Abstract
Background
A limited number of jurisdictions worldwide authorize independent pharmacist prescribing. Legislative reforms in Quebec (Canada) now empower pharmacists to independently perform certain professional activities. Since 2021, Advanced Practice Agreements at the McGill University Health Centre have enhanced pharmacists’ clinical autonomy, though the scope of each agreement varies across units. Limited evidence exists on how these agreements influence prescribing practices. This audit aimed to describe pharmacist-written prescriptions at the McGill University Health Centre, and examine inter-unit differences in prescribing trends.
Methods
This retrospective audit of three hospitals included 300 patients admitted in February, May, and October 2024 across five inpatient units. For each included patients, all prescriptions were reviewed. Descriptive metrics were developed and included total prescriptions, prescriptions per patient per day, proportion of patient days with pharmacist prescriptions, types of interventions, and medication classes.
Results
Seventy-nine percent of patients screened had at least one prescription written by a pharmacist. Pharmacists authored 32% of the 18 960 recorded prescriptions with 95% of these written independently. The proportion of pharmacist-written prescriptions varied across services. Most interventions involved prolongation or transfer orders, and initiation of new pharmacotherapies. Pharmacists actively prescribed high-risk medications, including anticoagulants, anticancer therapy, and hypoglycemic agents, highlighting their clinical expertise.
Conclusion
Advanced Practice Agreements support pharmacists’ autonomous prescribing and integration into patient care, with substantial involvement in complex therapies. Variation across units likely reflect differences in agreements’ scope and timing of implementation, namely. Future research should evaluate the impact of pharmacist-led prescriptions on issues such as hospital readmissions.
Introduction
A limited number of jurisdictions worldwide, including the United Kingdom, Australia, and certain provinces in Canada, authorize pharmacists to prescribe independently under defined conditions [1–6]. The scope of this authority differs substantially across regions, with notable variation in pharmacists’ prescribing rights in both community and hospital settings [1, 4].
In Quebec (Canada), legislative reforms over the past decade have similarly recognized pharmacists as authorized prescribers within their professional scope. All pharmacists may now prescribe for minor ailments, prolong prescriptions and adjust medications in various contexts. However, expanded prescribing privileges such as independently initiating a broader range of drug therapies can be granted through Advanced Practice Agreements (APAs) established with physicians. Under these agreements, pharmacists are able to practice to their full scope.
Since 2021, several APAs have been implemented at the McGill University Health Centre (MUHC). Their structure and content, however, vary across departments due to differences in clinical environments, patient needs, and interprofessional models of care. Although other healthcare institutions in Quebec have begun adopting similar agreements [7], little is known about how this variability in APA design affects pharmacists’ prescribing practices.
Furthermore, while independent prescribing by hospital-based pharmacists for in-patients has been described previously [2, 3, 5, 7], there remains no standardized framework or metrics to characterize prescribing practices, quantify pharmacists’ workload, or support meaningful comparisons across healthcare institutions.
This audit aimed to describe prescriptions written by pharmacists working under APAs at the MUHC. Inter-sectorial differences in prescribing patterns were examined and descriptive metrics were developed to better characterize prescribing intensity and variability.
Methods
Study design
This retrospective audit was undertaken in three tertiary care hospitals affiliated with the MUHC.
Setting
The Royal Victoria Hospital (RVH, 368 beds) includes a 35-bed medical-surgical intensive care unit (ICU), a hematology-oncology unit (24 beds), a medical oncology unit (12 beds), and two internal medicine (IM) units (45 beds). The Montreal General Hospital (MGH, 282 beds) is a level one trauma center with a 26-bed medical-surgical-trauma ICU. The Montreal Neurological Hospital (MNH, 85 beds) has 12 ICU beds focused on neurocritical care and neurosurgery. These in-patient units were selected because they had fully implemented APAs at the time of the audit. To simplify data collection at the RVH, only one IM unit was reviewed. The hematology-oncology and medical oncology units (HMO) were grouped together, as they share the same floor and APA. Supplementary Appendix S1 outlines the pharmacy services available on these units. Supplementary Table S1 reports information about pharmacy staffing on these units and their experience level.
