PELVIC INFLAMMATORY DISEASE (PID) comprises a spectrum of inflammatory disorders of the upper female genital tract (e.g., endometritis, salpingitis, and tubo-ovarian abscess). The serious complications of PID (e.g., infertility, ectopic pregnancy, and chronic pelvic pain) are responsible for a substantial proportion of non-HIV sexually transmitted disease (STD) morbidity in the United States. 1–4 The greatest proportion of preventable PID is attributable to sexually transmitted infections that ascend from the cervix or vagina, such as Chlamydia trachomatis and Neisseria gonorrhoeae infections. 3,5–7
The clinical benefits of screening for chlamydial infection (e.g., reduction in the incidence of PID) 8–13 have motivated the development of national screening guidelines and policy recommendations. 14–16 In part, the motivation for investment in STD control to prevent PID stems from economic evaluations that have revealed screening to be “cost-saving” (i.e., the resources invested are more than recouped by the savings from averted costly complications). 17 However, many of these evaluations have utilized static analytic models, involved only a one-time screening, adopted relatively short time horizons, 18–23 or yielded outcomes in terms of cost per case of PID averted. 18,19,24–30
The studies to date of costs associated with PID (and saved through PID prevention) have utilized different sources for direct medical costs, adopted different assumptions about how unit costs are combined (e.g., the proportion of patients treated in an inpatient versus outpatient setting), and made different assumptions about the timing of clinical events, the age-related risk of events, and the consideration of risk over time. To conduct analyses that are consistent with the recommendations of the Panel on Cost-Effectiveness in Health and Medicine, 31 there is a pressing need to better characterize the uncertainty in the estimated costs attributable to PID. 5 We have capitalized on a formidable body of earlier work 18–30,32–42 to develop a comprehensive natural history model of PID. We used this model to estimate a plausible range for the average lifetime costs of PID and its major complications in order to improve future cost-effectiveness analyses of STD screening programs.
Methods
We developed a Markov model using DATA 3.5 software (TreeAge Software, Williamstown, MA) to simulate the natural history of PID in a hypothetical cohort of U.S. women of reproductive age, incorporating the severity of infection, number of recurrent episodes, treatment setting, and timing and risk of long-term complications (Fig. 1). The model is used to project major complications resulting from PID, such as ectopic pregnancy, chronic pelvic pain, and infertility, as well as life expectancy, quality-adjusted life expectancy, and average per-person lifetime costs. To estimate the average per-person lifetime cost for women with PID between the ages of 15 and 34 years, model simulations are conducted for cohorts of women between the ages of 15 and 19, 20 and 24, 25 and 29, and 30 and 34 years, and the projected clinical and economic outcomes are weighted according to age-stratified data on the prevalence of infection. 43 We adopt a societal perspective and incorporate time preference by discounting costs and benefits 3% annually. Sensitivity analyses are performed to assess the robustness of conclusions to uncertainty regarding costs, the natural history of PID, and the timing and duration of major clinical complications.
;)
Fig. 1: Model overview. Health states in the model are chosen to be descriptive of the woman’s current health and relevant history as well as predictive of clinical prognosis and future resource use. In each 3-month time period, women with PID have a probability of developing complications (chronic pelvic pain, ectopic pregnancy, or infertility) from a primary or recurrent upper genital tract infection. See Methods section for details.
Natural History Model
The model employs a state-transition framework, wherein progression of disease over time in an individual woman is characterized as a sequence of transitions from one health state to another. Health states in the model are chosen to be descriptive of the woman’s current health and relevant history as well as predictive of clinical prognosis and future resource use. Transition probabilities, extrapolated from data in the literature, are used to move women through different health states over time. The analysis incorporates a woman’s entire lifetime and is divided into equal increments of time, referred to as Markov cycles.
In each 3-month cycle, women with PID have a probability of developing complications from a primary or recurrent upper genital tract infection. In the absence of screening, women with asymptomatic upper genital tract infections face a future risk of any of the three major complications of PID: ectopic pregnancy, chronic pelvic pain, and tubal infertility. Women with symptomatic upper genital tract infections, or those identified by screening, are assumed to be treated according to current national guidelines for the treatment of sexually transmitted diseases 2; such women still face a probability of long-term complications (e.g., tubal infertility), albeit at a lower risk than untreated patients. Women who receive outpatient treatment for initial upper tract infection face the same risks of developing long-term complications as those receiving inpatient treatment. 44 We conservatively assume that in any one cycle a woman may experience only one of these long-term complications. Deaths attributable to all-cause mortality are distinguished from those due to an ectopic pregnancy or surgical treatment related to other complications of PID.
