Search

Aim: To evaluate whether the feedback of laboratory use and cost data to clinicians modifies their request behaviour., Methods: Over two years the effect of monthly feedback of clinical chemistry test use and revenue expenditure to three consultant physicians on their clinical chemistry and haematology requesting patterns was evaluated. Two physicians who received no information served as controls., Results: Feedback over one year led to an immediate and sustained decrease of 15%, 27%, and 21% in clinical chemistry requests (p less than 0.01), tests (p less than 0.001), and revenue expenditure (p less than 0.001), respectively, and a 10% reduction in haematology tests (p less than 0.05) per outpatient visit. These changes persisted in the six months after the feedback was stopped., Conclusions: These results suggest that feedback of laboratory data to clinicians modifies their request behaviour and that supplying clinicians with information on what they do can influence the way they make decisions.

In a 1 year prospective study we evaluated the effect of feedback of laboratory data on the requesting behaviour of physicians in general medicine. Data on within-hours and out-of-hours clinical chemistry laboratory usage and revenue expenditure for inpatients and outpatients, expressed in terms of clinical workload, were supplied monthly to a group of three consultant physicians in general medicine. With these data the physician could monitor his performance over a period of time and compare it with that of his peers. Two consultants in general medicine who received no information served as controls. Over a period of 6 months, there was a 25%, 13% and 18% decrease in tests (P less than 0.01), requests (P less than 0.05) and revenue expenditure (P less than 0.01) per outpatient visit, respectively, in the intervention group of physicians following the introduction of feedback when compared to their baseline period and to the control group. The decrease (P less than 0.01) was in the commonly requested and 'seemingly cheap' tests. There was no significant change in laboratory use and expenditure on inpatients. The feedback of laboratory data was acceptable to the physicians, raised their awareness of laboratory usage and costs and decreased laboratory workload and expenditure.

The effect of cimetidine on the risk of further bleeding shortly after acute gastrointestinal haemorrhage from peptic ulcer was investigated in a double-blind randomised trial. 34 patients were given cimetidine and 32 placebo, the two groups being matched for age, sex, and severity of haemorrhage. Further bleeding within a week of admission was detected clinically in 8 patients on cimetidine and 15 on placebo. Cimetidine had no effect on bleeding from duodenal ulcer, but only 2 of 14 patients with gastric ulcer treated with cimetidine bled again, compared with 10 of 19 patients on placebo. Cimetidine, therefore, may help to prevent haemorrhage from gastric ulcer but not duodenal ulcer.

High-density lipoprotein (H.D.L.) cholesterol has been measured by the 'Autoanalyzer', and apolipoproteins A-I, A-II, and B by an immunochemical method, in 100 patients with peripheral vascular disease (P.V.D.) and in 93 age and sex matched controls with an approximately similar prevalence of hyperlipidaemia. The patients with P.V.D. had significantly lower levels of the H.D.L. apolipoproteins (especially of apo A-I) than the controls. Further analysis of the data showed low H.D.L. levels to be related to the presence (but not to the severity) of the arterial disease and to be independent of concurrent hyperlipidaemia and smoking habits. The changes in H.D.L. apoproteins were not so clearly reflected by H.D.L.-cholesterol measurements in the same patients, possibly because of methodological reasons. It is therefore suggested that studies relating serum-H.D.L. to arterial disease may be more informative if both the lipid and protein portions of these lipoproteins are measured.

Investigation of the uptake and metabolism of drugs by organs such as the liver may allow assessment of specific aspects of organ function. Rifampicin, when orally administered, is transported into the hepatocyte from portal blood and thence passes, with its deacetylated metabolite, into the systemic circulation and into bile. This paper reports an investigation of the pharmacokinetics of a sub-therapeutic oral dose of rifampicin in healthy subjects, in patients with cirrhosis and in subjects with Gilbert's syndrome. The areas under the plasma concentration curves (AUC) in patients with cirrhosis were significantly greater than in healthy subjects. Subjects with Gilbert's syndrome had decreased AUCs compared with healthy subjects and were clearly distinguished from patients with cirrhosis. Rifampicin concentration in serum was measured by HPLC using a novel direct injection technique.

Endoscopic variceal sclerosis is effective at eradicating oesophageal varices and prolonging survival, but early rebleeding before varices have been obliterated remains a problem. A randomized controlled trial was therefore conducted to determine whether more rapid variceal obliteration and hence a lower morbidity and mortality in the first month could be achieved by compressing the varices after the first injection of sclerosant. Forty patients bleeding from previously untreated varices were studied. There was no demonstrable benefit from post-sclerosis variceal compression in terms of early death from rebleeding (compression, 3 of 19; no compression, 3 of 21 in the first month), total number of patients rebleeding (compression, 5 of 19; no compression, 6 of 21 in the first month), or speed of variceal obliteration (percentage of variceal columns obliterated at 1 month: compression, 13%; no compression, 26%). This study shows that post-sclerosis variceal compression by means of the Williams overtube and Sengstaken tamponade does not improve the efficacy of endoscopic variceal sclerosis.

In a randomised, double blind, parallel group study in patients with symptomatic gastric ulcer (94% greater than or equal to 5 mm diameter), 102 received omeprazole 20 mg om and 87 cimetidine 400 mg bd. After four weeks 73% and 58% (p less than 0.05) respectively had healed (eight weeks: 84% and 75%, ns). After four weeks, a greater proportion (81%) of omeprazole treated patients was symptom free than of those receiving cimetidine (60%; p less than 0.01). Over the first two weeks, patients receiving omeprazole had less day pain, less night pain and took fewer antacids than those receiving cimetidine (all p less than 0.05). The difference between omeprazole and cimetidine was not appreciably affected by age, smoking, size of the ulcer and trial centre. Tolerability was similar in the two treatment groups. In the treatment of symptomatic gastric ulcer, omeprazole relieves the symptoms more quickly than cimetidine and heals a greater proportion of ulcers within four weeks.

Seven cases of diffuse interstitial lung disease (DILD) are reported with an unequivocal temporal relationship between the development of the lung disease and treatment with gold (6 cases) and penicillamine (1 case). They were characterised clinically by the sudden onset of dyspnoea and crepitations and radiologically by diffuse bilateral pulmonary shadowing. Most showed evidence of hypersensitivity such as eosinophilia, a raised serum IgE level in response to gold, proteinuria, thrombocytopenia, or an immediate postinjection reaction. DILD is a serious complication of treatment with gold and penicillamine that is commoner than generally realised.