Houts et al. published a series of articles in Cancer describing the unmet psychosocial needs of people with cancer and their families in Pennsylvania in the mid‒1980s.1, 2 Utilizing the newly formed Pennsylvania Cancer Registry (PCR), the authors surveyed 629 people with cancer in Pennsylvania to determine the extent to which existing resources and services were meeting patients' psychosocial needs. Results indicated that 59% of people with cancer in Pennsylvania reported at least 1 unmet need in the year after diagnosis. The findings of these studies were used to develop a series of psychosocial and educational programs to address the unmet psychosocial needs of cancer patients.3–8 Subsequent needs assessments have focused on the general unmet needs of cancer patients and their families,9–11 site‒specific unmet needs,12–16 unmet needs related to cancer stage,17–20 relationship‒specific issues,21, 22 ethnicity‒based issues,12, 23, 24 and unmet needs related to socioeconomic status25 and geographic factors.26
Although we currently have a relatively detailed picture regarding the nature of unmet needs, it is unclear how advances in psychosocial oncology have affected the level and type of unmet psychosocial needs experienced by cancer patients. Progress over the past 20 years in the understanding and treatment of cancer may have changed the relative importance of some issues, but has likely led to new psychosocial needs that have been unrecognized until recently. Survivorship issues, including psychosocial adjustment, relationship factors, work issues, reproductive issues, insurance coverage, and secondary cancers, have emerged as important concerns, but the effects of this awareness are not to our knowledge well documented.27–30 Possible negative outcomes of information technologies, such as increased disparities, are beginning to be articulated, although health outcome data concerning the “digital divide” are scant.28, 29, 31–34 Finally, managed care and reimbursement issues have altered the way that cancer care is delivered, and a better understanding of how these changes have affected the needs of patients and families is warranted.
It is not clear whether psychosocial resources for people with cancer are distributed adequately or evenly across racial and age groups, geographic locations, or cancer sites.3544 Furthermore, to our knowledge, little is known regarding the factors that predict who is at higher risk for physical, psychological, or social problems after cancer treatment.27, 45, 46
To begin to address these gaps, this article describes a population‒based study aimed at identifying the unmet psychosocial needs of cancer survivors, understanding the distribution of these needs across subgroups of survivors, developing a model for predicting unmet needs, and comparing unmet psychosocial needs in 2005 with those reported by Houts et al. in 1986.1 We recognize that there are multiple definitions of the term “cancer survivor.” However, in this article, we will use the Institute of Medicine definition of a cancer survivor as a person who is in that phase of care after primary treatment.27
- Top of page
- MATERIALS AND METHODS
Nearly two‒thirds of cancer survivors in Pennsylvania reported experiencing at least 1 unmet psychosocial need 3.0 to 4.5 years from the time of diagnosis, particularly emotional, physical, and treatment‒related needs. Nearly half of the current study sample experienced >3 unmet needs. Greater unmet need was associated with younger age, female sex, the presence of comorbidities, lower income, and being single. Respondents who were younger and reported multiple comorbidities reported the most unmet needs. Somewhat unexpectedly, race, geographic location, and type of cancer were found to be unrelated to need. Results suggest that the overall level of unmet needs is at least as great as that experienced in the mid-1980s, and that insurance, employment, information, and homecare are especially likely to have increased. This is despite significant public health efforts to address the needs of cancer survivors.
A number of limitations must be considered. Our response rate was modest, and respondents were more likely to be younger, married, and a nonminority, and less likely to live in an urban setting and have an initial diagnosis of lung or colorectal cancer compared with nonresponders. Of these, age and marriage were correlated with unmet needs, although our weighting strategy should have decreased differences between our study sample and the PCR. Reporting biases also should be considered, although it appears equally likely that socially desirable responses could bias our results toward the null if needs are viewed as a sign of failure to cope adequately. In addition, many of our effects, although significant, are of moderate magnitude, suggesting that unmeasured variables account for the majority of variance in unmet needs.
That unmet needs have remained high since the survey by Houts et al.1 20 years ago may be due to a variety of factors. Expectations regarding the cancer experience, changes in the specific types of needs, and greater role disruptions resulting from the increased reliance on home‒based care may account for some need. In addition, changes in the ease with which survivors can access information without concomitant changes in how this information can be evaluated, weighed, and digested may result in greater levels of informational need now than in the recent past, and data from the current study are consistent with this theory. Similarly, much has been accomplished in the past 20 years to improve the number and variety of resources available to cancer patients and survivors, but this increase in access may not be reflected in decreased need. Working with the American Cancer Society, we identified numerous cancer‒related resources in areas of high need.55 Nevertheless, cancer care services that are rendered primarily to outpatients frequently are not covered by third‒party payors. Even when “quality‒of‒life” services are reimbursed, accessing services may be difficult given the competing demands of role functions.
