Increasing cardiomyopathy screening in childhood cancer survivors: A cost analysis of advanced practice nurse phone counseling compared to a print survivorship care plan

Abstract

Purpose:

Despite the significant risk of cardiac disease after pediatric cancer treatment with anthracyclines and radiation, adult survivors are not participating in recommended screening. We documented the per-survivor and per-additional-survivor-screened costs of a mailed survivorship care plan with (SCP+C) or without (SCP) advance practice nurse telephone counseling.

Design:

Randomized longitudinal clinical trial.

Setting:

St. Jude Children’s Research Hospital (SJCRH).

Sample:

411 at-risk pediatric cancer survivors (26-59 years old; mean = 41 years, SD = 7.68), stratified by age (<30 years vs. ≥30 years), recommended screening frequency (every one, two, or five years), sex, and cancer diagnosis (hematological vs. solid tumor).

Methods:

Clinical and resource data costs were derived from trial data and external estimates.

Findings:

The per-survivor costs of SCP (n = 206) and SCP+C (n = 205) were $74.91 and $224.69, respectively. The estimated costs of SCP and SCP+C per additional survivor screened over two years disseminated in a medium-sized clinic (n = 101 survivors annually) were $345.41 and $293.85, respectively.

Conclusions:

Adding advanced practice nurse counseling to a print care plan may help preserve cardiac health at little or no cost per additional survivor screened.

Implications for Nursing:

Advance-practice nurse counseling is cost-effective and superior to the standard of care in supporting at-risk survivors’ cardiac screening participation.

Knowledge Translation:

Autonomy-supportive counseling strategies are easily added to advance-practice nurses’ skill set for immediate implementation. The intervention can be extended to other at-risk cancer survivors for whom regular screening is crucial. These intervention approaches could become part of the infrastructure of emerging survivorship programs.

BACKGROUND AND PURPOSE

Despite anthracyclines’ contribution to childhood cancer survival, they confer an excess risk of asymptomatic left ventricular dysfunction, cardiomyopathy, congestive heart failure, and death (Lipshultz et al., 2013; Mulrooney et al., 2009; Pein et al., 2004; van Dalen, van der Pal, Kok, Caron, & Kremer, 2006). Irradiation of cardiovascular (CV) structures is associated with various adverse CV outcomes, including cardiomyopathy, constrictive pericarditis, and accelerated atherosclerosis, predisposing survivors to early-onset coronary artery disease, myocardial infarction, and stroke (Van der Pal, Van Dalen, Kremer, J., & van Leeuwen, 2005; Van der Pal et al., 2012). Unfortunately, these CV effects can be progressive, harmful, and frequently subclinical in the early stages (Mulrooney et al., 2009; van Dalen et al., 2006; Van der Pal et al., 2005).

All available long-term follow-up (LTFU) guidelines for pediatric cancer survivors (Armenian et al., 2015; Children’s Oncology Group (COG); Dutch Childhood Oncology Group; Scottish Intercollegiate Guidelines Network; United Kingdom Children’s Cancer Study Group Late Effects Group) recommend evaluating left ventricular systolic function through echocardiography or comparable imaging. Screening frequency is based on age at cancer diagnosis and the cumulative dose of cardiotoxic therapies; recommendations range from annual screening to screening every 5 years. Full details are available at www.survivorshipguidelines.org.

Yet, most childhood cancer survivors are not receiving risk-based preventive care for cardiac disease. The Childhood Cancer Survivor Study (CCSS) demonstrated that only 511 (28.2%) of 1810 childhood cancer survivors identified as being at high risk for cardiomyopathy reported undergoing screening during the previous 24 months (Nathan et al., 2008). Cancer treatment centers are encouraged to provide all patients with a survivorship care plan (SCP) that summarizes treatment information and outlines health-screening recommendations for LTFU of late effects. However, the single prospective pilot study evaluating the effect of SCPs on adherence to cardiac screening among survivors demonstrated that even with a care plan detailing risks and follow-up recommendations, only 20% of patients underwent cardiac screening within the next two years (Oeffinger et al., 2011).

