Exploring factors influencing initiation, implementation and discontinuation of medications in adults with ADHD

Abstract Background Adherence to ADHD medication is a complex phenomenon as the decision to adhere is influenced by a range of factors. To design tailored interventions to promote adherence, it is important to understand the factors that influence adherence in the context of its three phases: initiation, implementation and discontinuation. Objective The objective of this study was to explore the phase‐specific factors that influence adherence to medication in adults who have a diagnosis of ADHD. Methods Three focus groups (FGs) were conducted with twenty adults with ADHD in different metropolitan areas of Sydney, Australia. FGs were transcribed verbatim and thematically analysed. Results Participants’ decision to initiate medication (the initiation phase) was influenced by their perceived needs (desire to improve academic and social functioning) and concerns (fear of side‐effects) about medication following a similar process as defined by the Necessity‐Concerns Framework (NCF). The balance between benefits of medication (needs) and side‐effects (concerns) continued to determine participants’ daily medication‐taking (the implementation phase) and persistence (or discontinuation) with their medication. Forgetfulness and stigma were reported as concerns negatively impacting the implementation phase, while medication cost and dependence influenced the discontinuation phase of adherence. Conclusions Adults’ decision to initiate, continue or discontinue medication is influenced by a range of factors; some are unique to each phase while some are common across the phases. Participants balanced the needs for the medication against their concerns in determining whether to adhere to medication at each phase. It appears that the NCF has applicability when decision making about medication is explored at the three phases of adherence.


| BACKG ROU N D
Attention-deficit hyperactivity disorder (ADHD) is a neurodevelopmental disorder marked by persistent inattention and/or hyperactivity-impulsivity. 1 Although ADHD is predominantly a childhood disorder, it often continues into adulthood, with reports that approximately two-thirds of children with ADHD continue to have signs and symptoms of ADHD into adulthood, either unabated or with partial improvement. 1 The reported prevalence in adults varies from 1.2% to 7.3% 2 and may even be higher as the disorder may be underdiagnosed in adults. 3,4 Evidence suggests that about 90% of adults with ADHD have at least one comorbid psychiatric disorder. 5 Moreover, these adults are at increased risk of alcohol dependency, drug misuse, car accidents and suicide. 6,7 Lack of focus, disorganization, forgetfulness, impulsivity, time management problems and relationship issues are all symptoms of ADHD in adults. 8 These symptoms have been associated with academic, occupational and social difficulties, 9,10 warranting the need for ADHD management. Despite the availability of behavioural management, pharmacotherapy plays an important role in the management of ADHD. 11 Attention-deficit hyperactivity disorder medications reduce the symptoms and improve daily functioning and quality of life of adults with ADHD. 12 However, adherence is a key factor associated with the effectiveness of ADHD medication and treatment outcomes.
Evidence suggests that adherence to medication among adults with ADHD is sub-optimal. A study reported that 88% of adults with ADHD were inconsistent in taking their medication, 13 and 41.2% of adults were non-adherent to ADHD medication. 10 Furthermore, adult patients with lower adherence to medication (less than 80%) had considerably higher ADHD severity than those with higher adherence (more than 80%). 14 Medication adherence is influenced by a number of factors, which can change as the patient progresses through the different phases of medication-taking. 15 To better understand the complexity of this journey, Vrijens et al 16 proposed the Ascertaining Barriers to Compliance (ABC) taxonomy. The ABC taxonomy provides a more transparent and rigorous approach to examine adherence in the context of its three phases: initiation, implementation and discontinuation. The initiation phase was defined as consumption of the first dose of a prescribed medication. The implementation phase was defined as the extent to which patients adhere to the prescribed dosing regimen from initiation until the last dose of the medication. The discontinuation phase was defined as the cessation of the prescribed medication for any given reason. 16 Furthermore, the ABC taxonomy proposed to identify all possible causes of non-adherence, including intentional and unintentional non-adherence, to better understand the complex and dynamic nature of adherence.
Patients' beliefs about medication are an important factor that determine patients' adherence to medication. 17 The Necessity-Concerns Framework (NCF) argues that patients' beliefs about medication adherence are based on their perceived need (necessity) and concerns about medication. 17 Patients are more likely to adhere if their perceived needs for the medication outweigh their concerns and vice versa. The NCF accounts for both intentional and unintentional non-adherence. 18 This framework has been successfully applied across a range of medical conditions 18 ; however, little is known about its use in adults with ADHD.
Despite the evidence for poor medication adherence in adults with ADHD, only a small number of studies have examined the factors influencing medication adherence. 10,12,[19][20][21] These studies have reported psychiatric comorbidity, treatment settings (specialist vs general physician), forgetfulness, gender, education, age and family history as the main factors that influence medication adherence. 10,12,[19][20][21] However, none of these studies have examined adherence in view of its dynamic nature, or in the context of its influence on the three phases of adherence. Evidence from different disease populations suggests that different factors can influence different phases of adherence. [22][23][24] The objective of this study was, therefore, to explore the factors that influence adherence to ADHD medication in the context of the three phases of adherence (initiation, implementation and discontinuation) in adults with ADHD.

