Development and Acceptability Testing of a Patient Decision Aid on Levodopa Intestinal Gel for Parkinson Disease

Abstract

Background

Parkinson’s disease (PD) is a neurodegenerative disease characterized by motor and nonmotor symptoms that worsen over time. In some cases, an advanced treatment may be needed. The use of levodopa-carbidopa intestinal gel (LCIG) is one of these options. However, deciding whether to receive it can be difficult. A patient decision aid (PDA), a tool designed to inform about treatment options, can help and promote patients’ participation in decision making.

Objectives

This study was conducted to develop a PDA on LCIG and assess its acceptability.

Methods

The International Patient Decision Aid Standards framework was used to develop the PDA. An advisory committee (n = 5) gave feedback on the PDA prototype. Acceptability was evaluated using a cross-sectional descriptive design. A convenience sample of 36 participants (including persons with PD receiving and not receiving LCIG, caregivers, and health care professionals) was used. Acceptability data, sociodemographics, and health literacy were collected using questionnaires and a focus group.

Results

Sample characteristics were a mean age of 64.4 y (s = 14 y), university level of education (46.7%), and duration of illness of less than 10 y (80%). The health literacy score was judged as very good ( $\overline{x}$ = 55.2/70, s = 7.3). Qualitative data analysis allowed for final adjustments to the published PDA version. Conclusions. This study is the first to report the development of a PDA on LCIG and its acceptability testing. Participants found the PDA to be useful and would recommend it. All health care professionals indicated they intended to use it in their practice. More research will be needed to evaluate the PDA’s implementation and its effects on users.

Highlights

• Deciding whether to opt for an advanced Parkinson’s disease treatment such as levodopa-carbidopa intestinal gel can be challenging for many patients.

• A patient decision aid on levodopa-carbidopa intestinal gel for Parkinson’s disease persons, caregivers, and health care professionals was developed to support a decision-making process.

• This article summarizes the steps used for its development and its acceptability testing.

Parkinson’s disease (PD) is a neurodegenerative disorder associated with a progressive loss of midbrain dopaminergic neurons, a neurotransmitter involved in the control of movement.^1,2 It is characterized by a range of motor and nonmotor symptoms that appear gradually and worsen over time, including rigidity, bradykinesia, tremors, postural instability, and cognitive decline.^1,3 Perhaps most alarmingly, the global age-adjusted prevalence of PD has increased by ∼22% from 1990 to 2016, ² leading some authors to describe the disease as an emerging pandemic.^3,4 As it progresses, PD affects quality of life, strains caregivers (CG), and contributes to a substantial socioeconomic burden. ⁴

Currently, no treatment can cure this disease or halt its progression. Thus, the first aim of treatment is to control symptoms and maintain the quality of life of patients. ¹

Therapeutic approaches for PD are based on several factors: severity of symptoms, level of functional autonomy, comorbidities, possible side effects, and the individual’s preferences.^5,6 It includes both pharmacologic and nonpharmacologic approaches. The latter refers to interventions to support adaptation and quality of life, ⁷ including physiotherapy, nursing care, occupational therapy, and speech therapy.

The pharmaceutical arsenal comprises different classes of medication, mostly acting on the dopamine system. Levodopa (L-dopa), taken as oral tablets, remains the first choice of treatment as it is particularly effective in controlling motor symptoms.^6,7 However, even with optimal dosage, this treatment can lead to motor fluctuations and dyskinesia. ⁶ As the disease progresses, symptoms become more difficult to manage. The therapeutic window in which the person is relieved of parkinsonian symptoms then becomes narrower. ⁸ Higher doses and combinations of treatments are then required to reach and maintain a target therapeutic dosage. ⁶ Also, the appearance of motor fluctuations and dyskinesias implies more frequent pharmacologic adjustments. ⁸ Alternatively, advanced treatment options may need to be considered during the course of the disease. These include deep brain stimulation (DBS), which requires surgery, and subcutaneous infusions of lisuride, apomorphine, or foslevodopa-foscarbidopa, which are not available in every country. ⁷ A more accessible advanced option is the use of LCIG administered continuously via a duodenal pump. This treatment can significantly improve motor symptoms and quality of life and alleviate the burden of care on the CG. ⁷ However, it is also associated with a relatively high frequency of adverse events, including those related to the endoscopic procedure required to insert the gastrojejunal tube or to the equipment used to administer it. Compared with DBS, little information is available concerning infusion therapies such as LCIG. ⁹ In fact, a lack of sufficient information and misconceptions appear to be prevalent among persons with PD, and this remains one of the most important barriers to achieve a shared decision-making (SDM) process.^10,11 Thus, only a small proportion of patients are informed about PD’s advanced treatment options, especially in the earlier course of the disease. ¹² For those reasons, deciding whether to opt for this treatment can be challenging for many patients.

