Effects of a question prompt sheet on active patient behaviour: A randomized controlled trial with depressed outpatients

Abstract

Background:

A deeper engagement into medical decision-making is demanded by treatment guidelines for patients with affective disorders. There is to date little evidence on what facilitates active behaviour of patients with depression. In general medicine ‘question prompt sheets’ (QPSs) have been shown to change patients’ behaviour in the consultation and improve treatment satisfaction but there is no evidence for such interventions for mental health settings.

Aims:

To study the effects of a QPS on active patient behaviour in the consultation.

Methods:

Randomized controlled trial (involving N = 100 outpatients with depression) evaluating the effects of a QPS on patients’ behaviour in the consultation.

Results:

The QPS showed no influence on the number of topics raised by patients (p = .13) nor on the external rater’s perception of ‘Who made the decisions in today’s consultation?’ (p = .50).

Conclusions:

A QPS did not change depressed patients’ behaviour in the consultation. More complex interventions might be needed to change depressed patients’ behaviour within an established doctor–patient dyad. Patient seminars addressing behavioural aspects have been shown to be effective in other settings and may also be feasible for outpatients with affective disorders.

Keywords

Shared decision-making patient autonomy communication adherence depression health services research

Introduction

There is good evidence to suggest that successful communication between patients and doctors may result in better patient satisfaction, better adherence and better health outcomes (Roter, 2000; Stewart, 1995). In addition, an active engagement of patients in medical decisions (‘shared decision-making’) is currently promoted as a promising approach to meet ethical standards as well as to improve patient satisfaction and adherence (Hamann, Leucht & Kissling, 2003).

From general medicine we know that these goals (good communication, engagement of patients in decisions) may be reached by the implementation of communication training for doctors (Elwyn et al., 2004) or the distribution of decision aids (O’Connor et al., 2009), and also by relatively simple interventions (question prompt sheets, leaflets) that prepare patients for their consultations with doctors (e.g. Fleissig, Glasser & Lloyd, 1999). ‘Question prompt sheets’ (QPSs) are simple leaflets in which patients are encouraged to pose questions to their physicians. In many cases patients are provided with samples of frequently asked questions (e.g. ‘What is my diagnosis?’, ‘What treatment options are available for my condition?’ etc.) to give them cues and to make asking questions easier. Among other advantages, QPSs are low-cost interventions that do not require the special training of patients or physicians. In a medical context other than mental health QPSs have been shown to increase the number of questions asked by patients (Brown, Butow, Boyer & Tattersall, 1999), to improve patient satisfaction with the consultation (Little et al., 2004) or to change some communication patterns in the consultation towards more active patient behaviour (Albada, van Dulmen, Ausems & Bensing, 2012). Results from studies on QPSs, however, are inconsistent and setting variables as well as the behaviour of the consulting physician may modulate effects (Harrington, Noble & Newman, 2004).

Despite the importance attributed to an increased engagement of psychiatric patients in medical decision-making (e.g. LUD, 2009), there are to date no studies that have investigated the feasibility and effects of a QPS for patients with affective disorders (or any other psychiatric disorder).

As QPSs may constitute a cost-effective intervention to increase patient participation, we were interested in whether or not QPSs might also prove to show beneficial effects in a mental health setting. If so, they might serve as a helpful adjunct to interventions that have already been shown to be implementable in mental health (e.g. communication training for doctors (Loh et al., 2007) or decision aids (Hamann et al., 2006)) and boost their effects.

In the current study we tested the hypothesis that a QPS for depressed outpatients would increase active patient behaviour in the consultation.

Methods

Study design

We performed a randomized, open-label, single-centre, controlled study comparing the effects of a QPS with usual care for outpatients with depression.

Participants, recruitment, randomization

Patients were recruited in one large psychiatric practice in the city of Munich, Germany. This practice has a caseload of approximately 1,200 patients per quarter, most of them insured by public health insurance, which covers about 90% of the German population. In this practice approximately 800 outpatients with an affective disorder are seen by the same psychiatrist (male, 38 years) at least once every quarter.

All patients visit the practice in which the psychiatrist mainly offers psychiatric counselling, but also does psychotherapies for a minority of the patients (these patients were not included in the present trial, because we expected psychotherapeutic consultations to be distinctly different, e.g. longer, than the majority of consultations).

