Managing complex patients on a medical psychiatric unit: An observational study of university hospital costs associated with medical service use, length of stay, and psychiatric intervention

Objective

Although there is a suggestion that the medical psychiatric unit (MPU) may reduce length of hospital stay (LOS), little is known about costs in terms of medical service use and psychiatric interventions in MPU care.

Method

A record linkage study was conducted, linking cost data of hospital medical service use, LOS, and hospital psychiatric interventions to patients admitted to the MPU of the Maastricht University Medical Centre (MUMC) between 1998 and 2004. The data set was analyzed to enable comparison between cost changes of the same complex patient population following either MPU index admission or index admissions to reference MUMC medical wards.

Results

Comparisons revealed lower costs of medical service use in favor of the MPU (−€104; 95% CI −€174 to −€35; P<.01). However, cost of psychiatric intervention and cost of LOS were higher after MPU admission (respectively, +€165; 95% CI +€25 to +€305; P<.05; and +€202; 95% CI +€170 to +€235; P<.001). Total costs were higher after MPU admission compared to medical ward admission (+€263; 95% CI +€68 to +€458; P<.05). These differences were not moderated by somatic diagnosis or previous pattern of admissions.

Conclusion

The findings suggest that patients at the interface of psychiatric and somatic morbidity are diagnosed and treated adequately at the MPU, leading to a decrease in medical service use and an appropriate increase in exposure to psychiatric interventions. These results are specifically generalizable to MPUs with a focus on psychosomatic conditions, for instance, somatoform disorders or affective disorders with comorbid somatic diseases. However, failure to show cost savings in terms of LOS compared to medical wards outweighs cost-benefit derived from lower medical service use, suggesting that MPU activities may gain in cost-effectiveness if shifted more to outpatient psychosomatic care solutions.     

Motor and sensory function of the proximal stomach in reflux disease and after laparoscopic Nissen fundoplication

OBJECTIVE: After Nissen fundoplication, dyspeptic symptoms such as fullness and early satiety develop in >30% of patients. These symptoms may result from alterations in proximal gastric motor and sensory function. METHODS: We have evaluated proximal gastric motor and sensory function using an electronic barostat in 12 patients after successful laparoscopic Nissen fundoplications (median follow-up; 12 months). Twelve age- and gender-matched patients with severe gastroesophageal reflux disease (GERD) and 12 healthy volunteers served as controls. Studies were performed in the fasting state and after meal ingestion. Gastric emptying tests were performed in all patients. Vagus nerve integrity was measured by the response of pancreatic polypeptide (PP) to insulin hypoglycemia. RESULTS: Minimal distending pressure and proximal gastric compliance were not significantly different between post-Nissen patients, GERD patients, and healthy controls. Postprandial relaxation of the stomach, however, was significantly (p < 0.05) reduced post-Nissen (267 +/- 34 ml), compared with controls (400 +/- 30 ml) and GERD (448 +/- 30 ml). Postprandial relaxation was significantly (p < 0.01) prolonged in GERD patients. Postprandial relaxation of the stomach correlated with gastric emptying of solids (r = 0.62; p = 0.01). Gastric emptying of solids became significantly (p < 0.05) faster after fundoplication. Postprandial fullness was significantly (p < 0.05) increased in the operated patients. CONCLUSIONS: Post-Nissen patients have a significantly reduced postprandial gastric relaxation and significantly accelerated gastric emptying, which may explain postoperative dyspeptic symptoms. The abnormalities result from fundoplication and not from vagus nerve injury or reflux per se, because in reflux patients gastric relaxation and gastric emptying are prolonged.     

