Older Patients' Aversion to Antidepressants

Abstract

BACKGROUND

Depression is common among older patients yet is often inadequately treated. Patient beliefs about antidepressants are known to affect treatment initiation and adherence, but are often not expressed in clinical settings.

OBJECTIVE

To explore attitudes toward antidepressants in a sample of depressed, community-dwelling elders who were offered treatment.

DESIGN

Cross-sectional, qualitative study utilizing semi-structured interviews.

PARTICIPANTS

Primary care patients age 60 years and over with depression, from academic and community primary care practices of the University of Pennsylvania Health System and the Philadelphia Department of Veterans Affairs. Patients participated in either the Prevention of Suicide in Primary Care Elderly: Collaborative Trial or the Primary Care Research in Substance Abuse and Mental Health for the Elderly Trial. Sixty-eight patients were interviewed and responses from 42 participants with negative attitudes toward medication for depression were analyzed.

MEASUREMENTS

Interviews were audiotaped, transcribed, and entered into a qualitative software program for coding and analysis. A multidisciplinary team of investigators coded the transcripts and identified key features of narratives expressing aversion to antidepressants.

RESULTS

Four themes characterized resistance to antidepressants: (1) fear of dependence; (2) resistance to viewing depressive symptoms as a medical illness; (3) concern that antidepressants will prevent natural sadness; (4) prior negative experiences with medications for depression.

CONCLUSIONS

Many elders resisted the use of antidepressants. Patients expressed concerns that seem to reflect their concept of depression as well as their specific concerns regarding antidepressants. These findings may enhance patient-provider communication about depression treatment in elders.

At least twelve percent of primary care patients over the age of 60 experience major or minor depression.¹ Despite evidence that both psychotherapy and pharmacologic treatments are effective in older adults, depression is often untreated or undertreated in this population.^2–5 While medication for depression is often prescribed in primary care, adherence is low and may be related to the negative views toward antidepressants expressed by both patients and the general public.⁶ Beliefs about medication are cited by patients as important determinants of depression treatment acceptance and are also known to affect adherence.^{7, 8}

Antidepressants are well established as the recommended treatment for major depression in the elderly⁹; however, their effectiveness in treating minor depression and dysthmia is less certain and psychological treatments may be equally effective.^{4, 10} In these situations where the optimal treatment is uncertain, patient beliefs and preferences regarding antidepressants are particularly salient.

Underlying our study is the premise that beliefs affect behavior, an assumption supported by cognitive models of health behavior such as the Health Belief Model and the Theory of Reasoned Action. In these models, health-related behaviors are explained by knowledge, beliefs, and attitudes. The Health Belief Model stresses beliefs about perceived susceptibility to illness and severity of illness. The Theory of Reasoned Action includes perceived social approval or disapproval as a determinant of behavior.¹¹

Evidence that patients view medications with caution and resist taking them is increasing.¹² The most common negative attitudes toward psychotropic medication are fears about side effects and addiction.^{7, 13} Attitudes felt to prevent depressed older patients from using antidepressants include the associated stigma, concerns about side effects, and lack of education or support from providers.^{2, 14} In order to further explore potential barriers to antidepressant use in older depressed adults, this qualitative study investigates reasons for resisting the use of pharmacologic treatment for depression and builds a thematic framework for understanding these concerns in order to facilitate patient-provider communication in a clinical setting.

METHODS

Design

This qualitative study follows 2 larger quantitative research projects in a sequential mixed methods design.¹⁵ Participants were drawn from a cohort of older adults (age 60 years and over) with depression who participated in 1 of 2 multicenter studies of depression care delivery. The Prevention of Suicide in Primary Care Elderly: Collaborative Trial (PROSPECT) was a community-based randomized trial comparing usual care for depression to use of a clinical algorithm tailored to the elderly. The Primary Care Research in Substance Abuse and Mental Health for the Elderly (PRISM-E) study was a randomized, multisite investigation comparing the effectiveness of integrated care to enhanced referral systems for the treatment of behavioral health issues in the elderly. Depressed participants were identified through primary care screening after which all patients underwent diagnostic interviews. Details of the methods for these trials are published elsewhere.^{16, 17}

