Mock Viva Questions

Question 1

Tell us about your thesis - What are the key findings?

Answer 1

We found that mHealth interventions are an acceptable, informative and supportive streategy that appears to help people with physical disabilities improve their physical activity levels. MHealth is cost-effective, does not require too much expertise, can be tailored towards the individuals needs, can be used anywhere and can help people with disabilities overcome many PA barriers.

Question 2

Tell us about your thesis - What does it add to the literature?

Why does it matter/why do we care?

Answer 2

The reason we care is people with physical disabilities are physically inactive, they are not meeting recommended guidelines due to numerous barriers that exist and our thesis adds to the literature because we are the first to evaluate the Accessercise application in terms of it’s theoretical underpinning, usability and feasibility before we proceed into a full-scale trial.

David: Yes, so you are doing that preparatory work before you go into a full-scale trial, which takes away some of that risk and uncertainty.

The Accessercise application is a novel application. So, current application that exist mainly focus on a singular disability. However, more recently apps that focus on multiple disabilities struggle to tailor the content around the users disability. Accessercise is unique because you can tailor the content around the disability you select.

Question 3

Summarise your approach, why did you opt for the MRC and not an alternative? (think about the MRC framework, mixed methods)

Answer 3

The MRC framework has been widely used to develop and evaluate behaviour change interventions.

As we know we need to develop more interventions in this population, it’s recommended because it’s a systematic, logical and reliable framework to help develop and evaluate interventions because it goes through the development, feasibility, evaluation and implementation..

Question 4

What motivated you to look at disability/SCI, physical activity, and mHealth?

Answer 4

1. I undertook my MSc dissertation looking at the barriers and facilitators to PA in young adults with physical/sensory disabilities. Based on these barriers, I noticed that there is a need to develop suitable interventions that can help improve physical activity in this population, which can help overcome sedentary behaviours
2. Going into my PhD, I wanted to develop an intervention. Unfortunately, this plan changed because of COVID etc but we got approached by the Accessercise team, which developed a novel intervention and they seeked us to evaluate their intervention. Therefore, based on the expertise of the PHC and my supervisors, we thought that SCI were a suitable population to work on.
3. Talk about my Care Jobs
4. Dyslexia

Overall, I have a wealth of experience when it comes to working with people with disabilities in a physical activity context.

Question 5

I said that Accessercise is the only app that looks at multiple disabilities. Is this true?

Answer 5

No. So, the SUPERHEALTH targets multiple disabilities (e.g., arthritis, SCI etc), WHEELS focuses on several different SCI disabilities.

However, the tailoring part of the Accessercise app is what makes it unique. So, Accessercise allows you to select disability and the content of the application changes and is tailored towards the individuals impairment. Other apps don’t really have that function!

The SUPERHEALTH and WHEELS app are not really highlighted much in my PhD thesis but if I mention it to my examiners it will show my wider understanding of the literature

Question 6

I mention about the recommended guidelines. Which ones do I refer to?

Answer 6

I mainly talk about SCI related guidelines (e.g., 40-minutes of moderate to vigorous PA once per week).

However, I need to be mindful that my PhD doesn’t just look at SCI but I could use the WHO guidelines because they do have specific guidelines for physical disabilities, which are quite similar to non-disabilities.

PA guidelines (SCI): 40 minutes of moderate to vigorous aerobic PA once per week

PA Guidelines (Disabled): 150-300 moderate intensity PA once per week

Question 7

The examiners may ask you a real life question like we have a new government and imagine you are speaking to the minister of disability. You have two minutes so what would you say to them, what do they need to know?

Answer 7

Think about the key points from my PhD thesis!

Avoid my answers always being around Accessercise. Instead, say mobile enabled health (mHealth apps) are accessible and useable for this demographic. Therefore, we should invest more money and time into developing these.

I need to think much bigger here.. So, we found that mHealth is an accessible way, can help overcome barriers for people with disabilities, acceptable mode of delivery for people with physical disabilities to increase physical activity and we really should be investing more time and money into delivering these.

Think about the advantages of mHealth apps… (1) Cost effective, (2) Can help overcome barriers, (3) provide real time feedback, (4) other modes present barriers to this population, (5) mHealth can be tailored towards individual needs..

Question 8

Why did I opt to use the Medical Research Council framework and not another framework? Are there any other ones that I could have used?

Answer 8

Not that I am aware of but I feel like the MRC framework has been widely used in this population because looking at existing literature it seems to be a useful way to evaluate interventions, it’s robust and universally accepted (David really liked this answer because I said I wasn’t aware of anything and because I said the MRC framework has been widely used)

Question 9

Chapter 1 (Introduction): You describe in detail the different models of disability. Why did you do this and how did they inform your research? Do you have a favoured model, and why?

Answer 9

The reason I spoke about the models of disability is because at the beginning of the thesis I wanted the readers to understand exactly what disability is before we jump into SCI, Physical Disabilities and so on. So, overall I wanted them to understand what is the definition of disability.

Based on the literature and what I wrote, it’s still difficult to define what disability is because everyone has different perspectives of what disability is. To help us understand what disability is, there are different models can help us define what disability is. Within my literature review, I only used the Medical Model, Social Model and the ICF model because these are the most prominent ways of understanding disability.

The Social Model builds on the limitations of the Medical Model and then followed by the ICF model which uses a combination of the Medical and Social Model to give us a better understanding of disability.

Question 10

Did these models help inform my research I did throughout my PhD rather then just help to define/understand disability?

Answer 10

No, they didn’t feed into my research but they gave me a wider perspective on understanding disability which helped me to form and structure my thesis (David liked this answer- He said perfect) These models helped develop my understanding of disability.

Question 11

Do you have a favoured model of disability and if so why?

Answer 11

I would say the ICF model.

The reason for selecting the ICF model is because I feel like disabilities can be a social problem as well as a medical problem. The ICF helps us to understand that it isn’t one or the other but it’s a combination of both and I feel that the society that we live in today is that it’s a combination of social barriers (e.g., accessibility issues, stigma) but also internal because the individual is disabled because of their condition, which makes them disabled.