Decision analytic models aim to bear insight in a setting of uncertainty. Accordingly, in the absence of complete and perfect data, analyses using these models require a series of assumptions. Table 1 depicts the assumptions made in previously published models of chlamydia screening about the timing of major complications following an episode of PID. 18,24–26,28,33–36,38,39,42 We made the following assumptions for our base case: (1) in an average representative cross-section of a cohort with PID, 80% of cases represent first infections, 15% represent second infections, and 5% represent third or more infections;45 (2) a woman with PID who is promptly treated has a lower risk of complications than one who is not treated or whose treatment is delayed;46 (3) the risk of ectopic pregnancy begins 3 months following an upper genital tract infection and extends over a subsequent 15-year time frame; (4) each woman may experience up to three ectopic pregnancies;47–50 (5) women who are at risk for infertility but do not desire or pursue pregnancy will not seek treatment for their infertility; (6) the risk of infertility (defined as failing to conceive after 1 year of unprotected sexual intercourse) begins 5 years following an upper genital tract infection and extends over a subsequent 10-year period; and (8) the risk of chronic pelvic pain starts 3 months following an upper genital tract infection and extends over a 2-year time frame.
TABLE 1: Time Horizon Assumptions Utilized in Prior Studies
Figure 2 summarizes the assumptions made about the timing of PID sequelae for the base case. To explore the uncertainty about the age-specific risk of major complications attributable to PID, we evaluated several different scenarios in which the duration of time between an upper genital tract infection and a major complication was varied. For a given duration of time between the acute episode of PID and a specific complication, we also explored the implications of a constant rate of sequelae versus a rate that changed over time.
Fig. 2: Timing of sequelae following PID in a cohort of women aged 20 to 24 years. The time period during which women are at risk for developing a specific major complication as a result of PID varies. Shown are the main assumptions made in the base case. Alternative assumptions were explored in sensitivity analyses. See Methods section for details.
Clinical Data
The values used for the base case analysis are summarized in Table 2. 20,21,24,27,29,34,35,37,45–63 Specifying the quantitative and temporal relationships between PID and each of the major complications is complex since clinical studies vary in design and duration and are inconsistent in their definitions of different clinical criteria (e.g., chronic PID) and use of laparoscopic verification. 64 Further complicating our understanding of its epidemiology, PID is often asymptomatic and caused by different microbial organisms, and its major complications are not exclusive to the disease. 5 Probabilities and their plausible ranges for infertility, ectopic pregnancy, and chronic pelvic pain were derived from two cohort studies of women with laparoscopically verified PID. 5,45,55 While some data suggest that no differences exist between women with chlamydia- and gonorrhea-associated infections, chlamydia-associated upper genital tract infections may be associated with a less-favorable fertility prognosis than nonchlamydial pathogens. 5 Because of the limitations in the available data, we assigned wide plausible ranges to each probability used in the base case.
TABLE 2: Estimated Probabilities Used in the Model
Cost Data
Several studies have estimated the costs associated with PID, 24,33–35 each of which has relied on different methods to identify, quantify, and value resource utilization (Table 3). Since these studies varied in their strengths and limitations, we conducted four separate analyses, using the data from each. For pedagogical purposes we present a base case analysis using data from Magid et al. that represented a midrange for the four studies.
TABLE 3: Cost Inputs Used in the Model
All studies included direct medical costs such as for doctor visits, laboratory tests, diagnostics, prescription drugs, hospital stays, and surgical procedures. Direct nonmedical costs, such as those associated with seeking supportive counseling for fertility impairment and exploring alternative parenting options (e.g., surrogate mother, adoption, etc.), were not included in any of the studies. Indirect costs, referring to lost productivity and representing the value of output forgone by women with PID or women who die prematurely of PID, were measured in one study. 34
In a cost-effectiveness analysis comparing the economic consequences of doxycycline and azithromycin treatments for chlamydial infections, Magid et al. estimated costs associated with PID and its sequelae by using Colorado Blue Cross and Blue Shield payments and allowed charges for medical services. 24 Resource use was identified by chart review, literature review, and expert opinion for each clinical outcome. Costs for each medical service were then estimated on the basis of Blue Cross and Blue Shield payments (hospital services), Blue Cross and Blue Shield-allowed charges (outpatient and physician services), and average wholesale prices (AWP; antibiotic costs). 24
The Institute of Medicine commissioned a committee to develop a quantitative model that could help decision-makers prioritize development of vaccines against 14 infectious diseases. 33 In its evaluation of Chlamydia trachomatis disease, resource utilization for each PID-related outcome was assessed by literature review and committee estimates. Data from the Health Care Financing Administration (HCFA) from 1995 were then used to estimate costs for the services specified (doctor visits, hospital stays, prescription drugs, diagnostics, and laboratory tests).
