Self-rehabilitation strategy for rural community-dwelling stroke survivors in a lower-middle income country: a modified Delphi study.

Background: More than half of stroke survivors in lower-middle income countries lack access to stroke rehabilitation services. The promotion of self-rehabilitation could be promising for addressing stroke rehabilitation inadequacies in lower-middle income countries. Self-rehabilitation interventions are more readily acceptable to community-dwelling stroke survivors, and therefore, have the potential to boost the successful realization of the Sustainable Development Goals and other WHO rehabilitation goals. We report a consensus-building process that sought to identify which task trainings are relevant to include in a task-specific self-rehabilitation strategy for rural community-dwelling stroke survivors.

Methods: An iterative two-stage mixed-method consensus-building approach was used: (1) focus group discussions (n =  5) with rural community-dwelling chronic stroke survivors were conducted to explore personal life experiences in performing daily activities, and the results were used to develop a list of candidate task trainings that could be included in a task-specific self-rehabilitation intervention model for improving functional ability of survivors; (2) a three-round Delphi exercise with a panel of stroke rehabilitation experts to establish consensus on the importance/relevance of the developed task trainings. Consensus was pre-defined to be the point where the proportion of items given a rating of 3 (quite relevant) or 4 (highly relevant) by expert panellists is ≥  0.8. Kendall's coefficient of concordance (W) was used to assess the level of agreement among the expert panellists.

Results: A list of 74 task trainings was generated from the results of the focus group discussions involving 29 chronic stroke survivors. The tasks were classified as follows: training for the upper extremity (37); lower extremity training (21); trunk training (7); and balance training (9). A panel of 13 stroke rehabilitation experts reviewed these task trainings using the Delphi method and consensus was reached on keeping 28 task trainings in the first round (Kendall's W =  0.252, p <  0.001) and an additional 7 in the second round (Kendall's W =  0.409, p <  0.001). In the study team's analysis of open text responses, several areas of debate were identified and some task trainings were modified. The exercise yielded 49 task trainings (66% of 74) on which there was consensus (the mean proportion of items given a rating of 3 or 4 by panellists was 0.93; Kendall's W =  0.291, p <  0.001) to keep 3 task training groups relating to: upper extremity (27), lower extremity/balance (8), trunk strength (4) as well as warm up exercises (10).

Conclusion: The study provides a consensus-based view of the features of a task-specific self-rehabilitation training strategy to improve outcomes following a stroke. This self-rehabilitation training strategy can be used as an intervention approach to augment and promote stroke rehabilitation among rural community-dwelling stroke survivors, especially in sub-Saharan Africa.