Participants
Patients were identified using the pharmacy software database. Eligibility criteria included: (i) Admission to an audited service in February, May, or October 2024; (ii) Had at least one prescription written by a pharmacist during their stay. Patients were excluded if: (i) They had no prescriptions written by a pharmacist during their stay; (ii) They were still hospitalized at the time of the audit (to ensure all handwritten prescriptions were scanned and available in the electronic medical record); (iii) They had a length of stay (LOS) of over 30 days (to avoid excessively long audits); (iv) Their electronic charts did not include prescriptions or their prescriptions had not been archived and scanned.
These time periods were selected because they typically represent times when the MUHC operates at full capacity.
Procedures
A retrospective audit of the medical records was performed using OACIS Electronic Medical Record (version 7.5.0, Telus Health Solutions GP, ON, Canada) and BDM pharmacy (version 10.40.9, BDM Healthware Inc., SK, Canada) as data sources. All data was collected by three authors (MAD, ATD, CC) using pre-piloted standardized case report forms on Microsoft Excel 365 (Microsoft Corporation, WA, USA). A validation of 10% of the dataset was performed by a second person to ensure accuracy of data entry.
For each patient, the date of admission and discharge from the unit as well as the total number of prescriptions written (by all prescribers and by pharmacists) was collected. For prescriptions by written by pharmacists, the following information was extracted: (i) Whether the prescription was a suggestion to the team or done as an independent prescriber. (ii) The type of intervention done (e.g. initiating a new medication). (iii) The medication class of the prescribed drug (e.g. analgesia).
Supplementary Appendix S2 describes in details the pre-established types of intervention and medication classes associated with each audited prescription written by a pharmacist. The types of interventions and the terminology used were derived from the Quebec’s provincial pharmacist professional practice guide [8] and from additional frequent interventions seen in clinical practice. Medication classes were chosen based on the perceived frequency of use on the audited units and by discussion between authors. Although anticancer therapy is typically prescribed collaboratively by pharmacists and physicians at the MUHC, the authors chose to attribute these prescriptions to pharmacists due to their predominant role in the prescribing process.
Each order for a medication, laboratory, or test was considered a prescription. This included transfer orders from reports generated on OACIS, pre-printed orders (pre-printed admission orders bundles, standardized insulin forms, etc.) and handwritten prescriptions. Orders for enteral nutrition, mechanical ventilation, blood products or anything other than a medication or a test were not assessed. There was no evaluation of clinical appropriateness.
To better understand pharmacists’ prescribing practices, workload, and presence across units, the authors sought to develop meaningful comparison metrics, as no standardized indicators currently exist. To address this gap, differences in pharmacist coverage (e.g. five versus seven days per week) were considered, along with the proportion of patient-days during which pharmacists actively prescribed. Through discussion, the authors developed specific audit metrics to be calculated for each patient, which are presented in Box 1.
Box 1.
Definition and description of metrics developed for the audits
-
Number of prescriptions per patient per day of stay (for all prescribers and for pharmacists)
Defined as the total number of prescriptions written for a patient (by all prescribers or by pharmacists) divided by the length of stay on the unit in days
For cases where the length of stay was less than one day, a minimum length of stay of one full day was applied in the calculation to avoid reporting extreme or misleading values.
-
Daily prescription count on days when pharmacist prescribes
Total of all prescriptions written by pharmacists for each separate calendar day pharmacists prescribe during a patient’s stay
To account for variations in pharmacy clinical coverage (5 days a week vs 7 days) on audited units
-
Proportion of stay dates where a pharmacist prescribed at least once
Number of separate calendar day of stay with at least one prescription by a pharmacist divided by the total number of separate calendar day of stay
Reported as a percentage
Sample size
The first 20 patients of each targeted period to fit inclusion and exclusion criteria were selected for each audited ward.
Outcomes
The primary outcome was the description of pharmacist prescribing activities.
Statistical methods
Results were summed up descriptively using proportions or median and interquartile range, as appropriate.
Results
A total of 422 patients were screened (Fig. 1). The most frequent reason for exclusion was the absence of a pharmacist-written prescription (87 patients). Overall, pharmacists prescribed at least once on 335 patients (79% of screened patients). Ultimately, 300 patients were included in the study.
Figure 1.
Study flow diagram.
Table 1 presents data on patients’ length of stay and prescribing intensity on audited units by all healthcare professionals, including pharmacists. A total of 18 960 prescriptions were recorded. Pharmacists were responsible for 32% of all prescriptions, and 95% of these were written independently. The RVH ICU had the highest proportion of pharmacist-written prescriptions (46%). Independent prescribing by pharmacists was consistently high across most units, except in IM, where only half of the prescriptions were written independently, and the other half were suggestions to the treating team.
