Journal of Gerontology 1976. Vol. 31, No. 3, 324-326
Aging and Depression: Some Unanswered Questions Lissy F. Jarvik, MD, PhD :
i CT TE is a fool that is not melancholy -tJ-once a day," states an old Russian proverb, reminding us of the first unanswered question about aging and depression: Where do we draw the line between depression and sadness, between depressive illness and the melancholy mood which is an appropriate response to life's stresses? While this question pertains to depression at any age, it is particularly cogent in old age where decline in physical strength, in health, in mental alertness, in earning capacity, in prestige at home and in society combine with economic deprivation, the loss of significant others, social isolation, and the ever more closely approaching prospect of self-death to provide adequate cause for dejection, despair, and despondency. As Epstein (this issue) has pointed out: "Depression has almost been viewed as a characteristic of senescence." In an elderly person even "seemingly obvious symptoms" may not result in a diagnosis of depressive illness; instead they may be ascribed to the aging process as such. Without a clear-cut universally accepted definition, it is not surprising that prevalent figures vary from survey to survey as do reviewers' conclusions. Thus, Busse& Pfeiffer (1973) are quoted by Epstein (this issue) as citing a range from 10% to 65%, while 'Revision of presentation made at the Symposium on Age Differentiation in Depressive Illness (J. Zubin, PhD, Chairman; Lissy F. Jarvik, MD, PhD, Organizer) at the 25th Annual Scientific Meeting of Gerontological Society, San Juan, Puerto Rico, Dec. 18, 1972. 'Veterans Administrations Hospital, Brentwood, and UCLA Dept. of Psychiatry (for reprints: 760 Westwood Plaza, Los Angeles 90024).
324
Gurland (this issue) considers the most likely prevalence to be less than 5% for the age group above 65 years using psychiatric diagnoses rather than symptom checklists as criterion measures. Klerman (this issue) estimates that 8 10% of men and 16 - 29% of women will develop depression sometime during their lives, while according to Essen-Moller & Hagnell (1961) the cumulative risks to age 80 in a population of 2550 (ascertained and examined psychiatrically with 99% completeness) were 17.7% and 8.5% for men and women, respectively. The uncertainty regarding its prevalence is paralleled by the confusion concerning types of depressive illnesses. While there are few advocates today of the position that the diagnosis of depression connotes a homogeneous entity (differences in severity accounting for apparent clinical subgroups), there is as yet no satisfactory scheme for the systematic classification of depressive disorders. The lack of an adequate nosology leads to unanswered questions such as the following: Should we regard depressions occurring for the first time in the seventh decade or later as separate entities from earlier onset depressions? Can we assume that a patient whose first depression occurred at age 35 is experiencing the same disorder when he has another depressive episode at age 75? How justified are we in regarding the age at first diagnosis of depression as the age of onset, when the detection of mild depression is so
Downloaded from http://geronj.oxfordjournals.org/ at The University of British Colombia Library on June 20, 2015
The subject of aging and depression leaves many unanswered questions, and the lack of precise and universally accepted definitions (not to mention an inadequate nosology) further complicates the issue. Little is known regarding the differentiation of depressive illness from a melancholic response to the stressful aging process, and equally little regarding the natural history of depressions with onset in the teens, 20s, or 30s. Studies are focusing on biochemical and physiological aspects of depression, but at present biochemists suffer from the uncertainties of the clinicians, and the clinicians and geneticists from the limitations of the biochemists. However, despite our uncertainties about the condition, several effective forms of therapy have been developed, ranging from a focus on the therapeutic milieu to the use of pharmacologic antidepressants (particularly lithium). Ultimately, the quesion remains: Why are not all elderly persons suffering from depression? The answer may lie in the interaction of environment, life stresses, and the internal adaptive capacities of the individual.
