Skip navigation



The Bandolier conference on Chlamydia was really good. Copies of the conference report will be available soon and can be obtained from Eileen Neail by fax on 01865 226978. They will be posted on Bandolier 's Internet site. But more information becomes available every month, with some good thinking on cost-effectiveness recently [1].


Because Chlamydial infection is often asymptomatic, especially in women, approaches to treating it are often on a population basis; this amounts to screening. Screening has had a bad press, partly because screening programmes (of which there are over 300 operating in the UK) have often been started without the exhaustive thought and testing needed to show that they work and can be effective and cost-effective.

The effectiveness of a screening programme depends on a combination both of test accuracy and therapeutic effectiveness. But screening is not straightforward. It has been said that all screening programmes do harm; some can also do good. And screening programmes are just that - a programme rather than just a test or just a treatment, and as programmes they need good management to be effective and efficient in the long term.

One way of approaching screening is to use evidence from various sources to model effectiveness and cost-effectiveness on a real population. A study on 7,700 women in Baltimore does just that [1], and is a useful read for anyone contemplating Chlamydia screening, or any screening programme, come to that.

Cost-effectiveness study

Using information from the literature, and from 7,700 asymptomatic women attending family planning clinics in Baltimore, researchers calculated the costs of several putative screening strategies and the costs of the effects of Chlamydial infection. These include pelvic inflammatory disease, ectopic pregnancy, pelvic pain and infertility for the women, urethritis or epididymitis for their male partners, and conjunctivitis and pneumonia for children they bear.

The screening strategies examined were:

  • No screening
  • Screening using CDC criteria (mucopurulent cervicitis, all women <20 years, women 20-23 years not using barrier contraceptives or with new sex partner in last 3 months, and older women not using barrier contraceptives and with new sex partner in last 3 months)
  • Screening women <30 years
  • Screening all women
The testing strategy involved testing with polymerase chain reactions (PCR) of cervical swabs or urine. These new DNA tests have very high sensitivity and specificity, with likelihood ratios often of over 100. Treatment was with doxycycline for seven days.


The total cost of the medical consequences of Chlamydial infection in the 7,700 asymptomatic women was $676,000 (in 1995 dollars). All the screening strategies reduced the total costs since costs of the screening strategies were more than offset by reduced costs of the medical consequences of infection (Table).
Cost-effectiveness of screening strategies based on 7,700 women in Baltimore
Screening method Screening costs ($) Medical costs ($) Total cost ($) Number of PID cases PID cases prevented
No screening 676,000 676,000 152
CDC criteria 55,000 390,000 446,000 88 64
Women < 30 years 75,000 297,000 372,000 67 85
Universal screening 120,000 270,000 391,000 61 91
The most cost-effective strategy was age-based screening of all women under 30 years.

Sensitivity analyses in this study showed that screening was cost effective at any prevalence of Chlamydia in asymptomatic women of 1.1% or above, and that at prevalence of 11% or above universal screening became most cost-effective. Other sensitivity analyses showed that the conclusions were robust with regard to many different variables.


There are several additional pieces of evidence which help to make the results of this study credible. Firstly the PCR tests have been shown to be excellent in a number of different studies, so reliable and robust testing is possible on urine specimens.

Treatment is effective. Both a single dose of azithromycin and a seven day regimen of doxycycline give high rates of cure above 90% in nine randomised trials (Table, Bandolier 28 ).
Outcome Time Azithromycin Doxycycline
weeks % %
Clinical cure 91 93
Microbial cure 93 96
Adverse effects 18 19
But drop-outs were high, and this may reflect the population of people inevitably included in such trials (younger, STD patients) or in part non-compliance with seven day regimens of doxycycline [2,3]. Single-dose azithromycin may give better results without much affecting the overall costs.

The prevalence of Chlamydial infection in women in the UK has been the subject of a literature review [4]. Figures ranged from about 2 to 12%, with a weighted mean average of 5.3% (Figure).
Methods used varied, and probably, if anything, underestimated prevalence, but UK prevalence seems not to be much different from that in the USA [4].

Most important is that we now have a randomised trial of screening which shows that it can be effective [5]. In it, 2600 women at high risk (7%)(age <=24; black race; nulligravid; douching; >=2 sexual partners) were randomised to screening plus treatment, or usual care; patient characteristics were similar at baseline. In the following 12 months, PID occurred in 9/1009 women screened & treated and 33/1598 women given usual care. This is an NNT of 85 (95% CI 48 - 375), with a reduction of about 60% in relative risk.


Screening is a complex issue, even when, as here, it appears to produce overall healthcare savings of $50 per woman screened. Muir Gray's book [6] is instructive when considering screening, as well as being a good read generally. In any complex organisation, and especially for screening programmes, performance is the key, and it depends on a number of factors:

Performance = ((Motivation x Competence)/Barriers)

A good strategy is to assume people are well motivated, to ensure that they have the necessary skills and to knock down the barriers which can hinder performance. Those barriers include inaccessible information, information overload, information irrelevance and insufficient time.


  1. MR Howell, TC Quinn, CA Gaydos. Screening for Chlamydia trachomatis in asymptomatic women attending family planning clinics. Annals of Internal Medicine 1998 128:277-84.
  2. AC Haddix, SD Hillia, WJ Kassler. The cost effectiveness of azithromycin for Chlamydia trachomatis infections in women. Sexually Transmitted Diseases 1995 22:274-80.
  3. AP Lea, HM Lamb. Azithromycin. A pharmacoeconomic review of its use as a single-dose regimen in the treatment of uncomplicated urogenital Chlamydia trachomatis infections in women. Pharmacoeconomics 1997 12:596-611.
  4. T Stokes. Screening for Chlamydia in general practice: a literature review and summary of the evidence. Journal of Public Health Medicine 1997 19:222-32.
  5. D Scholes, A Stergachis, F Heidrich et al. Prevention of pelvic inflammatory disease by screening for cervical Chlamydial infection. New England Journal of Medicine 1996 334:1362-6.
  6. JA Muir Gray. Evidence-based Healthcare - how to make health policy and management decisions. Churchill Livingstone 1997. ISBN 0-443-05721-4

previous or next story in this issue