Participants

At the termination of their participation in either the PROSPECT or PRISM-E studies, participants with depression who were recruited from primary care practices affiliated with the University of Pennsylvania Health System and at the Philadelphia Department of Veterans Affairs (VA) were invited to participate in follow-up qualitative studies (n=322). Of those who consented (n=201), a purposive sample (n=68) was selected composed of participants from both the PRISM-E (n=29) and PROSPECT (n=39) studies. The purposive sampling strategy sought to include roughly equal numbers of participants from each of 4 outcome response categories defined as follows based on comparison between baseline and termination scores from the Center for Epidemiologic Studies Depression scale (CES-D)¹⁸ in PRISM-E or the Hamilton Depression Rating Scale (HDRS)¹⁹ in PROSPECT: (1) no response (baseline CES-D>16 or HDRS>10 with <25% reduction in score at termination); (2) partial response (baseline CES-D>16 or HDRS>10 with >25% reduction but score still over 16 or 10 at termination); (3) remission (baseline CES-D>16 or HDRS >10 with scores below this at termination); (4) low distress (CES-D<16 or HDRS<10 at both baseline and termination). Each sampling category included participants who both engaged in treatment (had at least 2 treatment visits) as well as those who did not have at least 2 treatment visits but remained in the study. This paper presents results from the 42 participants who expressed reluctance or refusal to use antidepressant medication during the qualitative interview.

Data Collection

A graduate student conducted interviews lasting 60–90 minutes in the patient's home. Questions for the interview were developed using Kleinman's Explanatory Model²⁰ as a framework. This model allows for a description of how patients describe the nature, cause, and course of their illness and has been used by others in the study of mental illness.²¹ For this study, participants were asked about their experience of depression and of their treatment during the studies. The 3 questions about treatment relevant to this paper were: (1) what kind of treatment did they recommend for you?; (2) when you heard about the treatment you would be getting, what did you think?; (3) after you got started with your treatment what did you think about it? The average time from completion of the quantitative parent study to the qualitative interview was 9 months.

Analysis

Interviews were audiotaped and transcribed by a professional transcription service. Two research assistants checked the transcripts for accuracy against the audiotapes prior to data entry into QSR N6 (QSR International, Melbourne, Australia), a software package for qualitative analysis. The constant comparative method, in which original themes are revised after comparison with newer themes generated during the coding process, was used for analysis.²² Broad codes reflecting patients' responses to questions about depression and treatment were created. For this paper, narratives expressing a reluctance or refusal to use medication for depression were reviewed and, within these, finer codes were created to identify specific themes of aversion. Using the constant comparative approach, once the themes relating to attitudes toward treatment were identified, themes were compared across cases to ensure that they were both representative and inclusive of all cases. A multidisciplinary team of investigators from anthropology, psychiatry, primary care, psychology, and psychiatric nursing were involved in coding and analyzing the transcripts. Discrepancies in coding were discussed during bi-monthly meetings and were resolved by consensus.

Quantitative comparisons of participant characteristics between those who expressed negative attitudes toward antidepressants and those who did not were completed using Pearson's χ² tests. The University of Pennsylvania Institutional Review Board approved this study and all participants signed informed consent forms.

RESULTS

Of the 68 adults who participated in qualitative interviews, 42 expressed a reluctance or refusal to use antidepressant medication. These participants did not differ significantly from the remaining 26 participants with respect to age, sex, race, education, level of depression, engagement in treatment, or response to treatment when compared using the Pearson's χ² test (Table 1). Participants who underwent qualitative interviews had major depression, minor depression, depression not otherwise specified (NOS), or dysthymia. Of the 42 participants who expressed reluctance or refusal, the majority had major depression (n=25), and did not differ significantly from those with the less severe forms of depression with respect to race or gender when compared at the .05 level using the Pearson's χ² test (data not shown).

Table 1.

Characteristics of Participants Who Stated Reluctance to Take Antidepressants Compared to Those Who Did Not Express Reluctance

Demographic	Expressed Reluctance N=42, n (%)	Did Not Express Reluctance N=26, n (%)	P Value*
Female	27 (64)	20 (77)	.27
Race			.17
African American	17 (40)	16 (62)
White	23 (55)	10 (38)
Asian	2 (5)	0 (0)
Age (y)			92
60–69	10 (24)	6 (23)
70–79	21 (50)	12 (46)
Over 80	11 (26)	8 (31)
Educational level			.26
Less than high school	6 (14)	8 (31)
High school	14 (33)	7 (27)
Greater than high school	22 (52)	11 (42)
Current marital status			.83
Married	8 (31)	14 (33)
Not married†	18 (69)	28 (67)
Depressive illness			.46
Major depression	25 (60)	15 (58)
Minor depression	12 (28)	7 (27)
Depression NOS‡ dysthymia	5 (12)	4 (15)
Response to treatment			.46
No response	9 (35)	11 (26)
Partial response	5 (19)	7 (17)
Remitted	9 (35)	22 (52)
Low distress	3 (12)	2 (5)

^*χ² test of two proportions.