So, I think if we view it from that view its better to understand disability as a collective problem rather then one or the other. It helps us to understand from a bigger picture.

Question 12

Are you aware of any limitations of the ICF model?

Answer 12

The ICF model incorporates over 1,400 categories which can make it time-consuming and labour intensive for any new researchers to apply this framework to their research.

Question 13

So, you mentioned that the ICF has a lot of categories so then how can we use this then? How is it usable?

Answer 13

It has been widely used in the literature now.

It has been summarised and there is a suitable manual which can help researchers to use it.

Even though it has a lot of these categories, there is a manual which does split it up by conditions (e.g., SCI etc) to make it more usable.

Question 14

So, let’s talk about your paradigms and philosophies. What is your underlying approach to science, how does it underpin your PhD?

Answer 14

So, my philosophical underpinning is pragmatism. I selected pragmatism because of the key reasons.

We applied the MRC framework, which consists of different stages so it has a combination of qualitative/quantitative approaches, so therefore my overall methodology based on my view of research paradigms is that I am more of a mixed-methods pragmatist researcher going into this PhD because I wasn’t just an interpretivist or a positivist but I felt that I was in the middle with a pragmatic view to help answer the research question on how to understand how to increase PA in people with disabilities.

Overall, the pragmatist approach aligns closely with the MRC framework, a triangulated approach by using the two together because they both have their advantages and disadvantages.

We agreed to use pragmatist approach as a supervisory team because it follows the MRC framework which is key to my PhD!

Question 15

So, you describe SCI in my literature review at length. Why did you initially focus on that condition but you then focus on physical disability more generally?

Answer 15

1. So, among the physical disability population, SCI is the least physically active and are at the lower end of the PA spectrum.
2. Additionally, based on the expertise of colleagues at Loughborough, Peter Harrison Centre and my supervisors, we felt it would be a good population to address in terms of developing PA interventions. (Perfect answer)

Question 16

Why did you jump from SCI to Physical Disabilities?

Answer 16

The reason for this comes down to 2 key reasons..

1. When we were recruiting participants for our study, we found it quite difficult to recruit participants with SCI, so we then extended out the recruitment to more people with physical disabilities (Perfect Answer)
2. As my thesis developed and the intervention that I had available to me incorporated multiple disabilities.
3. To make my research/findings to be more inclusive, establish equality and to ensure our work is more generalisable to a wider population within the UK.

Overall, this is why I have jumped from SCI to Physical Disabilities.

Question 17

Why did I focus my PhD around using the Behaviour Change Wheel (BCW) and what about other theories of behaviour change

Answer 17

We know changing behaviour is challenging and requires a lot of work.

Even though there are many theories that exist, it’s always difficult to work out which theory to use because they overlap. However, I think the majority of the theories exist don’t always consider the interpersonal, intrapersonal and environmental factors. So, therefore recently the BCW has been developed, which overcomes some of the limitations of the theories that I mentioned before (e.g., Health Belief Model, SDT).

Some of the benefits of the BCW it can helps us to understand the barriers and facilitators to PA in this population, helps us to understand the behaviour.

The BCW is a super theory, it incorporates 19 frameworks into one. Additionally, the BCW has a further advantages such as it allows you to develop and/or evaluate interventions. The BCW is very much intervention focused. Also, we chose the BCW because my supervisor has extensive experience with the BCW.

The majority of theories that exist only look at predicting behaviour and DO NOT look at understanding behaviour.

Question 18

You hone in on mHealth in the introduction chapter but what about other modes of delivery (e.g., in-person/telephone). Why have you only focused on mHealth? Are you aware of any other literature that assesses other modes in this population?

Answer 18

So, the majority of literature that looks at in-person, telephone counselling are mainly around strength training, which are effective at increasing PA in this population.

However, we know that in-person interventions are difficult for participants to attend in terms of transportation, time, cost and other barriers. The reason why I have gone straight into mHealth is because we know that since COVID individuals with disabilities had to do physical activity at home and mHealth is a suitable intervention approach to overcoming many barriers, can be undertaken in any location, is cost-effective, is easy to use, can be tailored towards the end-users needs and is ACCESSIBLE to a wider population.

We now know that people with disabilities are increasing the amount of smartphone use compared to previous years. Even though that some people with disabilities may struggle with accessing this technology, we know it’s a suitable information.

Question 19

So, the MRC framework that you describe in Figure 1.8 is not the latest version. Are you aware that it’s been updated and why did you not include that?

Answer 19

I’m not aware that there is a new framework.

My answer to this question is that I started my PhD in January 2021, I used the older version when I started my journey and the newest version (2021) had not been published yet.

Also, when I looked at the newer version, there wasn’t much difference between the two so I continued using the older approach throughout the thesis as it didn’t make sense to change. So, when I started my thesis the newest version had not yet been published.

It was published in 2021 during COVID and I had already started my thesis before that. Therefore, to ensure consistency in my work I followed the older version through to the end of my PhD but when I reviewed the newest there wasn’t much difference.

Question 20

How did I generalise my research questions? What were the reasoning’s and justifications for these?

Answer 20

Based on the systematic review, we know that theory is effective at increasing physical activity for people with SCIs when it comes to interventions. We know that only a couple of studies included in the systematic review used theory.

Therefore, we attached on the next study the BCW to assess the theoretical underpinning of a novel intervention, so that’s how I developed my first and second study

The first study was mainly around that interventions needed to be developed, we know our systematic review is based on the MRC framework.

For study 3, we know despite Accessercise appearing to be a promising intervention in this popualtion based on the findings from Study 2 unfortunately not much usability testing has been done on mHealth interventions for people with disabilities, so that’s the rationale for doing my 3rd study

For my feasibility study, when we were looking at the existing research for mHealth interventions, we know that the effectiveness of these interventions haven’t really been undertaken and we know that Accessercise has yet to be investigated.

(David says that for this answer I can also link back to the MRC framework - because my aims are very much structured around those aims)

Question 21

Which steps of the MRC framework did I address in my PhD?

