Technology assessment in healthcare: a review and description of a "best practice" technology assessment process.

BACKGROUND: Technology assessment has become a rapidly growing component of the healthcare system. It has assumed a functional role in operational settings and is rapidly impacting decisions involving purchasing, coverage, and reimbursement. This review is intended to assist the healthcare decision maker in considering the application of technology assessment in healthcare, so as to maximize the efficiency of future purchasing decisions. METHODS: This "best practice" was synthesized after identifying key institutions performing technology assessment in healthcare and analyzing their working processes, including literature review, consensus panel discussions, and expert opinion. RESULTS: We describe this best practice on a reiterative loop that consists of five processes: awareness, strategic appropriateness, analysis versus need, acquisition and implementation, and reassessment. Typical barriers to adoption of technology assessment are also identified and discussed. CONCLUSIONS: This review suggests a common terminology for the core processes involved in technology assessment, thereby facilitating a more uniform understanding among the different components of the healthcare system (i.e., payer, provider, and society) while recognizing their different perspectives.

Relationship between practice guidelines, formulary management, and pharmacoeconomic studies.

Pharmacy and therapeutics committees can use pharmacoeconomic and outcome studies as tools to evaluate and implement clinical guidelines for patient care. Results of studies help optimize the clinical effects and control the costs of drug therapy. Such data also assist in positioning products in competitive environments. A four-part classification of research studies is offered as an aid to strategic research planning.

Clinical experience with timolol maleate monotherapy of hypertension.

Beta-adrenergic blocking drugs are gaining acceptance as initial therapy for patients with mild to moderate hypertension. In a postmarketing surveillance study, 5,190 hypertensive patients received timolol maleate monotherapy and were evaluated by 1,355 physicians. A total of 1,057 patients did not complete the study: 28% of these patients experienced an adverse event. Mean systolic and diastolic blood pressure readings were reduced 20 and 13 mm Hg, respectively. Mean diastolic blood pressure was reduced 11% for patients with mild hypertension; larger mean reductions were noted for patients with moderate (17%) and severe hypertension (22%). The effect in black and elderly patients was less than in other groups. Although 22% of all patients experienced an adverse event, less than 2.2% of all patients experienced events related to beta-adrenergic blockade, ie, respiratory difficulty, heart failure, bradycardia, and cold extremities. Fatigue, dizziness, and nausea were the most frequently reported adverse events requiring discontinuation of therapy. Timolol monotherapy is a well-tolerated and effective treatment for a broad range of hypertensive patients.

Postmarketing study of timolol-hydrochlorothiazide antihypertensive therapy.

A postmarketing surveillance study was conducted to determine the safety and efficacy of a fixed-ratio combination containing 10 mg of timolol maleate and 25 mg of hydrochlorothiazide, administered twice daily for one month to hypertensive patients. Data on 9,037 patients were collected by 1,455 participating physicians. Mean systolic blood pressure decreased 25 mmHg and mean diastolic blood pressure declined 15 mmHg after one month of timolol-hydrochlorothiazide therapy (P less than 0.01, both comparisons). Age, race, and sex appeared to have no influence on the decrease in blood pressure. The antihypertensive effect of the drug was greater in patients with more severe hypertension. Overall, 1,453 patients experienced a total of 2,658 adverse events, the most common being fatigue, dizziness, and weakness. Treatment in 590 patients was discontinued because of adverse events.

Pharmacokinetic study of sisomicin in humans.

