key: cord-347207-1u4i6qmc authors: Almomani, Huda Y.; Pascual, Carlos Rodriguez; Al-Azzam, Sayer I.; Ahmadi, Keivan title: Randomised controlled trial of pharmacist-led patient counselling in controlling hypoglycaemic attacks in older adults with type 2 diabetes mellitus (rose-adam): A study protocol date: 2020-07-29 journal: Res Social Adm Pharm DOI: 10.1016/j.sapharm.2020.07.012 sha: doc_id: 347207 cord_uid: 1u4i6qmc INTRODUCTION: Hypoglycaemia is one of the most serious adverse effects of diabetes treatment. Older adults are at the highest risk to develop hypoglycaemia. Several studies have established the important positive role of educational interventions on achieving glycaemic control and other clinical outcomes, however, there is still a lack in studies that evaluate the impact of such type of interventions on hypoglycaemia risk in elderly patients with type 2 diabetes. The purpose of this research is to evaluate the effectiveness of pharmacist-led patient counselling on reducing hypoglycaemic attacks in older adults with type 2 diabetes mellitus. METHODS: and analysis: This study is an open-label, parallel controlled randomised trial, which will be conducted in the outpatient clinics at the largest referral hospital in the north of Jordan. Participants who are elderly (age ≥ 65 years), diagnosed with type 2 diabetes mellitus, and taking insulin, sulfonylurea, or any three anti-diabetic medications will be randomly assigned to intervention (SUGAR Handshake) and control (usual care) groups. The SUGAR Handshake participants will have an interactive, individualised, medications-focused counselling session reinforced with a pictogram and a phone call at week six of enrolment. The primary outcome measure is the frequency of total hypoglycaemic events within 12 weeks of follow up. Secondary outcomes include the frequency of asymptomatic, symptomatic, and severe hypoglycaemic events, hypoglycaemia incidence, and time to the first hypoglycaemic attack. We will also conduct a nested qualitative study for process evaluation. ETHICS AND DISSEMINATION: The Human Research Ethics Committee of the University of Lincoln and the Institutional Review Board of King Abdullah University Hospital approved this protocol. The findings of this study will be presented in international conferences and published in a peer-reviewed journal. TRIAL REGISTRATION NUMBER: The study protocol has been registered with ClinicalTrials.gov, NCT04081766. Background 36 Hypoglycaemia is the major limiting factor in diabetes management. 1 Hypoglycaemia has 37 been found to be associated with cardiovascular events such as myocardial infarction, 38 arrhythmias and cardiovascular mortality as well as cerebral complications such as 39 dementia. 2-6 . Additionally, hypoglycaemia can impact patients' quality of life. 7 Patients with 40 moderate or worse symptoms of hypoglycaemia are less satisfied with their treatment and 41 report poorer adherence to their medications. 8 The great burden of hypoglycaemia is largely 42 presented by the considerable health care cost resulting from hospitalisations, ambulance 43 services, emergency department visits, and absenteeism from work. 9, 10 44 Older adults are the most susceptible age group to develop hypoglycaemia and to 45 experience hypoglycaemia-related complications. 11 The correlation between ageing and 46 hypoglycaemia in type II diabetes is multifactorial. [11] [12] [13] [14] [15] Factors such as physiological changes 47 in elderly that would affect the pharmacokinetics profile of anti-diabetic drugs, 48 comorbidities that affect heart and kidneys, nutrition changes and cognitive impairment that 49 could affect concordance and compliance with treatment regimen [11] [12] [13] [14] [15] . As the population of 50 older adults is increasing, globally; it is expected to see an increase in the prevalence of 51 T2DM in older adults. 16,17 52 In Jordan, the prevalence of diabetes in people aged 60 years and over increased 53 significantly over ten years period. 18 It is estimated that the number of elderly persons in 54 Jordan will be three times higher in 2050 than it was in 2017, consequently more older 55 adults will be at risk of developing type II diabetes and diabetes-related complications. 19 56 There are several studies on the association between hypoglycaemia and patient 57 characteristics such as patients' perspectives and attitudes towards diabetes management 58 skills, self-monitoring of blood glucose, and non-adherence to anti-diabetic medications in 59 Jordanian population. 20-23 However, there is a dearth of data on the interventions that could 60 potentially prevent hypoglycaemia in such patients, especially the older population. 