Answer 21

2

Step 1 (Development)
- Systematic Review
- Theoretical Underpinning stidy
- Usability

Step 2 (Feasibility)
- Feasibility RCT study

Question 22

How did you ensure the integrity of the evaluation of the Accessercise app when working with the app developers? In other words how did you know that the developers did not influence the research?

Answer 22

We had a collaboration agreement with the app developers.

They approached us because they wanted an independent evaluation where they weren’t involved, so although we worked closely with to gain access to the app, and understand the app and we had meetings with them to understand the design ultimately the decisions were ours.

They cannot embargo anything that we publish. They have 30 days to comment/provide feedback.

However, they were not involved in the study design, I was not involved in the development of the app etc, they were not involved in the study design, it was very much looking at independently evaluating the app.

The app developers had an involvement but

Question 23

Why did I undertake a systematic review?

Answer 23

1. I wanted to understand what is out there already
2. I followed the MRC framework which suggests undertaking a systematic review first.

Question 24

Why did you only focus on manual wheelchair users and why adults?

Answer 24

Previous sytematic reviews (e.g., Kathleen Martin Ginis) looked at a range of different physical disabilities and Watson et al (2023) looked at a range of different SCIs

However, we know this sample is heterogenous with their requirements being different.

However, Manual Wheelchair Users with SCI are the most physically inactive spinal cord injured population and therefore we wanted to address this issue by developing/evaluating interventions in this population.

Question 25

So, several similar systematic reviews exist that have been published in this area. Are you aware of these and why have you replicated these. Did you not check this before starting your review? What are the similarities/differences between your reviews?

Answer 25

Differences: Ma and Ginis only used RCTs. Whereas, we used RCTs, non-RCTs, before and after etc! Ma and Ginis (2018) and Watson et al (2023) focus on a range of different physical disabilities or SCIs. Whereas, I just focus on a singular disaiblity. However, the Watson et al (2023) paper which is quite similar to mine (SCI specific) they registered their systematic review after ours on PROSPERO, so actually they copied you. We registered our review on PROSPERO and then subsequently 2 months later the Watson (2023) paper registered their review. So, at the time of my registration of my review there was nothing else out there on this specific/unique topic! They must of likely seen my review and then registered it and then quickly published it before I had the option to publish it. I thoroughly checked PROSPERO and there was no other records there that showed the same thing and it wasn't until this one was published that I came across it and saw that it was registered on PROSPERO, 2 months after my registration. It was disappointing that they didn't contact me regarding it. We could have worked together on it.

Question 26

Given that your literature review focuses on mHealth, why didn't you focus on this exclusively for your intervention? (e.g., systematic review - Behaviour Change Intervention)..

Answer 26

There is currently limited research on mHealth intervention research for people with SCI. So, if I did mHealth as the intervention it could have just been an empty review.. I am aware that there is limited evidence on mHealth. Instead, I wanted to look at everything (all interventions).

Question 27

Why did I choose those specific outcomes for your systematic review (e.g., self-reported PA, Behavioural PA, Adverse effects)? Were there any others you could have included? Why did you not include these?

Answer 27

1. Physiological - We wanted to look at heart rate, blood pressure because these are obviously health outcomes related to PA 2. Adverse effects (e.g., pain, injury) are commonly experience in people with SCI and adverse effects is good practice to include in systematic review according to Cochrane, which is a gold standard systematic review process. 3. Self-reported PA and Behavioural PA is obvious 4. Functional fitness is related to Self-reported PA and Behavioural PA because PA makes you fitter. 5. Health-related QoL related to PA All these outcomes are associated with PA. If you see improvements in PA you are likely to see improvements in General-Health Related QoL, reductions in adverse effects etc. Overall, we wanted to include a lot of outcomes, a lot of research designs because we didn't want to have an empty review.

Question 28

You searched a lot of databases for your systematic review. Why did you select these and not others? Were there any criteria that you used to inform your decisions?

Answer 28

I think based on previous systematic reviews in the area they have only included several, so I wanted to provide an holistic approach to really search databases for new and existing research and really help us in answering the research question.

Question 29

Great, so you used a lot of databases which is great but all of your databases are health based. Why did you only focus on health databases?

Answer 29

As we are only looking at health conditions, I was only looking into health. I did not include any Arts or Humanities because you really diged down into health I also have two supervisors which are experts in systematic reviews, so I went on their guidance.

Question 30

How did you divide your search strategy? Did you get any additional support and were you confident that your search strategy was comprehensive and included all the correct terms? Do you think it's possible that any relevant literature could have been missed?

Answer 30

I received support from my supervisors as well as the academic librarian from my school. In terms of missing any relevant research, we can never be certain but I ensured that I did the searches all within one day, overseen by my supervisory team, searched each database clearly using each database, I used Covidence (gold standard software) for storing papers for my systematic review. However, a strength of the search strategy is that they can be replicated. So, I can always do them again to ensure if I am happy with how I am done it. If needed, my supervisors can also double check too! I followed the highest support provided from my supervisory team and obviously it's possible I could miss something but I used the most rigorous approaches to avoid that.

Question 31

Why did you include Risk of Bias 2, GRADE, Downs and Black for risk of bias, study quality Why do you think it's important to do that?

Answer 31

I think every study incorporates some risk of bias and I think it's important to know the quality of each paper you include in a review. The reason I used RoB2 is basically I originally used RoB1, which looks at a range of different study designs but I obviously wanted to follow the most up to date research. For GRADE, I used this because I wanted to know the quality of each study. Lasty, Downs and Black (1998) checklist was chosen because it provides 27 clear questions on the studies (e.g., does it provide a hypothesis, are participants randomised). Overall, Downs and Black is good because it's suitable for any study design RoB2 non-RCTs score very low because it's all about randomisation. So, Downs and Black gives you that variability in terms of a list of different questions it asks. RoB1 is simpler to use and I liked it more but I wanted to ensure my research was up to do date and used suitable approaches. Also, the RoB2 was selected because the reviewers asked me to do it.

Question 32

Why did you use standard mean differences and random effects?

