Agreement between HADS classifications and single-item screening questions for anxiety and depression: a cross-sectional survey of cancer patients.

BACKGROUND: We assessed agreement between reported anxiety and depression levels of cancer patients using (i) single self-report items and (ii) the Hospital Anxiety and Depression Scale (HADS). We also explored whether anxiety and depression assessment by (i) single self-report items or (ii) the HADS was most strongly associated with a preference to be offered professional assistance. The proportion of patients indicating that they would accept (or were currently using) professional support if they were experiencing anxiety or depression was also examined. PATIENTS AND METHODS: A consecutive sample of cancer patients undergoing radiotherapy at four metropolitan public hospitals in Australia completed a touch screen computer survey. A consecutive subsample of patients attending three of these treatment centres answered additional questions about psychological support preferences. RESULTS: Of 304 respondents, 54% [95% confidence interval (CI) 48% to 60%] perceived that they were currently experiencing mild to severe anxiety and depression. 22% (95% CI 18% to 27%) indicated a preference to be offered professional help. There was moderate agreement between the HADS and single-item responses for categorisation of anxiety and depression. Patient-perceived mild to severe anxiety and depression levels appeared to be the best measure for identifying those with a preference to be offered professional assistance. Of a subsample of 193 respondents, 89% (95% CI 84% to 93%) indicated that if they were experiencing anxiety or depression, they would accept (or were currently using) professional support. CONCLUSIONS: Single-item screening in a cancer care setting may not adequately capture clinical anxiety and depression. However, single-items assessing patients' perceived levels of anxiety and depression are useful indicators of whether patients want to be offered, and are likely to accept, psychosocial care.

Psychological distress in cancer patients undergoing radiation therapy treatment.

PURPOSE: The objective of this study was to examine the likely presence of, and factors associated with, anxiety, depression and overall psychological distress in cancer outpatients undergoing radiation therapy treatment in Sydney, Australia. METHODS: A touchscreen computer survey was conducted in four radiation therapy treatment centre waiting rooms. Patients waiting to receive treatment completed the survey which included questions about demographic and disease characteristics, survey acceptability and the Hospital Anxiety and Depression Scale (HADS). RESULTS: A total of 454 patients (70 %) completed the touchscreen computer survey. The likely presence of anxiety (HADS-A ≥11), depression (HADS-D ≥11) and overall psychological distress (HADS-T ≥15) was 15, 5.7 and 22 %, respectively. Cancer type was found to be associated with each HADS screening category. The majority of patients reported high survey acceptability and willingness to complete similar touchscreen computer surveys in the treatment centre waiting room on additional occasions. CONCLUSIONS: As radiotherapy patients frequently attend the radiation oncology department, routine screening and intervention for elevated levels of psychological distress in this setting appears to be feasible. High survey completion rates and high patient-rated acceptability support this approach to screening. The likely presence of psychological distress is reported for this sample; however, the selection of HADS threshold scores is likely to have influenced the reported rates. Further research is needed to identify how cancer type impacts on likely caseness across the different HADS classifications examined.

Cancer patients' willingness to answer survey questions about life expectancy.

PURPOSE: This study aimed to determine the proportion and characteristics of radiation oncology outpatients who were willing to answer questions about their life expectancy. METHODS: A cross-sectional patient self-report survey was conducted using touch screen computers in Australian radiation oncology treatment centers. The primary outcome was the respondent's willingness to complete a survey subsection about life expectancy. Demographic and disease characteristics were also collected, and level of anxiety and depression was assessed using the Hospital Anxiety and Depression Scale. RESULTS: Of the 469 oncology outpatients who completed the survey, 327 (70 %; 95 % CI, 65 %, 74 %) indicated that they were willing to answer questions about life expectancy. Being female (p < 0.001), older (p < 0.05), born in Asia (p < 0.05), and being diagnosed with cancer types other than breast and prostate cancer (p < 0.01) were associated with lower odds of answering life expectancy questions. CONCLUSIONS: The opportunity to opt-out of survey questions about sensitive issues such as life expectancy is a feasible method for accessing important information about patient preferences while minimizing burden. Further research may be needed to improve acceptability of life expectancy research to some patient groups.

Current state of medical-advice-seeking behaviour for symptoms of colorectal cancer: determinants of failure and delay in medical consultation.

