Healthcare experience among patients with type 2 diabetes: A cross‐sectional survey using the IEXPAC tool

Abstract Aim To assess the experience with health care among patients with type 2 diabetes (T2DM) and to evaluate patients’ demographic variables and healthcare‐related characteristics which may affect their experience. Methods A cross‐sectional survey was delivered to T2DM adults. Patient experiences were assessed with the ‘Instrument for Evaluation of the Experience of Chronic Patients’ (IEXPAC) questionnaire, a validated 12‐item survey, which describes patient experience within the last 6 months (items 1–11) and hospitalization in the last 3 years (item 12), with possible scores ranging from 0 (worst) to 10 (best experience). Results A total of 451 T2DM patients responded to the survey (response rate 72.3%; mean age 69.5 ± 10.1 years, 67.8% men). The mean overall IEXPAC score was 5.92 ± 1.80. Mean scores were higher for productive interactions (7.92 ± 2.15) and self‐management (7.08 ± 2.27) than for new relational model (1.72 ± 2.01). Only 32.8% of patients who had been hospitalized in the past 3 years reported having received a follow‐up call or visit after discharge. Multivariate analyses identified that regular follow‐up by the same physician and follow‐up by a nurse were associated with a better patient experience. Continuity of healthcare score was higher only in those patients requiring help from others. Conclusions The areas of T2DM care which may need to be addressed to ensure better patient experience are use of the Internet, new technologies and social resources for patient information and interaction with healthcare professionals, closer follow‐up after hospitalization, and a comprehensive multidisciplinary approach with regular follow‐up by the same physician and a nurse.


| INTRODUC TI ON
Patient experience with health care is a key component for the provision of a patient-centred healthcare model, as both clinical effectiveness and safety are correlated positively with patient experience. 1 In patients with chronic conditions, a more positive patient experience is associated with improved care quality with the interaction of patients with healthcare professionals 2 particularly general practitioners, being important for patient well-being. 3 Effective chronic illness management also depends on multidisciplinary care teams, including nurses and pharmacists, with clinical experience. 4 Diabetes is a major public health problem that is approaching epidemic proportions globally. Annually, >3 million (5.2%) deaths are attributable to diabetes making it a leading cause of death worldwide. 5 Diabetes is also associated with poor quality of life (QoL) and disability. 6,7 Despite being largely preventable, type 2 diabetes mellitus (T2DM) accounts for around 90% of diabetes cases, affecting 294.3 million people in 2017 and rising globally, and predicted to affect 394.2 million people by 2045. 8 In Spain, the prevalence of T2DM is estimated to be 13.8% (with almost half of these cases being undiagnosed DM) 9  The annual cost per diabetic patient averages close to €1,660 for direct costs and €916 for productivity losses. 10 As a result, ascertaining healthcare experience in diabetic populations is important.
Previously, we have reported the outcomes of a survey to assess the experience of a diverse group of patients with four different chronic conditions (T2DM, human immunodeficiency virus infection, inflammatory bowel disease and rheumatic diseases) with health care using the Instrument to Evaluate the EXperience of PAtients with Chronic diseases (IEXPAC). 11 IEXPAC is a validated questionnaire, developed in Spain, with several advantages over other available questionnaires (namely, focusing on the overall interaction of patients with the healthcare system and not with specific professionals, and the inclusion of a broader notion of integrated care, including social care, patients' self-management, new technological interventions and patients' interactions with other patients). 12 Herein, we focus on the cohort of patients with T2DM, with the objective of describing patient perception of health care, to identify the main areas for improvement and to assess potential variables affecting patient experience including demographic variables and healthcare-related characteristics. and main outcomes for the overall population have been described previously. 11 The main objectives of the current study were to describe patients' experience with health care, to identify the main areas for improvement and to assess potential variables affecting patient experience among a population with T2DM. The study was reviewed and approved by the Clinical Investigation Ethics Committee of the Gregorio Marañón Hospital, Madrid, Spain. Patients provided written informed consent before entering the study.