Male and female patients with a diagnosis of an affective disorder according to ICD-10 and between 18 and 60 years of age were eligible for the study. Only patients who had at least German basic language skills and only patients who were already familiar with the practice (i.e. not patients attending for a first appointment, because these patients constitute only a minority in this practice) were recruited for the study by a research assistant.

At the days when the research assistant was at the practice, all patients fulfilling inclusion criteria were consecutively recruited for the study. After a complete description of the study was delivered to the participants, written informed consent was obtained. Afterwards patients were randomized to the intervention and the control condition. Every patient was given a numbered, sealed, allocation concealment envelope that contained allocation to their group and all study materials.

Interventions

The QPS for outpatients with an affective disorder was developed by four experienced clinicians (two psychiatrists, two psychologists). Its content was derived from: (1) theoretical considerations about which issues may be of importance in the treatment of depression (e.g. from treatment guidelines (LUD, 2009)); (2) an adaption of related approaches from somatic medicine (e.g. Brown et al., 1999; Fleissig et al., 1999); and (3) pilot testing and revising the QPS in the setting of our study. The first author had tested the QPS in the psychiatric practice using the concurrent-think-aloud method.

The final version of the QPS was a one-page leaflet in which patients were encouraged to behave actively in the consultation (‘Make the best out of the consultation’), to write down notes about their wishes for today’s consultation and to tick up to 15 standard questions that were provided on the QPS (e.g. ‘What is my diagnosis?’, ‘What treatment options are still available for my complaints?’ etc.). Finally, the QPS stated that patients could refer to the leaflet during the consultation (see Appendix 1 for the QPS).

All patients in the intervention group were provided with the QPS prior to the consultation by the research assistant and were asked to work through it in a separate room. The research assistant was present while patients were completing the QPS and offered help in case patients had any questions.

Patients in the control condition went to the consultation without receiving the QPS.

Measurements and outcomes

Before the consultation patients were requested to provide the following data:

Socio-demographic data (age, gender, education etc.) and whether the present appointment was a planned one or an emergency one.

Participation preferences: here we applied the Autonomy Preference Index (API; Ende, Kazis, Ash & Moskowitz, 1989), a self-report measure devised to measure patients’ desire to participate in medical decision-making. For the study we used the validated four-item German version of the API as suggested by Simon et al. (2010) that leads to a sum score from 4 to 20, indicating the patient’s preference for active participation in decision-making.

Satisfaction with their treatment in this psychiatric practice (five-point scale).

Trust in the psychiatrist (‘How much do you trust your doctor?’, ranging from 1 = ‘not at all’ to 5 = ‘very much’).

After the consultation, patients were asked to provide the following data:

Satisfaction with the consultation (‘How satisfied are you with today’s consultation?’ ranging from 1 = ‘not at all’ to 5 = ‘very much’).

Amount of time pressure during the consultation.

An estimation of to what extent psychiatrist and patient were involved in the decision-making process (‘Who made the decisions in today’s consultation?’, ranging from 1 = ‘only the doctor’ to 5 = ‘only the patient’).

How helpful they considered the QPS to have been, on a five-point scale (intervention group only).

The psychiatrist provided the diagnosis according to ICD-10 and the duration of illness. He assessed the illness severity according to the Clinical Global Impression Scale (CGI-S) (ranging from 1 = ‘not at all ill’ to 7 = ‘the most severely ill patients’). In addition, he rated various aspects of the consultation (time pressure, satisfaction with the consultation, how demanding he had perceived the patient and who made decisions during the consultation) from his point of view on five-point scales.

All consultations were audio-taped and subsequently analysed. A random sample of 10 tape recordings of the consultations were analysed by four experienced clinical psychiatrists/psychologists, leading to a list of 18 clusters of topics mentioned by either the patients or the psychiatrist (e.g. diagnosis, medication, problems with family members). Following the Global Affect Ratings from the Roter Interaction Analysis System (RIAS) (Roter & Larson, 2002), a few further items were added concerning pressure of time, dominant behaviour of the patient/psychiatrist, active behaviour such as asking questions of the patient/psychiatrist (six-point scales) and to estimate who was involved in decision-making (‘Who made the decisions in today’s consultation?’, ranging from 1 = ‘only the doctor’ to 5 = ‘only the patient’). Finally, the duration of the consultation was documented.

Ten other interviews were then analysed following this scheme by two raters (JH, PI), leading to an inter-rater reliability for the thematic categories of 0.55 ≤ κ ≤ 0.85.