AGING AND DEPRESSION: UNANSWERED QUESTIONS
insonism appeared late in the disease, which proved fatal 5 to 6 years after onset. The biochemical and physiological study of depressive illness might be expected to soive a number of the problems posed by the clinical approach. Unfortunately, as Lipton points out (this issue), we understand remarkably little of the processes underlying depression, despite the amount of data that has been collected. The technical difficulties of the biochemical investigations and the lack of reliable detailed information on normal cerebral biochemistry are serious obstacles. It is also difficult to distinguish those changes which are fundamental to depression from accessory changes such as the consequences of age associated altered activity levels. Advances will depend on further progress in the clinical as well as biochemical field. At the present stage biochemists suffer from the uncertainties of the clinicians, and the clinicians and geneticists from the limitations of the biochemists. When treatment is considered, we find that — as so often in the history of medicine — several reasonably effective forms of therapy and even prophylaxis (Schou, 1973) have been developed while we remain uncertain of the nature of the condition being treated. We understand neither the mechanism of the spontaneous recovery nor the reasons for the cyclical course so often seen in affective disorders. And yet, as discussed by Vickers (this issue), the appropriate milieu can prove to be therapeutic for the older patient, regardless of diagnostic category. Fann & Krai (also this issue) reviewed the current status of somatic and pharmacologic antidepressant treatments in the elderly. They both turn to clinical criteria as the basis for choice of treatment modality but as Lipton (this issue) has pointed out, response to drugs does not provide consistent support for any one hypothesis about the nature of depression. Nonetheless, there may be some specificity in the response to lithium in that the prophylactic use of this drug tends to be more successful in patients with bipolar disease than in any other group, excepting perhaps their unipolar relatives. Sleep deprivation, especially deprivation of REM sleep, a new form of treatment introduced in Europe, lends itself to trial in a wide variety of clinical situations — including elderly patients whose physical health precludes more conventional measures — because it is
Downloaded from http://geronj.oxfordjournals.org/ at The University of British Colombia Library on June 20, 2015
unreliable? As pointed out by Epstein (this issue), it is possible that many individuals hospitalized for depression in old age suffered similar episodes earlier in life but managed to avoid hospitalization because of better resources Gob security, family protection, and the like). Such earlier episodes might not be recalled in retrospective histories either by the patient or by those surviving relatives who happen to be interviewed. Lack of certainty regarding age at onset of first depressive episode in those suffering from depression in old age finds its counterpart in our ignorance of subsequent depressions in those whose first episode occurs in their teens, 20s or 30s. What proportion will have recurrent depressions? What proportion will reach the 70s, 80s, and 90s without a recurrence? Or even, what proportion will reach an advanced old age and is that proportion similar to or different from expectation based on general population data? In other words, is there a survival advantage, or disadvantage, for those with depressive disorders early in life? In all likelihood the answers will vary according to the type of depressive disorder and the adaptive capacities of the affected individuals; the confusion concerning subtypes is a major handicap. The extent of the confusion is reflected in the inconsistent nomenclature so clearly illustrated in the preceding papers. Endogenous vs. reactive, psychotic vs. neurotic, primary vs. secondary, unipolar vs. bipolar are the most popular dichotomies, but not one of them corresponds to an etiologically welldefined category. Although bipolar disease may approach such definition more closely than any of the others, even bipolar illness may be genetically heterogeneous (Mendlewicz, this issue) and even for bipolar disease there is little information concerning its natural history and life-course. Gathering such information would be greatly simplified and the reliability considerably enhanced were it possible to identify affected, or predisposed, individuals by one or more valid markers. One such marker, a taurine deficiency inherited as an autosomal dominant, recently described by Perry, Bratty, Hansen, Kennedy, Urquhart, & Dolman (in press), may differentiate a small group of unipolar depressions unresponsive to antidepressant drugs or ECT. In the one pedigree studied, symptoms of Park-
325
326
JARVIK
1971), while Seligman (1974) and colleagues noted marked differences among dogs in susceptibility to the interference with adaptive responses produced by inescapable shock and termed "learned helplessness." Whether such differences are mediated by hormonal, enzymatic, or other factors remains to be determined: the search is underway (Hamburg, 1972). In humans, too, the biological correlates have remained elusive, and even when uncovered, are likely to yield but part of the answers. It is most likely that the appearance of depressive illness will be found to depend upon the severity and number of stresses (physical-chemical, intrapsychic, and psychosocial) relative to the adaptive capacities (biological and psychological) of the individual developed from genotypic and experiential givens. In attempting to answer some of the questions on aging and depression it behooves us, therefore, to look at and learn from the aged who do not exhibit symptoms of depression as well as from those who do.
