SEARCH

SEARCH BY CITATION

Keywords:

  • measurement;
  • adolescent;
  • young adult;
  • oncology;
  • program development

Abstract

  1. Top of page
  2. Abstract
  3. General Characteristics of Metrics Relevant to AYA
  4. AYA With Cancer
  5. Survivorship
  6. Conclusion
  7. FUNDING SOURCES
  8. REFERENCES

Against a background of poorly coordinated provision of holistic care to the adolescent and young adult (AYA) cancer population, the Canadian National Task Force on Adolescent and Young Adult Oncology, which is supported by the Canadian Partnership Against Cancer and the C17 network, convened a workshop to formulate the components of a systematic approach to care for this age group. Because such a program will deflect scarce resources, it must be validated and justified by reproducible metrics. A subgroup of experts was convened, comprising attendees at the AYA workshop, including AYA cancer survivors. A substantial number of key, feasible, and consistent metrics were identified and are systematized, justified, and presented in this article. Prioritization from within this range will be necessary. Cancer 2011;117(10 suppl):2342–50. © 2011 American Cancer Society.

The particular issues pertaining to the adolescent and young adult (AYA) cancer patient population have only recently been recognized. In part, this oversight has resulted from their relatively good prognosis. In Canada, the observed 5-year survival proportion in this age group is estimated to be 83%.1 This age group contributed only 0.9% of potential years of life lost because of cancer in 2004 and, by contrast, contributed disproportionately to potential years of life gained from treatment by virtue of longevity. Despite these remarkable statistics, the impact of this age group on the health care delivery system is not generally appreciated. Optimizing the quality of the years of life gained is crucially important both to the survivor group and to payers. Thus, with the introduction of a structured, programmatic approach to providing care to AYA with cancer, the selection of outcomes and metrics appropriate to the validation and vindication of such an approach is particularly important.

Like in all evaluations of cancer outcomes, metrics that address both the quantity and the quality of these outcomes must be considered. Furthermore, these outcomes must reflect the interests of multiple stakeholders to optimize buy-in from patients, survivors, families, health care providers, and, particularly important, decision makers and funders. In the context of the larger cancer enterprise, it must be demonstrated that indicators of favorable “return on investment” or “bang for the buck” (ie, the efficient use of resources allocated to the AYA population) justify the additional resources required to optimize the care of the AYA population—resources that likely will come from other areas and, thus, will reduce opportunities in other health care domains.2

Successful collection and analysis of outcome metrics will depend on reaching a consensus around clear definitions of data elements and ready availability and feasibility of data collection. It is likely that an overall blueprint for service delivery will be established; however, because health care is a provincial jurisdiction in Canada, the implementation of that blueprint may vary in detail from province to province. So, it will be critical to have in place a systematic plan for the measurement of outcomes, and validation and comparability of metrics across these jurisdictions will be particularly important. Yet comparability should not trump the validity of such measures (ie, their ability to capture what we want them to measure). Many outcome metrics are possible and will need to be prioritized using comprehensive stakeholder input to gain acceptability. The potential metrics discussed are summarized in Table 1.

Table 1. Summary of Potential Outcomes and Relevant Metrics
CategoryOutcomesMetrics
  1. AYA indicates adolescents and young adults; HPV, human papillomavirus; QOL, quality of life; HRQOL, health-related quality of life; HUI, Health Utilities Index; PedsQL, Pediatric Quality of Life Inventory.

