A low-cost approach to prospective identification of impending high cost outcomes.

OBJECTIVES: The overall objective of this study was to define and evaluate patterns of use of medical services in the care of patients with chronic illness that represent circumstances which, if modified, might lead to reduction in risk of acute-level care. METHODS: This was a retrospective observational study. The study population consisted of Kaiser Permanente enrollees at four sites during January 1993 through June 1995, who were 20 to 64 years of age and had two of three chronic diseases (diabetes, circulatory disorders, obstructive pulmonary disorders). Using logistic regression, the effect of primary care visit patterns and therapeutically risky drug combinations on likelihood of hospital admission in a subsequent 3-month period is adjusted for age, gender, and disease state in the prior 12-month period. RESULTS: Enrollees with visits to three or more different primary care physicians were 46% more likely to be admitted than expected (P < 0.01) according to their age, gender, and disease state, and those with therapeutically risky drug combinations were 34% more likely to be admitted (P < 0.01). CONCLUSIONS: The risk adjustment models evaluated in this study defined care processes associated with increased risk of subsequent acute-level services. Those processes may represent nascent acute disease states or suboptimal organization of care delivery. The results of these models can be used to inform changes in organization and delivery of outpatient care that might improve patient outcomes.

Successful outpatient management of acute upper gastrointestinal hemorrhage: use of practice guidelines in a large patient series.

BACKGROUND: Acute upper gastrointestinal hemorrhage is a common reason for hospitalization. Clinical and endoscopic characteristics predict outcome. The aim of this study was to determine the characteristics and outcome of patients with acute upper gastrointestinal hemorrhage cared for without hospitalization. METHODS: One hundred seventy-six consecutive patients in a staff-model health maintenance organization were selected for outpatient care based on absolute endoscopic and non-absolute clinical criteria. Clinical and endoscopic characteristics, British national audit "risk scores," and rates of recurrent bleeding, hospitalization, and mortality were determined. RESULTS: Mean patient age (+/- SD) was 56.4 +/- 16.0 years, and 106 patients (60%) were men. One hundred one (57%) had endoscopy within 2 days of the onset of hemorrhage. The mean initial hemoglobin concentration was 11.7 +/- 2.3 mg/dL. Ninety-seven patients (55%) had a peptic ulcer, and 57 (32%) had a British risk score greater than 2. Hospitalization, recurrent bleeding, and mortality occurred in two (1%), one (1%), and zero (0%) patients, respectively, during 16.0 +/- 10.8 months of follow-up. CONCLUSIONS: Many patients with acute upper gastrointestinal hemorrhage can be safely treated as outpatients using endoscopic and clinical guidelines.

Outpatient care of selected patients with acute non-variceal upper gastrointestinal haemorrhage.

Patients with upper gastrointestinal haemorrhage (UGIH) are usually cared for in hospital. To evaluate the efficacy and safety of outpatient care of selected patients with acute non-variceal UGIH who had endoscopy, we retrospectively analysed 4.5 years' experience of patients treated without hospital admission. We developed practice guidelines for outpatient care, and prospectively studied patients treated during the first 6 months of their use. 78 (8.4%) of 933 patients in the retrospective series and 34 (24.1%) of 141 in the prospective series received outpatient care. The guidelines comprised early notification of a gastroenterologist, urgent endoscopy, clinical, laboratory, and endoscopic criteria for outpatient care, and details of care. In the prospective study patients treated as outpatients were younger than those admitted (52.8 [SE 3.6] vs 63.0 [1.5] years) and had a slightly longer time from onset of bleeding to endoscopy (2.4 [0.2] vs 2.1 [0.2] days). Outpatients were less likely to have alcoholism, other major concomitant disease, syncope or presyncope, or supine tachycardia. Outpatients had higher haemoglobin concentrations than inpatients (125 [4] vs 106 [3] g/L). Most patients in both groups had peptic ulcers. There were no complications in the retrospective series; 1 of the 34 prospective outpatients was admitted with rebleeding. All outpatients survived. The estimated hospital cost saved per outpatient was about $990. A substantial proportion of carefully selected patients with acute non-variceal UGIH can be effectively cared for without admission to hospital.

Hospital care of acute nonvariceal upper gastrointestinal bleeding: 1991 versus 1981.

Developments important to the care of upper gastrointestinal (UGI) bleeding occurred in the 1980s. We compared the features, care, and outcome of patients hospitalized with acute nonvariceal UGI bleeding in 1991 (n = 216) and 1981 (n = 105). Most patient characteristics and the hemoglobin values were similar. Final diagnoses differed, mainly because there were more peptic ulcers and fewer unknown diagnoses in 1991. More patients had endoscopy in 1991 (94%); 92% of endoscoped patients had the procedure within 24 h of admission. The hospital stay of patients without surgery or comorbidity in 1991 was 2.8 +/- 0.1 days, significantly shorter than that in 1981. In 1991 only one half of patients received blood transfusion. The hemoglobin values before transfusion (7.2 +/- 0.1 g%) and at discharge (9.8 +/- 0.1 g%) were lower in 1991; transfused patients received similar amounts of blood in the 2 years. Endoscopic hemostatic therapy was used in 25% of patients with ulcers in 1991 and in none in 1981. Rates of rebleeding (including patients readmitted for recurrent bleeding within 1 week of discharge) and surgery were similar. Mortality was 2.8% in 1991, similar to the 1981 death rate. Because of the excellent 1991 patient outcome, we continue to care for most patients with acute nonvariceal UGI bleeding with urgent endoscopy and short hospitalization.

pH threshold for human duodenal bicarbonate secretion and diffusion of CO2.