Answer 32

We used Standard Mean Differences and Random Effects because we assumed there would be variability. Fixed Effect you would use if the studies all used the same outcome measure and their interventions and populations were very similar. It means very little difference. Standard mean difference: Always used when the outcome measures are measured slightly differently, which I expected would be because of the heterogeneity in outcome measures and research designs and that's why I also used a random effect model. We determined at apriori (before we did the meta-analysis) because we knew there would be loads of variations in our outcomes, how interventions were implemented. WE KNEW THERE WOULD BE HIGH HETEROGENEITY! Mean Difference is only used when the outcome measure is the same and is measured in the same way (units of measurements are identical), so we can make that comparison. However, we didn't expect to find that so we used the Standardised Mean Difference because it doesn't matter how something is measured (e.g., differerent questionnaire) it's STANDARDISED. Random Effect: We expect there to be heterogeneity in the sample, study designs and therfore that's why we did a RANDOM EFFECT.

Question 33

Why did you do moderator analyses and how did you do those?

Answer 33

We did moderator analyses to assess whether the study characterstics, intervention characteristics, mode of delivery, BCTs and Theory were more effective in the intervention relative to the control groups. We did it using RevMan, we worked out which studies used BCTs vs studies that didn't. OVERALL, WE SPLIT THE STUDIES FROM THE META-ANALYSIS INTO SUB-GROUPS AND WE THEN RE-RAN THE ANALYSIS BASED ON STUDIES THAT DID/DIDN'T USE THEORY IN THE INTERVENTIONS VS THE CONTROL GROUPS TO HELP COMPARE THE DIFFERENCES BETWEEN THE GROUPS. We did the moderator analyses based on the recommendations from Ma and Ginis and Watson et al (2023) also did it and was recommended by a reviewer!

Question 34

How/why did you focus on intervention characteristics, study interventions, use BCTs and theory for my moderator analyses? Why did you not look at anything else like gender, ethnicity or level of disability?

Answer 34

Overall, we were looking at the overall effectiveness of the interventions in terms of mode of delivery, who delivered the interventions etc was helpful! The use of theory - We know some interventions that use theory are more effective then other theories so we wanted to test that claim.

Question 35

I initially had a lot of records to screen for the systematic review. How did I accomplish that task because it was very detailed with over 11,000 papers? What did you use to inform your judgments for including inclusion, did you get any help and why didn't you have an additional screener?

Answer 35

1. Completed the searches all within one day. 2. Supervisors/librarian saw everything I did and ensured they were involved in the screening process. 3. Searched each database 1 by 1, downloaded the files onto my PC and then imported them onto Covidence. 4. To ensure I screened each paper correctly, I had the methods next to with the PICOS in view of my eye so that I knew I was accepting/removing the most suitable studies. 5. Adam Dickinson and David Maidment were involved in screening and provided input when it was most necessary.

Question 36

Were meta-analyses were not possible, how did you undertake your narrative synthesis, what strategies did you use and why?

Answer 36

I looked through each study and their results for what they found and I descriptively summarised what they found. So, did they find a statistically significant finding/or not and then I basically just grouped them all together based on the studies that found an improvement and the ones that didn't.

Question 37

Why do you think your moderator analyses were not statistically significant?

Answer 37

This is mainly due to a small sample of studies included in my review and possibly because I focused on a singular disability rather then grouped (e.g., tetraplegia/paraplegia/manual wheelchair users). If I had a bigger number of studies (rather then 6) this may have led to a bigger and more reliable finding.

Question 38

Study 1: Given your null results, do you think your review actually adds anything substantial to the literature?

Answer 38

Yes, I would say the review was a timely piece of work and it was done at a time where it needed more understanding of the effectiveness of existing interventions in this specific population. We also know that the use of theory is not as effective compared to similar studies. Overall, I think the study is still contributing something because the majority of systematic reviews focus on multiple disabilities whereas this focuses on a novel area of just one. We are highlighting the gaps in the literature which supported the rest of my PhD.

Question 39

For Study 2, Summarise your approach and steps undertaken for this research?

Answer 39

We followed the Behaviour Change Wheel (BCW) guidance for developing/evaluating Behaviour Change Interventions. We also did this study during COVID-19, so it considered a helpful study while I couldn't collect data. A lot of the work I did at the beginning of my PhD were all done during lock down.

Question 40

Why did I do Study 2? Why did I use the BCW framework?

Answer 40

I did study 2 for the following reasons: 1. Unable to collect data during COVID due to lock down so this was seen as a study that was possible while I wait to collect data. 2. The Accessercise app was yet to have it's behaviour change potential assessed, so we wanted to check this to see what the app is currently like.

Question 41

Why retrofit the BCW onto an existing intervention (or app). Shouldn't this all be done before you develop the app?

Answer 41

As a supervisory team, we were not involved in the intervention development. David recommended that we follow this approach to fully understand the problem in behavioural terms and help understand what Accessercise is all about, what BCTs it includes, what's it behaviour change potential.

Question 42

Why did you use Accessercise? Why this app and why not use another app that has already been developed (e.g., WHEELS)?

Answer 42

1. Other apps (e.g., WHEELS) do not just focus on PA. 2. COVID was unable to develop an intervention. Accessercise approached us. 3. Accessercise is yet to be fully investigated and is NOVEL

Question 43

How did you make the decisions of judgements at each stage when you're using the BCW? So how did you decide what BCTs, how did you make the judgement for the APEASE criteria especially as they are quite subjective? Was there anything that you could have missed and did you actually consider everything?

Answer 43

Yes, so the BCW book helped us go through everything in a structured way steps 1-8. Any decision that I made I worked closely with my supervisor team, and seeked guidance from the Peter Harrison Centre. In the BCW book, it says when you are deciding and making some of these judgements, expert opinion (PHC, Supervisors) that supported me. It's evidence based and the systematic review supported!

Question 44

For Study 2, the BCW suggests that you do a barriers and facilitators study. Why did you not do that?

Answer 44

David and I discussed this and we noticed that there is a lot of research in the area looking at that so there was not a need for me to do that (I NEED TO REMEMBER THIS)

Question 45

So, for study 2 why are the results important? Why does it actually matter if Accessercise has good behaviour change potential?