61 Therefore, it is imperative to develop diabetes-related care strategies targeted to this broad 62 population of patients to cope with the growing figures in the future. 63 In the context of diabetes, pharmacist-led care interventions appear to have a pivotal role in 64 glycaemic control, improving self-care activities, medication adherence, improving quality of 65 life, and reducing related complications. 24-28 Pharmacist-led care interventions can be 66 individualised to each patient to achieve glycaemic control. 24,29,30 67 However, a few randomised controlled trials (RCTs) have investigated the impact of such 68 interventions on hypoglycaemia in adults diagnosed with T2DM. 31 Although elderly people 69 are considered heterogenous group with different characteristics from younger adults, none 70 of these trials has explored the effect of educational interventions in this age category. 71 Jordanian pharmacy education equips the pharmacists with robust clinical knowledge and 72 clinical skills to work with other healthcare professionals to provide optimal quality care to 73 the patients 32 . Therefore, the aim of this study is to evaluate the effectiveness of 74 pharmacist-led patient counselling on preventing hypoglycaemia in older adults with T2DM. 75 Educational interventions are considered to be "complex" interventions compared with 76 classic examples of drug interventions in RCTs. 33 That is an educational interventions' 77 success or failure could be attributed to a myriad of factors besides the interventions' 78 effectiveness. The factors such as the delivery of the intervention, understanding of the 79 intervention by the patients and implementation of intervention would decide the fate of an 80 educational intervention. 34 For this reason, a process evaluation is valuable to identify 81 whether an intervention works and how, barriers for its implementation, and how to 82 improve it in the future. 34 Undertaking qualitative studies to evaluate the interventions 83 during the implementation stage helps in modifying the ongoing interventions as well as the 84 study design to make them more feasible and effective. 35 The study will be a single-centre, two groups, 1:1 parallel, open-label, pragmatic randomised 102 controlled trial with a nested process evaluation embedded; and will be conducted in two 103 clinics at a referral tertiary hospital in Jordan. Participants will be randomly assigned to 104 either the intervention group hereinafter referred to as the "SUGAR Handshake" group or 105 the control group. Figure 1 illustrates the detailed study flow chart. hypoglycaemic episodes pose some challenges as the frequency of hypoglycaemic events is 114 the primary factor in calculating the sample size. 37, 38 We referred to the most relevant study 115 that used a similar outcome measure, methodology and intervention to our research to 116 calculate the needed sample size. 39 The previous study found that the mean total number of 117 hypoglycaemic attacks in the control and the intervention groups were 5.26±6.5 and 118 2.58±2.3 per patient in 24 weeks, respectively. We used the reported frequency of 119 hypoglycaemic attacks in both groups to calculate the minimum required sample size. 120 Therefore, we need to recruit at least 184 patients to achieve a significance level of 0.05 and 121 a study power of 80%. 40 Accounting for 10% to compensate attrition rate and missing data, 122 the final sample size wished to be recruited is 204 participants (102 in each group). 123 We will recruit older adults who are 65 years and above, diagnosed with T2DM and being 125 prescribed sulfonylurea, insulin, or any three anti-diabetic medications. Exclusion criteria 126 include patients unable to understand instructions or to give consent, diagnosed with 127 haemolytic anaemia or haemoglobinopathies, on palliative care for cancer, with advanced-128 stage or end-stage diseases who are terminally ill, diagnosed with psychosis or severe 129 depression, or with life expectancy < 6 months, impaired mental capacity, unwilling to take 130 home glucose measurements or to use the glucose meter, or unwilling to return for follow 131 up. Patients who have a partner or a first-degree relative who has been enrolled in the 132 study, are excluded as well. 133 We will use two recruitment methods to reach potentially eligible patients: an 134 advertisement placed in the reception room where patients wait for their appointments, 135 and through direct identification of potentially eligible patients at the recruitment sites. HA 136 will be responsible for recruiting participants in the endocrinology and diabetic foot care 137 clinics meanwhile the research assistant (RA) will recruit from the cardiology clinics. HA and 138 the RA will explain the trial purpose and processes and provide participant information 139 sheets (supplementary file 1) to the interested patients. They will also confirm the eligibility 140 of patients who are willing to participate and will ask them to sign a written consent form 141 We will randomly assign participants to the intervention and control groups on a 1:1 basis. 