| Study design
Given the exploratory nature of this study, a qualitative approach was used. 25 Focus groups (FGs) were preferred as they provide an opportunity to generate richer data through interactions and discussions between participants compared to individual interviews. 26 The reporting structure of this study was guided by the consolidated criteria for reporting qualitative research (COREQ) 27 (Appendix A).

| Study participants
After receiving approval from the institution's Human Research Ethics Committee (Project no. 2018/271), participants were recruited through a market research company. Convenience sampling was used to recruit participants from different geographical locations (centre, south and west) of Sydney to ensure diversity in the participants' socio-economic characteristics. Each participant was reimbursed AUD80 for their time and travel expense. FGs were conducted in English. Participants were recruited if they were aged between 18 and 65 years, diagnosed with ADHD, prescribed ADHD medication and able to participate without the need for an interpreter. Twenty adults with ADHD participated in three FGs (Table 1).
Two FGs had seven participants each while one FG had six participants. Most participants initiated medication during adulthood (n = 8) while some initiated in their childhood (n = 6) and some in their adolescence (n = 6).

| Focus group discussions
Prior to attending the FGs, potential participants were provided with the participant information statement (which included detailed information about the study, and other issues such as confidentiality, anonymity and voluntary nature of participation) and consent form. They were provided the same information at the beginning of the FGs to ensure that they had read the information and had the opportunity to ask questions about the study. Participants were asked to complete a brief demographic information form (Appendix B). All participants provided written consent prior to commencing the FG discussions.
FGs were conducted between August 2018 and January 2019 at venues intended for FGs. Efforts were made to provide an informal and relaxed but professional environment to ensure that participants felt comfortable. Two FGs were moderated by an experienced facilitator (PA). MUK acted as an observer and took field notes on the key points raised during the group discussions. One FG was moderated by MUK under the supervision of PA. To minimize moderator bias and ensure consistency between the FGs, a semi-structured FG protocol was developed by reviewing previous studies 10,12,[19][20][21] in line with the study objectives (Appendix C). The format and style of the questions were in accordance with the criteria proposed by Krueger and Casey. 28 The content of the questions was based on the ABC taxonomy. 16 Each FG lasted approximately 60 minutes. All FGs were audio-recorded with the permission of the participants and transcribed verbatim.

| Data analyses
The data were analysed by MUK and PA using a framework for thematic analysis. 29 The NVIVO 11 Pro computer software was used for data management. The process of data analysis started during the phase of data collection. After each session, MUK and PA discussed the themes that arose during the discussions. No new codes were generated from the 3rd FG discussions, and it was decided by the authors that data saturation had been reached. 30 Thematic analysis was performed in six steps. In the first step, MUK read the FG transcripts and listened to the audio-recordings numerous times to become familiar with the data. In the second step, the initial codes were generated by MUK through inductive coding. A latent approach was used for coding the transcripts to capture the conceptual and underlying meaning of the data. Each code was grouped under one of the three phases of adherence based on the ABC taxonomy. In the third step, the codes were merged to form broader themes within each of the three phases following a discussion between MUK and PA. A few themes overlapped between the implementation and discontinuation phase, which were combined under the two phases to avoid duplication.
In the fourth step, the themes were reviewed by MUK and PA for further refinement in the context of the overall data set. The iterative process of rearranging codes and themes was reflexive, 31 TA B L E 1 Summary of participants' demographic information (n = 20)

Median (IQR)
Age (in years) 34 (11.5) Two participants did not provide a response. c According to the Australian Family Law Act 1975, de facto is defined as a relationship between two people who are living together on a genuine domestic basis and are not legally married or related by family. d Some participants chose more than one response. moving back and forth through the transcripts to search for common and recurrent themes, which were then named and finalized after a thorough discussion between the researchers in the fifth step of the analysis. Disagreements between the researchers were resolved by discussion until a consensus was reached. In the final step (manuscript preparation), the themes were carefully presented to ensure that the research objectives were answered in the context of the original data set. The NCF was used to identify participants' needs for and concerns about medication at the three phases of adherence.

