Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

1 The History of Attachment Theory in Medicine

Attachment theory provides a biopsychosocial model to explain how individual differences in experience and behavior are related to interpersonal proximity and distance as well as to the regulation of affect and stress (Bowlby 1977). Whereas Bowlby’s original intention was to develop a theory for the assessment and treatment of emotional disorders, he was clearly disappointed that his work – for a long time – did not take root in the clinical realm of adult psychoanalysis and psychotherapy (Sandler 2011). Instead, his ideas were assimilated into developmental psychology, leading to numerous studies that validated concepts from attachment theory such as Mary Ainsworth’s investigations of infant attachment patterns using the Strange Situation (Ainsworth et al. 1978). These studies provide a solid basis for the extension of our knowledge of human development as well as its influence on personality and psychopathology (Bowlby 1988).

Although attachment theory was initially resisted by the psychoanalytic community, later psychoanalytically oriented authors in the United Kingdom (e.g., Holmes 1994) and Germany (e.g., Köhler 1991) reintegrated attachment theory and the results of developmental psychology research into clinical applications. Since the early 1990s, many psychotherapy research studies based upon attachment theory have been conducted which show, for example, that attachment characteristics and attachment-related interpersonal expectations predict both the quality of the therapeutic alliance (Diener and Monroe 2011) and the outcome of psychotherapy (Levy et al. 2011). There is also evidence that attachment security increases during psychotherapy (Taylor et al. 2015) and that the attachment patterns of therapists might play an important role in establishing therapeutic rapport (e.g., Schauenburg et al. 2010).

Since the attachment system plays an important role in the regulation of both the emotions and the interpersonal patterns activated in stressful situations, the concept has consequently been applied to the fields of psychosomatic medicine (Maunder and Hunter 2001) and health psychology (Pietromonaco et al. 2013), reflecting a “return” of the theory into the medical field.

Over the past years, the importance of attachment theory has grown immensely in many fields of medical care. Table 4.1 lists a selection of issues related to medical illness where attachment is believed to be important.

Table 4.1 Examples of how attachment theory helps health-care practitioners
Full size table

Attachment theory has increasingly been applied to understand the development of disease and chronic illness, the behavior of patients in medical care (Maunder and Hunter 2001), and the doctor-patient relationship. Attachment theory has been used to understand the behavior of patients with chronic illness (e.g., Sirois and Gick 2014), pain (Meredith et al. 2008; Meredith 2013; Costa-Martins et al. 2014), and cancer (Hunter et al. 2006; Rodin et al. 2007; Lo et al. 2010; Henselmans et al. 2010; Nicholls et al. 2014) and patients who depend heavily on medical providers, such as in intensive or palliative care (Petersen and Koehler 2006).

Researchers found evidence in diabetic patients that insecure attachment was associated with poorer diabetes self-management (e.g., lower adherence to recommendations related to oral hypoglycemic medications, diet, exercise, foot care, and smoking) and negative outcomes (elevated glycosylated hemoglobin levels, Ciechanowski et al. 2004a, b). In our own studies, we also found a connection between insecure attachment and low self-management skills and behavior in patients with diabetes. Attachment anxiety was significantly linked to impaired coping and lower self-efficacy, hope, dietary control, and physical activity. Attachment avoidance, on the other hand, was associated with lower levels of social support and health-care use (Brenk-Franz et al. 2015).

Similarly, in the context of chronic pain, insecure adult attachment was clearly connected to disability levels (McWilliams et al. 2000) and depressive symptoms (Ciechanowski et al. 2003; Meredith et al. 2007). In the National Comorbidity Replication Survey, McWilliams and Bailey (2010) found that a wide range of health conditions, including several cardiovascular disorders (i.e., stroke, heart attack, high blood pressure), were associated with anxious attachment, whereas secure attachment was unrelated to the health conditions (McWilliams and Bailey 2010). These examples indicate how promising an attachment perspective might be to improve medical treatment.