| Survey instrument
The survey mainly included the IEXPAC questionnaire plus additional multiple-choice questions in order to provide information on patient demographics, healthcare and treatment-related characteristics. The survey was drafted by expert physicians and reviewed and finally endorsed by the Spanish Diabetes Federation (FEDE) among other patients' associations.
Details of the IEXPAC questionnaire have been published. 12 The questionnaire was in Spanish. Briefly, IEXPAC is a self-administered 12-item questionnaire with patient responses made using a 5-point Likert scale: always (score 10), mostly (7.5), sometimes (5), seldom (2.5) or never (0). An overall score is given by summing the scores of items 1-11, which describe patient experience within the last 6 months, ranging from 0 (worst experience) to 10 (best experience).
Item 12, describing continuity of health care after hospitalization in the last 3 years, is reported separately.
Three factors are derived from IEXPAC items 1-11. Factor 1 (productive interactions) refers to the content and characteristics of interactions between patients and healthcare professionals and is the mean score of items 1, 2, 5 and 9. Factor 2 (new relational model) refers to new forms of patient interaction with the healthcare system through the Internet or with peers and is the mean score of items 3, 7 and 11. Factor 3 (patient self-management) captures the ability of individuals to manage their own care and improve their well-being based on healthcare professional-mediated interventions and is the mean score of items 4, 6, 8 and 10.

| Other variables measured
Beliefs about medication were determined using the Beliefs About Medicines Questionnaire (BMQ). 13,14 The BMQ evaluates an individual's opinion about medicines in general (abuse and damage) and about specific drugs for his/her disease (need and concern). This 10item questionnaire covers two domains-Necessity and Concerns, with five statements per domain. Patients respond on a 5-point Likert scale, ranging from strongly disagree (scored 1) to strongly agree (scored 5). Scores are summed for each individual item, and total scores for the Necessity and Concerns domains (each ranging from 5 to 25) are also calculated. Higher scores in the Necessity and Concerns scales indicate stronger beliefs in the necessity for the prescribed medication and more concern about taking the medication, respectively. The overall BMQ score (presented in this study) is calculated as the difference between the Necessity Scale and Concerns Scale scores, with a possible range of −20 to 20. Higher overall scores indicate stronger beliefs.
A visual analogue scale (VAS) was used to determine health status. It is a psychometric response scale across a continuum of values that ranges from 0 (worst health status) to 100 (best health status). 15 The Barthel scale was used to measure performance in activities of daily living. The Barthel disability index ranges from 0 to 100: 0-20: 'total' dependency; 21-60: 'severe' dependency; 61-90: 'moderate' dependency; 91-99: 'slight' dependency; 100: independent. 16

| Statistical analysis
This was an exploratory study with no formal hypothesis nor prespecified sample size. A conservative approach was adopted to calculate sample size based on a qualitative variable with an expected prevalence of 50%, 95% confidence interval and with 6% precision giving an initial calculated sample size of 267 patients, plus 15% of variables completed incorrectly (an additional 47 patients; total, 314 patients), and accounting for an expected response rate of approximately 50%, as found in other surveys handed to patients by clinical teams 17,18 to give a total sample size of 628 patients. Then, 48 primary care centres were selected to be representative of the Spanish population. As a result, it was calculated that at least 13 patients for each primary care centre were required to complete the sample size.
Descriptive information is displayed as mean and standard deviation for quantitative variables, and frequencies or percentages for qualitative variables. The results of the IEXPAC questionnaire were calculated as overall mean and standard deviation score and mean and standard deviation scores for Factors 1-3. The distribution of responses to each item was also displayed, as well as the mean score for each item.

| Description of the sample
In the overall study population, 1 Only 66.8% of patients were generally followed-up by the same physician and most patients (82.7%) received additional follow-up by a nurse. Support from non-healthcare workers (relatives, friends or caregivers) for patients' health care was received by 45.2% of patients, and 56.1% of patients had been hospitalized at least once in the past 3 years. With regard to medications, patients were taking a mean of 6.5 ± 3.2 different pills daily (ie if someone took a medicine twice a day, the number of different medicines would be two), and 23.9% were receiving injectable medications.

