implementation)
2. Safety of the deprescribing algorithms using monitoring tools developed for each medication
3. The impact of the deprescribing tools on patient satisfaction

Methods: In this quasi-experimental interventional cohort study, the nephrology healthcare team will assess medications for all patients as per usual practice in their respective HD units, using the deprescribing algorithms to assist in clinical decision making and patient education tools to explain rationale to patients. Participating patients will be followed for 6 months for outcomes.

The primary outcome will be proportion of individuals successfully deprescribed at least one of the 9 target medications.

Additional outcomes include:

• Adverse events associated with deprescribing and medication class specific safety outcomes (e.g. for furosemide, the investigators will be tracking blood pressure, potassium, intradialytic weight gain and heart failure admissions)
• Proportion of identified candidates who began a deprescribing trial
• Proportion of deprescribing trials declined by medical team and patient, respectively
• Patient satisfaction (using a patient survey)
• Average number of medications per patient before/after implementation
• Average medication cost savings per patient

Expected Results/Impact on Health Research: This study will determine the efficacy of the deprescribing algorithms on reducing polypharmacy in HD patients. It will also provide insights on knowledge translation, as investigators aim to educate providers and patients on the harms of polypharmacy and influence prescribing patterns in HD units nationally. This study will encourage other institutions to incorporate similar tools into their practice and encourage comprehensive and team based re-assessment of patient's medications.

Inclusion Criteria:

• 18+ years
• Have been receiving outpatient HD treatment at one of the four study sites for at least the past three months
• Able to read and understand English and provide consent

Exclusion Criteria:

- Acute starts to HD

There will be approximately 1,200 patients in the outpatient HD units from Toronto (300), Vancouver (200), Winnipeg (400) and Halifax (300). Based on a previous pilot study, it is assumed that 80% of patients have been prescribed at least one of the nine target drugs (i.e., n=960). Of those, it is assumed that 50% will be eligible for the study (i.e., n=480). Of eligible individuals, it is assumed 88% will initiate a De-prescribing Trial (Intervention Group), resulting in an anticipated cohort of n=420.