Financial Analysis of an Intensive Pediatric Continuous Positive Airway Pressure Program

Abstract

Study Objectives:

Continuous positive airway pressure (CPAP) is effective in treating obstructive sleep apnea in children, but adherence to therapy is low. Our center created an intensive program that aimed to improve adherence. Our objective was to estimate the program’s efficacy, cost, revenue and break-even point in a generalizable manner relative to a standard approach.

Methods:

The intensive program included device consignment, behavioral psychology counseling, and follow-up telephone calls. Economic modeling considered the costs, revenue and break-even point. Costs were derived from national salary reports and the Pediatric Health Information System. The 2015 Medicare reimbursement schedule provided revenue estimates.

Results:

Prior to the intensive CPAP program, only 67.6% of 244 patients initially prescribed CPAP appeared for follow-up visits and only 38.1% had titration polysomnograms. In contrast, 81.4% of 275 patients in the intensive program appeared for follow-up visits (p < .001) and 83.6% had titration polysomnograms (p < .001). Medicare reimbursement levels would be insufficient to cover the estimated costs of the intensive program; break-even points would need to be 1.29–2.08 times higher to cover the costs.

Conclusions:

An intensive CPAP program leads to substantially higher follow-up and CPAP titration rates, but costs are higher. While affordable at our institution due to the local payer mix and revenue, Medicare reimbursement levels would not cover estimated costs. This study highlights the need for enhanced funding for pediatric CPAP programs, due to the special needs of this population and the long-term health risks of suboptimally treated obstructive sleep apnea.

INTRODUCTION

Continuous Positive Airway Pressure (CPAP) therapy is used in the pediatric population to treat the obstructive sleep apnea syndrome (OSAS), especially in children who do not respond to adenotonsillectomy, or have underlying medical conditions such as craniofacial anomalies or Down syndrome. OSAS is associated with short-term and long-term health consequences including neuropsychological dysfunction and cardiovascular disease.¹ In addition, the economic burden of disease treated with CPAP is high. The medical costs of children with OSAS are 215% higher in the year prior to diagnosis, compared to healthy peers, due to hospitalizations, emergency department visits and high rates of upper respiratory tract infections.^2,3 Although no studies have quantified the comprehensive cost of untreated or undertreated pediatric OSAS, the annual cost of treating medical consequences of OSAS in adults is estimated at $3.4 billion.⁴

In adults, randomized controlled trials have shown the benefits of CPAP therapy for OSAS.^5,6 In the pediatric population, CPAP therapy has been shown to be associated with improvements in attention, sleepiness, quality of life and school performance,^7,8 as well as polysomnographic improvements.⁹ Yet, rates of adherence to CPAP therapy in the pediatric population are very low. In part, this may be due to the presence of comorbid conditions and syndromes resulting in developmental delays (such as trisomy 21, Prader-Willi syndrome and other genetic conditions) in many children requiring CPAP.⁹ There is a high drop-out rate, and for those who do use it, nightly use tends to be low.^9–14 Research has demonstrated the importance of team-based interdisciplinary care for patients requiring pediatric CPAP, and shown that the addition of respiratory therapists can lead to improvements in adherence.¹⁵ Therefore, it is critical to develop a more complete understanding of how to provide cost-effective and interdisciplinary solutions to improve rates of pediatric CPAP adherence. The objectives of this study were to estimate the efficacy, cost, revenue and break-even point in a generalizable manner for an intensive CPAP program compared to the prior standard approach.

METHODS

We used the CPAP program at The Children’s Hospital of Philadelphia (CHOP) to examine how models of care impact CPAP adherence and to understand the cost-revenue structure of an intensive (or “high touch”) program structure. Although we focus on the CHOP CPAP program, we constructed an economic model using national estimates to ensure that centers across the country could appreciate the economic impacts of an intensive care model at their institutions.

Structure of the Intensive CPAP Program

The CPAP program coordinates the care of children with OSAS requiring CPAP treatment. In 2012, CHOP restructured their CPAP program to provide more comprehensive treatment for patients and their families. Initial changes included the hiring of a dedicated behavioral psychologist/program coordinator at 0.7 full-time equivalent (FTE), and a 0.5 FTE respiratory therapist. Previous staff, who continued within the new model, included nine sleep physicians who spend approximately 50% of their clinical time seeing CPAP patients (with the proportion of total clinical time varying between 20%–100%); and three registered nurses (RN), each dedicating 0.65 FTE to the program.

