Background and aims: Hypoglycaemia is an important concern for patients with diabetes and physicians when setting glycaemic targets. The Hypoglycaemia Assessment Tool (HAT) study, the largest and most comprehensive of its kind, assessed self-reported hypoglycaemia and associated predictive factors in a global population of patients with insulin-treated diabetes.
Materials and methods: HAT was a non-interventional, multicentre, 6-month retrospective and 1-month prospective study of hypoglycaemic events in 24 countries using self-assessment questionnaires and patient diaries (for 28 days) in people aged ≥18 years with type 1 (T1D) or type 2 (T2D) diabetes using insulin for ≥12 months attending routine clinics. Associations between predictive factors and hypoglycaemia were examined using negative binomial regression models adjusted for period and country.
Results: 27,585 patients completed the study (Table 1). 83.4% of patients with T1D and 50.8% of patients with T2D experienced ≥1 hypoglycaemic event in the 4 weeks before baseline (51.5 and 16.5 events per patient year). Higher (p < 0.001) incidence rates were reported in the 4 weeks after baseline (73.3 [T1D] and 19.3 [T2D] events per patient year). A greater percentage of patients with T1D vs. T2D reported any (83.0 vs. 46.5%), nocturnal (40.6 vs. 15.9%) or severe (14.4 vs. 8.9) hypoglycaemia in the prospective period.
Conclusion: In this large, multinational population of patients (insulin-treated) with T1D or T2D, rates of overall, nocturnal and severe hypoglycaemia were higher than previously published. An increased incidence of overall hypoglycaemia in the prospective study indicated significant under-reporting of hypoglycaemia.
Depressive symptoms in diabetes are associated with reduced self-care,
glycaemic control and health-related quality of life (hrQOL) and increased
diabetes-specifi c distress. We analysed if the recovery from depressive
symptoms would be associated with improvements in these aspects. 182
diabetes patients (age 46 ± 13 y.; 56% female; BMI 30 ± 7; 62% type 1
diabetes; illness duration 15 ± 10 y.; 95% with insulin; HbA1c 8.7 ± 1.7%)
with subclinical depressive symptoms (CES-D score ≥ 16 without meeting
criteria for clinical depression; mean CES-D score: 23 ± 8) participated in a
prospective study. Recovery was defi ned as CES-D score < 16 at 12-month
follow up. Dependent variables were diabetes self-care (SDSCA), glycaemic
control (HbA1c), diabetes distress (PAID) and hrQOL (SF-36). We
compared baseline-to-follow up changes between recovered versus nonrecovered
patients using ANCOVA (adjusted for baseline values). At follow
up, 85 patients (47%) showed recovery. The mean reduction of depressive
symptoms in this group was -13 ± 9 CES-D scale points; the mean change in
the 97 patients remaining depressed (53%) was +2 ± 9 CES-D scale points.
Recovered patients compared to unrecovered ones showed signifi cantly
greater improvement (baseline-to-follow up change) regarding self-care
(+0.14 ± 1.11 vs. -0.19 ± 1.05 SDSCA scale points, Δ = 0.31, P = 0.014), glycaemic
control (-0.78 ± 2.19 vs. -0.56 ± 1.53 HbA1c %-points, Δ = 0.12, P =
0.042), diabetes distress (-13.6 ± 18.8 vs. -4.4 ± 17.1 PAID scale points, Δ =
0.51, P < 0.01) and hrQOL (physical hrQOL: +2.0 ± 8.7 vs. -1.1 ± 9.6 T scores,
Δ = 0.341, P = 0.005; mental hrQOL: +14.5 ± 11.9 vs. +0.1 ± 12.2 T scores, Δ =
1.19, P < 0.01). This study provides evidence that recovery from depressive
symptoms may have positive impact on diabetes control, diabetes-specifi c
distress and quality of life.