AIM: There are few data on the factors associated with healthcare-seeking behaviour for symptoms of colorectal cancer. This study describes the determinants of failure and delay in seeking medical advice for rectal bleeding and change in bowel habit. METHOD: In total, 1592 persons (56-88 years) were randomly selected from the Hunter Community Study and mailed a questionnaire. RESULTS: In all, 18% (60/332) of respondents experiencing rectal bleeding and 20% (39/195) reporting change in bowel habit had never consulted a doctor. The rate of delay (>1 month) for each symptom was 18% and 37%. The reasons for delay included the assumption that the symptoms were not serious or that they were benign. Triggers for seeking medical advice varied. Healthcare-seeking behaviour for rectal bleeding had not significantly improved compared with a previous community-based study. CONCLUSION: The seriousness of symptoms, importance of early detection and prompt medical consultation must be articulated in health messages to at-risk persons.

The prevalence and characteristics of alcohol-related presentations to emergency departments in rural Australia.

BACKGROUND: The burden of harm associated with alcohol misuse is disproportionately high in rural areas of Australia, and a considerable proportion of this burden is borne by the health system. The health impact of alcohol in rural areas has been measured in terms of the contribution of alcohol to hospital inpatient admissions and mortality rates, despite many more alcohol-related cases presenting to emergency departments (EDs). This study aims to estimate the proportion of presentations to EDs in rural Australia that are alcohol-related and to identify the associated patient and presentation characteristics. METHODS: Patients aged ≥14 years presenting to four EDs in rural NSW were assessed on two measures: (1) Clinician judgement of alcohol consumption, and (2) patient self-report of alcohol consumption in the 6 h preceding the onset of their condition. RESULTS: Preliminary analyses revealed sample selection biases in two of the EDs, and these samples were consequently excluded from further analyses. In the two remaining EDs, 46% of presentations were assessed, of which 9% were identified as alcohol-related. Presentations for mental disorders, those with more urgent triage categories and those occurring on weekends or at night were more often alcohol-related. CONCLUSIONS: The prevalence of alcohol-related ED presentations observed was at the lower end of the documented range, probably due to methodological differences and limitations, as well as geographic variation. Despite this, alcohol-related presentations were associated with a substantial impact on the ED. Policies and programs to reduce the impact of alcohol on rural emergency departments are needed.

Patient or treatment centre? Where are efforts invested to improve cancer patients' psychosocial outcomes?

The psychosocial outcomes of cancer patients may be influenced by individual-level, social and treatment centre predictors. This paper aimed to examine the extent to which individual, social and treatment centre variables have been examined as predictors or targets of intervention for psychosocial outcomes of cancer patients. Medline was searched to find studies in which the psychological outcomes of cancer patient were primary variables. Papers published in English between 1999 and 2009 that reported primary data relevant to psychosocial outcomes for cancer patients were included, with 20% randomly selected for further coding. Descriptive studies were coded for inclusion of individual, social or treatment centre variables. Intervention studies were coded to determine if the unit of intervention was the individual patient, social unit or treatment centre. After random sampling, 412 publications meeting the inclusion criteria were identified, 169 were descriptive and 243 interventions. Of the descriptive papers 95.0% included individual predictors, and 5.0% social predictors. None of the descriptive papers examined treatment centre variables as predictors of psychosocial outcomes. Similarly, none of the interventions evaluated the effectiveness of treatment centre interventions for improving psychosocial outcomes. Potential reasons for the overwhelming dominance of individual predictors and individual-focused interventions in psychosocial literature are discussed.

The validity and reliability of an interactive computer tobacco and alcohol use survey in general practice.