As patients will be asked to make numerous decisions for symptoms management while the disease progresses ¹³ and to actively involve them in SDM, HCP need to rely on a supportive tool that could concisely present all the best scientific evidence regarding treatment options. ⁷ A patient decision aid (PDA) is a well-recognized tool for improving knowledge and expectations on the treatment options and increasing the level of satisfaction with the decision-making process, the accuracy of risk perception, and the congruence between the decision and patient’s personal values.^14,15 A PDA can be a written document such as a booklet, leaflet, or other media formats such as a video, an audio, or an online tool. ¹⁶ They are designed for and with patients to encourage their active participation in decision making about treatment or screening options. It explains the advantages and disadvantages of treatment or screening options and helps patients clarify their personal values regarding the characteristics and outcomes of a specific treatment. ¹⁵

As a PDA specifically on LCIG is not available in the literature, the development of such a tool would meet an unmet need for HCP, potential LCIG users, and their CG.

This article aims to summarize the steps used for the development of a PDA on LCIG and its acceptability testing.

Based on a user-centered approach, this project was conducted pursuing 3 main objectives: 1) develop a PDA prototype, 2) evaluate its acceptability, and 3) review and update its final version according to the results and feedback obtained from the acceptability testing.

Methods

Development of the PDA on LCIG

The authors referred to the International Patient Decision Aid Standards (IPDAS)^17,18 as well as the Université Laval’s “Conception d'outils d'aide à la prise de décision partagée” training course ¹⁹ as an evidence-based framework to support the development of the PDA prototype. Guidelines recommend involving patients and stakeholders when designing and developing a PDA. ²⁰ To develop a well-adopted PDA, an advisory committee (AC) was formed. It comprised 2 people with PD (1 of whom was in their third year of LCIG), 2 CG of PD persons receiving LCIG, and a nurse specialized in advanced PD (HCP). When forming the AC, the choice was made not to include a neurologist in order to limit social desirability bias. Indeed, neurologists available at this time were also involved in the care of some participants. In this context, those might have felt less free to express themselves in front of their doctor. However, as 1 neurologist was part of the research team, he was involved, with the other 2 research members, in the validation of the different phases of the PDA prototype development.

Guided by the research team members, the AC was called upon to provide feedback and preferences throughout the development of the PDA. This committee was asked to share comments and suggestions during 4 online group meetings spread out along the development process, from February 2021 to February 2022. As discussed by Bientzle and colleagues, ²¹ it is crucial to understand and consider the individual preferences of persons with PD and the impact of those on treatment decisions. Therefore, the information needs on advanced therapies for PD identified in the study by Nijhuis et al. ²² served as a starting point for discussion between the authors and the AC to better understand the user’s needs and to identify themes or topics that should be included in the PDA. Subsequently, to cover all the themes and topics identified to be included in the PDA, a supplementary documentary search and literature review were conducted to identify relevant publications. Among those, a literature review on the treatment of LCIG, published by the CADTH, ²³ was chosen as a flagship reference because of its rigor, relevance, and impartiality regarding recommendations of this treatment. Overall, AC went through 6 iterative cycles of review to refine the PDA prototype. As a second step, a focus group later walked through the draft version of each component of the PDA. As suggested by Vaisson et al., ²⁰ they were asked to comment on the wording, format, visual layout, and their ease of interacting with it. The draft was adjusted in accordance with their comments. Once the feedback reached saturation and AC appraised the final prototype, it was then ready for its initial acceptability field testing in the PD population. Detailed development steps of the PDA prototype and time frame are summarized at Figure 1.

Figure 1.

Development steps of the patient decision aid (PDA) prototype and time frame.