Table 1.
Description of patients and prescriptions audited.
| Overall | RVH internal medicine | RVH hematology-oncology/medical oncology | RVH ICU | MGH ICU | MNH ICU | |
|---|---|---|---|---|---|---|
| Number of patients screened (n) | 422 | 81 | 91 | 68 | 69 | 113 |
| Patients screened with at least one prescription by a pharmacist (n, %) | 335 (79%) | 71 (88%) | 74 (81%) | 66 (97%) | 63 (91%) | 61 (54%) |
| Number of patients included (n) | 300 | 60 | 60 | 60 | 60 | 60 |
| Patient length of stay on audited unit, median (IQR) (days) |
|
|
|
|
|
|
| Prescriptions written by all prescribers (n) | 19 960 | 2931 | 2990 | 4692 | 4788 | 3559 |
| Prescriptions written by pharmacists (n, %) | 6075 (30%) | 537 (18%) | 899 (30%) | 2170 (46%) | 1643 (34%) | 826 (23%) |
| Written as independent prescriber (n, %) | 5798 (95%) | 267 (50%) | 894 (99%) | 2168 (100%) | 1643 (100%) | 826 (100%) |
| Suggestion for treating team (n, %) | 277 (5%) | 270 (50%) | 5 (1%) | 2 (0%) | 0 (0%) | 0 (0%) |
| Prescriptions per patient by all prescribers, median (IQR) (n) |
|
|
|
|
|
|
| Prescriptions per patient by pharmacists, median (IQR) (n) |
|
|
|
|
|
|
| Ratio of pharmacists’ median prescriptions per patient to the median prescriptions per patient from all prescribers (%) | 27.5% | 16.2% | 31.6% | 43.9% | 35.5% | 13.5% |
| Prescriptions per patient per day of stay written by all prescribers, median (IQR) (n) |
|
|
|
|
|
|
| Prescriptions per patient per day of stay written by pharmacists, median (IQR) (n) |
|
|
|
|
|
|
| Daily prescription count on days when pharmacist prescribes, median (IQR) (n) |
|
|
|
|
|
|
| Proportion of stay dates where a pharmacist prescribed at least once, median (IQR) (%) |
|
|
|
|
|
|
ICU, Intensive care unit; IQR, Interquartile range; MGH, Montreal General Hospital; MNH, Montreal Neurological Hospital; RVH, Royal Victoria Hospital.
Across all prescribers, the median number of prescriptions per patient was 54.5, with the RVH ICU (69.5) and MGH ICU (69.0) showing the highest prescribing intensity. Pharmacists at the RVH ICU also had the highest median number of prescriptions per patient (30.5), while the IM unit had the lowest (6). When considering prescriptions per day of stay, ICUs had higher rates (RVH: 12.6/day, MGH: 14.7/day, MNH: 14.0/day) compared to IM (4.8/day) and HMO (6.0/day). Pharmacist-specific prescribing rates per day of stay were highest at the RVH ICU (5.1/day) and MGH ICU (4.3/day). Additionally, the median proportion of stay dates where a pharmacist prescribed at least once was notably higher in the RVH ICU (70%) and MGH ICU (50%) compared to IM and HMO (25%).
Table 2 outlines the types of interventions pharmacists carry out. The most common interventions involved prolonging a prescription or issuing transfer orders (36.5%), prescribing new medications (23.9%), specifying the duration of therapy or discontinuing medications (15.8%), and adjusting the dose, frequency, or posology (10.2%). The HMO (66.1%) and IM (29.1%) units had the highest rate of new medication prescriptions. In contrast, prolongation and transfer orders were predominantly observed in the ICU settings (38.3%–56.1% of prescriptions). Interventions involving therapeutic drug monitoring (TDM) and the prescription of tests or labs were relatively rare, representing only 2.5% and 0.8% of orders, respectively.
Table 2.
