- Email: [email protected]
A Cost-Effectiveness Analysis of Rapid Yeast Detection Kits
Women’s Health Issues 20 (2010) 75–79
www.whijournal.com
A COST-EFFECTIVENESS ANALYSIS OF RAPID YEAST DETECTION KITS Surabhi K. Gaur, MBAa, Kevin D. Frick, PhDb, and Vani Dandolu, MD, MPHc,* a
Temple University School of Medicine, Philadelphia, PA 19140 Johns Hopkins School of Public Health, Baltimore, Maryland c Department of Obstetrics and Gynecology, Department of Public Health, Temple University School of Medicine, Philadelphia, Pennsylvania b
Received 25 November 2008; revised 28 July 2009; accepted 9 September 2009
Objective. To determine the cost effectiveness of the utilization of over-the-counter yeast infection detection kits in the diagnosis of vaginal candidiasis. Methods. A cost–benefit analysis based on a group of 70 adult women from a previous prospective study who presented with vaginitis symptoms. By constructing two decision trees, one in which the kits are an option to the women and one in which they are not, we predict the cost for diagnosing vaginal candidiasis in this group of women. Results. For a group of 70 women presenting with vaginitis symptoms, the total cost of diagnosing their infections without the use of kits is predicted to be $7,051.10. For the same 70 women, the total of cost of diagnosing their infections with the use of kits is predicted to be $5,941.02. Conclusion. We conclude that the use of yeast infection detection kits could reduce the cost of diagnosis by 16%. The introduction of kits could save patients the time, money, and other resources involved in visiting a physician to confirm the diagnosis. Moreover, the sensitivity of yeast kits is superior to the traditional wet mount (77% vs. 52%), so there may be a role for the kits in the physician’s office as well.
Introduction
A
lthough not a life-threatening condition, vulvovaginal candidiasis is a common problem. In one study, 6.5% of women older than 18 reported a least one episode of presumed C vaginitis in the 2 months prior (Foxman, Barlow, D’arcy, Gillespie, & Sobel, 2000). Many of these cases are self-diagnosed—which has been found to be only 35% sensitive—and few more are diagnosed over the telephone (Abbott, 1995). Many women with symptoms of vaginitis (i.e., itching, discharge, vaginal soreness) would have to see a physician for accurate diagnosis of a Candida
Supported by a grant from Savyon Diagnostics Ltd, Kiryat Minrav, Israel. The funder had no role in reviewing and/or approving the contents of this manuscript and none of the authors have a conflict of interest. * Correspondence to: Prof. Vani Dandolu, Temple University School of Medicine, Temple University Hospital, Department of Obstetrics and Gynecology, Department of Public Health, 3401 N Broad Street, Philadelphia, PA 19140; Phone: 215-707-3046; fax: 215-707-1389. E-mail: [email protected]. Copyright Ó 2009 by the Jacobs Institute of Women’s Health. Published by Elsevier Inc.
infection. Physician visits are expensive and, according to 2004 Household Component of the Medical Expenditure Panel Survey data, contribute to 16% of overall health care expenditures. Such data suggest that an expedient and effective method of diagnosis that detects more cases would be useful. The goal of our investigation was to determine which option for candidiasis diagnosis and treatment minimizes the cost associated with management of this condition. With 77% sensitivity, the kits have the potential to eliminate costly trips to the doctor’s office and expedite the healing process (Chatwani et al, 2007). This analysis focuses only on the costs associated with kits for diagnosing simple candidiasis. Complicated candidiasis or other vaginal infections are outside the scope of our investigation and are not among the approved uses of the kit. Methods We analyzed the cost on a set of 70 women from another study, where the prevalence of yeast infections 1049-3867/09 $-See front matter. doi:10.1016/j.whi.2009.09.002
76
S. K. Gaur et al. / Women’s Health Issues 20 (2010) 75–79 Dx confirmed, OTC administered, + wet mount pt. presumably cured Women with candidiasis symptoms, goes to doctor, who performs wet mount must perform culture - wet mount to rule out candidiasis
+ culture - culture
Dx confirmed, OTC administered, pt. presumably cured Pt. definitively does not have candidiasis
Figure 1. Diagnostic process for a symptomatic woman without a rapid yeast detection kit.