PDA on LCIG

The PDA on LCIG is intended for PD persons whose motor symptoms are poorly controlled with oral medication and to whom treatment with LCIG may be indicated. It is also designed to reach close relatives of these persons. The decision being supported by this tool aid refers as “whether or not to take levodopa-carbidopa intestinal gel (LCIG).” Practical aspects of LCIG treatment include information on pretreatment, administration modes, treatment starting options and steps, dosage adjustments, drug format, as well as daily routine and its management. It also states the advantages and disadvantages of the LCIG for PD persons as well as users and the CG experience with LCIG. At the end of the PDA, a short questionnaire called the SURE Test ²⁴ helps patients to measure their level of comfort about the decision being made. Finally, the PDA contains documentation about levels of scientific evidence and a glossary to facilitate its comprehension. The configuration of the PDA, including tables and figures, aims to summarize and outline information to ease the patient’s decision-making process. Lastly, the PDA on LCIG is mainly delivered and used as a supportive tool during consultation with the neurologist or other HCP specialized in PD. It is meant to be handed out to persons with PD and CG during the medical encounter, with specific encouragement for them to be active in the decision-making process. The PDA on LCIG is provided as a paper handout as well as an electronic booklet.

Acceptability Testing of the PDA Prototype

As a preliminary step of the PDA implementation, it is paramount to assess its acceptability by the people who will use it. ²⁵ This will help to predict if it will be adopted and executed. ²⁶ Acceptability is defined as a concept that reflects the extent to which those providing or receiving a health care intervention consider it appropriate, based on an anticipated or experienced cognitive and emotional response to the intervention.^25,27 It aims to evaluate comfort and satisfaction toward content and format from the targeted audience. ²⁸ It is considered as a primary quality indicator prior to implementing a PDA in a clinical setting. ²⁸

Study design

A cross-sectional uncontrolled descriptive design was used to evaluate the acceptability of the PDA.

Study population, enrollment, and sampling

A nonprobabilistic convenience sample was chosen for the study. The sample size in acceptability testing studies usually includes 9 to 28 people living with the medical condition for which the PDA is intended and 9 to 13 HCP.^29–32 According to Faulkner, ³³ a minimum sample size of 10 participants would contribute to explain 80% to 95% of acceptability concerns or problems to be rectified in a PDA.

Enrolled participants had to be targeted as potential or prospective users of the PDA. As such, persons with PD receiving oral pharmacologic treatment and persons with PD already receiving the LCIG treatment, CG, and HCP were recruited. By selecting those participants, this study aimed to consider all populations concerned with this ongoing decision process.

The enrollment process led to a sample size of 36 participants, that is, 15 PD persons (note that 2 PD persons enrolled were CG as well), 5 PD persons under LCIG, 10 CG, and 6 HCP. Persons with PD, including those under LCIG treatment, were recruited primarily through an online advertisement distributed by the local Parkinson’s association located in Sherbrooke (Quebec, Canada). HCP were enrolled through the specialized PD clinical teams from the 5 health care facilities offering LCIG treatment in Quebec, Canada. Among those HCP, 1 was a neurologist and 5 were PD nurses. More precisely, an informative e-mail was sent by the researchers to HCP to briefly explain the research project and invite them to participate. The e-mail also contained the links to access the electronic informed consent form, PDA, and acceptability questionnaire. People who consented to participate were enrolled in the study. To be included, they had to be able to read and understand the French language at a functional level.

Questionnaires

Data were collected using self-questionnaires and a focus group meeting.

Health literacy (HL) level

PDAs are designed to meet patients’ information needs. The importance of taking HL into account during the development of a PDA is crucial and increasingly recognized. ¹⁶ Moreover, HL evaluation guides researchers to better tailor PDA to make it more readable and comprehensible, especially for those with lower scores. ³⁴ Thereon, authors also outline the importance of improving equity with respect to accessing decision support. ³⁴ As such, since 2005, IPDAS’s PDA quality criteria checklist includes a criterion pertaining to the use of “plain language,” stating it should be written at a level that can be understood by the majority in a target group.^16,35 In that respect, HL is defined as the knowledge, motivation, and competencies of individuals to obtain, understand, and apply health-related information required to make decisions. ³⁶ In this study, the HL of PD persons and CG was assessed using the valid French version of the Functional Communicative and Critical Health Literacy (FCCHL) questionnaire. ³⁷ According to Nutbeam’s model, ³⁸ the FCCHL refers to 3 levels of literacy abstraction: functional, communicative, and critical. The FCCHL questionnaire includes 14 questions using a 5-level Likert-type scale ranging from totally disagree to totally agree and then computed out of a score of 70: functional literacy (items 1 to 5, with a score ranging from 5 to 25), communicative literacy (items 6 to 10, with a score ranging from 5 to 25), and critical literacy (items 11 to 14, with a score ranging from 4 to 20). The higher the score, the higher the HL level (total score ranging from 14 to 70). The FCCHL showed satisfactory reliability (α = 0.77). ³⁷

