The LBGI is a validated percentage index of frequency and amplitude of glucose excursions below a conventional glucose threshold set at 6.25 mmol/L and allows the normalization of non-Gaussian distributions of both capillary and interstitial blood glucose measurements collected over a pre-determined time frame.

The HBGI is a validated percentage index of frequency and amplitude of glucose excursions over a conventional glucose threshold set at 6.25 mmol/L, as previously described.

The Beck Depression Inventory II is a 21-item questionnaire developed to assess the intensity of depression as defined by the Diagnostic and Statistical Manual of Mental Disorders-Fourth Edition (DSM-IV). The BDI-II items include cognitive symptoms such as sadness and pessimism, affective symptoms such as loss of pleasure and loss of interest, somatic symptoms such as loss of energy, appetite changes and changes in sleep pattern. The BDI-II is calculated by summing the ratings for the 21 items. The 4-point scale ranges from 0 to 3 for each item. The maximum total score is 63. The measurement of outcome variables will be conducted 6, 12 and 24 months after transition to CSII.

The difference of the BDI-II scores among baseline, at 6,12 and 24 months will be also calculated.

The Beck Anxiety Inventory is a 21- item questionnaire developed to assess the severity of anxiety by describing its emotional, physiological and cognitive symptoms. The 4-point scale of BAI score ranges from 0 to 3 for each item, with a maximum total score of 63. Outcome variables will be evaluated at 6, 12 and 24 months after transition to CSII. The difference of the BAI scores among baseline, 6, 12 and 24 months will be also calculated.

The Multidimensional Diabetes Questionnaire (MDQ) was designed to provide a comprehensive assessment of diabetes-related cognitive and social factors. It includes 41 items subcategorized in three sections. The first section assesses the general perceptions of diabetes and related social support. It is comprised of three scales. These scales targets:

perceived interference caused by diabetes to daily activities, work and social activities (9 items); perceived severity of diabetes (3 items); perceived diabetes-related social support from a significant other such as family, friends and health professionals (4 items).

Responses were rated on 7-point rating scales (from 0 to 6), with higher scores indicating higher levels of perceived interference, social support and severity.

This section of the MDQ measures social incentives in relation to self-care activities. The scales included in this section were designed to measure the frequency of positive reinforcing behaviour (8 items) and the frequency of non supportive behaviours (4 items). Subjects recorded their responses on 7-point Likert scales (from 0 to 6) with higher scores indicating higher levels of positive and negative reinforcement behaviours.

This section of the MDQ is used to assess self-efficacy expectancies (7 items) and outcome expectancies (6 items). The self-efficacy scale measured patients' confidence in their ability to perform behaviors specific to the diabetes self-care activities. Responses were rated on a 0 to 100 rating scale. The outcome expectancies scale assessed patients' perceptions of the effect of diabetes self-care behaviors on metabolic control. Responses were rated on a 0 to 100 rating scale.

The multidimensional Health Locus of Control Scale consists of three subscale with six items each. The three subscales are: Internal Health Locus of Control (IHLC), Powerful Others Externality (PHLC), Chance Health Locus of Control (CHLC).

Coping strategies will be measured with the Coping Orientation to Problems Experienced-New Italian Version (COPE-NVI). The instrument is a multidimensional 60-items coping inventory to assess the different ways in which people respond to stress and it was developed to assess a broad range of coping responses.