Type of interventions recorded for prescriptions by pharmacists.
| Type of intervention/unit | Overall | RVH internal medicine | RVH hematology oncology/medical oncology | RVH ICU | MGH ICU | MNH ICU |
|---|---|---|---|---|---|---|
| Prolonging the duration of a treatment or prescribing transfer orders (n, %) | 2119 (36.5%) | 24 (4.5%) | 0 (0%) | 831 (38.3%) | 922 (56.1%) | 345 (41.8%) |
| Initiating a new drug therapy (n, %) | 1385 (23.9%) | 156 (29.1%) | 594 (66.1%) | 409 (18.8%) | 161 (9.8%) | 65 (7.9%) |
| Specifying the duration of therapy or stopping a medication (n, %) | 916 (15.8%) | 86 (16.0%) | 102 (11.3%) | 297 (13.7%) | 164 (10.0%) | 267 (32.3%) |
| Adjusting the dose, frequency or posology of a medication (n, %) | 590 (10.2%) | 72 (13.4%) | 49 (5.5%) | 258 (11.9%) | 176 (10.7%) | 35 (4.2%) |
| Represcribing home medication (n, %) | 236 (4.1%) | 63 (11.7%) | 5 (0.6%) | 80 (3.7%) | 57 (3.5%) | 31 (3.8%) |
| Interrupting (“holding”) a prescription (n, %) | 205 (3.5%) | 22 (4.1%) | 13 (1.4%) | 69 (3.2%) | 54 (3.3%) | 47 (5.7%) |
| Clarifying a prescription (n, %) | 185 (3.2%) | 45 (8.4%) | 67 (7.5%) | 41 (1.9%) | 24 (1.5%) | 8 (1.0%) |
| Prescribing therapeutic drug monitoring (n, %) | 144 (2.5%) | 37 (6.9%) | 13 (1.4%) | 44 (2%) | 40 (2.4%) | 10 (1.2%) |
| Prescribing or adjusting insulins (n, %) | 111 (1.9%) | 7 (1.3%) | 1 (0.1%) | 77 (3.5%) | 24 (1.5%) | 2 (0.2%) |
| Changing the route of administration of a drug (n, %) | 62 (1.1%) | 1 (0.2%) | 4 (0.4%) | 41 (1.9%) | 12 (0.7%) | 4 (0.5%) |
| Prescribing tests or labs (n, %) | 48 (0.8%) | 14 (2.6%) | 20 (2.2%) | 3 (0.1%) | 9 (0.5%) | 2 (0.2%) |
| Substituting a medication for another that is on the hospital formulary (n, %) | 23 (0.40%) | 8 (1.5%) | 2 (0.2%) | 3 (0.1%) | 0 (0%) | 10 (1.2%) |
| Othera (n, %) | 51 (0.9%) | 5 (0.9%) | 29 (3.2%) | 17 (0.7%) | 0 (0%) | 0 (0%) |
Other interventions include agreeing to suggestions from other services in closed units, prescribing medication for a temporary discharge from hospital (e.g. 24 h pass), prescribing tapering doses of medication for weaning purposes or changing the diluent for IV preparations.
ICU, Intensive care unit; IV, Intravenous; MGH, Montreal General Hospital; MNH, Montreal Neurological Hospital; RVH, Royal Victoria Hospital.
Table 3 presents the most frequently prescribed medication classes by pharmacists. The most common categories included electrolytes and intravenous fluids (12.4%), analgesics (10.4%), antacids, antidiarrheals, and antiemetics (10.2%) and antimicrobials (10.0%). Anticancer therapy prescriptions were almost exclusively limited to the HMO unit. Immunosuppressants were more commonly prescribed in IM and the RVH ICU compared to other units. Other high-risk drug classes such as anticoagulants (both prophylactic and therapeutic) and insulin or hypoglycemic agents accounted for 4.2% and 4.7% of pharmacist-written prescriptions, respectively.
Table 3.
Most frequent medication classes involved in pharmacists’ prescriptions.