Busse, E. W., & Pfeiffer, E. (Eds.) Mental illness in later life. American Psychiatric Assn., Washington, 1973. Essen-Moller, E., & Hagnell, O. The frequency and risk of depression within a rural population group in Scania. Acta psychiatrica scandinavica, 1961,suppl. 162,28-32. Hamburg, D. A. Projections for future research. In T. A. Williams, M. Katz, & J. A. Shields (Eds.), Recent advances in the psychobiology of the depressive illnesses. DHEW Pub. No. (HSM) 70-9053. USGPO, Washington, 1972. Kirkegaard, C , Norlem, N., Lauridsen, U. B., Bjorum, N., & Christiansen, C. Protirelin stimulation test and thyroid function during treatment of depression. Archives of General Psychiatry, 1975,32, 1115-1118. McKinney, W. T., Jr., Suomi, S. J., & Harlow, H. F. Depression in primates. American Journal of Psychiatry, 1971,727, 1313-1320. Perry, T. L., Bratty, P. J. A., Hansen, S., Kennedy, J., Urquhart, N., & Dolman, C. L. Hereditary mental depression and Parkinsonism with taurine deficiency. Archives of Neurology, in press. Schou, M. Prophylactic lithium maintenance treatment in recurrent endogenous affective disorders. In S. Gerson & B. Shopsin (Eds.), Lithium, its role in psychiatric research and treatment. Plenum Press, New York, 1973. Seligman, M. E. Depression and learned helplessness. In R. J. Friedman & M. M. Katz (Eds.), The psychology of depression: Contemporary theory and research. V. H. Winston (John Wiley), Washington, 1974. Vogel, G. W., Thurmond, A., Gibbons, P., Sloan, K., Boyd, M , & Walker, M. REM sleep reduction effects on depression syndromes. Archives of General Psychiatry, 1975,32, 765-777.
Downloaded from http://geronj.oxfordjournals.org/ at The University of British Colombia Library on June 20, 2015
without adverse effects, so far as has been established. Moreover, it has led to the postulate that a disturbance of circadian rhythm may be the basic dysfunction in affective disorders. In at least one study, improvement of depressive symptoms following REM sleep deprivation correlated highly with subsequent improvement on antidepressant medication (Vogel, Thurmond, Gibbons, Sloan, Boyd, & Walker, 1975). Thus, the question is raised whether the insomnia generally considered part of "normal" aging, should not be regarded instead as another biological factor in predisposing older persons to depression (not excluding increased levels of monoamineoxidase and other biochemical alterations summarized by Lipton in this volume and continuously being reported, e.g., increased free thyroxine [Kirkegaard, Norlem, Lauridsen, Bjorum, & Christiansen, 1975]). Indeed, perhaps the most important unanswered question may be: Given these biological changes, the rising frequency of somatic illness, physiological decline, physical debilities, malnutrition, overmedication (iatrogenic or selfinduced), sensory deficits, reduction in mental agility, economic deprivations, social losses, and the increasing proximity of death, all of which are associated with advancing chronological age, why is not every old person in a profound state of depression? When we consider these manifold stresses impinging upon the aging organism, an organism which, with the passage of time, has grown less plastic and less adaptable than it was in its youth, it is hard to conceive that there should be anyone in the upper age group without the clinical diagnosis of depressive illness. And yet, a significant proportion of the elderly manage to cope with life's ever-accumulating stresses without succumbing to depressive illness. The explanation may, in part, have been provided some 19 centuries ago by Epictetus (±60 A.D.) who held that men "were not born to be depressed and unhappy with others... for God made all men to enjoy felicity and peace." Clearly, not all men were born to enjoy felicity, nor all women either, and marked individual differences in proneness to depression extend beyond homo sapiens. Studies carried out by Harlow and his associates have demonstrated considerable variability in separation-induced depressive symptoms for both mother and infant monkeys (McKinney, Suomi, & Harlow,
Aging and Depression: Some Unanswered Questions Lissy F. Jarvik, MD, PhD :
