You are here

Article Text

Article menu
Original research
Pharmacy program to improve care for veterans with transient ischaemic attack: a pilot implementation evaluation
  1. Nicholas Rattray1,2,3,
  2. Teresa M Damush1,2,3,
  3. Laura Myers1,2,
  4. Anthony J Perkins2,4,
  5. Barbara Homoya2,
  6. Christopher Knefelkamp5,
  7. Breanne Fleming5,
  8. Andrea Kingsolver5,
  9. Amy Boldt5,
  10. Jared Ferguson2,
  11. Alan Zillich6,
  12. Dawn M Bravata1,2,3
  1. 1Internal Medicine, Indiana University School of Medicine, Indianapolis, Indiana, USA
  2. 2VA HSR&D Precision Monitoring to Transform Care (PRISM) Quality Enhancement Research Initiative, Richard L Roudebush VA Medical Center, Indianapolis, Indiana, USA
  3. 3William M. Tierney Center for Health Services Research, Regenstrief Institute, Inc, Indianapolis, Indiana, USA
  4. 4Department of Biostatistics, Indiana University Purdue University Indianapolis, Indianapolis, Indiana, USA
  5. 5Pharmacy Department, Richard L Roudebush VA Medical Center, Indianapolis, Indiana, USA
  6. 6Department of Pharmacy Practice, Purdue University, West Lafayette, Indiana, USA
  1. Correspondence to Dr Nicholas Rattray; nicholas.rattray{at}va.gov

Abstract

Background Early evaluation and effective communication to manage transient ischaemic attacks (TIA) may lead to a reduction of up to 70% in recurrent events for patients with TIA/minor stroke, along with reduced costs and lengths of hospital stay.

Methods We conducted a single site pilot evaluation of a clinical pharmacy programme to improve medication management among TIA patients. The programme included a structured protocol, online identification tool, and a templated discharge checklist. Primary effectiveness measures were change in systolic blood pressure (SBP) 90 days post discharge and prescription of high/moderate potency statins. Contextual aspects and clinical perspectives on the implementation process were evaluated through prospective semistructured interviews with key informants.

Results The analysis included 75 patients in the preimplementation group and 61 in the postimplementation group. The mean SBP at 90 days post discharge was significantly lower in the post implementation period (pre implementation, 133.3 mm Hg (SD 17.8) vs post implementation, 126.8 mm Hg (16.6); p=0.045). The change in SBP from discharge to 90 days post discharge was greater in the postimplementation period (15.8 mm Hg (20.5) vs 24.8 mm Hg (23.2); p=0.029). The prescription of high/moderate potency statins were similar across groups (pre implementation, 66.7% vs post implementation, 77.4%; p=0.229). Front-line clinicians involved in the pilot study reported positively on the acceptability, appropriateness and feasibility of implementing the protocol without additional cost and within current scope of practice.

Conclusions Implementation of a clinical protocol outlining medication management and provider communication to ensure rapid postdischarge treatment of TIA patients was associated with SBP improvements. The pilot evaluation demonstrates how clinical pharmacists may play a role in treating low frequency, high stakes cerebrovascular events where early treatment and follow-up are critical.

  • Pharmacists
  • Primary care
  • Chronic disease management
  • Quality improvement
  • Communication

Data availability statement

All data relevant to the study are included in the article or uploaded as online supplemental information. The United States Department of Veterans Affairs (VA) places legal restrictions on access to veteran’s health care data, which includes both identifying data and sensitive patient information. The analytic data sets used for this study are not permitted to leave the VA firewall without a Data Use Agreement. This limitation is consistent with other studies based on VA data.

http://creativecommons.org/licenses/by-nc/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.

https://doi.org/10.1136/bmjoq-2022-001863

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    WHAT IS ALREADY KNOWN ON THIS TOPIC

    • Following a transient ischaemic attack (TIA) or minor stroke, medication management and blood pressure control are fundamental to reducing the risk of recurrent cerebrovascular events.

    WHAT THIS STUDY ADDS

    • Few protocols exist for communication between pharmacists and health providers. This study offers pilot evidence about a feasible intervention that guides clinical pharmacists in improving care and outcomes for TIA patients through structured communication.

    HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

    • Clinical pharmacists may play an increasing role is treating low frequency, high-stakes cerebrovascular events where early treatment and follow-up are critical to improving outcomes.

    Introduction

    In the USA, a stroke occurs every 40 s. Among the 800 000 strokes that occur annually,1 approximately 15% are preceded by a transient ischaemic attack (TIA).2 Following a TIA, evidence suggests that a significant proportion of adverse events that follow discharge is drug related and may be preventable.3 4 Because more than half of the recurrent events that occur within 3 months of an index TIA event actually occur in the first 2 days, preventive actions must be applied early to maximise the benefit.5 Programmes that emphasise early evaluation and management may lead to a reduction of up to 70% in recurrent events for patients with TIA or minor stroke, along with reduced costs, reduced lengths of hospital stay, and improved vascular risk factor management.6–8 The extant literature on recurrent TIA events suggests that communication about discharge and follow-up care represent opportunities to improve stroke/TIA outcomes.9 10 Although the initial hospital episode and subsequent outpatient clinic visits are the most favourable opportunities to address vascular risk reduction, appropriate treatment and management of cerebrovascular patients is frequently prolonged until patients receive primary care follow-up, which may be weeks or months later.11–13 Discharge instructions and discharge orders are sometimes vague, inconsistent and incomplete; often little or no care coordination occurs between inpatient and outpatient providers.14

    System level changes in the structure of acute stroke care include the formation of stroke units and/or stroke teams, which have demonstrated improvements in mortality and recovery from stroke.15 Despite these improvements, evidence suggests that pharmacists are underused in transitions of care, particularly in the setting of stroke/TIA management.4 16 17 Importantly, the inclusion of pharmacists in long-term management of secondary prevention measures and multidisciplinary stroke teams can improve patient outcomes.17–26 Yet, in most medical centres, discharge communication specifically about stroke or TIA patients does not typically occur between inpatient pharmacists and the primary care team.14 The Veterans Health Administration is an ideal setting to examine communication between inpatient and outpatient pharmacists because pharmacists are embedded within the primary care clinics and are tasked with managing medications, patient education and care coordination.