^†Includes divorced, never married, separated, and widowed.

^‡NOS, not otherwise specified.

Four themes emerged from an examination of responses expressing reluctance or refusal to use antidepressants: (1) fear of addiction; (2) resistance to viewing depression as a medical illness; (3) concern that antidepressants will prevent feelings of natural sadness; and (4) prior negative experiences with medications for depression are an obstacle to treatment. We examined the demographic characteristics of patients articulating each of the 4 themes, and no pattern emerged with respect to age, gender, race, or level of depressive symptoms. Each theme is discussed below.

1. Fear of addiction: When asked how they responded to having been offered antidepressants, many participants expressed a fear of becoming dependent upon the medication. In addition to a global concern of dependence, there were 2 worries that seemed particularly relevant to antidepressants. One of these was a concern of needing prolonged treatment.

   I didn't want to start get myself hooked on a medication that I would have to be taking the rest of my life.

   Other participants were open to taking the medication but were concerned that it would be too effective and expressed a need to remain vigilant in order to monitor themselves for dependence.

   I think sometimes medication is wonderful but I think you can't escape from your problems that way so I watch when I take it. I don't want to get dependent on it.

   I stopped taking it on my own … I felt that I didn't want to stay on the medication. I didn't want to become addicted to it, an antidepressant.

2. Resistance to viewing depression as a medical illness: During the 90-minute interview, participants had the opportunity to explain their views on the etiology of their depression in great detail, and many participants articulated specific social causes as illustrated by the following quotes:

   I have (a) stressful time going, dealing with death in the family, losing my mother, losing my father a year ago. In fact, a year this February I lost six other family members in one year. And it just looked like it was just too much to cope with.

   It's not—I don't know whether it's the depression or not but I think when it changed why I feel that the death of my husband has changed me. He was the first man that I loved and I—even feel yet that a part of me is missing, that something—just something I feel that a part of me is missing because he is not around. …

   If you can't see and you feel like you're going to lose your eyesight, you know, it kinda gets you down. Especially when you don't have nobody.

   In keeping with our explanatory model in which the cause of illness is related to the understanding of treatment, some participants referred to the cause of depression when they were asked about their response to treatment. For some, the attribution of depression to social causes led to a resistance to pharmacologic treatment. For example, in the following quote, medication was not seen as addressing the root cause of depression, and so was not valued.

   I do think that there's a reason for my depression. I don't think it's just there like a cloud because nothing's wrong. I think there are things that are wrong and that's why I kind of don't like to take medication for it because the medication doesn't change the basics.

3. Antidepressants prevent natural sadness: A particular resistance relating to a belief in the transforming power of antidepressants emerged. Participants were concerned that the medication would take away their reality and make them unnaturally happy. For these depressed elders, the themes of loss were common, and the experience of genuine sadness was valued and seen as an appropriate normal occurrence. In addition, some did not want to be too happy, as this might connote betrayal or create distance from others.

   He prescribed Zoloft for me. Well I never took it. I mean, my feeling at the time was that I wasn't interested in the pill. I didn't want to do this because I couldn't just bury my husband and then go on and go out and party.

   I have to face reality and I think you have to feel some pain in life.

   I didn't want to stay on the medication … why should I be different than everybody else?

4. Prior negative experiences with medications for depression: Some participants described having taken medication for depression in the past, with troubling side effects, particularly related to sedation. This resulted in a refusal to accept treatment, even with newer medications. For example:

I didn't want to take them … 'cause I had taken tranquilizers when I was young … A doctor recommended that … I don't think they knew about antidepressants then … I never thought it was nerves but I couldn't take 'em, I slept.

I'm not interested in pills anymore. I get bad dreams. I mean, they gave me pills that left me waking up and not knowing where I was. I was still in a dream.

DISCUSSION

Our analysis reveals 4 themes that describe older adults' aversion to pharmacologic treatment for depression. Patients had concerns regarding addiction, resistance to viewing depression as a medical illness, fear that antidepressants prevent natural sadness, and prior negative experience with medication for depression. Our initial intent in this analysis was to answer the clinical question of why older patients resist the use of antidepressants. We came away with a richer understanding and appreciation for participants' experience of depression and an awareness of how depression treatment is imbued with meaning based on particular concerns about the treatment as well as patients' views of their illness.

