Introduction

Diabetes mellitus is a complex group of metabolic disorders characterised by high blood glucose levels due to an absolute or relative deficit in insulin production or action [1]. Globally, the prevalence of diabetes has increased significantly over the past three decades, with an estimated 537 million adults (10% of the global population) living with diabetes in 2021, and a continuous rise in diabetes prevalence expected [2]. Undiagnosed, untreated, or poorly controlled, diabetes can result in chronic microvascular (peripheral neuropathy, diabetic nephropathy, retinopathy) and macrovascular (cardiovascular, cerebrovascular and peripheral vascular) complications [3], all of which are associated with increased morbidity and mortality [4].

Diabetes-related foot disease (DFD) is recognised as a leading cause of hospital admissions, amputations, disability, and health care costs globally [5–8]. DFD is defined by the International Working Group on the Diabetic Foot as ulceration, infection, or destruction of foot tissues in people with diabetes, typically accompanied by neuropathy and/or peripheral artery disease [9]. In New South Wales (NSW), Australia, the treatment of DFD accounted for 35,000 hospitalisations and an inpatient cost of nearly $330 million in 2014–15 [10]. Notably, approximately 30,500 of these hospitalisations and 80% of the cost were dedicated to the treatment of infections and ulcers [10]. Despite intensive treatment, such as wound care, off-loading, revascularisation, and glycaemic control, diabetes-related foot ulcer (DFU) recurrence is common, with 40% of ulcers reoccur within 1 year following wound closure and 60% within 3 years [5]. DFUs are also very vulnerable to infection, with 40% of ulcers developing an infection [11] and around 20% of moderate to severe diabetic foot infections resulting in a limb amputation [5]. Given the significant burden of DFD in NSW, analysing factors associated with readmission is not only clinically beneficial but may help reduce the increasing costs to the health system.

A recent systematic review and meta-analysis of risk factors for DFD readmission to hospital showed 30-day readmission risk was associated with female gender, peripheral neuropathy, lack of private health insurance, and coronary artery disease [12]. Whilst these risk factors may not be modifiable, exploring patients’ views and experiences of DFD-related care may help identify factors associated with unplanned and potentially preventable readmissions. Increasingly, attention has focused on the association between patient-reported experiences of care and readmissions [13]. There have been several studies which have reported that patient experience scores were more predictive of hospital readmission than clinical performance measures, such as the health status of patients [14–18]. In a recent Australian study among adults with COPD or CHF, patients who offered the most negative ratings for overall care, communication, organisation, and their preparedness for discharge were at a 1.5 to more than 2 times higher risk of readmission within 90-days [19]. This finding is supported by a larger Canadian study of unplanned all-cause readmissions between 43 and 365 days post-discharge where patients who self-reported that they were not involved in care decisions or hadn’t received written documentation post-discharge were 1.3- and 1.2-times more likely to be readmitted, respectively [20]. Such findings support that delivering value-based healthcare which improves overall patient experiences may progressively reduce the risk of adults with chronic conditions, such as DFD, returning to acute care.

Therefore, the aim of this study was to determine the association between patient experience of DFD-related care and unplanned hospital readmission within 90-days through the linkage of de-identified survey and hospital administrative data. We hypothesised that there would be a significant association between the quality of patient experience during their index admission for DFD care and the likelihood of unplanned readmissions within a 90-day period.

Methods

Patient experience measures

Patient experience questions were prioritised based on key themes known to be important to patients, accreditation, and clinical outcomes, specifically compassion, respect, kindness, clear communication, involvement in decision making, and preparedness for discharge [29–31]. Survey questions were selected based on previous studies of patient experiences associated with readmissions to NSW hospitals [19, 32]. Survey questions assessed are included in Table 1.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Diabetes-related foot disease patient reported experiences and outcomes assessed for association with readmission to hospital within 90-days, by readmission type, 2019–2022.

https://doi.org/10.1371/journal.pone.0314895.t001

Results

A total of 4,247 DFD patients were initially included in the study. Approximately 93% of the 2019–2022 survey data were linked with a corresponding acute DFD index hospitalisation, forming an initial cohort of 3,959 patients. After applying the exclusion criteria, 3,173 (80%) patients were included in the final analyses (S2 Appendix).