EpidemiologyCancer incidence in AYA and subpopulations•Disease incidence and stage
  •Ethnicity
  •Age
  •Sex
  •Geographic residence
  •Socioeconomic status
Screening and preventionImpact of screening and preventive measures on cancer incidence•Rates of HPV vaccination
•Trends in cervical cancer incidence
Access to careDelays from first symptom to treatment•Time of:
   ○First symptom
   ○First health care contact
   ○First oncology visit
   ○Diagnosis
   ○Initiation of treatment
 Impact of patient and health care system resource factors on access•Specialty of first health care contact
 •No. of providers available
Locus of careImpact of demographics on site of care; impact of site of care on survival (for varying age, diagnosis)•Site of care (pediatric vs adult; teaching vs community; AYA focus)
 Impact of site of care on quality of care/patient satisfaction•Patient satisfaction with age and developmental appropriateness of care
 Use of clinical trial/protocol therapy•No. of disease-appropriate and age-appropriate clinical trials open in cooperative groups
  •No. of AYA-appropriate clinical trials open per center
  •Proportion of incident cases enrolled in clinical trial
  •Use of protocol-guided therapy
SurvivalSurvival•5-Y overall survival
  •5-Y event-free survival
  •Cause of death (malignancy, complication of therapy, or other)
  •Early mortality
Psychosocial healthIdentification of poor psychosocial health in patient and caregivers•Psychosocial screening assessment of self-efficacy, locus of control, anxiety, and mood at:
   ○Diagnosis
   ○Midtreatment
   ○End of treatment
   ○Survivorship
 Impact of model of care or intervention on psychosocial health•Evaluation of patient's family/spouse/supporters psychosocial health
 •Use of mental health and support services
HRQOLImpact of cancer on QOL; identification of areas of decreased HRQOL where intervention needed; impact of intervention on HRQOL•HUI Mark 3
 •Young Adult Version of PedsQL instrument
Palliative careUse, cost, and efficacy of palliative care•Involvement/timing of palliative care/hospice
  •Locus of end of life care
  •Pain control
  •Caregiver costs; work time and income lost
  •Medication and care-provision costs
  •HRQOL measures
  •Family satisfaction
EconomicCost/benefit of care (in measurement that can be compared to other possible uses of resources)•Quality-adjusted life years
 •Healthy years equivalents
  •Willingness to pay
SurvivorshipLate-effect medical and patient-related outcomes•Incidence of late effects and chronic disease
  •Creation and use of a survivor care plan
  •Stage of late effect illness or recurrence at diagnosis (including rates of preclinical diagnosis)
 Compliance with guidelines/monitoring•Attendance at follow-up care (post-transition from primary oncologic care)
  •Patient knowledge of health history and risk
  •Patient satisfaction with care
  •Patient self-empowerment
  •Compliance with recommended monitoring guidelines (cardiac post anthracycline, breast cancer after chest radiation)
 Health care use•Cost of surveillance for and treatment of late effects
  •Volume and type of health care services used (scheduled vs emergency)
 Impact of model of care on outcomes, monitoring, and use•Cause-specific mortality
 •HRQOL
  •Measures of educational, social, and vocational achievement

General Characteristics of Metrics Relevant to AYA

  1. Top of page
  2. Abstract
  3. General Characteristics of Metrics Relevant to AYA
  4. AYA With Cancer
  5. Survivorship
  6. Conclusion
  7. FUNDING SOURCES
  8. REFERENCES

Both outcome and process measures should be included in the evaluation of AYA programs; focusing only on outcome measures, such as survival, would fail to evaluate the quality and experience of the cancer journey from diagnosis to survival or palliative care. Compared with the group ages 0 to 14 yeas, more of the classic components of the spectrum of cancer control are relevant in the AYA group and require monitoring. In particular, metrics for AYA outcomes should encompass prevention and early detection/screening.

Metrics should provide information relevant to research and to service planning and evaluation. The outcomes measured must be both comprehensive and holistic and should include population-based, clinical, psychosocial, and health care system components. Stakeholders are particularly insistent on this holistic construct of relevant metrics.

Different metrics will be needed for different components of the AYA enterprise. These components are viewed conveniently as encompassing 1) cancer diagnosed and treated in the AYA age range and 2) cancer survivorship, including adult survivors of cancer in childhood and AYA cancer survivors.

AYA With Cancer

  1. Top of page
  2. Abstract
  3. General Characteristics of Metrics Relevant to AYA
  4. AYA With Cancer
  5. Survivorship
  6. Conclusion
  7. FUNDING SOURCES
  8. REFERENCES

Many potential metrics across the cancer control continuum must be considered, and a key challenge is prioritizing which measures to use. Below, we describe a range of areas for which the development and collection of data on AYA are deemed relevant and from within which prioritization could occur:

