The impact of a Personal Outpatient Worksheet and Educational Reference form on patient engagement and communication in the surgical clinic: multicentre randomized clinical trial

Abstract

Background

In contemporary medicine, time pressure, limited resourcing, human factors, and health literacy are barriers to delivering high-quality patient-centred care. Improving patient engagement may mitigate these barriers. This study evaluated the impact of the Personal Outpatient Worksheet and Educational Reference (POWER) form on patient engagement in the surgical outpatient setting.

Methods

A multicentre, randomized clinical trial was conducted in two metropolitan hospitals between July 2022 and January 2025. Patients over 16 attending a general surgical outpatient appointment were eligible for inclusion. Participants were randomized into either intervention or control cohorts using a random number generator with 1 : 1 allocation stratified by sex. All participants underwent a surgical outpatient appointment, which was audio-visually recorded. The intervention cohort received a copy of the POWER form, a document with five suggested questions participants could ask during their appointment. The control cohort received standard care without the POWER form. Participants were telephoned one week after their appointment to complete a survey. The primary endpoint was patient recall, assessed by comparing survey responses with the recorded discussion, using an intention to treat analysis.

Results

A total of 143 participants were recruited (71 participants to intervention, 72 to control). Ten participants (five in each group) withdrew and were excluded from analysis of the primary endpoint. Mean recall in the intervention cohort was 88% (95% confidence interval 83 to 92) compared with 31% (26 to 35) in the control group (P < 0.001).

Conclusion

Use of the POWER form in the general surgical outpatient setting significantly improves patient recall.

Registration number: ACTRN12625000075482 (https://www.anzctr.org.au/).

This is the first randomized clinical trial to evaluate the impact of a generic question prompt list (QPL) on general surgical patients. Outside of the pilot study it is the first of its kind to utilize the objectivity of video recording to assess the accuracy of patient recall. It demonstrates that use of a generic QPL (the POWER form) improves patient recall, reduces decisional conflict, and does not increase consultation length.

Introduction

Quality clinical practice requires a strong therapeutic relationship between doctor and patient^1,2. Failure to establish this can result in mistrust, poor compliance, and poor patient outcomes³. Historically, clinical practice was rooted in a paternalistic, authoritarian model of service delivery⁴. This construct of ‘doctor knows best’ resulted in loose interpretations of informed consent, a loss of patient autonomy, and substantial harm enacted with good intentions⁵. In recent times, the outdated coercive model of practice has been replaced with a more collaborative, patient-centred style². Shared decision-making (SDM) has become the norm, resulting in greater patient involvement in the healthcare journey. Patients feel empowered, and have reduced anxiety and greater compliance⁶. This results in better healthcare outcomes, reduced healthcare costs, and more efficient service utilization⁷. Despite this, use of tools designed to improve SDM has been poor⁸. Interventions are typically targeted at the clinician or the patient¹. Interventions targeting the clinician such as coaching, workshops, or SDM training are often costly, poorly attended, or found to be too cumbersome to implement in busy clinical practice⁹. Conversely, there is limited research regarding interventions that target patient engagement; however, tools that exist have been demonstrated to not only improve patient knowledge and reduce decisional conflict but also have fewer logistical constraints¹⁰.

As the body of evidence in this often-overlooked field grows, understanding of the complex human and system factors that impact clinical interactions improves. Despite this, patient engagement and understanding remains poor. In a typical clinical encounter, the rate at which a patient is able to remember important information accurately lies between 14% and 40%^10,11. A myriad of factors contribute to this, including time pressures, emotional stress, and limited patient health literacy. Global medical short-staffing forces clinicians to multitask and balance multiple competing commitments¹². With these external stressors, doctors often need to achieve a compromise whereby they can deliver the best care to large cohorts, and often must cut short the time spent with a single individual to serve the needs of the many¹³. In such resource-poor settings effectively implementing SDM practices can improve the quality of a clinical interaction, without having to compromise on consultation duration.

