Alcohol and Alcoholism Advance Access originally published online on August 10, 2006
Alcohol and Alcoholism 2006 41(6):643-649; doi:10.1093/alcalc/agl059
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
THE EFFECTIVENESS OF A TAILORED INTERVENTION FOR EXCESSIVE ALCOHOL CONSUMPTION PRIOR TO ELECTIVE SURGERY
1 School of Public Health, 2 Psychological Medicine and 3 Discipline of Medicine, University of Sydney, NSW 2006, Australia and 4 Drug Health Services, Royal Prince Alfred Hospital, Camperdown NSW 2050, Australia
* Author to whom correspondence should be addressed at: Drug Health Services, Level 5, Page Building, Royal Prince Alfred Hospital, Missenden Road, Camperdown, NSW 2050, Australia; Tel.: +61 2 9515 6419; Fax: +61 2 9515 8970; E-mail: phaber{at}mail.usyd.edu.au
(Received 21 December 2005; first review notified 10 April 2006; in revised form 6 June 2006; accepted 6 July 2006)
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Aim: To assess the effectiveness of a tailored pre-operative intervention for excessive alcohol consumption in reducing post-operative complications and alcohol consumption thereafter. Methods: Patients scheduled for elective surgery requiring at least overnight hospitalisation were screened for alcohol misuse. Consenting, eligible participants with
7 days to surgery at the time of screening were offered an intervention and those with <7 days to surgery were provided usual care. Results: Over a period of 2 years and 10 months, 3139 patients were screened to recruit 136 participants. Baseline analysis revealed a mean age of 53 (±15.8) years and a mean consumption of 71 g/day (±48.1). The intervention group (n = 45) did not differ significantly from controls (n = 91) in age, consumption, and number of current smokers, but there were significantly more women in the control group. There was no difference between the groups in major or minor complications experienced, or length of stay after controlling for age, gender, and baseline consumption. At 6-month follow-up there was a significant reduction in drinking for the entire study population. Conclusion: The study did not demonstrate any beneficial effect of the pre-operative intervention on post-operative complications. The relatively short time to surgery, intervention by a non-member of the surgical team, challenges to recruitment and reduced consumption in the control group may have limited the ability of the study to detect a significant effect of the intervention. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Similar to tobacco use (Peters et al., 2004
), chronic excessive alcohol consumption increases the risk of peri-operative complications among surgical patients (Tonnesen and Kehlet, 1999). These complications range from delayed wound healing and site bleeding to repeat surgery and alcohol withdrawal reactions, and contribute to an increased need for nursing care and prolonged hospitalisation (Tonnesen and Kehlet, 1999; Shourie et al., 2004). In addition to this distress, the cost per successful surgical procedure among those drinking excessively has been shown to be three times greater than that of matched controls (Little, 1980).
Abstinence from alcohol is associated with reversal of some of the adverse effects of chronic excessive alcohol consumption on organ systems including the immune, cardiovascular, gastrointestinal, and central nervous systems. The time required for reversal varies (Shourie et al., 2004). Platelet function and bleeding times can recover in a 7–14-day period (Rubin, 1999), cellular immunity can show an improvement in 2 weeks but may take up to 2 months to normalise (Tonnesen et al., 1992a), lymphocytes recover over 3–6 months (Tonnesen et al., 1990) but the negative effect on bone turnover and bone mass can take up to 2 years to resolve (Lindholm et al., 1991; Peris et al., 1994).
Efforts to minimise alcohol consumption prior to elective surgery might diminish complications related to alcohol misuse. Yet despite numerous interventions to reduce excessive alcohol consumption in other settings, there has been only a modest effort to address alcohol problems among pre-operative patients. One randomised controlled trial found that use of disulfiram to ensure complete pre-operative abstinence was associated with improved post-operative outcomes amongst men consuming 60 g of alcohol/day. In this study, the intervention group developed significantly fewer post-operative complications (31 versus 74%) than the control drinkers receiving usual care (Tonnesen et al., 1999). But such pharmacotherapy requires a high degree of compliance and may not be acceptable, applicable or safe or effective in all cases (Martin et al., 2004). A recent meta-analysis has confirmed that in the absence of dependence, brief intervention is effective in reducing excessive alcohol consumption (Bertholet et al., 2005), and is cost-effective (Wutzke et al., 2001), and for those with alcohol dependence more intensive treatments have been shown to be beneficial (Garbutt et al., 1999). People drinking at lower levels (i.e. men drinking 40–60 g daily and women drinking 20–40 g daily) could also benefit from a pre-operative intervention for alcohol consumption, as they too suffer poorer clinical outcomes following surgery (Felding et al., 1992; Tonnesen et al., 1992b).
