Data on probability of clinical success were obtained from RCTs, but the quality of these studies was weak, and the total number of patients in the studies used to construct the model was very small (Table 2). Also, some studies had patients with diabetic foot infections only as a subgroup, and the results may not necessarily be comparable.

The decision-tree model indicated that clindamycin was the primary cost-effective oral antibiotic for the treatment of mild diabetic foot infections. Clindamycin dominated the other oral antibiotics with a primary efficacy rate of 96.3% and an overall success rate of 99.4% (the additional successes being due to backup drugs, i.e., parenteral antibiotics administered when primary oral drugs failed). In terms of the cost of 10 days of oral antibiotic therapy, clindamycin was more expensive than cephalexin, cloxacillin, and amoxicillin-clavulanate ($31.20 versus $9.60, $15.20 and $20.10, respectively). However, because of the higher primary efficacy rate (96.3% versus 86.2%, 70.4%, and 71.3%, respectively), clindamycin was still the most cost-effective agent overall.

Table 3. Expected Rates of Clinical Success, Amputation and Death for Each Comparator




Clinical Success (%)




Antibiotic




Primary




Secondary



Total

(95%

CI)*




Amputations (per


1000)



Deaths
(per


1000)


Amoxicillin-clavulanate


71.3


24.1


95.4 (94.1-96.6)


43.6


2.7


Cephalexin


86.2


11.6


97.8 (96.9-98.7)


20.9


1.5


Clindamycin


96.3


3.1


99.4 (98.9-99.9)


5.6


0.3


Cloxacillin


70.4


24.8


95.2 (92.3-98.2)


45.0


3.0


Levofloxacin


69.2


25.8


95.0 (91.8-98.3)


46.7


3.0


*95% confidence
interval (CI) from sensitivity analyses.

Amputations and deaths were also substantially lower among patients receiving clindamycin than among patients receiving other oral antibiotics (Table 3).

The expected total cost, as estimated by the decision-tree model, was substantially lower for clindamycin ($361.33) than for the other oral antibiotics (Table 4). Clindamycin had a cost- effectiveness ratio of $363.50 per treatment success, and the ICERs for cephalexin, cloxacillin, amoxicillin-clavulanate and levofloxacin were all dominated by clindamycin, the primary comparator (Table 4).

Table 4. Results of Cost-Effectiveness Analysis of Oral Antibiotics Used for Mild Diabetic Foot Infections


Cost
(2007 Can$)



Success


(%)



Antibiotic



Expected



Difference*



Expected



Difference*


CE


ICER


(%)



(percentage



points)


Clindamycint


361.33


Comparator


99.4


Comparator


363.50


NA


Cephalexin


1239.99


878.66


97.8


-1.6


1268.23


Dominated


by clindamycin


Amoxicillin-clavulanate


2580.81


2219.48


95.4


-4.0


2706.25


Dominated


by clindamycin


Cloxacillin


2658.88


2297.55


95.2


-4.2


2792.52


Dominated


by clindamycin


Levofloxacin


2823.25


2461.92


95.0


-4.4


2970.89


Dominated


by clindamycin


CE =
cost-effectiveness ratio, ICER
=


incremental cost-effectiveness ratio, NA


= not applicable.


*Relative to clindamycin


tPrimary comparator.

Varying the cost of clindamycin in a one-way sensitivity analysis (Table 5) did not affect the treatment decision, except at the extreme (i.e., $950 per 300-mg capsule of clindamycin), which is an unlikely scenario. Likewise, changing the cost of cephalexin did not affect the final decision (Table 5). Although the 95% confidence interval for cephalexin success (69.4% to 94.5%) suggested that clinical success greater than 97% would be unlikely, it is important to note that this success rate was based on an observation of only 29 patients. In comparison, variation of the success rate for clindamycin showed that cephalexin would be the dominant agent if clindamycin had an efficacy rate less than 87%. Given the 95% confidence interval for success for clindamycin (81.7% to 99.3%, obtained from one observation of 27 patients),15 it is possible that clindamycin might have a success rate less than 87%. Two-way sensitivity analyses were also performed (Figures 2A and 2B).

Table 5. One-Way Sensitivity Analyses of Clindamycin and Cephalexin: Impact of Variations in Clinical Efficacy and Cost


Comparison of
Estimated Cost


of Treatment (2007
Can$)


Success or Cost


Clindamycin


Cephalexin


Difference



Clinical success
(%)





Clindamycin


50


4492.36


1239.99


+3252.37


75


2512.94


1239.99


+1272.95


87


1193.33


1239.99


-46.66*





Cephalexin


50


361.33


4470.76


-4109.43


75


361.33


2463.24


-2101.91


95


361.33


455.72


-94.39


97.5


361.33


232.66


+128.67t


Unit cost





Clindamycin


$0.10/300-mg capsule


330.23


1239.99


-909.76


$50.00/300-mg capsule


380.22


1239.99


-859.77


$950.00/300-mg capsule


1280.13


1239.99


+40.14t





Cephalexin


$0.10/500-mg tablet


361.33


1230.49


-869.16


$50.00/500-mg tablet


361.33


1255.48


-894.15


*Cephalexin no longer
dominates.


tClindamycin no longer
dominates.

The assumption that once oral antibiotics had failed, the patient would be admitted to hospital for 14 days of antibiotic therapy might have overestimated total cost; therefore, a sensitivity analysis was performed to determine if varying the length of hospital stay would change the result. When length of stay was varied from 1 day to 14 days, the overall cost- effectiveness ranking was still dominated by clindamycin; however, the per-patient cost was lower with shorter length of stay. The per-patient cost ranged from $159 to $364 for clindamycin, $494 to $1268 for cephalexin, $1055 to $2706 for amoxicillin-clavulanate, $1085 to $2793 for cloxacillin, and $1195 to $2971 for levofloxacin, depending on the number of hospital days required for secondary treatment.

Figure 2. A Two-way sensitivity

Figure 2. A: Two-way sensitivity analysis varying efficacy rates of clindamycin and cephalexin. B: Two-way sensitivity analysis varying costs of clindamycin and cephalexin.

When Monte Carlo simulations were performed with variations of means by 10%, the treatment decision was still dominated by clindamycin (Figure 3).

Figure 3. Sensitivity of cost-effectiveness

Figure 3. Sensitivity of cost-effectiveness results from the Monte Carlo simulation.

Category: Diabetes / Tags: antibiotic, Canada, cost-effectiveness, diabetic foot infection, oral therapy, pharmacoeconomics

Leave a Reply

Your email address will not be published. Required fields are marked *

CAPTCHA image
*