Our recently completed randomized controlled trial illustrated that the addition of telephone counseling by an advanced practice nurse (APN) to a printed SCP substantially increased cardiac screening participation (Hudson et al., 2014). This report describes the costs associated with providing personalized care plans and APN counseling to improve CV screening participation among at-risk survivors. To our knowledge, this is the first cost-effectiveness analysis (CEA) of a distance-based strategy to address cardiomyopathy screening in adult pediatric cancer survivors. CEA describes the costs of an intervention in terms of the cost to achieve a specific effect or benefit; in this case we compared the cost of two different interventions (standard of care vs. standard of care plus counseling) in terms of the number of survivors screened for cardiomyopathy.

Study Setting and Sample

The intervention and evaluation methods are described in detail elsewhere (Hudson et al., 2014). The study was initiated only after approval by SJCRH’s institutional review board in accordance with an assurance filed with and approved by the U.S. Department of Health and Human Services. Briefly, the study was conducted from April 2010 through August 2013. The study population was recruited from long-term childhood cancer survivors who were currently participating in the Childhood Cancer Survivor Study (CCSS), a 27-institution cohort study that is currently following more than 10,000 long-term survivors of childhood cancer diagnosed between 1970 and 1986 (For details, see: http://ccss.stjude.org). Survivors were eligible for the study if they met the following criteria: 1) were CCSS participants, 2) age ≥ 25 years, 3) had no cardiomyopathy screening in the past five years, 4) received anthracyclines and/or irradiation to cardiovascular (CV) structures, 5) were not being followed at a LTFU survivorship program, and 6) had a previous history of successful independent (non-surrogate) response to CCSS surveys. Eligible participants were randomized to either usual care control (SCP) or to usual care plus telephone counseling by an APN (SCP+C). The total sample was stratified by age (<30 years vs. ≥30 years), COG–recommended screening frequency group (every one, two, or five years), sex, and cancer diagnosis (hematological malignancy vs. solid tumor).

Usual Care Control

After baseline assessment and randomization, survivors were mailed a printed SCP outlining their specific cancer treatments, health risks, and behavioral recommendations to maintain health as per the COG Guidelines. The SCP was personalized by using the terms “you” or survivors’ first names throughout the document; offering specifics of treatment rather than referring globally to treatment exposures; and offering specific recommendations for follow-up and screening. The packet also included a laminated card with treatment exposures, future health risks, and recommendations for follow-up that the survivor could take to his or her primary care provider.

Telephone Counseling Intervention

Survivors assigned to telephone counseling (SCP+C) received the same mailed summary and laminated card as those in the usual care arm and participated in two telephone-counseling sessions, one and three weeks after receiving the treatment summary. The APN reviewed the treatment summary with the survivor clarifying any questions or concerns that may have been generated by the summary. The APN tailored the motivational interviewing strategy and content to motivation (CV-self-regulation, intrinsic motivation), affective response (depressed mood, stress, fear), and survivor-provider interaction (provider support for self-management, health care climate, decisional control). Thus APN counseling addresses not only specific health information content, but does so using therapeutic interaction dialogue (motivational interviewing techniques) that deliberately targets motivation to support CV-screening adherence.

Each survivor was sent a follow-up letter summarizing the content of each call. The APNs focused on supporting survivors’ competency by providing information about behavior-outcome contingencies; facilitating realistic expectations and self-selected goals; and providing positive, non-judgmental feedback. APNs supported survivors’ autonomy by avoiding confrontation and coercion, exploring behavioral options, identifying the discrepancy between current behaviors and behaviors needed for optimal CV health, and encouraging choice of course of action. Lastly, APNs supported survivors’ need for relatedness by expressing genuine interest and warmth, offering empathic, unconditional support, and avoiding criticism or blame (Cox, 2003; Deci, Eghrari, Patrick, & Leone, 1994; Williams & Deci, 2001).

Figure 1 summarizes APN activities in implementing the intervention.

Figure 1:

Brief Summary of APN Counseling Intervention Activities

The APNs relied on an interactive computer program developed specifically for the study by St. Jude IT staff. The program not only supported standardized delivery of the intervention, but allowed for complete access to the following information: 1) previously collected CCSS questionnaire data; 2) most recent CCSS-abstracted medical record; 3) print summary; 4) baseline questionnaire; and 5) scripts for the targeted behavioral indicators (e.g., knowledge, beliefs, motivation, fears/worries). The program generated scripts based on survivors’ specific responses to the behavioral indicators on the baseline questionnaire (See example computer scripts in supplemental materials); the APNs then used these scripts in their interactions with survivors.