Description of Standard Versus Intensive Model of Care

This staffing model changed the workflow and had many important clinical impacts (Figure 1). Some important changes are: First, the nursing staff dedicates time prior to the initial clinical visit to screen medical records to identify patients presenting to the clinic for CPAP initiation (eg, patients with an abnormal polysomnogram following adenotonsillectomy) and to prepare for CPAP therapy by filing letters of medical necessity and coordinating prior authorization with insurance companies prior to the initial clinic visit in order to obtain CPAP equipment by the time of the initial visit (see below). This decreases the time until CPAP therapy begins, and allows for early identification of insurance barriers. Second, at the initial visit, patients meet with the entire care team, including approximately 30 minutes with a sleep physician, 30 minutes for nursing education, 15 minutes for respiratory therapy education and mask fitting, and an additional 45 minutes with the psychologist to identify and discuss barriers to care. This contrasts to the standard care program where patients would meet with a physician and nurse only. Third, patients with certain insurance policies (54% of the patients) receive CPAP equipment from a Durable Medical Equipment (DME) company at the initial visit and spend 30 minutes meeting with the DME representative for training on device use and maintenance. Although obtaining CPAP equipment is rarely a problem for patients in our system, this structure is beneficial to patients because they receive their devices far sooner (as compared to 1–4 weeks with the standard system where the machine is delivered to the home by the DME company at a later date), resulting in earlier therapy at a time when the OSAS teaching and CPAP training is fresh in their minds. It is also beneficial to DME companies because it decreases travel expenses and logistics. Patients in both models then use CPAP at home on a low pressure in order to habituate to the equipment, followed by a CPAP titration polysomnogram several weeks later to determine the therapeutic pressure required. Patients in both models are seen in clinic for a follow-up visit prior to the titration polysomnogram in order to assess adherence and troubleshoot problems as the patient acclimatizes to CPAP. This is, in part, driven by the wait of several weeks for polysomnography at our laboratory. Fourth, all patients receive follow-up phone calls from the health care team (psychologist, nurses or respiratory therapist, as indicated). These phone calls address barriers to utilization, troubleshoot problems and side effects, and facilitate scheduling of clinical visits and titration polysomnograms. Families receive an average of two phone calls between the initial CPAP visit and their follow-up visit; one in the first week after the initial visit and then a subsequent call within the following 2 weeks. They also receive a personal reminder call the week of their follow-up visit. More frequent calls are made if indicated or if the follow-up visit is more than 6 weeks from the initial visit. Fifth, follow-up visits, beginning approximately 1 month after the initial visit, also include the entire interdisciplinary team. The respiratory therapist checks the equipment, downloads the data card to the electronic medical record and evaluates the equipment (eg, mask, filters, tubing, machine) to ensure it is working properly and at the prescribed settings. Nursing evaluates for side effects and reported benefits, responds to any questions or concerns, provides educational support if needed, and assists the physician in ordering supplies. If adherence is considered “excellent” (ie, wearing the CPAP for >80% of nights for > 6 hours per night on nights it is used), the psychologist briefly provides positive reinforcement. If adherence is not excellent, the psychologist reviews the downloaded data with the family, assesses motivation and barriers and develops a plan to improve adherence. The physician reviews and discusses all medical aspects of OSAS, any procedures performed prior to follow-up visit (eg, sleep studies), examines and evaluates the patient, and makes recommendations for continued follow-up care. The education program at the initial and follow-up visits includes: written information, direct demonstrations of the equipment and supplemental videos available to families online.

In the maintenance phase of the program, follow-up frequency is determined by the severity of illness and the degree of adherence. Patients are typically followed every 3 months and then spaced to 6 and 12 months depending on adherence and disease status. The maintenance portion of the program was not considered as part of our economic analysis as it was similar in the standard and intensive models of care.

Economic Model of Intensive CPAP Program

We constructed an economic model to estimate the cost-revenue structure of the intensive CPAP care model. No relevant model exists, as it is difficult to estimate the true costs of the program and construct a reimbursement framework given the variability of payment contracts. We used national benchmark figures to estimate costs, charges and reimbursement so that the data would be readily applicable to other sleep programs and generalizable beyond the experiences of our center. For contractual and proprietary reasons, we are unable to detail our local costs and reimbursement. In order to more fully understand how the intensive CPAP program improved adherence to clinical visits and return for follow-up visits, we used center-specific data on rates of clinical follow-up and completion of titration polysomnograms. CPAP adherence based on device downloads was not used as an outcome, as data from the standard care model (prior to the initiation of the database) were not available. Although we are a large center and prescribe CPAP for an average of 150 new patients per year, we have broken down each of the costs on a per-patient level to make it easier to scale our analysis based on the size of an individual center.

