Doctor–patient communication, knowledge, and question prompt lists in reducing preoperative anxiety – A randomized control study

Summary

Objective

The aims of the study were to assess factors responsible for the reduction of preoperative anxiety in patients undergoing breast and abdominal surgeries. In particular, we investigated whether question prompt lists (QPL), patients’ knowledge, or the communication skills of surgeons had effects on anxiety reduction.

Methods

Patients were randomly assigned to QPL and control groups. Anxiety was assessed on the State Trait Anxiety Inventory.

Results

Both groups showed significant reduction in anxiety between initial consultation and one day prior to surgery, with QPL patients showing a trend towards a greater reduction of anxiety after surgery and a significant reduction at the first outpatient follow-up. Satisfaction with consultation and the doctor’s ability to answer questions concerning diagnosis, and treatment were significantly associated with anxiety reduction.

Conclusion

Effective anxiety reduction hinged on doctors’ communication abilities and patients’ satisfaction with the consultation.

Keywords

communication;knowledge;patient satisfaction;preoperative anxiety;question prompt lists

1. Introduction

Studies indicate that the communication between patients and their doctors during consultations is generally poor.^{1; 2 ;  3} Maguire and Pitceathly cite various deficiencies in doctor–patient communication.⁴ Moreover, doctors often do not check how well their patients have understood the information given, and have observed that even when doctors provide information, they do so in an inflexible manner and tend to ignore what the individual patient wants to know.⁵ Unsatisfactory consultations lead to patient dissatisfaction, and in worst cases, may lead to misunderstandings, and even litigation.⁶ In surgical patients, poor doctor–patient communication leaves unanswered questions about the diagnosis, intervention and postoperative care, and may contribute to pre and postoperative anxiety. Conversely, communication could be enhanced when patients are given opportunities to clarify doubts on matters of greatest concern to them.⁷

To assist patients ask questions, some researchers have used question prompt lists (QPL) during the course of surgical consultation.^{8; 9 ;  10} A QPL is a structured set of questions to remind patients to seek answers from their doctors. McJannett and colleagues found QPL to be simple and inexpensive, and by obtaining answers to their questions, patients were less likely to be fearful of surgery.^{7 ;  11}

The aims of the present study were to assess the factors responsible for reducing preoperative anxiety in a tertiary general hospital. We hypothesized that patients who had better knowledge of their pre and postoperative surgical care and who used the QPL would have less pre and postoperative anxiety. To the best of our knowledge, such interventions have not been previously studied in Southeast Asia.

2. Materials and methods

2.1. Inclusion criteria

Patients should be between the ages 18 and 65 years, had been scheduled for surgery and were able to read the English or the Mandarin version of the question prompt list.

2.2. Exclusion criteria

We excluded those whose condition was terminal or in whom the tumor was so extensive as to be inoperable, and for whom no surgery was being offered. Those unable to read English and Mandarin were not recruited.

2.3. Methodology

Patients meeting the inclusion criteria were randomly assigned to either the experimental (QPL group) or the control group. The participants were asked to select one out of 10 envelopes. Five envelopes contained slips of paper stating “test” and the other five contained slips of paper stating “control”. We initially intended to recruit patients scheduled for head and neck, abdomen, and breast operations, but decided to concentrate on abdomen and breast patients as these two groups yielded the highest number of patients. The study was approved by the hospital’s Institutional Review Board.

The QPL group were shown a list of common questions (compiled by the researchers) which they could use for seeking clarification from their surgeons. A sample of the QPL is appended in Table 1. The QPL served as a guide, and patients were at liberty to ask additional questions of their own. The purpose was to encourage clarification of doubts about the operation and postoperative care, thus forming the basis for anxiety reduction.

Table 1. Question prompt list.

What is the diagnosis of my condition?

If the diagnosis is cancer

• What is the stage of my cancer?

• If we get rid of the cancer, what are the chances of recurrence?

What will happen to me during surgery?

Are there any dangers/risks during surgery?

How long do I have to remain in hospital after surgery?

How much pain will I experience after surgery?

What other treatments will I need in addition to the surgery?

