- Email: [email protected]
PR2 PERCEIVED UTILITY OF HEALTH STATES INDUCED BY CHRONIC HEPATITIS B: ESTIMATES FROM UNINFECTED PERSONS IN CANADA
Abstracts
A18 MC8 THE APPLICATION OF BAYESIAN METHODS TO EVALUATE THE IMPACT OF TIME-DELAY ON THE COST-EFFECTIVENESS OF PRIMARY ANGIOPLASTY FOR ACUTE MYOCARDIAL INFARCTION
Palmer SJ, Asseburg C, Bravo-Vergel Y, Fenwick E, Sculpher M University of York, York, North Yorkshire, UK OBJECTIVES: To apply Bayesian approaches to evaluate the impact of time-delay on the cost-effectiveness of primary angioplasty versus hospital-administered thrombolysis for patients with acute myocardial infarction. METHODS: A probabilistic model was developed to evaluate the life-term cost-effectiveness of primary angioplasty. A health service perspective was adopted with outcomes estimated using Quality-Adjusted Life Years (QALYs). Evidence on short-term event rates was obtained by updating a recent meta-analysis of randomised trials comparing angioplasty with thrombolysis (Lancet 2003; 361:13–20). Data from 22 trials were combined using Bayesian hierarchical modeling and meta-regression. This approach enabled the simultaneous estimation of posterior distributions and correlation structure for: 1) major cardiovascular events including death, reinfarction and stroke, and 2) different time endpoints (4–6 weeks and 6 months). The impact of PCI time-delay to treatment was analysed using mean time delay compared to thrombolysis as a covariate of the random effects model. Long-term costs and QALYs were estimated using a Markov model populated from UK registry data. RESULTS: Based on the average time-delay reported in the trials (54 minutes), primary angioplasty resulted in a mean gain of 0.29 QALYs and an additional cost of £2680 compared to thrombolysis. The associated incremental costeffectiveness ratio (ICER) was £9241 per QALY. At a threshold of £20,000 per QALY there was a 90% probability that primary angioplasty was cost-effective. Adjusting the time-delay resulted in considerable variation in the cost-effectiveness estimates. For a shorter delay of 30-minutes the ICER was £6,850 per QALY; increasing the time-delay to 90-minutes resulted in a marked increase in the ICER to £64,750 per QALY (98% and 36% probability cost-effective at £20 k, respectively). CONCLUSIONS: This study demonstrates the policy importance of time-delay when considering the cost-effectiveness of primary angioplasty, and the advantages of using Bayesian approaches to model multiple endpoints, treatment effects and baseline event rates. Preference and Utility Based Patient Reported Outcomes PR1 EVALUATION OF PREFERENCES IN GENITAL HERPES TREATMENT USING A DISCRETE CHOICE EXPERIMENT
Scalone L1, Watson V2, Ryan M2, Kotsopoulos N3, Patel R4 Center of Pharmacoeconomics, University of Milan, Milan, Italy; 2 Health Economics Research Unit, University of Aberdeen, Aberdeen, UK; 3GlaxoSmithKline, Brentford, UK; 4Department of Genito Urinary Medicine, Royal South Hampshire Hospital, Southampton, UK OBJECTIVES: Genital herpes (GH) is widespread, psychologically disabling and costly. GH is characterized by episodic outbreaks of genital and perigenital vesicles and ulcers, and is one of the most prevalent sexually transmitted diseases in the world today. The seroprevalence of herpes simplex virus (HSV) type-2 ranges from 13% to 40% in the United States and from 7% to 16% in Europe. Often GH is not adequately treated, with consequences on benefits and patients’ well being, and potentially important economic repercussions. GH patients play an important role in decision-making for their management strategies, and
1
