Consumers, providers, and payers all play a role in health care spending. Most changes to the health care delivery system are aimed at changing the behavior of consumers, providers, or payers. Of course, there are myriad other parties that can influence health care spending, such as medical device manufacturers, public health agencies, and pharmaceutical companies. Given the magnitude of the health care delivery system, any one initiative is not likely to have a significant impact. Many innovations and initiatives will be needed to have a substantial influence on health care spending in the United States.

The table below shows the percentage of National Health Expenditures by certain types of services. In an ever-changing health care system, it is interesting to observe how health care cost distributions change over time.

Table 1: Percent of National Health Expenditures by Type of Service

Below is an overview of items discussed in this brief:

• Primary care providers (PCPs): PCP care delivery is essential to an efficient health care system. Nurse practitioners and physician assistants can also be responsible for primary care services. PCPs are responsible for the majority of care for most individuals. Primary care services, when provided on a timely basis, can reduce more-costly services such as specialist visits, emergency room visits, and hospital admissions. This brief discusses initiatives that focus on primary care such as Direct Primary Care and increasing the number of PCPs.
• Providers: Providers drive many health care decisions. It has been shown that there is significant variation in provider practice patterns with no material impact on outcomes. There are also many cases where providers recommend treatment that is not optimal. This issue brief discusses high value networks, value-based care models of care, and comparative effectiveness initiatives that attempt to shift behavior to the most appropriate treatment.
• Consumers: Consumers can impact health care costs by choosing efficient, high-quality providers and seeking effective care at a low-cost setting. Historically, member cost-sharing was used in an attempt to manage costs by sharing the burden of the care and to reduce unnecessary utilization. However, this one-size-fits-all approach does not differentiate between high-value and low-value services and provides consumers little incentive to undergo essential treatment while forgoing low-value services. It can be difficult for consumers to get reliable, understandable, and timely information to help them make these decisions. This issue brief discusses initiatives such as transparency and comparative effectiveness, which are designed to make consumers smarter shoppers.

This brief aims to look at various proposed approaches and provide an understanding of the mechanics and the potential impact these approaches can have on the health care system. Some of these initiatives will be successful, and some may not. It is expected that health care delivery and financing will continue to evolve over time but that many of these evolutions will provide information necessary for future decisions related to changes in the health care delivery system.

Direct Primary Care

Direct Primary Care¹ (DPC) is a primary care practice model in which patients receive unlimited access to a defined set of primary care services in exchange for a monthly membership fee. DPC contrasts to concierge medicine in that it does not take payment from third-party payers such as insurers, Medicare, or Medicaid, and monthly DPC fees are much lower than typical concierge fees—monthly DPC fees for an adult average about $70 per month.² DPC is a growing primary care practice model, with proponents frequently highlighting the potential to improve patient outcomes, lower health care expenditures, and increase physician satisfaction.^3,4

In addition to unlimited access to a defined set of primary care services, DPC physicians typically negotiate deeply discounted cash prices with local labs and imaging centers, and they dispense prescription medications at wholesale prices (where states allow). For many patients, the savings on prescriptions more than offsets the membership fees. DPC physicians say that they dispense many medications at 90-95% off retail price.

DPC practices vary in how they approach their business model. Some focus mainly or exclusively on direct relationships with patients; others market to employers. Some practices use a pure DPC model and do not accept third-party payments from insurance companies, Medicare, or Medicaid for any of their patients. Other practices use a hybrid DPC model in which they accept payments from a small number of third-party payers for a subset of their patient panel – the total cohort of patients a provider serves - and they have a direct relationship with patients and/or employers for the remainder of their panel. Hybrid companies are emerging, which make DPC a more viable offering for employers with multiple locations as they essentially create a network of DPC practices that all employees can access.