What are my chances of recovery?

Will my condition affect my ability to work or perform other activities?

Although the surgeons would have already given explanations to both experimental and control group on the indications, nature, and postoperative care of the intended operations, the purpose of the study was to assess whether the use of QPL conferred any added advantage in reducing preoperative, or to some extent, postoperative anxiety. The opportunity to ask questions from the prepared list would arise during the ward round in the course of admission, usually one day before the scheduled operation. The time interval between the initial consultation when patients were informed about the need for surgery to the time they were admitted for surgery ranged between 1 and 3 weeks.

The psychiatrist investigator on the team performed independent clinical assessments of a random selection of about one in five patients and checked that the forms were correctly filled. He was initially blinded to the patients’ scores on an anxiety questionnaire, the State-Trait Anxiety Inventory (STAI).¹² The STAI are self-rated anxiety scales which comprise the State Anxiety scale (STAI Y-1) and the Trait Anxiety scale (STAI Y-2). The STAI Y-1 evaluates feelings of apprehension, tension, nervousness, and worry, which increase in response to physical danger and psychological stress. The STAI Y-2 measures trait anxiety, a relatively stable predisposition of an individual to being anxious. These scales have been extensively used by researchers studying anxiety in patients with physical conditions. The research coordinator met with the patients on four separate occasions.

2.3.1. Encounter time 1

The research coordinator approached patients who had been scheduled for surgery. Those patients randomly selected to receive the QPL were shown a list of questions they could ask their doctors prior to surgery. The control groups were given the usual information concerning admission procedures. Anxiety in both QPL and control groups were rated using the STAI.

2.3.2. Encounter time 2

One day before surgery, during the ward round QPL patients were encouraged to use the QPL to ask their doctors questions concerning their illness and forthcoming surgery. The anxiety levels in both QPL patients and controls were again rated after meeting their doctors. Both groups were asked about knowledge of their diagnosis, whether discussion with their doctor covered all the questions they had wanted to ask, whether they had unanswered questions concerning diagnosis, operation, and postoperative care following the ward round. Subjective knowledge of diagnosis, operative, and postoperative procedures were recorded, although the patients’ actual knowledge was not formally tested.

Patients were asked to rate on a 10-point Likert scale, their satisfaction with their consultation, and whether the doctor was able to answer all their questions. They were asked for their opinion concerning the QPL, i.e., whether it was useful, and whether they believed the QPL helped them to communicate with their doctor.

2.3.3. Encounter time 3

One to four days postoperatively, anxiety levels were again measured in both groups.

2.3.4. Encounter time 4

The research coordinator met with patients a final time when patients returned to the outpatient clinics for their first postoperative follow-up appointment. Patients were asked to rate their anxiety level.

2.4. Statistical analysis

2.4.1. Calculation of sample size

It was postulated that 45% of the control group and 20% of the QPL group would remain anxious postoperatively. A sample size of 50 in each group would have a power of 80% with a two-sided test to achieve a statistically significant result.

2.4.2. Analysis

All statistical analyses were performed using SPSS 17.0 (SPSS Inc., Chicago, USA).

Differences in quantitative demographical variables between the QPL and control groups were assessed using parametric two-Sample t test when normality and homogeneity assumptions were satisfied, otherwise the nonparametric Mann Whitney U test was used. Chi-square or Fisher Exact tests were performed for differences in qualitative variables. A repeated measurement analysis was performed on the percentage change in STAI scores for Encounter times 2–4 from the baseline Encounter time 1 when comparing the two groups, adjusting for age, gender, educational level, STAI, and operation site. To determine which items of the questionnaire (on knowledge of diagnosis, what their surgery entailed, and what would take place postoperatively and whether they had unanswered questions prior to operation) affected the participants’ level of anxiety at Encounter time 2, a linear regression analysis adjusting for age, gender, educational level, STAI was performed. The above multivariate analyses were also performed for the site-subgroups. Within-group analysis of pre and postoperative anxiety status was assessed using the McNemar test. Odds ratios with 95% confidence intervals (CI) will be presented. Statistical significance was set at p value of 0.05.

