Alcohol Policy Opinions in the United States over a 15-Year Period of Dynamic per Capita Consumption Changes: Implications for Today's Public Health Practice

Article excerpt

We examine 1989-2005 trends in United States (US) public opinion using 11 alcohol policy items (seven national telephone surveys) during a dynamic drop in per capita intake until the mid-1990s, followed by the first sustained upturn seen since 1980. In the 2000 U.S. National Alcohol Survey (NAS) survey 14 available policy items were factor analyzed, forming four factor-based scales replicated in 2005. Linear regression examined changing demographic correlates of support for the four policy areas between 2000 and 2005. Several empirically effective policy levers have weak support, which continues to erode. Between 2000 and 2005 support weakened by an average of 1% for all except two policies, both of which declined since 1989. Even support for alcohol warning labels has turned down for the first time since 1989. Implications of the declining endorsement of alcohol policies are discussed, given the upturn in consumption. Community-based strategies are needed to inform the public of benefits of alcohol policies.

KEY WORDS: Alcohol policy, public opinion, national surveys, alcohol control, treatment access.

In the United States, public opinion on alcohol policies has received only intermittent attention since the 1990s (Greenfield, Johnson & Giesbrecht 2004b; Wagenaar & Streff 1990). The need for more research on the role of public opinion in alcohol policy development has been identified (Greenfield 1994) because public support for a policy may make its enactment more politically viable and help sustain a policy once enacted (Room, Graves, Giesbrecht & Greenfield 1995). Although they are not likely to determine which policies will be established, public opinions may help legitimate political courses of action (Leedham 1987). Additionally, gaps are quite commonly seen between evidence-based alcohol policies (Babor, Caetano, Casswell, Edwards, Giesbrecht, Graham, et al. 2003) and public opinions regarding the same policies (Giesbrecht & Greenfield 2003). When this gap is large it may indicate the need for further educational efforts designed to improve people's understanding of policy rationales (Greenfield, Johnson & Giesbrecht 2004b).

The present analyses involve 14 survey items, of which 11 measures of alcohol-related policy opinions came from the 1989 Canadian National Alcohol and Other Drugs Survey (NADS) (Eliany, Giesbrecht, Nelson, Wellman & Wortley 1992) and were subsequently used in US national and Canadian provincial and national surveys. For example, these items were included in a series of US surveys from 1989 through 1994, the Impact of Alcoholic Beverage Warning Labels (WL) surveys. These cross-sectional national surveys were conducted by the Alcohol Research Group (ARG) in California. In this article we use the 1989 to 1994 WL surveys and two later surveys conducted by the ARG Center, the National Alcohol Surveys (NASs) conducted in 2000 and 2005. Earlier reports have concerned aspects of the WL policy-opinion data series (Greenfield 1997a; Greenfield, Johnson and Giesbrecht 2004b; Hilton and Kaskutas 1991; Kaskutas 1993; Room, Graves et al. 1995) and the 2000 NAS (Greenfield, Ye & Giesbrecht, 2007). Here we extend the series by adding data from the 2005 NAS, extending the series to 15 years. All included surveys (WL and NAS) were conducted by telephone. The latest two NAS surveys included three additional items, one mandating warnings on alcohol advertisements and two about policies designed to enhance alcohol treatment access: (1) Requiring that health insurance policies cover alcohol treatment and (2) Providing free alcohol treatment.

Given our focus on US policy opinions, for brevity we will not review policy-opinion research in other countries, but these have included Canada (Giesbrecht, Ialomiteanu, Room & Anglin 2001, Giesbrecht & Greenfield 1999), New Zealand (Casswell, Gilmore, Maguire & Ransom 1989), European and Scandinavian countries (Ahlström & Österberg 1992; Moskalewicz & Tigerstedt 1998) and Puerto Rico (Harwood, Bernât, Lenk, Vazquez & Wagenaar 2004). …