Visiting a doctor can be an incredibly challenging and emotional time for a patient. Patients are often in a state of vulnerability and are expected to process a vast amount of information, often delivered by a stranger, in a short period of time. Although patient-centric interventions have been demonstrated to be efficacious, the design, useability, and cognitive burden must be considered. Question prompt lists (QPLs) are interventions that are designed to reduce the mental load of both a clinician and a patient¹⁴. They are a cost-effective, non-burdensome means enabling the patient to organize their thoughts and act as an aide memoir to trigger discussion about the important parts of the consultation. This ensures that the patient’s treatment is tailored to their individual values and priorities. By ensuring that patients have these needs met, the mental load of both doctor and patient are reduced. Bidirectional exchange of information is subsequently made easier and the onus of this is shared between two parties¹⁴.

Interventions of this type have been demonstrated to improve patient engagement in medical and oncological settings, though the current literature on QPL use in surgical settings is limited¹⁵. In the surgical setting, studies have focussed primarily on preoperative consultations, included small cohorts, and have not directly measured the impact of QPL on SDM and decisional conflict¹⁴. Furthermore, many tested QPLs were developed with limited input from consumers. Of the tools that are available, almost all are designed for a specific pathology or specialty, with none able to be generalized to a broad patient cohort¹⁴. A recent systematic review¹⁴ identified only seven randomized clinical trials evaluating QPLs in surgery, and all studies had flaws resulting in significant bias. A pilot study¹⁶ conducted after this review reported favourable outcomes in the preoperative outpatient surgical cohort. This provided the framework for the development of the Personal Outpatient Worksheet and Educational Reference (POWER) form—a generic QPL designed for all surgical patients, in all surgical settings.

This study hypothesized that use of a 5-point question prompt sheet, the POWER form, would impact patient engagement, recall, decisional conflict, and consultation dynamics in the surgical outpatient setting.

Methods

Study design

A multicentre, single blinded, randomized clinical trial with 1 : 1 allocation was conducted at The Royal Adelaide and Queen Elizabeth Hospitals involving six surgical units. Ethical approval was granted by the Central Adelaide Local Health Network Human Research and Ethics Committee (Ref 15447). This study was conducted in accordance with the Australian Code for the Responsible Conduct of Research and the National Statement on Ethical Conduct in Human Research and adhered to the CONSORT reporting guidelines^17,18.

Participants

Participants over the age of 16 attending surgical clinic during the trial period who had capacity to consent, did not require an English language interpreter, and without memory impairment were eligible to participate. They were identified from clinic booking lists and were contacted by phone the day before their appointment and invited to participate in the research study. Those interested were subsequently approached by a member of the research team where they were given the opportunity to ask questions and if willing to participate gave formal written consent. Basic demographic data, including sex, was collected using electronic health records and gender was determined by asking patients their preferred pronouns.

General surgeons (upper gastrointestinal, hepatobiliary, colorectal, breast, and surgical oncology) who had clinics at recruitment sites occurring during the recruitment period were invited to participate by inviting expressions of interest at departmental meetings. Interested surgeons were later approached by a member of the research team and gave formal written consent.

Randomization and masking

Randomization was achieved using a random number generator and block randomization (blocks of 4 and 6) in a 1 : 1 ratio, stratified by patient sex. Patients were allocated to either intervention, where they were provided with a physical copy of the POWER form and pen, or standard care, where no form was provided. Allocation concealment was maintained using sealed opaque, tamper-proof envelopes, each opened in chronological order upon patient enrolment. This was performed by two members of the research team who had no patient contact, and no involvement in data collection or video review. Both cohorts then underwent a surgical consultation with their treating clinician, which was audio-visually recorded. Surgeons were not blinded to patient allocation as they could see the presence of the intervention; however, they were not aware of the study outcomes. Surgeons were not coached to encourage use of the form. Participants were not aware that they were allocated to a cohort and patients in the control cohort were not aware of the existence of the POWER form and underwent a regular surgical consultation. Participants who received the intervention having previously had a regular consultation may have noticed the difference from their previous consultation.

