Assessments: Hand-held computer, HbA1, Psychosocial Questionnaire

Monthly Diaries: Severe hypoglycaemia, motor vehicle violations, missed work days Figure 7.9. Repeated baseline design for study of BGAT-II

extreme levels48, from the handheld computer data. A reduction in Low BG Index indicates that hypoglycaemic episodes have decreased in frequency and/or severity. After BGAT-II, the HA group showed a reduction in the Low BG Index, while the RHA group showed a reduction in the High BG Index. Thus, even though BGAT-II improved detection of hypoglycaemia, it did not appear to reduce the frequency and severity of low BG fluctuations in patients with RHA. The number of symptoms related to hypo- and hyperglycaemia did not increase after training, suggesting that patients were not identifying new symptoms but rather using pre-existing symptoms and external cues more effectively. Glycosylated haemoglobin did not change in either group. Importantly, there were no differences between the three different training sites, indicating that BGAT can be effectively generalized to other patient populations, geographic regions and trainers.

Even though BGAT-II improved detection of hypoglycaemia in RHA patients, its failure to decrease the frequency and severity of low BG excursions was somewhat disappointing. To address this problem, BGAT-II was revised to include training specifically targeted toward reducing the frequency of hypoglycaemia. This revision (BGAT-III) incorporated more information on the impact of frequent hypoglycaemia on counter-regulation and hormonal symptoms. Research was included that showed that avoidance of hypoglycaemia, even for only a few days to a week, can improve counter-regulation and strengthen symptoms49'50, and patients were given explicit instructions to try to reduce the frequency of SMBG readings < 3.9 mM. BGAT-III is currently being used by hundreds of patients in several different countries. However, only one controlled study testing its efficacy has been published51. Researchers at the Joslin Diabetes Center at Harvard University assessed BGAT in patients undergoing intensive insulin therapy to improve diabetes control which, as shown by the DCCT and subsequent studies, greatly increases the risk for severe hypoglycaemia52,53, presumably because the increase in frequency of low BG impairs counter-regulation. These researchers tested whether BGAT would reduce this risk. After intensive therapy training, half of the patients were assigned to BGAT-III and the other half were assigned to a control group who attended a class on cholesterol awareness. Before intensive therapy training and after BGAT, patients underwent a hypoglycaemic clamp study to assess symptoms, counter-regulation and ability to detect extreme BGs, then used the handheld computer for 70 home trials to assess symptoms, accuracy and glucose profiles.

The results of this study are exciting, albeit somewhat mixed. As expected, intensive insulin therapy improved metabolic control in both the BGAT and control groups. However, the frequency of low BG readings also increased equally in both groups, indicating that BGAT-III did not reduce this risk factor for severe hypoglycaemia. In terms of ability to recognize hypoglycaemia, BGAT patients did not show an improvement in accuracy compared to the control group during inpatient testing. However, when hand-held computer data were analysed, BGAT patients showed the expected improvements in detection of low glucose levels and had fewer undetected low BG readings than control patients. Perhaps most importantly, this study yielded the first evidence that BGAT can have an impact on physiologic response. Even though intensive insulin therapy improved metabolic control and increased frequency of low BG, BGAT patients did not show the expected reduction in epinephrine response to hypoglycaemia. In contrast, control patients showed the well-documented decrease and delay in counter-regulation. Because BGAT did not improve ability to reduce the frequency of low BG readings, the mechanism by which counter-regulatory integrity was preserved is unclear. Nonetheless, this study provides intriguing preliminary evidence that BGAT, through some indirect mechanism, can maintain the efficiency of hormonal counter-regulation during intensive therapy.

Another question of recent interest is the long-term effects of BGAT. In other words, do the benefits of BG persist over time? To test this question54, we recruited 28 BGAT patients and 12 control group patients from the first study37 an average of 4.9 years after their original participation. Half of the BGAT group received 'booster training', which involved keeping BG Awareness Diaries for 2 weeks prior to assessment of accuracy. The remaining BGAT patients, as well as the control group, received no booster training before accuracy assessment. Patients also had blood drawn for glycosylated haemoglobin measurement, and filled out a questionnaire asking them how many days of work they had missed due to hypoglycaemia or DKA, as well as how many automobile accidents they had had since their participation in the original study. BGAT + booster patients demonstrated significantly more accuracy in BG detection than either the BGAT + no booster or control patients, and BGAT + no booster showed a trend toward better accuracy than control patients. There were no group differences in lost work days, but BGAT patients reported significantly fewer automobile accidents than control patients. Only 15% of BGAT patients reported accidents, while 42% of control patients reported one or more. There were no differences in metabolic control.