Answer 45

It can reach a wider population and more likely that the intervention will help people with disabilities change their behaviour. In other words, we are more likely to change their behaviours successfully in the longer term. We are doing all this development work to sure that this is a worthy app that we should invest money and time into evaluating because it's going to work.

Question 46

If the app is already available. Did the developers not already think it was suitable and usable? If it's already commercially available why did we need to do this study if the developers could think it's been used?

Answer 46

Unfortunately, the app was not evidence-based and apps should be evidence-based. We wanted formal evidence that it was usable, which is what they wanted.

Question 47

Why only focus on Accessercise and not other apps?

Answer 47

This app is novel and has yet to be investigated which other apps have. Accessercise is the only one in English. The majority of apps are in dutch.

Question 48

Why have you only looked at 12 participants?

Answer 48

Research has indicated the between 12-15 participants is suitable enough to determine usability. Manageable database to use!

Question 49

Why have you not compared/contrasted other apps?

Answer 49

It's out of the scope of my PhD. Quite time consuming. Future research could do this! Accessercise was new, Accessercise was the only one in English (I was aware of).

Question 50

The think aloud participants (Study 3) are the same as the feasibility participants. Why did you do that, why didn't you recruit other participants?

Answer 50

Due to time constraints of my PhD, as we did these studies alongside one another, it made sense to use the same participants. We only had like 4 months. I used a mixture of intervention and control groups but when we assessed usability it was the first time they used it.

Question 51

Why did you do Think-Aloud? Are there alternatives methods you could have used?

Answer 51

1. Think-Aloud interviews can help us overcome some existing options such as interviews, questionnaires because of recall bias and social desirability bias. 2. Think-Aloud Interviews are helpful because they can help us to detect usability issues in real-time, which can help us determine the source of problem. 3. Think-Aloud Interviews help us to understand what the participants are thinking when using the app for the first time.

Question 52

Did you have any problems using the think aloud approach? Were participants quite open with you about talking about the app? Was it hard to get them to think aloud?

Answer 52

1. First of all, I did a think-aloud pilot test to ensure the questions were suitable for the participants to answer/check for any ambiguities. 2. Yes, I think Think-Aloud was a suitable better to get the data we needed. 3. I built rapport with them, I told them I was not an app developer so they could talk freely about the app.

Question 53

Why did you use the uMARS as your framework for your content analysis?

Question 54

Study 3: Did you do Content or Framework Analysis?

Answer 54

I did Content Analysis! Content Analysis is quick, easy, suitable for think aloud, I’m looking at it very surface level does not need that epistemology/ontology The research for the think aloud is very surface level. This method was appropriate for the research usability. I just wanted to see how it was used on a semantic level (what were common frequency of use, I did not need to look at it in depth), it was deemed the most suitable for the research question. We used that FRAMEWORK (Umars) because it’s validated, robust, commonly used. Time – I had 4 months to do all of this because it was towards the end of the PhD Content Analysis – My research question was just looking at usability this was deemed the most suitable method. I acknowledge that I could have used another methodology but for the purpose of time, research question, I considered this the most approach methodology.

Question 55

Why did I use Deductive and why uMARS?

Answer 55

uMARS - Relatively easy framework to understand, reduced number of questions. It aligns closely with the Content Analysis approach.

Question 56

But why Content Analysis for your usability study and not thematic analysis?

Answer 56

It’s ok to say I did a modified version of content analysis. I didn’t want to do frequency because I didn’t want to be too quantitative because I am following this pragmatic paradigm, which is why I didn’t count the number of quotes but it was very much looking at the frequency of how often something was said Content Analysis is quick, easy, suitable for think-aloud because I am looking at it very surface level. It doesn’t need all that epistemology/ontology because it’s just looking at how people are using something.

Question 57

Why did you do reflexive thematic analysis for your last study?

Answer 57

Overall, reflexive thematic analysis was the most important method to answer my research question as I was the researcher in the research. I am the researcher in the research and I could then reflect on what was being said because I was also delivering the study.

Question 58

Why in your SRQR you mention you are doing thematic analysis and the the findings are content analysis?

Answer 58

I'm aware that it seems confusing. However, when I start familirisation, I did thematic analysis because we did it inductive. We used inductive thematic analysis because it allowed us to be reflective because I was the researcher in the research because I was administrating the project so it gave me a bit of that reflexivity We did inductive reflexive thematic analysis because it enabled me to reflect on the process as well because I was involved. Also, there is a content analysis map in the supplementals but that only informed my analysis, so I appreciate that it could be viewed as confusing and suggests that it could be content analysis but that's how I started the analysis but I actually followed the thematic analysis procedure.

Question 59

Why did you do Deductive Content Analysis for your Think-Aloud Interview Study?

Answer 59

It’s ok to say I did a modified version of content analysis. I didn’t want to do frequency because I didn’t want to be too quantitative because I am following this pragmatic paradigm, which is why I didn’t count the number of quotes but it was very much looking at the frequency of how often something was said Content Analysis is quick, easy, suitable for think-aloud because I am looking at it very surface level. It doesn’t need all that epistemology/ontology because it’s just looking at how people are using something.

Question 60

Study 4: Why did you use a stop-go (traffic light system) for your feasibility study?

Answer 60

1. Early Decision Making - Helps determine whether a full-scale trial is worth doing. 2. Minimises risk: Stop-go criteria help minimise the risk to participants by ensuring that interventions are safe, reliable and won't cause harm to participants. 3. Cost-effectiveness - It helps to determine whether the cost of undertaking a full-scale trial are worth doing.

Question 61

Study 3: How does the usability of Accessercise compare to other apps in this area (better/worse)?

Answer 61

The usability of the Accessercise app compares similarly to other apps that exist for example WHEELS also reported positive usability from their mHealth app.

Question 62

Study 3: What were your key findings and why are these important? What do they add to the literature? What about the other aspects of the uMARS that weren’t identified – are these important and do you think these need to be addressed?

Answer 62

Overall, the usability study demonstrated that the Accessercise app provides a simple platform that, despite several troubleshooting issues can be efficiently operated by various users with different physical disabilities.

Question 63

Study 4: Why did you opt for a feasibility study?