Of these, 297 (9%) had an unplanned emergency DFD-related readmission within 90-days of discharge, while 520 (16%) had an all-cause unplanned emergency readmission. Response rates and consent to data linkage varied annually with 30% of adults with DFD completing the survey and 82% consenting to data linkage in 2022 (34%, 85% in 2021, 36%, 79% in 2020 and 40%, 75% in 2019) [23, 25–27].

The respondents to the survey were compared with the general population admitted to NSW public hospitals with DFD in terms of age and sex. It was found that survey respondents were more likely to be male (65% vs 61%) and older, with a mean age of 74.0 years (SD = 10.7) compared to 67 years (SD = 18.7) for the general population. However, since the primary object of the study was to explore associations while adjusting for risk factors, including those typically used for sampling weights, the samples were not weighted.

The socio-demographic and clinical profiles of respondents are provided in Table 2. DFD-related readmissions were associated with Charlson comorbidity score, time of survey, and smoking. In comparison, all cause readmissions within 90-days were associated with Charlson comorbidity score, time of survey, age group, marital status, smoking and obesity.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Demographic and clinical characteristics for DFD patients, 2019–2022.

https://doi.org/10.1371/journal.pone.0314895.t002

In terms of annual admissions to hospital, 761 adult DFD respondents in 2022 (i.e., most recent year) had a total of 2311 acute periods of care (i.e., admissions, excluding transfers), with nearly 1 in 2 (49%) admissions principally for DFD. Throughout the year, approximately 2 in 5 had one acute admission (41%), 1 in 4 had two acute admissions (28%), and 1 in 3 had three or more acute admissions (31%). Similar patterns of annual hospital admissions were observed for other years (i.e., 2019–2021).

Overall ratings of care among adults with DFD were found to be associated with DFD-related readmissions within a 90-day period (Table 3). Respondents who would express a critical view of their hospital experience when discussing it with friends and family had an 85% (HR 1.85, CI: 1.21–2.82) higher DFD-related readmission risk than those who would speak highly of their hospital experience. Respondents who reported that their overall care as ‘poor or very poor’ (HR 2.14, CI: 1.22–3.76) or ‘neither good nor poor’ (HR 1.72, CI: 1.10–2.67) had a higher DFD-related readmission risk compared to those who reported the care provided was ‘very good’. Those reporting that the care and treatment they received in the hospital did not help at all had almost twice the DFD-related readmission risk (HR 1.96, CI: 1.21–3.17) compared to those who reported that it ’definitely helped’.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Risk-adjusted statistically significant patient experiences associated with 90-day DFD-related readmissions for adults with DFD, 2019–2022.

https://doi.org/10.1371/journal.pone.0314895.t003

Ineffective communication was identified as a significant factor associated with higher risk-adjusted DFD-related readmissions within 90-days. Specifically, respondents who reported that health professionals did not explain things in a way they could understand had more than twice the risk of readmission, compared to those who stated that health professionals ’always’ explained things clearly (HR 2.43, CI: 1.47–4.00). While DFD-related readmission risk was 71% (HR 1.71, CI: 1.33–2.20) higher among respondents who reported that health professionals ’sometimes’ explained things in a way they could understand. Respondents who reported that they did not receive enough information about their condition or treatment had a 51% (HR 1.51, CI: 1.14–2.00) higher DFD-related readmission risk compared to those who were given the ‘right amount’. The risk of DFD-related readmission was also higher among respondents who felt they did not receive enough home care management information, with a 83% (HR 1.83, CI: 1.24–2.70) higher readmission risk compared to those who were ‘completely’ informed.