Gastro-intestinal surgery and gastroenterology. V. Chronic pancreatitis: gastroenterologic aspects

Chronic pancreatitis causes irreversible damage to the pancreatic parenchyma and ultimately leads to exo- and endocrine insufficiency. In the Western world, alcohol is the main cause of chronic pancreatitis; part of the patients with idiopathic or alcoholic pancreatitis conceivably have a raised sensitivity to the toxic effects of alcohol because of a certain genetic predisposition. The most striking symptom, severe recurrent or persistent pain is often difficult to manage. Various forms of imaging examination provide complementary information on lesions of the pancreatic parenchyma, the pancreatic duct, the bile ducts and adjacent structures; ultrasonoscopy is the technique of first choice in case of suspicion of a pancreatic disorder. Endoscopic therapies are booming: sphincterotomy, calculus extraction, insertion of an endoprosthesis in the pancreatic duct and drainage of pseudocysts (transpapillary or through the jejunum or stomach).     

The effect of motilin on the rectum in healthy volunteers

AIMS: The role of motilin in the regulation of upper gastrointestinal (GI) motility is well defined. However, little is known about the effects on the distal GI tract. To investigate the effect of exogenous motilin on rectal function, barostat measurements in the rectum were performed and lower abdominal symptoms were scored. METHODS: Eight fasted, healthy volunteers were infused intravenously with synthetic motilin or placebo over 90 min in a double-blind, randomized, cross-over design. Rectum volume was measured with a barostat device during constant pressure and during isobaric distensions. Lower abdominal symptoms were scored by visual analogue scales. Plasma motilin concentrations were measured by radioimmunoassay. RESULTS: Baseline rectum volumes were similar between treatments: 185 +/- 62 mL (motilin) and 136 +/- 41 mL (placebo). During the constant pressure procedure, motilin increased rectum volume [area under the effect curve (AUEC)] by 6%[95% confidence interval (CI) -3, 16] of baseline, compared with placebo. During isobaric distensions motilin increased rectum volume (AUEC) by 43 mL (95% CI 0.4, 85; P < 0.05) and compliance by 10 mL mmHg-1 (95% CI 0.3, 20; P < 0.05) relative to placebo. Motilin did not induce changes in the sensation of rectal feelings. CONCLUSION: Exogenous motilin increased rectal compliance in healthy volunteers, without affecting rectal sensations.     

Effects of somatostatin on proximal gastric motor function and visceral perception

Background:

Somatostatin affects gastrointestinal motility and secretion and visceral sensation, but little is known about its effects on the proximal stomach.

Aim:

To evaluate the effects of somatostatin on proximal gastric motor function and perception of symptoms.

Methods:

Six healthy subjects participated in two experiments performed in random order during continuous intravenous infusion of saline or somatostatin (250 μg/h). Proximal gastric motor function was evaluated using a barostat. We performed pressure and volume distensions and a barostat procedure (minimal distending pressure + 2 mmHg). Symptoms were evaluated at regular intervals using visual analogue scales (VAS).

Results:

Neither minimal distending pressure nor gastric fundal tone were significantly different between somatostatin and saline. Pressure–volume curves during distensions were not influenced by somatostatin. However, phasic volume waves were significantly (P < 0.001) reduced by somatostatin, and somatostatin significantly (P < 0.05) reduced symptom perception of fullness and abdominal pressure during stepwise distensions.

Conclusions:

Continuous infusion of somatostatin does not influence gastric compliance but it inhibits phasic volume waves and significantly reduces visceral perception.

     

Prospective study of the effect of laparoscopic hemifundoplication on motor and sensory function of the proximal stomach