Questionnaires assessing the satisfaction of patients already exist, but no
questionnaire assesses the satisfaction of physicians. However, physicians’
satisfaction with patient-communication and with diabetes therapy is a crucial
factor for the effi cacy of diabetes therapy. A relevant part of patientcommunication
is the discussion of blood glucose values with the patient. In
order to systematically assess the perspective of physicians, we developed
For author disclosure information, see page A810.
& Guided Audio Tour poster ADA-Funded Research
Behavioral Medicine, Clinical
Nutrition, Education, and Exercise
PSYCHOSOCIAL, BEHAVIORAL MEDICINE
a questionnaire that assesses satisfaction of physicians with diabetes therapy
in general (Sat-DT) and with the discussion of blood glucose values (SatBG)
in particular. Data from 188 physicians who completed the questionnaire
was analyzed. The fi nal Sat-DT scale comprised 13 items and achieved a
Cronbach’s Alpha of 0.93 wit a mean item-total correlation of r=0.71. The
fi nal Sat-BG scale comprised 10 items and achieved a Cronbach’s Alpha of
0.92 with a mean item-total correlation of r=0.73. Factor analysis (Varimax
rotation) revealed two factors within the Sat-DT scale (65% explained variance):
“Effects of diabetes therapy” and “costs and benefi ts.” The Sat-BG
scale could be divided into “Usage of blood glucose data” and “effi cacy of
the discussion” by factor analysis (76% explained variance). A greater satisfaction
in both scales was associated with a greater satisfaction with the
work as a physician (Sat-DT: r=0.2, p<.05; Sat-BG: r=0.32, p<.05), a greater
satisfaction with therapy outcomes (Sat-DT: r=0.34, p<.05; Sat-BG: r=0.3,
p<.05), and with a greater satisfaction with the results of working as a physician
(Sat-DT: r=0.27, p<.05; Sat-BG: r=0.25, p<.05). This new questionnaire
with its two scales is a reliable and valid assessment tool to measure the
satisfaction of physicians. In future studies, this questionnaire can be used
to investigate the infl uence of physicians’ satisfaction on diabetes therapy
as well as how physicians’ satisfaction can be altered.
This cross-over study used a CGM system (DexCom SEVEN PLUS CGM).
In a randomized order, participants had either no access (CGM blind) or real
time access to current glucose data (CGM open). One objective was to analyze
if type 1 diabetic patients with hypoglycemia problems (at least one
episode requiring third party assistance) could be identifi ed by the use of the
blinded CGM data. We also analyzed the impact of CGM use on biochemical
hypoglycemia. Type 1 diabetic patients with hypoglycemia problems had
signifi cant longer diabetes duration (17.0 vs. 11.0 yrs.), a higher unawareness
score (4.0 vs. 2.0) and lower thresholds for detecting hypoglycemia (50.0
vs. 65.0 mg/dl) than patients without hypoglycemia problems. During the
blinded CGM phase patients with hypoglycemic problems had a signifi cant
longer duration of low glucose phases 248 vs. 153 minutes per day (p=.037;
<70 mg/dl) respectively 173 vs. 96 minutes per day (p=.041; <60 mg/dl). Area
under the receiver operating curve (ROC 0.72 p=.03) indicated a suffi cient
screening performance of the duration of low glucose periods (< 70 mg/dl)
for the identifi cation of patients with hypoglycemia problems. A cut-off of
170 minutes per day of time spend in the low glucose range had a sensitivity
of 75% and a specifi city of 70.3%; the positive predictive value was 52.9%,
the negative predictive value was 86.4%. A comparison of blind vs. open
CGM showed that time spend in a low glucose range could be signifi cantly
more reduced in patients with hypoglycemia problems than in patients without
hypoglycemia problems during CGM open (< 60 mg/dl; - 67.8 min per day
p=.040; < 50 mg/dl; -50.6 min per day, p=.038; < 40 mg/dl; -41.4 min. per day,
p=.03). This study shows that CGM has an unused potential for identifying
type 1 diabetic patients at risk for hypoglycemia problems in clinical practice
as well as for avoidance of biochemical hypoglycemia, which plays a pivotal
role for the development of hypoglycemia associated autonomic failure.
