Emotional distress and psychiatric syndromes are prevalent in the breast cancer population at large. However, to date there is a paucity of literature specifically concerning presurgical breast cancer patients.
Emotional distress and psychiatric syndromes are prevalent in the breast cancer population at large. However, to date there is a paucity of literature specifically concerning presurgical breast cancer patients.
The authors assessed 236 newly diagnosed patients at the time of their presurgical consultation at the Comprehensive Breast Cancer Program of Dartmouth-Hitchcock Medical Center in Lebanon, NH.
Of patients in this study, 41% rated their distress in the clinically significant range on the Distress Thermometer (ie, >5, 0–10 scale). Nearly one-half (47%) of patients met established thresholds for positivity on 1 or more screens for distress or psychiatric disorders. Prevalence rates were 11% for major depression (60% of these patients were moderately severe to severely depressed) and were 10% for posttraumatic stress disorder (PTSD). Emotional symptoms markedly interfered with daily function in both groups. Of depressed patients, 56% were already taking a psychotropic medication, yet they still met screening criteria for major depression.
Emotional distress and psychiatric syndromes (major depression and PTSD) were prevalent in this population. Markedly impaired function was evident for both depressed and PTSD patients. Future research should refine current screening procedures and develop interventions to better address emotional distress and psychiatric disorders in newly diagnosed breast cancer patients. Cancer 2006. © 2006 American Cancer Society.
Approximately one-third of breast cancer patients experience significant distress and/or impairment of function,1–5 and symptoms of distress persist for as long as 10–20 years after treatment.4, 6 Distress is often accompanied by a psychiatric syndrome; the most common are adjustment disorder, major depression, minor depression, anxiety disorders, delirium and/or dementia, and substance abuse.1–3, 7–12 These data have been collected in samples of patients at postsurgical stages of treatment, thus not allowing generalization to newly diagnosed patients before surgery.
Unfortunately, women with newly diagnosed breast cancer who are highly distressed may not disclose emotional concerns,13 and clinicians may not always probe these areas.14 The reasons for this are many. From the patient side, patients are often reluctant to ask for help because of their fear of being stigmatized for having a psychological problem or because of their fear of distracting the oncologist from curing their cancer.15–17 This is not unique to cancer patients. Distressed patients in primary care are also reluctant to admit psychological distress.18, 19
On the healthcare system side, oncologists are not always trained or skilled at discussing emotional problems.15, 20, 21 Clinicians often assume that emotional symptoms are a normal consequence of having cancer, and, therefore, they are less likely to follow signs of distress.14 This problem has become more pronounced as more patient care has shifted to the busy ambulatory care setting, where physician visits are short and hurried, and time to exam emotional function is limited.7, 14 The result is that patients are at risk for unrecognized and inadequately treated distress.22, 23
The development of screening mechanisms to improve detection and management of emotional distress in breast cancer patients has been endorsed by the National Comprehensive Cancer Network (NCCN)7, 16 and the American Psychosocial Oncology Society (APOS).15 These guidelines emphasize initiating screening for distress at the time of cancer diagnosis and continuing screening through treatment and survivorship.
Efficient screening methods to identify patients with significant distress and psychiatric disorders have been developed. In 1 study with breast cancer patients, self-report measures consistently distinguished between patients with and without significant psychological distress compared with clinician-administered structured diagnostic interviews.24 Distress screens correlate with rigorous methods used to identify clinically significant distress, and these screens are sensitive to change as cancer-related problems increase.23 Identifying distressed patients is the first step toward implementing evidence-based treatments for distress management as outlined by the NCCN7, 16 and the APOS.15 Initiating screening at the time of initial consultation with the surgeon is perhaps ideal, as this is often the patient's first point of contact, and an initial screening will provide a baseline for comparison as cancer patients progress through treatment and follow-up.
In this article, we report prevalence figures for clinically significant emotional distress, psychiatric syndromes and substance use, and status of function in newly diagnosed breast cancer patients.
