Abstract
BACKGROUND:
Independent mobility is the most important determinant of quality of life after stroke and it is vital that training aimed at restoration of gait is based on contemporary evidence. Despite several practice guidelines for gait rehabilitation after stroke existing globally, their feasibility of application in low-resource settings is often questionable.
OBJECTIVE:
To investigate the current practices in gait training among Indian physiotherapists involved in the rehabilitation of stroke survivors.
METHODS:
A questionnaire on the various aspects of gait training was developed and the content was validated by experts. The survey was made available online and distributed among Indian physiotherapists working in the field of stroke rehabilitation, using snowball sampling. Frequency distribution was used to summarize responses to each component of the questionnaire.
RESULTS:
Responses were obtained from 250 practicing physiotherapists. The majority of the respondents (55%) reported that they initiate gait training within seven days after stroke. Gait training sessions ranged from 15–30 minutes (55%), once every day (44%), and the majority (89%) reported use of subjective outcome measures to evaluate gait. Although most respondents agreed on the use of assistive aids, 24% indicated that their use may deter gait, rather than improve it. Nearly 86% of the respondents reported that they do not follow standard guidelines pertaining to gait rehabilitation for stroke survivors.
CONCLUSION:
The findings of the study point toward a lack of evidence-based practice among Indian physiotherapists while training gait after stroke. This implied the urgent need for development and implementation of country specific guidelines for stroke rehabilitation.
Introduction
The global burden of diseases study in 2001 estimated that over 85% of the global burden of stroke was borne by Low and Middle Income Countries (LMICs) and recent WHO reports state that the number of Disability Adjusted Life Years (DALY’s) due to stroke in LMICs is almost seven times that in High Income Countries (HICs) (Feigin, 2007). Particularly in India, a recent systematic review found that the cumulative incidence ranged from 105 to 152/100,000 population per year, and the crude prevalence was 44.29 to 559/100,000 population, which was higher than those of HICs (Kamalakannan et al., 2017). In addition, 7.3% of all deaths and 3.5% of the DALY’s in India have been attributed to stroke (The India State-Level Disease Burden Initiative, 2021).
As a consequence of stroke, approximately 50% live with hemiparesis and 30% have mobility limitations (Mozaffarian et al., 2016). It is well acknowledged that loss of independent mobility after stroke is the most important determinant of the level of disability (Langhorne et al., 2009; Bowden et al., 2011). Reports indicate that only 37% of the individuals are able to walk in the first week after stroke (Bogey & Hornby, 2007). Although 79% of stroke survivors eventually regain the ability to walk independently, only 41% regain independent outdoor mobility and 7% achieve safe community ambulation (Hill et al., 1997).
Globally, surveys of individuals with stroke show that walking is the most important goal of rehabilitation (Bohannon et al., 1988; Harris & Eng, 2004; Khan F et al., 2012). Consequently, gait training occupies a higher proportion of time during the rehabilitation sessions, in comparison with other activities (Latham et al., 2005). Emphasis on gait training begins in the early stages of rehabilitation and continues post-discharge for several years (Latham et al., 2005; Lord et al., 2004).
Considering the influence of gait on the quality of life after stroke, it is imperative that clinical practice aimed at restoration of gait should be based on contemporary evidence. Best-practice recommendations provide a scientific basis for clinical practice and is designed to promote maximal functional recovery (Duncan et al., 2005). Regular adaptation of these recommendations improves patient outcomes and enhances the quality of life of patients and caregivers. They also provide support for the use of interventions that have a proven benefit, while increasing awareness about ineffective methods (Duncan et al., 2005). Despite several practice guidelines for stroke gait rehabilitation existing world-wide, the feasibility of their application in low resource settings is often questionable. This may be attributed to a number of geographical, cultural and environmental factors, along with the limited availability of resources. Particularly in India, stroke is a significant healthcare burden (Kamalakannan et al., 2017) and yet there is lack of evidence-based guidelines for rehabilitation. Notably, very few studies have investigated the practice of physiotherapists in LMICs and hence, there is uncertainty about the nature of clinical practice in India. Therefore, the aim of the present study was to investigate the current gait training practices for stroke survivors among Indian physiotherapists.
