Comparison of 7‐day retrospective and prospective alcohol consumption diaries in a female population in Perth, Western Australia—methodological issues

The summer alcohol consumption of women living in metropolitan Perth was measured using both a retrospective and prospective diary. Overall, 1356 women fully completed both the retrospective and prospective alcohol consumption diaries. The prospective diary yielded a significantly higher level (9%) of consumption (p less than 0.00001) compared with the retrospective diary. Although there was no significant difference in the percentage of women drinking at harmful levels (greater than 280 g absolute alcohol per week) between the two diary methods, there did appear to be an important percentage difference (1.9% versus 2.8%) suggesting that the prospective diary did detect a larger percentage of women at risk. This would be both practically and statistically significant in large populations. The heavier the drinker the greater the variability in reported consumption between the two diary weeks. However, it is interesting to note that the direction of the variation depended on which diary was used as the base for classifying women as different types of drinkers. Women who drank at hazardous (140-280 g per week) or harmful (greater than 280 g) levels during the retrospective diary week reported a lower level of consumption during the prospective diary week. Conversely women who drank at hazardous or harmful levels during the prospective diary week reported a lower level of consumption during the retrospective diary week. Possible explanations for these findings are explored and methodological considerations for future studies are discussed.