Disability is defined by the International Classification of Functioning, Disability and Health (ICIDH-2) as an umbrella term for “impairments, activity limitations or participation restrictions” [22]. Despite declining trends in chronic disability among US adults aged 65 and older, approximately one-fifth were disabled in 2005 [23]. The high prevalence of physical disability among older adults is mainly attributed to chronic diseases [24]. Depression and disability mutually reinforce the risk of each other, and adversely affect disease progression and prognosis [21, 25]. On the one hand, disability caused by medical conditions serves as a risk factor for depression [26]. When people lose their normal sensory, motor, cognitive, social, or executive functions, especially in a short period of time, they can become very frustrated or depressed. Inability to perform daily tasks as before decreases self-esteem, reduces independence, increases the level of psychological stress, and creates a sense of hopelessness. On the other hand, depression increases the risk for disability. Negative interpretation, attention bias, and learned hopelessness of depressed persons may increase risky health behaviors that exacerbate physical disorders or disability. Meanwhile, depression-related cognitive impairment also affects role performance and leads to functional disability [25]. For example, Egede [27] found in the 1999 National Health Interview Survey that the risk of having functional disability among patients with the comorbidity of diabetes and depression were approximately 2.5–5 times higher than those with either depression or diabetes alone.

A leading cause of disability among medical patients is pain and pain-related fears [28]. Pain is an unpleasant perception caused by diseases or injuries, which is particularly common among the elderly, especially institutionalized older adults [29]. The main medical reasons include back injury, arthritis, and cancer [30, 31]. Although a large proportion of pain complaints can be attributed to physiological changes from physical disorders, psychological factors (e.g., attention, interpretation, and coping skills) play an important role in perception of pain, as indicated in Fig. 4.1 [32]. Individuals with cognitive bias and poor coping strategies report more intense and persistent pain, as well as depressive symptoms [31, 33]. Bair et al. [31] indicated in a literature review that the prevalence of pain was higher among depressed patients than non-depressed patients, and the prevalence of major depression was also higher among pain patients comparing to those without pain complaints. The comorbidity of depression and pain may be explained by shared biological pathways [31]. Decreases in neurotransmitters among depressed patients (such as serotonin) amplify nociceptive signals and, therefore, increase the frequency and severity of pain symptoms [31]. Consistently, over half of the chief complaints from depression patients in primary care were somatic symptoms, mainly pain complaints, which impose difficulties for primary care physicians in recognizing mild or moderate depression and in preventing the occurrence of major depression [31].

While disability and pain are sources of stress, so is being diagnosed with a medical condition. Being diagnosed with diseases and undergoing treatments increase patients’ psychological stress, especially for those who do not have positive perceptions, effective coping strategies or adequate social support [34]. According to the diathesis stress model, major depression will occur once the magnitude of perceived stress increases to an extent at which one’s threshold for the disorder is reached [35]. Meanwhile, major depression is found to be associated with stress-induced activation of the inflammatory response (Fig. 4.1) that produces a series of physiological and psychological effects [36]. Immune activation releases cytokines (e.g., TNFα and IL-1β) in the brain that may mediate the link between medical conditions and depressive symptoms [36]. Hormonal response is another important stress-related biological pathway that links physical disorders and depression [37]. Maladaptation in stress response may cause the dysregulation of the hypothalamic-pituitary-adrenal (HPA) axis, characterized by elevated levels of cortisol and corticotropin-releasing hormone (CRH) [37]. Increased cortisol level is not only associated with stress and emotional disturbance but also is implicated in progression of a variety of medical conditions, such as cardiovascular disease, type 2 diabetes mellitus, ulcers, and loss of bone density [37, 38].

Risk behaviors (e.g., smoking, alcohol and drug abuse, poor adherence to treatment, malnutrition, and physical inactivity) could be maladaptive coping responses to stress, or unhealthy lifestyles continued from earlier life stages. For example, physical inactivity, as a consequence of depression, is also associated with increased risk for coronary heart disease, diabetes, stroke, and cancer [39–42]. Mattson [43] indicated in a review that physical activity activates adaptive stress responses in neurons, produces protective factors such as brain-derived neurotrophic factor (BDNF) that increase the brain’s capacity to buffer stress, and therefore decreases the risk for depression as well as other brain disorders among older adults. Findings from the Alameda County Study also supported that physical activity was a protective factor for subsequent depression among adults aged 50 and over [44]. People who participated in physical activities were more likely to have less risky health behaviors, such as smoking and overdrinking [44]. Alcohol use has more serious adverse health effects on older adults than other age groups, since aging-related physiological changes (e.g. reduced liver detoxification and renal clearance) affect alcohol metabolism, increase the blood concentration of alcohol, and magnify negative consequences. More importantly, alcohol interacts with a variety of frequently prescribed medications potentially influencing both treatment and adverse effects.

Loneliness is the perceived isolation from one’s surrounding relationships. According to the Health and Retirement Study, approximately one-fifth of the US adults aged 65 and older experienced loneliness [45]. It is considered as an independent risk factor for depression [46, 47], and has been demonstrated to be associated with low physical activity, increased cardiovascular risks, hyperactivity of the hypothalamic-pituitary-adrenal axis, and activation of immune response [48–51]. Physical disorders and aging-related functional decline, such as hearing loss and vision impairments, limit individuals’ mobility and capability for social involvement to different extents. A cross-sectional study conducted in the Finnish older population indicated that poor health conditions and functioning impairments were associated with higher risk for loneliness, and medical illness was also reported as one of the most common causes of loneliness [52]. Older adults may view loneliness as an important factor in the expression of depression [53].