    Given the importance of structured approaches to improving stroke outcomes through guideline-driven delivery processes, our team developed a formal protocol for inpatient–outpatient pharmacist TIA care coordination. This single site, pilot evaluation study was designed to assess the implementation of the pharmacy protocol aimed at improving clinical care and communication between inpatient pharmacists and primary care pharmacists involved in caring for TIA patients. The primary aim was to evaluate the efficacy of the programme by comparing TIA patients receiving care before versus after programme implementation. The specific research question was: did the clinical protocol lead to improvement in hypertension control for TIA patients? The second aim was a process evaluation of programme implementation focused on identifying critical elements that promoted successful adoption and determining which ‘core components’ enhanced programme effectiveness.

    Methods

    Study design

    We used a parallel mixed method study design.27 Quantitative data were collected through retrospective chart review of electronic medical record (EMR) data on clinical outcomes for patients with TIA who received care before and after the protocol was implemented. To understand context for the intervention, we conducted a prospective evaluation of the perspectives of key clinical staff. The study protocol was approved by the institutional review board and VA Research and Development Committee. The study draws on the Consolidated Framework for Implementation Research (CFIR)28 to understand the contextual factors that affect how the pharmacy programme was implemented.29 We followed Standards for Quality Improvement Reporting Excellence V.2.0 guidelines.30

    Study setting

    This study was conducted at a VA Medical Center (VAMC) that delivers inpatient and outpatient healthcare services to approximately 200 000 veterans annually. The VAMC is a teaching hospital affiliated with a university medical school. It is a tertiary facility with medical and surgical intensive care, stepdown, inpatient rehabilitation units; one of the acute care areas is designated as the stroke care unit. Neurology is an admitting service with medical residents involved in inpatient care and consults.

    Primary care services are provided through clinics, which are subdivided into patient-aligned care teams (PACT).31 A PACT teamlet is composed of a primary care provider (MD or advanced practice nurse), the nurse case manager (registered nurse) and the health technician (licensed practical nurse). Other healthcare professionals are shared between teamlets, such as clinical pharmacists, social workers and health psychologists. The Veterans Affairs (VA) scope of practice defines the clinical pharmacist’s prescriptive authority, routine duties, areas of responsibility and supervision by a physician; importantly, it includes lipid and hypertension management. PACT pharmacists encounter patients after they are seen by a primary care provider at a postdischarge follow-up visit and referred for hypertension, diabetes, hyperlipidaemia or anticoagulation related to atrial fibrillation.

    Prior to implementing this intervention, pharmacists who were assigned to an inpatient medical team made medication-related recommendations but did not systematically document the recommendation in the EMR. The standard of care for TIA or stroke patients cared for in the emergency department (ED) and discharged did not routinely involve pharmacists. Communication between the inpatient pharmacist and the PACT pharmacist was infrequent and generally reserved for complex patients. When communication between inpatient and outpatient pharmacists occurred, it was typically informal, through email, phone calls or instant message rather than through cosignature of clinical notes. Delays occurred because patients saw primary care providers within 1–2 weeks of discharge, with pharmacy follow-up 4–8 weeks later.

    Description of the intervention

    The pharmacy intervention is part of a programme entitled ‘Protocol-guided Rapid Evaluation of Veterans Experiencing New Transient Neurological Symptoms’,32 which seeks to improve care for TIA patients. The intervention addressed medication management for cerebrovascular disease risk factors including: hypertension, hyperlipidaemia, atrial fibrillation, diabetes and tobacco use. The written protocol was iteratively developed and refined by clinical champions from relevant services (ie, pharmacy, endocrinology, cardiology, vascular neurology, internal medicine). Every domain of care included a decision tree with an evidence-based clinical target and recommendations formatted as a table and as a flowchart (see online supplemental appendix A). The intervention sought to improve coordination of care for TIA patients by establishing communication pathways between inpatient and ambulatory care pharmacists. The protocol defined the roles and responsibilities of the ED, inpatient and PACT pharmacists, lists the PACT pharmacist assigned to each primary care team, provides directions on using an online tool for identifying TIA patients in real time and includes a templated TIA Discharge Checklist for documentation (see online supplemental appendix B).

    Supplemental material

    Supplemental material

    The protocol called for a ‘warm handoff’ between inpatient and PACT teams prior to patient discharge (see figure 1). The inpatient pharmacist used the patient identification tool to identify inpatients with a TIA that were currently admitted or recently discharged and lists the patient’s name, their primary care provider and their PACT pharmacist. The patient identification tool also identified patients who presented to the ED but who were not admitted. The inpatient pharmacist examined the medical record in conjunction with the protocol algorithm for each process of care for which the patient was eligible. The inpatient pharmacist then contacted the PACT pharmacist (generally by secure messaging them or adding them as cosigners to a note; rarely by calling them) with the goal of scheduling an appointment before the patient was discharged. Subsequently, the PACT pharmacist documented their communication in the EMR, scheduled an appointment if one had not already been made before discharge and addressed any further clinical issues (eg, discharging the patient home with a blood pressure cuff).

