Investigating strategies used by hospital pharmacists to effectively communicate with patients during medication counselling

Abstract Background Medication counselling opportunities are key times for pharmacists and patients to discuss medications and patients’ concerns about their therapy. Communication Accommodation Theory (CAT) describes behavioural, motivational and emotional processes underlying communication exchanges. Five CAT strategies (approximation, interpretability, discourse management, emotional expression and interpersonal control) permit identification of effective communication. Objective To invoke CAT to investigate communication strategies used by hospital pharmacists during patient medication counselling. Design This was a theory‐based, qualitative study using transcribed audiorecordings of patients and hospital pharmacists engaged in medication counselling. Setting and participants Recruited pharmacists practised in inpatient or outpatient settings. Eligible patients within participating pharmacists’ practice sites were prescribed at least three medications to manage chronic disease(s). Main outcome measures The extent to which pharmacists accommodate, or not, to patients’ conversational needs based on accommodative behaviour described within CAT strategies. Results Twelve pharmacists engaged four patients (48 total interactions). Exemplars provided robust examples of pharmacists effectively accommodating or meeting patients’ conversational needs. Non‐accommodation mainly occurred when pharmacists spoke too quickly, used terms not understood by patients and did not include patients in the agenda‐setting phase. Multiple strategy use resulted in communication patterns such as “information‐reassurance‐rationale” sandwiches. Discussion and conclusions Most pharmacists effectively employed all five CAT strategies to engage patients in discussions. Pharmacists’ communication could be improved at the initial agenda‐setting phase by asking open‐ended questions to invite patients’ input and allow patients to identify any medication‐related concerns or issues.

tal pharmacist-patient interactions and the identification of aspects of effective communication. 25,32,33 When speakers appropriately adopt these strategies, studies have shown that interactants rate the communication as effective. 34 CAT research typically dichotomizes outcomes as being either accommodative or non-accommodative.
Accommodative behaviours reduce communication barriers between those interacting whereas non-accommodation behaviours tend to create linguistic distances between speakers. 31,33 Approximation concerns how individuals adjust their speech patterns such as the pitch, rate, volume, tone, use of dialect or accents to converge towards or diverge from their partner's speech. 35,36 Appropriate approximation occurs when speakers perceive that their speech patterns complement each other. Interpretability strategies focus on each speaker's conversational competence. Interactants who modulate their language and word choice to ensure their words are understood demonstrate appropriate interpretability. While interpretability highlights conversation content, discourse management strategies involve conversation processes to promote conversation between interactants. Appropriate discourse management strategies accomplish this through turn-taking, changing topics as needed, responding to non-verbal cues and using conversational repair such as face maintenance that allows patients to maintain a positive selfimage and prevents interactions from becoming ineffective or negative. 25,36 Accommodative emotional expression takes place when pharmacists provide appropriate levels of reassurance and empathy in response to patients' emotional needs. 32 Interpersonal control focuses on the roles and power relations between speakers. Appropriate interpersonal control strategies used by pharmacists promote equality between themselves and patients, not constraining patients to passive patient roles. 25 However, power imbalances between patients and pharmacists will always be inherent, because pharmacists hold specialized knowledge and information not always accessable to patients. These relationships between health-care providers and patients have been described by researchers in health-care communication as intergroup. 25,37 Specifically, pharmacists self-identify as members of a "health-care professional" group while those they counsel belong to the "patient" group. CAT proposes that although health-care professionals may move between these intergroup identities (social and professional) and personal identities (individual likes and dislikes), it is often their intergroup identities that are most salient. 38 Invoking CAT provides a vehicle for interpreting the detailed patterns and flow of pharmacist-patient conversations and will help identify occasions of accommodation or non-accommodation. These may highlight areas of strength as well as areas in which pharmacists' communication skills require further development to improve the effectiveness of their exchanges with patients.
This research represents a portion of a PhD project that will include perspectives of pharmacists and patients about their shared medication counselling experience. However, this part of the study will focus on the observed pharmacist-patient interactions only. The aim of this study is to invoke CAT to investigate communication strategies used by hospital pharmacists during patient medication counselling.

| Recruitment and data collection
This study took place at a large 1000 bed tertiary teaching hospital with multiple specialties including inpatient wards and outpatient clinics.

| Inclusion criteria
Pharmacists: Interested hospital pharmacists whose current professional duties included the provision of clinical pharmacy services in an inpatient and/or outpatient setting.
Patients: Interested patients admitted to either an inpatient or outpatient setting in which clinical pharmacy services were provided by a pharmacist participating in this study; Patients who have been prescribed three or more medications to manage a chronic disease

| Recruitment
Pharmacists were recruited first. An electronic invitation was sent to approximately 50 departmental pharmacists. Eight pharmacists expressed interest initially, and the remaining four pharmacists responded following a reminder email sent 3 weeks later. Pharmacists' demographics were examined and compared to those from a national demographic survey to ensure an accurate representation. 40 Because the pharmacists' ages, qualifications, experience and practice site reflected demographics from within the hospital and nationally, further purposive sampling was not required.
A convenience sample of interested patients meeting the inclusion criteria were first identified by their nurse and then approached by BC who provided study details, obtained consent and gathered patient demographic data.