Timely and coordinated care, along with active involvement in decision making, was also found to be associated with risk-adjusted readmissions within a 90-day period. Respondents who perceived the care they received in the hospital as ‘not well’ organised were at more than twice the risk of DFD-related readmission compared to those who reported that care was ‘very well’ organised (HR 2.24, CI: 1.45–3.47). A higher DFD-related readmission risk was also found among respondents who perceived that the care received was ‘fairly well’ organised (HR 1.37, CI: 1.04–1.72). Respondents who reported how well health professionals worked together as ‘neither good nor poor’ had a 68% (HR 1.68, CI: 1.11–2.55) higher DFD-related readmission risk compared to those who reported ‘very good’. Notably, respondents who stated that they were ‘not well’ enough to leave the hospital had almost twice the risk of DFD-related readmission compared to those who indicated they were ‘well’ enough (HR 1.93, CI: 1.41–2.64). Conversely, the absence of a document summarizing the care received in the hospital was associated with a reduced risk of readmission (HR 0.45, CI: 0.25–0.80).

Similar patterns were observed for all-cause readmissions within 90-days, albeit with generally smaller effect sizes and confidence intervals (Table 4). For instance, critical views of hospital experience were associated with an 82% increase in all-cause readmissions (HR 1.82, CI: 1.32–2.52) compared to an 85% increase in DFD-related readmissions. Ineffective communication showed a similar trend, with failure to understand explanations associated with an HR of 1.91 (CI: 1.26–2.90) for all-cause readmissions compared to an HR of 2.43 for DFD-related readmissions. The perception of poorly organised care was associated with an HR of 1.96 (CI: 1.38–2.78) for all-cause readmissions, slightly lower than the HR of 2.24 for DFD-related readmissions.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 4. Risk-adjusted statistically significant patient experiences associated with 90-day all cause readmissions for adults with DFD, 2019–2022.

https://doi.org/10.1371/journal.pone.0314895.t004

Discussion

Reducing the burden of diabetes-related foot disease is a major public health priority in NSW. Even with appropriate care, persistent DFD is associated with serious complications, such as infection, lower limb amputation, and even death [39–42]. To our knowledge, this is the first comprehensive state-wide analysis of patient self-reported experiences of care related to DFD, linking administrative hospital data to quantify the likelihood of unplanned DFD-related readmissions within 90-days post-discharge. We found that a poorer experience of inpatient care during the index admission is associated with a higher risk of DFD-related and all-cause readmission within 90-days. Our analysis showed that the quality of communication from health professionals and coordination of care were critical factors influencing this risk.

Whilst adults admitted to public hospitals in NSW for DFD related care mostly reported that health professionals always explained things in a way they could understand, many offered neutral or negative ratings. Respondents who could not understand explanations offered by health professionals were at more than twice the risk of DFD-related readmission. Moreover, more than 1 in 7 respondents also felt they were not given enough information regarding their care and treatment during their hospital stay or how to manage care at home, both of which were associated with increased readmission risk. Poor communication is one of the most cited underlying causes of patient complaints about the healthcare system [43]. In a qualitative meta-synthesis of perceptions and experiences of DFU-specific care, it was found that health professional to patient communication regarding their ulcer and foot care was generally considered to be poor [44]. One hypothesis for these findings is that the aetiology of DFU is complex and often difficult to communicate to patients [44], which can be further complicated by varying levels of health literacy [45]. Communication that is simple, clear, and individualised is integral to improving patient understanding of their condition and promoting engagement in effective self-care to reduce recurrence [45]. The development of DFU-specific communication models, such as the “Fragile Feet and Trivial Trauma” model [45], may offer potential for improving patient satisfaction with health professional communication during hospital DFD-related care. More generally, satisfaction with patient-health professional communication has been shown to be associated with better understanding of diabetes [46], self-care [47], and related health outcomes [48]. It is possible that by improving the communication of DFU aetiology patients may become more engaged in self-care, resulting in lower DFD readmission rates. However, at present, there is a lack of randomised controlled trials which have investigated whether DFU communication models improve patient outcomes and contribute to lower DFD readmission rates. Future research is required to assess the efficacy of such communication models in the Australian healthcare system.