    Figure 1
    Figure 1

    Diagram of protocol for tracking and communication about TIA patients. EMR, electronic medical record; PACT, patient-aligned care teams; RN, registered nurse; TIA, transient ischaemic attacks.

    Data collection, outcomes and analysis

    Eligible patients were those with an index TIA seen in the ED or inpatient setting at the VAMC from May 2016 through September 2018. The preimplementation phase was May–December 2016; the implementation phase was from January 2017 to September 2018.

    Patient health data

    EMR data elements extracted included status of index event, demographic factors and several relevant conditions in the patients’ medical history (including prior stroke or TIA, medications, comorbidities; see table 1). Blood pressure and cholesterol measurements were extracted, along with key processes of care including discharge on high/moderate potency statins, hypertension control, antihypertensive medication intensification, timeliness of antithrombotic prescriptions, anticoagulation for atrial fibrillation, international normalised ratio measured, deep vein thrombosis (DVT) prophylaxis, glycosylated haemoglobin measurement and hypoglycaemic medication intensification. Healthcare utilisation included primary care and neurology visits within 30 and 90 days of discharge. We included incidence of mortality, stroke or TIA within 90 days of discharge.

    View this table:
    Table 1

    Baseline characteristics

    Outcomes

    Two effectiveness measures were evaluated: (1) the primary outcome was the difference in mean systolic blood pressure (SBP) at 90 days post discharge between the preimplementation and postimplementation groups and (2) the secondary outcome was the proportion of eligible patients who were prescribed high or moderate potency statins within 7 days of discharge. These two care processes were selected because they offered the greatest opportunities for improvement and conformed with existing studies.5 6 A secondary hypertension outcome was the change in SBP from the day of presentation to either the inpatient setting or ED, to the average systolic measurement in the 90 days post discharge. Categorical data were presented as percentages (n) and compared across time periods using Fisher’s exact test.

    Continuous variables were reported as means with SD or ranges. Means were compared with two sample t-tests while medians were tested using the Wilcoxon rank sum test. Two-sided p<0.05 was considered statistically significant and analyses were performed with SAS Enterprise V.7.13. Multivariable regression was used to adjust for differences in the baseline characteristics of the two groups. This pilot evaluation study was not powered to detect differences in patient outcomes; rather data were collected on all consecutive patients cared for at the medical centre during the pilot implementation phase.

    Implementation evaluation

    Behavioural change theories were used to establish working hypotheses that could explain how the communication protocol operated as an intervention. Drawn from organisational development theory, the framework of sense-making33 would suggest when influential leaders endorse new protocol and reframe and model new procedures, clinical staff may consider integrating these new practices into their workflow as they ‘rebalance’ and respond to an intervention that offers a relative advantage.34 Social network theory35 offers evidence that disseminating through existing social networks can aid in implementing new evidence-based programmes. The aim was to examine how existing relationships based on shared training and common scope of practice for inpatient and outpatient pharmacists would affect implementation.

    Eleven semistructured interviews were conducted (March 2017 to January 2018). A convenience sample was recruited for a single in-person interview (duration: 30–60 min). This purposive sample included each available provider involved in the pilot: inpatient and outpatient pharmacists, an ED physician, a neurologist and a nurse. The interview guide focused on key CFIR constructs (domains of intervention characteristics, inner setting and process) and selected implementation outcomes (acceptability, appropriateness, feasibility, fidelity, penetration). Fieldnotes were composed within 24 hours of the interview that summarised key themes and non-verbal behaviour in the interview encounter. Audio-recorded interviews were professionally transcribed. Transcripts were deidentified, checked for accuracy and imported into NVivo V.11 for data management and coding. A team of three analysts carried out an iterative thematic analysis. Open inductive coding in teams of two generated a codebook with definitions and inclusion/exclusion criteria. Refinement of the codebook continued until thematic saturation was reached (eg, no new codes emerged). Subsequently, each analyst separately applied the codes to the full set of interviews and then met to establish consensus on qualitative findings.

    Patient and public involvement

    The research question was aimed at improving TIA patient outcomes. Patients were not directly involved with the design or analysis of the study, primarily because the intervention is focused on pharmacists. Research results are intended to be disseminated through open access publication.

    Results

    Overall, 75 patients were included in the pre implementation group, with 61 in the post implementation group. Table 1 describes patient characteristics, medical history and medication use prior to index event. The majority were on statins and an antithrombotic prior to the index TIA event. The two groups did not differ demographically or in terms of inpatient admission or weekend presentation. There were some between-group differences on elements of medical history (ie, hemiplegia, sleep apnea, depression, history of DVT, and smoking). Except for a higher smoking rate in the preimplementation group, the postimplementation patients had similar or higher comorbidity burden. In both groups, approximately two-thirds of patients were admitted, with a minority presenting on weekends (17.3% pre implementation; 21.3% post implementation).

    Table 2 summarises each group in terms of vital signs, processes of care, utilisation of health services and outcomes. Because the intervention was not designed to change hypertension management in the inpatient setting, the primary hypertension measures were 90 days post discharge. The mean SBP at 90 days post discharge was significantly lower in the pos timplementation period (pre implementation, 133.3 mm Hg (SD 17.8) vs post implementation, 126.8 mm Hg (16.6); p=0.045). The change in SBP was greater in the postimplementation period than the pre implementation period (mean difference in differences 9.0, p=0.029). After adjustment for differences in the baseline characteristics (in table 1) that were either marginally significant or statistically significant (ie, age, race, current smoker, embolism/DVT, depression, hemiplegia, sleep apnea, baseline anticoagulant, baseline statin and baseline warfarin), the results for both the mean SBP and change in SBP between the groups remained essentially unchanged (see online supplemental appendix C). While not significantly different, prescription of high or moderate potency statins and antihypertensive medication intensification tended to be higher in the postimplementation period. (pre implementation, 66.7% vs post implementation, 77.4%; p=0.229). Antithrombotics and DVT prophylaxis were similar across the groups.