Twelve months after baseline assessment and completion of each study arm, the medical records of all survivors were requested to document physician encounters and adherence/non-adherence to CV screening during the year before and year after the intervention. The SCP and SCP+C interventions respectively motivated 22.3% and 52.2% of the participants to complete an LV function study. Survivors in the counseling group were more than twice as likely than those in the usual care control group to have the recommended cardiomyopathy screening (RR 2.31; 95% CI: 1.74-3.07)](Hudson et al., 2014).

METHODS

Cost Data Collection

Intervention cost data were collected by using standard methods (Gold, Siegel, Russell, & Weinstein, 1996; Graham, Corso, Morris, Sequi-Gomez, & Weinstein, 1998) and included accounting records, timers and time logs, and data-based estimates. Accounting records were used to estimate payroll and budget expenditures. Staff entered budget expenditures, such as capital equipment and office supplies, at the time of purchase. Payroll data were used to calculate average salaries on a per-minute basis for individuals providing the intervention (computer programmers, advanced practice nurses, those training the nurses, and office staff who sent questionnaires and performed data entry). These costs were calculated separately for all study personnel on the basis of their 2011 salaries and fringe costs; salary costs were averaged when multiple individuals worked on a single task. A timer was built into the study database to collect time expended for tasks related to the intervention (Table 2). Times for making each phone call, recording data in the database, and updating participants’ statuses were recorded by pressing start and stop from a dropdown menu on the database recorder indicating the task performed. If task time for an activity was not adequately captured in the database, then staff members manually tracked time expended per task (e.g., social worker assistance). An Internet timer was used to calculate time to complete larger tasks, such as compiling and mailing materials. The time per item (e.g., time to compile an introductory packet) was then calculated as the total time to complete the task divided by the total number of items. In those few cases (n = 11) in which the timers were not started or not turned off, staff estimates were used and did not differ from average calculated times for those activities. Overhead costs were not included: office space, heat, lights, computer access, and phone calls were treated as institutional costs and assumed to be available at other institutions that would seek to provide these services.

Table 2.

Time, personnel, and material costs associated with the phone counseling intervention

Task	Average time (min)	Personnel Cost	Materials cost
Create newsletter	55.59	$31.07	$0.16
Create health information card	45.94	$25.68	$0.66
Create survivorship care plan	38.75	$21.66	$0.00
PI’s review of patient’s health information	10.00	$15.80	$0.00
Review letter and card for accuracy	9.82	$5.49	$0.00
Review questionnaire responses	7.08	$3.96	$0.00
Strategy development for Call 1	39.87	$22.28	$1.20
Intervention Call 1	20.00	$11.18	$0.00
Develop Letter 1 to participant	5.00	$2.79	$0.00
Generate summary of Call1	14.68	$8.20	$0.08
Print letter 1	0.08	$0.05	$0.00
Strategy development for Call 2	20.00	$11.18	$0.00
Intervention Call 2	28.60	$15.98	$0.86
Develop Letter 2 to participant	5.00	$2.79	$0.00
Generate summary of Call 2	14.68	$8.20	$0.08
Print letter 2	0.08	$0.05	$0.00

PI = principal investigator

Cost Classification:

Both fixed and variable costs for the provision of usual care (SCP) and for SCP with the counseling intervention (SCP+C) were documented for the base model. Fixed, one-time intervention costs for usual care included laminating equipment, whereas SCP+C fixed costs included the laminator, the development of the interactive computer program, and APN training. The fixed, one-time costs unique to the SCP+C intervention (e.g., programming costs and nurse training) were divided among those in that study arm only to provide a cost per participating survivor. APN training costs in the base cost model included the salary of the investigator providing the training and the cost and time for regular meetings with the trainer to refresh the intervention approach and to check periodic audio-taped counseling sessions (nurse dialogue only) to assure optimal performance.

Variable costs related to personnel in the base model (Table 3) included supplies and the time intervention personnel in spent delivering the interventions. The unit cost of materials for both study arms (e.g., stationery, stamps, and laminating materials) was obtained from invoices. Personnel costs included time spent doing the following: reviewing survivor baseline questionnaires, medical records, and CCSS questionnaires; entering progress notes; and preparing post-counseling follow-up letters. Personnel costs for tasks were multiplied by the salary-per-minute of the required personnel.

Table 3.

