Screening for distress, the 6th vital sign: common problems in cancer outpatients over one year in usual care: associations with marital status, sex, and age

The National Comprehensive Cancer Network (NCCN) defines distress as an unpleasant psychological, social, and/or spiritual experience that interferes with effective coping [1]. Despite prevalence rates that can range from 35-60% in the cancer population [2‐5], distress often goes unrecognized [6‐10] and can have negative implications for patients including reduced health-related quality of life [11], greater long-term distress [12], poor satisfaction with medical care [13], and possibly reduced survival [14, 15]. Common psychosocial, practical and physical problems may amplify feelings of distress, and their assessment offers clinicians opportunities to refer patients specifically to professionals who can address these patient problems. In this paper, we examine predictors of these common problems in order to facilitate strategic targets for offering psychosocial service.

Patients most frequently report physical, emotional and informational problems [11, 16‐18]; and these problems consistently predict clinical distress [3, 11, 17‐19]. Some, but not all [18] researchers find that practical problems may also predict distress [3, 11, 19]. Other distressing problems may include social [20], financial [11, 16, 21], cognitive [3], sexual [22, 23] and family related problems [3, 11, 16, 24], and problems relating to the quality of cancer care received including care coordination [25] and relationships with health professionals [16, 24, 25].

Given the high prevalence of distress and common problems that may be present in the cancer population and the detrimental impact both can have on wellbeing, analyzing common problems and identifying characteristics of people who are more likely to report particular problems could facilitate planning for targeted clinical services [26]. Unmarried, younger and female patients report greater practical problems in mixed [21, 24, 27, 28], as well as diagnostically homogenous populations [23, 29, 30]. Marriage also appears to buffer the number of problems patients report [27, 31‐34]. These associations include problems with finances [21, 24, 27, 28], insurance [21, 27, 31], drug coverage [35] and employment issues [27, 30, 31]; as well as transportation [33], help around the home [27, 33] and childcare [24]. Others report no differences in problems according to demographic characteristics [36].

Women report greater psychological, patient care and support [24, 26, 37], and sexuality problems [37]. Younger people report more psychological [24, 27, 31, 37], patient care and support [38], information [24, 27, 31, 37], social [21, 27, 31], physical and treatment-specific [21, 31], sexual [23, 30, 37, 38] and spiritual problems [24, 27, 31, 37]. The association between marital status and psychosocial problems is less consistent; perhaps due to differing study settings and populations [16, 29, 39‐41]. Variations in the measures used to assess problems, the population assessed, and the timing of assessments may contribute to these conflicting results [42].

Very few studies examine the longitudinal prevalence of problems and the awareness or use of clinical programs by patients who report these problems [26, 38]. Of the studies that examine age, gender and marital status as predictors of a range of patient outcomes, none examines the interactions between these demographic variables. For example, do younger women experience psychological problems in the same way as older women and do these associations change over time? By examining how these demographic variables interact with each other, we may be better equipped to identify subgroups of people that may be more at risk for specific problems and facilitate the early identification and management of these issues before they become too overwhelming. As part of a comprehensive Screening for Distress the 6th Vital Sign in Cancer Care Program [43] adopted by a number of provinces across Canada, this study examined a naturalistic course of usual care without triage to document clinical outcomes and gaps in service. We have previously published usual-care baseline and longitudinal trajectories of distress, anxiety and depression, pain and fatigue [44], and this analysis not only adds to the literature, but also facilitates clinicians’ ability to directly modify the services they offer.

Because few studies examine common problems over time and their associations with distress, we first check these associations. We then test our primary hypotheses specifically examining associations between age, gender, and marital status as they interact and predict psychosocial and practical problems. Lastly, we examine secondary hypotheses relating to past, present, and future resource use.

We found that marital status buffers or reduces common problems, as it often buffers cancer patients’ distress [3, 40] and unmet needs [27, 31‐34]; however, age differences impacted this relationship. Similarly, older age buffers common practical problems but differences exist between males and females. Lastly, not only younger women, but also younger men reported higher past and future use of our Psychosocial Resources Program. Stage of disease and type of treatment did not explain our results.

Analyzing common problems over time and identifying people at risk for common problems may guide clinicians in targeting interventions toward people who need them most [26]. It allows us to deconstruct elements of need so that we can offer appropriate practical as well as psychological help. For instance, in this study our findings suggest that providing younger single males and females access to practical support (e.g. help with finances), and younger men but women of all ages greater access to psychological support may be beneficial. These associations and interventions based on these findings may also impact distress as we found that the more psychosocial and practical problems patients reported, the higher their distress, anxiety, and depression levels at baseline. Psychosocial and practical problems declined over time and correlated with declines in distress, depression and anxiety.

This study is the first longitudinal investigation of common problems in patients new to the TBCC. The large sample size and longer follow-up period have enabled us to refine previous knowledge in this area. Similar to others, we found that common practical problems include finances and drug coverage [21, 27, 29, 35], with work/school only a concern for younger people [27]. At baseline older people had considerably fewer problems, as did married people, resulting in the highest prevalence of practical problems in young, single people, particularly women. Marital status findings are consistent with other reports in the literature [21, 27, 33, 53], but we add interactions with age to extend this literature. Younger and single people may have lower incomes, less financial stability if they need to leave work for long periods of time, and greater responsibility for young children. Cancer may disrupt their ability to handle all of these competing demands, so practical help might lead to the most positive improvements.