    Supplemental material

    View this table:
    Table 2

    Group differences on vital signs, processes of care, utilisation and outcomes

    Implementation outcomes and contextual factors

    Analysis of interviews with front-line clinicians focused on the implementation outcomes of acceptability, appropriateness and feasibility of implementing the protocol. Additional themes that emerged included contextual factors such as culture, learning climate and intervention characteristics. Table 3 summarises participants’ perspectives on implementation outcomes and with representative quotations.

    View this table:
    Table 3

    Qualitative evidence on selected implementation outcomes and contextual factors

    Implementation outcomes

    Clinicians reported satisfaction with how well the protocol matches with local workflow, emphasising the feasibility of implementation. Pharmacists discussed how the protocol fits with current practices, as one pharmacist (P102) explained: “I think it’s pretty straightforward. It very much aligns with our usual job description for those disease states so it hasn’t necessarily put any extra strain on our pharmacists in that department and they’re all very familiar with diabetes, hypertension, smoking cessation management.” Moreover, the protocol gained wide acceptance and later penetration among inpatient and outpatient pharmacists because it was seen as “very concrete with published evidence” (P101). A pharmacy supervisor (P103) expressed how pharmacy could play a crucial role, “TIAs are these low frequency but high stakes conditions where it seems like it is sort of this invisible role to some degree that pharmacists are playing.”

    Pharmacists expressed satisfaction with their close involvement in designing the protocol and having their work ‘respected’. Pharmacists were unconcerned with additional workload and confident that the processes covered in the protocol would fit into their usual management. Pharmacists reported that use of the protocol had penetrated throughout the PACT teams. They indicated the primary modes of communication was through ‘view alerts’, which are EMR alerts that must be read. Other modes of communication required EMR cosigning and frequent instant messaging, which was common practice among facility pharmacists.

    Context

    Key factors that affected how the pilot study was implemented included the specific characteristics of the intervention, the learning climate and culture of the facility. Participants described how the dual format of the protocol in tabular and flowchart presentations made it easier to implement. Further, they described how early involvement in terms of synthesising the evidence and offering feedback on prototypes made them feel uniquely part of the process. Pharmacists appreciated how formal communication process and real-time patient identification tool offered advantage over current, ad-hoc practices. All interviewees emphasised that the medical facility emphasised a culture of quality improvement through reliance on lean techniques,36 and that Quality Improvement (QI) projects were embedded in training programmes for pharmacists. One pharmacist (P108) expressed the appropriateness of the project:

    As a department, [we] are trying to figure out: what can we do to help prevent those readmissions, or you know to meet all of those [facility report card] reports? What can we do to not just do what we’ve been doing for the last 10 years as pharmacists, but how do we step outside the box and look at some of those measures and help improve care for the veterans? (P108)

    Another primary care pharmacist described that awareness of protocol was high, but that the low patient load (2–4 patients/month) meant not all pharmacists used the protocol often. She further described both the importance of helping TIA patients with hypertension but the challenges of working across services:

    I think it seems like a good idea, catching those patients, because I have one right now I just started following, and her blood pressure’s still high. She’s been admitted to the hospital on the outside multiple times, so trying to catch that seems like a really good thing if we can catch it … It’s hard to implement things across multiple avenues of the hospital. (P111)

    During active implementation, a VA-wide policy change occurred that stipulated that medical support assistants must schedule outpatient appointments. This external change technically prevented pharmacists from following the protocol, but as one pharmacist described, they continued to attempt to call patients on the phone to discuss treatment.

    Discussion

    This study adds to the literature the importance of pharmacists in treating vascular disease and hypertension.18 25 37 Results suggest that this pilot clinical programme focused on deploying pharmacists to provide medication management for patients with TIA lead to improved hypertension control. These results align with other studies in hypertension and hyperlipidaemia management where the timely addition of a pharmacist within a team-based care model is associated with significant improvement in clinical outcomes.37 38 Several systematic reviews confirm the efficacy of the addition of a pharmacist to team-based care in conditions such as diabetes, hypertension and hyperlipidaemia.39–41

    Few prior studies about pharmacist care for vascular disease offer a detailed evaluation of how contextual elements may interact during programme implementation that includes clinicians’ perspective as well as data on clinical effectiveness. Evaluation of qualitative data on implementation indicated that front-line clinicians and leadership found the programme to be an appropriate, feasible and efficient use of existing resources. Due to shared professional norms and common training experience, inpatient and primary care pharmacists are well-positioned within local social networks to communicate in a timely, accurate way about TIA patients as they transition through the hospital and into the outpatient setting. The standardised procedures enabled improved consensus of treatment and monitoring through co-signing of electronic health records, real-time messaging and the scheduling of follow-up visits during or immediately after hospitalisation.

    The capability to detect at-risk patients in near-real time is a welcome development for conditions such as TIA that require rapid identification and treatment. However, distinct challenges remain with using information not documented in the EMR or related technological tools including the challenge of recording decision-making that takes place verbally or through other information channels.42 Evidence suggests that deliberate action may be necessary to effectively track patients as they transition through hospital services, a practice that has been labelled ‘chart stalking’.43 In this pilot programme, designated clinicians (inpatient pharmacist, internist, nurse, all with experience working with the neurology service) closely monitored all admitted TIA patients during and after hospital discharge.