The most critical outcome variables were predefined as: (1) number of topics brought up by the patient; (2) observer rating of who made the medical decisions (ranging from 1 = ‘only the doctor’ to 5 = ‘only the patient’); and (3) patient rating of satisfaction with the consultation. As patient satisfaction proved to be at the upper limit of the scale even before the consultation, we deleted it from the outcome measures.

Participant flow

A total of 152 patients with an affective disorder were consecutively approached for the study; 100 fulfilled all inclusion criteria and consented to participate in the study (19 patients did not speak German at all; 33 refused to participate). Of these 100 patients, 51 (51%) were randomly assigned to the intervention and 49 (49%) to the control condition.

Data analysis

Means and standard deviations were calculated for normally distributed data, and medians and ranges for non-normally distributed data. Categorical data are presented as absolute and relative frequencies. For group comparisons, t-tests and exact Mann-Whitney U tests were used for continuous data and χ² tests for categorical variables. Relations between continuous data were evaluated using Pearson correlation coefficients. A p value < .05 was considered significant.

Institutional review board approval and study registration

Institutional review board approval was obtained for the study (Ethikkommission der Technischen Universität München) and all patients gave their written informed consent after complete description of the study. The study was registered at clinicaltrials.gov (NCT01313013).

Results

Baseline data

As shown in Table 1, more women than men participated in the study and nearly half of the patients had a foreign background (mostly from Turkey, former Yugoslavia and Greece). Overall, patients were very satisfied with the psychiatric practice (before the intervention and the consultation). There were no significant differences (t-test, χ² test, all p > .05) between the intervention and the control group with regard to age, gender, education, participation preferences, duration of illness and severity of illness.

Table 1.

Baseline characteristics of the patient sample (N = 100).

	Intervention group(n = 51)	Control group (n = 49)	Z score/t values/χ²
Age, years, M (SD)	47.4 (9.6)	44.8 (9.4)	1.34 (0.18)^a
Gender, female	32 (63%)	29 (60%)	0.13 (.72)^b
Education			1.81 (.77)^b
– less than nine years	3 (6%)	3 (6%)
– nine years (Hauptschule)	17 (33%)	14 (29%)
– 10 years (Realschule)	14 (27%)	17 (35%)
– 12 or more years (Gymnasium)	15 (29%)	14 (29%)
Mother tongue, German	31 (61%)	25 (51%)	0.97 (0.33)^b
Diagnosis according to ICD-10			11.6 (0.64)^b
– Bipolar affective disorder (F31)	1 (2%)	3 (6%)
– Depressive episode (F32)	7 (14%)	7 (14%)
– Recurrent depressive disorder (F33)	36 (71%)	36 (73%)
– Persistent mood disorders (F34)	5 (10%)	3 (6%)
– Adjustment disorders (F43.21)	2 (4%)	0 (0%)
Duration of illness, years, M (SD)	3.9 (2.8)	3.2 (2.3)	1.22 (0.23)^a
Illness severity (CGI), M (SD)	3.7 (1.0)	3.7 (0.9)	0.06 (0.95)^a
Kind of appointment (%)			1.61 (0.20)^b
– Planned appointment	41 (80%)	34 (69%)
– Emergency appointment	10 (20%)	15 (31%)
Participation preferences (API score), M (SD)	9.2 (3.2)	9.6 (3.1)	−0.62 (0.54)^a
Satisfaction with psychiatrist, median (range)	5.0 (3–5)	5.0 (3–5)	−0.76 (0.45)^c

t-test.

χ² test.

Mann-Whitney U test.

Topics raised in the QPS and acceptability

All 51 patients randomized to the intervention group received the QPS and worked through it with the help of the research assistant. Thirty-six of the 51 patients of the intervention group (71%) followed the suggestion to make notes on the sheets about what they wanted to discuss with the psychiatrist during the consultation and 47 patients (92%) ticked one or more of the 15 questions listed on the QPS. Overall, patients ticked a mean of 3.7 (SD = 2.5) questions. The most frequently selected topics were questions about sleep disturbances (n = 25), request for a change in medication (n = 24), strategies to increase quality of life (n = 23) and coping with a worsening of the affective symptoms (n = 22). Questions about working or living conditions were rarely selected (n < 5).