their preferences can be crucial to improve treatment outcomes and resource allocation. This study evaluated patients’ preferences in GH treatment. METHODS: Preferences were elicited from 157 patients, recruited from the Harris Poll panel, who completed an online Discrete Choice Experiment (DCE) questionnaire. DCE data was analysed using multinomial logit regression models to estimate respondents’ preferences for GH medical treatment. RESULTS: Respondents, from the US (87.9%) and UK (12.1%), had a median age of 43 years (21–65), and 83.4% were women. Overall, respondents preferred medical treatment to no treatment of GH, with suppressive treatment preferred to episodic, assuming everything else equal. Willingness to pay (WTP) was $38.2/month for episodic treatment and $60.6/month for suppressive treatment, assuming everything else equal. Furthermore, respondents were WTP $56.6/month for a 1% reduction in the “chance of a GH recurrence” in 12 months. In terms of treatment, respondents’ preferences indicated that generally patients preferred the treatment they currently receive. Globally, more patients would choose to be treated (74.3%) than those actually treated (56.2%). CONCLUSIONS: Patients prefer suppressive treatment of GH. Such preferences are influenced by experience, knowledge and awareness of available options. The estimated model suggests that more patients than those actually receiving drug therapies would consider being treated. PR2 PERCEIVED UTILITY OF HEALTH STATES INDUCED BY CHRONIC HEPATITIS B: ESTIMATES FROM UNINFECTED PERSONS IN CANADA
Levy A1, Tafesse E2, Mukherjee J2, Iloeje U2, Poissant L3, Briggs AH4 1 Oxford Outcomes, Vancouver, Bristis Columbia, Canada; 2BMS, Wallingford, CT, USA; 3McGill University, Montreal, QC, Canada; 4 University of Glasgow, Glasgow, UK OBJECTIVE: With an estimated prevalence in Canada between 206,000 and 280,000 persons infected with the virus, persons with hepatitis B typically progress through increasingly severe disease states before death. The objective was to estimate preferences (ratings and utility weights) for six hepatitis B-related disease states among uninfected persons. METHODS: Three hepatologists characterized the typical effects of symptoms on health-related quality of life in terms of symptoms, frequency of tests, hospitalizations, procedures, and dimensions of health such as pain, ability for self-care, activities of daily living, psychological well-being and future outlook. From a convenience sample of 100 uninfected persons in Canada, we elicited ratings using a visual analog scale based on a ‘feeling’ thermometer and standard gamble utility weights using probability wheels with 2-color pie charts for the relative probabilities of perfect health and death. RESULTS: The mean age was 42 years (standard deviation (SD): 14.8, range: 18 to 80 years) and 29% were male. Mean utilities were: 0.80 (95% confidence interval (CI): 0.76; 0.83) for chronic hepatitis B; 0.79 (CI: 0.75; 0.83) for compensated cirrhosis; 0.41 (CI: 0.35; 0.46) for decompensated cirrhosis; 0.69 (CI: 0.65; 0.74) for the first year after liver transplant; 0.77 (CI: 0.73; 0.80) for subsequent years after liver transplant; and, 0.45 (CI: 0.40; 0.51) for hepatocellular carcinoma. The values using the visual analog scale were lower than utilities elicited using the standard gamble, but the relative ranking of each health state did not change. CONCLUSION: All six hepatitis B-related health states are associated with substantial loss of quality-of-life, with uninfected persons perceiving that decompensated cirrhosis and hepatocellular carcinoma causing the most severe drop in utility. This information provides a means of making a direct comparison between health states and esti-
A19
Abstracts mating quality-adjusted life years (QALYs) for use in an incremental cost-utility analysis. PR3 BIPOLAR PATIENTS’ STATED THERAPY PREFERENCES AND LIKELY ADHERENCE