Hopelessness is a key concept of major depression [54], and also an independent risk factor of suicidal ideation [55]. According to the hopelessness theory of depression, with the presence of negative life events (e.g., experiencing physical disorders), depression-prone individuals tend to process information with bias, attribute these events to global, constant, and internal causes, and make adverse inferences [56]. Hopelessness is also conceptualized as a distinct concept independent of depression, acting as a mediator of the association between physical disorders and depression [57]. Elevated hopelessness was found to be associated with the onset of myocardial infarction, cancer, and hypertension, and with mortality [58, 59]. More commonly, hopelessness can be a consequence of physical disorders among medical patients, especially those with terminal illness and life-threatening conditions [57]. Hopelessness reduces expectations for the future, and negatively affects judgment for making medical and behavioral decisions, including non-adherence to medical regimens or engaging in unhealthy behaviors.

Due to age-related changes in pharmacokinetics and pharmacodynamics, older adults are a vulnerable population to iatrogenic diseases caused by adverse drug effects. Drugs for treating medical conditions could affect neurotransmitters in the brain through biological pathways, and also may have a psychological impact on patients by causing somatic symptoms [60]. Celano et al. [60] summarize major categories of medications that have the potential to induce depression among medically ill patients, including neurologic medications, cardiovascular medications, anti-infective agents, oncologic medications, and miscellaneous medications. Although cases of drug-induced depression were reported for a variety of drugs, few prospective studies with good internal validity were conducted to examine the true incidence [60]. Even so, physicians are recommended to evaluate medical patients’ risk for depression, avoid depressogenic drugs for patients with prior or current depression episodes, and monitor for drug-induced depression [60]. Similarly, antidepressants can also cause physical disorders among older adults, such as falls, hip fractures, weight gain, and obesity, which may in turn increase the severity of depressive symptoms [61–63]. Adverse drug events are frequently due to failure to adjust dosage or to account for drug–drug interactions in older adults [64].

Co-occurring depression and medical conditions are associated with more functional impairment and mortality than expected from the severity of the medical condition alone. For example, depression accompanying diabetes confers increased functional impairment [27], complications of diabetes [65, 66], and mortality [67–71]. Frasure-Smith and colleagues highlighted the prognostic importance of depression among persons who had sustained a myocardial infarction (MI), finding that depression was a significant predictor of mortality at both 6 and 18 months post MI [72, 73]. Subsequent follow-up studies have borne out the increased risk conferred by depression on the mortality of patients with cardiovascular disease [10, 74, 75]. Over the course of a 2-year follow-up interval, depression contributed as much to mortality as did myocardial infarction or diabetes, with the population attributable fraction of mortality due to depression approximately 13 % (similar to the attributable risk associated with heart attack at 11 % and diabetes at 9 %) [76]. Investigators seeking to understand the biological mechanisms linking depression and medical conditions have focused on cardiovascular, immunologic, inflammatory, metabolic, and neuroendocrine pathways [77, 78].

Several intervention studies (i.e., among samples of older adults with depression) have examined mortality as an outcome [79–82]. In ENRICHD (Enhancing Recovery in Coronary Heart Disease) taking a selective serotonin reuptake inhibitor was associated with reduction in the risk of all-cause mortality [83], as was participation in group plus individual therapy [84] in secondary analyses that ignored randomization. SADHART (Sertraline Antidepressant Heart Attack Randomized Trial) reported a statistically non-significant beneficial trend on combined cardiovascular outcomes that included death at 24 weeks [82]. PROSPECT (Prevention of Suicide in Primary Care Elderly Collaborative Trial) reported diminished mortality after 8 years of follow-up among older persons in practices randomized to a depression care management program [79].

Universal, selective, and indicated prevention strategies apply to mental health [85, 86]. Universal prevention strategies are intended to reach the entire population, without regard to individual risk factors. For example, every person, without screening, in a community or practice might be provided with prevention skills. We do not consider universal prevention strategies here, since persons with medical conditions are de facto subgroups of the population at increased risk. Selective prevention strategies target subgroups of the general population that are determined to be at risk, for example, persons who have a medical condition. Indicated prevention interventions identify individuals who are experiencing depressive symptoms, but do not meet criteria for major depression. This potentially includes many persons who have medical conditions, since depressive symptoms are so common.

We carried out a literature review to identify prevention intervention trials with a focus on adults with medical comorbidity. Included randomized controlled trials of prevention recruited non-depressed patients with at least one physical disorder. We did not incorporate trials primarily directed at treatment of existing major depression (e.g., in the context of diabetes [87] or cardiovascular disease [80]), prevention of depression recurrence (e.g., in the context of diabetes [88, 89]), or reduction in depressive symptoms among both depressed and at-risk patients (e.g., in the context of cardiovascular disease [90] or cancer [91, 92]). Ideally, we would have required investigators to exclude persons who met criteria for major depression since persons who have major depression are no longer at risk. Rather than using diagnostic criteria, investigators may have excluded persons whose depression scores were above a threshold (e.g., using the Hamilton Depression Rating Scale [93, 94], the Hospital Anxiety and Depression Scale-depression subscale [95], or the Geriatric Depression Scale [96]. While it is true that persons whose scores were above a threshold would be more likely to meet criteria for major depression, persons below the threshold could still meet criteria for minor depression. The implication is that we would call a study “indicated” for persons with minor depression, but “selective” for persons who do not have depression. Investigators sometimes did not report the information required to make this distinction. So we decided to categorize interventions in the table based on our best justification for each individual study. We did not include studies specifically tied to the prevention of depression associated with the administration of alpha-interferon or similar substances. We summarize prevention trials meeting our inclusion criteria in Table 4.1 according to whether we categorized as indicated or selective prevention (we did not find any studies we would consider universal prevention in association with medical conditions).