Epidemiology and trends in incidence

There is an-ongoing and dramatic change in the demography of the Canadian population; new immigrants comprise a large and dynamic segment of Canada's population. Statistics Canada has indicated that, by 2031, nearly half (46%) of Canadians aged ≥15 years (up from 39% in 2006) will be foreign-born or will have at least 1 foreign-born parent.3 In Toronto, 63% of individuals will be from visible minority groups, as either immigrants themselves or as the first generation to be Canadian-born. A similar change is predicted for Montreal, Vancouver, and Calgary and probably is applicable in many major cities worldwide. Because some cancers have higher or lower incidence in particular ethnic populations, the increasing immigrant population may result in shifts in patterns of incidence. This shift, in turn, will have implications for resource allocation, both for direct care and in terms of ethnocultural perspectives on diagnosis, treatment, and consequences and the resultant requirement for appropriate psychosocial support. This changing demographic may be particularly relevant to the AYA population, because the next generation of AYA will represent the vanguard of offspring of parents of different ethnicities and may manifest a different distribution of malignancies than the current generation of AYA. Thus, careful monitoring of incidence trends will be important.

Implementation and appropriate use of screening approaches

There is little effective opportunity or need for screening in the group ages 0 to 14 years. Thus, the first opportunity to introduce screening into the cancer control spectrum occurs in the older AYA group. Cervical cancer screening is a prime example. Cervical cancer constitutes 9% of malignancies in the AYA group, although the incidence is declining.1 Although human papillomavirus (HPV) vaccination has been approved for use in Canada for females ages 9 to 26 years, the current AYA population may have been beyond the age of first sexual contact at the time of its introduction and, thus, remain at risk of developing cervical cancer. Because HPV vaccination is not mandatory in Canada, there also are issues of compliance. Rates of screening and demonstration of trends in incidence will be important to measure.

Access to care metrics

Access to care generally is acknowledged as an important AYA issue. Figure 1 depicts the continuum of access to care. The bolded italicized portion reflects host-driven, family-driven, and disease-driven determinants of access to care and lag times. These factors can be difficult to tease out, because the first symptom is notoriously difficult to define and is poorly recorded in hospital charts. However, this interval may represent a target for intervention through education aimed at the general public and specifically at the AYA age group. Consideration should be given to making the notation of first symptom a mandatory data field in any electronic health record that is developed.

thumbnail image

Figure 1. This chart illustrates the continuum of access to care.

Download figure to PowerPoint

Similarly, impact of the type of first health care contact merits study, because preliminary data suggest that the referral to oncology services happens more expeditiously when the first health care contact is the emergency room.4 Furthermore, although there are some data addressing the impact of age, disease, and geographic location on the interval between first symptom and first health care contact,4 further clarification of the role of these factors would be important.

The nonitalicized portion in Figure 1 addresses system and resource issues and represents areas that provide an opportunity for system improvement. Health care resources, care setting, and institutional factors can be studied through linkage to administrative databases or chart review and offer ways to define potentially remediable systemic contributions to delays or gaps in clinical management. Related metrics include quantification of resource limitations, such as the number of appropriate health care practitioners, and limitation on access to appropriate health care delivery sites.

Metrics addressing patterns and locus of care

The majority of adolescents ages 15 to 19 years are treated in adult oncology units.5 Although several models of care exist, including the free-standing units championed by the Teenage Cancer Trust in the United Kingdom,6 the optimal model for a system of AYA care delivery is not yet identified. Furthermore, outcome data for existing models are not yet sufficiently mature or compelling to favor a particular model over another.

The locus of care may have an impact on a wide range of outcomes as diverse as survival, late effects, psychological well being, quality of life, and patient satisfaction. With respect to survival, several studies have demonstrated survival advantages for adolescents treated for acute lymphoblastic leukemia (ALL) in pediatric settings on pediatric protocols compared with equivalent populations treated on adult-based protocols.7, 8 Similar survival advantages have been reported in Ewing sarcoma9 and other cancers.10 It is noteworthy that a reanalysis of these latter data from Georgia suggested that survival outcomes may require more sophisticated interpretation, because a survival advantage for the group ages 15 to 19 years was demonstrable for pediatric-type tumors treated in pediatric centers and for adult-type tumors treated in adult settings.11 This interpretation suggests that expertise rather than locus of treatment may be the key element in producing a survival advantage. Thus, metrics for this important domain will need to be defined, studied, and related to outcomes. Such metrics may include:

  • 1)
    Locus of care by age quintile, type of tumor, type of institution (pediatric vs adult), and nature of institution (adult cancer center vs general teaching hospital vs community hospital)—a further refinement of this measure could compare outcomes for programs identified as offering access to AYA-specific expertise with equivalent programs with different categories of expertise;
  • 2)
    Putative determinants of locus of care, including age, sex, geography of residence, socioeconomic status, and first health care provider/patterns of referral;
  • 3)
    In addition to survival outcomes, the quality of care and of the experience of treatment may be determined by locus of care, which may be the consequence of age-appropriate and developmental stage-appropriate supportive and psychosocial care; although hard to measure, it is critical to evaluate this dimension of care, and it probably is best addressed/assessed using a measure of patient satisfaction; and
  • 4)
    Clinical trial enrollment, protocol use, and adherence to and compliance with planned therapy.

Ninety percent of pediatric patients in North America are treated in institutions participating in National Cancer Institute (NCI)-funded clinical trials, and between 40% and 70% of patients at these institutions are entered on clinical trials. By contrast, only 20% to 35% of patients ages 15 to 19 years are treated in pediatric settings, and only 10% are entered on pediatric clinical trials. For the group ages 20 to 29 years, only 1% are entered in any trial.12 A survival advantage has been demonstrated for AYA patients with a range of malignancies treated on pediatric clinical trial protocols, as noted above.

Whether the determinant of a better outcome is the actual trial question under study or the rigor clinical trial participation bestows on participant institutions is not important; the degree of participation itself is an index of quality of care. Investigators from Seattle, for example, have demonstrated within their institution that there was no difference in any outcome metric for pediatric patients with ALL who were treated on a clinical trial compared with those not on a clinical trial who were treated according to a standard of care that was defined strictly by a protocol with a roadmap of similar structure to that of a clinical trial.13 However, it is well accepted that progress in clinical outcomes for a given diagnosis or population occurs only through the science of clinical trials.

The most accessible metric addressing clinical trial enrollment is the simple proportion of incident AYA cases enrolled on a clinical trial. This metric is complicated by the availability of collaborative group trials for any given disease and age range and, secondarily, the number of trials open at any given institution. Adjustment for these 2 variables must be considered in interpreting the simple proportion of incident cases enrolled, and the identification of trends over time also is important.

Survival outcomes

Survival outcomes are an important, if blunt, indicator of the efficacy of a system of care for AYA. Although Canadian published data for contemporary outcomes in AYA14 suggest that survival outcomes for the entire population are as good as those for the group ages 0 to 14 years, this is not true for individual diseases, for which the prognosis often decreases with age. It will be important to track changes in the following outcomes over time, examining variation by province and by populated versus remote and rural versus urban residence:

  • 1)
    Five-year overall survival;
  • 2)
    Event-free survival;
  • 3)
    Cause of death (disease recurrence vs complications or comorbidity); and
  • 4)
    Early mortality for sentinel diseases: This outcome is defined best for acute leukemia, for which death within an arbitrary 42 days of commencement of therapy is defined as early mortality, which may be divided into death before therapy and early treatment-related mortality. Although the definition is an area of controversy, the concept of early mortality not related to disease progression probably indexes the quality of supportive care.

Metrics of psychosocial health

The AYA patient group clearly identifies the management of psychosocial distress as a key determinant of successful passage through diagnosis and treatment and beyond, into survivorship. The early identification of those patients at highest risk of psychosocial distress, by the use of screening tools, would enable early intervention and optimization of resource deployment. If repeated at predetermined intervals, then this might provide an important index of efficacy of interventions undertaken. An example is the program of the British Columbia Cancer Agency (BCCA), which mandates psychosocial screening at entry into care for every new cancer patient. The BCCA has evaluated 50,000 adult patients at diagnosis (R. Doll, unpublished observations).

Because significant psychological distress may occur at any point along the treatment trajectory, it would be optimal to evaluate the AYA population at fixed intervals, including diagnosis, mid-treatment, end of treatment. and at some point in survivorship. The optimal instrument(s) will need to be determined, but some of the important constructs to measure include self-efficacy, locus of control, anxiety, and mood. The development of an appropriate screening tool that accommodates the different developmental stages of AYA will be important. The creation of an item bank, with validated components that would permit tailoring of the instrument to the target population, merits consideration.