Johnson FR1, Özdemir S1, Manjunath R2, Burch SP2 Research Triangle Institute, Research Triangle Park, NC, USA; 2 GlaxoSmithKline, Research Triangle Park, NC, USA OBJECTIVE: To quantify the effect on adherence likelihood of patients’ preferences for short-term outcomes and side effects of bipolar disorder treatments. METHODS: Patients with bipolar disorder (n = 469) completed a choice-format conjoint or statedpreference, web-enabled questionnaire that included a series of 11 hypothetical treatment choices. Each treatment alternative specified and varied the frequency and severity of manic episodes and the frequency and severity of depressed episodes, in addition to weight gain, cognitive difficulties, fatigue, and the risk of developing a life-threatening side effect. The patient’s current treatment was included as one of 3 alternatives in 6 of the choice tasks. Each choice task was followed by a question on likely adherence to the chosen treatment. RESULTS: Patients preferred milder episodes of mania and depression. Reduced frequency of manic and depressive episodes contributes significantly to patients’ satisfaction with treatments. The likelihood of choosing a treatment with an interval between mania episodes of 4–6 months is 2.5 times greater than the likelihood of choosing a treatment with an interval between manic episodes of 2–3 months. Patients were willing to sacrifice symptom control to avoid significant weight gain or cognitive effects. For example, the reduction in patient satisfaction from a 10–20 lb gain in weight compared to a 2–10 lb weight gain was similar to the difference in patient satisfaction between a severe and mild manic episode. Fatigue and risk of a life-threatening side effect were the least important attributes. Factors that reduce likely adherence are rapid cycling, significant weight-gain experience, and severe fatigue or cognition problems. CONCLUSION: In this study bipolar patients were willing to sacrifice mood-control benefits of therapy to avoid side effects, particularly weight gain and cognitive side effects. Preferred treatments encourage quantitatively significant improvements in stated adherence and thus should improve long-term treatment outcomes.
1
PR4 PATIENT PREFERENCE AND WILLINGNESS-TO-PAY FOR ANTICOAGULANT TREATMENT OPTIONS IN PATIENTS RECEIVING ORAL ANTICOAGULANT TREATMENT (OAT): A CONJOINT ANALYSIS EXERCISE
Monzini M1, Carpenedo M2, Moia M2, Mantovani LG1 Center of Pharmacoeconomics, Milan, Italy; 2IRCCS Maggiore Hospital and University of Milan, Milan, Italy OBJECTIVES: New anticoagulant drugs, alternative to vitamin K antagonist (VKA) are currently under clinical evaluation. Patients’ preferences should be considered in the development of new therapeutic strategies. Objective of this study was to elicit patients’ preferences on different treatment options. METHODS: A conjoint analysis exercise, a technique for establishing the relative importance of different characteristics in the provision of a good or service, was applied to 96 patients on stable treatment with VKA followed by our anticoagulation clinic. Ninety-four patients (49 male 52,1%; mean age 57 ds = 13) completed the questionnaire. Patients had to choose between two different scenarios in 9 pair-wise comparisons. The attributes considered had previously been selected using an ad-hoc 1
questionnaire administered to a sample of 20 patients and 6 physicians. The following attributes were selected: cost of treatment for the patient (€0 vs. €15 vs. €75/month), route and number of administrations, monitoring frequency, interactions with drugs/food (attention required vs. not required), dose adjustment (required vs. not required), minor bleeding (few vs. no). RESULTS: The variable “cost” was a significant determinant in patients’ choice. A monetary value could be assigned to each attribute. A significant monetary discrimination was reached for all attributes, except for dose adjustment. Patients are willing to pay per month: €62 for once/daily administration tablets vs. one subcutaneous weekly injection; €44 for once/daily administration tablets vs. two/daily administration tablets; €26 for once/monthly vs. twice/monthly visits; €22 for each 6 month vs. once/monthly visits; €22 for a drug which requires no attention to the interaction with other drugs/food; €14 for a drug without risk of minor bleeding. CONCLUSIONS: To our knowledge, our study is the first to elicit preferences from patients in OAT. The importance of this study is the achievement of patients’ preferences in a simply and well accepted method to allow planning optimal health care. Cost Evaluation Studies in Infectious Diseases IN1 COSTS ASSOCIATED WITH SHORTER DURATION OF ANTIBIOTIC THERAPY IN HOSPITALIZED PATIENTS WITH COMMUNITY-ACQUIRED PNEUMONIA