One of the most frequently cited references on PID is a study by Washington et al. 34 that focused on estimating annual PID costs and determining their payment sources. In this study, International Classification of Diseases (ICD-9) codes 65 from California acute care hospital discharge and physician charge data were used to identify costs associated with initial upper genital tract infection, ectopic pregnancy, and chronic pelvic pain. All ectopic pregnancies were treated on an inpatient basis. All costs estimated from discharge data were based on charges. Infertility diagnosis and treatment costs were from the literature. 66
A study on the direct medical costs of PID by Rein et al. provides the most recent estimates of PID costs. The study was based on actual patient and insurance payments from the MarketScan database, a commercial database that provides payment data for approximately 12 million privately insured persons. 35 To estimate the average cost per upper genital tract infection, they added all PID-related claims for a given episode in order to determine an average cost per episode. Outpatient and inpatient costs for chronic pelvic pain and ectopic pregnancy were similarly estimated. Infertility costs were estimated from the literature. 34,66 All costs were based on reimbursement rates, as opposed to charges. Inpatient pharmacy costs were included but not outpatient prescription costs.
We estimated that 30% of women experiencing chronic pelvic pain would require inpatient treatment, 37 50% of ectopic pregnancies would require inpatient treatment, 58 and 45% of infertile women would seek treatment. 24 Since each of the cost studies varied in the proportion of women who received a specific type of treatment (e.g., surgery versus no surgery) and the setting in which treatment took place (e.g., inpatient versus outpatient), we made several assumptions to facilitate comparisons between estimates. For example, Magid et al. did not provide estimates of resource use associated with outpatient treatment for ectopic pregnancy. To derive an estimate, we assumed that outpatient costs would be a proportion of inpatient costs (45%) based on data from an alternative source. 33 All costs were translated into year 2000 dollars with use of the medical care component of the Consumer Price Index for all urban users (United States Bureau of Labor and Statistics) before the 3% discount rate was applied. 67
Health-Related Quality-of-Life
To incorporate the quality-of-life decrements associated with the complications of PID, we applied quality weights obtained from a study commissioned by the Institute of Medicine on vaccine development to health states reflecting chronic pelvic pain (0.60) and infertility (0.82). 33 We conducted extensive sensitivity analyses in which we varied these quality weights widely, since there are no published data that provide econometric health-related quality-of-life measures for the individual sequelae associated with PID. We also explored the impact of a variety of assumptions about the duration of the quality-of-life decrements associated with PID sequelae such as infertility. As recommended by the Panel on Cost-Effectiveness Analysis in Health and Medicine, we applied population-based age and sex-specific quality weights for all other health states, based on data from the Beaver Dam Outcomes Study. 31,68
Results
Projected Clinical Outcomes.
In a cohort of 100,000 adolescent females between the ages of 20 and 24 years with acute upper genital tract infection followed over their lifetimes, 8550 total ectopic pregnancies (including recurrent cases), 16,800 cases of infertility, and 18,600 cases of chronic pelvic pain were projected to occur. The average per-person life expectancy was 58.91 years. For an otherwise identical cohort without PID, life expectancy was 58.92 years, reflecting the low mortality associated with PID. PID-associated mortality included eight deaths of ectopic pregnancy complications and 1 death of complications secondary to surgery for upper genital infection and chronic pelvic pain. In contrast, the quality-adjusted life expectancy for women with PID ranged from 55.5 to 56.9 years, depending on the duration of the quality-of-life decrement, following a complication of PID. In comparison with an average healthy cohort, the average reduction in quality-adjusted life expectancy ranged from 1.6 months to 18.8 months. To place these quality-of-life decrements into perspective, the life-expectancy gains associated with cancer screening with biennial mammography, Papanicolaou smears, and fecal occult blood testing range from 0.8 to 3.2 months. 69 Results for age cohorts of 15 to 19 years and 25 to 34 years were similar (data not shown).
Projected Lifetime Costs.
Among women with acute upper genital tract infection between the ages of 20 and 24 years, the discounted average per-person lifetime cost was $2150 ($2745 undiscounted;Table 4). Average lifetime costs for women who developed major complications ranged from $1270 to $6840, depending on the specific complication. For example, the average lifetime costs associated with chronic pelvic pain were $6350; with ectopic pregnancy, $6840; and with infertility, $1270. The lower costs associated with infertility reflect the fact that not all women with tubal infertility seek infertility treatment. Assumptions on the proportion of women who received each type of treatment directly influenced the overall projected cost for each complication (Fig. 3). Projected per-person lifetime costs were similar for the younger cohort (aged 15–19 years) and the older cohort (aged 25–34 years) and ranged from $2110 to $2310, respectively. On the basis of age-stratified population based data, the overall per-person lifetime cost for women between the ages of 15 and 24 years was $2150 ($2755 undiscounted). We repeated these analyses using data from the three other sources of cost inputs (Table 4). The average lifetime per-person cost for a woman with PID ranged from $1060 to $3180.
TABLE 4: Projected Lifetime Costs (Year 2000 U.S. $)
Fig. 3: Impact of treatment assumptions on costs. Assumptions about the proportion of women who receive each type of treatment directly influence the overall projected cost for each complication. See Results section for details.
