A comprehensive assessment of the avoidability of long-term complications of diabetes. A case-control study. SID-AMD Italian Study Group for the Implementation of the St. Vincent Declaration.

OBJECTIVE: To identify and quantify risk factors for the development of long-term diabetic complications (i.e., critical limb ischemia, amputation, chronic renal failure [creatinine > 3 mg/dl], dialysis treatment, proliferative retinopathy, blindness), with particular emphasis on those variables that, being related to quality of care, can be considered avoidable. RESEARCH DESIGN AND METHODS: We designed a case-control study that enrolled 886 patients with long-term diabetic complications and 1,888 control subjects without complications from 35 diabetic outpatient clinics and 49 general practitioners offices during a 6-month period. Selected socioeconomic, pathophysiologic, self-care, health care, and lifestyle information were collected for all patients. RESULTS: A logistic regression analysis showed that several factors are related to the development of major diabetic complications. Among patient characteristics, male sex (odds ratio [OR] = 1.8, 95% CI 1.4-2.3) and age (OR = 1.7, 95% Cl 1.2-2.4 for patients between 50 and 69 years of age as opposed to those younger than 50 years of age) were associated with an increased risk of complication. Among clinical variables, the type and the duration of diabetes were the most important predictors of diabetic complications. The presence of hypertension was also associated with the development of diabetic complications, particularly when it was poorly controlled by treatment (OR = 3.1, 95% CI 2.3-4.3). Patients who needed help to reach a health care facility and those who did not regularly attend such a facility were at higher risk of developing complications (OR = 1.5, 95% CI 1.2-1.9; OR = 1.7, 95% CI 1.3-2.2, respectively). Educational aspects were also related to the outcome: patients who did not receive any kind of educational intervention had an increased risk of developing complications (OR = 4.1, 95% CI 1.7-9.7), while self-management of insulin therapy had a protective effect (OR = 0.6, 95% CI 0.5-0.8). The summary attributable risk related to avoidable risk factors (i.e., uncontrolled hypertension, poor compliance with visit scheduling, inadequate diabetes education, no self-management of insulin treatment) was 0.39. CONCLUSIONS: Our data suggest that, by removing avoidable risk factors, the number of diabetic complications considered could be reduced by more than one-third. The case-control methodology represents an efficient way of monitoring clinical practice and relating it to important outcomes. It can be of help for policy makers in identifying the more effective strategies and in tailoring specific interventions aimed at improving the quality of the care delivered to diabetic patients.

Long-term survival following neoadjuvant chemotherapy and radical surgery in locally advanced cervical cancer.

The aim of this study was to analyse the long-term survival and the relationships between prognostic factors at presentation, chemoresponsiveness and disease outcome in patients with locally advanced cervical cancer treated by neoadjuvant chemotherapy and radical surgery (RS). Two consecutive studies of neoadjuvant chemotherapy containing cisplatin, bleomycin plus/minus methotrexate followed by radical hysterectomy and systematic aortic and pelvic lymphadenectomy were carried out between January 1986 and September 1990 on 130 patients with > or = 4 cm stage IB2-III cervical cancer. Survival analysis was performed using the Kaplan and Meier test and Cox's multivariate regression analysis. 128 (98%) of the patients enrolled were evaluable for clinical response and survival, 83% (106) of the patients responded to chemotherapy, with a 15% complete response rate. Logistic regression analysis demonstrated that International Federation of Gynecology and Obstetrics (FIGO) stage, cervical tumour size, parametrial involvement and histotype are highly predictive of response. Responding patients underwent laparotomy, but 8% were not amenable for radical surgery. The 10-year survival estimates were 91%, 80% and 34.5% for stage IB2-IIA bulky, IIB and III, respectively (P < 0.001). After Cox's regression analysis, the parameters significantly associated with survival were the same factors predicting response to neoadjuvant chemotherapy. No stage IB2-IIA bulky patient has so far relapsed, while 12% stage IIB and 56% stage III patients recurred. The 10-year disease-free survival estimates are 91% and 44% for stage IB2-IIB and III, respectively (P < 0.001). Metastatic nodes and persistent tumour in the parametria were the only two independent factors for disease-free survival after multiple regression analysis. After a long-term follow-up (median follow-up 98 months (20-129+)), our results give new evidence of the prognostic value of response to neoadjuvant chemotherapy and of a possible therapeutic benefit of the sequential treatment adopted which, however, must be verified in a randomised setting.

