Establishing collaborative partnerships to improve stroke care in India: a synopsis of IMPROVISE, a multi-site feasibility study

Background: Globally, five and a half-million people die from stroke each year and 13% of all stroke deaths occur in India. The programme had two main projects IMPROVIng StrokE care in India (IMPROVISE) and IMPROVIng Stroke care in India – Advancing The INSTRuCT Operations and Network (IMPROVIS-ATION), delivered through six workstreams. IMPROVISE aimed to explore the feasibility and acceptability of implementing three evidence-based care bundles into practice. IMPROVIS-ATION aimed to explore the feasibility of implementing care bundle 1 in four additional hospital sites; the provision of post-discharge care in seven hospitals; and the feasibility of establishing a multicentre ethics approval process within the Indian Stroke Clinical Trial (INSTRuCT) network. IMPROVISE methods: A multi-centre, feasibility study and nested process evaluation. Three care bundles were implemented sequentially at three hospitals. Care bundle 1: a Global Evaluation of Swallowing and a hydration ‘Osmolarity App’. Care bundle 2: a Standardised Neurological OBservation Schedule for Stroke. Care bundle 3: post-discharge patient/carer education in the form of animations. Process and outcomes were evaluated in each of four cohorts admitted between July 2019 and November 2021. IMPROVISATION methods: Semi-structured interviews (all work packages) and focus groups (work package 3) with purposive samples of health professionals, patients and carers (work packages 1 and 2), ethics committee members and principal investigators (work package 3). IMPROVISE findings: Of 707 patients screened, 515 were eligible and 379 (73.6%) were recruited to the study;118 (31.1%) female, mean age 59 years (12.8 SD). Overall median National Institute Health Stroke Scale was 9 (interquartile range 5–16); 285 (75.5%) of participants had a modified Rankin Scale ≥ 3. Global Evaluation of Swallowing swallow evaluations increased to 51 (56.0%) and calculated osmolarity was recorded in 67 (75.3%) in care bundle 1, maintained in care bundles 2 and 3. Standardised Neurological OBservation Schedule for Stroke was recorded at similar levels and maintained to care bundle 3. Four animations were provided in hospital to all relevant carers and patients. The process evaluation found that care bundle implementation resulted in improved decision-making, new roles and responsibilities. Barriers to implementation were patient/caregiver literacy, physical and workforce resources and organisational culture. IMPROVIS-ATION findings: Work package 1: 16 clinical staff and 12 patient/carers reported variation in practice arising from a lack of specialists and resources. Work package 2: 66 clinical staff and 102 patient/carers identified unmet needs related to swallowing problems and the psychological impact of stroke. Work package 3 (multiple stakeholders): 12 interviews and 2 focus group participants (n = 18) identified some streamlining of ethical approvals following the COVID-19 pandemic, but sustainable standardised procedures were needed for multicentre studies. Conclusion: Evidence-based, context-specific interventions to improve the basic elements of stroke care can be successfully implemented and sustained as an acceptable model of care. Future work: Future work is planned to explore the needs of patients and their carers to identify and develop lowcost rehabilitation services. Limitations: The impact of COVID on hospital systems and processes of care is likely to have influenced the results of the study, including fewer patients recruited than planned, the admission of more severe stroke patients, delays in patients being admitted Stroke Units and the rotation of trained staff to other hospital departments. Funding: This synopsis presents independent research funded by the National Institute for Health and Care Research (NIHR) Global Health Research programme as award number 16/137/16. A plain language summary of this synopsis is available on the NIHR Journals Library Website https://doi.org/10.3310/TTDD0404.