148 The random sequence will be generated using the website (www.randomization.com) to 149 generate the randomisation schedule. Envelopes will be used to conceal the allocation and will 150 be opened by the researcher sequentially at the time of each participant's enrolment. The study 151 envelopes will contain the study name, the participants' codes, the group to which the 152 participants are randomised. The randomisation sequence and the study envelopes will be 153 prepared by a third independent party who will not be involved in the study. The envelopes will 154 be closed and opaque and will be given to the researcher and the RA who are involved in 155 conducting the study. The envelopes that will contain the group allocation will be kept in a 156 locked cabinet in the hospital. 157 158 Blinding 159 160 Since this is an open-label study, the patients and the data collectors will not be blinded to 161 the assigned group. Eligible patients will be informed about the purpose of the study and the 162 study-related activities before they sign the consent form. Proper measures will be taken 163 along the trial duration to minimise performance and ascertainment bias that may result 164 from unblinding participants. 165 All participants will be unblinded to the study groups and to the real purpose of the trial at 166 the follow-up visit that would mark the end of the trial. Additionally, participants who are 167 assigned to the control group will receive the SUGAR Handshake intervention at the follow-168 up visit. 169 Intervention group (SUGAR Handshake) 170 The educational intervention, the SUGAR Handshake, is designed to promote behavioural 172 change to prevent hypoglycaemia. We applied the principles of the behaviour wheel theory 173 (BCW)to design the intervention. 41 Our educational intervention would enhance the physical 174 and psychological capabilities of the patients through improving their knowledge and skills in 175 managing and preventing hypoglycaemia. The intervention would also lead to the 176 behavioural change by a conducive environment to promote behaviour change by 177 addressing the physical and cultural needs of the patients. 178 We have structured the reporting of the intervention in line with the TIDieR (Template for 179 Intervention Description and Replication) checklist and guide. 42 Participants assigned to the 180 SUGAR Handshake group will receive individualised counselling regarding hypoglycaemia in 181 addition to the usual care provided at the trial sites. The intervention is designed by HA who 182 is a pharmacist with prior work experience in patient counselling and pharmacist-related 183 clinical services. HA has trained the RA on delivering the intervention. The intervention is 184 delivered in two steps i.e., a face-to-face conversation at the enrolment visit followed by a 185 phone call six weeks later. 186 Step One: Face-to-face conversation at the enrolment visit 187 The SUGAR Handshake intervention will cover comprehensive strategies to prevent and 195 handle hypoglycaemia with instructions related to anti-diabetic medications and managing 196 drug-related problems. Table 1 Step Two: Phone call at the 6 th Week: 206 Participants will receive a 20-minute follow up call at week six of enrolment; so that the first 207 step of the intervention would be reinforced as well as participants' queries/questions 208 would be answered. 209 Participants will also be asked about the number and timing of having hypoglycaemic attacks 210 during the first six weeks in the trial to re-consider modifying the intervention components. 211 Participants in this group will be offered guidance on hypoglycaemia diagnosis and the 214 proper use of the glucose meters in addition to the usual care. They will also be provided 215 with instructions on hypoglycaemia treatment. As participants in both groups will receive 216 the same information regarding hypoglycaemia recognition, they will have similar ability to 217 recognise hypoglycaemic attacks. At week 6 of enrolment, participants will receive a phone 218 call to remind them of using the glucose meters and documenting hypoglycaemic attacks. 219 Of special note, participants who complete the trial duration will receive the intervention at 220 the debrief visit and after returning the hypoglycaemia diaries. 