Original article
Pharmacoeconomics of treating uncomplicated urinary tract infections

https://doi.org/10.1016/S0924-8579(99)00024-2Get rights and content

Abstract

The frequency of urinary tract infections (UTIs) and mounting pressure for cost containment in medical care emphasize the need to consider costs of evaluating and treating UTIs. Many clinicians base antibiotic choice on drug cost, probably because this information is objective and readily available. However, the cost of treating UTI patients involves other factors, such as pathogenic susceptibility and consequences of inadequately treated infection. These factors and their associated costs can be difficult to assess and weigh against issues such as drug cost. The direct cost of treating a UTI patient includes not only initial medical evaluation and treatment, but what occurs subsequently. If initial treatment is provided with a drug for which a pathogen is not sensitive, patients will be likely to continue to experience symptoms and return for re-evaluation, resulting in a more thorough evaluation and a second antibiotic, generally a more expensive fluoroquinolone. The most important predictor of high cost-effectiveness is high efficacy against the most common urinary pathogen, Escherichia coli. The lower the effectiveness of antibiotics against this pathogen, the greater the number of revisits and cases of progression to pyelonephritis. Increased follow-up care results in diminished cost-effectiveness. Antibiotic cost is a poor predictor of cost-effectiveness, illustrated by the finding that the most and least expensive drugs, ofloxacin and trimethoprim–sulfamethasoxazole, are approximately equally cost-effective. Both of these are more cost-effective than other drugs, nitrofurantoin and amoxicillin, considered in this analysis.

Introduction

Urinary tract infections (UTIs) are one of the most common reasons for physician office visits, with more than seven million occurring; the estimated cost burden approaching one billion dollars each year in the United States [1], [2]. Most visits occur in women of reproductive age and represent superficial infection of the bladder and lower urinary tract without other complicating factors. Increasingly, evaluation of patients with symptoms suggestive of UTI includes medical history, physical examination, and urinalysis. If the findings are consistent with a diagnosis of uncomplicated UTI, infections are generally treated empirically, i.e. without culturing the urine in order to determine antibiotic sensitivity.

Empiric treatment is generally effective at treating infections because etiologic agents and their antibiotic susceptibility are well established. Escherichia coli, for example, is generally responsible for approximately 80% of such UTIs, with the balance accounted for by other organisms that include Staphylococcus saprophyticus, Proteus mirabilis, Klebsiella spp., and others. A variety of antibiotic options exist for treating these organisms, including aminoglycosides, beta-lactamase inhibitors, cephalosporins, quinolones, folate antagonists, and others. These drugs span a broad range of pricing, with a tenfold difference between the least and most expensive [2]. In practice, clinicians use a range of antibiotics and regimens for treating uncomplicated UTIs [3].

The frequency of UTIs and mounting pressures for cost containment emphasize the need to consider a range of costs in treating UTIs. This perspective recognizes the fact that treatment with an antimicrobial to which an infection does not respond will usually require patient re-evaluation. A return visit is more expensive than the initial visit because in addition to repeating laboratory and physical examination, a culture and sensitivity is generally performed and a second antibiotic prescribed, generally for more expensive fluoroquinolones [4]. Inadequately treated infections also increase the likelihood of progression to more serious infections such as pyelonephritis.

For the practitioner faced with choosing an antibiotic, weighing the trade-offs between drug expense, effectiveness, and subsequent events may be both complex and counterintuitive. While determining the medical consequences of inadequately treated infections may not be difficult, determining the cost and a probability for each is not. For example, resistance among E. coli and other strains has been steadily increasing to trimethoprim–sulfamethoxazole (TMP–SMX) [4], but how does a practitioner identify the point at which another drug may be preferred, either because ultimately it is less expensive or less likely to progress to pyelonephritis, or likely to result in fewer symptom days? In addition, different antibiotics have different, and a different frequency of, side-effects. To provide a perspective on these issues, this analysis discusses the cost and frequency of outcomes for several commonly used antimicrobials to focus on elements that enter into a cost-benefit analysis of UTIs.

Economic analyses relevant to antibiotic therapy of UTIs include cost-benefit, cost-effectiveness, and cost-minimization. Cost-benefit analysis, one of the most frequently used approaches, involves quantifying the dollar costs of a treatment option to provide a common denominator for comparing differences. Cost-effectiveness utilizes a similar approach except benefits are measured in events, such as physician visits or episodes of pyelonephritis. Finally, cost minimization analysis finds the lowest cost option among those found to be of equal benefit.

Pharmacoeconomic analyses draw on the idea of direct, indirect, and intangible costs. This analysis is limited to direct costs, which includes antibiotics, laboratory tests, and physician services. Indirect costs are those such as productive time lost by patients because of illness, including the time for seeking medical attention. Intangible costs are those attributable to patient suffering due to illness.

Defining the different outcomes and costs requires a decision tree, a means by which each possible outcome and its associated cost is specified. For each terminal node, the probability is calculated by multiplying the probability at each of the branch points along its path and specifying a cost for that outcome. For each antibiotic option, each probability at each terminal node is multiplied by its cost and summed to produce a total cost for the tree. In the case of antibiotic options, a separate tree is produced for each possible choice (Fig. 1).

Section snippets

Methods

This analysis is based on two economic analyses of empiric antibiotic treatment of UTIs in women [5], [6]. Because one analysis included detailed economic information, this analysis relies primarily on that [5].

The total cost for each treatment option includes the initial office visit and subsequent events. An initial visit includes urinalysis, medical evaluation, and cost of prescription with one of four antibiotics (amoxicillin, nitrofurantoin, ofloxacin, and TMP–SMX). Except for ofloxacin,

Results

The total cost of treating a UTI patient varies from a low of US$112 to a high of US$172; a 54% difference (Fig. 2). Cost of treatment is similar for ofloxacin and TMP–SMX, slightly more for nitrofurantoin, and considerably greater for amoxicillin. For most antibiotics, the cost of the initial visit—medical evaluation, urine dipstick, and prescription—comprises the largest single component. For amoxicillin, however, the high frequency of repeat visits makes that component the largest single

Discussion

The most important factor in determining cost-effectiveness is an antibiotic’s ability to cure E. coli. A lesser ability to do so (low antibiotic susceptibility) is reflected in increased numbers of costly return visits and cases of pyelonephritis, and diminished cost-effectiveness. Antibiotic cost by itself does not predict cost-effectiveness, reflected by the finding that the most expensive drug, ofloxacin, is the most cost-effective, while one of the least expensive drugs, TMP–SMX, is only

References (6)

  • J.P. Patton et al.

    Urinary tract infections: economic considerations

    Med Clin North Am

    (1991)
  • J.R. Johnson et al.

    Urinary tract infections in women: diagnosis and treatment

    Ann Int Med

    (1989)
  • A.O. Berg

    Variations among family physicians’ management strategies for lower urinary tract infection in women: a report from the Washington Family Physicians Collaborative Research Network

    J Am Board Fam Pract

    (1991)
There are more references available in the full text version of this article.

Cited by (0)

View full text