Acceptability questionnaire

The Ottawa Decision Support Framework (ODSF)^25,39 was used to guide the development of questionnaires to operationalize acceptability, as it provides an approach to measure its outcomes. Three versions of the acceptability questionnaire (PD persons/CG, PD persons under LCIG, and HCP) were developed. Acceptability questionnaires intended for PD persons, CG, and HCP included 11 core items. A subversion for PD persons under LCIG included 4 extra questions aiming at documenting specific issues about LCIG treatment and its daily-living management. The one intended for the HCP included 2 extra questions to assess the clinical management process specifically. (Acceptability questionnaires can be provided on request.)

Data Collections

PD persons, PD persons under LCIG, and CG (n = 30) were invited to attend a single 90-min on-site group meeting at the local Parkinson’s association setting in Sherbrooke. Participants were briefed and undertook different tasks in a specific order: 1) completion of the HL questionnaire, 2) reading of the PDA, 3) completion of the acceptability questionnaire (PD persons/CG version), 4) completion of the sociodemographic section of questionnaire, and 5) participation to the focus group with other participants and research team members to discuss the most important themes and items related to acceptability testing of the PDA.

The focus group interview was audio-recorded. A semi-structured interview guide was created that summarized the most important topics to address and discuss with the group related to acceptability testing of the PDA. It also supported the research team in maintaining focus on the main purpose of the interview within the allocated time. ⁴⁰

In addition, HCP participants had to complete the following tasks online, in an asynchronous mode: 1) reading of the PDA, 2) completion of the acceptability questionnaire (HCP version), and 3) completion of the sociodemographic section of questionnaire. The estimated time required to complete these tasks was 45 min.

Data Analysis

Quantitative data analysis was performed with SPSS data software v.28 in April 2023. Data were computed by the biostatistical department at CIUSSS de l’Estrie-CHUS Research Centre in Sherbrooke. Descriptive analyses were performed using the mean (standard deviation) for continuous variables normally distributed and the median (interquartile range) for continuous variables not normally distributed. Referring to the ODSF to support this study, ²⁵ acceptability testing data were extracted by mainly using a qualitative approach.^41,42 The total score of the FCCHL questionnaire was computed out of 70 (continuous variable). Explorative comparative analysis was also done using the chi-square test for categorical variables, the Kruskal-Wallis H test for categorical and continuous variables, and the Spearman correlation test for continuous variables. Explored association measures were between participant’s characteristics (such as age, gender, educational level, and HL score) and dependant variables of the acceptability questionnaire (i.e., length, usefulness, neutrality, intention to recommend PDA).

Quotations from the focus group were transcribed. Two researchers read the transcripts independently and classified the content as strengths, areas of improvement, or as suggestions. Content analysis was then compared and discussed during a consensus meeting.

Results

Questionnaires were successfully completed by all study participants (N = 36).

The characteristics of the sample are summarized by population subgroups in Table 1, reporting gender, age, ethnicity, and highest level of education. Job title and years of experience were also documented specifically for HCP. The duration of illness and self-report Hoehn and Yahr stage ⁴³ were also collected for persons with PD as well as duration of treatment for those under LCIG.

Table 1.

Sample Sociodemographic and Disease-Related Characteristics (N = 36)