Procedures

The POWER form was adapted from a previously developed QPL, refined through a modified Delphi process informed by the results of a prospective survey distributed to a group of stakeholders including doctors, nursing staff, clerical staff, and patients¹⁶. Healthcare consumers including surgeons, trainees, junior doctors, nursing staff, clerical staff, and lay persons were surveyed asking them to identify the most important things that a patient should know when visiting a surgeon. Based on these results, the content for a 5-point form was developed. This was then discussed and reviewed by a panel comprised of each group surveyed, and the final content was agreed upon by voting, with agreement requiring 80% consensus. Binary (yes/no) scoring was used to determine the Item Content Validity Index (I-CVI). The final prompts all scored I-CVI scores of > 0.9. The five points on the form were:

• What is my medical issue (or what do we suspect)?

• What are the treatment options (or further tests)?

• How does this impact my ability to work/eat/exercise?

• What are the risks/benefits of treatment (or what happens if I don’t have treatment)?

• Who is involved in my treatment and when will it occur?

This was then made visually appealing using Canva and given to persons with colour vision deficiencies to ensure legibility and usability¹⁹. The final POWER form then underwent further usability assessment by two independent parties—two consultant surgeons neither involved in the form development nor participants in this study. Usability was scored using the Patient Education Materials Assessment Tool for Printable Materials²⁰. Both reviewers found the forms to score 94% for understandability, usability, and actionability. The final form is presented in Fig. 1.

Figure 1.

The Personal Outpatient Worksheet and Educational Reference form

Surgical outpatient consultations were audio-visually recorded using a small camera (GoPro Hero 10^™). Based on previous studies^16,21, the camera was placed in an unobtrusive location designed to capture communication but not patient undressing or physical examination. Study participants, including the surgeon, patient, and companion (if present), were recorded with their explicit knowledge and written consent. Patients were contacted one week after consultation via telephone for a structured questionnaire (Appendix S1). This was performed by a researcher who had not seen the consultation video.

Outcomes

The primary outcome of this study was patient recall, evaluated by a structured questionnaire conducted one week after consultation (Appendix S1). Each patient’s responses were transcribed verbatim by a blinded investigator and compared with the consultation video. Secondary outcomes included markers of patient engagement, including surgeon and patient talk time, number of questions asked, decisional conflict, SDM, and evaluation of the Hawthorne Effect. Talk time and questions asked were counted manually by two investigators using digital timers. Talk time included total cumulative seconds of speech excluding active listening. Where timing fell outside a 20% error range, the video was extracted by a third reviewer until it came within the accepted margin of error¹⁶. Decisional conflict was measured using the SURE score during the follow-up questionnaire²². SDM was measured by calculating the mean score of two trained reviewers using OPTION 12 scoring²³. Evaluation of the Hawthorne Effect involved specific questions regarding behavioural alteration secondary to the presence of the camera (Appendix S1). Subcohort analyses were performed delineating patients by sex, presence of a companion, pathology, and timing of consultation.

Statistical analysis

Power analysis was performed using the published pilot study¹⁶. Two-sample t tests for mean difference with unequal variances were calculated for each of the primary outcome, with mean difference(standard deviation (s.d.)), alpha = 0.05, two-sided test, and equal weighting that has a power of 80% and requires a total sample size of 140, allowing for 10% loss of follow-up.

Analysis followed the intention-to-treat principle. Univariable linear regressions were performed for various outcomes versus predictor: Cohort (Intervention/Control) in separate models. Cross tabulations were performed for Cohort versus categorical variables (sex, presence of a companion, consultation timing, and pathology). The statistical software used was SAS On Demand for Academics (SAS Institute Inc. 2024, Version 9.4, Cary, NC, USA). A P value ≤ 0.05 was considered to be statistically significant. A data monitoring committee was not used due to the low-risk nature of the intervention.