Opmeer BC, el Moussaoui R, Bossuyt PM, Speelman P, Prins JM, deBorgie CA Academic Medical Center/University of Amsterdam, AMSTERDAM, Noord-Holland, The Netherlands OBJECTIVES: To compare costs associated with short (3-day) versus standard (8-day) antibiotic therapy in hospitalised adult patients with mild to moderate-severe community-acquired pneumonia (CAP) in an economic evaluation as part of prospective double-blinded randomised controlled trial (RCT). METHODS: Randomised, double blind, placebo-controlled non-inferiority trial. Nine secondary and tertiary care hospitals in the The Netherlands. A total of 186 adults with a mild to moderate-severe CAP (pneumonia severity index £ 110). Patients who had substantially improved after 72 hours (n = 119) were randomly assigned to receive 5 days of either oral amoxicillin or placebo thrice daily. Direct and indirect medical and non-medical costs associated with resource utilisation during treatment and follow-up until 28 days after randomisation. RESULTS: We randomised 56 patients to placebo and 63 to active treatment. Health outcomes in terms clinical success rates and symptom recovery of were comparable in both study groups. The average cumulative total costs generated during the first 10 days (treatment) were €3320 for standard versus €3352 for short antibiotic treatment (€32 in favour of standard duration); during the follow-up until day 28 these costs were €1072 versus €879, respectively. The overall difference in costs was €159 in favour of short therapy (€4391 versus €4232, respectively). CONCLUSIONS: Shorter antibiotic treatment is equally effective and does not generate additional costs, as compared to standard treatment in hospitalised patients with mild to moderate-severe CAP who have substantially improved after three days of treatment. Although clinical and economic outcomes for the individual patient are comparable, the relevance of the findings concern the societal level, as a potential reduction in the use of antibiotics may slow down the development of antibiotic resistance and need for costly development of new antibiotic therapies.
A18 MC8 THE APPLICATION OF BAYESIAN METHODS TO EVALUATE THE IMPACT OF TIME-DELAY ON THE COST-EFFECTIVENESS OF PRIMARY ANGIOPLASTY FOR ACUTE MYOCARDIAL INFARCTION
Palmer SJ, Asseburg C, Bravo-Vergel Y, Fenwick E, Sculpher M University of York, York, North Yorkshire, UK OBJECTIVES: To apply Bayesian approaches to evaluate the impact of time-delay on the cost-effectiveness of primary angioplasty versus hospital-administered thrombolysis for patients with acute myocardial infarction. METHODS: A probabilistic model was developed to evaluate the life-term cost-effectiveness of primary angioplasty. A health service perspective was adopted with outcomes estimated using Quality-Adjusted Life Years (QALYs). Evidence on short-term event rates was obtained by updating a recent meta-analysis of randomised trials comparing angioplasty with thrombolysis (Lancet 2003; 361:13–20). Data from 22 trials were combined using Bayesian hierarchical modeling and meta-regression. This approach enabled the simultaneous estimation of posterior distributions and correlation structure for: 1) major cardiovascular events including death, reinfarction and stroke, and 2) different time endpoints (4–6 weeks and 6 months). The impact of PCI time-delay to treatment was analysed using mean time delay compared to thrombolysis as a covariate of the random effects model. Long-term costs and QALYs were estimated using a Markov model populated from UK registry data. RESULTS: Based on the average time-delay reported in the trials (54 minutes), primary angioplasty resulted in a mean gain of 0.29 QALYs and an additional