BACKGROUND: Uncertainty regarding the accuracy of the computer as a data collection or patient screening tool persists. Previous research evaluating the validity of computer health surveys have tended to compare those responses to that of paper survey or clinical interview (as the gold standard). This approach is limited as it assumes that the paper version of the self-report survey is valid and an appropriate gold standard. OBJECTIVES: First, to compare the accuracy of computer and paper methods of assessing self-reported smoking and alcohol use in general practice with biochemical measures as gold standard. Second, to compare the test re-test reliability of computer administration, paper administration and mixed methods of assessing self-reported smoking status and alcohol use in general practice. METHODS: A randomised cross-over design was used. Consenting patients were randomly assigned to one of four groups; Group 1. C-C : completing a computer survey at the time of that consultation (Time 1) and a computer survey 4-7 days later (Time 2); Group 2. C-P: completing a computer survey at Time 1 and a paper survey at Time 2; Group 3. P-C: completing a paper survey at Time 1 and a computer survey at Time 2; and Group 4. P-P: completing a paper survey at Time 1 and 2. At Time 1 all participants also completed biochemical measures to validate self-reported smoking status (expired air carbon monoxide breath test) and alcohol consumption (ethyl alcohol urine assay). RESULTS: Of the 618 who were eligible, 575 (93%) consented to completing the Time 1 surveys. Of these, 71% (N=411) completed Time 2 surveys. Compared to CO, the computer smoking self-report survey demonstrated 91% sensitivity, 94% specificity, 75% positive predictive value (PPV) and 98% negative predictive value (NPV). The equivalent paper survey demonstrated 86% sensitivity, 95% specificity, 80% PPV, and 96% NPV. Compared to urine assay, the computer alcohol use self-report survey demonstrated 92% sensitivity, 50% specificity, 10% PPV and 99% NPV. The equivalent paper survey demonstrated 75% sensitivity, 57% specificity, 6% PPV, and 98% NPV. Level of agreement of smoking self-reports at Time 1 and Time 2 revealed kappa coefficients ranging from 0.95 to 0.98 in each group and hazardous alcohol use self-reports at Time 1 and Time 2 revealed kappa coefficients ranging from 0.90 to 0.96 in each group. CONCLUSION: The collection of self-reported health risk information is equally accurate and reliable using computer interface in the general practice setting as traditional paper survey. Computer survey appears highly reliable and accurate for the measurement of smoking status. Further research is needed to confirm the adequacy of the quantity/frequency measure in detecting those who drink alcohol. Interactive computer administered health surveys offer a number of advantages to researchers and clinicians and further research is warranted.

Are we there yet? The state of the evidence base for guidelines on breaking bad news to cancer patients.

The way clinicians break bad news to cancer patients has been retrospectively associated with poor psychosocial outcomes for patients. Education and practice in breaking bad news may be ineffective for improving patients' well-being unless it is informed by a sound evidence base. In the health field, research efforts are expected to advance evidence over time to inform evidence-based practice. Key characteristics of an advancing evidence base are a predominance of new data, and rigorous intervention studies which prospectively demonstrate improved outcomes. This review aimed to examine the progress of the evidence base in breaking bad news to cancer patients. Manual and computer-based searches (Medline and PsycINFO) were performed to identify publications on the topic of breaking bad news to cancer patients published between January 1995 and March 2009. Relevant publications were coded in terms of whether they provided new data, examined psychosocial outcomes for patients or tested intervention strategies and whether intervention studies met criteria for design rigour. Of the 245 relevant publications, 55.5% provided new data and 16.7% were intervention studies. Much of the intervention effort was directed towards improving provider skills rather than patient outcomes (9.8% of studies). Less than 2% of publications were rigorous intervention studies which addressed psychosocial outcomes for patients. Rigorous intervention studies which evaluate strategies for improving psychosocial outcomes in relation to breaking bad news to cancer patients are needed. Current practice and training regarding breaking bad news cannot be regarded as evidence-based until further research is completed.

Health inequities: the need for action by schools of medicine.

BACKGROUND: It is well recognised that marked inequalities in mortality and morbidity exist between populations particularly those in lower socio-economic groups, including Indigenous and some ethnic minorities. Academic medicine has not yet articulated a clear stance on reducing health inequity within communities. AIM: To develop criteria that medical schools can implement to reduce health inequity. These criteria will enable the performance of a medical school's commitment to health equity to be measured. RESULTS AND CONCLUSION: We suggest that the contribution to lessening health inequity should be seen as an integral and important role of undergraduate medical education and the academic institutions that provide such programs. Five strategies aimed at increasing the commitment of medical and other undergraduate health students to work with disadvantaged groups to improve their health are described. They include student selection to increase representativeness of students and importantly, support for retention and academic success; undergraduate curriculum, both core and elective, to address inequality and provide skills necessary to implement change in a range of areas that impact on health; academic physicians modelling the above by actively working in and for disadvantaged groups; developing centres of excellence carrying out research in health inequity, particularly intervention rather than solely descriptive research and creating high status academic appointments in key designated positions addressing inequity. Schools of Medicine could be rated on their action on these criteria so that benchmarking across institutions could occur.

Competency based teaching: the need for a new approach to teaching clinical skills in the undergraduate medical education course.

