Adherence of Patients with Type 2 Diabetes to Medications – The Role Of Risk Preferences

VA L U E I N H E A LT H 1 9 ( 2 0 1 6 ) A 8 0 7 – A 9 1 8

of these cases. Saxagliptin, administrated with metformin in patients with Type 2 diabetes inadequately controlled with metformin, has been gaining more attention in recent years. However, there is few cost effectiveness studies of saxagliptin in China. This review aims to be the gap analysis.  Methods: We searched for Chinese literatures in CNKI. Search pattern was “saxagliptin” AND “cost or economic or expense” in abstract. Publication deadline was Mar 5th, 2016. Cost analysis (CA), cost-effectiveness analysis (CEA), cost-utility analysis (CUA), cost-benefit analysis and disease burden study related to saxagliptin were included. Results: We retrieved 183 abstracts. After abstract and full-text analysis, 5 studies were finally included. Two are the reviews of foreign researches. Both show saxagliptin+metformin was associated with improvement in QALYs and medical expenses, compared with sulfonylureas, thiazolidinediones and NPH insulin. One study uses a Cardiff Model to estimate the long-term costeffectiveness of saxagliptin+metformin vs glimepiride+metformin. It finds saxagliptin dominates glimepiride with an ICER of ¥-43,883/QALY. One study is a CEA of saxagliptin+metformin and pioglitazone+metformin based on 96 outpatient in China and concludes that saxagliptin enjoys significantly higher effective rates in 2hPG and HbA1c (89.58% vs 60.42%, 89.58% vs 58.33%), and lower cost effectiveness ratios (¥15.57 vs ¥16.63, ¥15.57 vs ¥17.22) than pioglitazone. One study studies the direct medical cost when saxagliptin take the place of sulfonylureas based on 4628 inpatients all over China, and shows drug expenditure of over 50 million Yuan would be saved by saxagliptin annually.  Conclusions: Based on available evidences, our review demonstrates that saxagliptin + metformin is a cost-effective choice. More research is needed for decision making in Type 2 diabetes second line treatment, especially local real world cost-effective study. PDB25 The Longterm Cost Effectiveness of the “Coaching Patients on Achieving Cardiovascular Health” (Coach) Program in Type 2 Diabetes in Tasmania Palmer AJ1, Vale MJ2, Wells CL3, Ski C2, Thompson DR2, Si L4 Institute for Medical Research, Hobart, Australia, 2Australian Catholic University, Melbourne, Australia, 3Diabetes Tasmania, Hobart, Australia, 4University of Tasmania, Hobart, Australia

1Menzies

Objectives: The “Coaching patients On Achieving Cardiovascular Health” (COACH) Program is a structured telephone and mail-out delivered program that trains people with chronic diseases including type 2 diabetes to achieve guidelinerecommended target levels for their cardiovascular risk factors while working with their usual medical practitioners. Our aim was to evaluate the long-term cost-effectiveness of this innovative patient education program in type 2 diabetes patients in Tasmania.  Methods: Baseline cohort characteristics and treatment effects were derived from data collected by Diabetes Tasmania in a real life setting at baseline (N= 975) and 18 months after a 6-month COACH Program training period. A published and validated simulation model of diabetes (IMS-CORE Diabetes Model) projected 24-month improvements in glycemic control, blood pressure and lipids to quality-adjusted life years (QALYs) and total lifetime direct costs in type 2 diabetes patients with The COACH Program+usual care versus usual care alone. Australian-specific complication costs were taken from published sources. The costs of The COACH Program were included. Analyses were run over patient lifetimes and were performed from a third-party payer perspective. Future costs and QALYs gained were discounted at 5% per annum. Probabilistic sensitivity analysis was performed.  Results: The COACH Program cost AUD1,296/patient to implement. After 6 months, HbA1c improved by mean (standard deviation) 0.35%-points (1.00%-points) from baseline of 7.15% (1.50%), and had a positive effect on lipid profile, BMI and blood pressure, translating into reduced risks of micro- and macrovascular complications, and increased QALYs of mean (95%CI) 0.058(0.023-0.092) years and increased costs of AUD 215(-902 to 1,333) per patient versus usual care alone, with an incremental cost-effectiveness ratio of AUD3,737/ QALY gained.  Conclusions: Long-term projections indicated that improved glycemic control, blood pressure and lipid profile sustained over an 18 month period following the 6 month COACH Program is highly cost effective in Type 2 diabetes patients in Tasmania.