Answer 63

By adhering to the MRC framework, the next logical step was to undertake a feasibility study involving the Accessercise app. By doing this, we would determine whether the Accessercise app is cost-effective and is worth undertaking in a full-scale trial.

Question 64

Study 4: Did you experience any issues when undertaking this study and why? How could these be addressed in the future?

Answer 64

Recruitment and adherence were key challenges within this study. Below are some reasons: Problem 1: Recruitment - Recruiting participants into my research was challenging. As people with disabilities are a hard-to-reach population and have insufficient skills with smartphones, I had to apply multiiple methods to recruit them. However, I still struggle to overcome this. Solution 1: Working closely with GP's, NHS and community centres to get closer connections with people with disabilities to interest and engage them into my research. Problem 2: Maintaining people with disabilities into my research was challenging. People would drop out and not want to take part. Key reasons for this could be because of accessibility barriers, intervention not tailored towards their needs, and not enjoying the study. Solution 2: Providing incentives (i.e., money to overcome barriers), completing sessions at home and completing the physiological outcomes at hubs or the partivcipants home).

Question 65

Study 4: Are you aware of any other apps in this area? Are they effective/usable and how does Accessercise differ?

Answer 65

Apart from Accessercise, there is WHEELS and SUPERapp. WHEELS didn't report a change in PA post intervention but SUPERapp appears to be an effective application. Differences: Accessercise is the only one in English, focuses specifically on tailoring content to the individuals needs.

Question 66

Study 4: Can you justify/explain why you selected your population, comparator, outcomes, and study design?

Answer 66

Study Design: Followed the MRC Population: App focuses on people with physical disabilities Comparator: Not sure Outcomes: WHEELS and SuperAPP both focus on 12 weeks. Also, as this was the last study of my PhD I did not have much time to do anything longer then 12 weeks. My deadline and PhD funding was running out.

Question 67

Study 4: How did you decide your feasibility criteria (i.e., traffic light system)? Are you aware of any alternative approaches? Any issues given not all participants completed an interview?

Answer 67

Traffic light system (stop-go) was employed as it's previously been effective in previous research. The traffic light system was used because it helps to determine whether a full-scale trial is worth it. In addition, it helps to prevent participants from harm if a full-scale trial is unethical and helps determine whether the cost of undertaking a RCT is worth doing. Unfortunately, I'm not aware of any other approaches but based on feasibility studies, the traffic light system is highly recommended. Yes, unfortunatley, it doesn't provide a balanced view of the participants thoughts on the study design. As I was unable to interview the participants that dropped out I was unsure of their thoughts on the study and the key reasons behind their drop out.

Question 68

Study 4: Why include a qualitative component and why use inductive thematic analysis as opposed to another analytical method (e.g., content analysis, as in Chapter 3)? There is a content analysis map in the supplemental at the end of this chapter, so did you actually undertake content analysis and not thematic analysis?

Answer 68

Why we did this study: To understand participants’ experiences of taking part in the Accessercise feasibility trial, as well as intervention implementation and fidelity. Why did we do thematic analysis: This is because we were involved in the research as we were researchers in the research and we wanted to be reflexive and understand our position within the research.

Question 69

Study 4: Why only 12-weeks for the trial period, why not shorter/longer? Would this time be sufficient to see changes in outcomes, why/why not?

Answer 69

Based on previous research (e.g., WHEELS, SuperTheory) 12 weeks was suitable for the time I had left on my PhD. I only had 3-4 months left before I had to submit as I was running out of funding Feasibility studies are small and help to determine whether the study outcomes, research design, equipment used are suitable. Unfortunately, yes, you're right. We are unlikely to notice a change because it's a feasibility study and we are not powered to asses statistical differences. In addition, in order to notice behaviour change, this is likely to occur after 12 weeks are more likely within 12 months.

Question 70

Study 4: Can you explain the how/why you went from 551 invites to 87 assessed for eligibility, to only 20 taking part? Why such a low take-up overall?

Answer 70

I contacted participants using a range of different approaches. For example, word of mouth, social media posts, NGBs but most important I messaged potential participants through Facebook Messenger. From this 551 participants, a certain number of participants did not have a physical disability or mentioned a barrier that would make it unlikely for them to participate (e.g., lack of time, did not want to do in-person assessments, or did not respond to my initial request). Based on the 552, 87 participants shared an interest and I then assessed them against my inclusion criteria (e.g., had a physical disability, lived within the UK, had access to a smartphone device) and from this 87 only 20 participants were eligible and wanted to take part. 20 participants were then randomised into either the intervention (n=10) or the control (n=10).

Question 71

Study 4: Why do you think participants dropped-out? Did this have an impact on your results?

Answer 71

1. Lack of time 2. Other commitments 3. Barriers to accessing the research 4. Intervention was not tailored to needs of the participants 5. Lack of support/check in-s from the interventionist I don't think it impacted my final study that much. The reason for this is because the data from participants was not negative as I used the pre for the post data for those that dropped out. However, it did impact the qualitative study interviews because I was unable to determine why the participants dropped out and their experience of the trial.

Question 72

Study 4: Why assess usability again, when this was done in Chapter 3? Were these the same participants as Chapter 3? Why not recruit new participants?

Answer 72

The way usability was used was completely different. 1. Usability in the think-aloud study was used a theoretical framework for the Content Analysis and did not allow the participants to directly measure their usability of the app (Qualitatively) 2. Usability in the feasibility study - Quantitatively measured the participants usability based on the questions.

Question 73

Study 4: If not powered to do so, why test for differences in quantitative outcomes and why (non-parametric) Mann-W U tests?

Answer 73

Helpful to determine whether these outcomes would be suitable for a full-scale trial.

Question 74

Study 4: If the trial is feasible, but with modifications, what would be the next step? What modifications would you make to address some of the issues you experienced?

Answer 74

Before undertaking a full-scale trial, I think these modifications would need to be addressed: (1) App’s layout (placing content in alphabetical order), (2) Visual appeal (including an in-built colour-changing setting) (3) Interactivity (providing real-time in-depth feedback) (4) Quality of information (number or bullet pointing content) (5) Quantity of information (removing content such as equipment not related to a specific goal)

Question 75

Discussion: How do your findings from Accessercise differ from other similar apps already evaluated/available?

