Methods
1. Patients
There are no specific inclusion criteria: all adult (>17 yrs.) diabetes outpatients (Type 1 and Type 2) scheduled for their annual review are eligible. Those unable to read or complete questionnaires on the computer due to language and/or reading problems or blindness will have to be excluded from the study. Clinicians will seek appropriate alternative ways to evaluate and discuss well-being in these individuals. No data are collected from these patients for this project.
The number of patients included in MIND is dependent on frequency of monitoring during the observation period (12 mos.) and the inclusion period (6 mos.). It is estimated that centres will be able to include and monitor 4 to 6 patients on a weekly basis x 40 weeks = 160 to 240 patients at baseline and after 12 months. For all 12 centres the total would be 1920 to 2880 patients.
2. Patient Information
Patients will be informed of the project by letter and invited to participate by their treating physician. In the information it will be made clear that this project concerns implementation of international evidence-based diabetes guidelines, with the aim to protect and encourage well-being in people with diabetes. The primary aim of the monitoring procedure is not to make a ‘psychological diagnosis’, but rather to facilitate recognition and discussion of psychosocial issues in the context of diabetes self-management. Monitoring well-being may assist patients in their efforts to overcome psychosocial problems and barriers, with subsequent positive effects on diabetes control and future health.
Implementing a procedure to monitor well-being as part of diabetes care is according to clinical practice recommendations, not experimental and therefore not subject to ethical approval.
Patients will be asked to provide written informed consent in the month prior to their annual visit, to allow the use of the well-being data on group level (anonymously) for scientific purposes, in accordance with Good Clinical Practice regulations.
Patients may refuse to fill out the questionnaire by computer and will receive care as usual. Based on experience, we anticipate attrition rates of no more than 10%.
3. Monitoring Procedure (Appendix I)
The well-being assessment is scheduled prior to the consultation with the diabetes nurse specialist (DNS) or other Health Care Provider (HCP, in a room or a quiet, designated waiting area (table, chair, PC connected to printer) made available for this purpose. The PC can be a stand-alone (laptop) or a PC connected to the hospital network. Patients are directed by the DNS or an assistant to the computer that is set up (names entered and ready to start the assessment) for direct use. Self-administering the questionnaires will take 7-10 minutes and requires no more than basic computer skills from the patient (reading instruction, using the mouse).
The computerized programme runs on Windows (98 and onwards) and is based on HealthQuest® developed by Tilburg University and adapted for use in diabetes care by the Amsterdam team. The programme has shown to be safe and robust, and is freely available to all participating centres for the purpose of this project.
HbA1c values routinely determined at/around the annual review are entered later in the patients’ pc-files by the diabetes team. The patients’ data/scores are automatically entered in a separate data file that can be imported in SPSS. The files are send digitally periodically by the DNS/HCP, or other local team member on site.
4. HealthQuest® computer assessment (Appendix II)
The monitoring starts with a short self-report questionnaire on socio-demographics and some medical data (type of diabetes, year of diagnosis, treatment regimen, hypoglycaemic episodes, and co-morbidity [pre-coded]). Also, the patient is asked to indicate whether and which life events [pre-coded] s/he has experienced the previous 6 months.
The following domains of well-being are included:
- Emotional well-being is assessed with the widely used World Health Organization-5 Well-Being Index (WHO-5), a generic measure of emotional well-being. This short scale contains five positively worded items pertaining to positive mood (good spirits, relaxation). Vitality (being active and waking up fresh and rested) and general interest (being interested in things). The WHO-5 has good psychometric properties and has shown to be suitable as screener for depression (see www.who-5.org).
- As a measure of illness intrusiveness or suffering, the Pictorial Representation of Illness and Self Measurement-revised (PRISM-R) is administered, an innovative, patient-friendly non-verbal test. Studies so far show satisfactory psychometric properties and excellent clinical utility (e.g. Reimus & Vingerhoets, 2006). PRISM can foster open communication on psychosocial issues between patients and clinicians (Büchi et al., 1998, 2002), which is the main purpose of using PRISM in this context.