This paper will argue that undergraduate medical courses are failing to provide students with adequate training in the area of clinical skills. Some of the reasons for this failure include changes in the health system leading to increasing commitments for clinicians, fewer patients as clinical case examples and limitations in current teaching methods. Current clinical teaching methods measure progress through completion of clinical "blocks" of learning and one-off clinical examinations. This paper offers an alternative approach to clinical teaching in undergraduate medicine. In this approach, each student's progress is measured through the attainment of a predetermined level of competency in dealing with a range of clinical conditions. Some of the benefits of a new approach to clinical teaching in undergraduate medicine include flexibility for both students and clinicians in terms of when and where clinical learning can occur, a significant emphasis on active learning, and increased generalization and integration of learning.

The Needs Assessment for Advanced Cancer Patients (NA-ACP): a measure of the perceived needs of patients with advanced, incurable cancer. a study of validity, reliability and acceptability.

AIMS: To describe the psychometric evaluation of a measure designed to assess the perceived needs of patients with advanced, incurable cancer. METHOD: A questionnaire known as the Needs Assessment for Advanced Cancer Patients (NA-ACP) was developed based on a review of available literature and professional opinion. A sample of 246 patients (consent rate = 59%) completed the NA-ACP, 28 patients completed the acceptability questions, while 41 completed a retest copy of the NA-ACP. The construct validity of the questionnaire was examined via principal components analysis, while reliability was evaluated in terms of the internal consistency of domains and test-retest scores. RESULTS: The principal components analysis revealed seven domains assessing patients' psychological/emotional, medical information/communication, social, symptom, daily living, spiritual and financial needs. The test-retest reliability estimates were within accepted levels, as were all but one of the internal consistency scores. The NA-ACP was highly acceptable for this patient group. CONCLUSION: The NA-ACP is one of the first multi-dimensional instruments specifically designed to assess the needs of patients with advanced, incurable cancer. The present study provides evidence of the NA-ACP's validity, reliability, and acceptability. The NA-ACP has potential applications both as a research tool and within clinical settings.

A cost-effective approach to the development of printed materials: a randomized controlled trial of three strategies.

Printed materials have been a primary mode of communication in public health education. Three major approaches to the development of these materials--the application of characteristics identified in the literature, behavioral strategies and marketing strategies--have major implications for both the effectiveness and cost of materials. However, little attention has been directed towards the cost-effectiveness of such approaches. In the present study, three pamphlets were developed using successive addition of each approach: first literature characteristics only ('C' pamphlet), then behavioral strategies ('C + B' pamphlet) and then marketing strategies ('C + B + M' pamphlet). Each pamphlet encouraged women to join a Pap Test Reminder Service (PTRS). Each pamphlet was mailed to a randomly selected sample of 2700 women aged 50-69 years. Registrations with the PTRS were monitored and 420 women in each pamphlet group were surveyed by telephone. It was reported that the 'C + B' and 'C + B + M' pamphlets were significantly more effective than the 'C' pamphlet. The 'C + B' pamphlet was the most cost-effective of the three pamphlets. There were no significant differences between any of the pamphlet groups on acceptability, knowledge or attitudes. It was suggested that the inclusion of behavioral strategies is likely to be a cost-effective approach to the development of printed health education materials.

Print material content and design: is it relevant to effectiveness?

Printed materials are widely used in cancer education. There are a considerable number of guidelines in the literature on the content and design characteristics of print materials. However, there is little outcome-based evidence about whether materials containing these characteristics are more effective under real-world conditions. Six pamphlets were designed such that two had most of these characteristics, one had content characteristics only, one had design characteristics only and two had few of the characteristics. The pamphlets encouraged women aged 50-69 years to join a Pap Test Reminder Service (PTRS). The pamphlets were mailed to a randomly selected sample of 7668 women. Registrations with the PTRS were monitored. Receipt, readership, acceptability, knowledge and attitudes were assessed by telephone interview. There were no significant differences between the pamphlet groups in registrations with the PTRS, receipt, readership, acceptability, knowledge or attitudes. It was concluded that the inclusion of content and design characteristics does not result in more effective print materials.

Trialling a new way to learn clinical skills: systematic clinical appraisal and learning.

