The Measurement of Performance in Probabilistic Diagnosis

Owing to the inherent uncertainty of diagnostic tasks, diagnostic advice should be offered in a probabilistic, rather than deterministic form. Since the late fifties a lot of effort has been invested in constructing probabilistic diagnosis rules. Much less has been done to devise rational tools for evaluating them ; conventional error rates reflect but one aspect of performance in a rather crude way. The aim of this paper and its successors is to offer a body of evaluation tools. After defining a general framework and stating its limitations we apply some graphical techniques to the acute abdominal pain data that will serve as illustration in the next papers as well: dot diagrams, the triangular diagram for the three-disease case, and three tabular representations based on categorization of the probabilities, viz. the usual (forced) classification matrix, from which various classification rates are read off; the classification matrix with doubt, in which uncertain and confident diagnoses are distinguished; and the exclusion matrix, which spots diseases that are judged improbable. Together these matrices give a good first impression of the behaviour of a probabilistic diagnosis system. The outlined techniques of later papers are needed for a more complete analysis.

The Measurement of Performance in Probabilistic Diagnosis V. General Recommendations

Guidelines are offered concerning the use of evaluation tools, with emphasis on the simplest and most informative ones amongst those proposed in Parts I—IV of this series. Moreover, the main areas of future research and development, as the authors see them, are pointed out.

The Measurement of Performance in Probabilistic Diagnosis IV. Utility Considerations in Therapeutics and Prognostics

It is argued that it is preferable to evaluate probabilistic diagnosis systems in terms of utility (patient benefit) or loss (negative benefit). We have adopted the provisional strategy of scoring performance as if the system were the actual decision-maker (not just an aid to him) and argue that a rational figure of merit is given by the average loss which patients would incur by having the system decide on treatment, the treatment being selected according to the minimum expected loss principle of decision theory.

A similar approach is taken to the problem of evaluating probabilistic prognoses, but the fundamental differences between treatment selection skill and prognostic skill and their implications for the assessment of such skills are stressed. The necessary elements of decision theory are explained by means of simple examples mainly taken from the acute abdomen, and the proposed evaluation tools are applied to Acute Abdominal Pain data analysed in our previous papers by other (not decision-theoretic) means. The main difficulty of the decision theory approach, viz. that of obtaining good medical utility values upon which the analysis can be based, receives due attention, and the evaluation approach is extended to cover more realistic situations in which utility or loss values vary from patient to patient.

Use of Posterior Probabilities to Evaluate Methods of Discriminant Analysis

By means of a case study the choice between several methods of discriminant analysis is presented. Experimental data of a two-groups problem with one or two variables is analysed. The different methods are compared according to posterior probabilities which can be computed for each subject and which are the basis of discriminant analysis. These posterior probabilities are analysed graphically as well as numerically.

Test Selection in Jaundice: A Comparison between Physician Behavior and a Diagnostic Model

The results of an observational study aimed at a formal assessment of clinicians’ test-selection behavior are presented. We first make a proposal for diagnostic test usage in the latter phases of jaundice diagnosis. Next we compute a probabilistic estimate of the patient diagnosis, based on the COMIK algorithm. From the proposal and the probabilistic estimate we can predict the “test-selection behavior” of clinicians. The assessment follows from a tabulation of the predicted behavior against the tests selected by clinicians. It is shown that for most tests, the predictions are consistent with the observed test-selection behavior at a statistically significant level. Discussions of discrepancies between prediction and observation, and reasons for deviations from general guidelines, provide new dimensions for medical education. The methodology applied is a useful tool to improve medical care for the jaundiced patient.

The LYMFASIM Simulation Program for Modeling Lymphatic Filariasis and its Control

The LYMFASIM modeling framework for the transmission and control of the tropical parasitic disease lymphatic filariasis is described and its use in the context of an endemic community in north-eastern Brazil is illustrated. Lymphatic filariasis is a disease with a complex natural history with many unknowns. This complicates decision making with respect to control strategies. With LYMFASIM, a variety of hypotheses can be tested about the life history of the parasite Wuchereria bancrofti, its transmission from man to man through mosquitoes, the role of the immune system in regulating parasite numbers, the development of disease symptoms, and the effects of control measures (drug treatment or mosquito control). The implications of alternative assumptions and uncertainty about the quantification of parameters for the effectiveness of control strategies can be investigated. Thanks to the use of stochastic microsimulation, LYMFASIM is highly flexible and can be adapted and extended as new knowledge emerges.