3. Results

A total of 230 patients (114 QPL and 116 controls) were recruited. They had a mean age of 49.0 years (SD = 9.6). Of these, 226 patients completed at least the first two interviews (112 participants from the QPL group, 114 from the control group). Two hundred and seven patients completed all of the four required interviews (101 from the QPL group, and 106 from the control group). The demographic characteristics of the patients are appended in Table 2, which shows that the two groups are comparable. There were no statistically significant differences in educational level of both groups of patients. Among those who were administered the QPL, only 4% had received no education, 20% received primary school education (i.e., from ages 6 to 12), 45% had received secondary school education (from ages 13 to 16 years), and about 30% had received postsecondary education (preUniversity) or tertiary (University or Polytechnic) education.

Table 2. Characteristics of QPL and control participants.

Demographic characteristics	QPL	Control	p
Sex, n (%)
Male	27 (23.7)	30 (25.9)	0.702
Female	87 (76.3)	86 (74.1)
Mean age, (y), SD	49.34 (8.98)	48.70 (10.30)	0.614
Educational level, n (%)
No formal education	5 (4.4)	5 (4.3)	0.945
Primary	23 (20.2)	22 (19.0)
Secondary	52 (45.6)	49 (42.2)
Postsecondary/Tertiary	34 (29.8)	40 (34.5)
Monthly income, $ (%)
<1000	36 (37.9)	29 (33.0)	0.361
1000–2999	36 (37.9)	31 (35.2)
3000–5999	19 (20.0)	27 (30.6)
≥6000	4 (4.2)	1 (1.2)
Employment, n (%)
Retired	8 (7.1)	11 (9.5)	0.627
Self-employed	8 (7.1)	10 (8.6)
Part-time	6 (5.3)	12 (10.3)
Unemployed	35 (31.0)	32 (27.6)
Full-time	56 (49.6)	51 (44.0)
Abdominal, n (%)
Cancer	30 (52.6)	30 (46.2)	0.475
Noncancer	27 (47.4)	35 (53.8)
Breast, n (%)
Cancer	34 (59.6)	34 (66.7)	0.451
Noncancer	23 (40.4)	17 (33.3)

In the abdominal group there were 58 patients who had cancer, 63 with noncancer diagnoses. Of the breast patients 72 had cancer, 41 had benign tumors. We managed to accrue seven patients with head and neck cancer and seven without. Owing to the relatively small number of head and neck patients we decided not to include them in the data analysis.

Table 3 demonstrates statistically significant reductions in anxiety at Encounter time 2 compared to Encounter time 1. These reductions were evident for both abdominal, and breast surgery patients regardless of QPL or controls. However for the breast QPL patients, anxiety scores were reduced to a greater extent than compared to controls (p < 0.001).

Table 3. STAI scores at Encounter time 1 and 2: within groups, subgrouped by site.

		Time 1	Time 2	Difference (95% CI)	p
All participants	QPL (n = 112)	46.8 (14.1)	41.4 (11.9)	5.4 (3.6, 7.2)	<0.001
	Control (n = 114)	45.4 (13.4)	42.0 (11.1)	3.4 (1.9, 4.9)	<0.001
Abdominal	QPL (n = 56)	42.6 (11.8)	38.9 (11.8)	3.7 (1.2, 6.2)	0.005
	Control (n = 64)	41.8 (13.6)	39.2 (10.7)	2.6 (0.4, 4.8)	0.019
Breast	QPL (n = 56)	51.0 (15.0)	43.9 (11.6)	7.2 (4.6, 9.7)	<0.001
	Control (n = 50)	50.0 (11.8)	45.6 (10.5)	4.4 (2.4, 6.4)	<0.001

Values are mean (SD).

In Table 4, comparing reduction in anxiety over the three periods, the combined (breast and abdominal surgery) patients administered with QPL (vs. controls) showed significant reduction at time 4 (p = 0.010). Mirroring this reduction, was the significantly reduced anxiety at time 4 for breast QPL patients (p = 0.042). Anxiety reduction at time 3 for breast QPL administered patients approached significance (p = 0.054).