The aim of this study was to test the effectiveness of an intervention to reduce alcohol consumption in preparation for elective surgery. We hypothesised that this intervention would have a beneficial effect measurable in terms of decreased number of post-operative complications, length of in-hospital stay, and a longer-term reduction in consumption.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
This controlled trial was conducted at two tertiary referral hospitals, affiliated with The University of Sydney, The Royal Prince Alfred Hospital (RPAH) and Concord General Repatriation Hospital (CRGH), from November 2000 to September 2003. Ethical approval was obtained from the Sydney South West Area Health Service (SSWAHS) Human Research Ethics Committees (RPAH and CRGH zones).
Recruitment was undertaken in two steps. The initial step was a screening of all patients aged 18 years scheduled for elective surgery. Following verbal consent, all pre-operative patients completed a brief self-administered questionnaire about their alcohol consumption in the pre-admission clinic (PAC) of the two hospitals (Fig. 1). This questionnaire included demographic information and questions on date of scheduled surgery, and the quantity and frequency of tobacco use. The AUDIT-C (Bush et al., 1998) comprising the first three questions of the AUDIT (Babor et al., 1998) was used to assess alcohol consumption. Patients who were too ill to fill out the screening questionnaire, with insufficient English language skills and with physical or mental disabilities were excluded from screening.
|
Following initial screening at the PACs, patients with an AUDIT-C score
5 and whose surgeons had indicated that their surgery would involve at least an overnight hospital stay were offered further participation. At a cut-off of 5, the AUDIT-C has a sensitivity of 74% and a specificity of 83% for detecting at-risk drinking and a sensitivity of 88% and a specificity of 81% for detecting current alcohol dependence as per DSM III-R (Rumpf et al., 2002). The overnight stay was required because day-only surgery is associated with very few adverse events such that inclusion of these patients would not allow us to test the hypothesis that the intervention to reduce alcohol consumption reduces post-operative complications. At this stage, an information sheet was provided describing the purpose of the trial and the role of each participant. After obtaining written informed consent, a baseline interview was scheduled. Alcohol consumption was quantified using the World Health Organization ‘Tri-level’ method (Bradley et al., 1998), which measures average consumption on high, medium and low level drinking-days. Men drinking 60 g daily and women drinking 40 g daily according to this measure were recruited into the trial. The Composite International Diagnostic Interview (CIDI) (Witttchen, 1994) was used to assess the presence of current harmful alcohol use or dependence (i.e. within the last 12 months) using DSM-IV and ICD-10 criteria. The baseline questionnaire also included questions on tobacco use, general health, current medications, and medical conditions such as high blood pressure, diabetes, hepatitis, and the SF 36 as a physical health measure.
Initially it had been expected that patients would be randomly allocated to intervention or control groups, but because of difficulty in recruiting participants with at least 7 days to surgery (a minimum for any likely benefit from reduction in drinking according to previous studies), it was necessary to allocate to groups by interval between screening and surgery. Participants with 7 days to surgery at the time of screening in the pre-operative clinic were assigned to the intervention group and those with <7 days to surgery were assigned to the control (usual care) group.
Study groups
Intervention group
In addition to usual care as offered by the treating hospital, this group received a tailored intervention. The intensity of the intervention was based on the presence or absence of alcohol dependence.
Non-dependent drinkers
Participants who were not alcohol dependent received brief intervention using the World Health Organization-based ‘Drink-less’ package (Gomel et al., 1994) as recently revised (The Drinkless Program 2004, Proude et al., 2006). The goal of this intervention was to reduce alcohol consumption to within Australian drinking guidelines (<40 g/day for men and <20 g/day for women).
Dependent drinkers Participants who were alcohol dependent according to the CIDI were given the above brief intervention and warned of the possibility of alcohol withdrawal. They were offered an opportunity to consult a drug and alcohol specialist at RPAH, where the potential need for management of withdrawal symptoms would be assessed and pharmacotherapy and other treatment provided as per individual requirement.
Control (usual care) group This group received usual care as offered by the participating hospitals. This included a description of the process of admission, preparation for, and recovery from surgery and written material as routinely provided by each surgeon and/or hospital.