Although evidence about patient outcomes in DPC is limited, similar models provide evidence of possible DPC benefits. A model with strong similarities to DPC in terms of much smaller panels and increased access to primary care has shown significantly better patient outcomes for its patients compared to similar patients receiving traditional primary care, including 37%-56% fewer non-elective hospital admissions and 23%-49% fewer avoidable hospital admissions compared to similar patients receiving traditionally delivered primary care.⁵

Specific experience reported by DPC companies is promising:

• Data reported in 2010 by DPC practice Qliance showed that DPC patients had 35% fewer hospitalizations, 65% fewer emergency department visits, 66% fewer specialist visits, and 82% fewer surgeries compared to regional benchmarks.⁶ Qliance also reported 2013–2014 data for selected large employer clients that showed claims for DPC patients (including DPC fees but excluding prescription drug costs) were almost 20% lower than claims for patients that received traditional primary care.⁷
• DPC company Nextera published an employer case study in 2016 that reflects the experience of 205 employees and dependents enrolled with Nextera. The case study showed a 25% reduction in claims costs for Nextera members of the employer compared to a 4% reduction for the employer’s non-Nextera members.⁸
• Iora Health, a DPC practice with more than 25 locations that only care for patients aged 65+, saw inpatient hospital admissions drop by half and emergency department visits decline by 20% over an 18-month period.⁹ Importantly, the cohort of 1,176 Medicare enrollees in this study were 1.5 to 2 times sicker than the average Medicare beneficiary according to diagnosis-based risk models.

Little scholarly research regarding patient outcomes in DPC has been published; the Qliance, Nextera, and Iora Health results were self-published. It is not known whether DPC patients are different from patients of traditional delivery models in ways that are difficult to observe or measure, and the extent to which DPC practices tend to attract different sorts of patients than use traditional primary care is not known.

One key to lower health care spending and improved patient outcomes in the DPC model may be the availability of the physician and primary care services. DPC physicians offer same-day scheduling for urgent needs and visits of 30 minutes or more are standard. More than three-fourths of DPC physicians in a 2015 survey reported that they offer 24-hour access.¹⁰ DPC physicians typically have much smaller patient panels than primary care providers who practice in an insurance model. The American Academy of Family Physicians estimates that DPC panels have 600-800 patients and traditional panels have 2,000-2,500 patients. Another notable difference between DPC physicians and traditional PCPs is the amount of time spent on administrative tasks. One study found that traditional PCPs spend nearly an additional two hours on electronic health records (EHRs) and other administrative work during the clinical day for each hour of direct face-to-face time with patients. Also, outside of office hours, physicians spend another one to two hours of personal time each night doing additional computer and other clerical work.¹¹ Because Direct Primary Care providers do not accept payment from insurance companies or other third-party payers, they are freed up from many of the administrative tasks that their traditional peers must perform, such as pre-authorizations, extensive population of EHRs, and capturing of quality data that doesn’t directly improve patient care. This freedom from administrative tasks means that more of their time is available for patient care.

The lack of access to primary care in traditional insurance-based models manifests in unwanted delays in obtaining an appointment, inadequate length of physician visits, and unmet health care needs. Reports of a lack of adequate primary care access are common. In a 2008 survey, 31% of privately insured patients reported an unwanted delay in obtaining an appointment for routine care, and 20% reported an unwanted delay in obtaining an appointment for illness or injury.¹² In a 2006 California survey, two-thirds of patients who used an emergency department for a problem they believed could have been handled in primary care would not have gone to the emergency room had they been able to obtain a primary care appointment.¹³ The emergency room is one of the most expensive settings in which care is received. The increased access to primary care through Direct Primary Care may lower utilization in the emergency room and offer savings for individuals, employers, and governments by offering patients primary care availability and access, which reduces the utilization of many expensive downstream services and improves patient health.

Another key to lower health care spending in the DPC model may be that lower administrative burden leads to lower practice overhead. It is common for a traditional practice to employ up to four administrative employees for each clinician;^14,15 conversely, it is common for DPC practices to have no employees entirely devoted to administrative tasks such as insurance pre-authorizations. It is estimated that processing claims through insurance accounts for 40% of the cost of traditional (or insured) primary care.¹⁶ Lower overhead of DPC practices leads to lower price points for patients and employers.

Employers who replace traditional primary care with DPC in their benefit plans experience savings of 15% to 30% on medical benefits.^17,18 The most significant savings could be realized when a so-called wrap-around insurance plan is used to cover services not provided by DPC. The savings achieved when transitioning from a traditional insurance plan to a combined DPC + wrap-around insurance plan could be significant, but the actual level depends on a variety of factors, such as benefit levels under both programs, patient engagement, provider discounts and treatment practices. Savings in the DPC + wrap-around insurance plan would be a result of improved health, lower utilization of downstream services, lower prices of services or products, and lower administrative costs.