Table 4. Between group comparisons subgrouped by site in percentage reduction in anxiety at Encounter times 2, 3 and 4 with reference to Encounter time 1.

	Period	QPL group	Control	Difference (95% CI)	p
All participants	Time 2	-9.9 (23.9)	-5.9 (23.7)	-4.0 (-8.8, 0.7)	0.097
	Time 3	-13.7 (33.6)	-9.5 (32.9)	-4.2 (-10.7, 2.3)	0.125
	Time 4	-20.4 (30.8)	-12.7 (31.4)	-7.7 (-13.6, -1.8)	0.010
Abdominal	Time 2	-8.6 (23.2)	-4.9 (24.8)	-3.7 (-11.3, 3.0)	0.341
	Time 3	-9.0 (33.5)	-7.3 (34.6)	-1.7 (-12.5, 9.2)	0.626
	Time 4	-19.7 (28.1)	-13.4 (29.7)	-6.3 (-15.1, 2.5)	0.225
Breast	Time 2	-8.5 (19.5)	-5.0 (17.0)	-3.5 (-9.5, 2.5)	0.253
	Time 3	-17.8 (23.6)	-11.5 (21.2)	-6.3 (-13.7, 1.0)	0.054
	Time 4	-22.0 (25.7)	-14.3 (23.8)	-7.7(-15.8, 0.4)	0.042

For all participants, site was also added as a covariate for adjustment. Values are mean (SD) % change.

Time 2 = adjusted for age, sex, and educational level compared with Encounter time 1; time 3 = adjusted for age, sex, educational level, pain score compared with Encounter time 1; time 4 = adjusted for age, sex, and educational level compared with Encounter time 1.

We then examined factors contributing to change in anxiety for the combined group of breast and abdominal surgery patients, using linear regression analysis, adjusting for intervention, site, age, sex, educational level and STAI scores at time 1. Satisfaction with the consultation (p = 0.020), the ability of the doctor to answer all the patients’ questions (p = 0.035), leaving no unanswered questions about the operation, (p = 0.029) were significant predictors of anxiety reduction. Preoperative knowledge of diagnosis just failed to reach significance (p = 0.070). No unanswered questions about postoperative care or no unanswered questions about diagnosis (p > 0.05) and knowledge about what was going to happen during surgery were not significantly predictive (p = 0.143) of anxiety reduction.

Upon subgroup analysis by site, significant predictors of anxiety reduction for the breast surgery patients were knowledge about what was going to happen during surgery (p = 0.032), no unanswered questions about diagnosis (p = 0.003) and “satisfactory discussion with my doctor” (p = 0.001). Preoperative knowledge of diagnosis did not predict anxiety reduction for breast patients. There were no associations between educational levels and knowledge of what surgery entailed (p > 0.05), and satisfaction with consultation (p > 0.05). Instead, those with no formal education (t = 3.433, 95% CI 0.315–1.164, p = 0.001) and primary level education (t = 2.489, 95% CI 0.06–0.567, p = 0.013) were more likely to report that discussion with their doctors covered all the questions they wanted to ask.

4. Discussion

It is well established that effective patient education can reduce preoperative anxiety.^{13 ;  14} Anxiety levels were highest during Encounter time 1 after patients were informed of the diagnosis and the need for surgery. Statistically significant reduction in anxiety occurred between times 1 and 2. But the QPL group were marginally less anxious compared to the controls, although this was not statistically significant. Our view is that armed with a list of questions to ask, the QPL patients were more prepared to ask possibly better quality, more in-depth questions compared to the controls, hence eliciting more detailed, and more satisfying responses from their doctors. Although patients did not attribute usefulness to the QPL they might not have realised that there might be subtle benefits from this exercise.

During the intervening period of time from diagnosis to operation, patients could have consulted friends and relatives, researched in books or the internet, and found out more about their diagnosis and treatment. We found that patients from all educational levels consulted the internet for information (p < 0.05), whereas those attaining only primary level education were significantly more likely to turn to books (p = 0.02). This may indicate that those with higher education were more internet savvy and were prepared to use electronic media for information, but regardless of how the information was obtained, acquisition of knowledge resulted in increased confidence and reduction of anxiety.