| Medication class/unit | Overall | RVH internal medicine | RVH hematology-oncology/medical oncology | RVH ICU | MGH ICU | MNH ICU |
|---|---|---|---|---|---|---|
| Electrolytes and IV fluids (n, %) | 752 (12.4%) | 93 (17.3%) | 137 (15.2%) | 166 (7.7%) | 226 (13.8%) | 130 (15.7%) |
| Analgesia (n, %) | 632 (10.4%) | 26 (4.8%) | 43 (4.8%) | 237 (10.9%) | 229 (13.9%) | 97 (11.8%) |
| Antacids, antidiarrheals and antiemetics (n, %) | 617 (10.2%) | 52 (9.7%) | 196 (21.8%) | 169 (7.8%) | 94 (5.7%) | 106 (12.8%) |
| Antimicrobials (n, %) | 605 (10.0%) | 90 (16.8%) | 102 (11.3%) | 235 (10.8%) | 147 (8.9%) | 31 (3.8%) |
| Sedation and antipsychotics (n, %) | 403 (6.6%) | 22 (4.1%) | 25 (2.8%) | 157 (7.2%) | 162 (9.9%) | 37 (4.5%) |
| Antihypertensive and diuretics (n, %) | 372 (6.1%) | 27 (5.0%) | 11 (1.2%) | 184 (8.5%) | 97 (5.9%) | 53 (6.4%) |
| Vitamins (n, %) | 301 (5.0%) | 49 (9.1%) | 18 (2.0%) | 91 (4.2%) | 101 (6.1%) | 42 (5.1%) |
| Laxatives (n, %) | 292 (4.8%) | 0 (0%) | 0 (0%) | 95 (4.4%) | 69 (4.2%) | 128 (15.5%) |
| Hypoglycemic agents and insulin (n, %) | 286 (4.7%) | 36 (6.7%) | 3 (0.3%) | 152 (7.0%) | 67 (4.1%) | 28 (3.4%) |
| Anticoagulants (n, %) | 258 (4.2%) | 31 (5.8%) | 18 (2.0%) | 109 (5.0%) | 65 (4.0%) | 35 (4.2%) |
| Inhalers, bronchodilators, mucolytics and anti-leukotrienes (n, %) | 199 (3.3%) | 7 (1.3%) | 1 (0.1%) | 64 (3.0%) | 119 (7.2%) | 8 (1.0%) |
| Corticosteroids (n, %) | 192 (3.2%) | 10 (1.9%) | 62 (6.9%) | 47 (2.2%) | 35 (2.1%) | 38 (4.6%) |
| Chemotherapy (n, %) | 143 (2.4%) | 0 (0%) | 140 (15.6%) | 3 (0.1%) | 0 (0%) | 0 (0%) |
| Immunosuppression (n, %) | 124 (2.0%) | 27 (5.0%) | 9 (1.0%) | 85 (3.9%) | 1 (0.1%) | 2 (0.2%) |
| Vasopressors and inotropes (n, %) | 90 (1.5%) | 0 (0%) | 0 (0%) | 65 (3.0%) | 25 (1.5%) | 0 (0%) |
| Antiplatelets (n, %) | 72 (1.2%) | 3 (0.6%) | 3 (0.3%) | 42 (1.9%) | 13 (0.8%) | 11 (1.3%) |
| Antiepileptics (n, %) | 66 (1.1%) | 0 (0%) | 6 (0.7%) | 7 (0.3%) | 33 (2.0%) | 20 (2.4%) |
| Mouthwashes and anti-mucositis agents (n, %) | 42 (0.7%) | 0 (0%) | 25 (2.8%) | 14 (0.6%) | 2 (0.1%) | 1 (0.1%) |
| Granulocyte-colony stimulating agents (n, %) | 29 (0.5%) | 1 (0.2%) | 28 (3.1%) | 0 (0%) | 0 (0%) | 0 (0%) |
| Antiarrhythmics (n, %) | 25 (0.4%) | 2 (0.4%) | 1 (0.1%) | 20 (0.9%) | 2 (0.1%) | 0 (0%) |
| Other medication classesa (n, %) | 538 (8.9%) | 47 (8.8%) | 51 (5.7%) | 227 (10.6%) | 156 (9.5%) | 57 (6.9%) |
| Tests/labs prescribed not related to a specific medication class (n, %) | 37 (0.6%) | 14 (2.6%) | 20 (2.2%) | 1 (0%) | 0 (0%) | 2 (0.3%) |
Included all other classes of medication.
ICU, Intensive care unit; IV, Intravenous; MGH, Montreal General Hospital; MNH, Montreal Neurological Hospital; RVH, Royal Victoria Hospital.
Supplementary Table S2 presents the most common medication classes associated with selected pharmacist prescribing interventions. The medication classes most frequently initiated by pharmacists varied considerably across units. When pharmacists adjusted prescriptions, analgesics and antimicrobials were the medication classes most often involved.