| Data collection
Participating hospital pharmacists and patients were audiorecorded while engaged in medication counselling. Each pharmacist counselled four different patients to observe how they used and adjusted their communication skills to meet the conversational needs of unique individuals. BC observed the pharmacist-patient interactions in person and out of the direct view of the participants to record non-verbal communication taking place between the participants and to make notes about environmental conditions that might have affected the quality of their exchange. Prior to conducting audiorecorded medication counselling sessions, BC accompanied pharmacists interacting with patients on their ward. This was an attempt to normalize BC's presence and minimize its effect on pharmacists' performances during audiorecorded sessions. As well, pharmacists were told that the study's aim was to investigate whether effective communication takes place between hospital pharmacists and patients; however, pharmacists were intentionally not informed about CAT behaviours to be observed during medication counselling.

| Data analysis and coding
Both patient and pharmacist demographic information and questionnaire results were descriptively analysed. The actual names of pharmacists were not used, but were replaced by pseudonyms that appear in exemplars provided.
Audio recordings were transcribed verbatim and verified by comparing transcripts with original audiorecordings to reconcile any discrepancies, wherever possible. Transcripts were selectively coded for the five CAT strategies in pattern-based discourse analysis as described in the literature. 41 Samples of selectively coded transcripts and their corresponding audiorecordings were checked by co-researcher (BW) to ensure appropriate and consistent interpretation. Transcripts were coded twice, initially grouped according to study pharmacist and then rechecked and recoded, if necessary, in the order of patient enrolment into the study. Audiorecordings were referenced to verify correct interpretation of tone and intent of dialogue. NVivo ® software was used to assist in code organization. Observational field notes were collected during audiorecorded medication counselling sessions and reflexive note taking occurred throughout the study.

| Reflexivity
The main researcher (BC) conducted reflexive note taking throughout the study. BC is a Caucasian female with more than 20 years' experience as a hospital pharmacist in Canada. She holds strong opinions about best practice and professionalism and is aware of how these could possibly influence her perception and interpretation of the data as a researcher. As well, being aware of BC's experience as a hospital pharmacist may have made pharmacists more conscious or nervous having a colleague observe their practice, and some may have felt that their skills were being judged or critiqued. BC had not previously worked with any of the participating pharmacists.

| Participant characteristics
Study pharmacists' demographic characteristics are shown in Table 1.
These are fairly consistent with demographics reported for Australia except more study pharmacists had postgraduate training (58% vs 26%). 40 Table 2 includes participating patient demographic characteristics. Compared to overall hospital age and gender demographics obtained for the study period, the study patients were older (average age of 63.1 vs 52 years of age) and included fewer female study patients (44% vs 53%).

| Discourse analyses-Hospital pharmacists' utilization of CAT strategies
The following analyses examines how pharmacists accommodate, or not, to patients' conversational needs during medication counselling based on the five CAT strategies.

| Approximation
Study pharmacists demonstrated accommodative approximation behaviours by matching patients' speech patterns (volume, pace and accent). Pharmacists occasionally used colloquialisms when speaking to patients, likely to put patients at ease and to encourage engagement in the conversation. Phrases used were often greetings such as "G'day again…" or when conversations were winding down, for example "That's alright mate. All the best hey." Some pharmacists used expressions such as "Righto" when breaks occurred in the conversation. One pharmacist appropriately accommodated a patient by matching their accent after establishing a shared UK origin.
With a few exceptions, pharmacists spoke clearly and with adequate volume and pace to ensure patient understanding. This was based on observing patients' nod or making utterances to indicate they were following pharmacists' explanation. However, a few pharmacists spoke too quickly, potentially not accommodating patients' conversational needs. In rare occasions, underaccommodated patients were observed to respond to the excessive speech rate by disconnecting or engaging less with pharmacists. Interestingly, these pharmacists were aware of their behaviour but identified different reasons for doing so. One pharmacist attributed her rapid speech to her highly energetic nature, whereas two others remarked that time constraints and pressure to discharge patients necessitated speaking quickly as it served to prevent conversations from straying off topic.
Sometimes pharmacists varied their speech rate depending on the focus of conversation. They used a faster pace when discussing interpersonal issues, but a slightly slower, more thoughtful and pronounced pace when emphasizing key points in therapy such as sideeffect management.