Association of depression with glycemic control are not conclusive. While
some studies found a positive association, others found none or found that
diabetes distress is a mediating factor. These inconsistencies might be
due to the complex symptomatology of depression. Depressive symptoms
can range from sleep or appetite disorders to depressed mood and having
crying spells. Thus, depressive symptoms can be divided into somatic and
affective symptoms. This study investigated the associations of the different
depressive subtypes with glycemic control. 923 patients completed the
Center for Epidemiological Studies - Depression scale (CES-D) which offers
subscales for somatic and affective symptoms. Linear regression analysis
with HbA1c as dependent variable was conducted. Independent variables of
interest were the somatic and affective scores of the CES-D controlled for
demographic (age, gender, body mass index, education) and medical (diabetes
type, diabetes duration, number of SMBG, late-complications) variables
as well as diabetes distress. Both depressive subtypes were signifi cantly
associated with HbA1c. A greater somatic symptomatology was associated
with a higher HbA1c (β = .15; p = .001) whereas a greater affective symp-
For author disclosure information, see page A810.
ADA-Funded Research & Guided Audio Tour poster
Behavioral Medicine, Clinical
Nutrition, Education, and Exercise
PSYCHOSOCIAL, BEHAVIORAL MEDICINE
tomatology was associated with a lower HbA1c (β = -.15; p = .001). Greater
diabetes distress was associated with higher HbA1c (β = .13; p = .001). Linear
regression with depressive symptoms in general (CES-D total score) as independent
variable revealed no signifi cant association (β = -.01; p = .86). This
study demonstrated that depression is a complex condition and offers an explanation
for the inconsistencies in current literature regarding associations
with glycemic control. Only the differentiation of subtypes offered a more
complete picture of the associations of depression with glycemic control.
In clinical practice and further research, a closer look which symptoms of
depression are present may be helpful to better understand depression as
a vulnerability factor.
A new treatment program has to prove its effectiveness in a randomized
controlled trial (RCT). However, health care research trials (HCRT) show that
the effi cacy of education programs is far less convincing in daily routine.
We evaluated whether a new education and treatment program for type 1
diabetic patients (PRIMAS) had similar effects in daily routine as in the RCT.
255 people with type 1 diabetes from 42 practices took part in the PRIMAS
course and were observed in this HCRT. As in the RCT, PRIMAS consisted of
12 lessons and the outcomes were assessed 6 months after the education
course. Primary outcome in both studies was HbA1c assessed in the same
central laboratory. Improvement in hypoglycemia Awareness, depressive
symptoms, diabetes-related distress, self-effi cacy, and diabetes empowerment
were secondary outcomes. In order to compare the effects of PRIMAS
in the RCT with the effects in the HCRT, effect sizes of differences for each
study were contrasted. The difference of the two respective effect sizes
(RCT - HCRT) along with the 95% confi dence interval was analyzed. HbA1c
reduction in the RCT was -0.36% (-3.9 mmol/mol) as compared to -0.39%
(-4.3 mmol/mol) in the HCRT. The difference of the effect sizes didn’t exceed
the non-inferiority margin of 0.4 (difference of effect size: α -0.03, 95%
CI -0.29 to 0.22). Effect sizes in all secondary outcomes were also highly
comparable: Improvement of hypoglycemia awareness α 0.11, 95% CI -0.15
to 0.38; reduction of depressive symptoms α -0.08, 95% CI -0.34 to 0.18;
reduction of diabetes distress α 0.23, 95% CI -0.04 to 0.49; improvement
of self-effi cacy α 0.05, 95% CI -0.21 to 0.32; improvement of empowerment
α 0.04, 95% CI -0.23 to 0.36. PRIMAS proofed its effi cacy in daily routine
and showed similar effect sizes as in the RCT. PRIMAS can contribute to an
improvement of routine health care in people with type 1 diabetes.