More recently, we have analysed follow-up data from the multi-center study of BGAT-II reported previously42 (see Figure 7.9)*, which offers the most comprehensive evaluation of long-term effects to date, with hand-held computer assessments, prospective monthly diaries, and psychosocial measures 1 month, 6 months, and 1 year following training. Table 7.5 summarizes these results, which we find quite encouraging. Improvements in ability to estimate BG were maintained across the year for both HA and

^Manuscript in preparation.

Blood Glucose Awareness Training Table 7.5. BGAT II: 1-year follow-up results

Improvements maintained

Improvements not maintained

No improvement

Accuracy AI score

Increased A zone estimates Decreased D zone estimates Low BG detection

Decision making To treat when actual BG < 70 Not to drive when actual BG is < 70

Not to drive when estimated BG < 70

Risk factors High BG risk index Number of severe hypoglycaemia episodes Number of nocturnal hypoglycaemia episodes Number of automobile violations

Psychosocial Diabetes quality of life Diabetes knowledge


High BG detection

To treat when estimated BG is < 70

Low BG risk index Daytime severe hypoglycaemia

Dyadic adjustment scale Beck Depression Inventory Glycosylated haemoglobin

RHA patients on nearly every measurement of accuracy—AI scores, A zone estimates, D zone estimates, and percentage low BG episodes detected. Improvements in detection of high BG, however, were not maintained. BGAT did not reduce the Low BG Index; however, there were significant decreases in severe hypoglycaemia, nocturnal hypoglycaemia, and traffic violations after intervention. One perplexing finding, for which we have no explanation, was a decrease in the number of symptoms related to hypogly-caemia over time.

This study also provides the first evidence that BGAT has beneficial effects on decision-making and judgement. Hand-held computer data showed that decisions to self-treat actual BG levels < 3.9 mM increased in frequency, as did decisions not to drive when BG was low. RHA and HA patients differed, however, in changes in decision-making when BG was believed to be low (estimated BG < 3.9 mM). Decisions to self-treat when BG was estimated to be low increased in RHA patients but decreased in HA patients. This may reflect increased awareness in RHA patients that, whenever they feel symptoms, they are likely to be quite hypoglycaemic and in need of immediate treatment. In contrast, HA patients may learn that they cannot always trust subjective estimations of BG and, consequently, that they need to verify these with SMBG before taking action. The frequency of decisions not to drive when BG was believed to be low increased in HA patients but, unfortunately, did not change in RHA patients.

Taken together, these studies provide encouraging evidence that BGAT has benefits beyond simply improving BG detection, such as improvements in decision-making and reduced risk for severe hypoglycaemia and motor vehicle violations, and that many of these benefits are maintained over time, although some sort of booster training may be needed to preserve them over several years. Furthermore, these studies provide preliminary evidence that BGAT may improve counter-regulation and reduce the risk of severe hypoglycaemia without jeopardizing metabolic control. However, there remains 'room for improvement' to optimize the effects of this intervention. As yet, we have not been able to reduce the frequency of low BG excursions, a major risk factor for future severe hypoglycaemia, or the frequency of daytime episodes of severe hypoglycaemia. The frequency of low BG and severe hypoglycaemia depends, in large part, on the decisions patients make and the actions they take to manage their diabetes. For example, deciding to delay a meal can precipitate low BG, while deciding to delay treatment of low BG can lead to severe hypoglycaemia. Even though BGAT reduced risky decisions about treatment and driving when patients knew they were hypoglycaemic, RHA patients did not become less willing to drive when they believed their BG might be low. Unfortunately, the handheld computer program used in these studies did not assess whether or not patients planned to measure their BG to verify their subjectively perceived hypoglycaemia. This methodological limitation, which has been corrected in our current research, restricted our ability to make definitive conclusions about self-treatment decisions when BG is believed to be low. Nonetheless, these findings provide empirical evidence that decision-making and behavioural response play an important role in reducing the risk of hypoglycaemia and its negative consequences, and that these areas may be important targets for future intervention.

Diabetes 2

Diabetes 2

Diabetes is a disease that affects the way your body uses food. Normally, your body converts sugars, starches and other foods into a form of sugar called glucose. Your body uses glucose for fuel. The cells receive the glucose through the bloodstream. They then use insulin a hormone made by the pancreas to absorb the glucose, convert it into energy, and either use it or store it for later use. Learn more...

Get My Free Ebook

Post a comment