Under our base case assumptions, the majority of costs (79%) were incurred within 3 years after initial upper genital tract infection. Even when we assumed an extended time period of more than 10 years after acute PID for both chronic pelvic pain and ectopic pregnancy, nearly 40% of costs were still incurred within 3 years.
Sensitivity Analyses.
Results were most sensitive to the risk of major complications resulting from an upper genital tract infection and the cost of surgery for chronic pelvic pain (Fig. 4). Results were moderately sensitive to the probability of receiving inpatient treatment for the initial upper genital tract infection treatment, the cost of chronic pelvic pain, and the relative risk of major complications in women with asymptomatic PID rather than symptomatic upper genital tract infection. If all asymptomatic and symptomatic women faced the same risk for developing sequelae, the average per-person lifetime cost decreased to $1890 (base case, $2150). In contrast, if women with asymptomatic infections faced a twofold increase in risk, the average cost increased to $2360. Results were only minimally sensitive to the probability and cost of receiving treatment for infertility, probabilities of having additional ectopic pregnancies, and the costs associated with treating initial upper genital tract infection and ectopic pregnancy.
Fig. 4: Impact of varying selected model parameters on the average lifetime cost of PID. See Results section for details. *Probabilities were varied ±50%. †Costs were varied from ±50%. ‡Discount rate was varied from 0 to 10%. §Time period of infertility risk was varied from 5 to 20 years.
Figure 5 shows the impact of varying both age and discount rate on the average lifetime cost of PID. As the discount rate was varied from 0% to 10%, the average lifetime costs in a cohort of women aged 20 to 24 years with PID decreased from $2745 to $1760 (base case, $2150). To understand the impact of the time horizon in the context of a 3% discount rate, we first applied the same time horizon to all major complications and then varied the duration of time (e.g., 2, 5, or 10 years) between the upper genital tract infection and the subsequent major complication. For example, if all three sequelae of PID hypothetically were to occur immediately following an initial upper genital tract infection, the per-person average lifetime cost would approach $2645. If all major complications occurred after 5 or 10 years, the per-person average lifetime cost would be reduced to $2340 or $2025, respectively.
Fig. 5: Impact of age and discount rate on the average lifetime cost of PID. See Results section for details.
To assess the relation between the duration of time from upper genital infection to the major complication and the discount rate (Fig. 6), we conducted a two-way sensitivity analysis. Higher discount rates had a disproportionate effect when complications occurred further from the initial upper genital tract infection. For example, with a 10% discount rate, if the risk of infertility began immediately following the initial upper genital tract infection and that risk was spread over 10 years, the average per-person lifetime cost for a women with infertility was $940. In contrast, if the risk of infertility did not start until 5 years after the acute infection but was still spread over 10 years, the average per-person lifetime cost was only $530.
Fig. 6: Impact of time horizon and discount rate on the average lifetime cost of PID. See Results section for details.
Discussion
Policy-makers, clinical investigators, and decision analysts interested in the per-person lifetime cost of PID are confronted with multiple published studies that have used different sources for cost data, varying cost estimation methods, and different assumptions about the timing and risk of PID complications. To resolve these discrepancies, we developed a comprehensive state-transition model of PID to evaluate the impact of different assumptions about the future timing of clinical sequelae on the projected average per-person lifetime cost of PID. Incorporating the best data available, we estimated a plausible range for the average per-person lifetime cost of PID, capitalizing on four earlier, well-characterized sources of direct medical costs.
We estimate that the average per-person lifetime cost for a woman with PID ranges from $1060 to $3180, depending on the particular source of data used for costs. Each cost source was associated with particular strengths and limitations. 24,33–35 For example, while estimation of cost resource utilization by chart review and verification by expert opinion as conducted by Magid et al. can identify “best practice,” it may not necessarily capture real-life treatment patterns. 24 Similarly, while review of hospital discharge and patient bill data as conducted by Rein et al. can give insight into real-life practice, it relies on accurate record-keeping and ICD-9 diagnosis coding. 35 In particular, estimation of costs associated with chronic pelvic pain may be particularly difficult to estimate with use of data based on ICD-9 codes for three reasons: (1) there are no specific ICD-9 codes for chronic pelvic pain, 65 (2) there is no consensus on the clinical definition of the condition, 70 and (3) chronic pelvic pain is often misdiagnosed and treated as another disease. 4
Differences in cost estimates among different cost data may also stem from the evolution of treatment practices. For example, assumptions regarding the percentage of women receiving inpatient versus outpatient treatment for ectopic pregnancy had a substantial impact on the estimated cost per complication.