Two major approaches to adult attachment have guided research in attachment in medicine (Bartholomew and Shaver 1998). The developmental approach has mainly used the Adult Attachment Interview and derived measures to infer states of mind regarding childhood experiences with relevant caregivers (Main et al. 1985). This approach usually leads to a categorical classification of attachment. A second approach was mainly developed within social psychology and personality research and commonly relies on self-report measures of attachment and related thoughts and feelings in adult relationships measuring the degree of attachment anxiety and avoidance (Brennan et al. 1998). Most health-related research regarding attachment relied on self-report measures of adult attachment, involving ratings of particular attachment styles (Hazan and Shaver 1987) or scales assessing the attachment dimensions (Bartholomew and Horowitz 1991) that are thought to underlie attachment styles.

2 Activation of the Attachment System with Illness and Disease

Based upon attachment theory, one can assume that a person’s disease and the activation of the attachment system influence each other. On the one hand, suffering from a disease can be perceived as threat by the patient and causes him or her to activate the attachment behavior system (Shaver and Mikulincer 2004; Bowlby 1977, 1988). On the other hand, individually formed inner working models of attachment should have both direct and indirect influences on the health conditions, including the way health care is used (Box 4.1).

Box 4.1: The Relevance of Attachment for Medical Illness

  • Activation of the attachment system with illness and medical treatment

  • Illness and physical/psychological impairment as a consequence of insecure attachment

  • Attachment as a determinant of coping and illness behavior

  • Attachment as a determinant of self-management

  • Attachment as a predictor of adherence

  • Attachment as an important factor forming the doctor-patient relationship

Strategies of affect regulation and their association with the attachment system in patients with disease can be analyzed according to the model of Shaver and Mikulincer (2004). According to the model, being confronted with a disease can be interpreted as a threat, which activates the attachment system and, in turn, may lead to seeking proximity to an attachment figure. A person with secure attachment may find it sufficiently soothing to activate internal resources (“proximity to an internalized attachment figure”), whereas a person with dismissing attachment may disavow his or her need for contact with others. If the internal or external attachment figure is available and responsive, then access to either the caregiver’s attention and/or security-based self-representations can lead to deactivation of the attachment system. If this is not the case, the attachment system remains activated, and, depending on the internal working model, deactivating or hyperactivating strategies will be employed (Boxes 4.3 and 4.4).

Box 4.2: Examples of Patient Characteristics as a Function of Secure Attachment Style

Patients with secure attachment…

  • Develop higher self-esteem and self-acceptance (Bartholomew and Horowitz 1991)

  • Have a higher level of self-disclosure (Mikulincer and Nachshon 1991)

  • Feel more comfortable and flexible in interpersonal relationships (Ciechanowski and Katon 2006; Mikulincer and Nachshon 1991)

  • Assess personal resources and needs more realistically (Ciechanowski and Katon 2006)

  • Use social support systems and benefit from these and seek actively help (Florian et al. 1995; Li et al. 2008; Mikulincer and Florian 1997; Ognibene and Collins 1998)

  • Use more flexible coping strategies (i.e., active problem solving, accepting the situation, acquiring emotional support) (Schmidt et al. 2002; Mikulincer and Florian 1998)

  • Are more cooperative, optimistic, confident, reliable, and understanding (Klohnen and John 1998)

  • Are positively perceived in groups and perceive other persons to be more differentiated, benefit more from therapies, and develop closer working relationships (Strauß and Schwark 2007)

  • Have a higher treatment acceptance in the doctor-patient relationship (Dozier 1990)

Box 4.3: Examples of Patient Characteristics as a Function of Dismissing Attachment Style

Patients with dismissing attachment…

  • Are characterized by the pursuit of autonomy and independence, more interpersonal distance, and control (Allen et al. 2005)

  • Distrust others; expect them to be hostile, exploitative, andinsensitive (Ciechanowski and Katon 2006; Dozier et al. 1994)

  • Report lower self-disclosure (Mikulincer and Nachshon 1991)

  • Develop uncooperative working relationships and rarely seek social support (Hesse 2008; Mikulincer and Shaver 2007; Ognibene and Collins 1998)

  • Use more rigid coping mechanisms,prefer cognitive distancing of emotions (Feeney 1995; Lopez et al. 2001)

  • Respond to anger, irritation, or frustration with interpersonal distance (Jellema 2002; Rholes et al. 1998)