Although resources such as support groups, information services, patient education tools, and homecare programs exist, many gaps persist. It remains unclear whether identified gaps between services and need is due to a mismatch between available services and perceived need, services that cannot keep apace with “new” needs, access issues related to resources, or a lack of information regarding resources. Patient expectancies also may play a role in psychosocial adjustment.46 Clinical guidelines mandating that no cancer patient should experience distress that goes unrecognized and untreated (National Comprehensive Cancer Network [NCCN] Clinical Guidelines, available at: http://www.nccn.org/professionals/physician_gls/PDF/distress.pdf [Accessed June 4, 2007]) may alter survivors' expectations concerning the adequacy with which their needs can be met within the current healthcare system.
Our finding that younger cancer survivors with comorbid conditions experience the highest levels of unmet need deserves comment. Houts et al.46 reported similar associations. Younger age is consistently associated with increased psychosocial distress among cancer patients,56 perhaps because a life‒threatening illness is more likely to disrupt family and occupation roles and is generally unexpected. Physical comorbidity also has been associated with decrements in quality of life among cancer patients, although the majority of studies have focused on its influence on well‒being among older adults.57, 58 Increased needs in the presence of a comorbid condition may be due in part to combining needs due to cancer with those due to other illnesses. However, this combination is not simply additive. Resources needed to support people with preexisting illnesses may be stretched thin before the cancer diagnosis, and adding another serious illness may stretch them to the breaking point. Physical limitations, fatigue, or pain may be exacerbated by the cancer diagnosis as well. While receiving active cancer treatment, the person with cancer may have access to cancer support services; however, after the initial treatment, patients return to their primary care physicians, where specialized services may be absent. Lack of attention to the specialized needs of younger cancer survivors in these settings may, in part, explain the correlation at 3.0 to 4.5 years after diagnosis.
Although women report more unmet needs than men on univariate analyses, this correlation was found to be nonsignificant once variances attributable to age, income, staging, and comorbidity are taken into account. This suggests that need is correlated with deficits in resources and disruptions in role function, for which gender is only a crude marker. Regardless of gender, decreased resources, increased demands, and more psychosocial stressors result in greater unmet need.
Although we did not directly assess differences in the absolute level of need between our sample and that of Houts et al.1 it does not appear that unmet need has decreased, and in fact needs may have increased in some domains. Differences may be attributed in part to the fact that our survey took place between the third and fifth year after diagnosis and the study by Houts et al.1 took place in the year after diagnosis. However, this difference most likely underestimates the increased amount of unmet need because one might expect distress to be more acute during the period of active treatment. In addition, we found no correlation between the level of unmet need and time since diagnosis.
The study by Houts et al.1 found no association between race and unmet needs, but did report that lower‒income patients reported more economic unmet needs. We also did not find differences related to race; however, lower income was found to be associated with 10 categories of unmet need. This difference suggests that income may have become more important in coping with cancer over the 20‒year period between the study by Houts et al.1 and the current report. We need to better understand what the pathways are through which income affects unmet need. Access is an obvious explanation, but more covert (and less easily captured) factors such as literacy and stigma also must be considered.
Rates of perceived unmet needs among persons with cancer have remained high over the 20‒year span between the study by Houts et al.1 and the current survey. There is evidence suggesting that unmet needs for insurance, employment, information, and homecare have increased during this period. These high rates exist despite increased attention to controlling the side effects of treatment and managing pain, and an awareness of the psychosocial effects of cancer and cancer treatments on patients and their families. The reasons for the continuing high rates of reported unmet needs are complex and should be the subject of future study. Future research should include objective as well as subjective measures of the need to clarify the meaning and causes of patient reports.
Correlates of unmet needs also were found to be similar in the 2 surveys, although there is some suggestion that income may be playing an increasingly important role in coping with cancer. The prominence of comorbid conditions as a predictor of unmet needs in the years after cancer treatment suggests that the primary care setting may be 1 place in which service referral can occur, because the majority of cancer survivors return to their primary care providers after acute treatment. Future research also needs to consider the longitudinal implications of unmet needs on psychosocial outcome, physical problems, and survival, especially for younger patients.