AIM: To describe and evaluate the effectiveness of a new method of teaching clinical skills designed to increase students' active and self-directed learning as well as tutor feedback. METHODS: A total of 22 consenting Year 4 medical students undertaking general practice and general surgery clinical experience were involved in a pre- and post-test research design. In the initial period of the study, students were taught clinical skills in a traditional manner. In the second phase a clinical teaching strategy called systematic clinical appraisal and learning (SCAL) was utilised. This learning strategy involved active and self-directed learning, holistic care and immediate feedback. Students independently saw a patient and were asked to make judgements about the patient's potential diagnosis, tests required, management, psychosocial needs, preventive health requirements, and any ethical problems. These judgements were then compared with those of the clinical supervisor, who saw the same patient independently. Students recorded details for each consultation. Comparisons were made of the two study periods to examine whether the use of SCAL increased the number of students' independent judgements, perceived student learning, tutor feedback and self-directed learning. RESULTS: During the SCAL learning period, students reported making a greater number of statistically significant independent judgements, and receiving significantly increased tutor feedback in both general practice and general surgery. The number of learning goals set by students was not found to differ between the two periods in surgery but significantly increased in general practice in the SCAL period. Students' perceptions of their learning significantly increased in the SCAL period in surgery but not in general practice. During the traditional learning period in both settings, there was limited student decision-making about most aspects of care, but particularly those relating to prevention, psychosocial issues and ethics. CONCLUSIONS: The SCAL approach appears to offer some advantages over traditional clinical skills teaching. It appears to encourage active and independent decision-making, and to increase tutor feedback. Further exploration of the approach appears warranted.

Which common clinical conditions should medical students be able to manage by graduation? A perspective from Australian interns.

The objectives of the study were to report the development of a core curriculum that details the clinical conditions medical students should be able to manage upon graduation; and to canvass the opinion of interns (first-year postgraduate doctors) regarding their perceptions of the level of skill required to manage each condition. Literature relating to core curriculum development and training of junior medical officers was reviewed and stakeholders in the education and training of medical students and junior doctors in the state of New South Wales, Australia (intern supervisors, academics, registrars, nurses and interns) were consulted. The final curriculum spanned 106 conditions, 77 'differentiated' and 29 'undifferentiated'. Four levels of skill at which conditions should potentially be managed were also identified: 'Theoretical knowledge only'; 'Recognize symptoms and signs without supervision'; Initiate preliminary investigations, management and/or treatment without supervision'; and 'Total investigation, management and/or treatment without supervision'. The list of conditions in the curriculum was converted to a survey format and a one-in-two random sample of interns (n = 193) practising in New South Wales who graduated from the state's three medical schools were surveyed regarding the level of skill required for managing each clinical condition at graduation. A total of 51.3% of interns responded to the survey. Interns felt they should be able to initiate preliminary investigation, management and/or treatment for most conditions in the curriculum, with more than half acknowledging this level of management for 53 of the differentiated and 28 of the undifferentiated conditions. It is concluded that developing core curricula in medical education can involve multiple stakeholders, including junior doctors as the consumers of educational experiences. The data gathered may be useful to medical schools revising their curricula.

Translating learning principles into practice: a new strategy for learning clinical skills.

AIM: There are data to suggest that medical school may not adequately prepare doctors for practice and that there are deficiencies in undergraduate teaching of skills in history taking, physical examination, diagnosis and management (clinical skills). There is a need to re-evaluate methods by which we can teach clinical skills effectively. This aim of this review was to describe the literature concerning the important principles underpinning effective clinical learning. Subsequently a structured learning tool and teaching process was developed in order to support these principles. METHOD: The principles of effective clinical learning were derived after a search of the medical education and relevant behavioural science literature. Consequently, a structured learning tool and teaching process was developed in order to potentiate the translation of these principles into practice for medical school training in clinical skills. RESULTS: Ten principles were derived from the 68 articles referred to in this review. These were: making active decisions, an individual focus to learning, gaining experience, feedback to the learner, reciprocal learning, holistic care, relevant learning, feasibility, cost efficiency and mentoring. A process for history taking, physical examination and management plan was developed for medical students which incorporated these principles. CONCLUSION: Relevant literature can provide the foundations for teaching and learning methods in medical education. We plan to trial this method and evaluate the impact on student learning outcomes.

Interactional skills of students from traditional and non-traditional medical schools before and after alcohol education.