Methodology
Study protocol was approved by the Institutional Ethics Committee and the trial was registered in the Clinical Trial Registry of India prior to onset of the study. The study was conducted in two phases. In the first phase, a questionnaire that focused on the various aspects of gait training for stroke survivors was developed and validated for content. Existing practice guidelines, evidence-based rehabilitation strategies and published surveys in stroke rehabilitation from the last 20 years were reviewed for development of the questionnaire. Five physiotherapists working in the field of stroke rehabilitation with a minimum of five years’ experience were identified as experts for content validation of the questionnaire. These experts evaluated the questionnaire for relevance, clarity and appropriateness. The percentage level of agreement was set at 80%, i.e. agreement from 4 out of 5 experts was required for inclusion of a question. Additional comments and suggested modifications were incorporated in the final questionnaire. The questionnaire was composed using Google Forms and the link to the form was made available.
In the second phase, a cross-sectional survey using the validated questionnaire was conducted among physiotherapists involved in stroke rehabilitation. In lieu of a simple random sample from an existing sample frame, a multimodal recruitment strategy was considered suitable to the present study. Since professional data banks of physiotherapists to identify practitioners involved in stroke rehabilitation do not exist in India, the snowball sampling method was used to identify the survey participants. This method involves the survey respondents themselves further circulating the survey link among other physiotherapists working in stroke rehabilitation. Physiotherapists practicing in India, with a minimum of one year of experience in the rehabilitation of stroke survivors, were considered eligible for the survey.
The link to the questionnaire, along with a brief study introduction and details of informed consent, were sent to physiotherapists in India through email and social network platforms. The therapists who were willing to participate in the survey provided informed consent and were asked to proceed by clicking on the link. Responses were anonymous and the respondents were assured confidentiality of the information provided by them. The survey link remained active for a period of six months from 1st October, 2018 to 31st March, 2019.
The data from the completed surveys was compiled and imported into SPSS (version 16.0) for statistical analysis. The data from open ended questions were coded into recurrent themes and frequency analysis was used to summarize the results of the survey.
Results
The main domains, number and type of questions in the questionnaire are enlisted in Table 1. The final questionnaire consisted of 30 questions outlining various aspects of gait training. In the three months that the survey remained accessible, 250 physiotherapists across India completed the survey. All the respondents were considered a part of the eligible sample as there were no incomplete responses.
Contents of the questionnaire
Contents of the questionnaire
Majority of the respondent physiotherapists belonged to the southern zone of India (n = 140; 56%), followed by western (n = 45; 18%), central (n = 37; 15%), northern (n = 20; 8%) and eastern (n = 7; 3%) zones. Experience level of therapists ranged from one to 23 years, with majority having a post graduate level of training (n = 165; 66%). The respondents were employed mostly in hospital settings (n = 132; 53%) followed by rehabilitation clinics (n = 82; 33%) across the country.
Survey respondents reported treating one to five stroke patients per week (n = 165; 66%). Additionally, respondents reported that their patients spanned across all the stages of recovery i.e. acute (n = 30; 12%), sub-acute (n = 154; 62%) and chronic (n = 66; 26.4%). Respondents practicing in hospitals and home-based settings reported treating patients in the early sub-acute stage of recovery (44% and 59%, respectively) while respondents in rehabilitation clinics reported treating mostly chronic patients (44%). However, subgroup analysis showed that gait training practices followed by the respondent physiotherapists did not vary with respect to their practice setting.
Time to initiate gait training
Of the survey respondents, 47% (n = 119) reported that they commence gait training between the third- and seventh-day post stroke. Figure 1 shows the time of initiation of gait training reported by study participants.

Time of initiation of gait training as reported by the respondents in the acute stage of recovery.
Ninety-three percent of the respondents (n = 233) indicated the need for pre-gait training programs prior to the initiation of gait training for stroke survivors. Basic abilities such as sit to stand transfers (n = 195; 83.7%) and standing balance (n = 225; 81.1%) were most commonly trained by therapists before ambulation, while strength training (n = 146; 62.7%) and stepping activities (n = 157; 67.4%) were reported only in lesser frequencies. The various pre-gait training techniques deemed necessary by the respondents are summarized in Fig. 2.

Pre-gait activities as reported by the respondents.
Nearly half of the respondent physiotherapists (n = 137; 55.6%) provided gait training for 15–30 minutes, followed by one-fourth (n = 62; 25%) providing sessions lesser than 15 minutes. These sessions were most commonly provided once daily (n = 110; 44%) or two to five days/week. (n = 70; 28%)
Gait training strategies
Therapist assisted walking (n = 206; 82.4%) and parallel bar training (n = 169; 67.6%) were the most commonly used gait training strategies (Fig. 3) with the main focus being improvement in quality of gait (n = 213; 85.2%). Distance or number of steps walked (n = 113; 45.4%) and quantitative parameters such as step length and step width (n = 92; 37%) were reported in much lesser frequencies. Other strategies used by the respondents for gait training are summarized in Fig. 3.