Number of participants
Similar to the concept of validity and reliability of qualitative study instruments, scholars emphasize establishing trustworthiness in qualitative research. In this study, trustworthiness of the findings was established during each step of the thematic analyses by following the strategies suggested by Nowell et al, 32 and guided by the concept of trustworthiness presented by Lincoln and Guba. 33 The means used to establish trustworthiness are presented in Table 2.

| RE SULTS
Numerous factors were reported that influenced participants' perceived need (necessity beliefs) and their concerns about medication ( Table 3). The findings are presented under two major themes in accordance with the NCF. Within each theme, factors have been categorized based on their influence on the three phases of adherence.
The two major themes are the following: 1. participants' need for medication 2. participants' concerns about medication

| The initiation phase
At the initiation phase, participants' perceived need for medication was influenced by two factors: challenges of dealing with ADHD and trust in physician. Both factors increased the necessity beliefs of participants that encouraged them to commence therapy.

Desire to improve ADHD outcomes
The persistent academic, occupational and social challenges of living a life with ADHD increased the necessity beliefs of the participants to take medication. For some participants, these challenges started from childhood with problems at school due to inattention and aggressive behaviour. During adolescence, lack of concentration, aggressive behaviour and the strong feeling of rebellion were the common challenges. During adulthood, one of the most common issues was poor interpersonal communication. Participants reported that they got very loud while talking and their voice (tone and pitch) changed during a conversation without realizing that they were sounding different. This consequently impacted their social interactions. Participants felt there was no empathy for people suffering from ADHD. This was not only limited to the general public but also to family members and work colleagues.
Mind wandering was another worrying issue that participants had to deal with, which they felt impacted their ability to perform daily chores, and often put them in an embarrassing situation, especially during conversations. Participants reported that these challenges were draining their energy daily, and they were looking for help. They wanted to improve their behaviour and their academic and social performance. The desire to improve outcomes enhanced their need for medication.

Trust in physician
Trust in physician was another factor that positively contributed to participants' perceived need for medication. Those who initiated their medication during adulthood (n = 8) reported that their trust in the physician helped them in deciding to initiate medication as they believed that the physician would not prescribe any medication that would harm them.
Yes, it was fine with me (to start medication). I trusted he wouldn't prescribe a drug that is bad for me.

| The implementation and discontinuation phase
The factors influencing participants' need for medication at the implementation and discontinuation phase have been presented together as there were a few factors that overlapped between the two phases, while some factors were unique to each phase.

Benefits of medication
Once the medication was initiated, some participants experienced the benefits of the medication and it reinforced their beliefs about the need for medication. Participants who benefited were more inclined to continue taking their medication. However, participants differed in what they regarded as beneficial outcomes of medication-taking. For example, improved concentration and more control over ADHD and their life were two important beneficial outcomes reported.
It improves concentration quite exponentially and feels like more control. I can be more organised.
Some participants mentioned that they take medication as it makes them more normal, settles their anxiety and helps them function better.

Medication stigma
Stigma experienced by participants because of the medication they took, negatively impacted their adherence to their medication.
Participants who took medication as a child or adolescent reported facing stigma mainly at school. As adults, participants discussed the stigma felt by people with ADHD, and specifically, the stigma they felt from people who knew they were on medication such as people at their workplace.
As you were saying about kids at school, the most com-