i CT TE is a fool that is not melancholy -tJ-once a day," states an old Russian proverb, reminding us of the first unanswered question about aging and depression: Where do we draw the line between depression and sadness, between depressive illness and the melancholy mood which is an appropriate response to life's stresses? While this question pertains to depression at any age, it is particularly cogent in old age where decline in physical strength, in health, in mental alertness, in earning capacity, in prestige at home and in society combine with economic deprivation, the loss of significant others, social isolation, and the ever more closely approaching prospect of self-death to provide adequate cause for dejection, despair, and despondency. As Epstein (this issue) has pointed out: "Depression has almost been viewed as a characteristic of senescence." In an elderly person even "seemingly obvious symptoms" may not result in a diagnosis of depressive illness; instead they may be ascribed to the aging process as such. Without a clear-cut universally accepted definition, it is not surprising that prevalent figures vary from survey to survey as do reviewers' conclusions. Thus, Busse& Pfeiffer (1973) are quoted by Epstein (this issue) as citing a range from 10% to 65%, while 'Revision of presentation made at the Symposium on Age Differentiation in Depressive Illness (J. Zubin, PhD, Chairman; Lissy F. Jarvik, MD, PhD, Organizer) at the 25th Annual Scientific Meeting of Gerontological Society, San Juan, Puerto Rico, Dec. 18, 1972. 'Veterans Administrations Hospital, Brentwood, and UCLA Dept. of Psychiatry (for reprints: 760 Westwood Plaza, Los Angeles 90024).
324
Gurland (this issue) considers the most likely prevalence to be less than 5% for the age group above 65 years using psychiatric diagnoses rather than symptom checklists as criterion measures. Klerman (this issue) estimates that 8 10% of men and 16 - 29% of women will develop depression sometime during their lives, while according to Essen-Moller & Hagnell (1961) the cumulative risks to age 80 in a population of 2550 (ascertained and examined psychiatrically with 99% completeness) were 17.7% and 8.5% for men and women, respectively. The uncertainty regarding its prevalence is paralleled by the confusion concerning types of depressive illnesses. While there are few advocates today of the position that the diagnosis of depression connotes a homogeneous entity (differences in severity accounting for apparent clinical subgroups), there is as yet no satisfactory scheme for the systematic classification of depressive disorders. The lack of an adequate nosology leads to unanswered questions such as the following: Should we regard depressions occurring for the first time in the seventh decade or later as separate entities from earlier onset depressions? Can we assume that a patient whose first depression occurred at age 35 is experiencing the same disorder when he has another depressive episode at age 75? How justified are we in regarding the age at first diagnosis of depression as the age of onset, when the detection of mild depression is so
Downloaded from http://geronj.oxfordjournals.org/ at The University of British Colombia Library on June 20, 2015
The subject of aging and depression leaves many unanswered questions, and the lack of precise and universally accepted definitions (not to mention an inadequate nosology) further complicates the issue. Little is known regarding the differentiation of depressive illness from a melancholic response to the stressful aging process, and equally little regarding the natural history of depressions with onset in the teens, 20s, or 30s. Studies are focusing on biochemical and physiological aspects of depression, but at present biochemists suffer from the uncertainties of the clinicians, and the clinicians and geneticists from the limitations of the biochemists. However, despite our uncertainties about the condition, several effective forms of therapy have been developed, ranging from a focus on the therapeutic milieu to the use of pharmacologic antidepressants (particularly lithium). Ultimately, the quesion remains: Why are not all elderly persons suffering from depression? The answer may lie in the interaction of environment, life stresses, and the internal adaptive capacities of the individual.