The assessment was administered to newly diagnosed breast cancer patients consecutively presenting at the comprehensive breast cancer clinic of the Norris Cotton Cancer Center of Dartmouth-Hitchcock Medical Center between April 2004 and March 2006. Patients completed the assessment battery as part of standard clinic procedure before their initial consultation with a surgical oncologist. Assessments were administered, and responses were self-reported by means of a touch-pad computer. Validated questionnaires were used when possible. The Dartmouth College Institutional Review Board approved this study and deemed informed consent unnecessary because of use of de-identified health information.
Participants included female patients from Northern New England (primarily New Hampshire and Vermont). All patients were newly diagnosed by a biopsy that resulted in a positive diagnosis of stage I, II, or III breast cancer. Women were eligible to complete the questionnaire if they could read, write, and speak English and were capable of making healthcare decisions for themselves. Only 1 woman was ineligible on these grounds, and 4 women failed to complete the assessment because scheduling problems, thus leaving a sample of 236 patients.
The Distress Thermometer (DT),25 endorsed by the National Comprehensive Cancer Network (NCCN) Psychosocial Distress Practice Guidelines Panel,16 is a brief self-report screening tool that measures sources of distress and distress levels over the past week. The DT is a visual analogue scale with scores from 0 (no distress) to 10 (extreme distress) and a midpoint anchor labeled “moderate distress.”7, 16 In accord with NCCN guidelines, we consider a DT score of >5 as an indication of clinically significant distress. In addition to rating distress, the DT screens causes of distress within 5 categories: practical problems, family problems, emotional problems, spiritual and/or religious concerns, and physical problems. We included an additional category, “Uncertainty”, because the intake assessment is also used as a component of an informed decision-making study.
The Patient Health Questionnaire26 (PHQ) was used to measure depression, panic disorder, and generalized anxiety disorder. The PHQ is a self-administered questionnaire that has been shown to have diagnostic validity for the Diagnostic and Statistical Manual of Mental Disorders—Version IV27 comparable to the PRIME MD, a structured diagnostic interview for medical settings administered by clinicians.28 There are no published thresholds for diagnostic indicators for the panic disorder and generalized anxiety modules of the Patient Health Questionnaire. We developed criteria for these disorders in consultation with 1 of the developers of the questionnaire (Kurt Kroenke, personal communication, 2004).
The Nine Symptom Depression Scale (PHQ-9) measures both presence and severity of 9 depression symptoms of the DSM IV over the past 2 weeks, and it is capable of yielding a diagnosis.29 Responses for each question range from “not at all” (scored as 0) to “nearly every day” (scored as 3), with a highest possible score of 27. In keeping with established norms,29 severity categories were assigned as follows: 0–5 = none; 6–10 = mild depressive symptoms; 11–14 = moderate depressive symptoms; 15–19 = moderate to severe depressive symptoms; 20+ = severe depressive symptoms. The score used in this study (PHQ-9 > 10) to indicate a screen positive for a major depressive episode has high specificity for major depression, strong convergent validity with other established measures of depression and function, and strong construct validity for the general population.26–30
For the panic subset of the PHQ, the patient responds to a screening question, “In the last 4 weeks, have you had an anxiety attack—sudden feeling of fear or panic?” If she answers in the affirmative, the patient is asked the following question, “Has this ever happened before?” If she answers in the affirmative, then the patient is asked to respond to 3 more questions with a yes or no answer, 1) “Do some of these attacks come suddenly out of the blue?”, 2) “Do these attacks bother you a lot, or are you worried about having another attack?”, 3) “During your last bad anxiety attack, did you have symptoms like shortness of breath, sweating, your heart racing?” A positive score of >4 for panic disorder results from an affirmative response to the first 2 questions, plus affirmative responses to 2 or more of the 3 questions following from the first 2 questions (Kurt Kroenke, personal communication, 2004).