Discussion
Statement of principal findings
There was considerable variation across services in the proportion of pharmacist-written orders. At the RVH and MGH ICUs, pharmacists authored 46% and 34% of medication prescriptions on patients that they prescribed on, respectively. Very few patients were excluded due to the absence of a pharmacist-written prescription on these units, suggesting these proportions closely reflect actual practice. In contrast, 113 patients were screened at the MNH ICU, but most exclusions were due to a lack of pharmacist-generated prescriptions, indicating the true proportion is likely lower than the reported 23%. To a lesser extent, the actual proportion of pharmacist-issued prescriptions on the IM and HMO units is also likely lower than reported.
Differences in prescribing frequency between units may reflect variations in patient populations, acuity of care, length of stay, clinical priorities, pharmacist coverage, pharmacist-to-patient ratios, and the specific scope of each APA. This evaluation also does not capture the substantial time that non-ICU pharmacists spend preparing discharge prescriptions, while it does include transfer orders within ICUs.
Pharmacists’ experience may also have contributed to the observed variability in independent prescribing. Prescribing comfort may have been lower among pharmacists who had recently joined the rotation, as well as among those with extensive experience who may find it challenging to modify long-standing practice habits. Additionally, APAs were not implemented simultaneously across units. For example, the IM APA had been in place for less than a year at the time of the audit, whereas ICU APAs had been active for nearly three years. As a result, some pharmacists may not have fully adapted their practice to incorporate more independent prescribing at the time of the audit. This may be why a higher number of suggestions made to the treating team rather than autonomous prescriptions on that specific unit.
Although clinical pharmacists had been part of these units for over 30 years at the time of the audit, physician comfort with pharmacist-led independent prescribing likely varied. While some of these differences are reflected in the design of certain APAs such as in IM, it is possible that, even when physicians are not opposed to pharmacist prescribing, preferences may still influence practice. For instance, some ICU medical staff may prefer that medical residents preferably write prescriptions during rounds so they can practice and develop prescribing skills.
While imperfect, the suggested metrics offer valuable insight into the intensity of pharmacists’ involvement in prescribing activities. When they are present on the unit, pharmacists working under an APA intervene frequently and make multiple clinical decisions per patient each day, highlighting their strong integration in patient care.
A wide range of pharmacist interventions were reported. In ICUs, these primarily involved transfer orders, reflecting the dynamic nature of patient care, as patients are transferred to ICUs from lower-acuity wards, then sent back once stabilized. Few TDM interventions were captured despite pharmacists’ extensive involvement in TDM, likely due to electronic ordering of drug levels not transcribed onto paper prescriptions. Similarly, other laboratory tests and clinical assessments were underrepresented due to variations in documentation practices. The significant rate of independent initiation of new drug therapies by pharmacists highlights their clinical autonomy and assurance in therapeutic decision-making.
The diversity of medications prescribed by pharmacists in this study reflects their integration into the clinical management of patients. The data highlights pharmacists’ active role in prescribing high-risk medications, including anticoagulants, anticancer therapy, hypoglycemic agents, and analgesics, underscoring their clinical expertise and the trust placed in their judgment within complex therapeutic areas.
Strengths and limitations
This study has several strengths, including a comprehensive review of prescriptions across multiple inpatient units and a detailed analysis of pharmacists’ prescribing behavior. It introduced novel metrics aimed at capturing pharmacists’ workload and level of involvement in patient care to provide a richer understanding of pharmacists’ clinical contributions and may help future evaluations of their impact as prescribers.
This study has some limitations. This retrospective analysis reflects prescribing practices at the MUHC, which may differ from those in other hospitals or jurisdictions. To enhance feasibility, only one of the two IM units with an APA was audited. Although pharmacists rotate between these units and patient populations are comparable, this may have introduced unmeasured heterogeneity. Three time periods were reviewed, and only the first 20 patients of each period were included. This limited sampling may not reflect seasonal variations, and because patients were not selected randomly, the results may not fully represent overall prescribing practices. The study also focused exclusively on units with APAs, excluding others where clinical pharmacists practice without this designation. The reported workload and number of interventions may be underestimated, particularly in IM and HMO as discharge prescriptions were not included. Some misclassification may have occurred in distinguishing home medications from new prescriptions, particularly when chronic therapies were not explicitly identified. Patients with a LOS >30 days were excluded to improve audit feasibility. Although few patients met this criterion, pharmacists may have exhibited different prescribing behaviors in extended-stay cases. Verbal recommendations from pharmacists to physicians that resulted in prescriptions were not captured. There is also a possibility that some prescriptions may not have been scanned into the electronic chart. This descriptive evaluation of prescriptions by pharmacists was done without an assessment of clinical appropriateness. Finally, as the study was conducted in a paper-based order system, generalizability of some elements to electronic prescribing environments may be limited.
