Main findings
Effective allocation of rehabilitation resources and appropriate distribution of patients across healthcare institutions remain critical challenges for LMICs. To address this, our study investigated the misalignments among patients’ healthcare-seeking behaviour, doctors’ clinical judgement and RTS tool results, using the first-hand data from the field survey in China. Our analysis revealed two key findings: (1) significant discrepancies exist between patients’ preferences, doctors’ decisions and RTS-guided pathways and (2) these gaps are influenced by factors from the institution, doctor and patients’ level characteristics.
Discrepancies among patients’ healthcare-seeking behaviour, doctors’ clinical judgement and RTS tool results
The findings of our study highlighted significant discrepancies between patients’ healthcare-seeking behaviours, doctors’ clinical judgements and the RTS tool results. First, we found a 21% gap between patients’ healthcare-seeking behaviours and doctors’ clinical judgements; several patients’ healthcare-seeking behaviours were not unreasonable. Notably, patients frequently bypassed primary or secondary care facilities—even for minor conditions deemed suitable for lower-tier care by clinicians—and sought treatment directly at tertiary hospitals. This gap could be explained by the higher quality of care at a tertiary hospital, lack of trust in primary healthcare, lower health literacy of patients, etc.40 41 Second, we also found that a more pronounced gap emerged between doctors’ clinical judgements and the results of the RTS tool, with almost half (49.3%) of doctors’ judgements deviating from the results of the RTS tool (the RTS tool results represent the level of rehabilitation experts). This gap could be explained by insufficient capacity and awareness to implement RTSs, even though healthcare quality has improved in recent years.40–43 These gaps underscore the complexities inherent in China’s tiered service system for rehabilitation.
Notably, supplementary analyses in online supplemental table S1 revealed that patients’ health-seeking behaviour aligned more closely with the RTS tool. This underscores the tool’s reliability in reflecting patient health-seeking behaviours and priorities. Given that over 55% of physicians deviated from the RTS recommendations—while patients adhered more closely—this discrepancy highlights a critical gap in clinical decision-making. These findings suggest physicians need to adopt a more patient-centric approach, ensuring that patient concerns, preferences and behaviours are actively integrated into care plans. By bridging this misalignment, healthcare providers can enhance trust, improve adherence and deliver more personalised care, aligning clinical practice more closely with the RTS framework and patient expectations.
Determinants of observed gaps
Our analysis identified multiple factors significantly associated with these gaps. The first gap (patient-doctor misalignment) demonstrated higher prevalence among doctors with advanced education, those who majored in western medicine, older patients and patients presenting with neurological, geriatric or childhood diseases. Conversely, doctors employed in public institutions, those with more medical training and patients with cardiopulmonary diseases showed better alignment. We found mixed results from previous studies. A study in East Africa indicated that rural residence, working mother, unmarried women, media access, the richest, age of child 7–23 months, age of child 24–59 months and family size >10 increase healthcare-seeking behaviour.24 Another study in China showed that patients’ healthcare-seeking behaviour was associated with age, self-rated health status and distance to a medical care facility.20 We found higher levels of ‘patient-doctor misalignment’ in certain diagnostic groups. The increased misalignment observed in neurological diseases (eg, stroke, brain trauma) may stem from several inter-related factors. The progressive nature of neurological disorders frequently necessitates multidisciplinary care and specialised rehabilitation protocols,44 which may not be fully available in primary/secondary facilities. Patients and families may perceive tertiary hospitals as better equipped to manage these conditions, despite the doctor’s clinical judgement suggesting lower-tier care. Furthermore, the ‘time-sensitive’ perception of neurological recovery—particularly post-stroke—often drives families to seek maximal care intensity,45 even when tiered service guidelines recommend step-down rehabilitation. The increased misalignment in geriatric diseases (eg, hypertension, Parkinson’s) may reflect unique challenges in managing age-related multimorbidity. The elders are often present with intertwined physical, cognitive and social vulnerabilities that complicate tiered service triage.46 47 For instance, while hypertension alone might be managed in primary care, coexisting conditions (eg, frailty, polypharmacy) may prompt patients to seek tertiary-level comprehensive geriatric assessment, over-riding clinical judgements about care appropriateness.48 Collectively, our findings underscore how doctors’ qualifications, medical training and specific patient characteristics critically influence the alignment between patient behaviour and doctors’ judgement.
The second gap (doctor-RTS tool misalignment) exhibited distinct patterns, which are more prevalent among primary healthcare doctors, male doctors, those who majored in western or traditional Chinese medicine and older patients. In contrast, doctors in public institutions and those with higher incomes are less likely to encounter this discrepancy. A study in Bangladesh indicated that the risk of violence by patients’ relatives, better management and physicians’ professional and personal social networks are key factors that influence physicians’ clinical decision-making process.32 A study in the USA proved that implementing the intervention to support doctors in making more transparent, consistent, patient-centred and ethically justified decisions would improve doctors’ decision-making.34 These convergent findings highlight how institutional environments, doctors’ demographics and socioeconomic status significantly influence the concordance between clinical judgement and RTS tool recommendations.
To address the variability in alignment rates across the six recommended pathways, we conducted subgroup analyses to identify factors influencing misalignment between physicians’ decisions and the RTS tool (online supplemental table S2). The results revealed distinct patterns: for example, outpatient treatment misalignment was associated with male physicians, certain specialties (eg, Chinese and western medicine) and neurologic diseases, while public institution affiliation and higher income reduced gaps. Conversely, for tertiary hospital admissions, orthopaedic/geriatric conditions increased misalignment, whereas older patient age reduced it. Notably, factors like policy/procedure awareness paradoxically increased misalignment in primary healthcare referrals, suggesting contextual misunderstandings despite familiarity. These findings highlight the need for targeted education—particularly for primary healthcare admission criteria and policy implementation—while affirming physician competence in tertiary care decisions. The subgroup-specific insights will guide future interventions to address gaps in knowledge and workflow alignment.
These findings have important implications for policy and practice. They suggest that targeted interventions to improve communication and education for patients and doctors could help bridge the gaps. For example, training programmes for doctors could focus on enhancing their understanding of the RTS tool and how it integrates with clinical judgement. Additionally, patient education initiatives could be implemented to better inform patients about the RTS system, encouraging them to seek care at the appropriate level of service. Furthermore, our results emphasise the need for continuous professional development and updates on best practices for rehabilitation specialists. Ensuring that all healthcare providers have access to up-to-date training and resources, to minimise discrepancies and improve the overall effectiveness of the RTS system. This study is, however, not without limitations. First, this study used cross-sectional measurements; we could not evaluate some potentially critical psychometric properties, such as test-retest reliability or sensitivity to change. Second, a large percentage of the sample was comprised of patients with orthopaedic disease, which might affect the generalisation of our findings, and it can only represent outpatient rehabilitation in the sampled cities in China.
This post was originally published on https://bmjopen.bmj.com