To assess presence of generalized anxiety disorder, the screening test asks the patient to respond to the screening question, “Over the last 4 weeks, how often have you been bothered by any of the following problems, feeling nervous, anxious, on edge, or worrying a lot about different things?” (0 = not at all, 1 = several days, 2 = more than half the days). If the patient checks “several days” or “more than half the days,” she is asked to respond to 5 more questions concerning symptoms (ie, restlessness, fatigue, poor sleep, muscle tension, and concentration). A score of 8 or above indicates a positive screening score for generalized anxiety disorder (Kurt Kroenke, personal communication, 2004).
To screen for post traumatic stress disorder, we used the 4-item Primary Care PTSD Screen (PC-PTSD),31 which has demonstrated high specificity and sensitivity when compared with other PTSD screening tools such as the PTSD Symptom Checklist (PCL)32 and the Clinician Administered Scale for PTSD (CAPS).33 The PC-PTSD has a binary (yes or no) response pattern. The patient is asked, “In your life, have you ever had any experience that was so frightening, horrible, or upsetting that, in the past month, you… a) Had nightmares about it or thought about it when you did not want to?, b) Tried hard not to think about it or went out of your way to avoid situations that reminded you of it?, c) Were constantly on guard, watchful, or easily startled?, d) Felt numb or detached from others, activities, or your surroundings?” An affirmative response to any 3 questions, generating a score of ≥3, indicates PTSD.31 Several months after initiating the screening procedure, we added a question to identify trauma symptoms specifically related to a cancer diagnosis, “If you answered ‘yes’ to any of the previous 4 questions, is the event(s) that you were referring to related to your current breast cancer diagnosis, or some other previous event, or both?”
Past and current psychotropic medication use was assessed by asking for current and lifetime use of “medication for anxiety, depression, or stress” (yes or no). A history of depression prior to cancer diagnosis was assessed by the 2 stem questions from the PHQ-9 that ask the patient whether in the past she had ever experienced depressed mood or loss of interest or pleasure for at least 2 weeks (yes or no). Psychiatric inpatient treatment history was assessed by asking the patient if she had ever been admitted to a hospital for stress or mental health disorders (yes or no).
Current smokers were asked to indicate the number of packs smoked each day and number of years they have smoked. Past smokers were asked to indicate the number of packs smoked each day, the number of years they smoked, and the amount of time since they last smoked. Patients were asked how many drinks of alcoholic beverages they consume each week. Those who responded ≥5 drinks were asked also to completed the 4-item CAGE questionnaire34 used to detect alcohol dependence. Items include, 1) “Have you ever thought you should cut back?; 2) “Have you ever felt annoyed by people criticizing your drinking?”; 3) “Have you ever felt guilty or bad about your drinking?”; 4) “Have you ever had a morning eye-opener to relieve hangover or nerves?” Items are scored 0 (no) or 1 (yes), and a total score of 2 or above is coded positive for alcohol dependence.
The SF-8 (short form with 8 questions) is derived from the Medical Outcomes Short Form 36 (SF-36), a multidimensional measure of health-related function. The SF-8 comprises 8 subscales: Physical Functioning, Role-Physical, Bodily Pain, General Health, Vitality, Social Functioning, Role-Emotional, Mental Health,35, 36 and 2 summary scores for physical and mental health-related ability to function. The SF-8 is scored using a normative algorithm, based on a general US population, with a mean score of 50 and standard deviation of 10. There are 5 or 6 possible responses to each question, answered for the past 4 weeks, and responses are weighted based on the algorithm.
Comorbid medical conditions were collected through use of the Self-Reported Charlson Comorbidity Index.37 The Charlson Comorbidity Index is a validated instrument created to assess comorbid conditions that may alter the risk of mortality. This index was developed empirically, based on 1-year mortality from an inception cohort, and then tested for its ability to predict risk of death in a cohort of 685 patients who were treated for primary breast cancer. Whereas the original index was based on chart abstraction from the medical record,37 the version used in our screening was a self-reported version validated in a subsequent study.38 The questionnaire used in our clinical assessment was modified to exclude questions on human immunodeficiency virus (HIV) and acquired immune deficiency syndrome (AIDS) because of concern for patient privacy. Scores ranged from 0–29 in the original self-reported version, but there was an upper limit of 23 when HIV and AIDS were excluded.