Answer 75

My findings are similar to other app studies (e.g., WHEELS/SUPERapp) which find mHealth apps as an effective, accessible and usable solution to help people with disabilities undertake physical activity.

Question 76

Discussion: Do you think mHealth is best to increase PA in people with physical disabilities, why/why not?

Answer 76

I think every solution has it's strengths and weaknesses. While some participants for example may like mHealth interventions (work for them) and others not, it provides people with disabilities with a relaible, suitable and accessible solution that may support them in changing behaviours and increasing their physical activity levels.

Question 77

Discussion: What is the value of co-design/participatory approaches – how would you recommend going about this? Are you aware of any frameworks that help guide this process?

Answer 77

Co-design (or participatory design) involves stakeholders, including the end-users (in this case, people with disabilities), in the design process to ensure that the intervention meets their needs and preferences. Some recommended co-design frameworks can be the following: 1. The Behaviour Change Wheel (BCW) 2. The Person-Based Approach The Person-Based Approach emphasises the importance of understanding the perspectives and experiences of the target population. This framework involves: Step 1: Qualitative Research: Conducting interviews, focus groups, and observations to gather insights into the needs, preferences, and barriers faced by individuals. Step 2: Development of Guiding Principles: Creating principles based on qualitative research findings to guide the intervention design. Step 3: Iterative Refinement: Continuously refining the intervention through feedback from the target population.

Question 78

Discussion: Do you think the MRC framework was useful, why/why not?

Answer 78

Yes, the MRC framework was valuable. The MRC framework provided a systematic approach to developing, evaluating, and implementing complex interventions.

Question 79

Discussion: What might you do differently?

Question 80

Discussion: What key things have you learnt from undertaking this research?

Answer 80

1. Changing behaviour is difficult 2. Recruiting people with disabilities remains a challenge in research 3. mHealth is an accessible, suitable and support method to improve PA in this area 4. Accessercise has good behaviour change potential

Question 81

Discussion: What more could be done in the future, where/what could this research lead to?

Answer 81

1. High Quality RCTs, which includes objective measures of PA, bigger sample sizes and longer period for time. 2. To examine whether different modes of delivery can improve outcomes in people with disabilities such as quality of life, adverse effects. 3. To evaluate the overall effectiveness of the Accessercise app in a full-scale trial progressing from Stage 2 to Step 3 (MRC) 4. To improve recruitment and retention in mHealth interventions among people with disabilities through NHS, GP services etc 5. To improve the access to mHealth for people with disabilities that reduces stigma etc

Question 82

What is passive (comparator)?

Answer 82

The participants receive nothing - They are required to continue with their day to day.

Question 83

What is active (comparator)?

Answer 83

We will give participants part of the intervention (e.g., some of the intervention such as leaflets, guidelines etc) – Change behaviour slightly

Question 84

What are the main issues and debates in this subject area?

Answer 84

Enter here

Question 85

What are limitations of telephone physical activity interventions for people with disabilities compared to mHealth (app) based interventions?

Answer 85

1. Lack of Visual Content: mHealth apps can provide video demonstrations, visual instructions, and graphical feedback, which are crucial for understanding and performing exercises correctly. Telephone interventions lack this visual component, making it difficult to convey detailed exercise instructions. 2. Interactivity and Engagement: mHealth apps often include interactive features such as progress tracking, goal setting, reminders, and gamification elements that can enhance user engagement and motivation. Telephone interventions tend to be more static and less engaging. 3. Accessibility and Usability: While telephone interventions can be accessible to those without smartphones, mHealth apps are often designed with accessibility features such as voice commands, text-to-speech, and compatibility with assistive devices. This makes them more usable for individuals with various disabilities compared to a standard telephone conversation. 4. Real-Time Feedback and Adjustments: mHealth apps can provide instant feedback and make real-time adjustments based on the user's performance and data input. Telephone interventions usually involve delayed feedback and less immediate adjustments.

Question 86

Why is recruiting people with SCI difficult?

Answer 86

1. People with SCI have unique needs, problems, limits, minority issues, and demographic distribution. 2. SCI is difficult to recruit due to the sporadic or unpredictable nature of SCI onset and the heterogeneity of impairments and comorbid health conditions.

Question 87

What are the implications of your research 'for practice'?

Answer 87

Overall, our research is encouraging for the community of people living with physical disabilities because our finding have indicated that Accessercise is feasible, adhered to and well perceived among people with physical disabilities. Obviously this intervention will work for some individuals but not others, but it provides a strategic approach to overcoming physical activity barriers and help in addressing low physical activity rates in this population. In addition, our findings have help people with disabilities develop physical activity habits or to increase self-regulation. Lastly, our research has shown the benefit of engaging in research with people with disabilities, which may support other research when designing/evaluation interventions in this population.

Question 88

What are the implications of your research 'for theory'?

Answer 88

We know that getting individuals to perform in a behaviour is difficult and changing behaviours is a complex construct because behaviours are habitual, normative and preventative. We know that interventions that aim to increase PA in people with disabilities need to use strategies to enable engagement in a particular behaviour. Overall, our research has shown the benefit of using BCW which has indicated that Accessercise incorporates the necessary components for individuals with SCI to be physically and psychologically capable of undertaking PA, provides physical and social opportunities to reduce sedentary behaviour and provides automatic and reflective motivation. This PhD thesis was under pinned using systematic procedures such as the MRC and the BCW, which are strategies and strategies for developing/evaluating behaviour change interventions in people with disabilities. Therefore, researchers in the future may consider the advantages of using such frameworks when designing and evaluating similar health behaviour change interventions.

Question 89

What are the implications of your research 'for healthcare

Answer 89

Whilst we know that health care professionals have limited input when it comes to promoting/encouraging physical activity due to a lack of knowledge, awareness, time and confidence, the findings in my PhD thesis could be used as guidance within a health care professional setting. For instance, health care professionals working in an area of physical activity for people with disabilities could use my work to help people with disabilities to make sustained changes to their physical activity behaviour using mHealth interventions.