Patients with a mother tongue other than German (Pearson correlation, r = 0.40, p = .004), patients with more severe symptoms according to CGI-S (r = 0.31, p = .03) and patients with lower participation preferences (r = 0.38, p = .02) selected more questions from the list of topics than German patients, patients with fewer symptoms or patients with stronger participation preferences according to their API scores. Fifteen patients rated the QPS as not helpful, while 16 rated it as somewhat helpful and 20 patients as helpful.

Effects of the QPS on patients’ behaviour in and rating of the consultation

According to Table 2, all the comparisons between the intervention and the control group showed non-significant differences, indicating that the QPS did not have any effect on any of our measures from patient self-report, physician report or our ratings of the consultation.

Table 2.

Post-intervention group comparisons.

	Intervention (median, range)	Control (median, range)	Z score/t value (p value)
Patient self-report after the consultation
How much pressure of time was there in the consultation?^a	2.0 (1–5)	2.0 (1–4)	−1.40 (.16)^e
How satisfied are you with the consultation?^b	5.0 (4–5)	5.0 (3–5)	−1.11 (.27)^e
What influence did you have on what had been decided during the consultation? ^b	4.0 (2–5)	4.0 (1–5)	−1.13 (.26)^e
Who made the decisions in today’s consultation?^c	3.0 (1–5)	3.0 (1–5)	−1.97 (.84)^e
Physician’s report after the consultation
How much pressure of time was there in the consultation?^a	3.0 (2–5)	3.0 (2–4)	−1.30 (.19)^e
How satisfied are you with the consultation?^b	3.0 (3–4)	3.0 (2–4)	−0.73 (.46)^e
How demanding was the patient today?^a	3.0 (2–5)	3.0 (1–4)	−0.33 (.74)^e
Who made the decisions in today’s consultation?^c	3.0 (3)	3.0 (3–4)	−1.02 (.31)^e
Third-party assessment of the consultation
Duration of consultation (min:sec)	8:56 (4:25) (M, SD)	7:16 (4:56) (M, SD)	1.71 (.09)^d
Pressure of time^a	1.0 (0–4)	1.0 (0–4)	−2.18 (.03)^e
Dominant behaviour of physician^a	2.0 (0–4)	3.0 (0–4)	−2.15 (.03)^e
Dominant behaviour of patient^a	1.0 (0–4)	1.0 (0–4)	−0.75 (.46)^e
Physician shows interest, raises questions^a	2.0 (0–3)	2.0 (0–4)	−0.62 (.54)^e
Patient shows interest, raises questions^a	2.0 (0–4)	1.0 (0–3)	−1.01 (.31)^e
Who made the decisions in today’s consultation?^c	2.0 (1–4)	2.0 (1–4)	−0.97 (.33)^e
Number of topics discussed in the consultation	5.9 (2.6) (M, SD)	5.5 (2.3) (M, SD)	0.85 (.40)^d
Number of topics raised by the patient	3.5 (2.3) (M, SD)	2.9 (2.3) (M, SD)	1.51 (.13)^d

Note: The two main outcome variables are shaded in grey.

Answers range from 0 = ‘not at all’ to 5 = ‘markedly’.

Answers range from 1 = ‘not at all’ to 5 = ‘completely’.

Answers range from 1 = ‘only the physician’ to 5 = ‘only the patient’.

t-test.

Mann-Whitney U test.

Regarding the question of ‘Who made the decisions in today’s consultation?’, patients as well as the psychiatrist rated a mean of about 3 (shared decision), whereas the external raters of the audio-taped consultations rated a mean of 2 (more power on the doctor’s side) (Table 2).

There were also no significant differences between intervention and control patients regarding the topics discussed (with the exception of ‘somatic illnesses’, χ² = 6.8, p = .02) or the number of topics brought up by the patients (all χ² tests: p > .05; Table 3). Overall, the most frequently discussed topics during the consultation were medication, patients’ complaints, somatic illnesses and problems with family members, authorities or insurance companies. As can be seen from Table 3, many of these issues (especially the more seldom ones) were brought up by the patients.

Table 3.

Issues discussed in the consultation.