| D ISCUSS I ON
The findings of this study provide insight into the factors that influence the decision to adhere to medication in adults with ADHD. Adults reported medication-taking as a journey that started from their negative experiences with ADHD including academic, occupational or social challenges, which encouraged them to initiate medication. These experiences formed the necessity beliefs that they need medication to improve ADHD outcomes and these potential outcomes encouraged them to initiate medication. However, intentional non-adherence was also noticed at this phase as some participants had more doubts to- shows that a good relationship with a physician can successfully influence a person's decision to initiate medication. 37,38 Once initiated, the decision to continue medication was influenced by participants' experiences with medication. The results show that participants constantly balance their positive experiences (benefits) against negative experiences (concerns) to decide on continuing medication. However, this decision is not straightforward for adults as they struggle to find the right balance and often trade-off between benefits and concerns about medication, consistent with previous research. 39 The positive experience is mainly influenced by the benefits of medication that improve their educational, occupational, social and family functioning while their negative experience is primarily related to their experience of having medication side-effects and stigma. These findings suggest that adults whose positive experience with medication outweighs the negative experience are more likely to adhere to medication at the implementation phase. Adherence at the implementation phase could be improved by ensuring that the most appropriate medication and dosing regimen has been prescribed. For example, non-adherence to medication due to stigma at the workplace could be addressed by giving the person a long-acting medication for ADHD.
As the balance shifts more towards the negative experiences, that is, if the desired benefits have not been achieved or the frequency or intensity of side-effects increases and are not worth the benefits, adults consider discontinuing medication. Our findings suggest that side-effects are the primary reason for discontinuation. This qualitative study confirms previous findings 14,44,45 and adds other factors such as dependency on medication, self-control and the availability of alternative options as factors influencing medication discontinuation. Fear of dependency was a unique factor that influenced participants' decision to discontinue taking medication.
Previous studies have reported that people who use stimulants have a high potential for abuse [46][47][48] ; however, there is limited research on whether adults with ADHD are aware of potential dependence on stimulants. Adults' fear of dependency contributes to their concerns about medication which may result in early discontinuation of medication. Various strategies, such as education about stimulants and the use of non-stimulants, could be employed to address adults' concerns about dependency to stimulants and improve adherence.
Self-regulation was another factor that shaped the needs of participants at both the implementation and discontinuation phase of adherence. Adults' capacity to self-control symptoms of ADHD is relatively better than children and adolescents with ADHD. 49 Our findings suggest that adults may feel better off the medication compared to being on the medication if they can self-regulate their symptoms. This is particularly important for those whose concerns are higher than their perceived need for medication. Those who can self-regulate could be trialled for a period without medication; however, the decision to go off medication should be taken in consultation with a physician. This argument is supported by other researchers that suggest discontinuation of medication, particularly if the medication is sub-optimal in terms of its benefits or if the patient is experiencing severe side-effects. 17 The findings of this study reveal factors that are unique to each phase, and factors that are common across the phases. This is

| Implications of findings
This study has several clinical and research implications. This study reports the factors that influence medication adherence at the three phases of medication-taking, which can help physicians in implement-

| Limitations
The findings of this study should be interpreted in the context of some limitations. As a qualitative study, the findings are not generalizable to other populations not represented in the study sample. Another potential limitation of this study is the small number of focus groups (n = 3) required to reach saturation; however, this is in line with previously published studies on adherence research. [50][51][52] Focus groups were used which may have prevented some people from talking about certain issues that they may feel are too private to divulge in front of others. Yet in these cases, the focus groups may have also acted as social support and allowed some participants to divulge more. Although efforts were made to ensure that all participants had an equal opportunity to share their opinions, the possibility of dominance bias cannot be completely ignored.

| CON CLUS IONS
Unique factors were identified that influenced each of the three phases of adherence while some common factors between the phases were also noted. The decision to adhere to medication was based on the balance between adults' perceived needs and concerns about medication as expected from the NCF. Desire to improve ADHD outcomes was the main motivation for adults to initiate medication.
Benefits of medication and its side-effects were the common factors that influenced the implementation and discontinuation phase of adherence. The negative factors of forgetfulness and medication stigma were unique to the implementation phase while dependency and cost were unique to the discontinuation phase of adherence.

CO N FLI C T O F I NTE R E S T
The authors have no conflict of interest to declare.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions. The codes were rearranged and grouped based on their relevance to one of the three phases of adherence described in the 'Ascertaining Barriers to Compliance' (ABC) framework for examining medication adherence.

O RCI D
The Necessity-Concerns Framework (NCF) was used to explore participants' needs for and concerns about medication at the three phases of adherence.

Sampling
How were participants selected? eg purposive, convenience, consecutive, snowball Participants were recruited by a market research company through convenience sampling.

Method of approach
How were participants approached? eg face-to-face, telephone, mail, email Participants were recruited by a market research company. The researchers were not involved in the recruitment process.
12. Sample size How many participants were in the study?
A total of three focus groups were conducted with twenty participants.

Non-participation
How many people refused to participate or dropped out? Reasons?
Data were not collected 14. Setting of data collection Where were the data collected? eg home, clinic, workplace Focus groups were conducted at venues intended for group discussions, and within easy access of the participants.