Total cost per participant in each study arm (BASE Model)

Task	One-time costs	Cost/participant	Phone Counseling + Survivorship Care Plan (SCP+C)	Survivorship Care Plan only (SCP)
Questionnaire
Create and mail questionnaire packet		$3.56	$3.56	NA
Questionnaire packet return mailing		$1.28	$1.28	NA
Open, review, enter questionnaire data		$5.87	$5.87	NA
Educational Materials
Cover letter		$2.10	$2.10	$2.10
Create health information card		$26.34	$26.34	$26.34
Create SCP		$21.66	$21.66	$21.66
PI’s review of patient’s health information		$15.80	$15.80	$15.80
Review letter and health card for accuracy		$5.49	$5.49	$5.49
Health card & SCP materials		$1.14	$1.14	$1.14
Card and summary mailing		$1.29	$1.29	$1.29
Laminator for health information card	$448.00	$1.09	$1.09	$1.09
Phone counseling
Review questionnaire responses		$3.96	$3.96	NA
Strategy development for Call 1		$23.48	$23.48	NA
Intervention Call 1		$11.18	$11.18	NA
Develop Letter 1 to participant		$2.79	$2.79	NA
Generate summary of Call1		$8.28	$8.28	NA
Print letter 1		$0.05	$0.05	NA
Strategy development for Call 2		$11.18	$11.18	NA
Intervention Call 2		$16.84	$16.84	NA
Develop Letter 2 to participant		$2.79	$2.79	NA
Generate summary of Call 2		$8.28	$8.28	NA
Print letter 2		$0.05	$0.05	NA
Mailing supplies and stamps (both letters)		$1.30	$1.30	NA
Study nurse training for phone counseling	$3419.84	$16.68	$16.68	NA
Study nurse intervention computer program – programming costs	$6807.80	$32.21	$32.21	NA
N in each group			205	206
Total Costs			$46,061.45	$15,431.46
Cost per participant				$224.69
Cost per additional participant screened				$432.10

PI = principal investigator

Cost Modeling:

We describe the costs associated with replicating the intervention at medium-sized or larger LTFU clinics (Table 4)(Eshelman-Kent et al., 2011). For this purpose we modeled two sizes of clinic populations: 202 and 1000 survivors for a two-year program. Differences in the Year 1 fixed costs and variable costs in subsequent years are noted. The fixed costs associated with developing the interactive computer program were dropped because it was assumed that clinics would be able to download the program from the CCSS website at no charge. APN training costs were left as in the base model, allowing for one-time training of staff as they entered the project and for staff turnover every 2 years as necessary. The laminator was budgeted for use over a 2-year period for 202 survivors in the medium-sized clinic model and for 1000 survivors in the larger-sized clinic model but would be available for other purposes and likely could be used for a considerably longer period of time in either case.

Table 4:

Total cost per participant in each study arm (DISSEMINATION Models)

Tasks and supplies	Large clinic: SCP+C	Large clinic: SCP only	Medium clinic: (SCP+C)	Medium clinic: SCP only
Educational Materials
Create cover letter	$2.10	$2.10	$2.10	$2.10
Create health information card	$26.34	$26.34	$26.34	$26.34
Create SCP	$21.66	$21.66	$21.66	$21.66
Review patient health information	$15.80	$15.80	$15.80	$15.80
Review letter and card for accuracy	$5.49	$5.49	$5.49	$5.49
Health card & SCP materials	$1.14	$1.14	$1.14	$1.14
Mail card and summary	$1.29	$1.29	$1.29	$1.29
Laminator for health information card	$0.45	$0.45	$2.17	$2.17
Phone Counseling
Abbreviated phone interview + Intervention Call 1	$22.36	NA	$22.36	NA
Develop Letter 1 to participant	$2.79	NA	$2.79	NA
Generate summary of Call1	$8.28	NA	$8.28	NA
Print letter 1	$0.05	NA	$0.05	NA
Prepare for & make Intervention Call 2	$13.98	NA	$13.98	NA
Develop Letter 2 to participant	$2.79	NA	$2.79	NA
Generate summary of Call 2	$8.28	NA	$8.28	NA
Print letter 2	$0.05	NA	$0.05	NA
Mailing supplies and stamps	$1.30	NA	$1.30	NA
Training study nurse for phone counseling (does not include travel/per diem)	$3.41	NA	$16.93	NA
Total Costs	$137.56	$74.27	$152.80	$75.99
N in each group	1000	1000	202	202
Total Costs both years	$137,560.00	$74,270.00	$30,865.60	$15,797.98
1st year start-up costs:	$70,713.92	$37,360.00	$17,366.72	$8,122.99
Subsequent year costs:	$66,846.08	$36,910.00	$13,498.88	$7,674.99
Cost per additional participant screened	$264.54	$337.59	$293.85	$345.41