Interpretation with the context of the wider literature
The definition of APA in this audit differs from the one reported in previous studies in hospitalized patients [3, 5]. In Quebec, pharmacists can prescribe for minor ailments, prolong prescriptions and adjust medications in various contexts without the need for an additional attestation of competence. Each APA define additional prescribing rights for pharmacists. Whether they are specific to certain conditions or drugs or are very “liberal” is up to the pharmacists and physicians involved. This contrasts with additional prescribing authorizations in some jurisdictions (e.g. Alberta) as these do not allow different prescribing rights from a unit to another. Another difference is also that APAs are not mandatory in Quebec. Therefore, it requires an initiative from the pharmacy department or a pharmacist team in order to make this happen. Finally, given certain recent accommodations of the federal law, pharmacists working under an APA in Quebec can adjust narcotics and controlled drugs prescriptions and even prescribe them for hospitalized patients.
Because of differences in sampling methods across studies, it is not possible to directly compare the annual number of prescriptions written for in-patients by clinical pharmacists or by full-time equivalent [3, 5]. As for weekly prescribing rates by pharmacists, Heck and colleagues mentioned a median of 4.2 self-reported prescriptions per week per 10 patients, while Saunders and colleagues found a median of 2.6 prescriptions per pharmacist per week [5, 9]. In the present audit, within the RVH ICU, pharmacists wrote a median of 5.1 prescriptions per patient per day of stay. Assuming a typical daily workload of 10–12 patients per pharmacist on that unit, this corresponds to an estimated 51–61 prescriptions per pharmacist per day (357–427 per week assuming the pharmacist is present 7 days). As previously discussed, this does not capture fully the laboratories ordered by pharmacists. Therefore, the prescribing volume by pharmacist, including prescriptions, laboratories and tests, is likely higher. While prescribing rates on other units of our institution differ, it is reasonable to say that pharmacists working under APAs at the MUHC appear to prescribe more than what has been previously reported.
Finally, the drug classification system used in this audit differed from those in previous studies, limiting the ability to identify individual prescribed drugs rather than broader drug classes. Nevertheless, as previously reported, pharmacists most often intervened on analgesics, antimicrobials, and gastrointestinal medications [3, 5]. Interestingly, on certain units in this audit, antihypertensives, sedatives, and antipsychotics were the most frequently prescribed drug classes, marking a notable contrast with earlier findings.
Implication for policy, practice, and research
As Quebec moves toward granting autonomous prescribing privileges to most hospital pharmacists, this audit offers valuable insights into what pharmacists currently do when they prescribe independently at full scope of practice. It also proposes metrics that can support benchmarking across institutions or jurisdictions. Further research should explore the frequency of medication errors when pharmacists prescribe, the clinical appropriateness of their prescriptions, and adherence to clinical guidelines. As for outcomes-guided research, future studies should also aim to define the impact of pharmacist-led prescriptions on issues such as hospital readmissions, patient satisfaction and attainment of clinical targets.
Conclusions
This study highlights the growing role of pharmacists as independent prescribers following the implementation of APAs on inpatient units at the MUHC. It confirms the value of APAs in deepening pharmacist involvement in patient care and enhancing the quality of pharmaceutical services. The audit a broad range of clinical interventions, and substantial pharmacist engagement in managing complex and high-risk therapies. Future studies assessing compliance with clinical guidelines and comparing electronic prescribing environments could further clarify pharmacists’ prescribing roles and guide professional policy development.
Supplementary Material
Contributor Information
Marc-Alexandre Duceppe, Department of Pharmacy, McGill University Health Centre, Montreal, Québec, Canada; Faculté de Pharmacie, Université de Montréal, Montréal, Québec, Canada.
Anh Thu Do, Department of Pharmacy, McGill University Health Centre, Montreal, Québec, Canada.
Catarina Campanella, Department of Pharmacy, McGill University Health Centre, Montreal, Québec, Canada.
Laura Benichou, Department of Pharmacy, Centre Hospitalier Sud Francilien, Corbeil-Essonnes, France.
