Do Health Care Regulation and Physician-Industry Interaction Influence Antibiotic Resistance Rates? The Example of Antimicrobial Prescribing and Dispensing in Japan Harbarth S, Oberländer C University of Geneva Hospitals, Switzerland; University of Tokyo, Japan

BACKGROUND Widespread antimicrobial resistance (AMR) amongst bacterial pathogens has compromised traditional therapy with narrow-spectrum antibiotic agents and may result in adverse outcomes. Among all industrialized countries with populations of at least 30 million, Japan has one of the highest proportional frequencies of AMR in community- and healthcare- associated infections. No previous study has attempted to elucidate reasons to explain this worrisome finding.

AIMS & METHODS The purpose of the current study was to describe the pharmaceutical reimbursement system, regulatory practices and physician-industry interactions that may influence antibiotic use and AMR patterns in Japan. A computer-based literature review was undertaken using the MEDLINE database from 1975 to the present. Additional references were identified from the references cited in retrieved reports and personal files. Papers in English and Japanese were reviewed.

CONSUMPTION OF ANTIBIOTICS IN JAPAN The use of antimicrobial drugs in Japan is remarkably high. In 1999, per-capita pharmacy sales of antimicrobial agents amounted to $US 17 in Germany, $US 32 in the USA and $US 46 in Japan. Japan used expensive broad-spectrum antibiotics (e.g., cephalosporins, fluoroquinolones) more frequently than any other industrialized country. Worldwide fluoroquinolone use (prescriptions/ 1000 pop/ year)

ANTIMICROBIAL RESISTANCE AMR is a serious problem in Japan, especially among Gram-positive bacteria. In a recent survey, 15% of Japanese children (n=270) were MRSA carriers. Percent penicillin non-suscept. pneumococci

Pharmaceutical reimbursement system For historical and cultural reasons, Japanese health care providers traditionally rely on the profit derived from the difference between the reimbursement rate for pharmaceuticals and the market prizes of drugs to subsidize their activities. This is known as “provider pharmaceutical profit”, which may vary between 10 and 40% of the drug cost reimbursed by the fee schedule. As a result, it has been the custom for physicians to compensate for relatively low medical service revenue by prescribing a high volume of drugs. This incentive to overprescribe leads to high expenditures on antibiotics.

Separation of prescribing and dispensing In Europe and the USA, separation of dispensing and prescribing is a well- established system. However, in Japan hospitals and clinics relied on profits from drug price differences as a source of revenue. Recently, progress had been made in separating prescribing and dispensing (=iyaku bungyo). New data show that the rate had reached 44% in 2001.

Pharmaceutical profit levels (1) Since the early 1980s, the Japanese government has concentrated on reducing providers’ profits from dispensing drugs by the use of rigorous surveys of market prizes to justify sharp reductions in fee-schedule prizes. As a result, the average prize of drugs dropped substantially and the ratio of spending for drugs to total medical spending was reduced from 44% in 1980 to 29% in 1992 (Ikegami, 1995). However, the major pharmaceutical companies have responded by introducing more drugs that are only marginally “new” but for which higher prices can be charged. One result is the premature release of third-generation cephalosporins and new broad-spectrum fluoroquinolones.

Pharmaceutical profit levels (2) Moreover, medical institutions have been shifting to more expensive drugs, consequently increasing the unit price of pharmaceuticals. For example, in the 1990 and 1994 prizing reviews, there was a respective 17% and 13% reduction in the prize of antibiotics. However, during this period, possibly due to a shift towards newer, higher prized products, there was a 20% increase in the average unit price of these drugs (Ikegami, 1999). Thus, the reported reduction in drug profit margins may thus have been largely offset by hospitals switching to new broad-spectrum antibiotics with higher prices that offer greater scope for making profit (Ikegami, 1999).

Marketing strategies and public health policies Pharmaceutical promotion: Aggressive marketing of new broad- spectrum antibiotics with large sales forces and failure of policy makers to reduce physician-industry interactions: In Japan, there are an estimated 43,000 drug salesmen and almost as many wholesale representatives. An average Japanese doctor receives 450 sales calls a year. Public health policy: Lack of epidemiologic surveillance and initiatives to stop over-prescribing of antimicrobial agents, which may be explained by the common misconception among Japanese physicians and opinion leaders that AMR is not a threat to patients and public health in Japan.

DISCUSSION Evaluating policy-relevant determinants of antibiotic overuse in industrialized countries is a complex task. It involves numerous variables, including regulatory practices, marketing strategies and the patients’ own behavior and expectations. In previous cross-country comparisons looking at Germany, France and the United States, we have suggested that regulatory practices and socioeconomic characteristics of populations may have a major impact on antibiotic prescribing practices at a national level [Harbarth, 2001 & 2002].

CONCLUSIONS Antibiotic consumption in Japan is directly affected by regulatory practices. Financial incentives linked to the pharmaceutical reimbursement system and physician-industry interactions strongly influence antibiotic prescribing and increase AMR in Japan. This worrisome problem should be further addressed at the level of policy and education.