in that population was 44% (Chatwani et al., 2007). We use a group of women instead of one individual woman so we can examine the impact of sensitivities and specificities of the different diagnosis methods. Although this lengthens our analysis, it overall makes the analysis more representative of the probabilities faced in real life. The basic assumptions for this analysis are as follows. 1. Culture is the gold standard for diagnosis. We assume that culture has nearly 100% sensitivity and specificity. We use culture results to establish the true infection status of our patients despite the delay in obtaining the results. 2. The end points of the study are to rule in or rule out candidiasis. The patient reaches an end point when 1) negative culture rules her out (or positive culture rules her in), 2) the results of treatment are evident: success or failure of treatment is tantamount to a positive or negative diagnosis of candidiasis, respectively. 3. Patients do not give up. Once patients become symptomatic, their symptoms do not spontaneously resolve nor do they suffer in silence—if the kit does not help, they automatically proceed to the doctor for a workup. 4. Treatment is over-the-counter (OTC) medication. OTC drugs have been proven highly effective for treating candidiasis. In fact, it has been shown that even in the absence of simple candida infection, traditional OTC drugs at least alleviate the
+ kit
Dx confirmed, OTC administered, pt. presumably cured
- kit
Pt. still symptomatic but now does not believe that she has a yeast infection so pt. goes to doctor, who performs wet mount
Women with candidiasis symptoms first purchases a kit
pruritis and inflammation that women experience with most vaginal infections. Also, we assume that there are no indirect costs associated with this treatment, especially because it is very accessible and relatively quick and convenient to use. 5. Positive wet mount is good enough. We assume that most clinicians have enough faith in wet mount to treat based on its results, despite its 23% false-positive rate. Again, there are no directly adverse side effects of treating the falsely positive patients. 6. If wet mount is negative, the clinician proceeds to Candida culture. We assume that testing other differential diagnoses never supersedes complete verification or nullification of a candidiasis diagnosis. Operating on these assumptions, we constructed a flow chart of steps a symptomatic woman would take to obtain a diagnosis of vaginal candidiasis, in the absence of any yeast infection detection kits and then in the presence of kits (Figures 1 and 2). Results First, we examine the costs in a process without kits (Table 1) The following is a step-by-step explanation of where costs are incurred and their amount. Upon developing symptoms, these women proceed to the doctor’s office. This produces three costs:
Dx confirmed, OTC administered, + wet mount pt. presumably cured
+ culture
Dx confirmed, OTC administered, pt. presumably cured
- culture
Pt. definitively does not have candidiasis
- wet mount
Figure 2. Diagnostic process for a symptomatic woman with a rapid yeast detection kit.
S. K. Gaur et al. / Women’s Health Issues 20 (2010) 75–79
77
Table 1. Assumptions Assumption
Comments
Culture is the gold standard for diagnosis.
We assume that culture has nearly 100% sensitivity and specificity. We use culture results to establish the true infection status of our patients despite the delay in obtaining the results. Patient reaches an end point when 1) negative culture rules her out (or positive culture rules her in), 2) the results of treatment are evident: success or failure of treatment is tantamount to a positive or negative diagnosis of candidiasis, respectively. Once patients become symptomatic, their symptoms do not spontaneously resolve nor do they suffer in silence—if the kit does not help, they automatically proceed to the doctor for a work-up. OTC drugs have been proven highly effective for treating candidiasis. In fact, it has been shown that even in the absence of simple candida infection, traditional OTC drugs at least alleviate the pruritis and inflammation that women experience with most vaginal infections. Also, we assume that there are no indirect costs associated with this treatment, especially because it is very accessible and relatively quick and convenient to use. We assume that most clinicians have enough faith in wet mount to treat based on its results, despite its 23% false-positive rate. Again, there are no directly adverse side effects of treating the false-positive patients. We assume that testing other differential diagnoses never supersedes complete verification or nullification of a candidiasis diagnosis.
The end points of the study are to rule in or rule out candidiasis. Patients do not give up.
Treatment is OTC medication.
Positive wet mount is good enough.
If wet mount is negative, the clinician proceeds to Candida culture.