Our model showed an average per-person lifetime cost for PID to be $2150, approximately midway between the highest and lowest estimates of prior studies. While the study by Washington et al. is probably the most widely cited cost study of PID, treatment practices have changed and shifted to the outpatient setting during the decade since its publication. For example, the CDC estimates that 50% of ectopic pregnancies are now treated on an outpatient basis. 58 Although we use more recent data 33 to estimate outpatient treatment resource utilization, this method may still overestimate ectopic pregnancy treatment costs and consequently result in higher cost estimates than in the other studies. In addition, Washington et al. relied on charge data, which included a margin of profit in prices and thus may have overestimated true treatment costs. We therefore suggest that at this time the cost projection ($3180) based on data from Washington et al. represent an upper-bound average per-person lifetime cost for a woman with PID. 34 We would also suggest that cost projections based on data from the Institute of Medicine ($1410) 33 and Rein et al. ($1060) 35 represent a lower-bound estimate for the lifetime cost of PID for the following two reasons: (1) The Institute of Medicine costs were based on Medicare and Medicaid reimbursement rates, which are typically lower than privately insured rates for medical services, 33 and (2) Rein et al. relied heavily on diagnosis codes to estimate PID-associated costs. 35 As a result, their estimates may fail to capture some costs and thus underestimate the actual resource utilization associated with PID. On the basis of unpublished data from a Seattle study that compared patient insurance claims and charts, Rein et al. speculated that their costs may underestimate PID costs by as much as 30%.
While PID and its sequelae are associated with low mortality, quantifying its morbidity is crucial to understanding the disease’s impact on women. We found that the reduced quality of life a woman experiences on account of chronic pelvic pain and infertility was associated with a decrease in quality-adjusted life expectancy, ranging from 1.6 months to 4.8 months, depending on the duration of the quality-of-life decrement within a range of 5 to 15 years. The Institute of Medicine estimated that quality-of-life decrements would last for the remainder of a woman’s life. If this assumption is true, quality-adjusted life expectancy would be reduced by 18.8 quality-adjusted months for women with complications secondary to PID.
We also found that the total treatment costs for chronic pelvic pain and ectopic pregnancies account for a larger percentage of overall PID-related costs than do costs for infertility. Typically, clinical symptoms of infertility are not apparent until a woman chooses to conceive, whereas chronic pelvic pain and ectopic pregnancy both have clinical symptoms that require a woman to seek medical attention sooner following her acquisition of PID. The further into the future the costs are incurred, the more they are attenuated by discounting effects. 31 In addition, some women with infertility do not desire children and consequently do not seek treatment for their condition. As a result, treatment costs for initial upper genital tract infection, chronic pelvic pain, and ectopic pregnancies collectively have a greater impact on the per-person average lifetime cost of PID.
Our results were most sensitive to the risk of complications and the likelihood of receiving treatment for chronic pelvic pain in an inpatient setting. Although the age at which PID occurred had minimal impact on costs, assumptions about the timing (or time horizon) for each complication affected the average per-person lifetime costs to a greater degree. The impact of these assumptions was exaggerated by varying the discount rate, resulting in an average per-person lifetime cost of PID ranging between $1800 and $2700 as a function of the time horizon assumed for each complication and the discount rate. This is especially important given the emphasis placed on discounting health costs and benefits in the cost-effectiveness analysis framework.
While cost-effectiveness analyses from a societal perspective require a lifetime perspective, many health care decision-makers are more interested in shorter time horizons, as the average person changes health care plans every 3 to 5 years. Under our base case assumptions, we found that nearly 80% of costs associated with PID are incurred within 3 years, given that the time lag for many sequelae is often only a few years after the initial acute upper genital tract infection. Even if the delay for chronic pelvic pain and ectopic pregnancy is increased to 10 years, almost 40% of costs are still incurred in 3 years. Consequently, in addition to the clinical benefits for patients, given this relatively short time frame for accrual of PID-related costs, managed care organizations may have financial incentives to implement screening programs that prevent PID.
This study has several limitations. First, all of the limitations associated with our primary data sources became our limitations as well. By utilizing data from multiple sources, however, we may have attenuated the impact of these limitations. Second, we made the simplifying assumption that women were at risk to develop only one of the major complications of PID. While PID sequelae are clearly not mutually exclusive and in fact are likely to coexist in a portion of women, there are insufficient data to model them probabilistically. The impact of having made this simplifying assumption is only to make our cost estimates more conservative. Third, we did not include costs associated with hysterectomy and its complications, despite the fact that women with PID may undergo such a procedure. As a result, we may have underestimated the cost of surgery and, consequently, the average per-person lifetime cost of PID. Fourth, we did not include time costs in our analyses, a limitation that adds to the conservative nature of our cost projections.
Fifth, while studies have shown that rates of hospitalization for PID and ectopic pregnancy increase with the number of chlamydial infections, 71 we assumed that rates of hospitalization were similar for all women, regardless of the number of prior infections. Sixth, we did not assign different treatment costs for women based on their history of infections, since detailed cost data are not available to estimate resource use conditional on the number of past upper genital tract infections. Finally, because our primary data sources consisted largely of U.S.-based studies, we acknowledge that our results may not be generalizable to other clinical settings or health care systems outside of the U.S. Despite these limitations, a distinct advantage of a model such as this one is that as better information becomes available the model can be revised, new data can be integrated, and the impact on the conclusions can be evaluated.
