CONSUMER‐DIRECTED HEALTH PLANS: A REVIEW OF THE EVIDENCE

• Author: M. Kate Bundorf
• DOI: http://doi.org/10.1111/jori.12141
• Date: 01 January 2016
• Published date: 01 January 2016

CONSUMER-DIRECTED HEALTH PLANS:AREVIEW OF THE

EVIDENCE

M. Kate Bundorf

ABSTRACT

The number of people enrolled in consumer-directed health plans (CDHPs)

has increased dramatically over the last decade. We review the empirical

literature on the effects of CDHPs, which is based on the experience of these

plans in the employer-sponsored market. Studies indicate that CDHPs

reduce health care spending by approximately 5–15 percent relative to

similar plans with lower deductibles and without spending accounts.

Spending reductions are concentrated among healthier enrollees and are

driven by reductions in the use of outpatient services and pharmaceuticals.

There is little evidence on whether they reduce the use of low-value services.

INTRODUCTION

Consumer-directed health plans (CDHPs) emerged in the late 1990s in the wake of

the public backlash against managed care and the subsequent rise in health care

expenditures. In response to the perception among consumers that managed care

plans were limiting access to potentially beneficial care (Blendon et al., 1998),

consumer-directed plans were intended to control costs by shifting responsibility for

health care decision making from insurers to consumers (Gabel et al., 2002). The

vision was that consumers, exposed to the financial consequences of their decisions

and armed with sophisticated information tools, would drive value-based innovation

in health care delivery (Herzlinger, 2002). CDHP enrollment has grown rapidly

over the last 10 years, rising from 4 to 20 percent of covered workers between 2006 and

2014 (Claxton et al., 2014).

While there is no consensus on the precise definition of a CDHP, these plans are often

associated with three features: a relatively high deductible, a personal spending

account, and the availability of information tools for enrollees. The purpose of the

high deductible, the amount that the insured consumer is responsible for paying out-

of-pocket before the plan provides coverage for services, is to encourage consumers to

make more cost-conscious treatment decisions by exposing them to the financial

consequences of their choices. Like all types of cost sharing, however, high

deductibles create a well-known trade-off for consumers: cost sharing provides

M. Kate Bundorf is at the Department of Health Research & Policy, Stanford University,

Stanford, CA. Bundorf can be contacted via e-mail: bundorf@stanford.edu.

© 2016 The Journal of Risk and Insurance. Vol. 83, No. 1, 9–41 (2016).

DOI: 10.1111/jori.12141

9

greater control of moral hazard in the use of insured services but also reduces the

amount of risk protection insurance provides (Zeckhauser, 1970). To address this,

CDHPs often combine a high deductible with a personal spending account, which the

enrollee can use to fund health expenditures not covered by the plan, to provide

enrollees with greater financial protection.

The third feature often associated with CDHPs is the availability of information tools

to facilitate enrollee decision making. Although the intent of CDHPs is to create

incentives for patients to consider both cost and quality when using medical care,

patients often do not have access to the types of information necessary to make

these decisions, such as the price and likely outcomes of treatment alternatives. The

purpose of information tools is to provide enrollees with the resources they need to

make informed decisions.

Proponents of CDHPs emphasize the potential for these plans to promote greater

value in health care spending and to accommodate diverse consumer preferences

(Cogan et al., 2005; Baicker, 2006; Feldstein, 2006). Critics, in contrast, raise three types

of concerns. First, while consumers may respond to high deductibles by using

less medical care, they may not differentiate effectively between more and less

valuable care when making those reductions, ultimately reducing quality of care, and

greater cost-sharing places an excessive financial burden on low-income and/or less

healthy enrollees (Davis, 2004; Woolhandler and Himmelstein, 2007). Others point to

the potential for greater risk segmentation in health insurance markets if CDHPs

disproportionately attract favorable risks due to their lower premiums and higher

cost sharing (Robinson, 2004; Rosenthal and Daniels, 2006). Finally, doubts also exist

over whether CDHPs, in practice, reduce health care spending. While the 1970s Rand