OBJECTIVE: To compare alcohol-related intervention and general interactional skills performance of medical students from a traditional (Sydney) and a non-traditional (Newcastle) medical school, before and after participation in an alcohol education programme about brief intervention. DESIGN: In two controlled trials, students received either a didactic alcohol education programme or didactic input plus skills-based training. Prior to and after training, all students completed videotaped interviews with simulated patients. SETTING: The Faculties of Medicine at the University of Newcastle and the University of Sydney, Australia. SUBJECTS: Fifth-year medical students (n=154). RESULTS: Both alcohol-related intervention and general interactional skills scores of the Newcastle students were significantly higher than those of the Sydney students at pre-test but not after training. Although alcohol-related interactional skills scores improved after training at both universities, they did not reach a satisfactory level. The educational approach used had no effect on post-test scores at either university. CONCLUSIONS: Significant baseline differences in interactional skills scores favouring non-traditional over traditional students were no longer evident after both groups had been involved in an alcohol education programme. Further research is required to develop more effective alcohol intervention training methods.

Preventive and other interactional skills of general practitioners, surgeons, and physicians: perceived competence and endorsement of postgraduate training.

BACKGROUND: Perceived competencies and support for formal postgraduate training across a range of preventive and other interactional skills were examined in three medical groups. METHODS: All eligible final year students and recent graduates of the three major Australian medical colleges (n = 767) were mailed a questionnaire examining communication skills in four domains: preventive, educational, therapeutic, and general. RESULTS: Overall consent rate was 45%. For most items, at least one-third of each group reported low competence. On preventive items, low competence ratings ranged from 5 to 39% in general practice, 38 to 67% in surgery, and 33 to 51% in the speciality physician group. Significant intergroup differences occurred on eight competence items. Agreement with training on preventive topics ranged from 80 to 91% in general practice, 48 to 69% in surgery, and 72 to 82% in the specialty physician group. On all 11 training items where significant differences occurred, the general practice group reported the highest level and the surgeon group the lowest level of endorsement for formal training and assessment. CONCLUSIONS: Substantial proportions in the general practice, surgery, and the physician specialty report lack of competence in common interactional skills. There were high levels of support for formal training in preventive and other interactional skills. The strong endorsement supports the development of effective, tailored interactional skills training programs.

Identifying medical school learning needs: a survey of Australian interns.

OBJECTIVES: To survey interns regarding their opinion of medical school learning needs for a range of core skills. METHODS: A random sample of interns practising in New South Wales, Australia, who graduated from the state's three medical schools were surveyed two-thirds of the way through their first hospital year. They were asked whether there was a need for further medical school education for each of 226 core skills. Skills were grouped into five themes: management of clinical conditions; clinical investigations; clinical procedures; core practice; and professional development. RESULTS: Frequency distributions weighted for age, gender and medical school background were calculated for each item. The 20 most frequently identified needs related to examinations of the eye and ear, nose and throat; managing uncooperative patients and difficult patient interactions; prescribing; writing not for resuscitation orders and death certificates. Also included were procedural needs related to ear, nose and throat; plastering and wound management; and needs for more education in the management of clinical conditions related predominantly to "acute" cases such as anaphylaxis and diabetic ketoacidosis. CONCLUSION: Interns were able to discriminate between their needs for different skills and identified many core skills for which they perceived there was a need for more medical school education. The implications for medical education are discussed.

Benzodiazepine prescribing by GP registrars. A trial of educational outreach.

BACKGROUND: Despite ongoing concern about adverse effects and dependence on benzodiazepines, approximately 2% of Australians are still taking them on a regular basis. The aim of this study was to evaluate the effectiveness of an educational outreach or 'academic detailing' program about prescribing of benzodiazepines. METHOD: In this randomised trial general practice registrars (n = 157) in New South Wales were allocated to an intervention group (n = 79), which received a 20 minute educational outreach visit; or a control group (n = 78) which received an intervention on an unrelated topic. Prescribing behaviour was monitored by a pre-intervention and two post-intervention practice activity surveys. MAIN OUTCOME MEASURES: These were the rate of benzodiazepine prescribing for all indications, for anxiety and for sleep disorders. RESULTS: Overall benzodiazepine prescribing by the intervention group declined from 2.3 to 1.7 per 100 encounters, while the control group also declined from 2.2 to 1.6 per 100 encounters. Analysis of variance showed this was a significant drop over time (P = 0.042) but there was no difference between groups (P = 0.99). The prescribing decrease observed was in continuing rather than initial prescriptions. CONCLUSIONS: A marked decrease in benzodiazepine prescribing was seen over the course of the study in both intervention and control groups but no differential effect due to the educational outreach visit was found.