At least 75% of health care services can be provided in a primary care setting. Using insurance to finance primary and other care that is readily affordable by all but the poorest patients can often double the cost. The primary care component of insurance premiums typically includes 15%-45% for carrier overhead and profit,¹⁹ plus higher provider reimbursements to cover the overhead of administrative employees and coding and billing systems. Additionally, the presence of a third-party payer increases utilization; people consume more care when low out-of-pocket payments make it appear to be less expensive than it really is. Fundamental actuarial principles tell us that it is not economically efficient to use an insurance mechanism to finance primary care and other health care expenses that are common, predictable, and relatively inexpensive.²⁰

Value-Based Payments

The value-based payment model, in which health care providers are paid based on outcomes, has emerged as a prime example of smarter spending. Medicare has taken a lead in setting the tone and agenda on delivery system reform, which is intended to address excessive spending on health care in the United States. The Affordable Care Act (ACA) created Medicare accountable care organizations (ACOs). The trend continues with the Medicare Access and CHIP Reauthorization Act of 2015 (MACRA). 

In a value-based payment model, provider payments are based on the quality of care that they provide rather than exclusively on the quantity of services performed, as is the case in fee-for-service models. Outcome criteria used to determine quality include the reductions in the number of unnecessary procedures performed, quality of care management markers for individuals with chronic illnesses, and the use of evidence-based health care management. The value-based payment model is designed to meet several criteria used to assess new models of care: improving patient experience of care (quality and experience), physician satisfaction, improving the health of a population, and reducing per capita cost.

The value-based payment model has several objectives. It is designed to focus more attention on preventive care and management of diseases. The model will increase efficiency by combining various services under one episode of care as a means of paying for value rather than volume. The increased efficiency can be achieved by decreasing the number of unnecessary services performed or shifting services to a more efficient site of care. The emphasis on preventive care and condition management also aims to reduce the overall cost of care and increase quality of life. This increased focus on patient outcomes is intended to lead to more effective use of resources, such as substituting prescription drugs and medication therapy management for medical cost which reflect an emphasis on prevention and maintenance rather than treatment for an episodic event.

Currently, value-based payments, appearing in the form of bundled payments, are becoming more common in all payer areas, Medicare, and commercial insurance, and are emerging in Medicaid as well, especially in ACOs. Patient-centered medical homes (PCMHs) also use value-based payments or capitation rates to pay providers based on certain quality criteria. The team and patient focus under PCMHs have significant potential to reduce costs and improve outcomes especially for patients with chronic including or mental health conditions.

Value-based payment models incentivize evidence-based care in order to reduce wasteful utilization, especially for services that provide little value. Those value-based payment models that focus on total cost of care or cost of an episode may reduce costs for implants or specific treatments, as well as mitigate the impact of high drug prices, as providers are encouraged to favor cost-effective drug therapies. This has encouraged pharmaceutical to reduce prices on drugs that offer little or no additional clinical benefits over less expensive alternatives.

Value-based payment models have a direct impact on improved patient experience, especially as member cost sharing for those who have high deductible coverages expose them to financial risk for the care they receive. 

Value-based payment models with real consequences for cost overruns should have an incentive to provide the highest quality of care for the lowest price. Providers are incentivized not to order services unless the services are justified by better outcomes. For example, the C-section rates in some regions in the United States are higher based on the delivery system. To the extent that high C-section rates are not clinically justified by superior health outcomes, comparison of data within the United States would show high C-section rates in certain regions as an issue. Comparative analysis of United States national data may not show the full extent of “value” enhancement possibilities if practice patterns are similar across the country, so international benchmarks may be desirable for some services.

Another source of excessive health care costs, avoidable hospital admission, may be reduced by the use of value-based payments. The use of admission data in the calculation of “value” would make the calculation more complicated, although valuable. Everything else being the same, the idea is to reward providers who help patients avoid inpatient admissions and costly emergency room visits. Value-based models that focus on the reduction in unnecessary admissions often target specific diagnosis-based admissions, readmissions, and evidence that hospital care is being used in place of more timely and less acute maintenance care.