Appropriateness and variation of surgical treatment of breast cancer in Italy: when excellence in clinical research does not match with generalized good quality care. Progetto Oncologia Femminile.

To assess appropriateness of surgical care delivered to breast cancer patients in Italy and quantify the use of unnecessary radical procedures, a retrospective charts review of patients treated in 1988-1989 was conducted. A series of 1724 consecutive patients (median age 61 years; range 17-89) treated in 63 hospitals selected from within 8 regions with newly diagnosed operable breast carcinoma was evaluated. Overall, 541 (38%) patients had inappropriate surgery with more than two thirds of it being accounted for by the use of unnecessary mutilating Halsted mastectomy. Substantial geographic variation emerged in the overall rates of appropriateness (range 88-52%) which were not substantially affected by allowance for imbalances in patient- and hospital-related variables. Despite the important contribution given by Italian clinical researchers to the demonstration that less radical surgery can be as good as more radical procedures, still a substantial proportion of breast cancer patients are treated too aggressively. Besides pointing to the urgent need of interventions aimed at facilitating the process of technology transfer in order to promote more appropriate surgical care, these results suggest that efforts to increase patients' participation into treatment decision and awareness about alternative treatment options are warranted.

Randomized clinical trials on medical treatment of glaucoma. Are they appropriate to guide clinical practice?

A systematic quantitative and qualitative overview of published randomized clinical trials was undertaken to assess the yield of medical treatment on the outcome of patients with primary open angle glaucoma. Reports of 102 randomized clinical trials were published between 1975 and 1991, totalling about 5000 patients. Only 16% (16/102) of the trials were, however, properly designed (ie, comparing an active treatment with a placebo-treated or untreated control group) to answer the question of whether any medical treatment can effectively cure patients with primary open angle glaucoma. Pooled analysis showed a moderate yet statistically significant reduction in mean intraocular pressure (-4.9 mm Hg; 95% confidence interval [CI], -7.3 to -2.5 mm Hg); however, data on long-term visual field changes were available in only three randomized clinical trials, and their statistical combination failed to show a significant protective effect of active treatment (odds ratio, 0.75; 95% CI, 0.42 to 1.35). All of the remaining 86 randomized clinical trials looked at the effectiveness of one drug vs another in lowering intraocular pressure and were thus of no use in the overview. Practicing ophthalmologists should be aware that the effectiveness of pressure-lowering agents in the treatment of primary open angle glaucoma is still to be determined and that the vast majority of published trials are not appropriate to guide clinical practice. It is urgent to plan trials with end-point and follow-up duration that is fully relevant for the health of patients.

Epidemiology and determinants of blood glucose self-monitoring in clinical practice.

The aim of this study was to describe the epidemiology of self-monitoring of blood glucose and to identify specific characteristics of those subgroups of diabetic patients treated with insulin that are most likely to monitor their blood glucose according to medical recommendations. Data were collected on 1384 insulin-treated patients, enrolled from 35 diabetic outpatient clinics and 49 general practitioners' offices between December 1993 and June 1994. Seventeen Italian regions out of 20 were included in the study. Our data show that 418 (31%) diabetic patients treated with insulin had never practised blood glucose self-monitoring. In addition, only 242 patients (18.2%) self-monitored their glycemia with a mean frequency of at least once a day (29.7% among insulin-dependent diabetes mellitus (IDDM) and 13.9%, among insulin-treated non-insulin-dependent diabetes mellitus (NIDDM-IT) patients). Patients' characteristics associated with a higher probability of practising blood glucose self-monitoring were age below 50 years, being treated at a diabetic outpatient clinic, hypertension, need of three or more insulin injections per day, history of hypoglycemic episodes, ability to self-manage insulin doses. Our study calls for vigorous efforts aimed at promoting the incorporation of clearly-defined educational programs at each level of care, in order to improve the motivation and self-care of diabetic patients. Furthermore, studies are necessary to identify subgroups of diabetic patients that truly need to self-monitor blood glycemia, and to assess the efficacy of the practice of self-monitoring of blood glucose in improving metabolic control and reducing acute and long-term diabetic complications.

Breast cancer surgery in 30 Italian general hospitals.

The utilization of limited surgery in patients with breast cancer operated between September 1986 and July 1988 was assessed using information collected within a cohort subsequently enrolled in a randomized clinical trial testing the efficacy of post-surgical follow up. Overall 30% had limited surgery, 61% had other more radical procedures and 9% are still undergoing an unnecessary Halsted mastectomy. Several factors were related to the lower likelihood of getting a conservative procedure: geographic distribution, age, level of education, quadrant and nodal state. The paper discusses the implications of these findings in view of the otherwise growing consensus that more radical surgery should be abandoned.