Characteristic	Participants
	PD Person (n = 15)	PD Person under LCIG (n = 5)	CG (n = 8)	CG + PD Person (n = 2)	HCP (n = 6)
Gender, n (%)
Men	8 (53.3)	2 (40.0)	2 (25.0)	1 (50.0)	1 (16.7)
Age, y, $\overline{x}$ (s) 64.4 (14.0)	69.9 (8.0)	67 (9.7)	68.5 (7.4)	75 (1.4)	39.3 (11.2)
Ethnicity
Caucasian	15 (100)	5 (100)	8 (100)	2 (100)	6 (100)
Highest educational level, n (%)
High school (12 y)	4 (26.7)	2 (40.0)	1 (12.5)
College (14 y)	4 (26.7)	1 (20.0)	2 (25.0)	1 (50.0)	1 (16.7)
University (17 y or more) a	7 (46.7)	2 (40.0)	5 (62.5)	1 (50.0)	5 (83.3)
Duration of illness, y, n (%)
<5	7 (46.7)
5–9	5 (33.3)			2 (100)
10–14	1 (6.7)	1 (20.0)
15–19	1 (6.7)	3 (60.0)
≥ 20	1 (6.7)	1 (20.0)
Self-reported Hoehn and Yahr stage, n (%)
1 (unilateral)	6 (40.0)	1 (20.0)
2 (bilateral)	3 (20.0)
3 (mild disease)	3 (20.0)	2 (40.0)		2 (100)
4 (severe disease, unassisted)	1 (6.7)	2 (40.0)
5 (severe disease, needs assistance)	2 (13.3)
Duration under LCIG treatment, mo, $\overline{x}$ (s)		41.4 (28.3)
Job title, n (%)
Neurologist					1 (16.7)
Clinician nurse					4 (66.7)
Nurse					1 (16.7)
Years of clinical practice with PD, $\overline{x}$ (s)					9 (6.99)

CG, caregivers; HCP, health care professionals; LCIG, levodopa-carbidopa intestinal gel; PD, persons with Parkinson’s disease; PD-LCIG, persons with Parkinson’s disease under LCIG.

^aType of diploma not assessed.

The HL skills were documented for PD and CG participants (n = 30). Overall, skills were judged as very good, with a mean score of 55.2 (s = 7.3) (of a maximal score of 70). Figure 2 illustrates the HL frequency histograms of the total score and its 3 respective dimensions: functional literacy ( $\overline{x}$ = 18.00, s = 5.08), communicative literacy ( $\overline{x}$ = 20.7, s = 2.15), and critical literacy ( $\overline{x}$ = 16.50, s = 3.38).

Figure 2.

Scores for total health literacy (HL) and its 3 dimensions (persons with Parkinson’s disease, persons with Parkinson’s disease under levodopa-carbidopa intestinal gel, and caregivers), n = 30).

Table 2 highlights the main results of the PDA acceptability testing of the study based on the descriptive analysis from the acceptability questionnaires. Population subgroups were added for a qualitative appreciation of how they contribute to those data. Because of the small sample sizes, it is not yet suitable for further statistical testing. Nevertheless, because a PDA should be used as a collaborative tool with the participation of PD persons, their CG, and their HCP, referring to the results of the whole sample might be more representative (N = 36).

Table 2.

Highlighted Results of the PDA Acceptability Testing

		Subgroups (n)					Total Sample Size
Acceptability Characteristics		PD Persons (n = 15)	PD Persons under LCIG (n = 5)	CG (n = 8)	CG + PD (n = 2)	HCP (n = 6)	%	n (N = 36)
Time required for reading	20 min or less	11	3	7	1	6	78	28
Content sections are clear:	Options displayed “What are my choices”	9	3	6	2	4	67	24
	Practical aspects of treatment	8	3	4	1	3	53	19
	Effects of treatment (advantages/disadvantages)	7	2	4	1	3	47	17
	Option grid “Choosing an option based on your priorities”	6	4	4	1	3	53	18
	Levels of scientific evidence	3	3	2	0	1	25	9
Length	Adequate	12	5	7	2	5	86	31
Amount of information	Sufficient	6	3	6	2	3	56	20
Neutrality of information	Neutral	14	4	6	1	4	81	29
Usefulness	Useful/very useful	13	2	7	2	5	81	29
Pleasant visual layout	Agree/totally agree	12	4	8	2	5	86	31
Would recommend it	Strongly/very strongly	9	4	5	2	4	67	24
PD persons under LCIG (subgroup n = 5)
When you first decide to undergo LICG treatment, would have PDA eased your decision-making process?	Totally agree						20	1
Back then, would your decision remain the same?	Yes						100	5
HCP (subgroup n = 6)
Easy to use in clinical practice							100	6
Have the intention to use it							100	6
Both neurologist and nurses would be the best to use it							50	3
Neurologist would be the best to use it							10	1
Nurses would be the best to use it							40	2

CG, caregivers; HCP, health care professionals; LCIG, levodopa-carbidopa intestinal gel; PD, persons with Parkinson’s disease; PDA, patient decision aid; PD-LCIG, persons with Parkinson’s disease under LCIG.