Health Insurance Experiment (HIE) provides strong evidence of the potential for

high-deductible health plans to reduce health spending (Newhouse, 1993), not only

was the level of the deductible in the HIE much higher than what is typically observed

in the market today (Newhouse, 2004), but the alternative to a high-deductible plan in

today’s market may provide care relatively more parsimoniously than the “free care”

plan in the HIE, making it more difficult for CDHPs to deliver cost savings of the

magnitude observed in the HIE (Ross, 2006). Many plans have adopted supply-side

care management techniques intended to control utilization and others already

incorporate substantial cost sharing. Remler and Glied (2006) estimate that the group

responsible for half of all medical spending in a given year would see either no change

or a decline in their cost sharing under a high-deductible health plan with a savings

account.

Given the differing views over the desirability and potential effects of CDHPs, the

objective of this article is to synthesize the evidence to date on their performance. The

number of studies on the effects of CDHPs has grown along with the presence of

these plans in the marketplace, resulting in a significant body of research.

1

This

literature review summarizes the evidence on who enrolls in CDHPs and how they

1

This review of the literatures updates an earlier study supported by the Synthesis Project of the

Robert Wood Johnson Foundation reviewing the literature published through 2011 to include

studies published through July 2015 (Bundorf, 2012).

10 THE JOURNAL OF RISK AND INSURANCE

differ from enrollees in other types of plans, the extent to which CDHPs experience

favorable risk selection and its implications, the effects of CDHPs on health

care spending and use, differences among CDHPs in the market and the effects of

these differences, and the types of information CDHPs provide to enrollees and the

extent to which enrollees are using it for health care decision making.

BACKGROUND

The consumer-directed umbrella initially covered a variety of innovations in benefit

design intended to promote greater consumer engagement in health care decision

making by exposing consumers to stronger financial incentives at the point of care

and providing them with the types of information necessary to make cost-conscious

decisions (Christianson et al., 2002; Gabel et al., 2002; Robinson 2002). The early

products, which emerged during the “dot-com” era in early 2000, emphasized the role

of the Internet in the development of information tools to facilitate consumer

comparisons of the cost and quality of alternative treatment options. The type of plan

that eventually gained the most traction in the marketplace combined a relatively

high deductible with a personal spending account (Regopoulos et al., 2006).

The development of federal tax policies played an important role in the evolution of

CDHPs. Employer payments for health insurance are treated as a business expense

for employers and not included in the taxable income of workers, creating a subsidy

for employer-sponsored coverage (Pauly, 1986). When CDHPs were initially

conceived, it was unclear whether personal spending accounts could be treated as

tax exempt for the purpose of employee compensation. Because the favorable tax

treatment of employer-sponsored coverage traditionally applied to premiums but not

to out-of-pocket expenditures, high deductibles effectively shifted health care

spending from pre- to posttax compensation. Thus, the favorable tax treatment of

spending accounts from which enrollees could finance out-of-pocket expenditures

was important for preventing less favorable tax treatment of high-deductible plans

relative to other types of coverage (Cogan et al., 2005).

In 2002, the Internal Revenue Service (IRS) clarified that health reimbursement

arrangements (HRAs), an employee-specific account established and funded by an

employer from which employees could be reimbursed for medical expenditures,

could be excluded from the taxable income of employees (IRS, 2002). HRAs can be

funded only by employer contributions, the employer may allow the funds to

accumulate over time, and the funds can be used only for qualified medical

expenditures for employees or their dependents. A key limitation of these accounts is

that they are tied to a particular employer and often a particular health plan. An

employee leaving a firm or changing plans within a firm will not necessarily have

access to any unused balances in the account and will be unable to make additional

future contributions.

2

2

The employer may choose whether to allow former plan enrollees to have access to remaining

account balances.

CONSUMER-DIRECTED HEALTH PLANS 11