Estimated Cost, Revenue and Breakeven Points

The aggregate costs associated with the CPAP program were constructed using a variety of sources in order to ensure that they were generalizable across institutions and help to shed light on the complexities of hospital-based billing practices. The costs can be broken down into three main categories: provider salary costs, allocated overhead and titration polysomnogram costs. National provider salary estimates were obtained from the 2014 US Bureau of Labor statistics data¹⁶ and the 2015 Medical Group Management physician compensation report.¹⁷ Benefits costs were estimated to be 27% of the base annual salary and were added to the base cost to reflect an aggregate provider cost. Allocated overhead costs were extrapolated from a 2003 report published in Medical Economics that enumerated the estimated annual expenditures for a private practice physician.¹⁸ These figures are broadly relevant because they include direct overhead costs that might not otherwise be included in hospital-based cost accounting systems, but are critically important when evaluating overall cost of a pediatric CPAP program.

It is more difficult to estimate the cost of titration polysomnograms in a generalizable manner because the costs of these services include hospital-specific facilities fees; professional billing and ancillary support services are included in hospital cost accounting systems which for the most part are not public.¹⁹ Therefore, cost data were extrapolated from data from the Children’s Hospital Association Pediatric Health Information System (PHIS). The PHIS data uses Clinical Transaction Classification (CTC) codes that enable billing comparisons across individual hospital billing systems, therefore constructing an aggregate charge figure. We used the CTC code 5150311, which includes all studies listed in the PHIS database as polysomnogram with titration or CPAP polysomnogram. In order to determine the study cost rather than the aggregate charge for these studies, an average cost to charge ratio²⁰ was applied to determine the aggregate study cost.

It is difficult to construct an aggregate reimbursement model because there is dramatic variability in payment levels for pediatric health care services.²¹ Due to the fact that Medicaid often reimburses below the level of cost and many hospitals have very variable payer mixes, we chose to use Medicare reimbursement figures from the 2015 Medicare Fee Schedule.²² We analyzed the physician billing practices at our institution for CPAP initiation visits and follow-up visits in order to find the most commonly used CPT codes. We then used these CPT codes to determine the associated Medicare reimbursement levels. Respiratory therapists and psychologists do not bill independently at our institution; therefore, there is not additional revenue associated with these services. The psychologists in our program do not bill because many patients do not have mental health insurance coverage. The respiratory therapists do not bill independently because, as per Centers for Medicare & Medicaid Services regulations, they can only bill independently for treatments but not education. These figures provide a benchmark that will allow individual health care systems to understand more fully where their reimbursement rates fall relative to these figures. Patient age has a large impact on revenue as the reimbursement for polysomnograms differs for children less than 6 years old and children 6 and older. Studies done on children less than 6 years old are more complex and therefore require a higher level of reimbursements. Our estimates are based on our center where 31% of CPAP patients are younger than 6 years of age.

Using the baseline costs and the Medicare Reimbursement structure described above, we constructed a break-even analysis to look at the level of reimbursement that would be required to break-even on each element of the intensive program and for the program as a whole. This provides a way for individual systems to judge how their average payments (based on their payer mix) stand relative to the Medicare pay rates.

RESULTS

Our center serves a largely urban patient population. The mean age of our patient cohort was 9.8 ± 5.6 (SD) years, with 31% of patients less than 6 years old. 57% were male. 43% were white, 40% African American, 5% Asian and remaining mixed race or other; 8% were of Hispanic ethnicity. As CPAP is usually the second line of therapy for children (after adenotonsillectomy), many children had comorbidities contributing to OSAS including obesity (33%), trisomy 21 (15.5%) and other conditions including genetic syndromes (33%).