DIABETES/ENDOCRINE DISORDERS – Patient-Reported Outcomes & Patient Preference Studies PDB26 Self-Care Behavior And Medication Use Of Buddhist Monks With Diabetes Mellitus In Phasi Charoen, Bangkok Yensabai J1, Torthear P1, Neamsripetch V1, Karndee A1, Pongmesa T1, Luckanajantachote P2 1Faculty of Pharmacy, Silpakorn University, Nakhon Pathom, Thailand, 2Department of Pharmacy, Samutsakhon Hospital, Samut Sakhon, Thailand

Objectives: To determine self-care behavior and medication use among Buddhist monks with diabetes mellitus (DM) from the temples in Phasi Charoen district in Bangkok, Thailand.  Methods: This qualitative study was conducted among all available monks with DM aged 21 years or over staying in the temples in Phasi Charoen district (altogether 28 temples). Data were collected through face-toface interviews with a semi-structured questionnaire from February to March 2015.  Results: A total of eighteen monks with type 2 DM (median age [range]: 65.1 [42-79] years) were recruited from 12 temples. Most monks took two meals a day, breakfast and lunch (before 12 p.m.). Almost every monk received recommendations on diabetic diet (88.89%), and hence chose DM-appropriate items from the foodstuff offered. However, several of them did not follow diet recommendations believing that they could not choose. Regarding DM knowledge, most monks lacked information on the causes of DM (61.11%), appropriate glycemic level (50.00%) and

A899

hypo-/hyperglycemic symptoms (33.33%/61.11%) because the lack of recommendations from healthcare professionals. Regarding exercise, most of them generally exercised by going on an alms-round or performing a monk’s duties (50.00%), whereas 3 monks (16.67%) said exercise was prohibited for monks. With regard to DM medication, the majority received combination therapy (77.78%) with sulfonylureas and metformin (61.11%). Drug-related problems were detected in 15 monks (83.3%), and the most commonly identified problems were non-compliance (72.22%), especially due to forgetfulness, and adverse effects of medications (27.8%), including hypoglycemia.  Conclusions: Some DM monks still had inappropriate self-care behavior and medication use, while most of them lacked DM knowledge. Accordingly, healthcare professionals should realize more the importance of providing DM education to the monks for optimal self-care behavior and medication use under the limitations of Buddhist disciplines. PDB27 Adherence of Patients with Type 2 Diabetes to Medications – The Role Of Risk Preferences Simon-Tuval T1, Shmueli A2, Harman-Boehm I3 1Ben-Gurion University of the Negev, Beer-Sheva, Israel, 2The Hebrew University of Jerusalem, Jerusalem, Israel, 3Ben-Gurion University of the Negev; Soroka University Medical Center, Beer-Sheva, Israel