In the screening and assessment process, consideration should be given to ensuring that the support system for AYA patients (eg, family of origin, partner/spouse, and children) also is evaluated. An appropriate screening instrument will need to be defined and will vary with the developmental stage of the AYA and family.

Health-related quality of life

Health-related quality-of-life (HRQL) tools provide a critical means to measure patient-reported outcomes that are more comprehensive than those that measure only psychological distress. These instruments quantify a more holistic measure of the impact of cancer on the lives of patients, and they have become ever more important in both clinical practice and research. The use of these measures across the trajectory of care may help to identify times of reduced HRQL when supportive care intervention may be most appropriate. In addition, comparison of self-reported HRQL before and after the introduction of a systematic AYA intervention may provide justification of expenditure. The ideal instruments will need to be identified. Again, consideration should be given to existing, easily applied, reliable, clinically relevant, validated instruments. The Health Utilities Index Mark 3 is a widely used instrument,15 and there are some data to support the validity of the young adult version of the Pediatric Quality of Life Inventory instrument.16 It is possible that the creation or use of an item bank of validated components to measure the various constructs important to this population would allow some degree of both discretion and uniformity. In addition to defining the important elements for the AYA population, it will be important to reach a consensus on timing and frequency of administration of HRQL measures.

Palliative care

Particularities of palliative care provision have been outlined in an accompanying article by Pritchard et al.17 Currently, metrics to assess the adequacy, effectiveness, and utility of palliative care for AYA are poorly defined. Such metrics must include both qualitative and quantitative measures and should address both the optimization and the financial implications of palliative care for this population. Quantitative measures may include:

  • 1)
    Involvement of a palliative care or hospice team, including timing of involvement;
  • 2)
    A measure of locus of care and its appropriateness, both in terms of cost and patient preference (eg, days in an acute care setting compared with days in a hospice setting or in home care);
  • 3)
    A measure of pain control;
  • 4)
    Caregiver costs, including lost work time and income;
  • 5)
    Out-of-pocket costs to families, including direct costs (such as medication and care provision costs) and indirect costs (including forgone income);
  • 6)
    HRQL measures for which consistent instruments should be used to enable comparison across jurisdictions; and
  • 7)
    The family's perception of and satisfaction with palliative care.

Outcome measures to be used in economic evaluations

The selection of optimal measures of outcome to determine whether resources allocated to this area represent the efficient use of scarce health care resources is paramount to the validation and sustainability of any proposed systematic program for AYA with cancer. The nature of the economic challenge is explained elsewhere in this supplement2 and will not be repeated here. Instead, we will outline briefly the requirements for such measures of outcome.

The use of resources will be judged as efficient (ie, getting the most value from available resources or “the biggest bang for the bucks”) if—and only if—the value of the program's benefits (ie, benefits produced) exceeds the value of what is foregone by not using them in all other ways. This approach means that the following must occur:

  • 1)
    The measure of outcome should be able to capture the different domains of health outcomes of all interventions and combine them into a unidimensional measurement scale to enable comparison and determination of whether what was gained by moving resources from 1 use to another outweighs what was lost.
  • 2)
    Because what matters in economics is the “value” of what was gained and lost, this measure, by definition, should be a preference-based measure.
  • 3)
    Because the outcome measure should capture the aggregate health benefits gained or lost by individual patients, it also should be consistent with equity criteria that guide the decision-making process.

Various measures of outcome exist (eg, quality-adjusted life years, healthy-years equivalents, willingness to pay), and there are various ways to measure each of them. Which measure to use should be discussed further. However, the merit of using 1 or more measures should be judged by taking into account the requirements described above. There are centers of excellence in health economics in Canada that may be persuaded to identify AYA oncology as a high-priority theme area for evaluation and could be involved in helping to identify the measure to use.

Survivorship

  1. Top of page
  2. Abstract
  3. General Characteristics of Metrics Relevant to AYA
  4. AYA With Cancer
  5. Survivorship
  6. Conclusion
  7. FUNDING SOURCES
  8. REFERENCES

Survivors of cancer in all age groups represent a steadily increasing population with substantial health, psychological, and quality-of-life issues. Some of these outcomes are carried forward from the active treatment period, and some manifest first after completion of therapy, constituting late effects. Survivorship has been defined typically in childhood cancer survivor cohort studies as beginning 5 years after completion of therapy. This definition excludes a substantial group of patients who also are cancer survivors. Thus, there is need for a consensus definition that would pertain to outcomes and metrics. A practical approach may be to focus on the period beginning with completion of treatment. Because the issues relevant to AYA survivors of cancer in childhood and survivors of cancer in AYA differ somewhat, it is reasonable to consider them separately.