Risk factors for lower limb complications in diabetic patients. The Italian Study Group for the Implementation of the St. Vincent Declaration.

Diabetic lower extremity complications may be influenced by a number of factors, including those related to the interaction between patients and the health-care system. Our objective is to identify risk factors for the development of lower limb complications, by looking for classical clinical variables and those related to quality of care. A case-control study was carried out between December 1993 and June 1994 by interviewing 348 patients with lower-limb diabetic complications and 1050 controls enrolled from 35 diabetes outpatient clinics and 49 general practitioner's offices in Italy. Among sociodemographic characteristics associated with increased risk of lower limb complications were male gender [odds ratio (OR) = 2.5, confidence interval (CI) 1.6-3.9], age between 50 and 70 years as opposed to younger than 50 (OR = 3.6, CI 2.1-6.3) and being single as opposed to married (OR = 1.4, CI 1.1-1.8). Among clinical variables, treatment with insulin for IDDM and NIDDM patients was an important predictor of lower extremity complications compared to NIDDM patients not being treated with insulin. Cardio-cerebrovascular disease and presence of diabetic neuropathy were associated with a higher risk of being a case (OR = 1.4, CI 1.2-1.8 and OR = 3.0, CI 2.1-4.2, respectively). Patients who needed help to reach the health facility before the onset of the complications and those who did not attend health facilities regularly were more liable to develop complications (OR = 1.5, CI 1.1-2.2 and OR = 2.0, CI 1.3-3.0, respectively). Patients who had never received educational intervention had a threefold risk of being a case as compared to those who received health information regularly. The study identifies factors most likely to be related to adverse outcome and permits to discriminate between avoidable and unavoidable factors.

[Geographic variability in the appropriateness of surgical treatment of breast carcinoma]. / Variabilità geografica nell'appropriatezza della terapia chirurgica del carcinoma della mammella.

To assess appropriateness of surgical care delivered to breast cancer patients in Italy and quantify the use of unnecessary radical procedures, a retrospective charts review of patients treated in 1988-1989 was conducted. Information about hospital characteristics was collected directly from administrative departments as a part of a national survey. A series of 1724 consecutive patients (median age 61 years; range 17-89) treated in 63 hospitals selected from within 8 regions, with newly diagnosed operable breast carcinoma was evaluated. Overall, 541 (38%) patients had inappropriate surgery with more than two thirds of it being accounted for by the use of unnecessary mutilating Halsted mastectomy. About one-fourth of patients with stage I-II disease undergone mastectomy according Halsted technique and conservative surgery in patients with small tumors (i.e. tumor size < = 2 cm) was under utilized. Substantial geographic variations emerged in the overall rates of inappropriateness (range 12-48%) which were not substantially affected by allowance for imbalances in patient- and hospital-related variables. Despite the important contribution given by Italian clinical researchers to the demonstration that less radical surgery can be as good as more radical procedures, still a substantial proportion of breast cancer patients are treated too aggressively. Besides pointing to the urgent need of interventions aimed at promoting more appropriate surgical care these results suggest that efforts to increase patients' participation into treatment decision and awareness about alternative treatment options are warranted.

[The adjuvant treatment of operable breast carcinoma. The current status and outlook]. / Il trattamento adiuvante del carcinoma mammario operabile. Attualità e prospettive.

Breast cancer represents one of the most frequent neoplasm: every year, in Italy, 25,000 new cases are diagnosed with more than 9,000 deaths. In Italy--and also in other countries--has been shown a broad variability in the use of diagnostic or therapeutic procedures caused by different hospitals' characteristics, patients' conditions and physicians' opinions. Among the different therapeutic options, post-surgical therapy, that extends the disease-free interval and survival, fills an important position in the cure of breast cancer in relation to the overview published on Lancet the 4th and the 11th January 1992. According to this overview, patients who had surgery for breast cancer benefitted by an adjuvant therapy--hormonal or cytotoxic--with a reduction of mortality between 15 and 20%. That means, in Italy, a reduction of one thousand deaths per year for this neoplasm. Particularly the effectiveness of hormonal treatment in postmenopausal patients (not only in the oestrogen receptor (ER) positive but in the ER poor too), of ovarian ablation and chemotherapy in premenopausal patients was confirmed. In node negative patients the present overview recommends the adjuvant treatment for the risk reduction in recurrence and mortality. Nevertheless important research implications persist about treatment and a better definition of high risk patients; therefore more patient should be included in the randomized clinical trial.