cost of £2680 compared to thrombolysis. The associated incremental costeffectiveness ratio (ICER) was £9241 per QALY. At a threshold of £20,000 per QALY there was a 90% probability that primary angioplasty was cost-effective. Adjusting the time-delay resulted in considerable variation in the cost-effectiveness estimates. For a shorter delay of 30-minutes the ICER was £6,850 per QALY; increasing the time-delay to 90-minutes resulted in a marked increase in the ICER to £64,750 per QALY (98% and 36% probability cost-effective at £20 k, respectively). CONCLUSIONS: This study demonstrates the policy importance of time-delay when considering the cost-effectiveness of primary angioplasty, and the advantages of using Bayesian approaches to model multiple endpoints, treatment effects and baseline event rates. Preference and Utility Based Patient Reported Outcomes PR1 EVALUATION OF PREFERENCES IN GENITAL HERPES TREATMENT USING A DISCRETE CHOICE EXPERIMENT
Scalone L1, Watson V2, Ryan M2, Kotsopoulos N3, Patel R4 Center of Pharmacoeconomics, University of Milan, Milan, Italy; 2 Health Economics Research Unit, University of Aberdeen, Aberdeen, UK; 3GlaxoSmithKline, Brentford, UK; 4Department of Genito Urinary Medicine, Royal South Hampshire Hospital, Southampton, UK OBJECTIVES: Genital herpes (GH) is widespread, psychologically disabling and costly. GH is characterized by episodic outbreaks of genital and perigenital vesicles and ulcers, and is one of the most prevalent sexually transmitted diseases in the world today. The seroprevalence of herpes simplex virus (HSV) type-2 ranges from 13% to 40% in the United States and from 7% to 16% in Europe. Often GH is not adequately treated, with consequences on benefits and patients’ well being, and potentially important economic repercussions. GH patients play an important role in decision-making for their management strategies, and
1
their preferences can be crucial to improve treatment outcomes and resource allocation. This study evaluated patients’ preferences in GH treatment. METHODS: Preferences were elicited from 157 patients, recruited from the Harris Poll panel, who completed an online Discrete Choice Experiment (DCE) questionnaire. DCE data was analysed using multinomial logit regression models to estimate respondents’ preferences for GH medical treatment. RESULTS: Respondents, from the US (87.9%) and UK (12.1%), had a median age of 43 years (21–65), and 83.4% were women. Overall, respondents preferred medical treatment to no treatment of GH, with suppressive treatment preferred to episodic, assuming everything else equal. Willingness to pay (WTP) was $38.2/month for episodic treatment and $60.6/month for suppressive treatment, assuming everything else equal. Furthermore, respondents were WTP $56.6/month for a 1% reduction in the “chance of a GH recurrence” in 12 months. In terms of treatment, respondents’ preferences indicated that generally patients preferred the treatment they currently receive. Globally, more patients would choose to be treated (74.3%) than those actually treated (56.2%). CONCLUSIONS: Patients prefer suppressive treatment of GH. Such preferences are influenced by experience, knowledge and awareness of available options. The estimated model suggests that more patients than those actually receiving drug therapies would consider being treated. PR2 PERCEIVED UTILITY OF HEALTH STATES INDUCED BY CHRONIC HEPATITIS B: ESTIMATES FROM UNINFECTED PERSONS IN CANADA