Catherine Sicard, Department of Pharmacy, McGill University Health Centre, Montreal, Québec, Canada.
André Bonnici, Department of Pharmacy, McGill University Health Centre, Montreal, Québec, Canada.
Caroline Beauchamp, Department of Pharmacy, McGill University Health Centre, Montreal, Québec, Canada.
Author contributions
M.-A.D., A.T.D., L.B., C.S., and C.B. contributed to the research design. M.-A.D. and A.T.D. wrote the study protocol, prepared, and tested the data collection forms. M.-A.D., A.T.D., and C.C. collected the data. M.-A.D. and A.T.D. performed statistical analysis. M.-A.D. and A.T.D. drafted the manuscript. All authors participated in interpretation of the data and revision of the manuscript. All authors have read and approved the final version of the manuscript.
Supplementary material
Supplementary material is available at IJQHC online.
Conflicts of interest
The authors have no relevant financial or non-financial interests to disclose.
Funding
Marc-Alexandre Duceppe received a salary grant from the Fonds de recherche du Québec en santé (DOI: 10.69777/357525). The funders were not involved in the study design nor the writing of the report. There were no restrictions regarding the submission of the report for publication.
Ethics and other permissions
A waiver was obtained by the McGill University Health Centre Research Ethics Board on 18 August 2025 given the use of de-identified data for this quality improvement study.
Data Availability
The authors confirm that the data supporting the findings of this study are available within the article and its supplementary material.
References
- 1. Ali A, Mishra S, Waddell K et al. Rapid Evidence Synthesis: Impacts of Pharmacist Prescribing on the Equity-Centred Quadruple-Aim Outcomes. 2025. https://www.mcmasterforum.org/docs/default-source/product-documents/rapid-responses/impacts-of-pharmacist-prescribing-on-the-equity-centred-quadruple-aim-outcomes.pdf? sfvrsn=2a260fc7_3 (24 February 2026, date last accessed).
- 2. Barton A, Dersch-Mills D, Saunders S et al. Pharmacist prescribing in pediatric and neonatal acute care: an observation study. J Pediatr Pharmacol Ther 2020;25:600–5. 10.5863/1551-6776-25.7.600 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Charlton A, Hill C, Dersch-Mills D et al. Changes in acute care pharmacist prescribing and laboratory ordering over time: CAPLET study. Can J Hosp Pharm 2024;77:e3565. 10.4212/cjhp.3565 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Public Health Wales NHS Trust. Pharmacist Independent Prescribing—A Review of the Evidence. 2015. https://primarycareone.nhs.wales/files/pharmacy-resources/independent-prescribing-a-review-of-the-evidence-2015-pdf (24 February 2026, date last accessed).
- 5. Saunders S, Dersch-Mills D, Mysak T et al. CAPABLE: Calgary zone usage of additional prescribing authorization by pharmacists in an inpatient setting: review of the prescribing landscape and environment. Res Social Adm Pharm 2020;16:342–8. 10.1016/j.sapharm.2019.05.021 [DOI] [PubMed] [Google Scholar]
- 6. Poh EW, McArthur A, Stephenson M et al. Effect of pharmacist prescribing on patient outcomes in the hospital setting: a systematic review. JBI Database Syst Rev Implement Rep 2018;16:1823–73. 10.11124/JBISRIR-2017-003697 [DOI] [PubMed] [Google Scholar]
- 7. Huskanovic I, Le DH, Zhang J et al. Évaluation des répercussions de l’entente de pratique avancée en partenariat sur le travail du pharmacien du CIUSSS du Centre-Ouest-de-l’Île-de-Montréal. Pharmactuel 2025;58:102–8. 10.63209/2025.1588 [DOI] [Google Scholar]
- 8. Ordre des Pharmaciens du Québec. Guide d’exercice: Les activités professionnelles du pharmacien. 2022. https://www.opq.org/wp-content/uploads/2020/12/Guide_exercice_nouv_act_fev_2022.pdf (24 February 2026, date last accessed).
- 9. Heck T, Gunther M, Bresee L et al. Independent prescribing by hospital pharmacists: patterns and practices in a Canadian province. Am J Health Syst Pharm 2015;72:2166–75. 10.2146/ajhp150080 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The authors confirm that the data supporting the findings of this study are available within the article and its supplementary material.