Descriptive statistics were computed for all variables. Demographic and clinical characteristics were summarized as the frequency (percentage) of endorsed responses. The Patient Health Questionnaire, PC-PTSD, CAGE, SF-8, and self-reported Charlson Comorbidity Index instruments were summarized according to recommended algorithms. Student t-tests were used to assess group differences in continuous outcomes; reported P values were associated with 2-tailed tests of significance. SAS statistical software, version 9.1 (SAS Institute, Cary, NC) was used for all statistical analyses and results.
The mean age of patients was 57.4 years (SD, 12.3 years) (Table 1). The majority of patients were married (65%), employed full or part time (55%), and white/non-Hispanic (96%) with at least some college education (71%). Other than breast cancer, patients reported few co-occurring medical conditions on the Charlson Comorbidity Index; the most common were asthma (8%) and diabetes (6%).
|Mean age, 57.4 y; SD, 12.3 y|
|Divorced or separated||34 (14)|
|Full time||96 (41)|
|Part time||34 (14)|
|White, Non-Hispanic||227 (96)|
|White, Hispanic||3 (1)|
|Asian/Oriental/Pacific Islander||2 (1)|
|Some high school or less||9 (4)|
|High school or GED||60 (25)|
|Some college||73 (31)|
|College or graduate degree||94 (40)|
Measured on the 10-point Distress Thermometer scale, distress levels were relatively high (mean = 4.8, 2.6), with 41% of patients rating themselves >5, which indicated clinically meaningful distress levels (Table 2). Of the patients scoring >5 on the distress thermometer, the most commonly reported source of distress was “emotional causes” (100%) related to breast cancer. Categories of emotional distress endorsed by these patients included, worry (89%), fear (82%), nervousness (78%), sadness (61%), and depression (50%). Emotional sources were followed by distress related to uncertainty about treatment (96%), physical symptoms (81%), practical life problems (63%), family problems (52%), and spiritual crises (9%).
|25||Distress screening, scale: 0–10|
|Mean (SD) distress, 4.8 (2.6)|
|Distress thermometer >5||96 (41)|
|Major depression (PHQ-9 >10)||25 (11)|
|Moderate (11–14)||10 (4)|
|Moderate to severe (15–19)||9 (4)|
|31||Severe (≥20)||6 (3)|
|26||PC-PTSD (≥3)||24 (10)|
|26||Generalized anxiety disorder (≥8)||24 (10)|
|1 positive||70 (30)|
|2 positive||18 (8)|
|3 positive||14 (6)|
|4 positive||7 (3)|
|5 positive||1 (0)|
|Patient total, ≥1 positive||110 (47)|
The prevalence of positive screens for psychiatric disorders ranged from 7% to 11%, with major depression as the most prevalent diagnosis (n = 25; 11% of entire sample) (Table 2). Of these patients, 15 scored >15 on the PHQ-9, which indicated moderately severe to severe depression. Of the entire sample, 21% were taking psychotropic medications. Of the 25 patients who met screening criteria for major depression, 14 (56%) were taking a psychotropic medication.
Of the 24 (10%) patients meeting screening criteria for PTSD, 8 (33%) were taking psychotropic medication. Because of adding questions regarding the specific event linked to PTSD several months after initiating the screening, data for these items were available for only 21 of the 24 patients that screened positive for the disorder. Of these 21 patients, 15 (71%) indicated that their symptoms were related to a combination of their cancer diagnosis and a previous event, 4 (19%) indicated their symptoms were related specifically to the breast cancer diagnosis, and 2 (10%) indicated a previous event.