Detailed analyses of the pharmacokinetics of sisomicin administered at doses of 25, 50 and 100 mg intravenously and intramuscularly to healthy volunteers established that the drug is handled by a two-compartment open model system with a disposition (elimination) half-life of 2.6 hr. The kinetic estimates over this dose range are linear and independent of dose and were verified by a 60-min infusion experiment in which dose and the maximum serum concentration achieved (5 microgram/ml) were predicted correctly. Sisomicin was rapidly distributed to the tissue compartment, and equilibrium between the central and the tissue compartment was established by 30 min after dosing. Renal clearance (55 ml/min) of sisomicin was about 30% less than total body clearance (78 ml/min). Total urinary excretion of sisomicin during a 24-hr period following drug administration was about 70% of the dose. The disposition kinetics of sisomicin following intramuscular administration are similar to those obtained following rapid intravenous administration. Intramuscular bioavailability of sisomicin for the doses of 25, 50, and 100 mg was greater than 95%. Based on these results, various initial loading infusion doses and maintenance infusion rates were calculated to provide specific desired peak and steady-state serum sisomicin concentrations rapidly. The purpose was not to expose patients to potentially toxic high peak concentrations of drug while maintaining these concentrations during the current therapeutic dosing intervals of 8 to 12 hr.

Pharmacology of enantiomers and (-) p-OH metabolite of indacrinone.

Indacrinone, a racemic mixture, is a loop-blocking diuretic with effects on uric acid elimination that differ from those of furosemide. A series of studies in healthy men was undertaken to characterize the pharmacologic activity of the positive (+) and negative (-) enantiomers (E) of indacrinone and its (-) p-OH metabolite, (-) MET. All subjects were on sodium- and potassium-controlled diet; each experiment was similar in design and included placebo and positive controls. Oral (-)E and (-)MET exerted dose-related natriuretic and diuretic effects; intravenous doses of (-)E were more effective than (-)MET. The effects of (-)E and (-)MET on serum uric acid were the same as those reported with indacrinone. After (-)E, both (-)E and generated (-)MET appeared to contribute to the natriuresis. (+)E induced dose-related decreases in serum uric acid up to 24 hr after dosage; at the higher doses of (+)E, the hypouricemic effects were of the order of those after 500 mg of probenecid. Thus, indacrinone is a novel loop diuretic with enantiomers and a (-)MET, each of which has a different pharmacologic profile.

Comparison of oral indacrinone with furosemide.

The oral dose response and time course of action of indacrinone was compared with that of furosemide in six healthy men on a sodium and potassium-controlled diet. The single doses were 5, 10, 20, 40, and 80 mg indacrinone and 20, 40, and 80 mg furosemide. Diuretic, natriuretic, and kaliuretic effects revealed that indacrinone was more potent, had a longer duration of action, and induced a greater sodium for equivalent potassium loss during its period of peak activity than furosemide. During the 8 hr after drug, all doses of indacrinone decreased serum uric acid levels and increased uric acid clearance while furosemide generally increased serum uric acid and decreased uric acid clearance. After 24 hr, serum uric acid and uric acid clearance were the same for the two drugs. A rise in plasma renin activity was observed 2 hr after an 80-mg dose of furosemide but not after indacrinone.

Effect of orally administered probenecid on the pharmacokinetics of cefoxitin.

To characterize the effect of orally administered probenecid on the pharmacokinetics of cefoxitin in healthy male volunteers, we administered to one group of six subjects 2 g of cefoxitin by intravenous (i.v.) bolus either alone, with 1 g of probenecid concomitantly, or when 1 g of probenecid was administered 1 h previously by using a crossover design. Likewise, we administered to a second group of six subjects 2 g of cefoxitin intramuscularly (i.m.) together with 1 and 2 g of probenecid. Probenecid increased the mean terminal half-life and the area under the serum cefoxitin concentration-time curve (AUC0-24) and decreased renal clearance, but did not alter the volume of the central compartment or the total urinary recovery of i.v.-administered cefoxitin; pretreatment with probenecid produced a greater increase in cefoxitin AUC0-24 and a constant decrease in renal clearance compared to concomitant probenecid. The AUC0-24 after i.m.-administered cefoxitin was greater after 2 g than 1 g of probenecid; the AUC0-24 after i.v.-and i.m.-administered cefoxitin was similar after 1 g of probenecid was given concomitantly. Cefoxitin AUC0-24 was increased further when 1 g of probenecid was given before i.v.-administered cefoxitin or when 2 g of probenecid was given with i.m.-administered cefoxitin. The effect of probenecid was related to both timing and dose.