  • Are more rational, autonomous, emotionally independent, and sarcastic (Klohnen and John 1998)

  • Tend to trivialize their problems, emotions, and symptoms (Jellema 2002; Strauß and Schwark 2007)

Deactivating strategies are commonly associated with the denial of attachment needs, downplaying of risks in terms of trivializing symptoms of diseases, repression of negative emotions and cognitions, as well as the avoidance of contact with an attachment figure. In contrast, hyperactivating strategies probably increase sensitivity to threatening situations. A chronically activated attachment system might lead a patient to continuously perceive danger in his or her environment, so symptoms might be dramatized and exaggerated. As a consequence, there is a strong demand for a sense of protection and security from potential attachment figures, which may include not only partners but also family members and health-care providers such as physicians and nurses. It is assumed that these processes happen automatically, i.e., a patient is not fully aware of them and cannot really reflect upon their actions.

3 Illness and Physical/Psychological Impairment as a Consequence of Insecure Attachment

Within medical care, deactivating and hyperactivating strategies of patients with insecure attachment are of paramount interest for understanding health behavior. It is also of interest to understand how insecure attachment correlates with the formation and maintenance of disease. Based on empirical evidence, Maunder and Hunter (2001) (see Fig. 1.2) developed a model of the association between insecure attachment and heightened risk for disease, based on three different paths. The first path describes the relationship between insecure attachment and various impaired mechanisms of stress regulation, which can have a direct influence on the formation and maintenance of disease caused by an increase in physiological stress response. A second path shows the association between insecure attachment and increased externalization of emotional regulationrelated to problematic health behavior and the role of specific risk factors such as substance abuse (nicotine, alcohol), disturbed eating behavior, or risky sexual behavior. The third path, indicates that insecure attachment is associated with an inadequate use of protective factors such as social support, adherence, and self-care of patients. This has a direct effect on the disease as well as an indirect effect through suboptimal use of the medical care system and an inadequate description of disease-related symptoms.

Insecure attachment has clearly been shown to be a risk factor for the development of mental and physical disorders, whereas secure attachment works as a protective factor associated with greater life satisfaction (Maunder and Hunter 2001; Thompson 1999; Wensauer and Grossmann 1998). Insecure attachment is also seen as an important risk factor for the development of chronic diseases, such as chronic pain, stroke, heart attack, high blood pressure, and ulcer disease (McWilliams and Bailey 2010). While it is estimated that the percentage of securely attached individuals in representative samples might fluctuate between 50 and 60 %, the percentage in clinical samples, especially those of patients with psychological disorders, is approximately 20 % (Bakermans-Kranenburg and Van Ijzendoorn 1993, 2009), underlining the risk potential of attachment insecurity.

4 Attachment as a Determinant of Coping

Theories of coping refer to both treatment-related and intrapsychic efforts to tolerate or minimize stressful situations (Lazarus and Launier 1978). Several studies have demonstrated that patients’ management of diseases depends on their attachment characteristics (Mikulincer and Florian 1998; Schmidt et al. 2002; Turan et al. 2003). In particular, attachment-based patterns of affect regulation influence patients’ coping behavior. Patients with secure attachment assess stressful situations more flexibly, solve problems more actively, and use their social network (Seiffge-Krenke 2004).

Patients with avoidant attachment tend to suppress emotions, such as fear or anger, have a need to exert control, tend not to seek social support (e.g., Kotler et al. 1994), tend to show avoidance and passive resignation (Turan et al. 2003), and generally employ threat-reducing, repressive, and deactivating coping strategies (Schmidt et al. 2002). In keeping with these strategies, avoidant patients seek less medical help and avoid contact with their physicians (Brenk-Franz et al. 2015, Mikail et al. 1994).

In contrast, patients with preoccupied attachment focus more on the threatening aspects of their disease. They report more negative emotions and disease-related symptoms and tend to catastrophize (Ciechanowski et al. 2003). They also employ more diverting strategies and more negative emotional coping (Schmidt et al. 2002). In self-report, preoccupied patients also revealed hyperactivating tendencies in their coping behavior (Box 4.4).