    The team sought to improve early post-TIA management in order to improve outcomes assessed in the 90-day postevent period in order to promote guideline-based medication prescription as soon as possible after the index–TIA event and to provide continuity in care as patients transitioned from the ED or inpatient setting into the outpatient setting. The two key processes of care varied in terms of the time period over which they were assessed: SBP at 90 days post discharge, and the prescription of high or moderate potency statins within 7 days of discharge. The blood pressure metric was assessed later post TIA for two main reasons: (a) clinical concerns about blood pressure lowering in the acute event period are reflected in current guidelines which emphasise getting patients to goal blood pressure only after the acute event period and (b) for patients with poorly controlled blood pressure post discharge, stepwise approaches to increasing antihypertensive intensity require some time to be implemented and to be reflected in clinic-based blood pressure measurements. In other words, the goal of the programme—with regard to hypertension management—was to meet goal blood pressure targets as quickly as possible post discharge. Given that primary care pharmacy visits are often conducted via telephone, the intention was for the pharmacists to engage with patients in the postdischarge period. However, only clinic-based blood pressure measurements are considered as ‘vital signs’ within the VA data systems. Therefore, the process measure was assessed over the 90-day period allowing patients time to return for follow-up assessments.

    Although there was an observed improvement in SBP in the 90 days post discharge after implementation of the programme, a statistically significant change in the prescription of guideline-concordant high or moderate potency statins at discharge was not observed. To observe significant changes, 138 patients per group would have been needed to see a difference in rates between 65% and 80%. Most TIA patients were on a low-potency statin prior to the index event and it may have been that clinical inertia (not wanting to change an existing medication)44 or patient preferences (eg, wanting to stay on the medication that they perceived was ‘working’ for them) contributed to a lack of a statistically significant increase in the proportion of patients who received high/moderate potency statins. Clinical inertia is well documented with physician providers but less understood with pharmacists. In addition, a reluctance to prescribe statins for patients over 75 years, which was typically clinical decision based on assessment of risks and benefits may have contributed to the observed results.

    There were several limitations in this single site quality improvement pilot. First, a relatively small sample size of patients as well as the limited number of clinicians working in this area limit the ability to make generalizable claims about these findings. Second, this facility had a well-development culture of quality improvement and infrastructure to support cerebrovascular care, therefore findings may not extend to settings with different quality improvement cultures. Third, observational design limits inferences about causal relationships outcomes and pharmacist behaviour involved in the intervention. Fourth, although age, blood pressure and history of diabetes were available on all patients, the ABCD2 (age, blood pressure, clinical features, duration, diabetes) score could not be calculated because clinical features and symptom duration were not available in the electronic health record data. Finally, large artery atherosclerotic aetiology and capsular warning syndrome have been associated with increased risk of early recurrence among patients with TIA45; however, our dataset did not include event aetiology. Given that risk factor management may play a differential role based on event type (eg, hypertension management among patients with lacunar events), future research should explicitly examine the differential benefits of medical management among TIA patients with varying event aetiologies.

    Early participation in programme development and leadership involvement contributed to wide penetration in outpatient clinics throughout the facility, demonstrating how clinical pharmacists may play an increasing role is treating low frequency, high-stakes cerebrovascular events where early treatment and follow-up are critical in improving outcomes.

    Data availability statement

    All data relevant to the study are included in the article or uploaded as online supplemental information. The United States Department of Veterans Affairs (VA) places legal restrictions on access to veteran’s health care data, which includes both identifying data and sensitive patient information. The analytic data sets used for this study are not permitted to leave the VA firewall without a Data Use Agreement. This limitation is consistent with other studies based on VA data.

    Ethics statements

    Patient consent for publication

    Ethics approval

    This study involves human participants. This study was approved by the Indiana University Institutional Review Board (approval #1602800879). All participants gave consent for participation in this study. Participants gave informed consent to participate in the study before taking part.

    Acknowledgments

    The authors would like to acknowledge the time donated by providers who participated in the quality improvement evaluation, and special thanks to Ariel Chatham and Lauren Murphy with data collection. We thank Ali Sexson, Sean Baird and Jennifer Myers for assistance with manuscript preparation.