Abbreviations: SCP, educational materials only; SCP+C, phone counseling + educational materials

Cost-effectiveness Analysis:

To assess cost-effectiveness, costs of each study arm were calculated on a per-survivor basis and then divided by the effectiveness of that arm (Gold et al., 1996). To determine the incremental treatment effect on costs, we calculated 2 types of predicted expenditures. The first were expenditures related to SCP, and the second were expenditures related to SCP+C; the only incremental costs were those associated with the telephone-counseling portion of the intervention. This approach provides information on the per-participant cost of each arm and the cost of the additional intervention activities associated with counseling and can be used with study-arm effectiveness to calculate the cost basis and per-additional-survivor-screened costs associated with the addition of counseling to print care plans (Andersen, Hager, Su, & Urban, 2002; Andersen, Urban, Ramsey, & Briss, 2004)

RESULTS

Table 1 characterizes the study sample. Table 3 describes the per-survivor costs of both study arms. The total costs were $15,431.46 for the SCP-only arm and $46,061.45 for the SCP+C arm, resulting in a per-survivor cost of $74.91 to implement SCP and $224.69 to implement SCP+C. Subtracting the costs of the SCP intervention from those of the SCP+C revealed an incremental cost of $149.78 per survivor for APN-led counseling. When examined with the effectiveness of the SCP and SCP+C interventions (Table 2) that respectively encouraged 22% and 52% of previously non-adherent survivors to participate in screening, the costs per survivor screened were $340.50 (SCP) and $432.10 (SCP+C). The addition of telephone counseling increased the rate of screening by 30%, making the costs of the two interventions considerably more comparable per survivor screened.

Table 1.

Descriptive summary of the sample

Characteristics		Control group		Intervention group		P value
		N	%	N	%
COG-recommended screening frequency	Every 2 years	43	20.9	43	21.0	0.73
	Every 5 years	36	17.5	30	14.6	.
	Every year	127	61.7	132	64.4	.
Gender	Female	112	54.4	115	56.1	0.73
	Male	94	45.6	90	43.9	.
Age at randomization	<30 years	18	8.7	23	11.2	0.40
	≥30 years	188	91.3	182	88.8	.
Race	White non-Hispanic	186	90.3	181	88.3	0.46
	Black	1	0.5	3	1.5	.
	Other	17	8.3	21	10.2	.
	Unknown	2	1.0	0	0.0	.
Education	High school or less	20	9.7	17	8.3	0.94
	Post-high school study/some college	55	26.7	59	28.8	.
	College graduate	81	39.3	79	38.5	.
	Postgraduate level	50	24.3	50	24.4	.
Household income	<$20,000	12	5.8	16	7.8	0.68
	$20,000–$60,000	62	30.1	57	27.8	.
	>$60,000	126	61.2	125	61.0	.
	Unknown	6	2.9	7	3.4	.
Health insurance	Yes/Canadian	187	90.8	186	90.7	0.86
	No	17	8.3	18	8.8	.
	Unknown	2	1.0	1	0.5	.
Diagnosis	Bone cancer	33	16.0	40	19.5	0.05
	Central nervous system tumor	1	0.5	0	0.0	.
	Hodgkin lymphoma	42	20.4	29	14.1	.
	Wilms tumor	24	11.7	10	4.9	.
	Leukemia	65	31.6	74	36.1	.
	Non-Hodgkin lymphoma	16	7.8	24	11.7	.
	Neuroblastoma	11	5.3	7	3.4	.
	Soft-tissue sarcoma	14	6.8	21	10.2	.
Age at cancer diagnosis	0-4 years	58	28.2	52	25.4	0.60
	5-9 years	38	18.4	46	22.4	.
	10-14 years	49	23.8	54	26.3	.
	15-20 years	61	29.6	53	25.9	.
Years since diagnosis	28 or fewer	97	47.1	83	40.5	0.18
	More than 28	109	52.9	122	59.5	.
Chemotherapy	Yes	183	88.8	191	93.2	0.13
	No	23	11.2	14	6.8	.
Radiation	Yes	137	66.5	140	68.3	0.70
	No	69	33.5	65	31.7	.
Both chemotherapy and radiation	Yes	114	55.3	126	61.5	0.21
	No	92	44.7	79	38.5	.
Chest radiation	Yes	62	30.1	50	24.4	0.16
	No	139	67.5	153	74.6	.
	Unknown	5	2.4	2	1.0	.
Brain radiation	Yes	37	18.0	55	26.8	0.04
	No	164	79.6	148	72.2	.
	Unknown	5	2.4	2	1.0	.
Alkylating agent chemotherapy	Yes	140	68.0	153	74.6	0.14
	No	66	32.0	52	25.4	.
Anthracycline chemotherapy	Yes	163	79.1	173	84.4	0.17
	No	43	20.9	32	15.6	.
Surgery	Yes	164	79.6	165	80.5	0.90
	No	41	19.9	40	19.5	.
	Unknown	1	0.5	0	0.0	.
Amputation	Yes	15	7.3	20	9.8	0.38
	No	190	92.2	185	90.2	.
	Unknown	1	0.5	0	0.0	.
Completed cardiovascular screening form	Yes	206	100	205	100	0.38
Grade 1-4 chronic condition at any time	No	39	18.9	33	16.1	0.45
	Yes	167	81.1	172	83.9	.
Two or more grade 3-4 chronic conditions at any time	No	183	88.8	180	87.8	0.75
	Yes	23	11.2	25	12.2	.
Health status	Excellent/very good/good	194	94.2	187	91.2	0.44
	Fair/poor	12	5.8	16	7.8	.
	Unknown	0	0.0	2	1.0	.