Physician charge: $50.32 (Carr, Rothberg, Friedman, Felsenstein, & Pliskin, 2005) Wet mount preparation: $8.06 (Carr et al., 2005) Patient’s cost of attending a 1-hour office visit (lost wages, transportation costs, etc): $24.48 (Carr et al, 2005) This adds to $82.86 for each patient visit, so a total cost so far of $82.86 3 $70 ¼ $5,800.20. Population prevalence in symptomatic women is 44%. So 31 of these women actually have a yeast infection. Sensitivity of wet mount is 52%, meaning 16 of these 31 women will be diagnosed as such simply by wet mount (Chatwani et al., 2007). These 16 women can use OTC treatment to presumably cure their infection. Each treatment costs $11.89, for a total of 16 3 $11.89 ¼ $191.67 (Mardh, Wagstrom, Landgren, & Holmen, 2004). The other 15 would have cultures performed to determine diagnosis. Each culture costs $23.85, for a total of 15 3 $23.85 ¼ $357.75 (culture cost was obtained from Temple University Hospital laboratories). The culture would presumably be positive for Candida and then these women would also buy an OTC treatment for again a total cost of $179.78. Thirty-nine of the symptomatic women do not have an infection at all. However, 10.3% of them will have a false-positive result on the wet mount. These women would also purchase OTC treatment, but unnecessarily. This cost totals: 0.103 3 39 3 $11.89 ¼ $47.76. These women do not, by the terms of our model, need to proceed to culture—if the treatment for the suspect infection failed, it seems unnecessary to perform a culture. In other words, the failure of treatment is tantamount to a negative candidiasis diagnosis. This is the end point for them in the study.
Of women actually without infections and with negative wet mounts, 89.7% would still be cultured according to the assumptions of our model. This is a cost of 0.897 3 39 3 $23.85 ¼ $834.34. So the total costs, of these 70 women, without the option of a rapid yeast detection kit, and only clinical workup, would incur a total cost of $7,051.10 (Table 1). Now we examine the costs in a process with kits. If these 70 symptomatic women all had access to a rapid yeast detection kit, then presumably they would all use that first. Each kit costs $10, for a total of $700.00 (Chatwani et al, 2007). Again, with a 44% prevalence rate of vaginal candidiasis in this population, 31 of these 70 women actually have a yeast infection. Of women with an infection, the kit picks up 77.4% of infections. So, 24 of the 31 women have a positive test and proceed to purchase the OTC medication: 24 3 $11.89 ¼ $285.29. The other seven women—who have symptoms and an infection but incorrectly had a negative test—go to a physician and incur the costs of the clinical workup. As above, the unit cost of a trip to the doctor’s office is $82.86. This is a total of $82.86 3 7 ¼ $580.02. Again, the wet mount picks up only 52% of infections. So 48% of these seven women will presumably also have a culture done: 0.48 3 7 3 $23.85 ¼ $80.14. All seven of these women had a yeast infection all along. So eventually, they all purchase an OTC regimen: 7 3 $11.89 ¼ $83.23. Of the 39 women without an infection, the kit is accurate for 76.9% (n ¼ 30). So 30 of the 39 women correctly test negative for candidiasis. However, according to the assumptions of our model, these women will see a physician. Moreover, the nine women with false positives will presumably first incorrectly purchase an OTC treatment: 9 3 $11.89 ¼ $107.01. These women have some symptoms that, because they do
78
S. K. Gaur et al. / Women’s Health Issues 20 (2010) 75–79
Table 2. Costs of Diagnosis and Treatment in a Symptomatic Woman Without Use of Rapid Yeast Detection Kit Number of Applicable Unit Cost Patients Total Cost
Category Office Visit (wet mount, doctor’s fee, patient opportunity cost) OTC Medication for all true and false positives Culture costs (true and false negatives on wet mount) Total cost
$82.86
70
$5,800.20
$11.89
35
$416.15
$23.85
35
$834.75
Table 4. Costs of Diagnosis and Treatment in a Symptomatic Woman by Patient Self-Diagnosis Using OTC Medication
Category
Unit Cost
Number of Applicable Patients
OTC medication for all symptomatic patients Culture costs (true and false negatives on wet mount) Total cost
$11.89
70
$832.30
$23.85
35
$834.75
Total Cost
$7,051.10
$7,051.10