Presence of non-participants Was anyone else present besides the participants and researchers?
Only participants and researchers were present in the room where focus groups were being held. No other persons were allowed to enter the room during the discussions.

Description of sample
What are the important characteristics of the sample? eg demographic data, date Adults (18-65 years) who were diagnosed with ADHD and prescribed medication for ADHD were eligible to participate. See Table 2 in the manuscript for further information.

Interview guide
Were questions, prompts, guides provided by the authors? Was it pilot tested?
To minimise moderator bias and ensure consistency between the focus groups, a semi-structured focus group protocol was developed by reviewing previous studies (

Repeat interviews
Were repeat interviews carried out? If yes, how many?
No repeat focus groups were carried out. 19. Audio/visual recording Did the research use audio or visual recording to collect the data?
All focus groups were audio-recorded only after permission had been provided by the participants.

Field notes
Were field notes made during and/or after the interview or focus group?
MUK a male researcher observed and took field notes in two focus groups while PA was the observer in one focus group discussion.

Duration
What was the duration of the interviews or focus group?
Focus group discussions lasted approximately 60 minutes.

Data saturation
Was data saturation discussed?
No new codes were generated at the 3rd FG, and it was decided that data saturation had been reached after discussion between authors

Transcripts returned
Were transcripts returned to participants for comment and/or correction?
Transcripts were not returned to participants for comment or correction.

Number of data coders
How many data coders coded the data?
The transcripts were coded by MUK, and the coding discussed with PA.

A PPEN D I X B
Exploring factors influencing adherence to medication at initiation, implementation and discontinuation of medications in patients with ADHD

D EM O G R A PH I C I N FO R M ATI O N O F A D U LTS
Could you please complete the following questions:

Data and findings consistent
Was there consistency between the data presented and the findings?
Yes, see the Results section of the manuscript. Participants' quotes are consistent with the findings.

Clarity of major themes
Were major themes clearly presented in the findings?
The study findings are presented within two broad themes: 1. Participants' need for medication 2. Participants' concerns about medication

Clarity of minor themes
Is there a description of diverse cases or discussion of minor themes?
Within each broad theme, findings are discussed in the context of the three phases of adherence (initiation, implementation, discontinuation).

A PPEN D I X C
Exploring factors influencing adherence to medication at initiation, implementation and discontinuation of medications in patients with ADHD

I NTRO D U C TI O N
Scientific evidence suggests low levels of adherence to medicines by patients with attention-deficit hyperactivity disorder (referred to as ADHD). These low levels of adherence to medicines can result in poor health outcomes. In order to design interventions to improve adherence to medicines, researchers have to first understand the factors which contribute to non-adherence. There is limited information currently available about the factors that influence a person's medicine taking, from when a person starts taking their medicine, to their daily medicine taking, and to when they stop taking their medicine.

Aims
This study therefore aims to explore what factors affect medicine taking, from starting medicines to when they are stopped, in patients with ADHD.

Question 1
Think back to that time when you were first prescribed ADHD medications. How did you feel about starting medications?

Prompts
• What were the barriers?
• Name some barriers that first come to your mind and how would you rank them based on their influence?
• What encouraged you to start taking medications?
• Name some facilitators that first come to your mind and can you rank them in order of their influence?

Question 2
When did you start your medications?

Prompts
• How long did it take you to start your medication after you received the prescription?
• What were your thoughts about starting medications in that period?

Question 3
What was your first experience taking those medications?

Prompts
• Was it hard to take? Why?
• What made you comfortable taking those medications?

Question 4
What made you continue medications after initial use?

Prompts
• What were the facilitators?
• What were the barriers?
• Name one each facilitator and barrier that first comes to your mind

Question 5
Did you always take the medications as prescribed? Why?

Prompts
• Skipping doses intentionally or unintentionally?

Question 6
What encourages you to continue your medications as prescribed?

Prompts
• Observed medication effects • Education about disease and medications • Refill reminder

Questions 7
Have you discontinued taking medications? Why?

Prompts
• What were the facilitators?
• What were the barriers?
• Name one each facilitator and barrier that first comes to your mind • Is it for certain period? Why • Why would you restart medications?

Question 8
What can help you to start and continue taking medications as prescribed? Why?

Prompts
• What type of professionals you think can help you?
• Do you think healthcare professional can help?
• Which health professional you think can help you the most?
• If that professional contact you periodically, would that be helpful or a nuisance?
• How pharmacist can be helpful in the whole process of medication taking?