- Diabetes-specific emotional distress is measures using the widely used 20-item Problem Areas In Diabetes scale (PAID),with proven good psychometric properties, both in Type 1 and Type 2 diabetes. This scale provides a total score and 4 subscale scores (Welch et al., 1997, 2003; Snoek et al., 2000).
- To invite the patient to help set the agenda and promote active patient participation, a one-item question concludes the assessment, offering the patient the possibility to tick one of more areas of interest to de discussed today (Medication/treatment, Symptoms/complaints [incl. pain, hypo’s], Lifestyle [incl. diet, alcohol, smoking], Mood/stress, Other) that s/he would particularly like to address today.
5. Feedback and discussion of outcomes
After filling out the questionnaire, the patient presses the “Finish” button and the computer instantly generates a report summarizing the results using standardized scores (bars 0-100), along with a representation of the pictorial test (circles), the potentially experienced life events and one or more items that the patient has identified for the agenda. Both the patient and DNS/HCP receive a copy of the print-out.
The results are then discussed with the patient as part of the consultation (annual review) and further actions are considered and agreed.
The discussion of well-being is scheduled for 15 minutes. If necessary, an extra consultation is scheduled or the patient is referred for supportive care/counselling.
It is of great importance that the results of the well-being assessment are fed back to the patient in an empathic, non-judgemental way. The participating teams receive instructions and training on assessment and communication in a 1-day workshop. Materials will be made available to assist teams in interpreting and discussing well-being outcomes. The patients’ (first) experience with the assessment is explored and his/her answer regarding today’s agenda is recognized.
S/he is then invited to comment on the Pictorial test (choice of circles and position). Experience has shown that this is particularly helpful to open up the conversation and have the patient talk about his/her coping with diabetes and related concerns. The DNS/HCP then continues discussing the WHO-5 and PAID-scores, using established cut-off scores for low, average and high. The leading questions are 1) whether the identified problem(s) is (are) related to and/or complicate the diabetes self-management and 2) whether the patient is in need of/receptive to help.
To assist the DNS/HCP, a 1-page instruction sheet is provided, to effectively communicate well-being outcomes in the context of diabetes management. On one side of the page the items of the WHO-5 and PAID are displayed, to assist in discussing specific problem areas; on the other side a simple algorithm is shown for WHO-5 and PAID scores, indicating that < 28 on the WHO-5 and > 40 on the PAID further psychological examination is advised. Depending on the local resources and patient preference, this could mean a referral to the mental health specialist (MHS) of the diabetes team, to a MHS unit within the centre or a MHS in primary care.
The conclusions and actions agreed following the assessment are included in the medical chart and in a separate Excel-sheet. Also, some additional data on HbA1c, complications, and ethnicity will be entered (Appendix III). If warranted, the outcomes are discussed with other health care providers a part of clinical management.
At the second assessment (after 12 months), the procedure is identical, with the report of the 1st assessment available to the patient and DNS/HCP to highlight changes and new issues.
6. Data Management (see also Appendix I)
After each assessment, scores/outcomes are automatically entered in a data file on the computer, to be downloaded periodically per centre by the local study coordinator and transferred to the centre system file. The assessment-files show green bars when incomplete (ready to start), and red for complete data. This will help diabetes team to ensure only complete data files are transported/uploaded in the central data-base.
Data analysis will be performed at the VU University Medical Centre (VUMC), according to pre-agreed plan, aimed to answer the research questions underpinning this project, in accordance with international standards, ensuring data quality and confidentiality. For this purpose centre patient names and registration numbers are re-coded at the VUMC by the research assistant (NZ) with unique numbers attached to each patient by centre (A-L). In agreement with medical ethical regulation securing privacy, the conversion key is only available to the VUMC and is stored apart from the data files.
Centres receive back results from the total data-base and hold ownership over their local data, that may be used for clinical and research purposes. Data can be used for benchmarking.
A general publication plan is agreed upon before the start of the project.