Modes of gait training as reported by the respondents. BWSST- Body Weight Support Treadmill Training; FES- Functional Electrical Stimulation; EMG - Electromyography; TMS- Transcranial Magnetic Stimulation; TDCS - Transcranial Direct Current Stimulation.
The majority of the respondent physiotherapists agreed with the use of lower limb assistive aids (n = 190; 76%) to facilitate ambulation. One hundred and thirty-six respondents (55%) reported that they trained their patients in community environments and the frequency of training was either once a week (n = 63; 25.2%) or once in two weeks (n = 57; 22.8%). Forty-one respondents (16.4%) reported that they provide community gait training more than once a week.
Use of outcome measures
Observational Gait Analysis (OGA) was the most commonly used outcome by the respondent physiotherapists (n = 222; 89%). Timed Up and Go (TUG) test (n = 96; 38.4%), Six Minute Walk Test (6MWT) (n = 65; 26%), Gait Velocity (n = 51; 20.4%) and 10-meter Walk Test (10mWT) (n = 36; 14.5%) were also reported to be used as outcome measures. Majority of the respondents (n = 123; 49.2%) measured improvement in gait post-intervention and only 45 respondents (18%) measured improvements at the end of each session.
Barriers to gait training
The barriers most commonly reported by the respondent physiotherapists are summarized in Fig. 4. Significant barriers to gait training reported by respondents included fear of falling (n = 198; 79.2 %), mental or physical fatigue (n = 144; 59.2 %) and lack of patient motivation. (n = 115; 46%).

Barriers to gait training as reported by the respondents.
Majority of the respondents (n = 205; 82%) reported that they did not follow any guideline as part of their routine clinical practice for gait training. Of the respondents who reported following guidelines (n = 45; 18%), only nine quoted existing Clinical Practice Guidelines (CPGs) such as American Heart association, Australian and NICE stroke rehabilitation guidelines. However, 96% of the respondent physiotherapists felt the need for development of clinical guidelines for gait rehabilitation specific to India.
Discussion
With the recognition of stroke as a leading public health problem across LMICs, efforts are in progress to devise and implement programmes for prevention, improved management and rehabilitation of stroke survivors. Although the National Programme for Prevention and Control of Cancer, Diabetes, cardiovascular diseases and Stroke (NPCDCS) in India is a first step towards improving survival rates (Pandian & Sudhan, 2013), very little emphasis is given to rehabilitation. This first national survey has brought to the fore the gaps in current practices and the challenges faced in gait rehabilitation after stroke highlighting the need for developing and implementing evidence based guidelines.
The results from this study demonstrated that time to initiation of gait training is inconsistent across physiotherapy practice. While it is encouraging to note that majority of the respondent physiotherapists reported initiating gait training early, seven percent reported initiation of gait training within 48 hours after stroke. It is well established that very early mobilisation (within 24 hours) reduces the odds of favourable outcomes (Bernhardt et al., 2015) and therefore, practice recommendations in the future may need to specifically address the time of initiation of gait training after stroke. At the other extreme, 16% reported that they initiate gait training after 15 days. One of the factors delaying the time to initiation of gait training may be assumption for training prerequisite skills before commencing ambulation.
Standing balance was considered to be the most important pre-requisite by a majority of respondents and was consequently trained along with transfer to standing prior to gait training. Although beneficial in reducing ataxia and risk of fall in stroke survivors, standing balance is not a direct indicator of improved gait (Winstein et al., 2016). A more important determinant of gait is motor strength of the paretic lower extremity (Winstein et al., 2016; Eng & Tang, 2007). Considering that strength training among stroke survivors may be dependent on the level of functioning of the paretic lower extremity, it may be important to recommend preferable pre-gait training modes when gait training is not possible.
Mounting evidence recommends repetitive task-specific gait training that occurs at high intensity to be effective. Intensity of training has earlier been defined as both frequency of repetitions and practice time (Kwakkel, 2006). A large number of repetitions (Winstein et al., 2016) and a minimum of 45 minutes of practice five days a week (Dworzynski et al., 2013) is recommended for optimal walking recovery. Contrarily, majority of the respondents to our survey were providing low intensity gait training.