Box 4.4: Examples of Patient Characteristics as a Function of Preoccupied Attachment Style

Patients with preoccupied attachment …

  • Show their stress or anger excessively and vigorously seek social support (Mikulincer and Shaver 2007)

  • Have more unbalanced relationships, because the need for support outweighsthe possibility of giving social support (George and West 2001)

  • Do not feel resilient and resistant (Meredith et al. 2005)

  • Are more demanding, unstable, dependent, moody, and frustrated (Klohnen and John 1998)

  • Tend to catastrophize the description of symptoms (Ciechanowski et al. 2003)

  • Develop strong dependent bonds to the doctor/therapist (Strauss and Schwark 2007)

  • Report their situation with ambivalent and diffuse language (Hesse 2008)

  • Want more intense and frequent contact with the therapist, test boundaries, and keep their therapists involved (Strauss and Schwark 2007)

5 Attachment as a Determinant of Self-Management

Self-care and self-management are increasingly important in medical care as patients become more involved in the treatment of their diseases. Health-promoting behaviors are taught to patients, on the assumption that by supporting empowerment and shared decision-making, patients will be more active and autonomous within the medical care system (Kanfer et al. 2006). Self-management as a behavioral and cognitive strategy helps people to structure their behavior and to achieve their goals (König and Kleinmann 2006). It can therefore be regarded as an expression of the self as a representational agent, which is closely related to secure attachment (Fonagy et al. 2004). Self-efficacy is a key component of cognitive self-management (Bandura 1977). It describes the expectation that one will be able to successfully perform a specific behavior (Schwarzer 2002). Self-management strategies are essential elements of evidence-based medical treatment of patients with chronic diseases in primary care (Wagner et al. 2001). Self-management programs have shown clear benefit for patients with diabetes mellitus (Cochran and Conn 2008; Duke et al. 2009), coronary heart disease (Barth et al. 2006), arterial hypertension (Glynn et al. 2010), and depression (Gensichen et al. 2011; Khan et al. 2007).

Some recent studies from primary care have explored the relationship between attachment and self-management. For example,diabetic patients with avoidant attachment showed less behavioral self-care (such as regular foot care, adherence to diet, avoiding smoking, adherence to medical health-care use) than patients who are securely attached (Brenk-Franz et al. 2015; Ciechanowski et al. 2004a).

6 Attachment as a Predictor of Treatment Adherence

Treatment adherence and nonadherence are another important focus. Only half of patients with chronic disease take their medication as prescribed (WHO 2003) even though strong adherence leads to better treatment outcomes (Simpson et al. 2006). However, adherence and nonadherence are subject to different influences. Factors which promote adherence include self-efficacy and the presence of symptoms (Dunbar-Jacob and Mortimer-Stephens 2001). Some of the factors known to influence nonadherence are poor education, unwanted adverse reactions, polypharmacy, and the presence of chronic diseases (Claxton et al. 2001; Hernandez-Ronquillo et al. 2003).

Concepts of developmental psychology have only recently been used to explain treatment nonadherence. Secure attachment is associated with more health-promoting behavior (Scheidt and Waller 2002) and various foci of self-efficacy and so is expected to promote treatment adherence. With respect to nonadherence, in diabetes, dismissing attachment is associated with low overall adherence and leads to poorer glucose control, especially if the communication with the physician is subjectively perceived as unsatisfactory (Ciechanowski et al. 2001). In lupus patients, a study tested if patients’ attachment styles could predict patients’ adherence and health-related quality of life. Attachment avoidance was again seen to have a negative effect on a patient’s adherence, whereas attachment anxiety was seen to have a negative impact on health-related quality of life (Bennett et al. 2011). Attachment avoidance has also been associated with not using seat belts when driving (Ahrens et al. 2012). Thus, the avoidant dimension of attachment insecurity is consistently associated with nonadherence.

