Psychosocial Care for People with Diabetes. Deborah Young-Hyman

of FOH in individuals with T1D or insulin-treated T2D, as well as in parents of youth with T1D, can contribute to treatment behaviors that maintain higher BG levels (Cox 1987, Cox 1990, Wild 2007, Patton 2008). Survey data have also shown that FOH is the most common reason for not adjusting insulin doses, as hypoglycemia is the most feared complication of intensive insulin therapy (McCrimmon 1994, Reach 2005). FOH is also a primary barrier to exercise in patients with type 1 and type 2 diabetes (Brazeau 2008, Shahar 2008).

       Risk Factors for Hypoglycemia

      All hypoglycemic episodes in diabetes are caused by a surplus of insulin or other BG-lowering medications relative to food intake and physical activity. Physiologically, factors such as greater insulin sensitivity, higher BG variability, and impaired renal function increase the risk for hypoglycemia, as do some medications that delay gastric emptying (Chelliah 2004, Workgroup on Hypoglycemia ADA 2005, Honkasalo 2011). Numerous studies have documented that glycemic control and variables that lower A1C (e.g., more intensive insulin therapy) result in increased risk for hypoglycemia in both youth and adults (Davis 1997, DCCT Research Group 1997, Allen 2001). Some studies have attempted to go beyond these broad relationships and identify the immediate, discrete precursors of hypoglycemic episodes. This research shows that diabetes management plays a critical role in hypoglycemia risk, with 75% or more episodes being behaviorally induced in people with both T1D and T2D. For youth and adults, the majority of episodes were attributed to reduced food intake, including missed meals, and increased physical activity (Bognetti 1997, Davis 1997, Murata 2005). A recent retrospective study of 84 children and adolescents found that most SH episodes occurred when meals (carbohydrates) were delayed, followed by intense physical activity and excessive insulin administration (Cosmescu 2008). Another retrospective study involving >1,300 adults with T1D and T2D identified daily exercise as a major risk factor for SH (Honkasalo 2011). The occurrence of unexpected vigorous activity has also been highlighted as a common precipitant (Clarke 1999).

      Clinical experience indicates that many patients have only a rudimentary understanding of the effects of food and exercise on BG levels. Clinicians cannot assume that patients and their caregivers know about more complex behavioral risk factors for hypoglycemia, such as eating foods high in fat content, which can result in a delayed and/or depressed glycemic response, or engaging in elongated sessions of moderate physical activity, which can utilize as much (or more) glucose than shorter bouts of intense exercise. Maladaptive decisions and behaviors leading to hypoglycemia can also occur in patients who are nonadherent or nonconscientious about their diabetes management, including those who may skip meals or snacks, or fail to count carbohydrate content when calculating meal boluses. Psychiatric conditions that result in disordered behavior can also greatly increase patient risk, especially disordered eating behavior characterized by undereating, or binge eating accompanied by purging (Young-Hyman 2010).

      For some patients, frequent and recurrent episodes of hypoglycemia may be related to emotional or cognitive problems that contribute to poor decision making and judgments, or risky behaviors in dealing with low BG. Patients may minimize or dismiss the potential seriousness of episodes, experience secondary gain through the care and concern of family members, or perceive delaying treatment as a form of “winning” in a power struggle with diabetes.

      Another understudied but important risk factor is patient cognitions: beliefs, attitudes, and judgments about hypoglycemia and its treatment. Clinically, it is not uncommon to encounter patients who believe they can function fine at very low BG levels (e.g., <54 mg/dl) or that hypoglycemia is a necessary part of having good diabetes control. Other beliefs and attitudes can lead patients to have lower personal glycemic thresholds for treating hypoglycemia or a tendency to delay treatment, which greatly increase risk.

      Some patients also engage in overtreatment of hyperglycemia, or what is sometimes called diabetes “micro-management,” often resulting in insulin overbolusing or “stacking.” These individuals may be highly conscientious in their diabetes management and highly motivated to avoid hyperglycemia and long-term complications. However, high anxiety about hyperglycemia can become maladaptive and lead to extreme behaviors to avoid high BG (e.g., increased insulin dosing), which may increase hypoglycemic episodes (Ritholtz 2008, Singh 2010).

