We evaluated activity and participation levels in each participant using valid and reliable tools. Their functional ability was measured using AbilHand questionnaire and participation levels using Lawton Instrumental Activities of Daily Living scale.28,29

We measured the amount of arm use, using Motor Activity Log supplemented by their personal daily activity log. The scores were used to categorically identify participants with high and low functional arm use. Their current quantity of functional task practice was ascertained from the duration for which they attempted activities as shown in Table 1. This additional categorization as high, moderate and low amount of practice was used to identify those who were optimally exercising.30

To compare amount of arm use between participants with different levels of motor ability and spontaneous recovery potential, we used a categorisation framework by operationally defining these constructs (Table 1).31 Their perceived magnitude recovery was documented using perceived recovery component of stroke impact scale. The functional scales and categorisation of the participants were utilized to clearly define each participant in terms of their ability, recovery potential and resultant amount of functional arm use.

We conducted face to face semi-standardized interviews with the participants along with caregivers in their home using an interview guide addressing capacity, opportunity and motivation for arm use and perceived recovery (Table 2). List of factors influencing physical activity among general and stroke population were identified by reviewing existing literature by the team members. Two experienced stroke physical therapy clinicians and one psychologist not directly involved with the study validated the domains and sub domains for questioning. The caregivers were questioned about the nature of support such as physical, emotional, informational support that they provided. Their presence helped to complement and corroborate the information provided by the survivors.

The semi-standardized interviews were conducted with participants, audio recorded and finally transcribed with the support of additional field notes. We refined the interview guide after analysing the initial interviews. The duration of interviews ranged from 30minutes to 2hours with an average of 45minutes/participant and was re-verified with participants.

We used qualitative content analysis software to perform analysis.32 We inductively coded physical capacity, environment and motivational contextual domains. Interview transcripts were read by the team members involved and parts of the transcript that were unclear were clarified with participants and they corrected informational discrepancies. Relevant inductive codes that were perceived contributory for arm use by proximity analysis formed subcategories. Prolonged engagement and member checks were used in the subsequent steps to understand and categorize the data. Next, the subcategories within each category were populated by content analysis and discriminated between those exhibiting positive versus negative arm use beliefs or behaviours. Eventually, the unique subcategories so identified were integral in framing final themes. Redundancy of major categories guided identification of data saturation. Data collection was stopped after interviewing 23 participants.

VR conducted all the interviews, coded and analyzed the transcripts. JMS, NM and JA verified the coding and interpreted the categories. All four framed the themes. VR, JMS and NM were physical therapists and JA was a clinical psychologist with all having more than 15 years of experience with stroke survivors. VR was research scholar and the other three were faculty with PhD degrees. Reflexivity was maintained against incorporating prior health behaviour theories during interpretation and theme formation.

Many participants were unaware that they have to exercise their arm to augment the natural recovery process. They believed that arm will automatically improve after stroke. [Participant 08], a chronic stroke survivor, expressed “I keep thinking when this arm and leg will start moving? When will it get alright? And when will I be able to perform my activities? I pray daily to God for improvement”. [Participant 12] another chronic survivor expressed, “We asked our doctor why am I not able to move my arm? He said it will be alright in due course. So we are waiting for improvement”. Some survivors, despite attaining negligible improvements were prospectively awaiting recovery even beyond two years without initiating any remedial measures. Certain others felt that the problem in arm was not due to the stroke. [Participant 6] described “When we came to hospital [acute hospitalization] problem in brain was treated. Now only the arm has to get alright”. A number of participants did not actively seek information on what measures to initiate for regaining arm function.

Very few participants [Participants 1, 2, 19, 20] could explain how to perform functional task practice to facilitate relearning of lost abilities. All the remaining participants were unclear on how to engage arm during activities or perform functional task practice. [Participant 7] informed that his doctor advised him to try doing activities using arm, yet did not describe how to perform. Those who practiced active movements felt practicing functional activities as superfluous and believed improvement of arm movements will directly translate into functional gains. Hence they did not engage their arm in performing functional activities. Nevertheless, none of them expressed how much they should practice in a day for achieving nominal gains, despite the fact that some of them were regular attendants to rehabilitation sessions.