    References

    1. ↵
      1. Benjamin EJ,
      2. Blaha MJ,
      3. Chiuve SE, et al
      . Heart disease and stroke Statistics-2017 update: a report from the American heart association. Circulation 2017;135:e146–603.doi:10.1161/CIR.0000000000000485pmid:http://www.ncbi.nlm.nih.gov/pubmed/28122885
      OpenUrlFREE Full Text
    2. ↵
      1. Kernan WN,
      2. Ovbiagele B,
      3. Black HR, et al
      . Guidelines for the prevention of stroke in patients with stroke and transient ischemic attack. Stroke 2014;45:2160–236.doi:10.1161/STR.0000000000000024
      OpenUrlAbstract/FREE Full Text
    3. ↵
      1. Graabaek T,
      2. Kjeldsen LJ
      . Medication reviews by clinical pharmacists at hospitals lead to improved patient outcomes: a systematic review. Basic Clin Pharmacol Toxicol 2013;112:359–73.doi:10.1111/bcpt.12062pmid:http://www.ncbi.nlm.nih.gov/pubmed/23506448
      OpenUrlPubMed
    4. ↵
      1. American College of Clinical Pharmacy,
      2. Hume AL,
      3. Kirwin J, et al
      . Improving care transitions: current practice and future opportunities for pharmacists. Pharmacotherapy 2012;32:e326-37.doi:10.1002/phar.1215pmid:http://www.ncbi.nlm.nih.gov/pubmed/23108810
      OpenUrlPubMed
    5. ↵
      1. Johnston SC,
      2. Gress DR,
      3. Browner WS, et al
      . Short-term prognosis after emergency department diagnosis of TIA. JAMA 2000;284:2901–6.doi:10.1001/jama.284.22.2901pmid:http://www.ncbi.nlm.nih.gov/pubmed/11147987
      OpenUrlCrossRefPubMedWeb of Science
    6. ↵
      1. Rothwell PM,
      2. Giles MF,
      3. Chandratheva A, et al
      . Effect of urgent treatment of transient ischaemic attack and minor stroke on early recurrent stroke (express study): a prospective population-based sequential comparison. Lancet 2007;370:1432–42.doi:10.1016/S0140-6736(07)61448-2pmid:http://www.ncbi.nlm.nih.gov/pubmed/17928046
      OpenUrlCrossRefPubMedWeb of Science
    7. ↵
      1. Kehdi EE,
      2. Cordato DJ,
      3. Thomas PR, et al
      . Outcomes of patients with transient ischaemic attack after hospital admission or discharge from the emergency department. Med J Aust 2008;189:9–12.doi:10.5694/j.1326-5377.2008.tb01886.xpmid:http://www.ncbi.nlm.nih.gov/pubmed/18601633
      OpenUrlPubMed
    8. ↵
      1. Lavallée PC,
      2. Meseguer E,
      3. Abboud H, et al
      . A transient ischaemic attack clinic with round-the-clock access (SOS-TIA): feasibility and effects. Lancet Neurol 2007;6:953–60.doi:10.1016/S1474-4422(07)70248-Xpmid:http://www.ncbi.nlm.nih.gov/pubmed/17928270
      OpenUrlCrossRefPubMedWeb of Science
    9. ↵
      1. Bushnell C,
      2. Zimmer L,
      3. Schwamm L, et al
      . The adherence eValuation after ischemic stroke longitudinal (AVAIL) registry: design, rationale, and baseline patient characteristics. Am Heart J 2009;157:428–35.doi:10.1016/j.ahj.2008.11.002pmid:http://www.ncbi.nlm.nih.gov/pubmed/19249411
      OpenUrlCrossRefPubMedWeb of Science
    10. ↵
      1. Bushnell CD,
      2. Olson DM,
      3. Zhao X, et al
      . Secondary preventive medication persistence and adherence 1 year after stroke. Neurology 2011;77:1182–90.doi:10.1212/WNL.0b013e31822f0423pmid:http://www.ncbi.nlm.nih.gov/pubmed/21900638
      OpenUrlCrossRefPubMed
    11. ↵
      1. Ovbiagele B,
      2. Saver J
      . Ensuring management of vascular risk factors after stroke. Rev Neurol Dis 2005;3:93–100.pmid:17047574
      OpenUrlPubMed
    12. ↵
      1. Ovbiagele B
      . Pessin Award Lecture 2008: lessons from the stroke protect program. J Neurol Sci 2008;275:1–6.doi:10.1016/j.jns.2008.07.014pmid:http://www.ncbi.nlm.nih.gov/pubmed/18722629
      OpenUrlCrossRefPubMedWeb of Science
    13. ↵
      1. Coull AJ,
      2. Lovett JK,
      3. Rothwell PM, et al
      . Population based study of early risk of stroke after transient ischaemic attack or minor stroke: implications for public education and organisation of services. BMJ 2004;328:326.doi:10.1136/bmj.37991.635266.44pmid:http://www.ncbi.nlm.nih.gov/pubmed/14744823
      OpenUrlAbstract/FREE Full Text
    14. ↵
      1. Bayley KB,
      2. Savitz LA,
      3. Rodriguez G
      . Barriers associated with medication information handoffs, 2005.
    15. ↵
      1. Kwan J,
      2. Sandercock P
      . In-hospital care pathways for stroke. Cochrane Database Syst Rev 2004:Cd002924.doi:10.1002/14651858.CD002924.pub2pmid:http://www.ncbi.nlm.nih.gov/pubmed/15495038
    16. ↵
      1. Myers J,
      2. Bravata DM,
      3. Sico J, et al
      . The quality of medication optimization among patients with transient ischemic attack or minor stroke. J Am Coll Clin Pharm 2020;3:36–46.doi:10.1002/jac5.1149
      OpenUrl
    17. ↵
      1. Schnipper JL,
      2. Kirwin JL,
      3. Cotugno MC, et al
      . Role of pharmacist counseling in preventing adverse drug events after hospitalization. Arch Intern Med 2006;166:565–71.doi:10.1001/archinte.166.5.565pmid:http://www.ncbi.nlm.nih.gov/pubmed/16534045
