Comparative evaluation of three continuous speech recognition software packages in the generation of medical reports.

OBJECTIVE: To compare out-of-box performance of three commercially available continuous speech recognition software packages: IBM ViaVoice 98 with General Medicine Vocabulary; Dragon Systems NaturallySpeaking Medical Suite, version 3.0; and L&H Voice Xpress for Medicine, General Medicine Edition, version 1.2. DESIGN: Twelve physicians completed minimal training with each software package and then dictated a medical progress note and discharge summary drawn from actual records. MEASUREMENTS: Errors in recognition of medical vocabulary, medical abbreviations, and general English vocabulary were compared across packages using a rigorous, standardized approach to scoring. RESULTS: The IBM software was found to have the lowest mean error rate for vocabulary recognition (7.0 to 9.1 percent) followed by the L&H software (13.4 to 15.1 percent) and then Dragon software (14.1 to 15.2 percent). The IBM software was found to perform better than both the Dragon and the L&H software in the recognition of general English vocabulary and medical abbreviations. CONCLUSION: This study is one of a few attempts at a robust evaluation of the performance of continuous speech recognition software. Results of this study suggest that with minimal training, the IBM software outperforms the other products in the domain of general medicine; however, results may vary with domain. Additional training is likely to improve the out-of-box performance of all three products. Although the IBM software was found to have the lowest overall error rate, successive generations of speech recognition software are likely to surpass the accuracy rates found in this investigation.

Understanding the relation between expectancies and drinking among DUI offenders using expectancy categories.

OBJECTIVE: In the context of several recent studies that have found negative expectancies to predict drinking-related variables, the present study was designed to examine the relationships among positive expectancies, negative expectancies and drinking, in a sample of DUI offenders. METHOD: Participants (N = 96; 75% male) recruited from a 3-day residential alcohol education program completed a demographic data sheet, the Negative Alcohol Expectancy Questionnaire (NAEQ), the Alcohol Expectancy Questionnaire (AEQ) and the Customary Drinking Record upon admission to treatment. At a 3-month follow-up assessment, these measures were re-administered to 49% of the sample. RESULTS: Hierarchical multiple regression revealed that both positive and negative expectancies were related to number of drinks in the previous 30 days. Stepwise multiple regression revealed that baseline drinking and negative expectancies were both related to number of drinking days at the 3-month follow-up assessment. To assess the interplay of positive and negative expectancies as they relate to drinking, participants were categorized as "low" or "high" on both the AEQ and the NAEQ. The low positive/low negative group drank more than the low positive/high negative group. There was no statistically significant difference in drinking between the high positive/low negative and the high positive/high negative group. CONCLUSIONS: This study provides additional evidence that negative alcohol expectancies are related to drinking. These findings suggest that the proposed restraining influence of negative expectancies on drinking may be reduced when a drinker has concurrent positive expectancies which are high.

Acceptance of moderate drinking by alcoholism treatment services in Canada.

OBJECTIVE: The present study was conducted to provide a nationwide survey of acceptance of nonabstinence goals and related alcoholism treatment practices by Canadian alcoholism treatment services. METHOD: A random sample of 335 Canadian alcoholism treatment service agencies were mailed a 4-page questionnaire designed to assess acceptance of moderate drinking as a drinking goal and related alcoholism treatment practices. RESULTS: Acceptance varied by type of service, with considerably more acceptance by outpatient programs (62%) and mixed inpatient/outpatient programs (43%) than inpatient/detoxification/ correctional facilities (27%) and halfway houses (16%). Two-thirds of the respondents who reported moderate drinking as unacceptable in their own agencies categorically rejected moderation for all alcoholism clients. CONCLUSIONS: Individuals seeking services in Canadian alcoholism treatment agencies are more likely to have a choice of drinking goals if they present to an outpatient program than a residential institution, and Canadian agencies appear more accepting of moderation goals than American programs, but less accepting than British and Norwegian service agencies.