Results

Between 7 July 2022, and 15 January 2025, 629 participants were screened for eligibility, of whom 486 were excluded for reasons including not meeting inclusion criteria (27), declining participation (19), or not answering their telephone (440). A total of 143 participants were randomly assigned to either the intervention (71) or control (72) groups (Fig. 2). One patient did not require a surgical consultation, one did not wait to be seen, one was seen by another clinician off camera, and one video was corrupted. A recorded video consultation was therefore undertaken with 139 participants. One patient later withdrew consent and 5 were lost to follow-up, resulting in a total of 133 participants, resulting in a completeness index of 93% for the primary outcome²⁴.

Figure 2.

CONSORT 2025 flow diagram

Baseline characteristics of participants are presented in Table 1. The mean(s.d.) age of participants was 60(15), with a range of 22 to 84, there was a female sex predominance of 76 (53%), and 37 participants (26%) had a companion present. A total of 103 participants (72%) had benign pathology, 27 (19%) malignant pathology, and 13 (9%) were undiagnosed. A total of 64 participants (45%) underwent a preoperative consultation (being booked for an operation), 44 (32%) were in the investigative phase of their treatment (being worked up before definitive management), 19 (13%) were postoperative (having undergone an operation within the last 3 months, and 14 (10%) were having a follow-up consultation (where patients had not had a previous surgical encounter within the last 3 months). Characteristics were similar across both groups and no adjustments were required for any analyses.

Table 1.

Demographic characteristics

	Control n = 72	Intervention n = 71
Sex
Male	33 (46%)	31 (43%)
Female	39 (54%)	37 (57%)
Age mean years (s.d.)	60(15)	60(15)
Pathology
Benign	52 (72%)	51 (72%)
Malignant	14 (19%)	13 (18%)
Undiagnosed	6 (9%)	7 (10%)
Timing of appointment
Investigative	22 (32%)	22 (31%)
Preoperative	29 (41%)	35 (49%)
Postoperative	12 (17%)	7 (10%)
Follow-up	7 (10%)	7 (10%)
Companion present	18 (25%)	19 (27%)

Values are n (%) unless otherwise stated.

Eleven surgeons participated in the study. Three were general surgeons, two were trauma surgeons, one specialized in oncology and sarcoma surgery, three were oesophagogastric surgeons, and two were hepatobiliary surgeons. Nine surgeons identified as men, and two as women.

Results are summarized in Table 2. Assumptions of a linear regression: normality of residuals and homoscedasticity were found to be upheld throughout by inspection of histogram and scatter plots of residuals and predicted values. The intervention improved patient recall compared with the control group (mean difference 57%, 95% confidence interval (c.i.) 50.2 to 63.4; P < 0.001). Mean recall in the intervention cohort was 88% (95% c.i. 83.6 to 92.6) versus 31% (26.5 to 35.9) in the control cohort.

Table 2.

Main results

	Control n = 67	Intervention n = 66	P*
Surgeon talk time (s)	489.97 (413.43, 566.51)	484.01 (418.46, 549.56)	0.907
Patient talk time (s)	212.31 (183.09, 241.53)	332.35 (288.03, 376.67)	< 0.001
Consult duration (s)	925.81 (813.52, 1038.10)	1044.86 (921.95, 1167.76)	0.158
Patient questions (n)	3.41 (2.80, 4.01)	10.68 (9.17, 12.19)	< 0.001
Surgeon questions	18.70 (16.63, 20.75)	19.63 (17.36, 21.91)	0.544
OPTION 12	7.70 (6.79, 8.60)	16.47 (15.34, 17.60)	< 0.001
Recall (%)	31% (26.5201, 35.90)	88% (83.56, 92.56)	< 0.001
SURE score	1.52 (1.22, 1.81)	3.55 (3.39, 3.72)	< 0.001

Values are mean (95% confidence intervals). s, seconds. *Univariable linear regression.