We quizzed them on availability, accessibility and utilization of objects they interacted with while performing ADL's. One participant complained about his inability to lift a drinking glass. [Participant 19] “I cannot lift drinking glass. I feel afraid that it might fall when I lift.” However, another participant with similar motor ability elaborated how he managed to perform the activity. [Participant 1] explained; “I used to fill water in big glass and try to drink. It was difficult. Then I started using thinner light weight glasses with little water.” Another participant [3] expressed “I found it hard to lift a bucket filled with water, and so I fill little amount of water and carry more times. He also said, morning, I take three milk packets two in this hand and one in this hand”. Participants modified the manner in which they used objects to decrease task difficulty during functional task practice. Four participants, performing home exercises had created exercise facilities such as pulley systems, custom made arm exercise equipment and hand railing to support when walking as in a physical therapy department. [Participant 3] had a hand railing inside a narrow pathway in his home enriching the immediate physical environment to support practice.

Caregivers were immediate family members. They reported providing physical assistance for performing activities and guided exercise sessions. Although the survivors acknowledged their contribution, they highly appreciated the emotional support extended during their most difficult times. Family members and care providers motivated them to attempt activities using affected arm. [Participant 5] explained “My wife asked me to eat using only the affected hand and not with a spoon”. She says, “even if it falls, it is ok, eat with this hand only.” Participant 20, wanted to open bottles. She expressed “My physio has told that I have to open bottle and show before Diwali [festival date]. She said, “If you really like me you will have to do this.” “I also feel that I will have to do it because she is the only person who talks nicely with me. I took it as a challenge, and I did it”. In contrast, few participants reasoned lack of emotional support from family members as a major deterrent to active life participation. Additionally, some perceived that family members were not competent to advise them. [Participant 7] explained “My family members support me a lot but I don’t rely on them for improving arm. How do they know about this?”

After being discharged from acute hospital care, most of our participants moved to centres offering complementary and alternative medicine [CAM] treatments. In the rural regions, these treatments were perceived as cheaper and to result in fewer side effects than rehabilitation treatments and hence the first choice not only for stroke but also for certain other diseases. In CAM therapy, medicinal oil and pastes are applied over the affected arm and patients are made to lie down for long hours. Advice to use arm actively were rare from CAM centres. Only three participants reported getting such an advice alongside the traditional treatments. Such advice was more common from practitioners of modern medicine. [Participant 1] described: “At CAM centre, doctor said it is good if you use the arm more. Even the other physician [neurologist who initially treated] also said, the more you use the arm it will be more beneficial”. Since participants underwent this treatment during sub-acute stage, a period of maximum natural recovery, whatever improvement ensued was attributed to these passive means and not to their active attempts.

Sustaining active task practice is indispensable to materialize the initial gains of practice. We inquired why they continued to practice. [Participant 1] explained; “I noticed that I was able to lift my hand and do activities more than before. Then I understood that I am getting alright. Initially, when I tried lifting tumbler to drink water, it used to fall. As I kept on doing, it improved. Later, I was also able to eat with this hand.” Another participant [Participant 2] reported; “After practicing daily activities it started improving. As I used more it improved more.”

However, other participants did not perceive such gradual increments. [Participant 7]: “Initially I was trying but it was slightly difficult. I tried till a certain point. I could not see improvement in any activities. Even now I want to do all daily activities but I don’t have confidence”. [Participant 6] PH: “Initially we were exercising more; we used to do exercise two times a day, more in the morning. Now it has reduced, only once”. Perceived improvements matching their expected recovery potential gradually shaped their beliefs over a period of time guiding them to ultimately sustain or stop their efforts.

Participants’ personal and environmental contexts did guide their choice to pursue active or passive means to regain the affected arm functions. Together, these factors modulated the degree to which they actively engaged their arm and sustained their efforts.