The developed DA was then field-tested during the SDM process in a clinical setting described below.

2.2.1 SDM programme for psychiatric outpatients

We conducted field testing of the DA using a clinical SDM process that we developed and reported previously.15 The SDM programme comprises the three steps as described below.

Step 1. Initial consultation: Option presentation consultation: After the psychiatrist informs the adult of the diagnosis as ADHD, further treatment options such as watchful waiting with coping skills or additional medication with coping skills are shared between the adult and the psychiatrist using the DA booklet.

Step 2. Deliberation outside the service: The participant goes home with the DA, and freely reviews and considers it.

Step 3. Decision-making consultation: At the next appointment, the participant sees the psychiatrist. Both parties discuss the treatment options using the DA and finally decide which treatment can be further taken by the participant.

2.2.2 Participants

Participants were recruited from three outpatient psychiatric services in Tokyo from March 2019 to February 2020. For the inclusion criteria, we included those who visited the psychiatric services and met the following criteria: (i) outpatients aged 20 years or older; (ii) received a first-time ADHD diagnosis followed by Diagnostic and Statistical Manual of Mental Disorders Fifth Edition (DSM-5).1 The exclusion criteria were as follows: (i) participants who have already been diagnosed with ADHD; (ii) current substance dependence and abuse; and (iii) unable to speak Japanese. Participants requiring hospitalization in the psychiatric ward due to symptom severity or suicidality were also excluded and referred to applicable psychiatric services. Seventeen individuals who fulfilled the inclusion criteria were approached after the initial consultation. The sample size was selected in accordance with the methods used in the DA literature for field testing.16-18 Two of the 17 individuals withdrew their agreements immediately after providing the written consent: one moved to another service and the other could not revisit the service owing to a weather disaster. Participants filled out the questionnaires immediately after the initial consultation (baseline), immediately after the decision-making consultation, and in the 1- to 2-month follow-up at the outpatient services.

2.2.3 Measurements

Baseline—filled the questionnaires immediately after the initial consultation

(a) Preferences for roles in decision making: We used the Control Preference Scale (CPS)19 to assess the participants’ preferences for roles in decision making. We used the Japanese translation version of the CPS. The CPS comprises the following items: (i) clinician should ‘make the decision using all that is known about the medicines’; (ii) clinician should ‘make the decision but strongly consider [patient's] opinion’; (iii) patient and clinician should ‘make the decision together, on an equal basis’; (iv) patient should ‘make the decision but strongly consider the clinician's opinion’; and (v) patient should ‘make the decision using all the patient knows or learns about the medicines’. The previous study using the Japanese version of the CPS demonstrated inter-rater reliability with a substantial level of agreement (κ coefficients = 0.61).20

(b) Decisional conflict: We assessed decisional conflict, defined as uncertainty about the course of action to take when choosing among several medical procedures,21 using the Japanese version of the decisional conflict scale (DCS).22 The DCS comprises 16 items and the following five subscales: feeling informed, having clear values about benefits and risks, support to take the decision, uncertainty, and perceived effectiveness of the decision. The items are scored from 0 (strongly agree) to 4 (strongly disagree), with higher scores indicating a higher decisional conflict. The total scores range from 0 (no conflict) to 100 (maximal conflict). Scores lower than 25 are associated with implementing decisions, and those above 37.5 are associated with decision delay or feeling unsure about implementation.21 The scores on the subscales of the Japanese version of DCS demonstrated high internal consistency (Cronbach's α = .84–.96).22

(c) Knowledge: Participants' knowledge regarding ADHD and its treatment was measured using a questionnaire containing 18 items with the response options ‘true’, ‘false’, and ‘don't know’. Of the 18 items, 13 items were presented in the DA and 5 items were not. The number of items, which were answered correctly within each subgroup, that is, the items presented in the DA and those not presented in the DA, was calculated. We developed this questionnaire for this study based on knowledge measures used in the previous ADHD study12 with the developer's permission. We used 13 items from measures in Brinkman et al.12 and 5 of our own items. Medication treatment options available for ADHD vary widely from country to country. Our own items were based on information regarding medication treatment options currently approved and marketed in Japan. The questionnaire was reviewed by the authors, including ADHD experts for content validity.

(d) ADHD severity: We assessed ADHD severity using Conners' Adult ADHD Rating Scales-Self-Report: Long version (CAARS-S: L).23-25 The CAARS-S: L includes 66 questions and consists of eight subscales: inattention/memory problems, hyperactivity/restlessness, impulsivity/emotional lability, problems with self-concept, DSM-IV inattentiveness symptoms, DSM-IV hyperactivity–impulsivity symptoms, DSM-IV ADHD symptoms total and ADHD Index. The answers were evaluated using a 4-point Likert scale with four options: 0: almost never, never; 1: occasionally, sometimes; 2: most of the times, usually; 3: very often, always. We used the Japanese version of CAARS-S: L.23-25 The scores on the subscales of the Japanese version of CAARS-S: L demonstrated good internal consistency (Cronbach's α = .73–.88).25

Post—filled the questionnaires immediately after decision-making consultation

(a) How decisions were made: We used the scales developed by Strull et al.19 to assess how the decision was made. The scales comprise the following items: (i) clinician ‘made the decision using all that is known about the medicines’; (ii) clinician ‘made the decision but strongly consider [patient's] opinion’; (iii) clinician and patient ‘made the decision together, on an equal basis’; (iv) Patient ‘made the decision but strongly consider the clinician's opinion’; and (v) patient ‘made the decision using all [the patient] knows or learns about the medicines’.

(b) Decisional conflict: We assessed decisional conflict using the Japanese version of the DCS.22

(c) Knowledge: Participants' knowledge regarding ADHD and its treatment was measured using exactly the same questionnaire that was administered after the initial consultation.

One- to 2-month follow-up

ADHD severity: Within the 1–2-month follow-up after the decision-making consultation, we assessed ADHD severity using the Japanese version of CAARS-S: L24, 25 as a long-term follow-up.

2.2.4 Data analysis

We assessed the concordance between preferences for a role in decision-making and how decisions were made. We combined the three categories reflecting participant involvement into one category and contrasted it with the categories of the clinician-alone and patient-alone decision-making. Gwet's Agreement Coefficient (AC1) was used to determine the concordance.

For the knowledge, decisional conflict and ADHD severity, we used a paired t-test to evaluate for significant differences between the before- and after-SDM process if the data were normally distributed. Otherwise, we used a Wilcoxon signed-rank test. The mean score and associated standard deviation (SD) were also calculated.

This study protocol was approved by the Ethics Board at St. Luke's International University (18-A055).