Outcome measures
Outcomes were measured at two time-points. The first time-point was the end of the hospital stay and the second was a 6-month follow-up interview. Towards the end of the hospital stay, post-operative complications were retrospectively recorded for the 5 days following surgery, along with benzodiazepine requirement and length of hospital stay from the hospital database. For the purpose of this study, a major complication was defined as one which debilitated the patient and required specific and significant treatment, and a minor complication as one which did not alter overall function significantly. Examples of major post-operative complications included local bleeding/anastomotic leakage requiring repeat surgery, sepsis (bacteraemia and fever), anaemia requiring transfusion, and conditions such as cardiopulmonary insufficiency requiring intensive care. Examples of minor post-operative complications included wound infections, wound leakage, and delayed healing. In case of doubt, a physician's opinion (one of the authors, P.H.) was obtained in deciding whether a complication was major or minor. In these cases, the physician made this decision without knowledge of the group to which the case was assigned. At the 6-month follow-up, AUDIT-C scores, average daily alcohol consumption (assessed by the Tri-level method), the number of visits to the general practitioner (GP), and ability to undertake physical activity were assessed (using the SF-36).
Power calculations
Based on the recruitment of 45 cases and 91 controls, the final study design had a power of 90% to detect a difference in complication rate of 30 percentage points (70% compared with 40%) at 0.05 level of significance (two-sided) between the control and intervention population. Tonnesen et al. (1999) found that post-operative complications fell from 74 to 31%, a difference of 43 percent points, so the present study was larger than the only other reported study of its type. These calculations were completed using ‘Power – version 1.21’ (POWER – Epicenter Software, 1985).
Statistical methods
Data were analysed using the statistical package SPSS for Windows (Release 10.0, 2000). After performing data checks for outliers, histograms for examining data distribution and scatter plots for associations; chi-squared tests were used to test significance of associations. Regression analysis was used to examine the effect of the independent variables, i.e. intervention status and baseline alcohol consumption on each of the dependent variables, including post-operative complications, alcohol consumption at 6-month follow-up, in-hospital stay, and general health of the participant. Continuous variables with skewed distribution underwent natural log transformation. In particular, natural log transformed values of average daily alcohol consumption were used in all analyses. Logistic regression was used for dependent variables that were dichotomous e.g. presence or absence of complications.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Demographic characteristics of screened elective surgery patients
Of the total 3783 patients approached for initial screening over a period of 2 years and 10 months, 3139 (83%) agreed to fill out the screening questionnaire (Fig. 1). Of the 541 excluded from initial screening, 68.4% had insufficient English language skills, 8.3% had major vision or hearing disability, 8.1% were too frail, 4.4% had disabling mental illness, 4.2% were <18 years old and 6.5% had returned for a repeat surgery and therefore had been previously screened. One hundred and three (2.7%) patients refused to fill the screening questionnaire.
The screened patients had a mean age 60.4 (±17.0) years, with a range of 19–100 years and 51.5% were men. A little over half of the screened patients (59%) were either married or in a de facto relationship, 
15% were single or never married, 12% were separated or divorced, 11% were widowed, and 3% did not answer this question. Overall 77% of the screened patients had a school certificate (completed 10 years of schooling), 43% were retired and 29% in paid work.
Of the 3139 patients screened, 617 had an AUDIT-C 5 (i.e. positive AUDIT-C score) and 1329 scored >3. The 617 with an AUDIT-C 5 were assessed further using the WHO Tri-level method. After excluding day-surgery patients and those who refused further participation (n = 114), a total of 136 (4.3% of those initially screened) were recruited to the trial (Fig. 1).
Demographic characteristics of study participants at baseline
There were no significant differences between control and intervention groups with regard to age, AUDIT-C scores, average daily consumption at baseline, prevalence of current DSM-IV dependence, tobacco use and cancer; or in average number of visits to the general practitioner and hospital admissions in the last 6 months (Table 1). There were more patients undergoing major surgery in the intervention group, but this difference was not significant. The proportion of women and the proportion of participants recruited from RPAH centre were both significantly higher in the control group, compared with the intervention group (Table 1). Subsequent analyses were adjusted for these differences. Of all participants, 78% were born in Australia, with no difference between the groups (intervention versus control group, 80 versus 77%; P = 0.7).