There is strong evidence that the successful implementation of standardised care protocols and multidisciplinary teamwork can improve DFD-related outcomes. Recent systematic reviews and meta-analyses have demonstrated the effectiveness of these approaches [49–51]. Meta-analyses have shown that implementation of multidisciplinary care teams resulted in a 39–56% reduction in major amputation rates among adults with diabetes [49]. Organisational arrangements including multidisciplinary teams and care pathways have been associated with up to a 48% reduction in the risk of lower extremity amputations [50]. Structured diabetic foot care services have been found to significantly reduce major lower extremity amputation rates [51]. The findings of our study also support the essential role of well-organised and coordinated care in reducing the risk of hospital readmissions for patients with DFD. While most respondents reported that health professionals worked well together and that their care was well-organised–a possible reflection of the implementation of the Standards for High-Risk Foot Services (HRFS) [52] and the establishment of 29 hospital based HRFS in NSW [53]–those who experienced less organised care had almost double the risk of readmission. Whilst it is beyond the scope of this study to evaluate HRFS implementation across NSW, disparities suggest that the implementation may not be uniform across all patient experiences. Consequently, there may be opportunities to review the implementation of HRFS in-hospital care across NSW to improve patient experiences and minimise hospital readmissions in this vulnerable population.

Interestingly, patient-doctor and patient-nurse ratings revealed that those who reported their care as “neither good nor poor” had a higher risk of readmission. Whilst this is likely the result of very few patients endorsing that their care was “poor”, this finding still points to a potential area for improvement. It is important that patients feel satisfied with the care provided by their healthcare team, as this could influence their post-discharge behaviours, including adherence to treatments, thus potentially impacting readmission rates [54].

A somewhat counter-intuitive finding was that the absence of a document summarising the care received in the hospital was associated with a decreased risk of readmission. Previous research among the general inpatient population reported that patients who did not receive written information post discharge were at increased risk of readmission [20]. A possible explanation for our finding could be that DFD patients who did not receive such documentation may have received more verbal explanations, or perhaps more personalized and detailed care that was not captured in a standardized document. However, further research is needed to test this hypothesis.

It is also important to note that the risk factors identified in this study for DFD-related readmissions were comparable to those for all-cause readmissions, but the effect sizes, as well as confidence intervals, were generally larger for DFD-related readmissions. It is possible that this readmission group represents patients who presented with more complex DFD or were more susceptible to DFD-related complications due predisposition to unmeasured risk factors.

The current study has several limitations which must be considered. While administrative hospital admission data are acknowledged to under-report outcomes, including in-patients receiving care for DFD [55], it is important to acknowledge that in NSW, the quality of coded administrative data is monitored, validated, and subject to audit. This process enhances the reliability of the data used for analysis and consequently the validity of the study results. There is potential for response bias, as survey respondents were more likely to be older and male compared to the overall DFD admitted population. Additionally, since surveys were conducted three months after the hospitalisation experience, recall bias may have influenced responses. However, it is important to note that our findings are consistent with results from other studies conducted on patients with chronic conditions such as CHF and COPD [19], suggesting that any potential bias may not have affected the validity of our results. Despite our cohort including over 3000 adults, the proportion of respondents who expressed the most unfavourable feedback was relatively small, especially when considering DFD-related readmissions. In some instances, this limited the statistical power of the study to detect differences between the most and least favourable responses (e.g., 2% of patients rating the nurses or doctors who treated them as ‘poor’). Finally, whilst the multivariable shared-frailty Cox regression analysis considered important health and socio-economic determinants, there were several factors not adjusted for in our analysis, such as ulcer severity and post-discharge self-care compliance, that could influence readmission rates.

The current study provides insight into the quality of in-hospital patient experiences among persons admitted with DFD in NSW. For most patient experience indicators, there was a significant ranked relationship between patient reported measures and readmission, where the degree to which an experience measure is suboptimal increases the risk of DFD-patients returning to hospital for both DFD-related and all-cause readmissions within 90-days. Together, these findings can provide useful guidance for healthcare providers, hospital administrators, and policymakers to reduce DFD readmission rates. Potential interventions to improve the experiences of patients admitted for DFD should focus on enhancing health professional to patient communications, assessing patient readiness for discharge, and improving overall coordination of care.

Supporting information

Acknowledgments

The authors wish to thank the patients who completed the survey and provided their consent for data linkage. We also acknowledge the Bureau of Health Information who administered and coordinated the survey.