Over the last years, the utilization of cloud resources has been steadily rising and an increasing number of enterprises are moving applications to the cloud. A leading trend is the adoption of Platform as a Service to support rapid application deployment. By providing a managed environment, cloud platforms take away a lot of complex configuration effort required to build scalable applications. However, application migrations to and between clouds cost development effort and open up new risks of vendor lock-in. This is problematic because frequent migrations may be necessary in the dynamic and fast changing cloud market. So far, the effort of application migration in PaaS environments and typical issues experienced in this task are hardly understood. To improve this situation, we present a cloud-to-cloud migration of a real-world application to seven representative cloud platforms. In this case study, we analyze the feasibility of the migrations in terms of portability and the effort of the migrations. We present a Docker-based deployment system that provides the ability of isolated and reproducible measurements of deployments to platform vendors, thus enabling the comparison of platforms for a particular application. Using this system, the study identifies key problems during migrations and quantifies these differences by distinctive metrics.
Background and aims: In a randomised, multi-centre trial, the effect of a structured education and treatment programme to promote self-management and empowerment for type 1 diabetic patients (PRIMAS) was compared to an established education programme (Structured education and treatment programme for type 1 diabetic patients) as an active comparator condition (ACC). Besides optimizing insulin treatment by testing basal insulin doses and carbohydrate factors as well as avoiding acute and late complications the PRIMAS programme also aims at motivational factors and the daily routine of living with diabetes.
Materials and methods: A total of 160 patients with type 1 diabetes recruited in diabetologist practices (PRIMAS: age 45.9 ±13.8 yrs, diabetes duration 19.6 ±12.8 yrs., 38.3% female, HbA1c 8.3 ±1.1%, 23.5% with CSII vs. ACC: age 45.2 ±13.4 yrs, diabetes duration 19.4 ±13.2 yrs., 49.4% female, HbA1c 8.1 ±0.9%, 27.8% with CSII) were randomised to either PRIMAS or the ACC. Main outcome was the HbA1c 6 months after participation in the respective programme. Secondary outcomes were diabetes related distress (Diabetes Distress Scale 17 items), extent of self-perceived empowerment (Empowerment scale 11 items), self-efficacy for diabetes management (Self-efficacy scale 10 items), diabetes knowledge (Knowledge test 11 items), and self-care behaviour (Summary of Diabetes Self Care Activity Scale). Baseline adjusted results are reported as mean ± SEM.
Results: Baseline adjusted HbA1c was significantly more improved in PRIMAS than in the ACC (-0.33 ± 0.1% vs. -0.03 ± 0.1%, p=.028) at 6 month follow-up. Also patients treated with PRIMAS reported lower baseline adjusted diabetes related distress levels (mean item score 0.9 ± 0.1 vs. 1.1 ± 0.1, p=.049), higher empowerment (27.4 ± 0.5 vs. 26.0 ± 0.5 p=.027) and self-efficacy levels (mean score: 23.3 ± 0.4 vs. 21.9 ± 0.4 p=.044) than patients in the ACC. Satisfaction with insulin treatment was also higher after 6 month in PRIMAS than in the ACC (mean score: 27.1 ± 0.6 vs. 25.3 ± 0.6; p=.028). Effects on knowledge (mean score 7.7 ± 0.2 vs. 7.9 ± 0.2 p=.555) and self-care behaviours (mean score: 3.7 ± 0.1 vs. 3.8 ± 0.1; p=.846) did not differ significantly between the education programmes nor did insulin dose (0.66 ± 0.02 vs. 0.63 ± 0.02 IU/kg; p=.284).
Conclusion: PRIMAS is more effective in lowering HbA1c than the established education programme in type 1 diabetic patients. In addition, PRIMAS has proven its efficacy regarding the improvement of empowerment, self-efficacy and satisfaction with insulin treatment as well as regarding the reduction of diabetes related distress. The effects on knowledge and self-care behaviour are similar to those of the established education programme. Thus, PRIMAS provides a good alternative for treatment and education of type 1 diabetic patients.