not have candidiasis, will not resolve with OTC treatment and then they, too, would presumably seek medical treatment. So, all 39 women end up at the doctor’s office for a unit cost of $82.86 and total cost of $3,231.54. Again, the accuracy of the wet mount plays a role. The wet mount gives false positives 10.3% of the time. So, four of the women have a false-positive wet mount and futilely purchase OTC treatment: 4 3 $11.89 ¼ $47.76. The other 35 women who are rightfully negative by wet mount but with symptoms still are cultured: 35 3 $23.85 ¼ $834.75. The culture presumably comes back negative and now these women are out of the scope of our investigation because candidiasis has been definitively ruled out. The women who tried the OTC and it presumably failed because they did not actually have a yeast infection do not need a culture for candida, practically speaking, because if the treatment failed, it seems illogical to perform a culture. The clinician must now consider other differential diagnoses and these women now, too, are out of the scope of our investigation. So the grand total cost for this group of 70 symptomatic women is estimated to be $5,941.02 (Table 2). Discussion Using the group of 70 women and comparing their outcomes with and without a rapid yeast detection kit, it seems that the kits result in a savings of nearly 16%
Table 3. Costs of Diagnosis and Treatment in a Symptomatic Woman With Use of Rapid Yeast Detection Kit
Category
Unit Cost
Number of Applicable Patients
Kit OTC Medication (true and false positives) Office visit costs Culture Total cost
$10 $11.89
70 44
$700 $523.16
$82.86 $23.85
46 38
$3,811.56 $906.30 $5,941.02
Total Cost
(Table 3). The kits impose an immediate cost on every patient. However, if every patient buys a kit instead of making a visit to the doctor—which costs about eight times more than a kit—it is a large savings. The alternative approach would be OTC treatment based on self-diagnosis by the patient (Table 4). The main drawback of this approach is overtreatment of at least two thirds of women who really do not have candidal infection. This further leads to development of azole resistance and infection by non-albicans strains. These are rare but extremely serious consequences. There are no data to quantify the cost related to these events; therefore, this algorithm was not entertained. Also, the sensitivity of yeast kits is superior to the traditional wet mount (77% vs. 52%), which is the clinician’s key diagnostic tool, so more women are able to treat sooner and more accurately with the advent of the kit. In addition, higher sensitivity leads to fewer cultures to be done, which also contributes to the savings with the kit. The false positives produced by the kit detract from its savings, because these patients suffered the additional futile cost of buying the kit plus the OTC. Were it not for the kit and the associated cost of unnecessary treatment, the patients would have had presumably lower costs, even if they did take their chances with the wet mount preparation. Overall, the kit is a relatively safe product and because vaginitis is not an emergent condition or one associated with high mortality, any initial misdiagnosis or delay in treatment results in very minimal costs to life and balance sheets. In conclusion, the kits present a savings and would be an efficient addition to the tools for diagnosing candidiasis. It would reduce incorrect self-diagnoses and drastically reduce expensive office visits, which still hold the potential for additional unnecessary testing. Moreover, with a higher sensitivity than the wet mount, it would be interesting to see if the kits could establish themselves as the mainstay of diagnosis even in the doctor’s office. References Abbott, J. (1995). Clinical and microscopic diagnosis of vaginal yeast infection: A prospective analysis. Annals of Emergency Medicine, 25, 587–591.
S. K. Gaur et al. / Women’s Health Issues 20 (2010) 75–79 Carr, P. L., Rothberg, M. B., Friedman, R. H., Felsenstein, D., & Pliskin, J. S. (2005). ‘Shotgun’ versus sequential testing: Costeffectiveness of diagnostic strategies for vaginitis. Journal of General Internal Medicine, 20, 793–799. Chatwani, A., Mehta, R., Hassan, S., Rahimi, S., Jeronis, S., & Dandolu, V. (2007). Rapid testing for vaginal yeast detection: a prospective study. American Journal of Obstetrics and Gynecology, 196. 309e.1–e.4. Foxman, B., Barlow, R., D’arcy, H., Gillespie, B., & Sobel, J. (2000). Candida vaginitis: Self-reported incidence and associated costs. Sexually Transmitted Diseases, 27, 230–235. Mardh, P. A., Wagstrom, J., Landgren, M., & Holmen, J. (2004). Usage of antifungal drugs for therapy of genital Candida infections, purchased as over-the-counter products or by prescription. Infectious Diseases in Obstetrics & Gynecology, 12, 99–108.
