Screening and clinical detection of alcohol disorder in general practice

Abstract

Aim: This thesis reports on the implementation of key alcohol research findings to the screening and clinical detection of alcohol risk and disorder in a general practice. it aims to provide a literature review which will explain the nature of alcohol risk and disorder, why alcohol screening is necessary, and how this screening could be undertaken in general practice using a model of screening by healthcheck. Alcohol risk is presented as the likelihood of an individual to develop alcohol-related complications. Definitions of alcohol disorder are discussed and the reasons for adopting the DSM111R criteria explained. A study is reported in which healthcheck screening is used to improve detection of alcohol risk and disorder in the author’s general practice. Selected cases are presented to highlight particular issues relevant to alcohol in general practice. The study findings are discussed with reference to the literature review and recommendations on the use of the healthcheck Screening model in general practice presented. Method: All English language medical journals reporting alcohol-related research were identified by contents review (specialist alcohol and New Zealand journals), medline search (using keywords and related articles/author-specific searching) or cross.referencing from other reports. other alcohol-related professional reports were identified from library searches or obtained from original sources. From the literature review, a format was developed for the healthcheck questionnaire which included a brief alcohol screen. This brief questionnaire included three questions on consumption of alcohol, and two questions identifying concern about alcohol use, modified from screening questionnaires available at the start of the study, particularly the Canterbury Alcohotism Screening Test (CAST) and the Self-Administered Alcoholism Screening Test (SAAST). Adult patients registered with the practice were offered a personal invitation by the doctor to attend a healthcheck appointment. Recruitment was undertaken over a 3 year period for 30-69 year olds (754 eligible patients), and a 5 year period for 18-29 year olds (339 eligible patients). Patients attending the healthcheck completed a self-administered questionnaire which was reviewed by the practice nurse, selected examination items performed, and the results were discussed with the patient by the doctor. Alcohol risk was assessed and laboratory tests arranged if appropriate with follow-up appointment offered to those requiring further alcohol assessment. Patients were classified at increased risk by reported consumption, and diagnosed with alcohol disorder by DSM111R criteria (alcohol abuse or dependence). At the completion of each study period, the clinical notes of all patients in the study were reviewed to determine the screening rate and to review all recorded alcohol information including clinical indicators of alcohol disorder, alcohol risk for all screened patients, and the effect of screening on the prevalence of alcohol disorder. Results: Alcohol information was obtained on 851 of 1093 study patients (78%). Screening increased the detection of alcohol disorder from 4% (44 patients) to 8% (84 patients). clinical detection of alcohol disorder was associated with physical, psychological and social factors. Clinical factors most commonly included abnormal liver function tests, gastro-intestinal symptoms, hypertension, gout and mental health symptoms. Personal and relationship problems were also commonly linked to detection in a clinical context. Recruitment for healthchecks was highest among women (78%) with a high level of screening in both 18-29 year old (80%) and 30-69 year old groups(76%). Recruitment of men (70%) was notably more successful in the 30-69 year old group (85%) than in the 18-29 year old group (44%).As a test for DSM111R alcohol use disOrder, the brief alcohol questionnaire included in the healthcheck had a sensitivity of 0.93, a specificity of 0.85, a positive predictive value of 0.25, and a negative predictive value of 0.99. The questions on alcohol consumption were the most sensitive items, identifying increased risk (intermediate or high risk consumption) in 86% of patients with alcohol disorder. The remaining patients were identified by abnormal liver function tests. A positive screening test (consumption male>20, female >15 standard drinks/week) indicated a 1 in 4 likelihood of alcohol disorder, increasing to 1 in 2 if concern was reported on a screening question, or a liver function tests was abnormal. The healthcheck screening programme increased the detected prevalence of alcohol disorder from 4% to 8%. DSM111R criteria identified 32 patients with alcohol dependence and 52 patients with alcohol abuse. lcDl0 criteria identified 16 patients with alcohol dependence and 54 with harmful use. Nearly half of the patients meeting DSM criteria for disorder (48%) were not drinking at high risk levels (consumption male >50, female>35 standard drinks/week). Case examples drawn from the clinical notes indicated that alcohol information may come from multiple sources, and that classification of this information may be difficult, influenced not only by the completeness of the data available but also by medical interventions and other influences on patient alcohol use over time. Classification was also more difficult in some patients due to the complexity of the particular case. There were advantages identified in classifying disorder including enhanced therapeutic options, increased awareness of related issues and completeness of medical problem lists, although interventions in both risk and disorder were possible without formal diagnosis. Conclusion: there is good evidence in the literature supporting the value of general practice screening for alcohol risk and disorder. This type of screening meets accepted criteria. The importance of alcohol as a health problem is supported by both the prevalence of alcohol-related problems in New Zealand, and the relationship between alcohol and morbidity and mortality. lt is possible to recognise a latent or early symptomatic phase (risk or early abuse/dependence) and it proved possible in the study to collect this information on most patients using the healthcheck screen. There is evidence supporting the value of intervention in heavy drinking and it is reasonable to conclude that intervention in a pre-symptomatic or early symptomatic phase is of more value than waiting until more severe disorder is established. There are good screening tests available including consumption measures, patient questionnaires and laboratory tests. These are relatively inexpensive, generally easy to deliver, and provide sufficient information to determine which patients require further assessment of alcohol use. The use of the brief alcohol screening questionnaire as part of the healthcheck screen proved acceptable to patients in the study and was effective at detecting a high proportion of patients with alcohol disorder. The literature suggests that the AUDIT questions have advantages as screening questions given its international acceptability and its use in several general practice-based New Zealand studies. Although the AUDIT was not available at the time the practice study commenced, it could be incorporated in the health check questionnaire. Laboratory screening has an established supportive role, particularly the selected use of liver function tests. The role of other laboratory tests including carbohydrate-deficient transferring remains unclear and general practice research is required to determine if it has a place in screening and in which patient groups. positive screening tests require further assessment. As indicated in the case examples, this assessment may be undertaken gradually as patients attend the surgery. However, there is benefit if the assessment can be completed more fully. An alcohol assessment package for general practice would be useful. The involvement of practice nurses was essential to the successful delivery of healthcheck screening and this role could be further developed. The practice study shows it was possible to implement evidence-based alcohol screening in a general practice. The healthcheck screening model was a successful method of providing screening though recruitment of younger men could be improved. Uptake of healthcheck screening by other New Zealand general practices would require attention to the organisational factors shown to limit screening and preventive health care delivery.