Issue	Intervention (n = 51)		Control (n = 49)
	Number (discussed in x% of consultations)	% raised by patient	Number (discussed in x% of consultations)	% raised by patient
Diagnosis	13 (25)	62	16 (33)	50
Prognosis	1 (2)	100	2 (4)	50
Other (somatic) illnesses	32 (63)	72	18 (37)	56
Inpatient treatment	15 (29)	53	22 (45)	64
Medication	43 (84)	28	44 (90)	30
Side effects	19 (37)	47	16 (33)	31
Psychotherapy	28 (55)	21	32 (65)	31
Specific complaints	33 (65)	79	27 (55)	81
Suicidal thoughts	4 (8)	100	2 (4)	100
Measures to increase quality of life	19 (37)	47	17 (35)	59
Alcohol, illegal drugs	7 (14)	86	3 (6)	33
Problems at work	16 (31)	69	17 (35)	65
Problems with housing	7 (14)	100	4 (8)	100
Loneliness	3 (6)	100	2 (4)	100
Sexual problems	1 (2)	100	1 (2)	100
Problems with family/partner	25 (49)	76	20 (41)	65
Financial problems	4 (8)	100	5 (10)	100
Contact with authorities, insurance companies	33 (65)	48	23 (47)	39
Sum	303	57	270	55

Finally, the number of items marked by the patients in the QPS did not correlate significantly with the number of topics brought up by the patients in the consultation (r = 0.07, p = .65)

Discussion

Contrary to some research from somatic medicine (Harrington et al., 2004), our intervention with a QPS had no significant influence on patient behaviour in the consultation or on their satisfaction with the consultation.

The reasons for this finding may be manifold. The first reason might be the patient sample studied. Since we systematically excluded all patients who had attended for their first appointment with the psychiatrist, the interaction between patient and psychiatrist might already have been well established and critical topics may have been frequently discussed in the past. This pattern might have reduced not only the number of issues to be raised by the patient in the actual consultation, but also the feeling of the patient that he/she had already provided the important information and could accordingly rely on the psychiatrist’s knowledge to a large extent.

In addition, it is often questioned whether patients suffering from depression (although our sample was only mildly to moderately ill) might by definition be rather passive and not motivated to exhibit active behaviour including speaking up, being assertive or contradicting the psychiatrist. While depressive symptoms might in fact interact with patients’ behaviour, other research has shown that depressed patients’ participation preferences do not considerably differ from those of other patient groups (Hamann et al., 2007) and that communication skills training for general practitioners works well with regard to engaging patients with newly diagnosed depression in medical decision-making to a larger extent (Loh et al., 2007).

Thus, the diagnosis of depression itself might not inhibit patients to be activated and a QPS might work better for patients outside a well-established doctor–patient relationship (e.g. newly diagnosed patients or inpatients). There, a QPS might help to establish an active communication between patients and physicians, similar to the findings from somatic medicine.

Second, many patients (as well as the psychiatrist) felt that in the consultations something like ‘shared decision-making’ had happened, although the external observers rated rather paternalistic decision-making. Therefore, patients’ motivation to participate more actively might be have been limited by their impression that they already were active decision-makers.

Finally, the reason for the lack of effects might be due to the intervention itself. On the one hand, many patients did not judge the QPS as helpful; they probably felt that they did not need any further support for their consultations and consequently did not use the QPS in the consultation, even if they had marked a significant number of questions that they wanted to address. On the other hand, QPSs have been shown to be more efficacious when physicians actively refer to the QPS and do not wait until patients come up with it (Brown et al., 1999).

Regarding our originally planned outcome measure ‘satisfaction’, which has been shown to improve after interventions with QPSs in some studies from general medicine, we had the ‘problem’ of a massive ceiling effect even before the consultations, which made any improvements after the consultation impossible.

The finding that there was a trend for less dominant behaviour of the psychiatrist and less time pressure in the intervention group should be treated with caution since we did not correct for multiple testing and differences were small. One might speculate that intervention group patients were more structured in their consultations (since they were prepared by the QPS) and therefore ‘wasted’ less time talking about less important issues. Control group patients might have induced more dominant behaviour of the psychiatrist who might have tried to keep the consultation time short.

Implications for clinical practice

As our intervention did not show any effects on the consultation and patients’ attitudes it cannot be recommended for use, at least for the setting (well-established and satisfying doctor–patient relationship) and the indication (patients with mild to moderate depression) studied.

Thus, it must be questioned whether interventions for patient activation are (1) at all desirable/necessary in the setting studied (or similar ones) and, if so, (2) how patients with depressive disorders could actually be motivated to behave more actively in the consultation.