Apart from obtaining information independently, it seemed more likely that satisfaction with the consultation played an even greater role in anxiety reduction. Abdominal and breast patients in both QPL and control groups reported that the doctor’s ability to answer all their questions was anxiolytic. Similarly, patients whose anxieties were reduced tended to report that discussion with their surgeons resolved all their uncertainties. Conversely, there was a trend for the combined breast and abdominal QPL group to be less anxious preoperatively. However, this effect was lost when the groups were examined individually.

This could be attributed to the individual group’s sample sizes being too small to allow any significant differences to show up. Notwithstanding, the breast patients who were given QPL showed a significant drop in anxiety at Encounter time 4 when compared with time 1. The precise reasons are unclear, although we can postulate that with Encounter time 4 representing the first postoperative outpatient visit, marked anxiety reduction at this stage compared to Encounter time 1, could be attributed to awareness of tumor removal and satisfaction with the results of surgery.

Breast patients’ anxiety reduction were also significantly related to knowledge of what was going to happen during surgery (p = 0.032), satisfaction with the surgical consultation (p = 0.001), the doctor’s ability to answer all the patients’ questions (p = 0.002), no further unanswered questions (p < 0.001), and no unanswered questions about diagnosis (p = 0.003).

In the case of the abdominal group, results from intra-abdominal procedures could not be readily perceived unlike in the case of a breast lump. In contrast to the breast patients, the abdominal group did not rate preoperative knowledge of diagnosis, or knowledge about what would be happening during surgery to be significant in reducing anxiety. It is possible such knowledge could have added to anxieties rather than decreased them, an example of knowledge triggering more anxiety.

Patients scheduled for abdominal surgery were more likely to retain unanswered questions (p = 0.60). It is likely that these patients felt that discussion with their doctor did not sufficiently cover all that they wanted to ask, leaving perhaps, some unanswered questions after consultation. In both groups of patients, knowledge of postoperative care did not reduce anxiety. Perhaps information about postoperative management was not sufficiently stressed in the preoperative setting.

Significant anxiety reduction in both the QPL and control groups in the time interval between initial consultation and surgery was attributable to a great extent to a satisfactory consultation, with additional sources of knowledge from books, friends, family and the internet. This reduction could have arisen because patients had time to process the information they received from their doctors, and from other sources. Whereas, the administration of QPL prior to surgery and the measurement of anxiety soon after it’s use after ward round may not have allowed sufficient time for the QPL to influence anxiety levels. Notwithstanding, it is important to stress that information from alternative sources cannot be a substitute for good doctor–patient communication.

We cannot be absolutely certain that QPL patients are less reluctant about raising their concerns with their surgeons compared to the control group. In fact, patients perceived the QPL as not so helpful (see Table 5). However, it still does not negate the importance of the study in assessing whether a simple intervention, for example, the QPL which has received favorable reports from mainly Western patients would be similarly welcomed in a Southeast Asian country. Our study shows that doctors’ interpersonal skills far outweighed the usefulness of the QPL.

Table 5. Predictors for percentage change of anxiety at Encounter time 2 with reference to Encounter time 1 for all participants.

	B (95% CI)	p∗
I am satisfied with the consultation	-2.2 (-4.0, -0.4)	0.020
The doctor was able to answer all my questions	-1.9 (-3.8, -0.14)	0.035
The QPLs were useful	-0.8 (-3.0, 1.4)	0.462
I believe the QPL helped me communicate with my doctor	-1.6 (-3.9, 0.7)	0.178
Preoperative knowledge of diagnosis	-1.7 (-3.5, 0.14)	0.070
Knowledge about what is going to happen during surgery	-1.3 (-3.1, 0.5)	0.143
Knowledge about what is going to happen after the surgery	-0.1 (-1.4, 1.3)	0.895
Total knowledge	-0.4 (-1.1, 0.24)	0.212
Discussion with my doctor	-1.7 (-1.9, 5.4)	0.352
No unanswered questions about diagnosis	-2.5 (-5.8, 0.8)	0.128
No unanswered questions about operation	-3.7 (-7.0, -0.4)	0.029
No unanswered questions about postoperative care	-0.7 (-3.9, 2.5)	0.668
Outcome of discussion with doctor	-1.0 (-2.1, 0.13)	0.082

• Adjusted for intervention, site, age, sex, educational level.