      Episodes of SH occur when milder or more moderate levels of hypoglycemia are not recognized and treated quickly enough or adequately. The probability of SH occurrence has been described in a biopsychobehavioral model that attempts to integrate the complex physiological, psychological, and behavioral processes that determine risk level (Gonder-Frederick 1997b, Cox 1999). The first step of this model is the occurrence of mild or moderate hypoglycemia, which triggers certain physiological reactions, including hormonal counterregulation and neuroglycopenia. These, in turn, cause adrenergic and neuroglycopenic symptoms, followed by patient detection and recognition of these symptoms, which ideally leads to appropriate decision making and self-treatment, thereby avoiding SH.

      However, the potential for SH is not avoided when hypoglycemic symptoms are not produced, recognized, or treated in a timely manner, which often occurs when a patient has hypoglycemic-associated autonomic failure (HAAF). With HAAF, frequent low-BG episodes lead to defective counterregulatory (epinephrine) responses to hypoglycemia and reduced hypoglycemic awareness (HA) (Cryer 2002, Cryer 2004, Kubiak 2004). This can cause BG to drop lower during hypoglycemia and fail to produce warning symptoms until levels are very low, which greatly increases the probability for significant neuroglycopenia and mental confusion, possibly preventing timely self-treatment. The relationship between reduced HA and SH risk has been well documented in people with both T1D and T2D, and it has been estimated that up to 60% of SH episodes are not associated with warning symptoms (DCCT Research Group 1991, Cryer 2002, Henderson 2003). Approximately 25% of patients with T1D appear to have some degree of reduced HA (Hepburn 1990, Geddes 2007).

      Factors contributing to HAAF, reduced HA, and SH risk include frequent hypoglycemic episodes, intensive insulin regimens, lower A1C levels, longer diabetes duration, physical exercise, suppressed nocturnal counterregulation, and alcohol consumption (Avogaro 1989, Kerr 1990, DCCT Research Group 1991, Davis 1997, DCCT Research Group 1997, Rewers 2002, Cryer 2003, Banarer 2004, Bulsara 2004, Sandoval 2004, Workgroup on Hypoglycemia ADA 2005, Richardson 2005, Schultes 2007). Smoking is another risk factor for SH, although it is unclear at this point whether it is related to HAAF (Hirai 2007). Even in the absence of HAAF and reduced HA, patients may fail to detect early warning signs of hypoglycemia. Field studies have found that both adults and children with T1D may fail to recognize half or more of BG levels lower than 70 mg/dl (3.9 mmol/l) (Clarke 1995, Gonder-Frederick 2008). Another field study prospectively followed adults with T1D with and without a recent history of SH (Cox 1999) over a 6-month period. The study replicated the relationship between HAAF-related factors (e.g., reduced HA, frequent low BG ) and SH risk; however, individuals in the high-risk group also showed cognitive and behavioral differences. Specifically, individuals with SH history showed more cognitive impairment during hypoglycemia and a tendency to treat episodes with food rather than faster-acting carbohydrates.

      Parents have been shown to demonstrate poor ability to detect low BG in their children (Gonder-Frederick 2008). Factors that can contribute to parents’ and their children’s failure to detect hypoglycemia include attention and perceptual processes, inaccurate beliefs about low-BG symptoms, and misattribution of symptoms to non-BG-related causes. Emotional distress may also have a negative impact; for example, school-aged children with depressive symptoms show poorer hypoglycemia detection (Gonder-Frederick 2008).

      Age is a major factor in SH, with older people and younger children at highest risk (Bognetti 1997, Bulsara 2004, Chelliah 2004). Increased risk in the elderly often occurs due to impaired renal function and/or medications (Chelliah 2004), and the elderly in poor glycemic control have more frequent episodes (Munshi 2011). Adolescents are at higher risk than adults with T1D (DCCT Research Group 1994), and children under 6 years of age are at higher risk than those over 6 years old (Davis 2001). One prospective study has documented the relationship between psychiatric disorders and higher SH risk in older children (Rewers 2002), and another investigated surreptitious intentional insulin overdosing by adolescents with emotional disturbance and family problems (Boileau 2006). A recent study of over 1,000 adults with T1D and T2D found that depression was a risk factor for SH (Honkasalo 2011). In the elderly, cognitive