AYA survivors of cancer in childhood

The issues for this AYA group are determined largely by the treatment modalities used, the era of treatment, the age at which treatment occurred, and the time elapsed since treatment. These issues are modulated by the degree of immaturity of physiologic and developmental systems at the time of treatment and the subsequent maturational trajectory of physiologic, psychological, neurocognitive, and social growth and aging.

There are existing cohort studies, notably the US-based Childhood Cancer Survivor Study (CCSS) and the United Kingdom Childhood Cancer Survivor Study (UKCCSS), which have been primarily descriptive and predictive. Morbidity and the incidence of chronic disease and/or late effects have been well documented.18, 19 There also are proposed population-based studies focused more on medical outcomes. Thus, in addition to documenting medical outcomes, measuring the impact of differing models of organization of care on definable health care use outcomes, on compliance with monitoring guidelines, and on the efficacy of such guidelines may be a priority.

The improvement in cancer survival rates among children over the last few decades has been the consequence mainly of better understanding of biologic and clinical predictors of treatment response and consequent stratification of patients by risk category. The identification of low-risk populations has permitted reduction in treatment intensity for this group, hopefully leading to reduced frequency of late effects for this stratum. However, the more recent improvement in survival has been the result of an increased survival rate for the higher risk groups. This increase in survival has been achieved by increased treatment intensity; therefore, it is reasonable to assume that these patients will manifest more late effects.

An important issue in the care of survivors of cancer in childhood is the success of transition from family-centered care in the pediatric context to individually focused care in the adult model. There is evidence to suggest a significant attrition rate at this juncture.20

Several models of care for the adult survivors of cancer in childhood exist: a centralized model in which survivors are encouraged to receive care at a clinic with expertise in survivorship; a model in which most adult survivors are discharged to primary care practitioners; and a model in which shared care is deployed, with some care being delivered by primary care practitioners and some by a specialized clinic.20 All models should use risk-driven guidelines, and the best known of these are from the Children's Oncology Group.21 Suggested important patient and system outcomes for AYA survivors of childhood cancer include the following:

  • 1)
    Evaluation of models of transition from care in a pediatric context to care in an adult context; potential indicators include initial attendance at care settings after transition and ongoing attendance as an indicator of perception of value; more direct evaluation would be provided by patient-reported outcomes, including knowledge of their medical history and risks, self-empowerment, and satisfaction with care;
  • 2)
    Evaluation of effectiveness of different models of care in achieving compliance with recommended monitoring guidelines for high-risk patients (eg, cardiac monitoring of patients exposed to significant anthracycline dose and breast cancer monitoring of young women who received chest radiation);
  • 3)
    Indicators of the efficacy of monitoring in reducing morbidity and mortality of late effects by frequency of preclinical diagnosis, stage of illness at diagnosis, and cost-effectiveness of the resultant intervention; the comparative impact of differing models of care provision and intervention on these parameters, measured either directly or by surrogate indicators, may clarify cost-effectiveness of the different models;
  • 4)
    Evaluation of efficacy of differing models of survivor care on survivor empowerment and on health care utilization and efficiency;
  • 5)
    An emerging standard of care for survivors is the creation, review, and implementation of survivor care plans, which ought to include a detailed history of disease and treatment22; a measure of the proportion of survivors provided with such a care plan will be important;
  • 6)
    Effectiveness of risk-based survivorship care in changing patterns of cause-specific mortality; mortality data from the CCSS indicate that early cause-specific mortality in 5-year survivors admitted to the study is predominantly because of disease recurrence; with longer time intervals from diagnosis, cause-specific mortality shifts to nonrecurrence causes, dominantly second cancers and cardiac death23; similar data have been reported from the UKCCSS,24 and screening and early intervention are intended to mitigate these causes of mortality; measurement of effective risk-based care in changing this pattern of mortality should be undertaken;
  • 7)
    Health care use studies quantifying the volume and type of health care services used; assessment of the real and potential economic costs of survival to the health care system, the distribution of those costs by disease and age category, and potential modifiers of these costs will be particularly significant as the anticipated wave of survivors with late effects increases steadily; the impact of planned services on the use of emergency services and on the severity of key outcomes may be assessable in this fashion; and
  • 8)
    Systematic, patient-related outcome measures should include HRQL measures, satisfaction measures, and measures of educational and vocational achievement, both direct (by self-report) and indirect (by employment attainment or other indicator).