A value-based payment model can also impact utilization of elective surgeries, such as invasive procedures for appendectomy, hysterectomy, back surgery, and knee and hip replacement. Several resources, including Choosing Wisely, can help frame the discussion about making sure that the intervention is the best option for helping the patient make a full recovery. These discussions can be included in a value-based payment model by using benchmark measures as a comparison for identifying optimal treatment patterns. 

Under a bundled payment approach, all services for an event are included in a single payment. Bundled payments have been used for a long time in the inpatient setting in the form of diagnosis related groups (DRGs) where reimbursement is a single payment, which varies by diagnosis, that covers all inpatient services associated with an admission. However, payment bundles are extending to a broader list of treatment episodes and more types of providers. For example, a bundled payment might be for hip or knee replacements or maternity deliveries in which the set of bundled services and potential complications are well-defined and -known. The single payment covers all related services; therefore, providers have an incentive to provide the most cost-efficient care and avoid high-cost complications in order not to exceed the payment they receive. 

Provider-Based Programs/Networks²¹

Efficient and effective care can be achieved when a provider network is organized and incentivized to deliver health care consistent with the goals described in the previous section. Such groups of providers can have a variety of structures and labels, be made up of various types of providers, and offer care to different populations. All are financially incentivized to achieve certain targeted results. This issue brief refers to them as provider-based programs or networks. Examples of these are: high-performance networks, accountable care organizations, patient-centered medical homes, Medicare Advantage or Medicaid managed care networks, and alternative or narrow networks. Appendix A provides some examples of how these networks and providers are organized. 

Results achieved by these provider-based programs vary widely. In general, better performance programs require the coordinated effort of a wide array of providers (physicians, hospitals, nurses, supporting staff, etc.) working together toward a set of targeted results that are aligned with known financial incentives, under a strong management team. 

There are three key drivers that enable programs/networks to achieve high performance—goals, management, and individual performance.

Performance starts with the goals of the providers. In traditional fee-for-service programs, provider revenue is generated by the volume of services and the fees charged per service. With few exceptions, reducing wasted services and smarter spending meant less revenue and net income. With the changing health care environment—such as the financial pressures on buyers, refined payment systems, legislative and regulatory changes, new management tools, and encouragement of stronger physician roles—providers have broadened their business strategies.²² When the financial goals of the providers (increased net income) and the payers (affordable care) are aligned under the framework of the Triple Aim, agreements on how to achieve these Triple Aim goals become easier. 

The catalyst comes from payers, through specific programs and alternative payment systems. The implementation must be driven by a small group of committed individuals and/or organizations that take up a broader role and responsibility for care, spending, and health results. A provider management team aligned with payers’ goals can have a major impact over time. The team builds capabilities and infrastructure, identifies and supports existing high performers, and strives to expand the program to improve the performance of other providers over time. 

Different parts of management’s role are customized by types of members and types of providers. The division of managerial roles between primary payer, carrier, and provider organization varies widely; each has strengths and weaknesses. Economies of scale, infrastructure, joint purchasing, and other support offered by organization often can improve performance. As one example, some Medicare Advantage programs have the primary payer set the payment structure and financial terms while the rest of the managerial duties are split between carrier, physician group, and hospital. In this situation, there is a formal agreement done through Definition of Financial Responsibility (DOFR)²³ agreements. 

Overall performance is the result of performances by individuals, which can vary widely. Often, individual providers do not have the information or support system to take a broader role. The primary payer or local manager would need to address the challenge of identifying and communicating the variability of performance across the system.

The primary payer must understand the goals and existing performance of provider groups, particularly if the payer has a management role. An uncommitted provider combined with weak financial agreement does not produce results, and often wastes time. The primary payer must continue to drive financial results, design an effective program through discussion with provider groups, utilize strong alternative payment systems, have ongoing monitoring, and require action plans when targets are not met. Additional information is available in the Academy issue paper, High-Performance Networks.²⁴

Examples of Initiatives and Levers Available to Improve Affordability and Produce Financial Results

Efficient health care requires real-time decisions and action; the provider community must be actively engaged in improving quality, service, and financial performance. There must be management of the local delivery system and actions to reduce wasted services and related expenses, with attention paid to all services and costs, from routine to high-cost interventions. Effective management has creative approaches to all types of services, such as providing care in alternative settings that are appropriate, effective, and efficient. It does not focus just on a few items like reducing admissions or emergency visits.