Prior psychiatric illness and/or treatment was relatively common, with 36% of patients reporting taking psychotropic medications in the past, 27% reporting at least 1 previous depressive episode, and 6% reporting at least 1 previous psychiatric hospitalization.
All mental health screening instruments (distress, depression, PTSD, panic disorder, generalized anxiety disorder) had established thresholds (see Materials and Methods section) to identify patients who could benefit from further assessment. In total, 110 of 236 (47%) presurgical breast cancer patients exceeded thresholds on 1 or more screening instruments (Table 2), whereas 40 (17%) exceeded thresholds on 2 or more instruments (Table 2). The associations between screening instruments were variable. For example, 24 of 25 (96%) depressed patients also exceeded the threshold on the distress thermometer; however, only 17 of 24 (71%) PTSD patients had a distress thermometer score of >5.
Examination of subscales from the SF-8 revealed that patients as a group were within the general population norms for indices of physical and mental function (mean = 50; SD = 10). However, patients who screened positive for major depression (PHQ-9 score of > 10) were significantly lower on the physical component score (PCS) than patients who screened negative for major depression (mean score, 46.0 versus 54.3, P < .01) (Table 3). This mean score is not in the range of clinical significance (PCS < 40), although a higher percentage of depressed patients than nondepressed (24% and 5%, respectively) scored in this range. Patients who screened positive for major depression were both significantly statistically and clinically lower on the mental component score (MCS < 40) than patients who screened negative (mean score, 31.1 versus 49.5, P < .01). Again, a significantly higher percentage of depressed patients than nondepressed (84%, 14%, respectively) scored below 40 on the MCS.
|Condition||Screen (−)||Screen (+)|
|Depression||(N = 211)||(N = 25)||t||P|
|PCS mean (SD)||54.3 (6.9)||46.0 (11.2)||3.6||<.01|
|MCS mean (SD)||49.5 (8.2)||31.1 (8.1)||10.6||<.01|
|PCS < 40||5% (n = 11)||24% (n = 6)||11.8||<.001|
|MCS < 40||14% (n = 30)||84% (n = 21)||64.3||<.001|
|PTSD||(N = 212)||(N = 24)||t||P|
|PCS mean (SD)||54.1 (7.3)||47.1 (9.9)||3.4||<.01|
|MCS mean (SD)||48.8 (9.1)||36.9 (10.5)||5.3||<.01|
|PCS < 40||6% (n = 12)||21% (n = 5)||7.4||<.01|
|MCS < 40||16% (n = 34)||71% (n = 17)||38.2||<.001|
Patients who screened positive for PTSD were significantly lower on the PCS than patients who screened negative for PTSD (mean score, 47.1 versus 54.1, P < .01), although, again, this mean score is not in the range of clinical significance (Table 3). PTSD patients, however, had a higher percentage who scored below 40 on the PCS than non-PTSD patients (21%, 6%, respectively). The PTSD patients were significantly statistically and clinically lower on the MCS than screen-negative patients (mean score, 36.9 versus 48.8, P<.01). In addition, PTSD patients had a higher percentage who scored below 40 on the MCS than non-PTSD patients (71%, 16%, respectively).
A minority of patients were current smokers (n = 28, 12%) with a mean (SD) pack-year history of 29.0 (21.2) years, although 53% of nonsmokers reported having smoked at one time with a mean pack-year history of 19.3 (22.2) years. Current drinking habits were moderate with a mean of 5.3 (4.2) drinks per week for the 49% of patients who did drink, with only 10 (4%) patients screening positive on the CAGE.
In this article we present our findings, collected in the context of routine care before surgical consultation, for psychosocial and health-related ability to function in newly diagnosed breast cancer patients. These data from research-based screenings are unique in 2 ways. First, nearly every (98%) patient who entered the breast care program was captured, so we did not have to rely upon a convenience sample or self-selected patients willing to sign an informed consent. Second, patients were screened before consultation with their surgeons, which allowed presurgical identification and appropriate intervention by clinical care managers.