Pharmacokinetics and comparative pharmacology of cefoxitin and cephalosporins.

Features of the distribution, metabolism, elimination, and pharmacokinetics of the cephalosporins and cefoxitin must be considered when concentrations of these drugs in biological fluids are interpreted. The extensive (approximately 86%) binding of cefazolin to plasma protein may account for the smaller volume of distribution and slower rate of renal clearance than are observed for cefoxitin, which is less extensively (73%) bound to protein. Results of microbiological assays of drug in urine may be influenced by the extent of metabolism of the drugs, which is 33% for cephalothin but less than 2% for cefoxitin. Elimination of cephalosporins and cefoxitin occurs by both glomerular filtration and tubular secretion and can be inhibited by the concurrent administration of probenecid. The pharmacokinetics of cefoxitin may be described by a linear, two-compartment, open model that has been used to predict levels of drug achieved in serum and urine after various dose regimens, including administration by intravenous bolus or infusion. The bioavailability of intramuscularly administered cefoxitin is equivalent to that of intravenously administered cefoxitin and is 90% complete within 3-4 hr after the dose is given.

The diuretic response in normal volunteers to a new benzylamine, 2-aminomethyl-4-(1,1-dimethylethyl)-6-iodophenol HCl.

A phase I dose-response study of 2-aminomethyl-4-(1,1-dimethylethyl)-6-iodophenol HCl (MK-447) was performed with the following oral doses: 6.25, 12.5, 25, 50, and 100 mg. Each volunteer served as his own control. The study was carried out in double-blind fashion on a 5-Gm Na and K diet with a minimum 2000 ml fluid intake. Urine was fractionated and analyzed for sodium, chloride, potassium, calcium, uric acid, and volume. Comparisons (MK-447 minus control values) of the 24-hour total sodium, calcium, potassium, and volume excretion rates at 6.25, 25 and 100 mg MK-447 were as follows: sodium, 195, 345, and 528 muEq/min; chloride, 191, 365, and 756 muEq/min; potassium, 77, -3, and 65 muEq/min; and volumes, 1, 3.4 and 11.7 ml/min. MK-447 did not alter calcium excretion. Uric acid excretion was observed to decrease as the dose of MK-447 was increased, however, the serum uric acid level always remained within normal limits. MK-447 did not alter the physiologic parameters but did produce symptoms of volume contraction at 100 mg. Because no further dose increase was attempted, a plateau in the dose-response curve was not reached. Comparison of 100 mg MK-447 with 80 mg oral furosemide revealed similar potency and a somewhat longer duration of action for MK-447.

Evaluation of the prison inmate as a subject in drug assessment.

The reasons for exclusion of prisoners from research studies on drugs were based mostly on a relatively limited group of laboratory parameters which could have been detected using a simple battery of screening tests. The answers to a medical history form added little to the evaluation of either the prisoner or student groups, were probably very unreliable, and could be just as well confined to a few selected questions regarding drug history as a matter of record. Students gave appropriate responses to a mood scale measurement test while prisoners characteristically did not comply. Because of a combination of various institutional and sociological factors, prisoners probably represent a special subgroup of research volunteers whose health status may not be representative of the total "healthy" population. They are unlikely to give accurate or reliable responses in testing situations which rely upon reporting of the subjective effects of drugs with regard to tolerance or pharmacologic effect. Studies of investigational drugs where the likelihood of potentiaal risk is significant should be avoided in such populations unless compliance has been assessed adequately.

Effect of clonixin and aspirin on platelet aggregation in human volunteers.

The effect of clonixin and aspirin on platelet function was assessed in healthy volunteers. Both drugs inhibited secondary platelet aggregation and prolonged bleeding time, but the effect of clonixin was significantly less than that of aspirin. Hemorrhagic complications are less likely after clonixin than after aspirin.