      OpenUrlCrossRefPubMedWeb of Science
    18. ↵
      1. Schmid AA,
      2. Kapoor JR,
      3. Miech EJ, et al
      . A multidisciplinary stroke clinic for outpatient care of veterans with cerebrovascular disease. J Multidiscip Healthc 2011;4:111.doi:10.2147/JMDH.S17154pmid:http://www.ncbi.nlm.nih.gov/pubmed/21594062
      OpenUrlPubMed
    19. ↵
      1. Allen KR,
      2. Hazelett S,
      3. Jarjoura D, et al
      . Effectiveness of a postdischarge care management model for stroke and transient ischemic attack: a randomized trial. J Stroke Cerebrovasc Dis 2002;11:88–98.doi:10.1053/jscd.2002.127106pmid:http://www.ncbi.nlm.nih.gov/pubmed/17903862
      OpenUrlCrossRefPubMed
    20. ↵
      1. Carter BL,
      2. Rogers M,
      3. Daly J, et al
      . The potency of team-based care interventions for hypertension: a meta-analysis. Arch Intern Med 2009;169:1748–55.doi:10.1001/archinternmed.2009.316pmid:http://www.ncbi.nlm.nih.gov/pubmed/19858431
      OpenUrlCrossRefPubMedWeb of Science
    21. ↵
      1. Kwan J,
      2. Sandercock P
      . In-hospital care pathways for stroke: a cochrane systematic review. Stroke 2003;34:587–8.doi:10.1161/01.STR.0000054673.28010.1Bpmid:http://www.ncbi.nlm.nih.gov/pubmed/12574581
      OpenUrlFREE Full Text
    22. ↵
      1. Dow RS,
      2. Dick HL,
      3. Crowell FA
      . Failures and successes in a stroke program. Stroke 1974;5:40–7.doi:10.1161/01.STR.5.1.40pmid:http://www.ncbi.nlm.nih.gov/pubmed/4811310
      OpenUrlAbstract/FREE Full Text
    23. ↵
      1. Hankey GJ,
      2. Warlow CP
      . Treatment and secondary prevention of stroke: evidence, costs, and effects on individuals and populations. Lancet 1999;354:1457–63.doi:10.1016/S0140-6736(99)04407-4pmid:http://www.ncbi.nlm.nih.gov/pubmed/10543686
      OpenUrlCrossRefPubMedWeb of Science
    24. ↵
      1. Stroke Unit Trialists' Collaboration
      . Organised inpatient (stroke unit) care for stroke. Cochrane Database Syst Rev 2013;2013:CD000197.doi:10.1002/14651858.CD000197.pub3pmid:http://www.ncbi.nlm.nih.gov/pubmed/24026639
      OpenUrlPubMed
    25. ↵
      1. Zillich AJ,
      2. Jaynes HA,
      3. Bex SD, et al
      . Evaluation of pharmacist care for hypertension in the Veterans Affairs patient-centered medical home: a retrospective case-control study. Am J Med 2015;128:539.e1–539.e6.doi:10.1016/j.amjmed.2014.11.027pmid:http://www.ncbi.nlm.nih.gov/pubmed/25534422
      OpenUrlCrossRefPubMed
    26. ↵
      1. Fischer MA,
      2. Choudhry NK,
      3. Bykov K, et al
      . Pharmacy-based interventions to reduce primary medication nonadherence to cardiovascular medications. Med Care 2014;52:1050–4.doi:10.1097/MLR.0000000000000247pmid:http://www.ncbi.nlm.nih.gov/pubmed/25322157
      OpenUrlPubMed
    27. ↵
      1. Creswell JW,
      2. Plano Clark VL
      . Designing and conducting mixed methods research. Los Angeles: SAGE Publications, 2011.
    28. ↵
      1. Damschroder LJ,
      2. Aron DC,
      3. Keith RE, et al
      . Fostering implementation of health services research findings into practice: a consolidated framework for advancing implementation science. Implement Sci 2009;4:50.doi:10.1186/1748-5908-4-50pmid:http://www.ncbi.nlm.nih.gov/pubmed/19664226
      OpenUrlCrossRefPubMed
    29. ↵
      1. Proctor E,
      2. Silmere H,
      3. Raghavan R, et al
      . Outcomes for implementation research: conceptual distinctions, measurement challenges, and research agenda. Adm Policy Ment Health 2011;38:65–76.doi:10.1007/s10488-010-0319-7pmid:http://www.ncbi.nlm.nih.gov/pubmed/20957426
      OpenUrlCrossRefPubMed
    30. ↵
      1. Ogrinc G,
      2. Davies L,
      3. Goodman D, et al
      . Squire 2.0 (standards for quality improvement reporting excellence): revised publication guidelines from a detailed consensus process. BMJ Qual Saf 2016;25:986–92.doi:10.1136/bmjqs-2015-004411pmid:http://www.ncbi.nlm.nih.gov/pubmed/26369893
      OpenUrlAbstract/FREE Full Text
    31. ↵
      1. Jackson GL,
      2. Powers BJ,
      3. Chatterjee R, et al
      . The patient centered medical home. A systematic review. Ann Intern Med 2013;158:169–78.doi:10.7326/0003-4819-158-3-201302050-00579pmid:http://www.ncbi.nlm.nih.gov/pubmed/24779044
      OpenUrlCrossRefPubMedWeb of Science
    32. ↵
      1. Bravata DM,
      2. Myers LJ,
      3. Homoya B, et al
      . The protocol-guided rapid evaluation of Veterans experiencing new transient neurological symptoms (prevent) quality improvement program: rationale and methods. BMC Neurol 2019;19:294.doi:10.1186/s12883-019-1517-xpmid:http://www.ncbi.nlm.nih.gov/pubmed/31747879
      OpenUrlPubMed
    33. ↵
      1. Weick KE,
      2. Quinn RE
      . Organizational change and development. Annu Rev Psychol 1999;50:361–86.doi:10.1146/annurev.psych.50.1.361pmid:http://www.ncbi.nlm.nih.gov/pubmed/15012461
      OpenUrlCrossRefPubMedWeb of Science
    34. ↵
      1. Gustafson DH,
      2. Sainfort F,