Objectives: To examine whether risk preferences are associated with adherence to oral hypoglycemic agents (OHA).  Methods: We performed a cross-sectional study among patients with type 2 diabetes (n= 308) aged 21–70 presenting for routine visits in the diabetes clinic at a university medical center. We used validated questionnaires to estimate: risk preferences, motivation, self-efficacy, impulsivity, perception about the disease and about the interpersonal process of care, and demographic and socio-economic characteristics, in addition to computerized patient medical records for retrieval of clinical indicators to estimate disease severity and a computerized database for retrieval of medication adherence one-year before the interview. Adherence was estimated using prescription-based measures of proportion of days covered (PDC). Concurrent adherence was calculated in three ways: PDC with ≥ 1 OHA; average PDC; PDC of ≥ 80% for all OHA. We classified patients into one of three groups using standard thresholds: ≥ 80%, 40-79%, and < 40%. Multivariable ordered logit and logistic models were specified to estimate predictors of adherence to each of the adherence measures.  Results: Multivariable analyses revealed that compared to others, risk-seeking patients had lower PDC with ≥ 1 OHA (β  =  -0.50, p≤ 0.1). Specifically, compared to the rest of study population, risk-seeking patients were 11.2 percentage points less likely to have ≥ 80% of the follow-up period covered with ≥ 1 OHA available (p≤ 0.1). In addition, risk-seeking patients had lower average PDC (β  =  -0.85, p≤ 0.05). Specifically, compared to others, risk-seeking patients were 19.5 percentage points less likely to have an average PDC of ≥ 80% (p≤ 0.05). Finally, risk-seeking was associated with lower probability to have PDC of ≥ 80% for all OHA in the follow-up period (OR; 90% CI: 0.59; 0.35 – 0.97).  Conclusions: Risk seekers are less adherent to OHA. Identifying these patients may enable practitioners to proactively tailor strategies to improve their adherence and health outcomes. PDB28 Disease Related Knowledge, Medication Adherence And Glycaemic Control Among Patients With Type 2 Diabetes Mellitus In Pakistan Nazir SU1, Hassali MA2, Saleem F2 of Pharmaceutical Sciences, Penang, Malaysia, 2Universiti Sains Malaysia, Penang, Malaysia

1School

Objectives: The purpose of this study was to investigate the association of diabetes-related knowledge and treatment adherence with glycaemic control among patients with type 2 diabetes mellitus (T2DM) in Pakistan.  Methods: The study was designed as a questionnaire-based, cross-sectional analysis. T2DM patients attending a public outpatient clinic in Sargodha, Pakistan, were targeted for the study. In addition to the demographic information, the Urdu version of Michigan Diabetes Knowledge Test and Morisky Medication Adherence Scale were used for data collection. Patients’ medical records were reviewed for glycated haemoglobin levels (HbA1c). Descriptive statistics were used to elaborate sociodemographic characteristics. The Spearman’s Rho correlation was used to measure association of disease-related knowledge and treatment adherence with glycaemic control. SPSS v 20.0 was used for data analysis and p <  0.05 was taken as significant.  Results: In total, 392 T2DM patients were included in the study. The mean age (SD) of these patients was 50.77 ± 9.671 years, 56.6% were males and 90% (n =  353) of respondents were married. The mean (SD) duration of disease was 5.58 (4.09) years with median HbA1c of 9.00 (IQR =  8.20–10.40). The median knowledge score was 8.0 (IQR =  6.0–10.0), while the median adherence score was 4.7 (IQR =  3.0–6.0). HbA1c had non-significant and weak negative association with diabetes-related knowledge (r = −0.036, p = 0.404) and treatment adherence (r = −0.071, p =  0.238).  Conclusions: There was negative association reported between HbA1c, treatment adherence and diabetes-related knowledge. Greater efforts are clearly required to investigate other factors affecting glycaemic control among T2DM patients in Pakistan.

PDB29 Effect Of A Pay-For-Performance Program For Diabetes On Health Status (Eq-5d) In Taiwan Chen T1, Lang HC2 1Fu Jen Catholic University, NewTaipei, Taiwan, 2National Yang-Ming University, Taipei, Taiwan

Objectives: The literature on pay-for-performance (P4P) has rarely studied healthrelated quality of life among type 2 diabetes mellitus using EuroQol-5D (EQ-5D). This study assesses the difference in health status and health-related quality of life among patients participating in a newly featured P4P program and those not participating in the program. Methods: Respondents at hospitals completed