BACKGROUND: Some 30 per cent of patients develop dyspeptic symptoms following antireflux surgery. These symptoms may result from alterations in the motor and sensory function of the proximal stomach. METHODS: Proximal gastric motor and sensory function was studied with an electronic barostat in 12 patients with reflux who underwent laparoscopic hemifundoplication. In addition, 24-h pHmetry, gastric emptying (scintigraphy) and vagus nerve integrity (pancreatic polypeptide response to hypo-glycaemia) were assessed. Fifteen healthy volunteers served as controls. RESULTS: Laparoscopic hemifundoplication significantly decreased total acid exposure time (P < 0.05). Vagus nerve function remained intact in all but one patient. The mean(s.e.m.) lag phase for emptying of solids was significantly shorter after operation than before (15(3) versus 21(3) min; P < 0.05). Proximal gastric compliance was not significantly different before and after fundoplication. However, mean(s.e.m.) postprandial relaxation was significantly reduced (P < 0.05) after hemifundoplication compared with the value before operation (3341(1105) versus 12 763(3616) ml over 90 min) and in controls (14 567(2358) ml over 90 min). Postprandial fullness was significantly increased after hemi-fundoplication (P < 0.05). Postprandial gastric relaxation correlated with the lag phase for emptying of solids (r = 0.55, P < 0.02). CONCLUSION: After hemifundoplication, proximal gastric compliance is not altered but postprandial relaxation is impaired and associated with sensations of fullness and shorter duration of the lag phase for emptying of solids.     

Effect of peptide YY on pancreatico-biliary secretion in humans

OBJECTIVE: The negative feedback on pancreatico-biliary secretion induced by ileal nutrients has been well documented but the role of the distal gut peptide YY (PYY) as mediator is less well defined. We determined the effect of PYY on basal, sham feeding and feeding-stimulated pancreatico-biliary secretion in humans. MATERIAL AND METHODS: Eight healthy volunteers participated in a placebo-controlled, double-blind, randomized study with intravenous infusion of a physiological dose of PYY. Subjects were intubated with a naso-jejunal tube. Duodenal samples were aspirated continuously while a recovery marker perfused the duodenum. Outputs were measured in the basal state, in response to modified sham feeding (MSF) and during jejunal perfusion with a liquid meal. Blood samples were drawn for plasma cholecystokinin (CCK) and PYY measurements. RESULTS: PYY infusion significantly (p<0.01) increased plasma PYY, from 16+/-2 to 55+/-6 pM. Basal outputs of amylase, lipase, trypsin and bilirubin were not affected by PYY. In the placebo experiment, MSF significantly increased amylase, lipase, trypsin and bilirubin outputs. However, during PYY infusion lipase, trypsin and bilirubin outputs did not significantly increase in response to MSF. Jejunal feeding, however, significantly (p<0.01) increased the outputs of bilirubin and enzymes equally during both the placebo and PYY infusions. CONCLUSIONS: Infusion of PYY to physiological plasma levels exerts an inhibitory effect on pancreatico-biliary secretion during the cephalic phase of digestion, but not in the basal state or during the intestinal phase. PYY suppresses pancreatico-biliary secretion during the cephalic phase but not during the intestinal phase of nutrient digestion.     

Role of cholecystokinin in relaxation of the proximal stomach

Ingestion of a meal causes proximal gastric relaxation (accommodation). The magnitude of accommodation is related to the fat content of the meal. A role for cholecystokinin (CCK) has been suggested. However, under fasting conditions intravenous CCK to postprandial levels does not induce a similar accommodation. This study further explores the role of CCK in accommodation. A gastric barostat was used in eight healthy persons to study accommodation in response to a carbohydrate meal with intravenous CCK (CH-CCK), carbohydrate meal with intravenous placebo (CH-placebo) and a fat rich meal with intravenous placebo (FAT). VAS scores for satiety and plasma CCK levels were obtained. In the first postprandial hour the FAT meal induced a relaxation of 112 +/- 29 ml, the CH-CCK meal 49 +/- 36 ml and the CH-placebo meal 12 +/- 32 ml (FAT versus CH-placebo P = 0.03; FAT versus CH-CCK P = 0.09). In the second postprandial hour, intragastric bag volume returned to baseline with all meals. The FAT meal had the most pronounced effect with respect to satiety, CH-placebo the least. In the first postprandial hour, plasma CCK levels increased with the CH-CCK and FAT meals but not with the CH-placebo meal; in the second postprandial hour, levels remained elevated with the CH-CCK meal. It is concluded that a carbohydrate meal with exogenous CCK does not induce fundic relaxation, whereas a fat-rich meal (endogenous CCK) does, despite similar plasma CCK levels.