      3. Eichler M, et al
      . Developing and testing a model to predict outcomes of organizational change. Health Serv Res 2003;38:751–76.doi:10.1111/1475-6773.00143pmid:http://www.ncbi.nlm.nih.gov/pubmed/12785571
      OpenUrlCrossRefPubMedWeb of Science
    35. ↵
      1. Valente TW,
      2. Palinkas LA,
      3. Czaja S, et al
      . Social network analysis for program implementation. PLoS One 2015;10:e0131712.doi:10.1371/journal.pone.0131712pmid:http://www.ncbi.nlm.nih.gov/pubmed/26110842
      OpenUrlPubMed
    36. ↵
      1. Mazzocato P,
      2. Savage C,
      3. Brommels M, et al
      . Lean thinking in healthcare: a realist review of the literature. Qual Saf Health Care 2010;19:376–82.doi:10.1136/qshc.2009.037986pmid:http://www.ncbi.nlm.nih.gov/pubmed/20724397
      OpenUrlAbstract/FREE Full Text
    37. ↵
      1. Smith MC,
      2. Boldt AS,
      3. Walston CM, et al
      . Effectiveness of a pharmacy care management program for veterans with dyslipidemia. pharmacotherapy: the Journal of human pharmacology and drug therapy. Pharmacotherapy: The Journal of Human Pharmacology and Drug Therapy2013;33:736–43.doi:10.1002/phar.1273
      OpenUrl
    38. ↵
      1. Flodgren G,
      2. Parmelli E,
      3. Doumit G, et al
      . Local opinion leaders: effects on professional practice and health care outcomes. Cochrane Database Syst Rev 2011;8:CD000125.doi:10.1002/14651858.CD000125.pub4pmid:http://www.ncbi.nlm.nih.gov/pubmed/21833939
      OpenUrlPubMed
    39. ↵
      1. Chisholm-Burns MA,
      2. Kim Lee J,
      3. Spivey CA, et al
      . US pharmacists' effect as team members on patient care: systematic review and meta-analyses. Med Care 2010;48:923–33.doi:10.1097/MLR.0b013e3181e57962pmid:http://www.ncbi.nlm.nih.gov/pubmed/20720510
      OpenUrlCrossRefPubMedWeb of Science
    40. ↵
      1. Proia KK,
      2. Thota AB,
      3. Njie GJ, et al
      . Team-based care and improved blood pressure control: a community guide systematic review. Am J Prev Med 2014;47:86–99.doi:10.1016/j.amepre.2014.03.004pmid:http://www.ncbi.nlm.nih.gov/pubmed/24933494
      OpenUrlCrossRefPubMed
    41. ↵
      1. Chisholm-Burns MA,
      2. Graff Zivin JS,
      3. Lee JK, et al
      . Economic effects of pharmacists on health outcomes in the United States: a systematic review. Am J Health Syst Pharm 2010;67:1624–34.doi:10.2146/ajhp100077pmid:http://www.ncbi.nlm.nih.gov/pubmed/20852164
      OpenUrlAbstract/FREE Full Text
    42. ↵
      1. Kirkendall ES,
      2. Ni Y,
      3. Lingren T, et al
      . Data challenges with real-time safety event detection and clinical decision support. J Med Internet Res 2019;21:e13047.doi:10.2196/13047pmid:http://www.ncbi.nlm.nih.gov/pubmed/31120022
      OpenUrlPubMed
    43. ↵
      1. Bowen JL,
      2. O'Brien BC,
      3. Ilgen JS, et al
      . Chart stalking, list making, and physicians' efforts to track patients' outcomes after transitioning responsibility. Med Educ 2018;52:404–13.doi:10.1111/medu.13509pmid:http://www.ncbi.nlm.nih.gov/pubmed/29383741
      OpenUrlPubMed
    44. ↵
      1. Phillips LS,
      2. Branch WT,
      3. Cook CB, et al
      . Clinical inertia. Ann Intern Med 2001;135:825–34.doi:10.7326/0003-4819-135-9-200111060-00012pmid:http://www.ncbi.nlm.nih.gov/pubmed/11694107
      OpenUrlCrossRefPubMedWeb of Science
    45. ↵
      1. Foschi M,
      2. Pavolucci L,
      3. Rondelli F, et al
      . Prospective observational cohort study of early recurrent TIA: features, frequency, and outcome. Neurology 2020;95:e1733–44.doi:10.1212/WNL.0000000000010317pmid:http://www.ncbi.nlm.nih.gov/pubmed/32727840
      OpenUrlPubMed

    Supplementary materials

    Footnotes

    • Contributors All authors made significant contributions to the manuscript. NR, TMD, AJP, LM and DMB analysed and interpreted data. NR and DMB drafted and revised the manuscript and prepared tables and figures with TMD, AJP, LM, DMB, BH and AJZ. CK, BF, AK, AB, JF, AZ, TMD and DMB made important intellectual contributions in the study design and in feedback on results. DMB is the guarantor of the article. All authors have read and gave final approval of the version submitted for publication.

    • Funding This work was supported by the Department of Veterans Affairs (VA), Health Services Research & Development Service (HSRD), Precision Monitoring to Transform Care (PRISM) and Quality Enhancement Research Initiative (QUERI) (QUE 15-280) and in part by the VA HSRD Research Career Scientist Award (Damush RCS 19-002) and Rattray Career Development Award-2 (Rattray D2642-W). The funding agencies had no role in the design of the study, data collection, analysis, interpretation or in the writing of this manuscript.

    • Competing interests None declared.

    • Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

    • Provenance and peer review Not commissioned; externally peer reviewed.

    • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.