For psychosocial problems, younger patients reported greater psychosocial problems than older patients, while women reported greater psychosocial problems than men at baseline. Few psychosocial resources target specifically young men and women with support groups or counseling interventions [54, 55], and anecdotally young patients often report that they feel they have nothing in common with older people with cancer in support groups. Again, other reports have consistently shown more psychosocial problems in women [24, 26, 37], perhaps due to an under-reporting bias in males [34, 56] or due to a greater tendency in women to focus on processing emotions [57].

The picture becomes more complex when we look at our novel data investigating changes over 12 months. In general, practical and psychosocial problems improved, with the sharpest improvement in the first three-months, although some problems did not ease on their own. Patients endorsed financial problems often and endorsements remained high over time, even though in Canada patients have public health benefits. Younger, single patients endorsed the most practical problems at baseline and remained the most elevated at 12 months. Younger females endorsed the most psychosocial problems at baseline and remained the most elevated at 12 months. Perhaps these results indicate that neither group currently receives the help they need. Interventions that include help with practical aspects of going through treatment for young single women might improve their distress.

Of the psychosocial problems, worry about friends and family, and difficulty with sleep, remained high as patients underwent a variety of treatments. It is interesting to note that the most common psychosocial problem endorsed was worry about friends and family, a difficulty rarely addressed by health practitioners. Endorsement of this worry also did not decline dramatically over time, with over 30% still reporting it at 12 months. Older males improved the least in psychosocial problems; however, they did not report high levels of problems at baseline.

Lastly, report of use of the Psychosocial Resources Department at baseline reflected some of these findings. More young single men than young single women reported using Psychosocial Resources, whereas the reverse was true in older participants. This is an unusual finding, perhaps reflecting the prominent erectile dysfunction services offered, and further investigation could document which services young men accessed throughout the year. More young and single men and women reported interest in future use. These results mimic their reports of greater problems. However, older patients may not access services due to experiencing greater barriers in transportation or low caregiver help [58]. Apparently, being married reduces rates of reported current and future interest in the use of services, reflecting some of the buffering seen in the low endorsements of practical problems in married people. Future research could also examine the tumour type of the younger people endorsing use of psychosocial resources which is beyond the scope of this paper. Additionally, future research could examine whether older patients perceived greater barriers to access.

These new findings have implications for cancer-care teams who may want to prioritize resources at the time of initial diagnosis toward helping young, single men and women access services to help with resource and financial concerns. For younger single patients, a diagnosis of cancer could result in significant loss in income while expenses such as child care, food, and transportation continue to accrue. These findings challenge us to examine whether resources provided are sufficient for the burden these groups experience. Linking with community programs, providing appropriate childcare or housekeeping services, and enhancing support for basic needs could help significantly reduce the burden and distress of these at-risk groups. Older single patients improved to a similar level as older married patients in this study--they may be more able to handle practical concerns as they suffer less income loss if retired, tend to be more financially stable, and have good medical coverage here in Canada. However, examining carefully barriers to access is important to consider when providing services to older people.

Women of all ages need psychosocial support at the time of diagnosis, but we should not neglect younger men, as they have high practical needs at diagnosis and are using our Psychosocial Resources even more than young women. Innovative psychosocial programs could target this group that is at-risk for sustained problems, since historically men are less likely to access supportive care services on their own. Our care team offers specific therapy programs for men with prostate cancer who have erectile dysfunction which may, in part, account for this unusual finding [59, 60]. Discussion groups or educational opportunities, that provide a point of entry into the care system, may also be attractive to men. When administrators better understand the needs of a range of patient groups, they can develop more suitable and effective programs. Clinically, these data identify problems and risk groups so intervention can happen earlier--which may result in improved coping and savings to the health care system as fewer crisis interventions may occur.

Although this study has strengths, including a large sample size, relatively high accrual rate, varied cancer diagnoses, and 12-month follow-up, only 72% of eligible participants consented to the study. There was a significant drop-out or missing data rate, which resulted in only 56.5% of the original sample (676/1196) being assessed at the final follow-up: some lost due to death or progression of illness, others missed follow-ups, a small group did not continue. However, HLM analysis mitigated this loss in generalizability by using data from all patients who provided at least one assessment in the analysis. The sample is also not representative of breast and lung patients because these tumour groups attended an outpatient clinic in a different location. This study used a measure that assessed the presence of problems in the week prior to questionnaire completion. Jacobsen et al. (2005) suggests that assessing those problems for which individuals want assistance may be more beneficial [18].

Lastly, because this is a mixed cancer sample in a usual-care cancer centre setting, the frequency of cancer types varied. Our goals were to provide evidence for a general cancer population and services offered in a general cancer setting. Certain cancers (testicular, prostate, breast) are gender-specific. It is not possible to adjust for this in an analysis examining gender. However, a larger number of participants in the current study had cancers that are not gender-specific (747 of 1196). Some cancers are age-correlated, and again, it is not possible to adjust for this in an analysis examining age. Other cancers are too rare in this sample for adjustment. As such, age effects may in part reflect the influence of those types of cancers that are age-dependent, and gender effects may in part reflect the influence of those types of cancers that are gender-specific. Further research could investigate larger samples of non-gender- and age-specific cancers.

As part of a comprehensive Screening for Distress program, we investigated these associations between common practical and psychosocial problems and interactions with marital status, sex, and age during usual care. Although patients completed a screening questionnaire, we did not triage with referral to psychosocial resources, as our goals were to provide a naturalistic picture. This work informs clinical programming decisions, and we have identified several risk groups (young single women, and young men) to target with innovative interventions.