Before discussing our results, we should point out some limitations of our study. All of the participants in our study took part in a depression treatment delivery trial and may be more receptive to depression treatment than other primary care patients. Strengths of our patient sample include purposive sampling procedures ensuring a diverse group of participants, all of whom had either major or minor depression. Our study design employs participant recall, which poses potential limitations to the validity of the results. However, we believe that patients' recall of their reactions to treatment options is likely to be stable over the time frame of the study. Last, although medication cost may be a barrier to antidepressant use for some patients, our study cannot provide reliable information on cost barriers because in some cases medications were offered free of charge.

FEAR OF ADDICTION

Patients in our study expressed strong fears of addiction, a finding consistent with other literature about antidepressants.⁶ This fear of addiction may indicate that patients expect to experience a “high” from antidepressants, or are concerned they will need to take them for an extended period of time. To some degree, both of these concerns are rooted in the reality of current practice. While newer antidepressants such as selective serotonin reuptake inhibitors (SSRIs) are not associated with typical features of dependence such as tolerance and craving, patients can experience potentially troubling discontinuation symptoms upon stopping the medication.²³ Current treatment practices also indicate that patients who have recurrent major depressive episodes may need extended treatment courses of antidepressants.²⁴ Therefore, it may be important for providers to discuss these issues with patients when considering antidepressant therapy.

Another interpretation of this concern is that patients believe that if they start an antidepressant they will become psychologically weak and unable to discontinue it. These patients may be concerned that they will essentially be controlled by the medication. This concern is in keeping with previous research on concerns about loss of control, which suggests that in contrast to treatments for physical illness, patients view psychotropic drugs as a threat to self.¹³

RELUCTANCE TO VIEW DEPRESSION AS A MEDICAL ILLNESS

Some participants in our study expressed a reluctance to view their symptoms as needing medical intervention. Other work has shown that patients who attribute depression to social or interpersonal causes are less willing to view their condition as requiring medication. Conversely, when depressive symptoms are seen as a medical disorder rather than as an emotional reaction to circumstances, there is a greater perception of severity and need to seek treatment.²⁵ Participants described events such as the loss of loved ones and medical illness as being the cause of their depression. Because such events are common among the elderly and have been associated with an increase in depressive symptoms,²⁶ inquiry about life events in the clinical encounter may be useful in diagnosing depression. Our results suggest that a discussion of these events and how they are related to depressive symptoms may also be important when offering depression treatment. Although patients may resist the use of medication, combination therapy with antidepressants and psychotherapy is recommended in expert panel findings for the treatment of depression in the elderly precipitated by a psychosocial stressor.²⁷

Participants also expressed the particular concern that antidepressants would not address the underlying cause of their distress, a view supported by the general public.²⁸ In a clinical setting, discussion with patients about how they view their depression can be helpful in identifying those patients who view their symptoms in a social context. These patients may be more receptive to psychotherapy than to pharmacologic treatment which may be appropriate especially for cases of minor depression.

ANTIDEPRESSANTS PREVENT NATURAL SADNESS

The third theme, that antidepressants may prevent natural sadness, also points to the importance of the social context of depression. Participants had strong feelings about wanting to be able to experience sadness, which had important meaning, especially in reaction to loss. Patients expressed the concern that taking antidepressants would prevent them from feeling this emotion. It is important to note that although many patients had experienced the loss of a loved one, all patients had depression; normal grief reactions would not qualify a patient for participation in the study. Patients may have a fear of feeling happy after the loss of a loved one out of a sense of guilt and not recognize the severity of their illness.

Beyond the specific situation of response to loss, these patients seem to be articulating the value of authenticity, of being true to one's emotions. The concern that antidepressants can make someone a different person is discussed in Peter Kramer's book, Listening to Prozac.²⁹ This book, which was popular in the mid-1990s, describes the belief in the transforming power of Prozac (fluoxetine), and how it could potentially alter a person's identity. In later writing, Kramer argues that this resistance to fluoxetine's potential to alter identity represents a “threat to melancholy,” essentially a concern about the inability to experience sadness while on the medication.³⁰ This concern appears to be similar to those articulated by some participants in our study.