Levy A1, Tafesse E2, Mukherjee J2, Iloeje U2, Poissant L3, Briggs AH4 1 Oxford Outcomes, Vancouver, Bristis Columbia, Canada; 2BMS, Wallingford, CT, USA; 3McGill University, Montreal, QC, Canada; 4 University of Glasgow, Glasgow, UK OBJECTIVE: With an estimated prevalence in Canada between 206,000 and 280,000 persons infected with the virus, persons with hepatitis B typically progress through increasingly severe disease states before death. The objective was to estimate preferences (ratings and utility weights) for six hepatitis B-related disease states among uninfected persons. METHODS: Three hepatologists characterized the typical effects of symptoms on health-related quality of life in terms of symptoms, frequency of tests, hospitalizations, procedures, and dimensions of health such as pain, ability for self-care, activities of daily living, psychological well-being and future outlook. From a convenience sample of 100 uninfected persons in Canada, we elicited ratings using a visual analog scale based on a ‘feeling’ thermometer and standard gamble utility weights using probability wheels with 2-color pie charts for the relative probabilities of perfect health and death. RESULTS: The mean age was 42 years (standard deviation (SD): 14.8, range: 18 to 80 years) and 29% were male. Mean utilities were: 0.80 (95% confidence interval (CI): 0.76; 0.83) for chronic hepatitis B; 0.79 (CI: 0.75; 0.83) for compensated cirrhosis; 0.41 (CI: 0.35; 0.46) for decompensated cirrhosis; 0.69 (CI: 0.65; 0.74) for the first year after liver transplant; 0.77 (CI: 0.73; 0.80) for subsequent years after liver transplant; and, 0.45 (CI: 0.40; 0.51) for hepatocellular carcinoma. The values using the visual analog scale were lower than utilities elicited using the standard gamble, but the relative ranking of each health state did not change. CONCLUSION: All six hepatitis B-related health states are associated with substantial loss of quality-of-life, with uninfected persons perceiving that decompensated cirrhosis and hepatocellular carcinoma causing the most severe drop in utility. This information provides a means of making a direct comparison between health states and esti-
A19
Abstracts mating quality-adjusted life years (QALYs) for use in an incremental cost-utility analysis. PR3 BIPOLAR PATIENTS’ STATED THERAPY PREFERENCES AND LIKELY ADHERENCE
Johnson FR1, Özdemir S1, Manjunath R2, Burch SP2 Research Triangle Institute, Research Triangle Park, NC, USA; 2 GlaxoSmithKline, Research Triangle Park, NC, USA OBJECTIVE: To quantify the effect on adherence likelihood of patients’ preferences for short-term outcomes and side effects of bipolar disorder treatments. METHODS: Patients with bipolar disorder (n = 469) completed a choice-format conjoint or statedpreference, web-enabled questionnaire that included a series of 11 hypothetical treatment choices. Each treatment alternative specified and varied the frequency and severity of manic episodes and the frequency and severity of depressed episodes, in addition to weight gain, cognitive difficulties, fatigue, and the risk of developing a life-threatening side effect. The patient’s current treatment was included as one of 3 alternatives in 6 of the choice tasks. Each choice task was followed by a question on likely adherence to the chosen treatment. RESULTS: Patients preferred milder episodes of mania and depression. Reduced frequency of manic and depressive episodes contributes significantly to patients’ satisfaction with treatments. The likelihood of choosing a treatment with an interval between mania episodes of 4–6 months is 2.5 times greater than the likelihood of choosing a treatment with an interval between manic episodes of 2–3 months. Patients were willing to sacrifice symptom control to avoid significant weight gain or cognitive effects. For example, the reduction in patient satisfaction from a 10–20 lb gain in weight compared to a 2–10 lb weight gain was similar to the difference in patient satisfaction between a severe and mild manic episode. Fatigue and risk of a life-threatening side effect were the least important attributes. Factors that reduce likely adherence are rapid cycling, significant weight-gain experience, and severe fatigue or cognition problems. CONCLUSION: In this study bipolar patients were willing to sacrifice mood-control benefits of therapy to avoid side effects, particularly weight gain and cognitive side effects. Preferred treatments encourage quantitatively significant improvements in stated adherence and thus should improve long-term treatment outcomes.