Gastric acid enters the proximal duodenum both as free and buffered H+. The procedures in this study were threefold: (a) to determine the pH threshold for duodenal mucosal bicarbonate secretion, iso-osmolar citric acid (pH 2.5-4.0; H+, 1.1 mmol) was infused; (b) to examine the effect of varying acid loads (H+, 0.4-5.1 mmol), citric acid (pH 3.0) was perfused; and (c) to quantitate duodenal diffusion of CO2, citric acid (pH 5.0) gassed with CO2 (PCO2, 0-210 mm Hg) was tested. Basal bicarbonate secretion was similar on each test day, 230 mumol/cm.h. Citric acid at pH 2.5 and 3.0 increased bicarbonate output equally to about 560 mumol/cm.h (similar to 2 mmol of 100 mmol/L HCl); citric acid at pH 3.5 and 4.0 had no effect. Varying the acid load increased bicarbonate output similarly. Duodenal loss of CO2 was minimal (4%) with infusion of 50 mm Hg PCO2 and increased to approximately 25% (15 mm Hg/min) at higher PCO2 values. It is concluded that (a) the pH threshold for human duodenal mucosal bicarbonate secretion is 3.0; (b) a pH-sensitive, rather than an acid load-sensitive, regulatory process exists; and (c) CO2 loss plateaus at 15 mm Hg/min at a PCO2 of 200 mm Hg.

Transient induction of c-jun during hepatic regeneration.

The cellular oncogene c-jun is transiently expressed in cultured cells stimulated to proliferate but has not been identified in normal liver. Because partial hepatectomy results in coordinated cell proliferation in the remaining liver, we investigated c-jun expression after partial hepatectomy in mice. Northern analysis of whole liver mRNA demonstrated a transient increased expression of c-jun within half an hour of the operation. The related gene junB increased only 50%, whereas c-jun expression increased 13-fold compared with sham-operated controls. To determine the cell of origin of the c-jun transcript, both in situ hybridization and Northern analysis of mRNAs from parenchymal and nonparenchymal cell fractions were performed 2 hr after partial hepatectomy. C-jun expression was found in both cell populations. To investigate the mechanism of increased c-jun expression, cycloheximide was given to some animals preoperatively. C-jun induction occurred with cycloheximide alone, but partial hepatectomy further increased c-jun expression, indicating that new protein synthesis was not required for this effect. Furthermore, run-on transcriptional assay demonstrated a twofold increase in c-jun expression. Thus c-jun expression increases after hepatectomy by transcriptional and posttranscriptional mechanisms. Because the extracellular matrix-degrading enzyme transin, which bears the recognition site for jun/AP-1, showed sustained induction after hepatectomy, we speculate that an important function of c-jun expression could be the remodeling of extracellular matrices to accommodate cell proliferation.

Expression of collagen genes in the liver.

Hepatic production of type I collagen is markedly increased in liver cirrhosis. Previous studies using primary liver cell cultures have demonstrated that hepatocytes, lipocytes and endothelial cells are all capable of producing collagen. In this study in situ hybridization and hepatic cell sorting have been used to identify which cells are expressing the type I collagen gene, alpha 1(I), in normal rat liver. Northern blotting of mRNAs from purified hepatic cell populations demonstrated that both hepatocytes and several types of non-parenchymal cells express the collagen alpha 1(I) gene. Calculations based on cell numbers, yields of mRNA, and cellular mRNA concentration demonstrated that the majority of collagen alpha 1(I) mRNA originates from the hepatocytes in the normal liver. Localization of a collagen alpha 1(I) mRNA by in situ hybridization confirmed that both hepatocytes and non-parenchymal cells express this gene. Furthermore, collagen alpha 1(I) gene expression in hepatocytes was obtained by transfecting a reporter gene driven by the collagen alpha (I) 5' regulatory segment in primary liver cell cultures. Future experiments will further characterize the regulation of collagen alpha 1(I) gene expression in the liver.

Tumor necrosis factor-alpha inhibits albumin gene expression in a murine model of cachexia.

The mechanisms responsible for decreased serum albumin levels in patients with cachexia-associated infection, inflammation, and cancer are unknown. Since tumor necrosis factor-alpha (TNF alpha) is elevated in cachexia-associated diseases, and chronic administration of TNF alpha induces cachexia in animal models, we assessed the regulation of albumin gene expression by TNF alpha in vivo. In this animal model of cachexia, Chinese hamster ovary cells transfected with the functional gene for human TNF alpha were inoculated into nude mice (TNF alpha mice). TNF alpha mice became cachectic and manifested decreased serum albumin levels, albumin synthesis, and albumin mRNA levels. However, even before the TNF alpha mice lost weight, their albumin mRNA steady-state levels were decreased approximately 90%, and in situ hybridization revealed a low level of albumin gene expression throughout the hepatic lobule. The mRNA levels of several other genes were unchanged. Hepatic nuclei from TNF alpha mice before the onset of weight loss were markedly less active in transcribing the albumin gene than hepatic nuclei from control mice. Therefore, TNF alpha selectively inhibits the genetic expression of albumin in this model before weight loss.