221 All participants in both groups will be given glucose meters and test strips with a 224 demonstration on proper use to measure their blood glucose levels at morning before 225 breakfast daily for 12 consecutive weeks. They will also be handed diaries and instructed to 226 Additionally, it is impractical for DM patients to measure their BG levels frequently to 254 diagnose hypoglycaemia. Hence, it is imperative to account for both symptomatic and 255 asymptomatic types of hypoglycaemia. 256 We will use diaries to measure types of hypoglycaemia and we will ask participants to fill in 259 the diary on a daily basis for 12 weeks. Participants will be asked to fill in the diary with the 260 date of each day and the fasting blood glucose reading. Additionally, they will be asked to 261 tick on the boxes for every time they experience severe hypoglycaemia, symptoms of 262 hypoglycaemia at the time of fasting blood glucose measurement, and symptoms of 263 hypoglycaemia during the rest of the day. Participants will document a symptomatic attack if 264 the symptoms resolve after receiving the corrective actions. 265 The rate of each type of hypoglycaemia will be measured according to the hypoglycaemia 266 diaries filled by the participants. Severe hypoglycaemia will be measured directly according worked and what didn't, and how the study could be improved in future research. 284 We will collect qualitative data using semi-structured interviews from a handful of 285 participants in each study group. We have prepared the interview guide (table 3) based on 286 the objectives of the study and the MRC domains for process evaluation. 287 Table 3 : Interview schedule 288 You have previously read in the participant information sheet that we are conducting a phone interview as a part of this study and you accepted to participate in it. Therefore, I would like to ask you some questions about your experience with the study processes that have been provided to you at the inclusion visit. The information will help us in improving several aspects of the study. The interview should take about 10 minutes. Are you available to respond to the questions at this time? 2. I would like to start by asking: how do you describe your participation in the study so far? If needed, the interviewers may explain the question by the follow-up question: How do you rate your participation in the study from good to poor and why? 3. What has worked for you from the study processes that you were asked to do? -why do you think they have? what hasn't worked for you from the study processes that you were asked to do? -why do you think they haven't? 4. From your perspective, how the study could be improved? Please consider any aspects of the study that you think could be improved. per-protocol analysis will be conducted including participants who are compliant to 80% or 293 more of the study protocol. 45, 46 The missingness in the primary outcome will be handled 294 under the assumption of "missing at random" rather than "missing completely at random", 295 because we are expecting that the probability of missing data depends on observed 296 covariates or outcomes rather than unobserved data. Therefore, the method chosen to 297 handle missing data in the outcomes is multiple imputations. 47,48 298 Descriptive analysis will be used across randomised groups and quantitative analysis will be 299 conducted in RStudio version 3.4.2 (28-9-2017). 49 Continuous variables will be presented as 300 means, standard deviation, median and interquartile range, meanwhile, categorical variables 301 will be presented as frequencies and percentages. The randomised groups will be examined 302 and compared for all variables. Categorical variables will be evaluated using the chi-square 303 test. Continuous variables will be tested for normality and based on the distribution of the 304 data, appropriate parametric or non-parametric tests will be used. For example, paired t 305 test or Wilcoxon test would be applied to assess the differences in baseline variables 306 between both groups and between participants who completed the trial as well as the 307 participants who will be lost to follow up. 308 The primary and secondary outcomes (total and types of hypoglycaemic attacks) will be 309 measured across the randomised groups and compared using the analysis of covariance 310 (ANCOVA), and the appropriate parametric or non-parametric tests dependent on the 311 normality of distribution. Subgroup analysis will also be performed using interaction terms in 312 regression models. 