The fear of not being able to feel sadness may also reflect a lack of distinction between sadness and depression. Medical providers are likely to believe that even when taking an antidepressant, a person can still experience sadness; the function of the antidepressant is to free patients from the hopelessness and incapacitation that characterizes depression. Patient education may alleviate some fears of losing the ability to experience emotions while on antidepressants. However, another reaction to these narratives is that patients may be appropriately immersed in their lives and interested in dealing with issues of grief and sadness. Patients such as these may benefit from psychotherapy either as a sole depression treatment or in addition to pharmacologic treatment.

PRIOR EXPERIENCE WITH MEDICATION FOR DEPRESSION

Last, concerns over prior experiences with medication for depression prevented some participants from using antidepressants. In particular, concern over the sedating qualities of prior treatment was common. Older patients such as those in our study may not be aware of the substantial changes that have taken place in the pharmaceutical treatment of depression. Newer medications such as SSRIs are less sedating than older antidepressants³¹ and although hypnotic and sedative agents have been used in the past to treat elders with depression,³² it is not a recommended treatment. While the general public tends to still view antidepressants as having sedating side effects,²⁸ patients who have been treated for depression in the past with older medications may be particularly concerned about sedation. Additionally, it may be particularly appropriate to discuss sedative qualities with older patients, as they may be more likely to experience this side effect than younger patients.³³

Our study focused on the attitudes of older patients, attitudes which likely reflect their particular life experiences. They may also reflect common experiences of an older generation's exposure to particular social forces. While the field of psychiatry has gained greater acceptance, older generations may still hold negative views. For this reason, some negative views of antidepressants may result from a deeper distrust of the field of psychiatry in general, particularly in the concerns about dependence and of surrendering control to a psychotropic agent. Patient age may also be a factor in the reluctance to medicalize depressive symptoms. Older adults may attribute symptoms of depression to the aging process itself, to the loss of loved ones, or to age-related physical illness and be less willing than younger patients to accept medication as treatment.³⁴

Our study has identified topics of concern among older patients with depression who expressed aversion to the use of antidepressant medications. Our findings highlight the need for patient-provider dialogue regarding the characteristics of current antidepressant therapy. Differences exist between how health care providers and patients conceptualize both depression and depression treatment.^{25, 35} These differences may hinder initiation or continuation of therapy, yet dialogue between patients and providers about such differences can lead to improved health outcomes.³⁶ Because patients may be reticent about expressing aversion to medicine in a clinical setting,³⁷ it is particularly important for health care providers to initiate discussions about patient beliefs and preferences. An open discussion between patient and provider about treatment options is especially important for mild-to-moderate depression, where many treatment approaches may be equally effective.²³

Our findings are a representation of what anthropologists refer to as the emic perspective that describes how insiders, in this case, persons experiencing depression, think about and categorize events such as illness and treatment. This perspective is important given the movement toward patient-centered practice and the need for better provider-patient communication.³⁸ Patient-centered practice is based on the premise that shared decision making is a valuable goal of the therapeutic encounter. Specific tasks in this realm include elicitation of patient views regarding treatment options, exploration of those views, discussion of the pros and cons of the treatments, and involvement of the patient in the treatment decision.³⁶ The Institute of Medicine (IOM) has recently identified the delivery of patient-centered care, including being responsive to individual patient preferences, needs, and values as a key component of improving quality of care.³⁹ Our findings may benefit patients and providers by offering a starting point for exploring the attitudes of an older individual being offered antidepressant therapy. Although some of the concerns raised by patients in this study may be changed through education, respect for patient views may lead to alternative treatment decisions. Specific concerns regarding dependence and whether a person can experience genuine emotions while on antidepressants may be important to discuss. For patients such as those in our study who wish to experience a full range of emotions, an honest discussion of how antidepressants may affect mood range and depth is essential. Patients with these concerns may benefit from a trial of medication with the goal of reassessing their ability to experience emotions. Educating patients about the reduced sedative qualities of newer antidepressants may be particularly useful when counselling older patients, and as newer agents differ in their sedative qualities,⁴⁰ choosing the appropriate medication and monitoring for this effect may be clinically important. Last, inquiring about patients' views of depression may allow for a discussion regarding whether antidepressants can be helpful to those who regard their depression as having social or interpersonal causes or whether psychotherapy would be more appropriate.

Our findings also suggest areas for future research. The concerns raised by patients in this study may represent isolated views, or may be representative of larger groups. Inclusion of questions related to these concerns in larger studies utilizing survey methods can help to determine how common these views are, and whether public health communication campaigns would be useful to help address these concerns.