Efficient programs/networks implement a wide range of initiatives and levers to achieve the goals of better care, healthier people, and smarter spending. These actions must be aligned with the payment structure.²⁵ Table 2 provides examples of some basic initiatives/levers that are used to manage clinical quality, admissions, emergency room visits, and patient support for common chronic illnesses, high-risk patients, high-cost patients, and pharmacy.

Table 2: Care Management Initiatives/Levers

Networks are not a concept that automatically works. Some networks produce better performance. Networks that produce higher performance and affordability typically have strong management. They also distinguish between the subset of providers who have aligned financial goals with payers and support more affordable programs and others who are primarily concerned with their own revenue growth. Payers must identify, support, and challenge provider groups/networks to meet their needs.

Administration and Retention

Administrative costs in our health care system often refer to the administrative costs of payers, such as insurance companies, Medicare, and Medicaid, but may also refer to administrative costs of providers and vendors. This section will mainly discuss payer administrative costs but also briefly discusses provider and vendor administration when applicable.

Payer administrative costs include the costs of membership enrollment, billing, information services, compliance, and claims handling. They also include other general overhead expenses such as building costs or rent, taxes, executive salaries, IT infrastructure, and actuarial support. Additionally, there may be items that increase administrative costs that are intended to decrease claims costs, such as utilization management and claims processing system pending of claims. A return-on-investment analysis would be needed to determine the cost-effectiveness of these items along with determination of the inconvenience to providers and consumers caused by these items. In general, standardization and automation would tend to lower administrative costs, but a corresponding analysis would be required to determine any effect on the claims due to these changes. Research and product development costs would be part of the administrative costs for many of the providers, vendors, and payers. Marketing costs can be a significant portion of administrative costs, especially for providers whose revenue streams have little limitation, such as when paid as a percent of charges by payers.

Compliance costs are highest for Medicare and Medicaid programs, but new transparency and state reporting requirements can be heavy burdens. These requirements include items such as reporting on appeals processes, formulary structures, readability of literature, loss ratio reporting, and market conduct studies.

There can be other non-claims cost expenses called retention, which includes profit and contribution to surplus loadings for providers, vendors, and payers. Limits may exist for payers but often do not exist for providers or vendors. Commissions may also be paid to brokers for group business, although there are some state regulations on amounts brokers may receive. Marketing costs, especially for providers and vendors, can be a substantial cost—especially, similar to profit, when revenue for the provider or vendor is based on a percent of charges.

These administration and retention charges are often 15% to 20% of total costs, although Medicare has a lower administration percent, in part, because the volume of Medicare claims per contract is often several times higher than the claims per contract of group insurance. For commercial insurance, the Minimum Loss Ratio (MLR) regulations require that at least 85% of premium dollars are spent on medical claims and quality improvement for large group plans and 80% for small group and individual plans. Because retention is often calculated as a percent of premium, as premium increases, so does administration. A per member per month (PMPM) cost for administrative expenses that increases at inflation may help moderate administrative costs and be more aligned with actual cost increases. Because claims are generally in excess of 80% of total costs, this brief concentrates more on the claims cost, but an overview of the administrative and retention cost may be helpful.

Comparative Effectiveness Research

Although precise definitions of the term vary, comparative effectiveness research (CER) compares various treatment paths for the same condition. The Institute of Medicine (IOM) uses this working definition:

"CER is the generation and synthesis of evidence that compares the benefits and harms of alternative methods to prevent, diagnose, treat and monitor a clinical condition, or to improve the delivery of care. The purpose of CER is to assist consumers, clinicians, purchasers, and policy makers to make informed decisions that will improve health care at both the individual and population levels."²⁶

While cost is not embedded in the definition of CER, by seeking to provide the most appropriate treatment path, providers will often (but not always) reduce the cost of care. Expanding the scope of CER to include cost, primarily when there are two equally effective treatment paths, is an obvious approach to lower health care spending with the same outcomes.