313 We will use logistic regression analysis to examine associations between the categorical 314 outcome variable ( frequent vs infrequent hypoglycaemia episodes) and independent 315 variables such as sex, age, educational level, living arrangements, duration of diabetes, 316 number of current medications, experiencing previous hypoglycaemia, the status of self-317 monitoring of blood glucose, types of anti-diabetic medications, baseline HbA1c and 318 interactions between the independent variables. The findings will be also presented as odds 319 ratio and 95% confidence intervals. A P value of less than 0.05 will be considered statistically 320 significant. Hypoglycaemia rates will be described with Kaplan-Meier survival curves, 321 considering the time to the first hypoglycaemic attack as the outcome measure. 50 322 Qualitative data analysis 323 The interviews will be audio-recorded then transcribed and translated into the English 324 language. We will use the thematic analysis approach to analyse the collected data for there is a need for protocol amendments through the embedded evaluation process, the 344 changes will be discussed by the supervisory team and will be communicated in writing to 345 the Human Research Ethics Committee of the University of Lincoln and the Institutional 346 Review Board of KAUH. 347 Upon completion of the study, we will provide the trial site with an executive summary of 349 the findings in the form of a report. Participants would be able to get the results of the study 350 from their health care professionals at the trial site no later than one year after the end of 351 data collection. We are planning to disseminate the study outcomes through peer-reviewed 352 publications and presentations in conferences. We will comply with the authorship eligibility 353 The SUGAR Handshake intervention is designed to be pragmatic and to facilitate 373 transferability of evidence into practice. Therefore, pharmacists can easily deliver it to the 374 patients in different working positions including hospitals and community pharmacies. 375 Moreover, the delivery of the SUGAR Handshake intervention is cheap and will not cost an 376 extra burden on patients. Previous studies concerning the attitudes, religious beliefs, and 377 self-management skills amongst patients with T2DM helped in assuring the appropriateness 378 of contextual and cultural delivery as well as the implementation of the study and the 379 SUGAR Handshake intervention. 20-23 Using blood glucose/hypoglycaemia diaries to 380 objectively report and measure several types of hypoglycaemia is another strength. This will 381 facilitate a more accurate measurement of hypoglycaemic events where the concern that 382 patients may forget to report the experienced episodes is reduced. 383 A plausible limitation that warrants consideration is the short follow up duration (12 weeks), 384 which may make it difficult to examine the sustainability of the intervention effect. 385 However, we anticipate that this duration will decrease the dropout rate. Moreover, we 386 expect the effect of our intervention to last up to at least six months as concluded by a 387 previous trial. 39 Another concern is the enrolment of relatives into different groups upon 388 randomisation, which will introduce contamination bias. Therefore, if a patient happens to 389 have a relative who has already participated in the trial, he would be excluded. Additionally, 390 participants may not fully adhere to the intervention during the follow-up period. For this reason, they will receive a phone call reminder at week six of enrolment. Individualising the 392 intervention according to each patient's lifestyle and potential causes of hypoglycaemia will 393 enhance the adherence to the intervention as well. 394 As this is an open-label study, performance bias and ascertainment bias may result from 395 unblinding participants and the data collectors, respectively. Participants in the control 396 group may be less adherent to the trial protocol and more likely to withdraw from the trial. 397 However, efforts will be made to standardise the trial protocol, frequency and time of follow 398 up, and treatment of experienced hypoglycaemia across both groups to minimise 399 performance bias. We also anticipate that the objective measurement of the outcomes 400 would minimise ascertainment bias. 401 The prevalence of diabetes in Jordan has been growing rapidly to reach 23.7% in 2017. 60 In 402 light of the lack of awareness regarding diabetes diagnosis, causes, and management we 403 would expect a further increase in the number of Jordanians who are diagnosed with 404 diabetes and who would suffer from diabetes-related complications. 20,60 While pharmacists 405 are easier to access than physicians, a possible strategy to mitigate the burden of diabetes is 406 to establish and support the pharmacist-led, patient-oriented services. 