Overall, this population was well-educated, medically uncomplicated, and enjoyed a high level of function. However, several notable findings deserve further discussion. In total, 47% of patients exceeded the established mental health threshold because of high distress, psychiatric, or substance abuse criteria. Distress levels were relatively high on the Distress Thermometer (mean = 4.8 on a 0–10 scale) with 41% of the population rating themselves above a 5, which is indicative of a clinically significant level of distress.16 For these patients, distress related to psychological issues such as worry, nervousness, and depression (100%) and uncertainty about decisions (96%) were the most commonly endorsed, although physical (81%) and practical (63%) problems were also prevalent.
Similar to research with other cancer and medical populations, 11% of our sample reported symptoms that suggested they were suffering from an episode of major depression, with 60% being moderately severely to severely depressed. This level of depression requires intervention, either pharmacologically or psychotherapeutically, or both. Although 56% of these patients were already being treated with a psychotropic medication, they still met screening criteria for major depressive disorder. Because of the nonspecific nature of the psychotropic medication assessment question, (“Are you currently taking any medication for anxiety, depression, or stress?”) we cannot determine the specific type of medication (eg, anxiolytic, antidepressant, etc), nor the specific reason for the prescription. That these patients were still highly symptomatic, however, suggests they were being only partially treated at best. The Distress Thermometer was associated with detection of a substantial percentage of psychiatric syndromes, especially depression (96%), and further examination of the Distress Thermometer as a screening instrument for psychiatric disorders seems indicated.
At our site, patients enter their responses to the mental health screening instruments on a touch-pad computer, which automatically generates a report summarizing the responses and flagging out-of-range responses. These reports are electronically mailed to members of the patient's care team, which includes social work care managers. If a threshold is exceeded, the care manager tailors interventions accordingly. Interventions can range from assisting the patient with practical issues, such as childcare, transportation, or health insurance, to providing educational materials, offering individual or family counseling, and making referrals to psychiatric services, substance abuse programs, or support groups. Our care managers have commented that the screening instruments have often successfully identified underlying emotional disorders that otherwise might have gone undetected.
Health-related ability to function for the entire cohort, as measured by the SF-8, was within the range expected for the general population. This is not surprising given the newness of the diagnosis, pretreatment status of the patients, and relative absence of medical comorbidities. However, for patients with comorbid depression, although their physical function was by and large not impaired, their emotional function was severely affected. These patients scored on average nearly 2 standard deviations below the population mean for impairment to function because of emotional problems, with depressed patients outnumbering by nearly 6 times the number of nondepressed patients scoring in the impaired range (84%, 14%, respectively). These patients were experiencing substantial emotion-related morbidity in their day-to-day lives. With an arduous treatment and recovery process ahead of them, immediate intervention and close follow-up would be indicated to minimize interference from comorbid depression with optimal cancer-related treatment outcomes.
Emotional and physical function impairment was also found for the PTSD subsample, although the degree was not as pronounced as for the depressed patients. It is possible that the rudimentary nature of the PC-PTSD screen (4 items) was less specific than the depression screen (9 items) and allowed less severe cases to be coded as positive. Studies of the sensitivity and specificity of the PC-PTSD screen with a breast cancer population should be a priority for the future.
In summary, this intake process identified about half of our patient population as having clinically significant distress levels or meeting criteria for a psychiatric syndrome. In particular, high prevalence rates were found not only for distress, but also for major depression and PTSD, along with clinically meaningful impairments to function related to psychiatric status. Although the use of psychotropic medication was prevalent in both sets of patients, these patients were still significantly symptomatic, suggesting that treatment needs to be improved. Future research should continue to develop and evaluate methods for detecting significant emotional distress and psychiatric disorders in cancer populations and to test innovative interventions for effectively treating and managing these disorders in the oncology setting.