In conclusion, we estimate that the average per-person lifetime costs for women with PID range from $1060 to $3180. Future cost-effectiveness analyses of STD screening programs can include this range as a reasonable upper and lower bound. By improvement of the caliber of cost-effectiveness estimates, the potential relevancy of such studies to public health policy can be enhanced. Our sensitivity analyses suggest that these lifetime costs are most influenced by the probability of long-term complications of PID, the decrement in quality of life associated with these complications, and the direct medical costs associated with chronic pelvic pain and ectopic pregnancy. Acquisition of these data should be research priorities.
References
1. Westrom LV.
Chlamydia trachomatis–clinical significance and strategies of intervention. Semin Dermatol 1990; 9: 117–125.
2. The 1998 guidelines for the treatment of sexually transmitted diseases. MMWR Morb Mortal Wkly Rep 1998; 47: 53–58.
3. Aral SO, Mosher WD, Cates W Jr. Self-reported pelvic inflammatory disease in the United States, 1988. JAMA 1991; 266: 2570–2573.
4. Paavonen J, Eggert-Kruse W.
Chlamydia trachomatis: impact on human reproduction. Hum Reprod Update 1999; 5: 433–447.
5. Westrom L, Eschenbach D. Pelvic inflammatory disease. In: Holmes KK, Mardh PA, Sparling PF, et al., eds. Sexually Transmitted Diseases. New York: McGraw-Hill, 1999: 783–809.
6. Pelvic inflammatory disease: guidelines for prevention and management. MMWR Morb Mortal Wkly Rep 1991; 40: 1–25.
7. Ross J. Pelvic inflammatory disease. BMJ 2001; 322: 658–659.
8. Nelson HD, Helfand M. Screening for chlamydial infection. Am J Prev Med 2001; 20: 95–107.
9. Scholes D, Stergachis A, Heidrich FE, Andrilla H, Holmes KK, Stamm WE. Prevention of pelvic inflammatory disease by screening for cervical chlamydial infection. N Engl J Med 1996; 334: 1362–1366.
10. Addiss DG, Vaughn ML, Ludka D, Pfister J, Davis JP. Decreased prevalence of
Chlamydia trachomatis infection associated with a selective screening program in family planning clinics in Wisconsin. Sex Transm Dis 1993; 20: 28–35.
11. Britten TF, DeLisle S, Fine D. STDs and family planning clinics: A regional program for chlamydia control that works. Am J Gynecol Health 1992; 6: 24.
12. Egger M, Low N, Smith GD, Lindblom B, Herrmann B. Screening for chlamydial infections and the risk of ectopic pregnancy in a county in Sweden: ecological analysis. BMJ 1998; 316: 1776–1780.
13. Kamwendo F, Forslin L, Bodin L, Danielsson D. Decreasing incidences of gonorrhea- and chlamydia-associated acute pelvic inflammatory disease: a 25-year study from an urban area of central Sweden. Sex Transm Dis 1996; 23: 384–391.
14. McGlynn E. Choosing and evaluating clinical performance measures. Jt Comm J Qual Improv 1998; 24: 470–479.
15. HEDIS 2000 measures zero in on women’s health issues. Healthcare Demand Dis Manag 1999; 5: 101–104.
16. US Preventive Services Task Force. Screening for chlamydial infection: recommendations and rationale. Am J Prev Med 2001; 20: 90–94.
17. Mangione-Smith R, O’Leary J, McGlynn EA. Health and cost-benefits of chlamydia screening in young women. Sex Transm Dis 1999; 26: 309–316.
18. Howell MR, Quinn TC, Gaydos CA. Screening for
Chlamydia trachomatis in asymptomatic women attending family planning clinics: a cost-effectiveness analysis of three strategies. Ann Intern Med 1998; 128: 277–284.
19. Howell MR, Quinn TC, Brathwaite W, Gaydos CA. Screening women for
Chlamydia trachomatis in family planning clinics: the cost-effectiveness of DNA amplification assays. Sex Transm Dis 1998; 25: 108–117.
20. Marrazzo JM, Celum CL, Hillis SD, Fine D, DeLisle S, Handsfield HH. Performance and cost-effectiveness of selective screening criteria for
Chlamydia trachomatis infection in women: implications for a national chlamydia control strategy. Sex Transm Dis 1997; 24: 131–141.
21. Phillips R, Aronson MD, Taylor WC, Safran C. Should test for
Chlamydia trachomatis cervical infection be done during routine gynecological visits? Analysis of the costs of alternative strategies. Ann Intern Med 1987; 107: 188–194.
22. Trachtenberg AI, Washington AE, Halldorson S. A cost-based decision analysis for chlamydia screening in California family planning clinics. Obstet Gynecol 1988; 71: 101–108.