Regarding the general need or desirability of patient activation, one needs to weigh the potential cost and benefits of a deeper patient engagement in medical decision-making. While patient participation can be seen as worthwhile per se (Drake & Deegan, 2009) it is often seen as a prerequisite for patient satisfaction and adherence (Hamann et al., 2003). Although we did not study adherence and non-adherence in our study, one may assume similar rates for our sample as those found in other studies (e.g. ten Doesschate, Bockting & Schene, 2009). Thus, of the highly satisfied patients in our study, about 50% could be expected to be or to become non-adherent and show poor outcomes due to non-adherence. Obviously, many patients, while being perfectly satisfied with their psychiatrist, sometimes decide to discontinue with medication. A better involvement of the patients in the consultation and especially the decision-making processes might, therefore, have positive consequences regarding adherence (or at least the detection of non-adherence).

As our study has shown, it is, however, not easy to make patients behave more actively. If one wants to implement more active patient behaviour, a more complex intervention than a QPS might be necessary. A first step could be the implementation of interventions involving both parts of the doctor–patient dyad to facilitate better discussions about, for example, unmet needs of the patient (e.g. Priebe et al. 2007; van Os & Triffaux, 2008). While such interventions have been shown to have more effects than just a QPS, they have the weakness that they must be implemented by doctors, meaning that in daily practice they may be implementable in only a few cases (Holmes-Rovner et al., 2000).

Another approach would be to offer more specific interventions such as decision aids that focus on selected decisions like choosing between psychotherapy and drug treatment or between different modes of drug treatment. Decision aids have been shown to be effective both in somatic medicine and mental health (Duncan, Best & Hagen, 2010; O’Connor et al., 2009), but they exist for very few conditions and have to be revised regularly as they should present current evidence to patients.

A possible third approach would be the offer of more intense interventions for patients, such as training in doctor– patient communication or patient seminars. In somatic medicine, such interventions have also been shown to yield impressing effects on patients’ adherence (e.g. Cegala, Marinelli & Post, 2000; Cegala, Post & McClure 2001). In addition, there is new evidence to suggest that patient seminars to develop communication skills can also be implemented in mental health settings (Hamann et al., 2011).

Limitations

Our study had several limitations that need to be acknowledged. First, we recruited patients at a single psychiatric practice led by a single psychiatrist. While this allowed us to reduce variance regarding the psychiatrist, it limits the generalizability of the results to other settings. In addition, the psychiatrist, who was not blinded to the study protocol, might have biased the study results by ‘over-sharing’ decisions with patients in both groups, which might have led to a sharing atmosphere without relation to the control or study condition.

Second, many patients refused to participate in the trial; therefore, the more dissatisfied patients had probably not been included in the study.

Finally, we had decided neither to oblige patients to present the QPS to the psychiatrist nor request the psychiatrist to discuss the QPS with every patient. We made this choice to meet ‘naturalistic’ conditions in which patients may present the QPS or not (as they wish). An obligation to discuss all the issues raised in the QPS might have resulted in more active patient behaviour.

Conclusion

A QPS seems not to be effective in inducing more active behaviour for depressed outpatients within an established doctor–patient dyad, but it may help save consultation time. Achieving more active patient behaviour in the consultation might require more complex interventions such as patient seminars, the introduction of more specific decision aids or a structured engagement of the consulting physician.

Footnotes

Appendix

Funding

This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.

References

Albada

van Dulmen

Ausems

M. G.

Bensing

J. M.

(2012). A pre-visit website with question prompt sheet for counselees facilitates communication in the first consultation for breast cancer genetic counseling: Findings from a randomized controlled trial. Genetic Medicine, 14, 535–542.

Brown

Butow

P. N.

Boyer

M. J.

Tattersall

M. H.

(1999). Promoting patient participation in the cancer consultation: Evaluation of a prompt sheet and coaching in question-asking. British Journal of Cancer, 80, 242–248.

Cegala

D. J.

Marinelli

Post

(2000). The effects of patient communication skills training on compliance. Archives of Family Medicine, 9, 57–64.

Cegala

D. J.

Post

D. M.

McClure

(2001). The effects of patient communication skills training on the discourse of older patients during a primary care interview. Journal of the American Geriatrics Society, 49, 1505–1511.

Drake

R. E.

Deegan

P. E.

(2009). Shared decision making is an ethical imperative. Psychiatric Services, 60, 1007.

Duncan

Best

Hagen

(2010). Shared decision making interventions for people with mental health conditions. Cochrane Database of Systematic Reviews, 20, CD007297. doi: 10.1002/14651858.CD007297.pub2.