The intervention cohort increased patient talk time (mean difference 120.04 seconds (s), 95% c.i. 67.34 to 172.74; P < 0.001), increased SDM (OPTION 12 Score) (mean difference 8.78, 7.34 to 10.21; P < 0.001), and reduced decisional conflict (Decisional Conflict Score) (mean difference −2.04, −2.38 to 1.70; P < 0.001) compared with the control group. The number of questions asked by patients (mean difference 7.28, 5.66 to 8.89; P < 0.001) were improved by the intervention, but the number of questions asked by surgeons (mean difference 2.94, 2.10 to 3.99; P = 0.544) or companions (mean difference 3.87, 0.08 to 7.82; P = 0.054) were not significantly altered. Talk time of the surgeon (mean difference 5.96, −94.08 to 106.01; P = 0.907) and consultation duration (mean difference 119.04, 46.23 to 284.31; P = 0.158) were not significantly altered.

There were no statistically significant associations comparing subcohorts delineated by sex, presence of a companion, pathology, or timing of consultation. All participants reported that the camera was neither distracting nor intrusive. Of the 133 participants surveyed, only 3 reported that they initially noticed the camera; 91 participants reported without prompting that they forgot the camera was present. There were no adverse events during this study.

Discussion

Use of the POWER form improved patient engagement, as demonstrated by improved talk time, number of questions asked, recall, and SDM. This is reflected in the reduced decision conflict in the intervention cohort. This benefit was not significantly altered with patient pathology, sex, or timing of their appointment, suggesting that the POWER form was beneficial across a wide range of patient and surgical variables. These findings are consistent with the current evidence¹⁰, suggesting that patient-focussed interventions improve communication and engagement in the surgical outpatient setting. A non-significant difference in the talk time of the surgeon and overall consultation duration suggests that minimal alteration to clinical practice is required to implement this intervention. This is a common barrier to adoption of novel communication interventions²⁵. Engagement of clinicians in the design process, simplicity of the intervention, and use of objective measures of evaluation (video recording), are where this tool differs from previous interventions¹⁴. These findings support the use of the POWER form in general surgical outpatient settings.

Clinicians often meet patients on the worst day of their lives^26,27. The shock that comes with receiving news of a life-threatening diagnosis, such as a new cancer diagnosis, causes a significantly heightened emotional state²⁶. Adding this to the mental capacity that is required by a patient to process the amount of information discussed during a consultation contributes to historically poor rates of patient understanding in surgery²⁸. Tools that are designed to promote dialogue, and ease this mental load, improve the quality of the doctor–patient interaction^28,29. Evidence³⁰ suggests that patients want more information. In addition to increasing SDM, tools like the POWER form also help patients filter what is relevant from a vast volume of information that is often discussed. This has the potential to reduce decisional conflict and improve patient care^31,32. The fact that the form showed improvement in recall in patients irrespective of consultation timing suggests that this aid, designed to reduce the mental load of a patient and organize their thoughts, can help to re-enforce already discussed topics, and is superior to standard care where no QPL is used.

The mental load of a doctor is significant³³. In a typical outpatient session, a clinician will see over a dozen patients, meeting most for the first time. They may have to inform a patient about a cancer diagnosis, troubleshoot rare symptoms, supervise and advise junior staff, and keep to a strict time schedule. It is often the latter aspect that is sacrificed and comes at the expense of the personal or family life of a clinician³⁴. These competing commitments contribute to fatigue and burnout amongst clinicians³⁴. These stressors are compounded when a patient, through no fault of their own or the clinician’s, fails to understand or follow a treatment recommendation. An intervention such as the POWER form does not significantly alter the duration of a clinical consultation; however, it demonstrably improves the quality of the interaction. As the patient becomes empowered and engaged in their healthcare, the mental load of the therapeutic relationship is shared, and the burden on the clinician is reduced³⁵.