In the 4 years since the initiation of the intensive CPAP program, the demand for CPAP therapy has increased dramatically (Figure 2; Spearman rank correlation = 1, p < .001). The intensive program has grown to meet the demands of this patient population and now begins CPAP therapy in more than 150 new patients per year. The intensive program has also led to a dramatic decrease in the time lag between the initial clinical visit and the start of CPAP therapy. 53% of patients get their devices through the consignment program and can start therapy immediately rather than waiting 1–4 weeks for DME delivery. Most importantly, the intensive program has led to increased rates of adherence to clinical follow-up, as reflected by the increase in the number of patients who return to clinic for scheduled follow-up after their initial CPAP visit (Figure 3). Prior to the intensive CPAP program, only 67.6% of 244 patients prescribed CPAP for the first time appeared for a follow-up clinic visit, and only 38.1% had titration polysomnograms. In contrast, 81.4% of 275 patients in the intensive program appeared for follow-up visits (p < .001) and 83.6% of those patients had titration polysomnograms (p < .001). Adherence data were not available prior to the start of the Intensive CPAP Program as there was no systematic schedule for downloading device adherence monitors, and no database to track it. Following implementation of the intensive CPAP Program, for patients initially prescribed CPAP from September 2013-December 2015 who returned to clinic at least one time, patients wore CPAP for an average of 63.6 ± 33.2% of nights during the first month of CPAP use, for an average of 4.3 ± 2.9 hours on nights used. 24% of these patients had “excellent” adherence.

Cost-Effectiveness of Intensive CPAP Program

Initial Clinic Visit

Based on our cost analysis, the initial intensive visit costs an additional $39.86 dollars per patient (Table 1). This cost is driven largely by the additional time spent on services provided by the team psychologist and respiratory therapist which are not billable. We did not include the costs of incorporating the DME representatives in these visits as the only cost incurred by the center is the space overhead cost, but the DME company is using space that would not otherwise be used for patient care. Our model demonstrates that Medicare reimbursement rates do not fully cover the cost of either the standard or the intensive initial visit. Therefore, either these rates are an underestimate of current reimbursement levels or public and private payers do not provide adequate reimbursement to cover care coordination services necessary to ensure proper initiation of therapy for children requiring CPAP therapy. At a rate of 1.44 times the based Medicare rate that we have included, a clinic with a cost structure similar to this template would break even on the intensive initial visits.

Follow-up Phone Calls

As there is currently no reimbursement structure for the follow-up phone calls, these costs must be borne entirely by the Sleep Center. Our analysis demonstrates that the follow-up phone calls cost approximately $17.61 per patient. Though these costs are not insignificant and are covered entirely by the center, they are thought to be one of the key drivers of increased adherence, improved follow-up and improved rates of titration polysomnograms, and therefore a critical component of the intensive program.

Follow-up Visits

As with the initial visits, there is an additional cost of $28.55 to facilitate the intensive follow-up visits and ensure that patients have the opportunity to meet with the entire team. In this model, Medicare reimbursement relative to the estimated costs for the follow-up visits is slightly more favorable as compared to the initial visits. However, the baseline estimate does not meet the template estimated costs. At a rate of 1.29 times Medicare payments, the center would break even on these intensive follow-up visits.

Titration Polysomnograms

Using the PHIS cost to charges data discussed above, we estimate that the average cost of a titration polysomnogram is $1777.63. This is roughly double the weighted average Medicare reimbursement level of $852.72. It is likely that this Medicare rate is significantly lower than the amount that most centers are reimbursed for titration polysomnograms, particularly those for patients under the age of 6 years. Due to the mismatch, it is most important to note that centers whose costs are similar to the template would need to be reimbursed at 2.08 times the Medicare rate in order to break even on titration polysomnograms.

For the intensive program as a whole to break even, centers would have to be reimbursed at 1.92 times the Medicare rates.

DISCUSSION

The intensive model of CPAP care at CHOP provides integrated clinical services to patients and families, has improved the time to obtaining CPAP therapy and boosted rates of clinical follow-up and titration polysomnograms. The rates of adherence demonstrated amongst patients in the intensive program compare very favorably to other reported pediatric data.^19–21 Based on local costs and reimbursement at our institution, we have been able to fully cover our costs and provide this improved patient care model. Depending on the costs, payer mix and reimbursement at other hospitals, this may also be the case.

The primary driver of the break-even point in this model of care is the cost and reimbursement ratio of the titration polysomnograms. This is due to the fact that the costs and payments for the titration polysomnograms are orders of magnitude greater than those for the office visits. Improvements in the percentage of patients who return for titration polysomnograms at our institution has ensured that our center can continue to provide a high level of interdisciplinary care to all of our patients. In order to assess the viability of a program locally, institutions must be aware of the way that polysomnography reimbursement impacts the financial feasibility of this care model.

