Construction of the variables

Hospital operating expenses. Total hospital operating expenses are of such vital concern that most governments want to control these through health policy channels. There has been a wide variation in operating costs even after adjusting for the varying complexity of patient needs, as patient are treated by each hospital at different regional wage levels. This difference indicates that thousands of dollars in additional expenses could have been saved if there had been an effective cost reduction regulation for those hospitals with higher expense structures. Total operating expenses means the operating expenses incurred as part of the delivery of care, including expenses on capital related costs, employee benefits, administrative and general expenses, housekeeping, dietary, cafeteria and so on. The operating expenses are transformed to logarithms to mitigate skewness in the dollar term.

APCDs adoption. Adoption and duration times were used as variables of interest to estimate the impact of the policy adoption itself as well as the history of policy adoption on hospital costs. States that enacted APCDs through legislative statute as well as states with voluntary reporting initiatives (mostly led by nonprofit organizations) are considered adopted and are coded as “adoption = 1.” Since the importance of informal networks for knowledge creation has been increasingly acknowledged, these voluntary efforts may be as influential as formal regulations [32]. Strong interests are treated as “adoption = 0,” as they have not adopted APCDs yet.

Market competition. The Herfindahl–Hirschman Index (HHI) is the standard measure used by the Department of Justice and the Federal Trade Commission to estimate the extent to which the market is concentrated and is frequently used by health care researchers. HHIs are calculated as the sum of the squared market shares of hospitals in a market; a lower value of HHI represents a more competitive market. In theory, markets with more comparable firms are inclined to be more competitive than those that are monopolistic [33]. Therefore, the magnitude of competition is inversely associated with this variable. HHIs are calculated as the sum of the squared market shares of hospitals in a hospital referral region (HRR), where a lower value of HHI represents a more competitive market. Market shares are calculated by dividing the number of beds in a hospital by the sum of beds in the HRR to which the hospital belongs.

Lobbying efforts. The interest group model of regulation finds that legislations are regarded as devices thorough which interest groups gain market power [34]. Interest groups like the AHA take advantage of specialized knowledge to frame legislation in favor of their interests [4]. Furthermore, those special interest groups are likely to affect the dynamics of competition by lobbying the Legislature to pass laws to limit competition [33, 34]. For instance, the AHA sought to support state Certificate of Need laws to restrict the expansion of freestanding ambulatory surgical centers [34]. In this sense, the amount of dollars spent on lobbying at the state level was included to control political influence on policy adoption and competition.

Characteristics of hospitals. The variables associated with the characteristics of individual hospitals are included to control for hospital specific characteristics of the samples. We used hospital size, which was determined by calculating the total number of beds, as larger hospitals may reduce costs in production as the result of economies-of-scale. If hospitals are located in urban areas, are nonprofit, or are teaching affiliated, they are coded as “urban = 1,” “non-profit = 1,” or “teaching = 1” respectively, otherwise they are coded as 0. Adjusted admissions include all patient care activities undertaken in a hospital such as inpatient and outpatient services.