More than half of the patients seen in a multidisciplinary outpatient lung cancer clinic reported clinically significant levels of distress. Distress was predicted by problems in the areas of family relationships, emotionalconcerns, information concerns, physical symptoms, and cognitive functioning. Specific symptoms predictive of distress included depression, anxiety, pain, and fatigue. Consistent with previous research, younger patients were more likely to report higher levels of distress along with an interest in obtaining help for distress [14
]. Distress levels were not associated with any other clinical variables, including stage of illness or treatment type. The lack of relationship between distress and stage and treatment variables is similar to previous reports [15
]. Specifically, illness prognosis and illness burden were unrelated to rates of distress in a sample of over 9000 cancer patients [15
]. These findings are also consistent with results from a recent study in which clinical variables were unrelated to depression in women treated for early-stage breast cancer [40
The high rates of distress reported by patients in this sample, many of whom were being seen for the first time or very early in their illness experience, are noteworthy for several reasons. First, patients may experience symptoms of distress beginning with the initial presentation to a lung cancer clinic, results consistent with findings reported by Cooley, Short and Moriarty [41
]. Second, in addition to physical symptoms, several other categories of problems (family relationship problems, emotional concerns, information concerns, and cognitive functioning) predict distress. The predictive power of these problem areas highlights the need to assess patients’ functioning beyond the usual practice of monitoring physical symptoms and performance status. The psychosocial nature of these problem categories also signal the importance of including symptom management and palliative care services in early symptom assessment. Third, the distress screening procedures followed in this study were easily implemented. Although not formally documented as a part of the present study, important information about symptoms experienced by patients seen in MLCP was conveyed quickly and efficiently to patients’ physicians and nurses. Clinic staff (CL, PM) and medical oncologists (SA) agreed that the screening improved delivery of patient care by facilitating referrals and calling attention to patient symptoms that may have otherwise gone unaddressed. Thus, although referral procedures and follow-up require further empirical evaluation, inclusion of routine distress screening, involvement of palliative care professionals or other staff dedicated to implementing the brief distress screening form, and treatment of distress symptoms based on a multidisciplinary approach are steps that will likely significantly impact lung patients’ symptom burden. Appropriate identification and treatment of distress and the symptoms contributing to that distress are likely to result in higher quality of life, satisfaction with care, smoking cessation efforts, and better adherence to treatment [29
]. Thus, documenting rates of referral based on distress screening, specific treatment interventions, and quality of life and clinical outcomes following distress screening are important next steps to improve clinical care and evaluate patient outcomes following utilization of distress screening.
In the present study, we responded to calls in the literature to expand the Symptom Problem list that accompanies the DT and to utilize a homogeneous sample of patients rather than a sample of patients with mixed cancer diagnoses [16
]. Two problem categories predictive of distress, information concerns and cognitive problems, were novel to the problem list used in this study. Because of the predictive utility of information concerns and cognitive problems for level of distress, future distress screening instruments should include assessment of these two areas in addition to the current categories of problems on the NCCN distress thermometer [39
Compared to other studies using the DT to screen for distress, our sample consisted of large numbers of patients residing in rural areas (56%) and had a sizable number of patients older than age 65 (35%). Our use of an expanded Symptom Problem List and assessment of the desire to help were also novel aspects to the study. Despite these strengths, several limitations should be noted. Although representative of the local geographic area in terms of patient ethnicity, our sample was limited in terms of ethnic diversity. In addition, our sample consisted of consecutive patients and distress screening was conducted as part of routine clinical service, and, as such, we did not collect clinical data on study decliners. Thus, we could not compare decliners to participants in terms of treatment type or stage of disease. Treatment type and stage of disease data for study participants consisted of broad rather than specific categories, and the number of individuals in certain categories (e.g., early stage disease) was small; thus future research should obtained detailed medical records to replicate the lack of relationship between clinical variables and distress found in the present study. Importantly, previous research has not found associations between clinical factors and distress ratings [15
], and thus our results are consistent with these prior reports. Finally, we did not systematically document and follow referral patterns and treatment outcomes for patients referred to the clinical services for treatment of distress. This information would be useful to draw more accurate conclusions about the clinical utility of distress screening. Keller and colleagues recently conducted a study investigating both identification of distressed patients and rates of referral to support services in a sample of patients with mixed cancer diagnoses. Their results suggest that even if distress is recognized by the medical staff, only a small percentage of patients with distress (31%) were actually referred for services [43
Similar to the rates of referral to services in the study by Keller and colleagues [43
], less than one-third of patients (29.3%) with clinically-significant levels of distress in the present sample reported an interest in receiving help for their symptoms. Our results indicate a relationship between distress level and interest in support services; however, previous research did not find such a relationship between distress level and interest in services [44
]. Support services may not be palatable for a number of reasons, including stigmatization of needing psychological support [19
], lack of knowledge of available support for distress symptoms, or lack of interest in what is perceived as traditional “talk therapy” [17
]. Certain characteristics of our sample may have further contributed to the relatively low percentage of individuals who desired help for their symptoms, including significant percentages of older individuals and individuals from rural areas. While age was the only significant factor related to wanting help (with younger patients reporting more interest in wanting help), other factors associated with rurality may be related to not wanting help for distress, including transportation concerns or misunderstanding of what support services may entail [45
]. Future efforts should be targeted toward clarifying additional barriers to seeking support as well as making services for treating distress as palatable as possible.
Routine screening for distress using brief self-report instruments has been criticized as resulting in elevated rates of distress and potentially inappropriate use of clinical resources [47
]. Importantly, however, using empirically established cut-off scores on the DT and following treatment recommendations established by NCCN guidelines for distress management will likely result in appropriate use of clinical resources. Specific areas of concern to the patient, such as uncontrolled pain or nausea, can be assessed, identified, documented, and then addressed through referrals and treatment according to clinical practice guidelines. Although certain symptoms that drive up a patient’s level of distress may remit on their own over time (e.g., state anxiety and/or uncertainty over medical treatment options), documentation of these symptoms is a first step toward encouraging members of the medical team to inquire about these symptoms at future appointments. Physician recommendation of and referral to support services is paramount to patient initiation and uptake of such services, as patients value oncologists’ opinions of these services [48
]. In the past, the majority of oncologists did not discuss support or symptom management services with their patients [48
], or, if these services were mentioned, many patients did not recall hearing about these services [49
]. Oncology professionals may not have received training in communication skills and likely have limited resources and time available [50
]. More recently, evidence points to patients taking a more active role in inquiring about support services [51
], and thus awareness of the specific symptoms that contribute to levels of distress would allow for more accurate and appropriate referral in response to patient inquiries for services.
The present study highlights the prevalence of clinically-significant distress in patients being seen in a multi-disciplinary lung cancer clinic and suggests that screening for distress should occur at initial patient visits to appropriately identify and refer patients experiencing significant levels of distress. Clinical implications from the present results are clear—1) screening for distress of consecutive patients in a multidisciplinary lung cancer clinic is feasible, 2) distress screening in this setting will identify distressed patients, and 3) a significant number of patients in an outpatient clinic can be expected to meet clinical criteria for distress. Future investigations can investigate the effectiveness of appropriate identification of distress by prospectively assessing patients’ distress and measuring rate of referral, intervention, and psychological and physical outcomes. Moreover, specific strategies can be elucidated for making distress management services palatable to patients, busy oncology health professionals, and cancer clinic administrators.