It is difficult to construct national estimates of cost and reimbursement in the pediatric population. This is due to the fact that there is tremendous variability in payer mix at every pediatric care center, and the fact that reimbursement rates vary widely between private payers and state Medicaid programs. Using national estimates for Medicare ensures that these numbers are generalizable; centers across the country can consider their local reimbursement experience relative to the Medicare level. Sleep centers may need to take into consideration the differentiation between hospital and professional streams of revenue when considering the costs and revenue of introducing an intensive CPAP adherence program

This economic analysis also highlights the importance of reimbursement for care coordination services. Spending just a small amount of money upfront ensures that patients are adherent to their therapies and return for clinic visits. Improved rates of adherence also ensure that patients have access to vital medical equipment as some insurance companies are refusing to pay for equipment that is not being consistently used and DME companies are removing the equipment if families are unable to pay for the devices out of pocket.

The use of behavioral therapists or licensed clinical social workers rather than psychologists might bring down the costs of the model. However, the therapists would require specific training to screen and appropriately address a wide range of psychosocial issues that may impact treatment adherence, as techniques often need to be individualized to the particular patient and need to be tailored for the specific medical intervention. In some places, DME companies may not provide a CPAP device without a polysomnography titration report. This is usually not a problem for pediatric programs as the literature has demonstrated a need for behavioral habituation,^14,23,24 and this is usually spelled out in the letter of medical necessity.

Further research is needed to determine whether an intensive, interdisciplinary approach as outlined here for CPAP adherence is beneficial to other chronic pediatric conditions where adherence is key.

A limitation of this study is the lack of data on CPAP nightly use prior to the institution of the intensive CPAP program as device adherence downloads were not obtained in a systematic manner prior to the development of the intensive program. However, it is assumed that the high number of patients not returning to clinic in the standard program were nonadherent to CPAP, and the clinical impression of the health care providers in the program is that adherence based on downloads has improved dramatically. Although CPAP adherence was high compared to other pediatric reports,^11,14,25 we do not yet have data on long-term adherence, and it is not known whether these benefits continue after months or years of CPAP therapy. However, in adults it has been shown that the pattern of adherence is established soon after treatment initiation, and predicts long-term use.²⁶ An additional limitation is that our cost estimates may not reflect the costs at other centers, including both direct costs and allocated overhead costs. Further studies are needed to objectively determine the impact of the intensive program on actual CPAP use. Our analysis is also limited because it is based on the volume, age distribution and type of patients referred to a single large academic medical center. Further studies are needed to compare experiences across centers, and refine the cost and reimbursement estimates used in this study.

CONCLUSION

An intensive CPAP program leads to substantially higher follow-up and titration rates, but costs are higher. While currently affordable at our institution due to the local payer mix and revenue, Medicare benchmark reimbursement levels would not be sufficient to cover estimated costs. Our economic model only looks at the short-term cost and reimbursement of the intensive CPAP program as it is difficult to estimate the true economic and health benefits of improved CPAP adherence over the course of a child’s lifetime. Studies have shown that children with OSAS have significantly higher rates of health care utilization as compared to controls,²⁷ but little research has been done to show how CPAP adherence changes health care utilization over multiple time points in childhood, adolescence and beyond. Measured improvements in school performance, neurocognitive outcomes and long-term economic and health indicators could add depth to models such as this one and further compel reimbursement for vital intensive CPAP services. Further research is also needed to appreciate whether this model of care lends itself to a bundled-payment system or value-based contracting.

FUNDING

This research was funded in part by the Leadership Education in Neurodevelopmental Diseases Fellowship at The Children’s Hospital of Philadelphia, which is supported by a grant T73MC00051 from the Maternal Child Health Bureau, Health Resources and Services Administration, and the US Department of Health and Human Services. Study data were collected and managed using Research Electronic Data Capture (REDCap).

DISCLOSURE STATEMENT

None declared.

ACKNOWLEDGMENTS

EBR built the financial model and drafted the manuscript. EF conceptualized the financial model and participated in manuscript writing. MSX conceptualized the project, analyzed center-specific adherence data and participated in manuscript writing. SEB assisted with data acquisition and manuscript revision. MKM assisted with data acquisition and manuscript revision. EM assisted with data acquisition and manuscript revision. CLM conceptualized the project, oversaw the care model innovation and participated in manuscript writing. All authors approved the final transcript as submitted and agree to be accountable for all aspects of the work. The authors thank Justine Shults, PhD, for statistical assistance.

REFERENCES

Author notes