61 International Hypoglycaemia Study Group IHS. Minimizing Hypoglycemia in Diabetes Diabetes Care Adverse Macrovascular Events, and 456 Inflammation in the Edinburgh Type 2 Diabetes Study. Diabetes Care Association of 460 hypoglycaemia and risk of cardiac arrhythmia in patients with diabetes mellitus: A 461 systematic review and meta-analysis Severe 465 hypoglycaemia, mild cognitive impairment, dementia and brain volumes in older adults 466 with type 2 diabetes: the Atherosclerosis Severe Hypoglycemia 470 and Risks of Vascular Events and Death 475 Effects of Severe Hypoglycemia on Cardiovascular Outcomes and Death in the Veterans 476 Affairs Diabetes Trial. Diabetes Care Predictors of Quality of Life and Other Patient-Reported Outcomes in the 480 PANORAMA Multinational Study of People With Type 2 Diabetes Impact of 484 symptomatic hypoglycemia on medication adherence Healthcare 495 resource use, direct and indirect costs of hypoglycemia in type 1 and type 2 diabetes, 496 and nationwide projections. Results of the HYPOS-1 study Management of hypoglycemia in older adults with type 2 diabetes Sircar et al M. Review of Hypoglycaemia in the Older Adult: Clinical Implications and 503 Cardiovascular Disease 507 Predicts Severe Hypoglycemia in Patients with Type 2 Diabetes Diabetes Management in Older Adults With Chronic 511 Curr Diab Rep Older Adults: Standards of Medical Care in Diabetes-2019 Diabetes Care The Department of Economic and Social Affairs of the United Nations. World economic 517 and social survey, 2007: development in an ageing world The burden of disease in older people and implications for 520 health policy and practice. The Lancet An increase in prevalence of 524 diabetes mellitus in Jordan over 10 years Personalized Diabetes Management: Moving from 572 Algorithmic to Individualized Therapy Self-efficacy-focused education in persons with 576 diabetes: a systematic review and meta-analysis The role of structured education in the management of 580 hypoglycaemia Pharmacy education in Jordan: updates When are complex interventions 'complex'? When are simple 586 interventions 'simple'? Developing and 590 evaluating complex interventions: the new Medical Research Council guidance Process 594 evaluation of complex interventions: Medical Research Council guidance SPIRIT 2013 Statement: Defining standard 600 protocol items for clinical trials Hypoglycemia: a 602 review of definitions used in clinical trials evaluating antihyperglycemic drugs for 603 diabetes Hypoglycemia Event Rates: A Comparison 606 Between Real-World Data and Randomized Controlled Trial Populations in Insulin-607 Treated Diabetes. Diabetes Ther Intensive Individualized 610 Reinforcement Education Is Important for the Prevention of Hypoglycemia in Patients 611 with Type 2 Diabetes The behaviour change wheel: A new method for 617 characterising and designing behaviour change interventions Better 621 reporting of interventions: template for intervention description and replication 622 (TIDieR) checklist and guide Hypoglycemia 625 and Diabetes: A Report of a Workgroup of the American Diabetes Association and The Diabetes Care Understanding 629 the new HbA1c units for the diagnosis of Type 2 diabetes Research Council (MRC) guidance. MRC Popul Health Sci Res Netw The prevention and handling of the missing data A Comparative Analysis of Generalized Estimating 647 Equations Methods for Incomplete Longitudinal Ordinal Data with Ignorable Dropouts Using thematic analysis in psychology Community-Based Peer-Led Diabetes Self-655 management Effectiveness of HypoAware, a Brief 658 Partly Web-Based Psychoeducational Intervention for Adults With Type 1 and Insulin-659 Treated Type 2 Diabetes and Problematic Hypoglycemia: A Cluster Randomized 660 Efficacy of structured education in patients with type 2 diabetes mellitus 663 receiving insulin treatment Diabetes X-PERT Programme makes a difference The effect of an education 670 programme (MEDIAS 2 BSC) of non-intensive insulin treatment regimens for people 671 with Type 2 diabetes: a randomized, multi-centre trial Exclusion of Older Adults from Ongoing Clinical Trials About Type 2 Diabetes 676 Using Shared 679 Decision-Making to Address Possible Overtreatment in Patients at High Risk for 680 Hypoglycemia: The Veterans Health Administration's Choosing Wisely Hypoglycemia 681 Safety Initiative. Clin Diabetes Publ Am Diabetes Assoc A Novel 685 Intervention Including Individualized Nutritional Recommendations Reduces 686 Hemoglobin A1c Level, Medication Use, and Weight in Type 2 Diabetes. JMIR Diabetes Time trends in 690 diabetes mellitus in Jordan between 1994 and 2017 Health Care and Pharmacy Practice in Jordan Pharmacist-Led Study in Controlling Hypoglycemia in Older Adults With 698 Type 2 Diabetes Mellitus (ROSE-ADAM) We would like to acknowledge the University of Lincoln and Isra University for their support 420 to conduct this study. We would also acknowledge the physicians and nursing staff working 421 at KAUH for facilitating recruitment and the study implementation.