A survey in Puglia: the attitudes and opinions of specialists, general physicians and patients on follow-up practice. G.S.Bio.Ca.M.

BACKGROUND: The follow-up for breast cancer patients is a well established routine practice requiring the organization and coordination of many professional figures and a large expenditure of national health funds. To study the problems in practical clinical management of such a complicated health program in a typical population, a questionnaire survey of the opinions and attitudes of specialists, general practitioners and patients directly involved in clinical follow-up was performed in the Puglia region of Southern Italy. MATERIALS AND METHODS: A representative sample of specialists (n = 285), general practitioners (n = 263) and patients (n = 284) involved in the management of follow-up practice for breast cancer in Puglia received different questionnaires to ascertain their behaviour, attitudes, opinions, perception of the disease and the organizational requirements for treatment. A total of 57.4% of questionnaires were returned. RESULTS AND DISCUSSION: The most important results were: (a) about one-third of cases complained of difficulties in follow-up management due to the lack of cooperation and integration of follow-up procedures among specialists; (b) the general practitioners preferred to have a more active role in follow-up; (c) the patients reported that being managed by more than one physician or living far from follow-up facilities resulted in an inferior quality of life.

Is proper methodology associated with the use of a clinically relevant outcome measure? The case of randomized clinical trials on medical treatment of open-angle glaucoma.

OBJECTIVE: The aim of this study was to assess critically the published literature concerning medical treatment of primary open-angle glaucoma (POAG), and to see whether trial methodologic quality was related to a clinically relevant outcome measure. METHODS: We identified and reviewed the methodologic quality of 102 published randomized clinical trials (RCTs) on treatment of POAG using an explicit protocol and explored the association between selected aspects of design and conduct and the studies' clinical relevance. RESULTS: Our analysis revealed serious methodologic problems with the trials reviewed. Areas of major concern were: use of unsatisfactory or unspecified methods of randomization (89% of the trials reported no information), exclusion of some patients from the analysis (53% of the studies), failure to provide evidence of having estimated the number of patients needed to detect a prespecified treatment difference (96% failed to provide such an estimate), and incomplete description of patient characteristics (in 39% of the RCTs information on this item was insufficient). Within this generally unsatisfactory picture we found, however, that those studies adopting a double-masked design and those not excluding patients after randomization followed patients for longer periods of time and assessed treatment effectiveness using a clinically relevant outcome (that is, visual field changes) compared to other studies. CONCLUSIONS: For clinicians to make use of the results of clinical trials, future studies must be adequately designed and conducted. In particular, proper method of randomization, masking of the observers, and inclusion of all randomized patients in the analysis must be used. Of perhaps even greater importance is the need for trials to measure clinically relevant outcomes.

Inappropriate Halsted mastectomy and patient volume in Italian hospitals.

To study whether Halsted mastectomy was used only when properly indicated, a prospective survey was undertaken on the process of care of 985 breast cancer patients seen consecutively at 62 general hospitals in Northern and Central Italy. Overall, 79% of Halsted mastectomies were performed inappropriately. The procedure was less likely to be performed on more educated patients and, other factors considered, on those seen at hospitals with larger volume. We conclude that the measurement of utilization of a surgical procedure for which only a few appropriate indications exist may help identify important relationships between hospital characteristics and quality of surgical care.

Variation in use of breast surgery and characteristics of hospitals' surgical staff.

OBJECTIVE: To assess whether and how the characteristics of hospitals' surgical staff can predict utilization of breast conserving surgery. We anticipated that the use of the procedure was more frequent in hospitals with a greater proportion of younger and female surgeons on their staff. SETTING: Sixty-two hospitals with different level of organization for the care of cancer patient from eight Italian regions were included. DESIGN: We collected information on the personal characteristics (age and sex) of all surgeons on the staff of the participating centers. A concurrent pattern of care study was undertaken, in order to assess the utilization of breast conserving procedures. RESULTS: The use of conservative surgery was 16%, 20%, 31%, in centers with 0%, < 20% and 20-50% female surgeons respectively (p = 0.003), while no associations emerged between the mean age of the surgical staff and the use of conservative surgery. This relationship also held true after adjusting for other patients' characteristics, such as stage of disease, age and presence of comorbidity. However, overall the use of patient-related and staff-related characteristics to predict the number of breast conserving surgeries performed in each area was able to explain only 16% of the observed variability in the use of breast conserving surgery. CONCLUSIONS: Even though in this study a large proportion of the observed variation in the use of breast conserving surgery was still left unexplained, this finding suggests that assessing the relationship between doctors' characteristics and the use of scientific evidence is an important topic for health service research and can be of some help in exploring variations in clinical practice styles.