79
Author Descriptions Surabhi Gaur, MD, MBA, was a student at Temple University School of Medicine when she wrote the manuscript. She subsequently joined residency at UMDNJ Cooper Hospital in NJ. Vani Dandolu, MD, MPH, is the residency program director and Associate professor in Obstetrics and Gynecology at Temple University Hospital. Kevin D. Frick, PhD, MA, is a health economist and Associate Professor at the Johns Hopkins Bloomberg School of Public Health in the Department of Health Policy and Management.
www.whijournal.com
A COST-EFFECTIVENESS ANALYSIS OF RAPID YEAST DETECTION KITS Surabhi K. Gaur, MBAa, Kevin D. Frick, PhDb, and Vani Dandolu, MD, MPHc,* a
Temple University School of Medicine, Philadelphia, PA 19140 Johns Hopkins School of Public Health, Baltimore, Maryland c Department of Obstetrics and Gynecology, Department of Public Health, Temple University School of Medicine, Philadelphia, Pennsylvania b
Received 25 November 2008; revised 28 July 2009; accepted 9 September 2009
Objective. To determine the cost effectiveness of the utilization of over-the-counter yeast infection detection kits in the diagnosis of vaginal candidiasis. Methods. A cost–benefit analysis based on a group of 70 adult women from a previous prospective study who presented with vaginitis symptoms. By constructing two decision trees, one in which the kits are an option to the women and one in which they are not, we predict the cost for diagnosing vaginal candidiasis in this group of women. Results. For a group of 70 women presenting with vaginitis symptoms, the total cost of diagnosing their infections without the use of kits is predicted to be $7,051.10. For the same 70 women, the total of cost of diagnosing their infections with the use of kits is predicted to be $5,941.02. Conclusion. We conclude that the use of yeast infection detection kits could reduce the cost of diagnosis by 16%. The introduction of kits could save patients the time, money, and other resources involved in visiting a physician to confirm the diagnosis. Moreover, the sensitivity of yeast kits is superior to the traditional wet mount (77% vs. 52%), so there may be a role for the kits in the physician’s office as well.
Introduction
A
lthough not a life-threatening condition, vulvovaginal candidiasis is a common problem. In one study, 6.5% of women older than 18 reported a least one episode of presumed C vaginitis in the 2 months prior (Foxman, Barlow, D’arcy, Gillespie, & Sobel, 2000). Many of these cases are self-diagnosed—which has been found to be only 35% sensitive—and few more are diagnosed over the telephone (Abbott, 1995). Many women with symptoms of vaginitis (i.e., itching, discharge, vaginal soreness) would have to see a physician for accurate diagnosis of a Candida
Supported by a grant from Savyon Diagnostics Ltd, Kiryat Minrav, Israel. The funder had no role in reviewing and/or approving the contents of this manuscript and none of the authors have a conflict of interest. * Correspondence to: Prof. Vani Dandolu, Temple University School of Medicine, Temple University Hospital, Department of Obstetrics and Gynecology, Department of Public Health, 3401 N Broad Street, Philadelphia, PA 19140; Phone: 215-707-3046; fax: 215-707-1389. E-mail: [email protected]. Copyright Ó 2009 by the Jacobs Institute of Women’s Health. Published by Elsevier Inc.
infection. Physician visits are expensive and, according to 2004 Household Component of the Medical Expenditure Panel Survey data, contribute to 16% of overall health care expenditures. Such data suggest that an expedient and effective method of diagnosis that detects more cases would be useful. The goal of our investigation was to determine which option for candidiasis diagnosis and treatment minimizes the cost associated with management of this condition. With 77% sensitivity, the kits have the potential to eliminate costly trips to the doctor’s office and expedite the healing process (Chatwani et al, 2007). This analysis focuses only on the costs associated with kits for diagnosing simple candidiasis. Complicated candidiasis or other vaginal infections are outside the scope of our investigation and are not among the approved uses of the kit. Methods We analyzed the cost on a set of 70 women from another study, where the prevalence of yeast infections 1049-3867/09 $-See front matter. doi:10.1016/j.whi.2009.09.002
76
S. K. Gaur et al. / Women’s Health Issues 20 (2010) 75–79 Dx confirmed, OTC administered, + wet mount pt. presumably cured Women with candidiasis symptoms, goes to doctor, who performs wet mount must perform culture - wet mount to rule out candidiasis
+ culture - culture
Dx confirmed, OTC administered, pt. presumably cured Pt. definitively does not have candidiasis
Figure 1. Diagnostic process for a symptomatic woman without a rapid yeast detection kit.