Doctors have previously been criticized for a lack of interpersonal skills and inapproachability^36,37. This has been especially highlighted in the surgical community³⁸. This fractures the therapeutic relationship and can promote error, leading to mistrust, patient complications, and litigation^37,39. The consequences of fractured rapport or misunderstanding directly impact patient outcomes. Patients failing to understand treatment decisions or pathology directly impacts treatment adherence⁴⁰. The trend is increasingly common in international hospitals, especially amongst vulnerable populations⁴¹. This adds to institutional mistrust and poor long-term healthcare engagement⁴¹. Preventable events such as this contribute to the reduced life expectancy seen in First Nations people⁴², and cause significant a financial burden on healthcare systems⁴³. The other factors contributing to these issues lie in social inequality, intergenerational trauma, and systematic oppression; however, there is the opportunity to make positive change in the healthcare space at the coalface by implementing cost-effective, validated interventions.

Limitations of this study include recruitment conducted only at two large metropolitan hospitals in general surgical specialties. Although the catchment area for these centres involves areas of both high and low rates of literacy, the results should be interpreted with caution for use in other specialties, and areas with poor literacy. Voluntary participation by surgeons and patients interested in communication has the potential to introduce selection bias to those who have higher health literacy or are more reflective in their thought processes, but the randomized design of the study, and lack of difference between cohorts, ensures that the cohorts are still comparable. Some 440 patients excluded from the study for not answering the recruitment call does potentially limit professionals or those with caring responsibilities; however, the age ranges in each cohort were diverse, and although the mean age was skewed towards the older population this is consistent with a realistic surgical outpatient cohort.

Any study that involves the use of a video camera has the potential to be impacted by The Hawthorne Effect—the phenomenon whereby a person alters their behaviour due to awareness they are being observed⁴⁴. Data gathered in the follow-up questionnaire suggest minimal impact of this phenomenon, and the study was designed specifically so that groups were evaluated under identical conditions to minimize the impact of the camera. Empirical masking was not possible in this study, but care was taken to conceal cohort allocation to data collectors where possible. The consultation duration result being non-significant despite the average consultation duration in the intervention cohort being 119 s longer than the control is possibly due to the vast variation in consultation times in both groups. As this was a secondary outcome, the study was not powered to this variable.

The POWER form has demonstrated efficacy in the general surgical outpatient setting. Future directions of research in this area should involve evaluation of the tool in multiple languages and in different clinical settings. This tool has the potential to be easily integrated into digital health platforms. The simplicity of the design uses minimal data and does not require intensive coding to be adapted to smartphone use. This allows for use in a telehealth setting to evaluate patient engagement in rural and remote practices. Similar QPLs have been effective in the inpatient setting, and this form should be evaluated in both inpatient and other specialty settings.

Use of the POWER form in the general surgical outpatient setting can improve patient engagement, recall, and SDM. It is a cost-effective, universally implementable intervention that can improve patient care, clinician workload, and hospital systems.

Funding

There was no funding source for this work. E.C.T., M.H., and J.D.E. received the University of Adelaide Research Scholarship and the Basil Hetzel Institute Higher Degree top up scholarship. M.H. also received the South Australian Hospital Research Foundation Higher Degree Scholarships.

Author contributions

Matheesha Herath (Conceptualization, Data curation, Design, Formal analysis, Investigation, Project administration, Visualization, Writing—original draft), Jesse D. Ey (Conceptualization, Data curation, Investigation, Methodology, Writing—original draft, Writing—review & editing), Ellie C. Treloar (Conceptualization, Data curation, Investigation, Methodology, Writing—original draft, Writing—review & editing), Jessie Martin (Methodology, Writing—original draft), Harriet McConville (Methodology), Martin Bruening (Conceptualization, Design, Oversight, Project administration, Supervision, Writing—reviewing & editing), and Guy J. Maddern (Conceptualization, Design, Oversight, Project administration, Supervision, Writing—reviewing & editing)

Disclosure

The authors declare no conflict of interest.

Data availability

The data underlying this article will be shared on reasonable request to the corresponding author.

Supplementary material

Supplementary material is available at BJS Open online.