As mentioned, qualitative appreciation of the PDA was also inquired from the focus group. Every PD person and CG went through the draft version of each component during the on-site session (n = 30). This led to the identification of strengths as well as areas for improvement. Online appreciation of PDA by HCP was also collected. The main considerations are reported in Table 3a and highlight the genuine verbatim responses of the participants. Areas requiring revision were considered, and adjustments have been made to the PDA for an updated version (Table 3b).

Table 3.

Qualitative Appreciation Summary of the PDA on LCIG during the On-Site Focus Group Discussion (PD, PD-LCIG, CG) and Online (HCP) (Verbatim)

3a. Strengths
Clear PD:“I liked the tool presentation and detailed explanations. Well explained.” CG:“As a caregiver, I think it’s very clear.”
Complete PD:“We find there the essential elements to initiate a reflection.” CG:“Quite a complete recap. Gives an overview of treatment.” HCP:“Very detailed document and not too long.”
Useful PD:“Seems like an interesting and useful tool.” CG:“The explanations enlightened me . . . documentation very useful.”
Relevant PD:“Will be very relevant information when available.” PD-LCIG:“Helps to know the facts about advantages and disadvantages of treatment.” HCP:“I like that the PDA includes evidence-based data because patients asked a lot about it, percentage . . . .”
Adapted PD:“Gives answers without falling into the medical jargon.” PD-LCIG:“Clear documentation, easy to understand.” HCP:“Clear, intended for patients, concise.”
Pleasant PD:“I like the layout, the visual aspects.” PD-LCIG:“The visual layout is pleasant.” CG:“Pleasant layout and ease to understandability.”
3b. Areas for Improvement	Actions/Modifications
Text and wording  HCP:“Be careful with nonmotor symptoms improvement ‘promises’!! I would be more neutral about this within the text.”	Wordings have been review where pointed out. Some statements have been nuanced.
Length of document and font size  HCP:“Text should be shorter with bigger font size.”  CG:“There is a lot to read! For a person with Parkinson, it can be hard to read all that, it needs a lot a concentration.”	Length of document was not modified. As guided by PDA development frameworks, essential components that summarize evidence-informed and best practices options are expected in such a document and enable clarification of the patient’s personal values and preferences that will influence their choices and decisions. 13 In such a scenario, caregivers could also assist the PD persons with the reading. Font size has been increased.
Need for supplementary information  PD:“So it’s a lot of information at the same time, all at once, about a treatment I’ve never heard before. A lot of professional information. That would have been helpful to get information before the group discussion meeting, or a video . . . you know, to be alerted, to be more prepared, to have a point of view.”	No supplementary documentation was available at time of this protocol study. However, for an optimal implementation and usability of the PDA in clinical settings, supplemental material could be considered prior encounter with neurologist (e.g., video or Web site). 26
Carry-on outfits and wearing  PD:“When reading the document, I asked myself, how do we wear this thing? Is there any wearing outfits? It must be heavy to carry-on? Can I play sport with it?”	Information has been added to the section “Practical Aspects of LCIG Treatment” and the “Daily routine” subsection. The section describing “One user’s experience with LCIG” refers to those aspects with details.
Cost of treatment  PD:“. . . not enough information on treatment cost. Is it covered by insurance or not? You know . . . this is an important aspect of the decision.”	No modification. Too many factors can influence the treatment cost (i.e., insurance coverage, state or province area of living, fundings).  Average cost of treatment was given for reference in the PDA.
History of drug approval  PD:“There is no approval historic data about the medication.”	Information about the history of drug approval by Health Canada has been added to the section “Practical aspects of LCIG treatment.”

CG, caregivers; HCP, health care professionals; LCIG, levodopa-carbidopa intestinal gel; PD, persons with Parkinson’s disease; PDA, patient decision aid; PD-LCIG, persons with Parkinson’s disease under LCIG.

Lastly, any significant or clinically relevant results were obtained by computing the exploratory comparative association measures between participants.