| Interpretability
Most pharmacists adopted appropriate interpretability strategies using easy-to-understand phrasing, explained rationale for medication therapy with fairly simple terms and effectively worded questions to elicit patient responses. Patient-specific medication lists were used by all pharmacists as visual guides to support their verbal explanations and often arranged to have medications at hand to emphasize their identity or demonstrate their use. Many pharmacists also used hand gestures or drew cartoon diagrams to indicate how medications work in the body.
In the following example, the pharmacist displayed appropriate interpretability using everyday language to describe how a patient might experience muscle weakness and soreness associated with certain cholesterol lowering agents: At times, pharmacists used medical jargon and terminology, but provided an explanation immediately afterwards, for example: "… in terms of analgesia or your pain relief …" Occasionally pharmacists demonstrated inappropriate interpretability strategies by using terminology that was not understood by patients who responded with quizzical facial expressions or directly asked for the term's meaning. In this example, the pharmacist used the term "neuralgia" as opposed to "nerve pain." "So this is the one that the pain team have suggested… with the kind of neuralgia that you've got." (Christine) Pharmacists typically adjusted the terminology they used in their interactions to match that understood by the patient. In this exchange, an inpatient pharmacist intentionally and appropriately used medical terms to describe the patient's heart condition to a patient, who was also a retired nurse.
"…the next one…Coversyl or Perindopril. That one… it is used for high blood pressure but it's also just used after cardiac events, to protect your cardiovascular system…" (Ingrid) Remind me what dose that one is."

| Discourse management
Many pharmacists also engaged patients by incorporating pauses at different points throughout the delivery of medication information with phrases such as "alright?" or "okay?". This not only verified patients' understanding, but also gave patients an opportunity to interject or ask questions about the information.
Sometimes pharmacists used conversation repair strategies such as face maintenance as demonstrated in the dialogue below by Ellen, an inpatient pharmacist, who down-played a mistake made by a patient to allow him to "save face." Patient: That is supposed to be two lots, not one! I was supposed to get two lots of 20 mL, once in the morning, two lots of 20 mL, twice a day, for the Kapanol.

| Interpersonal control
Numerous accommodative interpersonal control strategies were used by study pharmacists to promote equality between themselves and patients, and to empower or give patients agency, encouraging them to be active participants in their health care. However, pharmacists displayed a number of behaviours to promote patient autonomy and participation in their medication management. One strategy used by pharmacists was to incorporate the rationale for treatment with information and advice provided. This delivery method resembled an information-reassurance-rationale sandwich where medication information was followed by reassurance and then explanations for any medication changes made.
Here, information relates to "interpretability," reassurance to "emotional expression" and "rationale" is the "interpersonal control" CAT strategy.
An example where rationale was provided to encourage selfmonitoring is as follows: Pharmacists in this study were not observed to exaggerate their speech in this manner. Alterations in their tone and rate were subtle and made to emphasize key medication points.
In terms of interpretability, study pharmacists typically were conscientious about choosing language and terminology that was understood by patients. This is in contrast to findings from researchers who reported pharmacists frequently used medical terms not comprehended by patients during medication counselling. 16,17,22,43 Pharmacists adopted a number of accommodative discourse management strategies to engage patients in conversation. By interspersing questions such as "Okay?" or "Alright?" throughout the information provided, pharmacists gave patients an opportunity to interject or ask questions about the information as well as to check for patient understanding. Greenhill et al. depicted this activity as "chunk" and "check." However, they found that their study pharmacists delivered "chunks" of information, but did not check for patients' understanding. 22 Researchers studying pharmacists counselling patients in South African HIV clinics termed this as "response solicitations" where pharmacists used these prompts to both invite patients to ask questions and check their understanding. 44 Using open-ended questions was another technique utilized by our study pharmacists and identified by other researchers as an effective method to elicit patient engagement and highlight any patient concerns and preferences. [45][46][47] However, researchers studying hospital pharmacist communication with patients have found that phar- macists rarely use open-ended questions to engage patients during medication counselling. 14,22,43 Accommodative emotional expression (appropriate levels of empathy and reassurance) was nearly always demonstrated by study pharmacists. Validating patients' concerns is an important process for pharmacists to assist patients in dealing with negative emotions and worries about their medications and health care. 48 Sensitivity to patients' ability to afford their medications was another example of caring behaviour demonstrated by many study pharmacists. This contrasted with other Australian research that found hospital pharmacists did not explore financial reasons for patients' medication issues. 43 There were rare occasions where participating pharmacists ap- Another demonstration of inappropriate control observed was related to pharmacists' task-oriented approach to medication counselling. All pharmacists used patient-specific medication lists that acted as visual props and information sources for patients, but also as a mechanism to keep conversations on track.