AGING AND DEPRESSION: UNANSWERED QUESTIONS
insonism appeared late in the disease, which proved fatal 5 to 6 years after onset. The biochemical and physiological study of depressive illness might be expected to soive a number of the problems posed by the clinical approach. Unfortunately, as Lipton points out (this issue), we understand remarkably little of the processes underlying depression, despite the amount of data that has been collected. The technical difficulties of the biochemical investigations and the lack of reliable detailed information on normal cerebral biochemistry are serious obstacles. It is also difficult to distinguish those changes which are fundamental to depression from accessory changes such as the consequences of age associated altered activity levels. Advances will depend on further progress in the clinical as well as biochemical field. At the present stage biochemists suffer from the uncertainties of the clinicians, and the clinicians and geneticists from the limitations of the biochemists. When treatment is considered, we find that — as so often in the history of medicine — several reasonably effective forms of therapy and even prophylaxis (Schou, 1973) have been developed while we remain uncertain of the nature of the condition being treated. We understand neither the mechanism of the spontaneous recovery nor the reasons for the cyclical course so often seen in affective disorders. And yet, as discussed by Vickers (this issue), the appropriate milieu can prove to be therapeutic for the older patient, regardless of diagnostic category. Fann & Krai (also this issue) reviewed the current status of somatic and pharmacologic antidepressant treatments in the elderly. They both turn to clinical criteria as the basis for choice of treatment modality but as Lipton (this issue) has pointed out, response to drugs does not provide consistent support for any one hypothesis about the nature of depression. Nonetheless, there may be some specificity in the response to lithium in that the prophylactic use of this drug tends to be more successful in patients with bipolar disease than in any other group, excepting perhaps their unipolar relatives. Sleep deprivation, especially deprivation of REM sleep, a new form of treatment introduced in Europe, lends itself to trial in a wide variety of clinical situations — including elderly patients whose physical health precludes more conventional measures — because it is
Downloaded from http://geronj.oxfordjournals.org/ at The University of British Colombia Library on June 20, 2015
unreliable? As pointed out by Epstein (this issue), it is possible that many individuals hospitalized for depression in old age suffered similar episodes earlier in life but managed to avoid hospitalization because of better resources Gob security, family protection, and the like). Such earlier episodes might not be recalled in retrospective histories either by the patient or by those surviving relatives who happen to be interviewed. Lack of certainty regarding age at onset of first depressive episode in those suffering from depression in old age finds its counterpart in our ignorance of subsequent depressions in those whose first episode occurs in their teens, 20s or 30s. What proportion will have recurrent depressions? What proportion will reach the 70s, 80s, and 90s without a recurrence? Or even, what proportion will reach an advanced old age and is that proportion similar to or different from expectation based on general population data? In other words, is there a survival advantage, or disadvantage, for those with depressive disorders early in life? In all likelihood the answers will vary according to the type of depressive disorder and the adaptive capacities of the affected individuals; the confusion concerning subtypes is a major handicap. The extent of the confusion is reflected in the inconsistent nomenclature so clearly illustrated in the preceding papers. Endogenous vs. reactive, psychotic vs. neurotic, primary vs. secondary, unipolar vs. bipolar are the most popular dichotomies, but not one of them corresponds to an etiologically welldefined category. Although bipolar disease may approach such definition more closely than any of the others, even bipolar illness may be genetically heterogeneous (Mendlewicz, this issue) and even for bipolar disease there is little information concerning its natural history and life-course. Gathering such information would be greatly simplified and the reliability considerably enhanced were it possible to identify affected, or predisposed, individuals by one or more valid markers. One such marker, a taurine deficiency inherited as an autosomal dominant, recently described by Perry, Bratty, Hansen, Kennedy, Urquhart, & Dolman (in press), may differentiate a small group of unipolar depressions unresponsive to antidepressant drugs or ECT. In the one pedigree studied, symptoms of Park-
325
326
JARVIK
1971), while Seligman (1974) and colleagues noted marked differences among dogs in susceptibility to the interference with adaptive responses produced by inescapable shock and termed "learned helplessness." Whether such differences are mediated by hormonal, enzymatic, or other factors remains to be determined: the search is underway (Hamburg, 1972). In humans, too, the biological correlates have remained elusive, and even when uncovered, are likely to yield but part of the answers. It is most likely that the appearance of depressive illness will be found to depend upon the severity and number of stresses (physical-chemical, intrapsychic, and psychosocial) relative to the adaptive capacities (biological and psychological) of the individual developed from genotypic and experiential givens. In attempting to answer some of the questions on aging and depression it behooves us, therefore, to look at and learn from the aged who do not exhibit symptoms of depression as well as from those who do.