Long-term survivors of cancer in AYA

This population of survivors has been studied less systematically and currently constitutes a small proportion of most survivor cohorts. Although biologic maturation is largely complete by the age of treatment in this group, and the slope of the developmental trajectory to be followed is less steep, there are important health, psychological, and transitional issues faced by this group both at the end of treatment and in follow-up. Survivorship models are less developed for this group than for survivors of cancer in childhood. Some organizations define survivorship as beginning at diagnosis. If such a definition is accepted, then monitoring for recurrence becomes a more important part of survivorship that may lend itself to monitoring in a primary care setting.25

Although some of the outcomes and metrics developed for AYA survivors of cancer in childhood probably are applicable in this population, others may have to be developed specifically for survivors of cancer in AYA. This may be particularly true for psychosocial outcomes, because the challenges to this age group are fundamentally different from those confronting survivors of cancer in childhood. There also may be a variation in needs over the span between ages 15 to 29 years, because the developmental task progresses from the need to transition from adolescent behavior to adult behavior to the assumption of partnership and parental roles.

Conclusion

  1. Top of page
  2. Abstract
  3. General Characteristics of Metrics Relevant to AYA
  4. AYA With Cancer
  5. Survivorship
  6. Conclusion
  7. FUNDING SOURCES
  8. REFERENCES

The development of a systematic approach to the care of AYA with cancer has been transformed radically in the United Kingdom under the influence of the Teenage Cancer Trust. The potential for developing a unique model in the Canadian single-payer health care system is inviting, and attention must be paid to all the critical components of an integrated, holistic plan, the particulars of which remain to be defined. It is likely that an overall blueprint for service delivery will be established; however, because health care is a provincial jurisdiction in Canada, the implementation may vary in detail from province to province. So, it will be vitally important to have in place a systematic plan for the measurement of outcomes across jurisdictions. This report offers an initial description, based on input from attendees at the AYA workshop in March 2010, of what may be the range of possible metrics.

Canada and its provinces and territories are advantaged by the extensive, rich administrative databases available and the potential for linking these databases to cancer registries to provide surrogate markers of process and outcome of care delivery. Together with a single-payer health care system, these resources offer a unique opportunity to evaluate a new thrust in cancer medicine—an opportunity that it behooves the Canadian health care community to seize and to contribute to the burgeoning focus on cancer in AYA.

FUNDING SOURCES

  1. Top of page
  2. Abstract
  3. General Characteristics of Metrics Relevant to AYA
  4. AYA With Cancer
  5. Survivorship
  6. Conclusion
  7. FUNDING SOURCES
  8. REFERENCES

Funding for the National Task Force on Adolescents and Young Adults with Cancer has been made possible by a financial contribution from Health Canada through the Canadian Partnership Against Cancer. Funding for the workshop was provided by C17; the Advisory Board of the Institute for Cancer Research at the Canadian Institutes for Health Research (CIHR); the Public Health Agency of Canada; the Ontario Institute for Cancer Research; the Meetings, Planning, and Dissemination Grants program of the CIHR; the Terry Fox Research Institute; LIVESTRONG, formerly the Lance Armstrong Foundation; the Canadian Cancer Society Research Institute; Young Adult Cancer Canada; Hope and Cope; and the Comprehensive Cancer Centre at the Hospital for Sick Children, Toronto, in addition to the support provided by the Canadian Partnership Against Cancer to the Task Force on Adolescents and Young Adults with Cancer.

CONFLICT OF INTEREST DISCLOSURES

The authors made no disclosures.

REFERENCES

  1. Top of page
  2. Abstract
  3. General Characteristics of Metrics Relevant to AYA
  4. AYA With Cancer
  5. Survivorship
  6. Conclusion
  7. FUNDING SOURCES
  8. REFERENCES