Finally, as many patients frequently turn to online communities and social media to meet their information needs, the PDA on LCIG has been filed on the Parkinson Quebec Association Web site (https://parkinsonquebec.ca/wp-content/uploads/2023/10/OAD-GILC_V1.0_FR-Fillable-Version.pdf?_gl=1*1ql8a2w*_up*MQ..*_ga*NTc2MzUyNjg3LjE3NTU2MTI2NTg.*_ga_1KY7PMQMCX*czE3NTU2MTI2NTUkbzEkZzEkdDE3NTU2MTI3NzMkajU1JGwwJGgw).

Discussion

Because PDAs are meant to be used in clinical settings, we put persons with PD and CG at the center of our process. We asked for their feedback at multiple points of the development of the tool and adjusted subsequent iterations based on their insight.

As related earlier, no neurologist took part in the AC. However, the neurologist who was a member of the research team was engaged, with the 2 other researchers, in the validation steps of the PDA prototype. His input was imperative, knowing that the PDA is intended to be used, among others, during neurologist visits. The formal contribution of at least 1 neurologist should be considered in the AC in future PDA development research. However, to avoid desirability bias, it would be crucial to ensure that the neurologist is not involved in the care of any participant.

In addition, when designing a PDA, as pointed out earlier, greater attention to HL and socially disadvantaged populations is needed to ensure equity in decision support. ³⁴ High HL scores obtained in this study, especially those related to communicative ( $\overline{x}$ = 20.7, s = 2.15) and critical ( $\overline{x}$ = 16.50, s = 3.38) subdimensions gave the research team a strong indicator of participants’ capacities to read, understand, and communicate before undertaking the different steps in collecting data. Moreover, those results appear to be revealing as communicative and critical HL are reported to be stronger predictors of successful self-management in patients with chronic or neurodegenerative disease. ⁴⁴

As discussed by Heijmans et al., ⁴⁴ HL levels may also vary according to the sociodemographic and disease characteristics of the patients. Because 15 of 30 patients had a university educational level (even if the significant association was undetected), we might have encountered an educational bias thinking that the final version of the PDA could have been different with distinct sample characteristics of age, education level, and duration of disease, for instance. From this perspective, further validation of the PDA would benefit from a larger sample size as well as a purposive sampling strategy that would increase the representativity of the population with different HL levels. ⁴⁵

Furthermore, as recommend by the IPDAS ²⁵ and the SUNDAE checklist, ¹⁶ the use of “plain language” was of main concern when developing the PDA. Indeed, readability and comprehension should always be considered to reduce the “cognitive demand” as well (e.g., larger font size, plain text, no medical jargon, grade reading level). ³⁴ In our study, we first addressed those concerns with the AC. Those topics were also considered in the acceptability questionnaires and discussed during the focus group leading to final adjustments (i.e., wording clarifications, shorter sentences, larger font size, illustration support, etc.).

Moreover, as pointed out by Nijhuis et al., ²⁶ cognitive deficit in PD persons at an advanced stage of the disease or age could limit their reading ability and comprehension of the PDA. Therefore, measuring and monitoring cognitive impairment should be evaluated in future research as a baseline patient characteristic when assessing PDA outcomes.^26,42

In addition, as described earlier, data were collected at a single on-site group meeting in which participants (PD persons, PD persons under LCIG, and CG; n = 30) had to read the PDA, complete questionnaires, and finally take part in the focus group. This was initially intended to prevent participants from having to make multiple visits for the research project, as they can easily experience fatigue and lack of energy related to the disease. However, completing the questionnaires in such a nonprivate group setting might have caused a social desirability bias toward the research team members as well as peer pressure and stress to perform better toward others.

More importantly, when referring to the focus group method, another limitation is the sample size. According to Corbière and Larivière, ⁴⁶ the size of the focus group should not exceed 5 to 8 people, to allow everyone to express themselves. Hence, although it is unusual to gather such a great number of people at the same time (n = 30), the research team used this option mainly because of pragmatic issues. Indeed, while the local Parkinson Association meeting room is very practical for disabled persons with its accessible building facilities, it had a very limited daily availability during this period. Because we wanted to keep a mid-day schedule during which PD persons are more alert and have more energy to undertake such cognitive tasks (including prior completion of questionnaires), we decided to opt for a single focus group, despite its methodological limitations. Still, even if the research team had been fully prepared to guide this interview, there was the probability that the group size might have limited some discussions and led to less representativeness of viewpoints. In summary, a more private setting to read the PDA and multiple but smaller sized focus groups ⁴⁰ might have contributed to provide a more suitable environment to collect data, prevent inconvenience, and enhance ecological validity.

