Estimation of expected utility gained from the helmet law in Taiwan by quality-adjusted survival time

Abstract

The objective of this study was to empirically estimate the expected utility gained from the implementation of the 1997 helmet law in Taiwan by using quality-adjusted survival time (QAST). We randomly selected 400 out of 8221 registered cases of head injury and successfully interviewed 99 cases with the index of health-related quality of life (IHRQ). The function of IHRQ was then multiplied with the corresponding survival function to obtain the QAST for head injury. The total utility gained from the helmet law in Taipei within 1 year was estimated by multiplying the expected loss of utility per patient with the number of prevented cases. The results showed that after 80 months of follow-up, the QAST of the injured population was 66.3 quality-adjusted life-months (QALMs), while that of the reference population was 78.7 QALMs. We extrapolated the QAST for total life expectancy by simulating the survival of head injury cases using the life table data from the general population. The life-long utility loss of a head injury case was found to be 4.8 quality-adjusted life-years (QALY). The number of prevented cases during the first year of enforcement of the helmet law was estimated to be 1300 cases in Taipei, which amounted to 6240 QALYs gained. For lack of data, we were unable to calculate the possible gain from helmet on reduction of severity among nonfatal cases with head injury, and the estimation was only a lower bound. We concluded that the QAST approach is a feasible approach applicable to health policy decision-making, especially in cost-utility analysis.

Introduction

In many countries, the cost of health services has been growing rapidly for the last several decades (Sorkin, 1992). Although many remedies to control health care spending have been proposed, controversy exists on the long-term effectiveness of simply focusing on the supply side or changing the output of health services in the health care system (Patrick and Erickson, 1993).

Cost-utility analysis (CUA), a common method for evaluating health care programs (Drummond et al., 1987, Robinson, 1993), is used to measure the cost per unit of utility gained. Utility refers to the value or worth of a specific level of health status and can be measured by the preferences of individuals or society for any particular set of health outcomes. Current methods of estimating the utility of health (Morrow and Bryant, 1995) include the following approaches: the Ghana Health Assessment Team (GHAT) approach (Ghana Health Assessment Project Team, 1981), the quality-adjusted life-year (QALY) approach (Torrance, 1986), the disability-adjusted life-year (DALY) approach of the World Development Report (Murray, 1994), the quality-adjusted survival estimation with partitioned survival approach (Glasziou et al., 1990), and the quality-adjusted survival time (QAST) approach (Hwang et al., 1996).

In general, the GHAT approach measures the total utility remaining when an illness occurs, by assigning a specific quality of life value to patients and estimating their life expectancy from the general population’s life table. However, most chronic diseases, e.g. stroke or diabetes mellitus, will usually shorten patients’ life spans and progressively lower their HRQL (health-related quality of life) with time. As a result, the GHAT calculation may not be an accurate estimate of utility loss.

QALY is arrived at in each case by adjusting the length of time affected through the health outcome by the utility value (on a scale of 0–1) of the resulting level of health status. The usual QALY approach may be accurate and applicable for clinical trials with a short termination point. Yet, it is usually not feasible to continue measurements of HRQL throughout a patient’s survival period if it is too long, e.g. more than 10 years. The DALY approach avoids the above pitfalls by generating both a survival function and a disability-adjusted utility function over time from a group of experts. However, a democratic society may question the direct use of DALY in resource allocation because the people of a community must rely on an authority’s opinion, rather than their own, to measure survival function and a disability-adjusted utility function over time (Gold et al., 1996). Glasziou et al. proposed the partitioned survival approach to handle this problem (Glasziou et al., 1990). While their model treated the progress of illness as several discrete states, which is applicable for certain types of cancer, it may not be equally applied to other chronic diseases, of which disease states are often highly variable.

Hwang et al. proposed the QAST approach (Hwang et al., 1996) which simultaneously considers both survival and HRQL values over time directly from the patients. Thus, QAST seems suitable as a universal approach to evaluating health care. However, the QAST methodology has not yet been demonstrated in an empirical study.

Head injury has been a significant public health problem in Taiwan, occurring at a rate of about 182 per 100 000 population per year in Taipei and 304 per 100 000 population per year in Hualien, according to a comprehensive registration program, which included all cases came to the emergency rooms (Chiu et al., 1996). A sample was obtained from this registration program, and the survival function and HRQL for cases of head injury were calculated. Then, the average utility loss per case of head injury was estimated using the QAST method.

A helmet law was implemented in Taiwan on January, 1997 and enforced beginning on June 1, 1997. The proportion of motorcyclists with no helmet decreased dramatically, and the occurrence of severe head injury decreased thereafter. The purpose of this study was to calculate the number of head injuries prevented and the QALYs saved from the enforcement of the helmet law by using the QAST approach.