Similar low intensity rehabilitation practices have been observed in HICs of the world (Wellwood et al., 2009; Otterman et al., 2012). Possible reasons for an inability to provide high intensity gait training in a low resource setting are the lack of time within a session or lack of personnel assistance. In support of this, our survey showed that therapist assisted walking training and parallel bar training were used much more as compared to BWSTT and robotics. In the absence of adequate personnel support, evidence suggests that technology may be critical in providing high intensity task-oriented training with a large number of repetitions (Otterman et al., 2012; Teasell et al., 2020).
Majority of the respondents (76%) agreed with the use of lower limb orthosis for improving gait. Recommendations suggest that Ankle Foot Orthosis (AFO) may be used as an effective compensation for remedial impairments (Winstein et al., 2016). However, it is unclear if the respondents use the orthosis during the recovery period or as a remedial approach. Upon further enquiry, Posterior leaf spring AFO and Dynamic AFO were reported to be the most frequently used assistive aids, followed by walkers and hemi-rollators. Most commonly cited reasons by respondents for non-use of orthotics included possibility of dependency, inhibition of muscle activity, and alterations in gait patterns. Additionally, lack of availability and affordability of these aids were also stated as reasons, underpinning the common resource issues in most LMICs.
Community ambulation requires “adaptability” which is the ability of the individual to adjust to everyday task and environment demands and hence, requires additional training (Balasubramanian et al., 2014). In this study, it was unclear what type of gait training strategies respondent physiotherapists were using for community ambulation training and therefore may be an ambiguous finding.
Irrespective of the gait training strategy adopted, majority of the respondents focused only on the quality of gait i.e., improving gait pattern. However, contemporary evidence recommends high intensity training irrespective of the pattern of gait, pointing towards a more quantitative approach during gait training (Winstein et al., 2016). Quantitative gait measurement such as 10MWT and TUG is warranted post stroke (Sullivan et al., 2013). However our survey results showed that physiotherapists were not optimally utilizing objective gait measures. OGA is being used more commonly, which may be affected by subjective bias, thus influencing the gait training.
The most frequently reported barriers by respondents in this study were patient specific factors, which may be attributed to a lack of awareness about rehabilitation practices among stroke survivors in India (Mahak et al., 2018). Other identified barriers such as lack of time, staffing, space/infrastructure and lack of interdisciplinary care are known to be common problems in both HICs and LMICs (Bayley et al., 2012).
Lastly, gait training practices did not differ based on the practice setting and 86% of the respondent physiotherapists did not follow any guidelines during their clinical practice. Although not superseding patient preferences and clinical expertise, CPGs form an important component of Evidence-Based Medicine (Djulbegovic & Guyatt, 2017). Observational studies in HICs show approximately 70% adherence to CPGs (M.S et al., 2018). This indicates the need for promoting strict documentation practices in India which can further lead to audit based studies for CPG adherence analysis. Nonetheless, it is necessary to keep in mind the cultural, geographical and economic differences due to which a simple translation of guidelines from other HICs may not be suitable to meet the specific goals of Indian stroke survivors. In support of this, 94% of the respondent physiotherapists in this study expressed the need for rehabilitation guidelines specific to India. Most guidelines for gait rehabilitation primarily stem from the HICs and are largely based on national organizational needs and regional health-care systems (Platz, 2019).
Stroke rehabilitation requires multidisciplinary coordinated efforts and hence, we appeal to all relevant stakeholders to commence work towards this, at the earliest.
Limitations
Our study is not without limitations. The use of snowball sampling method was initially selected to expand the reach of the survey, however this led to an unequal representation from different zones of India limiting the generalizability of the results. Another limitation of this approach is that the source sample remains unavailable and hence, we were unable to calculate a response rate to the survey. Future research might usefully focus on incorporating more refined sampling methods.
Conclusion
In comparison with the current evidence-base, the practices of Indian physiotherapists point towards a general lack of evidence based practice. These findings may be reflective of practice not only in India, but other LMICs as well. Hence, there is an urgent need for country specific evidence-based gait rehabilitation guidelines that are relevant to low-resource settings.
Footnotes
Acknowledgments
The authors would like to thank the experts consulted for content validation of the questionnaire and the physiotherapists who participated in this study.
Conflict of interest
The authors report no conflicts of interest.