23. Humphreys JT, Henneberry JF, Rickard RS, Beebe JL. Cost-benefit analysis of selective screening criteria for
Chlamydia trachomatis infection in women attending Colorado family planning clinics. Sex Transm Dis 1992; 19: 47–53.
24. Magid D, Douglas JM Jr, Schwartz JS. Doxycycline compared with azithromycin for treating women with genital
Chlamydia trachomatis infections: an incremental cost-effectiveness analysis. Ann Intern Med 1996; 124: 389–399.
25. Postma MJ, Welte R, van den Hoek JA, van Doornum GJ, Jager HC, Coutinho RA. Cost-effectiveness of partner pharmacotherapy in screening women for asymptomatic infection with
Chlamydia trachomatis. Value in Health 2001; 4: 266–275.
26. van Valkengoed IG, Postma MJ, Morre SA, et al. Cost-effectiveness analysis of a population based screening programme for asymptomatic
Chlamydia trachomatis infections in women by means of home obtained urine specimens. Sex Transm Infect 2001; 77: 276–282.
27. Howell MR, Gaydos JC, McKee KT Jr, Quinn TC, Gaydos CA. Control of
Chlamydia trachomatis infections in female army recruits: cost-effective screening and treatment in training cohorts to prevent pelvic inflammatory disease. Sex Transm Dis 1999; 26: 519–526.
28. Welte R, Kretzschmar M, Leidl R, van den Hoek JAR, Jager JC, Postma MJ. Cost-effectiveness of screening programs for
Chlamydia trachomatis: a population-based dynamic approach. Sex Transm Dis 2000; 27: 518–529.
29. Haddix AC, Hillis SD, Kassler WJ. The cost effectiveness of azithromycin for
Chlamydia trachomatis infections in women. Sex Transm Dis 1995; 22: 274–280.
30. Shafer MA, Pantell RH, Schachter J. Is the routine pelvic examination needed with the advent of urine-based screening for sexually transmitted diseases? Arch Pediatr Adolesc Med 1999; 153: 119–125.
31. Gold MR, Siegel JE, Russell LB, Weinstein MC. Cost-effectiveness in Health and Medicine. New York: Oxford University Press, 1996.
32. Nettleman MD, Bell TA. Cost-effectiveness of prenatal testing for
Chlamydia trachomatis. Am J Obstet Gynecol 1991; 164: 1289–1294.
33. Chlamydia. In: Stratton KR, Durch JS, Lawrence RS, eds. Vaccines for the 21st Century: A Tool For Decision Making. Washington, DC: National Academy Press, 2000: 149–158.
34. Washington AE, Katz P. Cost of and payment source for pelvic inflammatory disease: trends and projections, 1983 through 2000. JAMA 1991; 266: 2565–2569.
35. Rein DB, Kassler WJ, Irwin KL, Rabiee L. Direct medical cost of pelvic inflammatory disease and its sequelae: decreasing, but still substantial. Obstet Gynecol 2000; 95: 397–402.
36. Paavonen J, Puolakkainen M, Paukku M, Sintonen H. Cost-benefit analysis of first-void urine
Chlamydia trachomatis screening program. Obstet Gynecol 1998; 92: 292–298.
37. Genc M, Ruusuvaara L, Mardh PA. An economic evaluation of screening for
Chlamydia trachomatis in adolescent males. JAMA 1993; 270: 2057–2064.
38. Genc M, Mardh PA. A cost-effectiveness analysis of screening and treatment for
Chlamydia trachomatis infection in asymptomatic women. Ann Intern Med 1996; 124: 1–7.
39. Gift TL, Pate MS, Hook EW3rd, Kassler WJ. The rapid test paradox: when fewer cases detected lead to more cases treated: a decision analysis of tests for
Chlamydia trachomatis. Sex Transm Dis 1999; 26: 232–240.
40. Sellors JM, Pickard L, Gafni A, et al. Effectiveness and efficiency of selective vs. universal screening for chlamydial infection in sexually active young women. Arch Intern Med 1992; 152: 1837–1844.
41. Petita A, Hart SM, Bailey EM. Economic evaluation of three methods of treating urogenital chlamydial infections in the emergency department. Pharmacotherapy 1999; 19: 648–654.
42. Mehta SD, Bishai D, Howell MR, Rothman RE, Quinn TC, Zenilman JM. Cost-effectiveness of five strategies for gonorrhea and chlamydia control among female and male emergency department patients. Sex Transm Dis 2002; 29: 83–91.
43. Chlamydia trachomatis genital infections–United States, 1995. MMWR Morb Mortal Wkly Rep 1997; 46: 193–198.
44. Ness RB, Soper DE, Holley RL, et al. Effectiveness of inpatient and outpatient treatment strategies for women with pelvic inflammatory disease: results from the Pelvic Inflammatory Disease Evaluation and Clinical Health (PEACH) randomized trial. Am J Obstet Gynecol 2002; 186: 929–937.