Dissemination Models:

To better understand the likely costs of each intervention when implemented in actual survivorship care settings, dissemination models were created (Table 4). Variable costs were similar to those of the base case model but did not include costs associated with sending, receiving, reviewing, and data entry of baseline questionnaires. Although we used the questionnaires for both research and counseling intervention purposes, secondary analyses of the intervention have documented the most influential variables to target in tailoring. Tailoring the intervention on these elements could be done with the assistance of the computer program during the call and on the basis of the participants’ responses, thereby eliminating the pre-call questionnaire. This change in protocol would significantly reduce the level of APN and office staff effort required for review, mailing, and data entry.

In dissemination, the estimated costs of SCP and SCP+C of two years in a medium-sized clinic (n = 101 survivors seen each year) were $75.99 and $152.80, respectively. After division by the effectiveness of the two interventions, the CEA per additional survivor screened were $345.41 (SCP) and $293.85 (SCP+C). Assuming a larger clinic (500 at-risk patients seen annually for two years), the CEA would be projected to be $337.59 for SCP and $264.54 for SCP+C per additional survivor screened.

DISCUSSION

The clinical trial from which the cost data were derived demonstrated that the addition of two autonomy-supportive phone counseling sessions led by oncology APNs substantially increased the rate of cardiomyopathy screening participation by previously non-adherent, at-risk adult survivors of pediatric cancer over that of a personalized care plan alone. APNs were selected as the interventionists based on our long-term care model at SJCRH and their familiarity with the implementation of research protocols. APNs assume primary responsibility for assuring that the long-term follow-up guidelines are the standard of long-term care; moreover, they assist physicians who rotate through our long-term care clinics, in learning and implementing the COG Guidelines. They have had many years of experience in following survivors using their advanced assessment skills and implementing behavioral change strategies. In telephone interaction with survivors, they were often required to address questions about symptoms, current and previous therapies, and screening recommendations beyond cardiomyopathy screening. Additionally, they provided the most current literature to primary care physicians who sought to gain greater understanding of survivors’ long-term care issues. All APN training related to the motivational communication strategies; additional training relative to long-term care of survivors was not necessary. With training in the motivational communication strategies, registered nurses, with similar experience in long-term management of adult survivors of pediatric malignancies, knowledge of presenting symptomatology and management, and the COG guidelines likely could implement the intervention with equal success.

Although using telephone counseling as an adjunct to the print care plan increases a clinic’s costs, those costs are substantially offset by the larger percentage of survivors who choose to participate in screening reducing the cost per survivor screened. Costs for the counseling intervention could be reduced substantially in dissemination depending on the size of the LTFU clinic and approach to implementation, potentially making the per-survivor-screened costs lower for counseling than for care plans alone. Additional strategies to further reduce the costs associated with the intervention include the use of non-APN staff to perform support activities (letters to survivors, medical record abstraction, scheduling calls) and completion of APN training partially through teleconferencing.