| IEXPAC responses and experience scores
The mean overall IEXPAC score was 5.92 ± 1.80 ( Figure 1 and Supplementary Table 1). Mean scores were higher for Factor 1 (Productive interactions score: 7.92 ± 2.15) and Factor 3 (Selfmanagement score: 7.08 ± 2.27) than for Factor 2 (New relational model score: 1.72 ± 2.01). The proportion of patients who responded 'always' or 'mostly' to items related to the Productive interactions score (items 1, 2, 5 and 9) was >70%. By contrast, regarding the New relational model score (items 3, 7 and 11), the majority of patients responded 'seldom' or 'never' to the 3 items. Regarding the Patient self-management score (items 4, 6, 8 and 10), except for being informed on health and social resources, >70% of patients responded 'always' or 'mostly'. Only 32.8% of patients who had been hospitalized in the past 3 years reported having received a follow-up call or visit after discharge (Table 2). Continuity of healthcare score was higher only in those patients requiring help from others, with no significant differences in the other healthcare-related variables.

| Multivariate analysis
Results of multiple linear regression analyses are shown in Table 3.
Factors associated with higher overall IEXPAC score (therefore indicating a better experience) were being followed regularly by the same physician (p < 0.001) and receiving additional follow-up by a nurse (p = 0.046). These variables were also associated with better self-management scores (p < 0.001 and p = 0.003, respectively).
Follow-up by the same physician was also associated with higher productive interactions scores (p < 0.001) and higher item 12 scores (continuity of health care after hospitalization) (p = 0.01). Regarding the new relational model score, it was lower with increasing patients' age (p = 0.001).

| BMQ and Health Status VAS scores
Mean overall scores for the BMQ and Health Status VAS were 6.38 ± 5.87 and 66.96 ± 17.12, respectively (both slightly above the average for each respective scale).

| DISCUSS ION
Improving healthcare experience among patients with chronic conditions, such as T2DM, should be considered as a therapeutic goal, as it is associated with higher clinical effectiveness and safety. 1 Using the IEXPAC tool 12 this study described the experience of T2DM patients with the health care received, identifying positive aspects of patient experience but also areas for improvement. As outlined previously, although there are other questionnaires that assess the experience of chronic patients, the IEXPAC scale may TA B L E 1 Patient demographics and healthcare-related characteristics of patients who completed the survey (n = 451).   trust. 28 In this context, regular follow-up by the same physician is desirable to improve patient experience. On the other hand, as our study showed, nurses play a key role in the care of diabetic patients. 29 These F I G U R E 1 Mean scores for each IEXPAC item. The limitations of this study have been described previously. 11 Since this was an anonymous survey, the profiles of patients who did not return the surveys were not known. Additionally, in general, these types of voluntary surveys tend to generate responses more frequently from motivated patients or patients who are particularly TA B L E 2 Patient responses (%) and mean scores for each IEXPAC item.

| Informed consent
As agreed by the Clinical Investigation Ethics Committee, the voluntary return of completed questionnaires was taken as implied consent to participate in the study. No clinical data were collected in this study.

| What's new?
• Improving healthcare experience among patients with chronic conditions (eg type 2 diabetes [T2DM]) may be considered a therapeutic goal as it is associated with better clinical effectiveness and safety.
• Using the IEXPAC tool in 451 patients with T2DM, we identified positive aspects of patient experience, regarding productive interactions, and self-management score, but not for the new relational model. Being followed regularly by the same physician and receiving additional follow-up by a nurse were associated with a better experience.
• Improvement areas of T2DM care to ensure better patient experience may include the use of the Internet, new technologies and social resources for patient information and interaction with healthcare professionals, closer follow-up after hospitalization and a comprehensive multidisciplinary approach with regular follow-up by the same physician and a nurse.

ACK N OWLED G EM ENTS
The following patients' association participated in the design of the overall study and endorsed it: Spanish Federation of Patients AGG and GF planned the study implementation.
KFC planned and coordinated the study implementation.
SAM, AHP, DCS, JLAJ, EAG and CVA substantially helped to collect the data.
DOB and KFC interpreted the results.
KFC wrote sections of the initial draft.
DOB, AGG, RLR, GF and KFC provided substantive suggestions for revision and critically reviewed subsequent iterations of the manuscript.
All the authors reviewed and approved final version of the paper.

E TH I C A L A PPROVA L
The study was reviewed and approved by the Clinical Investigation Ethics Committee of the Gregorio Marañón Hospital, Madrid, Spain.

Study documentation included printed instructions and information
for patients regarding the anonymous nature of the survey and aggregated data processing, which ensured that patient identification was not possible.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.