To date, CER has not been widely adopted. Two primary difficulties in achieving significant savings from CER research are:

• Wide dissemination of the information to physicians and patients to inform the decisions; and
• Changing provider (and to a lesser extent patient) behavior. Even with dissemination of CER, adoption is slow. Financial incentives may hasten adoption but have not been widely implemented.

For more information see the American Academy of Actuaries’ full issue brief, Comparative Effectiveness Research²⁷

Transparency

Price transparency can be an effective way to reduce health care costs if the provider or consumer has some incentive to use the additional knowledge and the knowledge is accessible and applicable. In a scenario where the consumer’s financial responsibility is limited due to low copays or out-of-pocket maxima and the provider has no incentive to choose a more efficient treatment or provider, there may be minimal savings. However, in a scenario where either the consumer or provider has an incentive to choose the more efficient service, then the transparency initiative may produce savings.

Transparency can be made more difficult as prices for a service can vary by provider, payer, product, or location, along with the benefit level of the individual seeking the service or product. Some standardization in pricing may assist in transparency.

Some employers have implemented reference pricing, where a price is set for a treatment, such as a hip replacement. The employer will reimburse up to the reference price but not above it. The reference price can be set using many methods but is often set at a percentile (such as the 50^th or 75^th percentile), so the member is able to find providers at or below the reference price relatively easily. This can also encourage providers above the reference price to reduce fees in order to maintain volume. Alternatively, it could encourage providers charging below the reference price to increase fees to the reference price. 

Reference pricing is only effective if there is transparency in the fees for services. Otherwise, the member could end up with a substantial bill for the procedure due to lack of information. Reference pricing is most effective in non-emergency, elective procedures that allow time for the member to comparison shop for services.

Price transparency could be especially effective if used in conjunction with comparative effectiveness research, allowing consumers and providers to work together to find the most effective and efficient treatments and products. 

Appendix A: Examples of Networks

The types of networks seen in different regions of the country vary according to the local business and legal environment. 

Networks developed by providers

Providers in this type of network can work together in many ways. Some are through formal arrangements and some can be formed through informal working relationships between providers. Some provider-based networks are major forces in their market. 

Hospital-ownership—The most common example is major hospital owning other parts of the health system. This can include outpatient facilities, physicians, nursing homes, and other part of the delivery system.

Physician groups—Groups of physicians often form an “association” or “clinically integrated network” for various business reasons including contracting and clinical improvement. This could be single or multi-specialty. Some physician groups are very extensive and serve as ACOs or building blocks for carrier-based networks.

Hospital-owned insurers—There are some very formal and extensive provider-based networks, such as a staff model HMO, hospital systems that owns physicians, or an insurance carrier owned by a hospital. 

Joint network development—Hospitals join to form a cooperative network covering an expanded geographic area or a member. Shared contracting and member engagement in a risk-taking structure represents a package for the payer. 

Depending on their goals and capabilities, these provider networks can drive smarter spending or block improvements in their communities. 

Networks developed by governments and carriers

Medicare—Medicare members get Original Medicare in all locations, where there are a broad range of participating providers. In many locations, members can join Medicare Advantage plans and access care through a network of providers, in exchange for better benefit packages and/or lower premiums.²⁸

Medicaid—The historic approach offers an extensive hospital network and a smaller physician network of physicians who accept the Medicaid fee schedule. However, in recent years, many different networks are being tested and used for Medicaid members. Many states use various alternative networks within Managed Medicaid. Some states also work with primary care physicians in patient-centered medical homes. Some states also use illness-specific bundled payments, support for high-risk patients, or other initiatives. 

Exchanges—As has been widely reported, alternative networks with a narrower set of participating providers in some states have produced much lower premiums and are the foundation for products used in some health Exchanges. 

Employers—In some states, employers offer High Performing Networks for insured and self-funded employers. For example, various alternative networks are core products offered to state government employees. Many are developed by insurance carriers, but some are developed by providers that have their own insurers. Some of these networks have produced higher performance including consistently lower premiums. 

New examples of networks

In recent years, there have been many voluntary networks developed and implemented. For example, accountable care organizations, patient-centered medical homes, etc. support populations with specific condition payments, and multiple programs are aimed at high-risk members or people. Some of these—for example, physician-based ACOs—see better financial performance. 

Hospitals, physicians, and other providers with the right support have demonstrated a positive impact in some parts of the country.