The influence of patient characteristics on the appropriateness of surgical treatment for breast cancer patients. Progetto Oncologia Femminile.

BACKGROUND: Within the framework of a multi-annual educational intervention sponsored by the Ministry of Health and regional health authorities, patterns of the care delivered to breast cancer patients in Italian general hospitals were monitored in order to identify areas of practice whose quality was in need of improvement. DESIGN: Information on the diagnostic and therapeutic procedures in 63 general hospitals in eight Italian regions performed in 1724 consecutive breast cancer patients were retrospectively gathered from medical records. Quality of care was assessed by a diagnostic and therapeutic score based on the observed degree of compliance with previously established courses of action. RESULTS: The median value of the overall diagnostic and staging score was 60%. About one-third of surgical operations (38%) were inappropriate: one-fourth (24%) of patients with stage I-II disease had unnecessarily radical surgery (i.e., Halsted mastectomy), and limited surgery in patients with small tumors (i.e., < or = 2 cm) was under utilized. Chronological age influenced physicians' behaviour: elderly patients were more likely to have a less intensive diagnostic work-up and less appropriate surgical treatment (with more frequent performance of an unnecessary radical operation and a less frequent utilization of limited surgery), independently of their overall health status. The presence of one or more co-existent diseases was associated with a failure to undergo axillary clearance and with a lower utilization of conservative surgery independently of age. CONCLUSION: In accord with others, this study confirms the existence of a clinically important effect of patient age on diagnostic and therapeutic behaviour and the use of unnecessarily radical surgery procedures. The paper discusses the implications of these findings for the next stage of the educational project, in which practice guidelines will be developed and implemented to improve the quality of care for breast cancer patients.

Patterns of care an Italian diabetic population. The Italian Study Group for the Implementation of the St Vincent Declaration, Società Italiana di Diabetologia, Associazione Medici Diabetologi.

We set out to describe patterns of care of an Italian diabetic population, with reference to the recommendations of the St Vincent Declaration. We investigated different aspects of care received by 2707 patients, of whom 2196 in the charge of 35 Diabetes Outpatient Clinics (DOCs) and 511 cared for by 49 General Practitioners (GPs). Data were collected by interviewing the patients, their physicians and by reviewing medical records. Our data show that diabetes care in Italy differs in many aspects from the recommendations of the St Vincent Declaration. Glycated haemoglobin measurement was lacking in 50% of the patients in the charge of GPs and in 15% of those attending DOCs. While the control of cardiovascular risk factors was satisfactory, information on albumin excretion was not available in one third of the patients. Overall, 79% of the patients had had an eye examination in the previous 12 months. More than one-third of the patients had not received adequate information on different aspects of care, with wide variations according to the setting of care. Forty-two per cent of the patients attending DOCs and 14% of those cared for by GPs practised blood glucose self-monitoring; similarly, insulin therapy self-management was performed by 50% and 19% of the patients attending DOCs and GPs, respectively. Our data call for vigorous efforts aimed at improving the awareness of the potential for reducing major diabetic complications. Therefore, it is essential to promote the incorporation of clearly defined clinical practice guidelines at each level of care.

A meta-analysis of trials on aldose reductase inhibitors in diabetic peripheral neuropathy. The Italian Study Group. The St. Vincent Declaration.