in that population was 44% (Chatwani et al., 2007). We use a group of women instead of one individual woman so we can examine the impact of sensitivities and specificities of the different diagnosis methods. Although this lengthens our analysis, it overall makes the analysis more representative of the probabilities faced in real life. The basic assumptions for this analysis are as follows. 1. Culture is the gold standard for diagnosis. We assume that culture has nearly 100% sensitivity and specificity. We use culture results to establish the true infection status of our patients despite the delay in obtaining the results. 2. The end points of the study are to rule in or rule out candidiasis. The patient reaches an end point when 1) negative culture rules her out (or positive culture rules her in), 2) the results of treatment are evident: success or failure of treatment is tantamount to a positive or negative diagnosis of candidiasis, respectively. 3. Patients do not give up. Once patients become symptomatic, their symptoms do not spontaneously resolve nor do they suffer in silence—if the kit does not help, they automatically proceed to the doctor for a workup. 4. Treatment is over-the-counter (OTC) medication. OTC drugs have been proven highly effective for treating candidiasis. In fact, it has been shown that even in the absence of simple candida infection, traditional OTC drugs at least alleviate the
+ kit
Dx confirmed, OTC administered, pt. presumably cured
- kit
Pt. still symptomatic but now does not believe that she has a yeast infection so pt. goes to doctor, who performs wet mount
Women with candidiasis symptoms first purchases a kit
pruritis and inflammation that women experience with most vaginal infections. Also, we assume that there are no indirect costs associated with this treatment, especially because it is very accessible and relatively quick and convenient to use. 5. Positive wet mount is good enough. We assume that most clinicians have enough faith in wet mount to treat based on its results, despite its 23% false-positive rate. Again, there are no directly adverse side effects of treating the falsely positive patients. 6. If wet mount is negative, the clinician proceeds to Candida culture. We assume that testing other differential diagnoses never supersedes complete verification or nullification of a candidiasis diagnosis. Operating on these assumptions, we constructed a flow chart of steps a symptomatic woman would take to obtain a diagnosis of vaginal candidiasis, in the absence of any yeast infection detection kits and then in the presence of kits (Figures 1 and 2). Results First, we examine the costs in a process without kits (Table 1) The following is a step-by-step explanation of where costs are incurred and their amount. Upon developing symptoms, these women proceed to the doctor’s office. This produces three costs:
Dx confirmed, OTC administered, + wet mount pt. presumably cured
+ culture
Dx confirmed, OTC administered, pt. presumably cured
- culture
Pt. definitively does not have candidiasis
- wet mount
Figure 2. Diagnostic process for a symptomatic woman with a rapid yeast detection kit.
S. K. Gaur et al. / Women’s Health Issues 20 (2010) 75–79
77
Table 1. Assumptions Assumption
Comments
Culture is the gold standard for diagnosis.
We assume that culture has nearly 100% sensitivity and specificity. We use culture results to establish the true infection status of our patients despite the delay in obtaining the results. Patient reaches an end point when 1) negative culture rules her out (or positive culture rules her in), 2) the results of treatment are evident: success or failure of treatment is tantamount to a positive or negative diagnosis of candidiasis, respectively. Once patients become symptomatic, their symptoms do not spontaneously resolve nor do they suffer in silence—if the kit does not help, they automatically proceed to the doctor for a work-up. OTC drugs have been proven highly effective for treating candidiasis. In fact, it has been shown that even in the absence of simple candida infection, traditional OTC drugs at least alleviate the pruritis and inflammation that women experience with most vaginal infections. Also, we assume that there are no indirect costs associated with this treatment, especially because it is very accessible and relatively quick and convenient to use. We assume that most clinicians have enough faith in wet mount to treat based on its results, despite its 23% false-positive rate. Again, there are no directly adverse side effects of treating the false-positive patients. We assume that testing other differential diagnoses never supersedes complete verification or nullification of a candidiasis diagnosis.
The end points of the study are to rule in or rule out candidiasis. Patients do not give up.
Treatment is OTC medication.
Positive wet mount is good enough.
If wet mount is negative, the clinician proceeds to Candida culture.