Elwyn

Edwards

Hood

Robling

Atwell

Russell

… Grol

(2004). Achieving involvement: Process outcomes from a cluster randomized trial of shared decision making skill development and use of risk communication aids in general practice. Family Practice, 21, 337–346.

Ende

Kazis

Ash

Moskowitz

M. A.

(1989). Measuring patients’ desire for autonomy: Decision making and information-seeking preferences among medical patients. Journal of General Internal Medicine, 4, 23–30.

Fleissig

Glasser

Lloyd

(1999). Encouraging out-patients to make the most of their first hospital appointment: To what extent can a written prompt help patients get the information they want? Patient Education and Counseling, 38, 69–79.

10.

Hamann

Langer

Winkler

Busch

Cohen

Leucht

Kissling

(2006). Shared decision making for in-patients with schizophrenia. Acta Psychiatrica Scandinavica, 114, 265–273.

11.

Hamann

Leucht

Kissling

(2003). Shared decision making in psychiatry. Acta Psychiatrica Scandinavica, 107, 403–409.

12.

Hamann

Mendel

Meier

Asani

Pausch

Leucht

Kissling

(2011). ‘How to speak to your psychiatrist’: A shared decision making training for inpatients with schizophrenia. Psychiatric Services, 62, 1218–1221.

13.

Hamann

Neuner

Kasper

Vodermaier

Loh

Deinzer

… Härter

(2007). Participation preferences of patients with acute and chronic conditions. Health Expectations, 10, 358–363.

14.

Harrington

Noble

L. M.

Newman

S. P.

(2004). Improving patients’ communication with doctors: A systematic review of intervention studies. Patient Education and Counseling, 52, 7–16.

15.

Holmes-Rovner

Valade

Orlowski

Draus

Nabozny-Valerio

Keiser

(2000). Implementing shared decision-making in routine practice: Barriers and opportunities. Health Expectations, 3, 182–191.

16.

Little

Dorward

Warner

Moore

Stephens

Senior

Kendrick

(2004). Randomised controlled trial of effect of leaflets to empower patients in consultations in primary care. British Medical Journal, 328. doi: https://dx-doi-org.web.bisu.edu.cn/10.1136/bmj.37999.716157.44.

17.

Loh

Simon

Wills

C. E.

Kriston

Niebling

Härter

(2007). The effects of a shared decision-making intervention in primary care of depression: A cluster-randomized controlled trial. Patient Education and Counseling, 67, 324–332.

18.

LUD (Leitliniengruppe Unipolare Depression) (ed.). (2009). S3-Leitlinie/Nationale Versorgungsleitlinie Unipolare Depression. Berlin: DGPPN.

19.

O’Connor

A. M.

Bennett

C. L.

Stacey

Barry

Col

N. F.

Eden

K. B.

… Rovner

(2009). Decision aids for people facing health treatment or screening decisions. Cochrane Database of Systematic Reviews, 3, CD001431. doi: 10.1002/14651858.CD001431.pub2.

20.

Priebe

McCabe

Bullenkamp

Hansson

Lauber

Martinez-Leal

… Wright

D. J.

(2007). Structured patient–clinician communication and 1-year outcome in community mental healthcare: Cluster randomised controlled trial. British Journal of Psychiatry, 191, 420–426.

21.

Roter

(2000). The enduring and evolving nature of the patient–physician relationship. Patient Education and Counseling, 39, 5–15.

22.

Roter

D. L.

Larson

(2002). The Roter Interaction Analysis System (RIAS): Utility and flexibility for analysis of medical interactions. Patient Education and Counseling, 46, 243–251.

23.

Simon

Kriston

Loh

Spies

Scheibler

Wills

Härter

(2010). Confirmatory factor analysis and recommendations for improvement of the Autonomy-Preference-Index (API). Health Expectations, 13, 234–243.

24.

Stewart

M. A.

(1995). Effective physician–patient communication and health outcomes: A review. Canadian Medical Association Journal, 152, 1423–1433.

25.

ten Doesschate

M. C.

Bockting

C. L.

Schene

A. H.

(2009). Adherence to continuation and maintenance antidepressant use in recurrent depression. Journal of Affective Disorders, 115, 167–170.

26.

van Os

Triffaux

J. M.

(2008). Evidence that the Two-Way Communication Checklist identifies patient–doctor needs discordance resulting in better 6-month outcome. Acta Psychiatrica Scandinavica, 118, 322–326.