1
PR4 PATIENT PREFERENCE AND WILLINGNESS-TO-PAY FOR ANTICOAGULANT TREATMENT OPTIONS IN PATIENTS RECEIVING ORAL ANTICOAGULANT TREATMENT (OAT): A CONJOINT ANALYSIS EXERCISE
Monzini M1, Carpenedo M2, Moia M2, Mantovani LG1 Center of Pharmacoeconomics, Milan, Italy; 2IRCCS Maggiore Hospital and University of Milan, Milan, Italy OBJECTIVES: New anticoagulant drugs, alternative to vitamin K antagonist (VKA) are currently under clinical evaluation. Patients’ preferences should be considered in the development of new therapeutic strategies. Objective of this study was to elicit patients’ preferences on different treatment options. METHODS: A conjoint analysis exercise, a technique for establishing the relative importance of different characteristics in the provision of a good or service, was applied to 96 patients on stable treatment with VKA followed by our anticoagulation clinic. Ninety-four patients (49 male 52,1%; mean age 57 ds = 13) completed the questionnaire. Patients had to choose between two different scenarios in 9 pair-wise comparisons. The attributes considered had previously been selected using an ad-hoc 1
questionnaire administered to a sample of 20 patients and 6 physicians. The following attributes were selected: cost of treatment for the patient (€0 vs. €15 vs. €75/month), route and number of administrations, monitoring frequency, interactions with drugs/food (attention required vs. not required), dose adjustment (required vs. not required), minor bleeding (few vs. no). RESULTS: The variable “cost” was a significant determinant in patients’ choice. A monetary value could be assigned to each attribute. A significant monetary discrimination was reached for all attributes, except for dose adjustment. Patients are willing to pay per month: €62 for once/daily administration tablets vs. one subcutaneous weekly injection; €44 for once/daily administration tablets vs. two/daily administration tablets; €26 for once/monthly vs. twice/monthly visits; €22 for each 6 month vs. once/monthly visits; €22 for a drug which requires no attention to the interaction with other drugs/food; €14 for a drug without risk of minor bleeding. CONCLUSIONS: To our knowledge, our study is the first to elicit preferences from patients in OAT. The importance of this study is the achievement of patients’ preferences in a simply and well accepted method to allow planning optimal health care. Cost Evaluation Studies in Infectious Diseases IN1 COSTS ASSOCIATED WITH SHORTER DURATION OF ANTIBIOTIC THERAPY IN HOSPITALIZED PATIENTS WITH COMMUNITY-ACQUIRED PNEUMONIA
Opmeer BC, el Moussaoui R, Bossuyt PM, Speelman P, Prins JM, deBorgie CA Academic Medical Center/University of Amsterdam, AMSTERDAM, Noord-Holland, The Netherlands OBJECTIVES: To compare costs associated with short (3-day) versus standard (8-day) antibiotic therapy in hospitalised adult patients with mild to moderate-severe community-acquired pneumonia (CAP) in an economic evaluation as part of prospective double-blinded randomised controlled trial (RCT). METHODS: Randomised, double blind, placebo-controlled non-inferiority trial. Nine secondary and tertiary care hospitals in the The Netherlands. A total of 186 adults with a mild to moderate-severe CAP (pneumonia severity index £ 110). Patients who had substantially improved after 72 hours (n = 119) were randomly assigned to receive 5 days of either oral amoxicillin or placebo thrice daily. Direct and indirect medical and non-medical costs associated with resource utilisation during treatment and follow-up until 28 days after randomisation. RESULTS: We randomised 56 patients to placebo and 63 to active treatment. Health outcomes in terms clinical success rates and symptom recovery of were comparable in both study groups. The average cumulative total costs generated during the first 10 days (treatment) were €3320 for standard versus €3352 for short antibiotic treatment (€32 in favour of standard duration); during the follow-up until day 28 these costs were €1072 versus €879, respectively. The overall difference in costs was €159 in favour of short therapy (€4391 versus €4232, respectively). CONCLUSIONS: Shorter antibiotic treatment is equally effective and does not generate additional costs, as compared to standard treatment in hospitalised patients with mild to moderate-severe CAP who have substantially improved after three days of treatment. Although clinical and economic outcomes for the individual patient are comparable, the relevance of the findings concern the societal level, as a potential reduction in the use of antibiotics may slow down the development of antibiotic resistance and need for costly development of new antibiotic therapies.