QPL = Question Prompt List.

As for whether increased frequency of contact with the researchers reduced patients’ anxiety, we have no conclusive evidence that this is the case. Our breast patients showed a trend towards anxiety reduction postoperatively at time 3 (see Table 4). In the final analysis, good communication skills and the ability to anticipate what patients needed to know concerning their diagnosis, their treatment, and their intra and postoperative care played an important role in anxiety reduction. Perhaps for breast patients, most of whom had cancer, knowledge that their cancer had been surgically removed contributed to a large extent to anxiety reduction at their postoperative outpatient visit at time 4 (Table 4).

Acknowledgments

This study was made possible by a grant from the Singhealth Foundation2005/047/A. We wish to thank the following surgeons viz. Drs Georgette Chan, Weng-Hoong Chan, Wei-Sean Yong and A/Prof London Ooi, and colleagues from the Department of General Surgery, Singapore General Hospital and National Cancer Centre who agreed to allow us to interview their patients. Thanks to Prof. David Kissane of Memorial Sloane Kettering Hospital, New York for his encouragement to carry out this study.

References

1. 1 M.K. Mavel, R.M. Epstein, K. Flowers, H.B. Beckman; Soliciting the patient’s agenda: have we improved?; JAMA, 281 (1999), pp. 283–287
2. 2 D. Roter, J. Hall; Physicians’ interviewing styles and medical information obtained from patients; J Gen Int Med, 2 (1987), pp. 325–329
3. 3 A.L. Suchman, K. Markakis, H.B. Beckman, et al.; A model of empathic communication in the medical interview; JAMA, 277 (1997), pp. 678–682
4. 4 P. Maguire, C. Pitceathly; Key communication skills and how to acquire them; BMJ, 325 (2002), pp. 697–700
5. 5 J. Silverman, S. Kurtz, J. Draper; Skills for Communicating with Patients; Radcliffe Medical Press, Oxford (1998)
6. 6 I. Munn; Poor communication main source of patient complaints in Maritimes, registrars report; Can Med Assoc J, 143 (1990), pp. 552–554
7. 7 J. Clayton, P. Butow, M. Tattersall, et al.; Asking questions can help: development and preliminary evaluation of a question prompt list for palliative care patients; Br J Cancer, 89 (2003), pp. 2069–2077
8. 8 R. Brown, P.N. Butow, M.J. Boyer, M.H. Tattersall; Promoting patient participation in the cancer consultation: evaluation of a prompt sheet and coaching in question-asking; Br J Cancer, 80 (1999), pp. 242–248
9. 9 P. Butow, R. Devine, M. Boyer, S. Pendlebury, M. Jackson, M.H. Tattersall; Cancer consultation preparation package: changing patients but not physicians is not enough; J Clin Oncol, 22 (2004), pp. 4401–4409
10. 10 R.F. Brown, P.N. Butow, S.M. Dunn, M.H. Tattersall; Promoting patient participation and shortening cancer consultations: a randomised trial; Br J Cancer, 85 (2001), pp. 1273–1279
11. 11 Butow P. McJannett, M.H. Tattersall, J.F. Thompson; Asking questions can help: development of a question prompt list for cancer patients seeing a surgeon; Eur J Cancer Prev, 12 (2003), pp. 397–405
12. 12 C.D. Spielberger, R.L. Gorsuch, R.E. Luschene; The State Trait Anxiety Inventory; Consulting Psychologists Press, Palo Alto, CA (1970)
13. 13 J. Beddows; Alleviating pre-operative anxiety in patients: A study; Nurs Stand, 11 (1997), pp. 35–38
14. 14 E.C. Devine; Effects of psychoeducational care for adult surgical patients: a meta-analysis of 191 studies; Patient Educ Couns, 19 (1992), pp. 129–142