Physician charge: $50.32 (Carr, Rothberg, Friedman, Felsenstein, & Pliskin, 2005) Wet mount preparation: $8.06 (Carr et al., 2005) Patient’s cost of attending a 1-hour office visit (lost wages, transportation costs, etc): $24.48 (Carr et al, 2005) This adds to $82.86 for each patient visit, so a total cost so far of $82.86 3 $70 ¼ $5,800.20. Population prevalence in symptomatic women is 44%. So 31 of these women actually have a yeast infection. Sensitivity of wet mount is 52%, meaning 16 of these 31 women will be diagnosed as such simply by wet mount (Chatwani et al., 2007). These 16 women can use OTC treatment to presumably cure their infection. Each treatment costs $11.89, for a total of 16 3 $11.89 ¼ $191.67 (Mardh, Wagstrom, Landgren, & Holmen, 2004). The other 15 would have cultures performed to determine diagnosis. Each culture costs $23.85, for a total of 15 3 $23.85 ¼ $357.75 (culture cost was obtained from Temple University Hospital laboratories). The culture would presumably be positive for Candida and then these women would also buy an OTC treatment for again a total cost of $179.78. Thirty-nine of the symptomatic women do not have an infection at all. However, 10.3% of them will have a false-positive result on the wet mount. These women would also purchase OTC treatment, but unnecessarily. This cost totals: 0.103 3 39 3 $11.89 ¼ $47.76. These women do not, by the terms of our model, need to proceed to culture—if the treatment for the suspect infection failed, it seems unnecessary to perform a culture. In other words, the failure of treatment is tantamount to a negative candidiasis diagnosis. This is the end point for them in the study.
Of women actually without infections and with negative wet mounts, 89.7% would still be cultured according to the assumptions of our model. This is a cost of 0.897 3 39 3 $23.85 ¼ $834.34. So the total costs, of these 70 women, without the option of a rapid yeast detection kit, and only clinical workup, would incur a total cost of $7,051.10 (Table 1). Now we examine the costs in a process with kits. If these 70 symptomatic women all had access to a rapid yeast detection kit, then presumably they would all use that first. Each kit costs $10, for a total of $700.00 (Chatwani et al, 2007). Again, with a 44% prevalence rate of vaginal candidiasis in this population, 31 of these 70 women actually have a yeast infection. Of women with an infection, the kit picks up 77.4% of infections. So, 24 of the 31 women have a positive test and proceed to purchase the OTC medication: 24 3 $11.89 ¼ $285.29. The other seven women—who have symptoms and an infection but incorrectly had a negative test—go to a physician and incur the costs of the clinical workup. As above, the unit cost of a trip to the doctor’s office is $82.86. This is a total of $82.86 3 7 ¼ $580.02. Again, the wet mount picks up only 52% of infections. So 48% of these seven women will presumably also have a culture done: 0.48 3 7 3 $23.85 ¼ $80.14. All seven of these women had a yeast infection all along. So eventually, they all purchase an OTC regimen: 7 3 $11.89 ¼ $83.23. Of the 39 women without an infection, the kit is accurate for 76.9% (n ¼ 30). So 30 of the 39 women correctly test negative for candidiasis. However, according to the assumptions of our model, these women will see a physician. Moreover, the nine women with false positives will presumably first incorrectly purchase an OTC treatment: 9 3 $11.89 ¼ $107.01. These women have some symptoms that, because they do
78
S. K. Gaur et al. / Women’s Health Issues 20 (2010) 75–79
Table 2. Costs of Diagnosis and Treatment in a Symptomatic Woman Without Use of Rapid Yeast Detection Kit Number of Applicable Unit Cost Patients Total Cost
Category Office Visit (wet mount, doctor’s fee, patient opportunity cost) OTC Medication for all true and false positives Culture costs (true and false negatives on wet mount) Total cost
$82.86
70
$5,800.20
$11.89
35
$416.15
$23.85
35
$834.75
Table 4. Costs of Diagnosis and Treatment in a Symptomatic Woman by Patient Self-Diagnosis Using OTC Medication
Category
Unit Cost
Number of Applicable Patients
OTC medication for all symptomatic patients Culture costs (true and false negatives on wet mount) Total cost
$11.89
70
$832.30
$23.85
35
$834.75
Total Cost
$7,051.10
$7,051.10