Assuming 101 survivors screened in a single year, a 35-hour work week, and a 45-week work year, an APN would need to devote less than 5 hours per week to the counseling intervention. This estimate is based on initial and second calls taking about 40 and 25 minutes, respectively, per patient and some additional time for call preparation and completing call summary letters. In a larger clinic with 1000 patients per year to be contacted, the counseling intervention might require 55% of an APN’s time but could likely be divided among a group of APNs so that a clinician would not have to devote most of his or her time to phone counseling.

It is quite possible that the participation rate for other LTFU clinics would be substantially higher than what we demonstrated in our clinical trial. None of our patients in the trial were treated at SJCRH (source of the printed care plans and APN counselors); moreover, they had to depend on their primary care physician to order the echocardiogram, and some physicians declined to order echocardiograms because of the survivors’ absence of symptoms and their own unfamiliarity¹⁴ with the COG Guidelines. Calls from the survivors’ own treating institutions and the ability to be followed at that institution would likely be a strong influence on the survivors and increase participation substantially. For patients who were originally treated for cancer at the institution implementing the counseling intervention but are now followed by a primary care provider, a phone call to the provider to request support for screening would be ideal and more likely to result in completion of screening.

Currently, there is consensus that survivors exposed to anthracyclines and chest radiation should begin LV function screening no later than two years after completion of therapy and continue to be screened at least every five years (Hunt et al., 2009). Some debate exists as to the frequency of screening needed for patients at high, moderate, or low risk. Wong and colleagues found that, based on risk profile, changing from every one, two, or five-year screening intervals to two, five, and ten-year intervals resulted in maintaining 80% of the health benefit of the COG Guidelines-recommended schedule while decreasing the costs by 50% (Wong et al., 2014). Yeh and colleagues proposed that survivors at high risk (treated with 250mg/m² or greater) should have screening echocardiography every two years or cardiac magnetic resonance imaging (cMRI) every five years; the preferred strategy for those at low risk was either no echocardiography or cMRI every ten years (Yeh, Nohria, & Diller, 2014). These future decisions could have significant implications for implementation of the telephone-counseling intervention within an LTFU clinic: 1) decreased screening frequency would reduce the number of annual survivor contacts, thereby reducing the overall costs of intervention implementation; and 2) longer intervals between screenings might contribute to increased survivor follow-up attrition, reinforcing the need for the counseling intervention.

The study must be interpreted in the context of some limitations. Our data are not generalizable to survivors who did not participate in CCSS. Cost data were based on a single institution’s personnel and materials costs and may not be representative of costs at other long-term care facilities or oncology practices. We made every effort to be completely transparent in our analysis, by providing both a base case and dissemination model; however, varying degrees of uncertainty and assumptions exist in all cost analyses (Raftery, 1999). To minimize uncertainty and assumptions, we relied on objective data or data based estimates, and we describe in our cost analysis our implementation of the current standard of care in the field.

Our results indicate that adding advanced practice nurse counseling is no more costly—and perhaps less costly—per additional survivor screened than is a printed care plan alone for increasing cardiomyopathy screening among at-risk survivors of pediatric malignancies. This intervention approach can very likely be extended to other at-risk childhood cancer survivors (e.g., screening mammograms of women exposed to chest radiation) and to individuals with other chronic illnesses that require regular screening. Additional cost-effectiveness research that would address participation in other high priority screening for survivors of adult cancer could be modeled on studies of the cost-effectiveness of cancer screening promotion (Andersen, et al, 2002; Andersen et al., 2004). Such studies have the potential to highlight the value of nurse implemented supportive care.

Implications for Nursing

Advanced practice nurses and oncology nurses with extensive experience in long-term care of adult survivors of pediatric malignancies are uniquely positioned to take leadership roles in the long-term management of now-adult survivors. While survivorship clinics are increasing in number and capacity nationwide, there is yet a significant gap between those survivors needing managed follow-up care and the availability of providers who are knowledgeable about long-term follow-up care. Advanced practice and oncology nurses, experts in long-term care guidelines and trained in the intervention approach we modeled, could extend the reach of their respective institutions to a far wider geographical target area. This would include direct interaction with the survivor who is now no longer followed at their institution and interaction with regional primary care providers, who may not be familiar with follow-up guidelines. Implementing this approach would reach a much larger at-risk population, potentially support screening participation at a level far exceeding the current standard of care, and could have far-reaching implications for the long-term health and well being of survivors.