Busse, E. W., & Pfeiffer, E. (Eds.) Mental illness in later life. American Psychiatric Assn., Washington, 1973. Essen-Moller, E., & Hagnell, O. The frequency and risk of depression within a rural population group in Scania. Acta psychiatrica scandinavica, 1961,suppl. 162,28-32. Hamburg, D. A. Projections for future research. In T. A. Williams, M. Katz, & J. A. Shields (Eds.), Recent advances in the psychobiology of the depressive illnesses. DHEW Pub. No. (HSM) 70-9053. USGPO, Washington, 1972. Kirkegaard, C , Norlem, N., Lauridsen, U. B., Bjorum, N., & Christiansen, C. Protirelin stimulation test and thyroid function during treatment of depression. Archives of General Psychiatry, 1975,32, 1115-1118. McKinney, W. T., Jr., Suomi, S. J., & Harlow, H. F. Depression in primates. American Journal of Psychiatry, 1971,727, 1313-1320. Perry, T. L., Bratty, P. J. A., Hansen, S., Kennedy, J., Urquhart, N., & Dolman, C. L. Hereditary mental depression and Parkinsonism with taurine deficiency. Archives of Neurology, in press. Schou, M. Prophylactic lithium maintenance treatment in recurrent endogenous affective disorders. In S. Gerson & B. Shopsin (Eds.), Lithium, its role in psychiatric research and treatment. Plenum Press, New York, 1973. Seligman, M. E. Depression and learned helplessness. In R. J. Friedman & M. M. Katz (Eds.), The psychology of depression: Contemporary theory and research. V. H. Winston (John Wiley), Washington, 1974. Vogel, G. W., Thurmond, A., Gibbons, P., Sloan, K., Boyd, M , & Walker, M. REM sleep reduction effects on depression syndromes. Archives of General Psychiatry, 1975,32, 765-777.
Downloaded from http://geronj.oxfordjournals.org/ at The University of British Colombia Library on June 20, 2015
without adverse effects, so far as has been established. Moreover, it has led to the postulate that a disturbance of circadian rhythm may be the basic dysfunction in affective disorders. In at least one study, improvement of depressive symptoms following REM sleep deprivation correlated highly with subsequent improvement on antidepressant medication (Vogel, Thurmond, Gibbons, Sloan, Boyd, & Walker, 1975). Thus, the question is raised whether the insomnia generally considered part of "normal" aging, should not be regarded instead as another biological factor in predisposing older persons to depression (not excluding increased levels of monoamineoxidase and other biochemical alterations summarized by Lipton in this volume and continuously being reported, e.g., increased free thyroxine [Kirkegaard, Norlem, Lauridsen, Bjorum, & Christiansen, 1975]). Indeed, perhaps the most important unanswered question may be: Given these biological changes, the rising frequency of somatic illness, physiological decline, physical debilities, malnutrition, overmedication (iatrogenic or selfinduced), sensory deficits, reduction in mental agility, economic deprivations, social losses, and the increasing proximity of death, all of which are associated with advancing chronological age, why is not every old person in a profound state of depression? When we consider these manifold stresses impinging upon the aging organism, an organism which, with the passage of time, has grown less plastic and less adaptable than it was in its youth, it is hard to conceive that there should be anyone in the upper age group without the clinical diagnosis of depressive illness. And yet, a significant proportion of the elderly manage to cope with life's ever-accumulating stresses without succumbing to depressive illness. The explanation may, in part, have been provided some 19 centuries ago by Epictetus (±60 A.D.) who held that men "were not born to be depressed and unhappy with others... for God made all men to enjoy felicity and peace." Clearly, not all men were born to enjoy felicity, nor all women either, and marked individual differences in proneness to depression extend beyond homo sapiens. Studies carried out by Harlow and his associates have demonstrated considerable variability in separation-induced depressive symptoms for both mother and infant monkeys (McKinney, Suomi, & Harlow,