not have candidiasis, will not resolve with OTC treatment and then they, too, would presumably seek medical treatment. So, all 39 women end up at the doctor’s office for a unit cost of $82.86 and total cost of $3,231.54. Again, the accuracy of the wet mount plays a role. The wet mount gives false positives 10.3% of the time. So, four of the women have a false-positive wet mount and futilely purchase OTC treatment: 4 3 $11.89 ¼ $47.76. The other 35 women who are rightfully negative by wet mount but with symptoms still are cultured: 35 3 $23.85 ¼ $834.75. The culture presumably comes back negative and now these women are out of the scope of our investigation because candidiasis has been definitively ruled out. The women who tried the OTC and it presumably failed because they did not actually have a yeast infection do not need a culture for candida, practically speaking, because if the treatment failed, it seems illogical to perform a culture. The clinician must now consider other differential diagnoses and these women now, too, are out of the scope of our investigation. So the grand total cost for this group of 70 symptomatic women is estimated to be $5,941.02 (Table 2). Discussion Using the group of 70 women and comparing their outcomes with and without a rapid yeast detection kit, it seems that the kits result in a savings of nearly 16%
Table 3. Costs of Diagnosis and Treatment in a Symptomatic Woman With Use of Rapid Yeast Detection Kit
Category
Unit Cost
Number of Applicable Patients
Kit OTC Medication (true and false positives) Office visit costs Culture Total cost
$10 $11.89
70 44
$700 $523.16
$82.86 $23.85
46 38
$3,811.56 $906.30 $5,941.02
Total Cost
(Table 3). The kits impose an immediate cost on every patient. However, if every patient buys a kit instead of making a visit to the doctor—which costs about eight times more than a kit—it is a large savings. The alternative approach would be OTC treatment based on self-diagnosis by the patient (Table 4). The main drawback of this approach is overtreatment of at least two thirds of women who really do not have candidal infection. This further leads to development of azole resistance and infection by non-albicans strains. These are rare but extremely serious consequences. There are no data to quantify the cost related to these events; therefore, this algorithm was not entertained. Also, the sensitivity of yeast kits is superior to the traditional wet mount (77% vs. 52%), which is the clinician’s key diagnostic tool, so more women are able to treat sooner and more accurately with the advent of the kit. In addition, higher sensitivity leads to fewer cultures to be done, which also contributes to the savings with the kit. The false positives produced by the kit detract from its savings, because these patients suffered the additional futile cost of buying the kit plus the OTC. Were it not for the kit and the associated cost of unnecessary treatment, the patients would have had presumably lower costs, even if they did take their chances with the wet mount preparation. Overall, the kit is a relatively safe product and because vaginitis is not an emergent condition or one associated with high mortality, any initial misdiagnosis or delay in treatment results in very minimal costs to life and balance sheets. In conclusion, the kits present a savings and would be an efficient addition to the tools for diagnosing candidiasis. It would reduce incorrect self-diagnoses and drastically reduce expensive office visits, which still hold the potential for additional unnecessary testing. Moreover, with a higher sensitivity than the wet mount, it would be interesting to see if the kits could establish themselves as the mainstay of diagnosis even in the doctor’s office. References Abbott, J. (1995). Clinical and microscopic diagnosis of vaginal yeast infection: A prospective analysis. Annals of Emergency Medicine, 25, 587–591.
S. K. Gaur et al. / Women’s Health Issues 20 (2010) 75–79 Carr, P. L., Rothberg, M. B., Friedman, R. H., Felsenstein, D., & Pliskin, J. S. (2005). ‘Shotgun’ versus sequential testing: Costeffectiveness of diagnostic strategies for vaginitis. Journal of General Internal Medicine, 20, 793–799. Chatwani, A., Mehta, R., Hassan, S., Rahimi, S., Jeronis, S., & Dandolu, V. (2007). Rapid testing for vaginal yeast detection: a prospective study. American Journal of Obstetrics and Gynecology, 196. 309e.1–e.4. Foxman, B., Barlow, R., D’arcy, H., Gillespie, B., & Sobel, J. (2000). Candida vaginitis: Self-reported incidence and associated costs. Sexually Transmitted Diseases, 27, 230–235. Mardh, P. A., Wagstrom, J., Landgren, M., & Holmen, J. (2004). Usage of antifungal drugs for therapy of genital Candida infections, purchased as over-the-counter products or by prescription. Infectious Diseases in Obstetrics & Gynecology, 12, 99–108.
79
Author Descriptions Surabhi Gaur, MD, MBA, was a student at Temple University School of Medicine when she wrote the manuscript. She subsequently joined residency at UMDNJ Cooper Hospital in NJ. Vani Dandolu, MD, MPH, is the residency program director and Associate professor in Obstetrics and Gynecology at Temple University Hospital. Kevin D. Frick, PhD, MA, is a health economist and Associate Professor at the Johns Hopkins Bloomberg School of Public Health in the Department of Health Policy and Management.