|
Outcome measures assessed in the 5 days following surgery
Significantly more major post-operative complication(s) (t = 7.16; P = 0.007) occurred in the intervention group than in the control group (Table 2). This difference disappeared after adjusting for baseline differences in gender using logistic regression. More patients in the control group experienced minor complications and a greater number required repeat surgery, but the differences were not statistically significant. There were no differences between the two groups with respect to post-operative administration of benzodiazepines (Table 2).
|
Age was the only factor significantly related to the presence or absence of post-operative complications after controlling for gender, major surgery and smoking status (t = 2.060; P = 0.026). Baseline alcohol consumption was not a predictor of post-operative complications in this study population, but this study was restricted to subjects who drank excessively. A logistic regression found no effect of smoking status on the presence or absence of complications after adjusting for confounders (P = 0.9). There were proportionately more male patients undergoing major surgery in the intervention group compared with the control group (65.9 versus 54.4%; P = 0.240), although the difference was not significant. Amongst male participants, after controlling for age, baseline alcohol consumption and smoking status, major surgery was correlated with the sum of post-operative complications (t = –2.6; P = 0.01), but there was no difference between the intervention and control groups in the number of post-operative complications.
For the entire sample, provision of intervention did not influence length of hospital stay (t = 0.354; P = 0.724), even after controlling for age, gender, baseline consumption, smoking status, and major surgery. As expected, the length of in-hospital stay was highly correlated with major surgery (t = –3.7; P < 0.0001) and the presence of post-operative complications (t = 5.1; P < 0.0001) after controlling for gender, baseline age, and smoking status. For the entire study population, age (P = 0.02) and a diagnosis of cancer (P = 0.045) were significant predictors of length of hospital stay, after controlling for gender, baseline consumption, and smoking status.
Outcome measures assessed at the 6-month follow-up
Average daily consumption fell in the entire study population from 70 g/day at baseline to 26 g/day at follow-up, or 63% (natural log transformed, t = 25.6; P < 0.0001). In the control group, the average daily consumption fell from 73 to 24 g/day and in the intervention group, from 67 to 29 g/day. No difference was found between the average daily consumption (using multiple backward regression model) at follow-up between intervention and control groups (t = 2.8; P = 0.9), even after controlling for gender, age, and average daily consumption at baseline (Table 3). The intervention group had fewer days off sick as compared with the control group following surgery but this difference was not statistically significant (4 versus 6 days; P = 0.39). The intervention group visited their general practitioner more often in the last six months compared with the control group (5.3 versus 4.5 times; P = 0.32) but again this difference was not statistically significant. The ability to do physical activities (measured using the physical component of SF 36) was similar in both groups (P = 0.84).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
This trial was conducted to determine whether a pre-operative intervention to reduce excessive alcohol consumption, tailored to the severity of the alcohol use disorder, could favourably influence post-operative outcomes and longer-term alcohol consumption. Over three thousand patients were screened to recruit approximately hundred patients to the trial and these challenges in recruitment limited the size of the enrolled sample. The response to the acceptance of the tailored pre-operative intervention was varied, with all participants accepting brief intervention but none of the dependent participants accepting our offer of a medical drug and alcohol consultation. We did not find evidence to suggest that the intervention reduced the short-term clinical outcomes of elective surgery or longer-term measures of alcohol consumption. These results suggest that other approaches should be explored to lessen the adverse impact of alcohol use disorders on the outcomes of elective surgery.
The trial found a large reduction in the average daily alcohol consumption at the 6-month follow-up time-point regardless of whether an intervention had been provided or not. This reduction could be an indirect effect of the surgery itself, or the underlying illness, where poor health following surgery does not allow the patient to drink so heavily. Alternatively patients may be more conscious of their health and choose to cut down, or be in closer contact with medical practitioners who offer advice about alcohol. It is also possible that participation in detailed assessment of alcohol consumption made the participants aware of their levels of drinking. One could infer from this study that there is less need for an alcohol intervention pre-operatively to reduce long-term alcohol consumption after surgery, but further studies would have to be undertaken to offer such advice conclusively.
In comparing our results with the only other pre-operative alcohol intervention study (Tonnesen et al., 1999), neither trial found any difference between the intervention and control groups in the number of major complications experienced, length of in-hospital stay or mortality. Tonnesen et al. (1999) found significant improvement in the physiological functioning among participants of the intervention group, such as arrhythmias and hypoxic events as detected by a 24 h monitor. Similar to the present trial, they found no effect of the improved physiological functioning on clinical outcome measures such as number of major or minor complications, length of hospital stay, or mortality. However, on combining the number of patients with major and minor complications (yielding the number of patients with any complications), Tonnesen et al. (1999) (unlike our trial) found a large and statistically significant difference between the intervention and control groups (5/16 versus 14/19, P = 0.02).
