Does it Matter Who Pays for Auto Injuries?

• Author: Eric Helland,Paul Heaton
• Date: 01 December 2019
• DOI: http://doi.org/10.1111/jori.12245
• Published date: 01 December 2019

DOES ITMATTER WHO PAYS FOR AUTO INJURIES?

Paul Heaton

Eric Helland

ABSTRACT

In many states, auto insurers rath er than health insurers pay for a

substantial fraction of t he medical care following aut o crashes. We

examine whether payer identi ty affects the care received by au to injury

patients. A 2003 Colorado r eform shifted a large fractio n of auto injury

patients from coverage thro ugh auto insurers to the traditio nal health

insurance system. Despite negl igible changes in auto injury charact eristics

during this period, treatment supply increased following the reform.

Procedure use rose by 5–10 per cent and billed charges rose by 5 per cent.

These changes reflect an incr ease in resources devoted to treatment, yet do

not improve mortality.

INTRODUCTION

Although policy discussions of medical payment reform in the United States

focus largely on Medicare, Medicaid, and private health insurers, property and

casualty (P&C) insurers play an important role in the provision of medical care

for certain types of injuries. For auto crash injuries in particular, P&C insurers

represent a significant payer, with data from the National Association of

Insurance Commissioners (2012) indicating that P&C insurers reimburse

roughly $45 billion per year in medical care. Because many states require

drivers to hold first-dollar medical insurance as a part of their mandatory

auto insurance, P&C insurers are much more involved in this portion of a

medical care market otherwise dominated by private and governmental health

insurers.

From a patient welfare perspective, the type and amount of medical care received

following an illness or injury should not depend on the financial arrangements used

to pay for the care. Nevertheless, abundant evidence indicates that physicians

respond to financial incentives of various sorts.

1

While the notion that payment

Paul Heaton is at the University of Pennsylvania and RAND. Heaton can be contacted via

e-mail: pheaton@law.upenn.edu. Eric Helland is at Claremont-McKenna College and RAND.

Helland can be contacted via e-mail: ehelland@cmc.edu.

2018 The Journal of Risk and Insurance. Vol. 9999, No. 9999, 1–26 (2018).

DOI: 10.1111/jori.12245

1

947

947

Vol. 86, No. 4, 947–972 (2019).

matters for the provision of medical services is not controversial, much less well

explored by researchers to date is whether there are any impacts arising from the

particular financial arrangements that exist when auto insurers versus traditional

health insurers pay for medical care. One reason this question is important is because

there is substantial variation across states in their requirements for medical coverage

related to auto injuries. Currently, 15 states require drivers to purchase personal

injury protection (PIP) as a part of their auto insurance, which covers medical care for

their injuries following an accident and which is typically primary to health

insurance. However, a number of these states have engaged in policy discussions that

could limit the role of PIP or otherwise alter the role of auto insurers vis-

a-vis

traditional payers in the provision of postinjury care.

The paucity of evidence regarding the impacts of P&C insurers in the medical care

market partly reflects the difficulties inherent in isolating the effect of the source of

payment from other factors that might determine the amounts or typesof care offered

patients.Cross-state analysesthat compare patients in statesthat require auto policiesto

cover medicalcare to states without such requirementsare likely to confoundthe role of

the payer with other factors that vary across states, such as the supply of medical

services, fee schedulesfor government payers, and nature of auto accidents.Although

within-state comparisons are also feasible, because some drivers in states without

mandatoryinsurance requirementspurchase electivemedical coverage as a component

of their auto policy, such comparisons are likely to face selection problems. The

population who chooses to purchase add-on insurance is likely not comparable to the

population withoutsuch insurance. At a conceptuallevel, the ideal comparison would

involve examiningthe medical care received by two otherwisesimilar pools of drivers,

one of which was required to carry medical coverage through an auto policy and the

other of which had coverage through traditional payers such as health insurance.

In this article, we exploit a legal reform to examine whether patientscovered through

auto insurancereceive different types of medical care frompatients with health or other

more traditional forms of coverage.Prior to 2003, drivers in Colorado were required to

hold medical coverage as a componentof their auto insurance, but a controversial law

change in that year removed the requirement, leaving future auto crash injuries to be

1

See, for example, Fuchs (1978), Rossiter and Wilensky (1983), and Cromwell and Mitchell

(1986), who examine differences in the number of physicians across locations as a proxy for

reimbursement. Subsequent research has demonstrated that these studies likely suffer from

omitted variable problems (Auster and Oaxaca, 1981; Feldman and Sloan, 1988; Dranove and

Wehner, 1994). More recently, Gruber and Owings (1996) use exogenous shocks to fertility to

identify the effects of availability of patients, a proxy for physician income, on physician use of

Cesarean sections. An alternative test examines how procedure intensity varies as fees for

particular types of treatment are adjusted. Examples of studies employing this methodology

include Rice (1983), Christensen (1992), Escarce (1993), Nguyen (1996), Yip (1998), Zuckerman,

Norton, and Verrilli (1998), Mitchell, Hadley, and Gaskin (2000), and Hadley and Reschovsky

(2006). McGuire (2000) provides a more thorough review of this literature. See also Iizuka

(2007) who shows that physicians are more likely to prescribe drugs for which they earn a

higher markup.

2THE JOURNAL OF RISK AND INSURANCE

2THE JOURNAL OF RISK AND INSURANCE

948