Question 90

What are the key findings from Study 1 (Systematic Review)?

Answer 90

Overall, our systematic review has shown that physical activity interventions have the potential to result in a broad range of beneficial outcomes in manual wheelchair users with SCI. For example, the meta-analysis has shown that physical activity favoured the intervention group relative to the control group. Moreover, based on our assessment of GRADE, we can be moderately "confident" that the true effect is likely to be close to the estimated effect. Unfortunatley, due to heterogeneity in terms of interventions, study designs and outcomes, we were unable to undertake meta-analyses on the remaining outcomes. Therefore, our narrative synthesis identified similar improvements in behavioural physical activity, functional fitness, physical health, psychosocial wellbeing, health related QoL. However, because of risk of bias and study quality varied across the study, we cannot be certain that these findings are correct and need to interpret these with caution. As a result, we need more high-quality research into this area. In addition, we also identified that most of the studies included in the systematic review incorporated health behaviour change theory (e.g., Health Belief Model, Social Cognitive Theory), which is valuable in designing and evaluating behavioural interventions. This finding is particularly valuable because previous research has shown that PA in people with physical disabilities significantly improves when theory underpins interventions compared to when it is not used. This finding is interesting given that our moderator analyses shows that interventions that were not underpinned by theory were more effective than when they are used. However, this finding is likely to be due to a small number of studies incorporated within the meta-analysis. In addition, another key finding is that we coded for BCTs across all studies, which is important because this can provide an understanding of mechanisms and assessment of intervention effectiveness. However, we noticed that only 75 BCTs were reported across all studies with an average of 9 BCTs per study, which is quite low. Future research should incorporate these BCTs to determine their effectiveness in improving PA in adult manual wheelchair users with SCI.

Question 91

What are the strengths of the systematic review (Study 1)?

Answer 91

1. We incorporated a range of different study designs in our research unlike existing reviews (e.g., Martin-Ginis). For example, we used RCTs, Non-RCTs, before/after. 2. We used many different outcomes, searched multiple databases (i.e., 9 and used many risk of bias and study quality assessments such as Risk of Bias 2, GRADE, Downs and Black (1998) checklist and also undertook moderator analyses.

Question 92

What are the limitations of the systematic review (Study 1)?

Answer 92

1. Interventions were no longer then 12 months with the majority of the outcomes being 5.5 months, which reduces our understanding on whether these interventions have long term behaviour change potential. 2. All the studies were undertaken in English, which limits other valuable research papers published in other languages from being included in our review. This may explain why most studies were undertaken in developed countries (e.g., UK, USA and Canada). 3. The majority of our outcomes were self-reported using a range of different questionnaires (e.g., LTPAQ-SCI, PARA-SCI), which can make the findings inaccurate or unreliable. Also, self-reported outcomes are at risk of recall bias, careless responding and social desirability bias.

Question 93

What are the key findings of Study 2?

Answer 93

1. Overall, our research has shown that Accessercise is an appropriate and practical intervention that can support people with SCI undertake more PA. The Accessercise app incorporates the necessary components for people with SCI to be physically and psychologically capable of undertaking PA, offers physical and social opportunities to reduce sedentary behaviour and provides automatic and reflective motivation. 2. Based on the ABACUS criteria, our study has revealed that Accessercise has high behaviour change potential, suggesting that Accessercise includes many BCTs which can address the behavioural issue (I.e., increasing PA). 3. Also, our research is promising given that Accessercise was not initially underpinned by theory, but this is not surprising given that many mHealth apps are not theory-based, making it challenging to determine the most effective intervention components that may be suitable in changing the behaviour. 4. In conclusion, "retrofitting" the BCW to the Accessercise app has been helpful, as it has allowed us to identify the BCTs that have/haven't been included that can facilitate physical activity behaviour change.

Question 94

What are the limitations of Study 2?

Answer 94

1. Despite the BCW being helpful in designing intervention, it lacks a formal guide in translating BCTs, which then relies heavily on the expertise of the researchers/app developers 2. Time consuming to complete all the worksheets across Stages 1-3 and Steps 1-8. 3. As Accessercise is under development and is constantly changing, it was difficult to determine the right time to evaluate the app's behaviour change potential 4. Designing/implementing/evalauting mHealth apps is time consuming and labour-intensive. End-users and technology can change over time.

Question 95

What are the key findings from Study 3?

Answer 95

1. Overall, our research adds to the limited number of studies that have assessed the usability of physical activity mHealth interventions in people with physical disabilities. 2. Accessercise offers a simple platform that despite several trouble shooting difficulties (e.g., layout, quality of information, quantity of information), can be effectively operated by people with disabilities. 3. Most participants liked the features of the Accessercise application in terms of entertainment, customisation, ease of use, navigation, which shows overall good usability. 4. Accessercise appears to be a suitable intervention that can help people with physical disabilities overcome barriers to PA. 5. Our study is one of the first to show that a think-aloud protocol is acceptable and can be informative in the context of increasing PA in people with physical disabilities.

Question 96

What are the limitations of Study 3`

Answer 96

1. The majority of participants were White British, male, had a SCI and lived within the East Midlands of England. 2. During testing, several components of the Accessercise application were under development (e.g., blogs, podcasts. nutrition), which may have impacted the users overall satisfaction with the app. 3. Interviewing participants in a controlled setting and audio-voice recording their responses, may have led to partcipants overly praising the app. 4. Participants completed a think-aloud interview in a controlled setting under the observation of a researcher, which is not normally how participants would interact with the app. Therefore, data on participants thoughts in a more ecological setting could have been missed.

Question 97

What are the key findings from Study 4 (Feasibility)?

Answer 97

1. According to the traffic light system, the adherence and retention to follow-up was 'amber' in other words 'moderator' which means that the study is feasible to proceed to a full-scale trial but warrants amendments. 2. In terms of implementation and intervention fidelity, the majority of participants were satisfied in terms of the researcher in the research, study equipment, length of intervention. However, whether the app can support users achieve their PA goals and whether users would recommend the app to other users were rated unfavorable by some participants.