The Danish randomised controlled trial employed a more intensive pre-operative intervention using disulfiram to ensure abstinence among its participants (Tonnesen et al., 1999), however, such a method often fails due to poor compliance (Nielsen et al., 2000). Non-dependent excessive alcohol users also suffer from poorer outcomes following surgery (Tonnesen et al., 1992b). Brief intervention, which is widely accepted, has been shown to be effective in a range of clinical and non-clinical settings among non-dependent drinkers (Bertholet et al., 2005; Bien et al., 1993) and is also cost-effective in terms of reduction in consumption (Wutzke et al., 2001). For these reasons, we adopted a tailored intervention for this trial, (i.e. brief intervention with or without referral to the drug and alcohol specialist for detoxification and pharmacotherapy depending upon the need of the patient). However, not one of the small number of alcohol-dependent people booked for elective surgery in this study (n = 17; Table 1) accepted the offer of consultation with a drug and alcohol specialist. One patient was referred directly to the drug and alcohol specialist by his surgeon. It was therefore difficult to test the full potential of our proposed ‘tailored’ intervention or to address the benefits of intervention specifically amongst an alcohol-dependent group booked for surgery. Brief intervention is more firmly established for non-dependent alcohol misuse and may not be effective in dependent drinkers. When dependent drinkers were excluded from the analysis, the results were unchanged (not shown). One clinical implication of this study is that intervention may be more likely to succeed when offered by a member of the surgical team, and integrated with surgical management.
One hypothesis underlying this study was that the time before elective surgery would be a ‘teachable moment’ when patients would be concerned about their health and more receptive to lifestyle advice. Several explanations may be offered to explain why brief intervention was ineffective in this context. It may be that brief intervention in this setting had lower salience due to impending serious surgery. Anecdotally, patients were focussed on preparation for surgery and often did not accept the information that their drinking might play an important role in surgical recovery. Further, there may have been inconsistent messages if the surgical team did not raise this issue, whereas the research team recommended significant lifestyle change to improve surgical outcome. As suggested above, stronger engagement of surgical team in providing pre-operative intervention for alcohol misuse may be required. Finally, the study population was relatively older, less well-educated, and relatively socially disadvantaged when compared with the subjects involved in brief intervention studies conducted in primary care settings. These demographic characteristics may have influenced the effectiveness of brief intervention.
In this intervention group, the mean interval to surgery was 10 (±10.7) days, reflecting the short interval between attendance at the pre-operative clinic and admission to hospital. The modal time interval was 1 day. This may have been insufficient period of reduced drinking to result in physiological improvements. In comparison, the intervention group in the Tonnesen et al. (1999) study had at least 30 days of abstinence prior to surgery. It may be that a longer period of reduced consumption is required to allow the physiological improvements to express themselves into clinically significant outcomes. It is also possible that total abstinence is required to achieve health gains, regardless of the degree of misuse of alcohol. Future studies have to be undertaken to assess if total cessation is required and to determine the optimal pre-operative duration of reduced drinking to achieve clinically significant benefits.
Although this trial was larger than the only other published pre-operative alcohol intervention study (Tonnesen et al., 1999), it still had only had adequate statistical power to detect a difference of 30 percentage points in complication rate between intervention and control groups. However, in view of the lack of any apparent trend towards a benefit for the intervention group, it did not seem reasonable to extend this study. Recruitment of participants for this study was a difficult task. It took nearly 3 years to recruit 136 participants. This may have been partly because relatively few people with problem drinking were scheduled for elective surgery. The relatively high mean age (53 years) of surgical patients could have contributed to low prevalence of hazardous alcohol use, as alcohol consumption in Australia falls substantially with age (Teesson et al., 2000). Under-reporting of alcohol consumption by the patients could have also added to the difficulty of recruiting participants. The use of AUDIT-C as the alcohol screening measure and the cut-off of 5 was selected as the most brief and accurate available instrument but the sensitivity and specificity are not entirely ideal. At the cut-off of 5, the reported sensitivity and specificity are 74 and 83%, respectively (Rumpf et al., 2002). Use of a lower cut-off, such as 3, would improve sensitivity to 99%, somewhat improving recruitment, but with substantially lower specificity of 42% (Rumpf et al., 2002). This would have almost doubled the number of baseline interviews required, thus increasing the labour and cost of this study. Further, recent changes in health care trends with increasing number of patients scheduled for day-surgery without an overnight hospital stay meant a number were excluded from the trial, further compromising recruitment. Tonnesen et al. (1999) appeared to find similar difficulties in recruitment, with the study taking two and half years to recruit 35 participants. This was in spite of the surgical team taking an active role in recruiting participants to the Danish trial.