7 Attachment as an Important Influence on the Doctor-Patient Relationship

Evidence shows that psychotherapists have the potential to act as an attachment figure by providing a patient a “secure base” and “safe haven” (Borelli and David 2004). Dozier and Bates (2004) indicated that “The client finds in the therapist someone who seems stronger and wiser than him- or herself. Thus, the client may interact with the clinician in ways that reflect expectations from other relationships,” a notion that could easily be extended to the field of medical treatment. Accordingly, attachment theory may provide a model for explaining why some patients have an intense need for their doctor. This would occur when a fundamental need for security, initiated by the threat posed by an illness, drives the patient to regard the physician as an attachment figure, with whom proximity needs to be maintained in order to feel safe.

Supporting this hypothesis is the fact that patients prefer continuous care by a single primary care physician (e.g., Pandhi and Saultz 2006) and continue to prefer continuity of care even when the severity of illness is increasing (Baker et al. 2007; Frederiksen et al. 2010; Guthrie and Wyke 2006). Patients commonly report problems when they have to change their primary care physician, even if the relationship was difficult (Frederiksen et al. 2010). Thus, the evidence suggests that continuity is desirable, and discontinuity problematic, for most patients. A continuous care provider may be well situated to act as a safe haven and a secure base with respect to threats related to health.

Studies in health-care research show correlations between patients’ attachment characteristics and the way they present themselves in the health-care system. Early attachment relationships influence behavior in all important relationships throughout life. Therefore, the doctor-patient relationship is likely to be affected by the deficits in social competencies and skills and deficits in perceived social support, which are common among patients with insecure attachment styles (Mallinckrodt 2000). Dysfunctional illness behavior can be understood as a result of such problems (Ciechanowski et al. 2002). Patients with preoccupied attachment, for example, express more attention-seeking behavior, report more symptoms, and overuse health services (Ciechanowski et al. 2002), whereas those with avoidant attachment are more likely to reject a practitioner, self-disclose less, underuse care services, and avoid regular contact with their physicians (Dozier 1990; Feeney and Ryan 1994; Brenk-Franz et al. 2015).

So far, we have concentrated on the attachment of the patient, but in any relationship there is a mutual influence between the individuals – what can we say about the attachment of the health-care worker? Unfortunately, studies to determine the attachment behavior of the treating physicians and their influence on the doctor-patient relationship are still rare. First results indicate that physicians who focus on a holistic treatment of patients often have characteristics of secure attachment and that medical students with more secure attachment often opt for primary care medicine and medical disciplines allowing more stable and long-lasting doctor-patient relationships (Ciechanowski et al. 2004b, 2006). There are also indications that securely attached case managers have a better understanding of the “hidden needs” of their patients, while case managers with insecure attachment primarily respond to the visible needs of their patients (Dozier et al. 1994). Physicians with a positive mental model of the self are more willing to go against their patients’ opinions and wishes for certain treatment (Salmon et al. 2007, 2008). Overall, it is clear that the attachment styles of physician and patient influence each other. Prospectively, there should be a stronger focus on the issue of doctor-patient fit and on the question how the doctor-patient relationship might change for the better over time (Salmon and Young 2009).

8 Conclusions: Benefits and Tasks of Attachment Classification in Medical Care

As shown in this introductory chapter, attachment theory – after its return into the medical world – has gained considerable interest in medical and psychosomatic research.

Attachment theory provides an explanatory model for different strategies of emotion regulation, coping, and the use of the health-care system. It is still not clear how the mechanisms related to different attachment styles influence health-related long-term outcomes. There is strong evidence that dismissing patients tend to suppress emotions, whereas hypervigilance and rumination are common characteristics of preoccupied patients. Both strategies result (mostly indirectly) in chronic health problems (e.g., Maunder and Hunter 2001; Mikulincer and Shaver 2007). There is clearly a need for long-term studies exploring in which ways attachment characteristics act as predictors of specific medical outcomes. Moreover, their underlying biological pathways should be further clarified (Miller et al. 2009).

For the future, it is also important to consider which interventions might strengthen secure attachment and improve basic mechanisms of affect regulation, enhancing pro-social behavior and improving long-term health-related factors (Simpson and Rholes 2010). So far, there is some evidence showing that crucial aspects of patients’ illness behavior, coping, and adherence as well as their self-management in face of (chronic) diseases relate to their attachment history. These results clearly have the potential to guide the development of strategies for improving patient treatment in primary care.