Peripheral neuropathy is one of the most common and disabling long-term sequelae of diabetes mellitus. Aldose reductase inhibitors (ARIs) have been proposed and are increasingly used in many countries for the prevention and treatment of diabetic neuropathy. The aim of this study was to review existing evidence on the effectiveness of ARIs in the treatment of peripheral diabetic neuropathy, with particular reference to the type and clinical relevance of the end point used and to the consistency of results across studies. Thirteen randomized clinical trials (RTCs) comparing ARIs with placebo, published between 1981 and 1993 were included in the meta-analysis. Nerve conduction velocity (NCV) was the only end point reported in all trials. Treatment effect was thus evaluated in terms of NCV mean difference in four different nerves: median motor, median sensory, peroneal motor, and sural sensory. A statistically significant reduction in decline of median motor NCV was present in the treated group as compared to the control group (mean 0.91 ms-1; 95% CI 0.41-1.42 ms-1). For peroneal motor, median sensory, and sural sensory nerves results did not show any clear benefit for patients treated with ARIs. When the analysis was limited to trials with at least 1-year treatment duration, a significant effect was present for peroneal motor NCV (mean 1.24 ms-1; 95% CI 0.32-2.15 ms-1) and a benefit of borderline statistical significance was also present for median motor NCV (mean 0.69 ms-1; 95% CI-0.07-1.45 ms-1). A heterogeneous picture emerged when looking at the results of different studies and serious inconsistencies were also present in the direction of treatment effects among nerves in the same studies. Although the results of 1-year treatment on motor NCV seem encouraging, the uncertainty about the reliability of the end-point employed and the short treatment duration do not allow any clear conclusion about the efficacy of ARIs in the treatment of peripheral diabetic neuropathy.

The effectiveness, safety and epidemiology of the use of acarbose in the treatment of patients with type II diabetes mellitus. A model of medicine-based evidence.

OBJECTIVE: To assess the efficacy, safety and extent of perceived indications of acarbose, a new antidiabetic agent, under routine clinical practice conditions in an unselected Northern Italian population of type II diabetic patients. METHODS: The study population was assigned to three different groups according to the physician's clinical judgement: group A (acarbose considered as an elective treatment); group B (acarbose considered to be of uncertain benefit): group C (acarbose deemed not to be appropriate). Group B patients were randomized either to continue their standard treatment or to add acarbose to it. Patients with type II diabetes mellitus were recruited from 17 diabetes outpatient clinics from one Italian region (Lombardy). A total of 1027 patients were recruited (group A: 283; group C: 494; group B: 250, of whom 124 were randomly assigned to standard treatment + acarbose and 126 to standard treatment alone). Acarbose was administered for 1 year at a median dose of 100 mg 3 times daily. Drug efficacy was evaluated in terms of mean HbAlc, pre- and post-prandial glycaemic values. Additional endpoints were the proportion of patients with HbA1c levels below 8% at the end of the study period and the proportion of subjects who needed a modification in the standard treatment. The safety and tolerability profiles of the drug were also investigated. Data on HbA1c, fasting and post-prandial blood glucose levels were analysed over time using repeated-measures analysis [Generalized Estimating Equation (GEE) models]. RESULTS: The analysis of Group B showed that, after treatment for 1 year, the mean reduction in HbA1c levels in the acarbose group with respect to the control group was 0.30% (95% confidence limits -0.60 +0.02; P = 0.07), while the mean reduction in post-prandial glycaemia was 17 mg-dl(-1) (95% c.l. -33.5 -0.8; P = 0.04). No difference resulted for fasting blood glucose levels. When looking at the baseline HbA1c levels, it emerged that the mean benefit associated with the use of acarbose was 0.14% (95% c.l. -0.6 +0.28; P = 0.5) in patients with HbAlc levels below 8%, 0.28% (95% c.l. -0.6 +0.05; P = 0.09) in those with values between 8% and 9.9% and 0.65% (95% c.l. -1.36 +0.06; P = 0.07) in those with values > or =10%. Only patients treated with diet+/-oral anti-diabetic agents (OAA) benefited from acarbose treatment (mean benefit = 0.37%, 95% c.l. -0.65 -0.08), while no effect was shown for insulin-treated subjects. The proportion of patients with HbA1c below 8% increased from 31% to 44% in the acarbose group and from 40% to 45% in the control group (absolute difference between baseline and end-of-study values = 8.0% in favour of acarbose-treated patients; P = 0.058). Patients treated with acarbose were significantly more likely to undergo a dose reduction in concomitant diabetic treatments compared with the control group; they were also less likely to require an increase in the dose of standard treatment and to start insulin during the study period. One third of the patients could not assume the drug for the whole study period, mainly due to gastrointestinal side-effects. CONCLUSIONS: The design adopted in this study allowed an integrated evaluation of the overall effectiveness of acarbose in clinical practice. The benefits of the drug in an unselected population of non-insulin-dependent diabetes mellitus (NIDDM) patients are significant but of marginal clinical relevance. Only a better definition of the subgroups of patients who are more likely to benefit from long-term treatment, particularly through possible postponement of secondary OAA failure, will allow a reliable definition of the cost-effectiveness of this complementary component of anti-diabetic strategy.