It is possible that the negative result might also have been in part due to unintended baseline differences between intervention and control groups, with a significantly greater number of women recruited to the control group and a greater number of individuals undergoing major surgery recruited to the intervention group. However, reanalysis of the data after adjustment for these variables did not influence the study result. In addition, there was no trend to improved outcome in the intervention group. A difference between the intervention and control groups might also have been obscured by the heterogeneity of the study sample, having patients from many different surgical specialties. It is likely that if the study group had been more homogenous, as in the Tonnesen et al. studies (Tonnesen et al., 1991, 1999) differences between the groups might have been more easily detected. However, the purpose of this study was to evaluate a tailored intervention among patients scheduled for elective surgery regardless of specialty and not among any one particular specialty.
Although the study did not detect any beneficial effect of this pre-operative intervention, it does not mean that pre-operative intervention lacks value. Pre-operative interventions for excessive alcohol consumption should be further explored in an effort to reduce the adverse effects of alcohol after surgery, but conduct of such studies appears to be more difficult to achieve than previously suspected. Until pre-operative intervention strategies are more fully evaluated, it would be advisable to systematically screen surgical patients prior to and on admission to hospital and offer appropriate treatment (Spies and Rommelspacher, 1999). Ultimately, effective pre-operative intervention should be linked to in-patient monitoring and care and appropriate follow-up with the general practitioner (or family doctor), following discharge.
|
ACKNOWLEDGEMENTS |
|---|
We would like to thank Kate Wilson and Mamta Porwal for assisting in screening potential participants for this study. This project was supported by a project grant from Department of Veterans' Affairs (DVA) and a post-graduate scholarship (Swati Shourie) from the National Health and Medical Research Council (NHMRC).
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Babor, T. F., de la Fuente, J. R., Saunders, J. et al. (1998) AUDIT—the Alcohol Use Disorders Identification Test: guidelines for use in primary health care. WHO/MNH/DAT/89.4. World Health Organization, Division of Mental Health, Geneva.
Bertholet, N., Daeppen, J.-B., Wietlisbach, V. et al. (2005) Reduction of alcohol consumption by brief alcohol intervention in primary care: systematic review and meta-analysis. Archives of Internal Medicine 165, 986–995.
Bien, T. H., Miller, W. R. and Tonigan, J. S. (1993) Brief interventions for alcohol problems: a review. Addiction 88, 315–336.[CrossRef][ISI][Medline]
Bradley, K. A., Bush, K. R., McDonell, M. B. et al. (1998) Screening for problem drinking: Comparison of CAGE and AUDIT. Journal of General Internal Medicine 13, 379–388.[CrossRef][ISI][Medline]
Bush, K., Kivlahan, D. R., McDonell, M. B., et al. (1998) The AUDIT alcohol consumption questions (AUDIT-C): an effective brief screening test for problem drinking. Archives of Internal Medicine 158, 1789–1795.
Felding, C., Jensen, L. M. and Tonnesen, H. (1992) Influence of alcohol intake on postoperative morbidity after hysterectomy. American Journal of Obstetrics and Gynecology 166, 667–670.[ISI][Medline]
Garbutt, J. C., West, S. L., Carey, T. S. et al. (1999) Pharmacological treatment of alcohol dependence: a review of the evidence. Journal of the American Medical Association 281, 1318–1325.
Gomel, M., Saunders, J. B., Burns, L. et al. (1994) Dissemination of early intervention for harmful alcohol consumption in general practice. Health Promotion Journal of Australia 4, 65–69.