As guided by PDA development frameworks (IPDAS), essential components that summarize evidence-informed and best practice options are expected and ineluctable in such a document as they are intended, based on scientific data, to clarify patients’ personal values and preferences that will influence their choices and decisions.^13,26 However, this content format, as reported by participants in the acceptability questionnaire, as well as in the focus group, tended to be perceived as a lot of information to assimilate and, at some point, overwhelming. This appears to be in line with some results reported in Table 2 about the clarity of the components of the PDA, for instance, where only 25% of participants rated the “Levels of Scientific Evidence” section as clear. While areas requiring revision were considered and adjustments were made for an updated version (Table 3b), no supplemental preparatory documentation was available at time of this study. Thereby, for an optimal implementation of the PDA in the future and ease of its usability in clinical settings, supplemental educational material should be considered for sharing before the encounter with the neurologist and PD care team. This could then be used to brief patients and CG with basic information, using a more friendly, easy-reading approach that popularizes essential topics about advanced therapies. Patients would then have time to read it at home, at their own pace, and prepare their questions in advance. In such a scenario, CG could also be of great help in assisting with the at-home readings.

Also, it appeared that testing the acceptability of the PDA might have led to a difficult projection exercise for PD persons. As discussed by Sokol et al., ⁴⁷ raising the topic of advanced treatments such as LCIG with PD persons, still at an early stage of the disease, may create emotional distress and anxiety and, at some point, threaten their hope of a cure. Indeed, this projection effect was reported during the focus group session by some participants. For future research, those persons with PD could be excluded to prevent this unnecessary projection exercise.

Likewise, the study also applied a similar projection exercise specifically for participants with LCIG to take a position on this treatment choice, knowing they already had made this decision in the past. This led to these participants being confronted with the same dilemma again. This issue could be reflected in our study results (Table 2), in which all LCIG participants (n = 5/5) reported that their decision would have remained the same if they would have had access to the PDA in the first place. This is also supported by the fact that persons with PD under LCIG rated the PDA as less useful in the decision-making process (80%, n = 4; P = 0.08, chi-square test). This raises the question as to whether more accurate feedback and introspection with those participants might have been observed if the decision to opt for LCIG treatment would have been more recent (within the past few months, for instance).

Although the acceptability testing revealed that all HCP intended to use the PDA in their practice, half (n = 3/6, 50%) reported that it would be best for both the neurologist and nurses to use it. Two others said this should be the nurse (n = 2/6, 40%) and 1 the neurologist (1/6, 17%). These mixed results may overlook the fact that the method for providing counseling to persons with PD might differ among health care settings. Larger affiliate or university centers may benefit from more on-site resources such as a deployed collaborative clinical team. In this team, a PD specialized nurse, also called pivot nurse, will lead this educational process, alongside the neurologist. However, in smaller health care facilities, the neurologist or nurses, with no access to a supportive team, might need to assume this on their own.

At this point, more than 1 y after the PDA on LCIG was released, a few questions remain. As supported by IPDAS ³⁵ and Hoffman et al., ¹⁶ we should inquire how HCP are trained in the delivery of the PDA as they have a direct impact and influence in the SDM process. We should also question how the PDA is used by HCP and patients. In other words, is it suitable in the clinical setting?

Conclusion

A PDA on LCIG for persons with PD, CG, and HCP was developed to support the decision-making process. To our knowledge, this study is the first to report the development of a PDA on LCIG and its acceptability testing. We elaborated on this PDA by adopting a user-centered approach with iterative revision cycles throughout all of the development steps. PDA acceptability was assessed as the first preliminary step for its validation. Overall, it was perceived as very good, satisfying with its content, and useful, and participants reported they would recommend it. Documenting the feedback and conducting qualitative appreciation data analysis allowed for adjustments to the PDA, which is now available to support SDM with those persons. The PDA was delivered for use in neurology clinics where LCIG is available in Quebec as well as on the Parkinson Quebec Association Web site. However, to ensure that the PDA on LCIG becomes a generalizable professional tool and leads the way as a referral guideline standard, more research will be needed to evaluate its level of implementation and its effects on users (persons with PD, CG, and HCP). To accomplish this, a larger and a more representative sample of the population will be required.