45. Westrom L, Joesoef R, Reynolds G, Hagdu A, Thompson SE. Pelvic inflammatory disease and fertility: a cohort study of 1,844 women with laparoscopically verified disease and 657 control women with normal laparoscopic results. Sex Transm Dis 1992; 19: 185–192.
46. Hillis SD, Joesoef R, Marchbanks PA, Wasserheit JN, Cates W Jr, Westrom L. Delayed care of pelvic inflammatory disease as a risk factor for impaired fertility. Am J Obstet Gynecol 1993; 168: 1503–1509.
47. Joesoef MR, Westrom L, Reynolds G, Marchbanks P, Cates W. Recurrence of ectopic pregnancy: the role of salpingitis. Am J Obstet Gynecol 1991; 165: 46–50.
48. Bronson RA. Tubal pregnancy and infertility. Fertil Steril 1977; 28: 221–228.
49. Makinen JI, Salmi TA, Nikkanen VP, Juhani Koskinen EY. Encouraging rates of fertility after ectopic pregnancy. Int J Fertil 1989; 34: 46–51.
50. Nagamani M, London S, Amand PS. Factors influencing fertility after ectopic pregnancy. Am J Obstet Gynecol 1984; 149: 533–535.
51. Westrom L. Gynecological chlamydial infections. Infection 1982; 10( suppl 1): S40–S45.
52. Washington AE, Arno PS, Brooks MA. The economic cost of pelvic inflammatory disease. JAMA 1986; 255: 1735–1738.
53. Nettleman MD, Jones RB, Roberts SD, et al. Cost-effectiveness of culturing for
Chlamydia trachomatis: a study in a clinic for sexually transmitted diseases. Ann Intern Med 1986; 105: 189–196.
54. Thompson SE, Washington AE. Epidemiology of sexually transmitted
Chlamydia trachomatis infections. Epidemiol Rev 1983; 5: 96–123.
55. Westrom L. Effect of acute pelvic inflammatory disease on fertility. Am J Obstet Gynecol 1975; 121: 707–13.
56. Washington AE, Sweet RL, Shafer MA. Pelvic inflammatory disease and its sequelae in adolescents. J Adolesc Health Care 1985; 6: 298–310.
57. Westrom L. Incidence, prevalence and treads of ectopic pregnancy in a population of women. BMJ (Clin Res Ed) 1981; 282: 15–18.
58. Ectopic pregnancy–United States, 1990–1992. MMWR Morb Mortal Wkly Rep 1995; 44: 46–48.
59. Hirsch MB, Mosher WD. Characteristics of infertile women in the United States and their use of infertility services. Fertil Steril 1987; 47: 618–625.
60. Anderson RN. United States life tables. Natl Vital Stat Rep 1999; 47: 1–37.
61. Randolph AG, Washington AE, Prober CG. Cesarean delivery for women presenting with genital herpes lesions: efficacy, risks, and costs. JAMA 1993; 270: 77–82.
62. Mrus JM, Goldie SJ, Weinstein MC, Tsvet J. The cost-effectiveness of elective Cesarean delivery for HIV-infected women with detectable HIV RNA during pregnancy. AIDS 2000; 14: 2543–2552.
63. Petitti DB, Cefalo RC, Shapiro SA, Whalley P. In-hospital maternal mortality in the United States: time trends and relation to method of delivery. Obstet Gynecol 1982; 59: 6–12.
64. Cates W, Wasserheit JN, Marchbanks PA. Pelvic inflammatory disease and tubal infertility: the preventable conditions. Ann N Y Acad Sci 1994; 709: 179–195.
65. Division of Quality Control Management, American Medical Association. International Classification of Diseases. 9th Revision. Clinical modification. Chicago: American Hospital Publishing, 1989.
66. Forrest JD, Gold RB, Kenney AM. The Need, Availability, Financing of Reproductive Health Services. New York: Alan Guttmacher Institute, 1989.
67. United States Bureau of Labor, Statistics. Consumer price indexes: Bureau of Labor and Statistics, 2000. Available at
http://www.bls.gov/cpihome.htm, accessed May 2001.
68. Fryback DG, Dasbach EJ, Klein R, et al. The Beaver Dam Health Outcomes Study: initial catalog of health-state quality factors. Medical Decision Making 1993; 13: 89–102.
69. Wright JC, Weinstein MC. Gains in life expectancy from medical interventions–standardizing data on outcomes. N Engl J Med 1998; 339: 380–386.
70. Zondervan K, Barlow DH. Epidemiology of chronic pelvic pain. Baillieres Best Pract Res Clin Obstet Gynaecol 2000; 14: 403–414.
71. Hillis SD, Owens LM, Marchbanks PA, Amsterdam LF, MacKenzie WR. Recurrent chlamydial infections increase the risks of hospitalization for ectopic pregnancy and pelvic inflammatory disease. Am J Obstet Gynecol 1997; 176: 103–107.