Lindholm, J., Steiniche, T., Rasmussen, E. et al. (1991) Bone disorder in men with chronic alcoholism: a reversible disease? Journal of Clinical Endocrinology and Metabolism 73, 118–124.[Abstract]
Little, J. M. (1980) The Henry Joseph Windsor Lecture: surgery for the alcoholic—no easy decision. Australian and New Zealand Journal of Surgery 50, 563–570.[ISI][Medline]
Martin, B. K., Clapp, L., Alfers, J. et al. (2004) Adherence to court ordered disulfiram at fifteen months: a naturalistic study. Journal of Substance Abuse Treatment 26, 233–236.[CrossRef][ISI][Medline]
Nielsen, B., Nielsen, A. S. and Wraae, O. (2000) Factors associated with compliance of alcoholics in outpatient treatment. Journal of Nervous and Mental Disease 188, 101–107.[CrossRef][ISI][Medline]
Peris, P., Pares, A., Guanabens, N., et al. (1994) Bone mass improves in alcoholics after 2 years of abstinence. Journal of Bone and Mineral Research 9, 1607–1612.[ISI][Medline]
Peters, M. J., Morgan, L. C. and Gluch, L. (2004) Smoking cessation and elective surgery: the cleanest cut. Medical Journal of Australia 180, 317–318.[ISI][Medline]
Proude, E. M., Conigrave, K. M. and Haber, P. S. (2006) Effectiveness of skills-based training using the Drink-less package to increase family practitioner confidence in intervening for alcohol use disorders. BMC Medical Education 6, 8.
Rubin, R. (1999) Effect of ethanol on platelet function. Alcoholism: Clinical and Experimental Research 23, 1114–1118.[CrossRef][ISI][Medline]
Rumpf, H-J., Hapke, J., Meyer, C. et al. (2002) Screening for alcohol use disorders and at-risk drinking in the general population: psychometric performance of three questionnaires. Alcohol and Alcoholism 37, 261–268.
Shourie, S., Haber, P. S., Boughton, R. and Conigrave, K. M. (2004) Clinical implications of alcohol misuse for surgery. In: Comprehensive Handbook of Alcohol Related Pathology, Vols 1–3, Preedy V. R. and Watson, R. R., eds, pp. 2192–3003. London, UK: Elsevier Academic Press.
Spies, C. D. and Rommelspacher, H. (1999) Alcohol withdrawal in the surgical patients: prevention and treatment. Anesthesia and Analgesia 88, 946–954.
Teesson, M., Hall, W., Lynskey, M. et al. (2000) Alcohol- and drug-use disorders in Australia: implications of the National Survey of Mental Health and Wellbeing. Australian and New Zealand Journal of Psychiatry 34, 206–213.[CrossRef][ISI][Medline]
The Drink-less program. (2004) Details available from the program's website. at http://www.cs.nsw.gov.au/drugahol/drinkless/ (accessed on 4.7.06).
Tonnesen, H. and Kehlet, H. (1999) Preoperative alcoholism and postoperative morbidity. British Journal of Surgery 86, 869–874.[CrossRef][ISI][Medline]
Tonnesen, H., Andersen, J. R., Pedersen, A. E. et al. (1990) Lymphopenia in heavy drinkers—reversibility and relation to the duration of drinking episodes. Annals of Medicine 22, 229–231.[ISI][Medline]
Tonnesen, H., Pedersen, A., Jensen, M. R., et al. (1991) Ankle fracture and alcoholism. Journal of Bone and Joint Surgery 73, 511–513.
Tonnesen, H., Kaiser, A. H., Nielsen, B. B. et al. (1992a) Reversibility of alcohol-induced immune depression. British Journal of Addiction 87, 1025–1028.[CrossRef][ISI][Medline]
Tonnesen, H., Petersen, K. R., Hojgaard, L. et al. (1992b) Postoperative morbidity among symptom-free alcohol misusers. Lancet 340, 334–337.[CrossRef][ISI][Medline]
Tonnesen, H., Rosenberg, J., Nielsen, H. J., et al. (1999) Effect of preoperative abstinence on poor postoperative outcome in alcohol misusers: randomised controlled trial. British Journal of Medicine 318, 1311–1316.
Wittchen, H. U. (1994) Reliability and validity studies of the WHO-Composite International Diagnostic Interview (CIDI): a critical review. Journal of Psychiatric Research 28, 57–84.[CrossRef][ISI][Medline]
Wutzke, S. E., Shiell, A., Gomel, M. and Conigrave, K. M. (2001) Cost effectiveness of brief interventions for reducing alcohol consumption. Social Science in Medicine 52, 863–870.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||