Comparison of two models predicting IVF success; the effect of time trends on model performance

STUDY QUESTION

How well does the recently developed UK model predicting the success rate of IVF treatment (the 2011 Nelson model) perform in comparison with a UK model developed in the early 1990s (the Templeton model)?

SUMMARY ANSWER

Both models showed similar performance, after correction for the increasing success rate over time of IVF.

WHAT IS KNOWN ALREADY

For counselling couples undergoing IVF treatment it is of paramount importance to be able to predict success. Several prediction models for the chance of success after IVF treatment have been developed. So far, the Templeton model has been recommended as the best approach after having been validated in several independent patient data sets. The Nelson model, developed in 2011 and characterized by the largest development sample containing the most recently treated couples, may well perform better.

STUDY DESIGN, SIZE, DURATION

We tested both models in couples that were included in a national cohort study carried out in the Netherlands between the beginning of January 2002 and the end of December 2004.

PARTICIPANTS/MATERIALS, SETTING, METHODS

We analysed the IVF cycles of Dutch couples with primary infertility (n = 5176). The chance of success was calculated using the two UK models that had been developed using the information collected in the Human Fertilisation and Embryology Authority database. Women were treated in 1991-1994 (Templeton) or 2003-2007 (Nelson). The outcome of success for both UK models is the occurrence of a live birth after IVF but the outcome in the Dutch data is an ongoing pregnancy. In order to make the outcomes compatible, we used a factor to convert the chance of live birth to ongoing pregnancy and use the overall terms 'success or no success after IVF'. The discriminative ability and the calibration of both models were assessed, the latter before and after adjustment for time trends in IVF success rates.

MAIN RESULTS AND THE ROLE OF CHANCE

The two models showed a similarly limited degree of discriminative ability on the tested data (area under the receiver operating characteristic curve 0.597 for the Templeton model and 0.590 for the Nelson model). The Templeton model underestimated the success rate (observed 21% versus predicted 14%); the Nelson model overestimated the success rate (observed 21% versus predicted 29%). When the models were adjusted for the changing success rates over time, the calibration of both models considerably improved (Templeton observed 21% versus predicted 20%; Nelson observed 21% versus predicted 24%).

LIMITATIONS, REASONS FOR CAUTION

We could only test the models in couples with primary infertility because detailed information on secondary infertile couples was lacking in the Dutch data. This shortcoming may have negatively influenced the performance of the Nelson model.

WIDER IMPLICATIONS OF THE FINDINGS

The changes in success rates over time should be taken into account when assessing prediction models for estimating the success rate of IVF treatment. In patients with primary infertility, the choice to use the Templeton or Nelson model is arbitrary.

Risk of cardiovascular disease after pre-eclampsia and the effect of lifestyle interventions: a literature-based study

OBJECTIVE

This study addresses the following questions. Do cardiovascular risk factors fully explain the odds ratio of cardiovascular risk after pre-eclampsia? What is the effect of lifestyle interventions (exercise, diet, and smoking cessation) after pre-eclampsia on the risk of cardiovascular disease?

DESIGN

Literature-based study.

SETTING

N/A.

POPULATION OR SAMPLE

N/A.

METHODS

Data for the calculations were taken from studies identified by PubMed searches. First, the differences in cardiovascular risk factors after pre-eclampsia compared with an uncomplicated pregnancy were estimated. Second, the effects of lifestyle interventions on cardiovascular risk were estimated. Validated risk prediction models were used to translate these results into cardiovascular risk.

RESULTS

After correction for known cardiovascular risk factors, the odds ratios of pre-eclampsia for ischaemic heart disease and for stroke are 1.89 (IQR 1.76-1.98) and 1.55 (IQR 1.40-1.71), respectively. After pre-eclampsia, lifestyle interventions on exercise, dietary habits, and smoking cessation decrease cardiovascular risk, with an odds ratio of 0.91 (IQR 0.87-0.96).

CONCLUSIONS

Cardiovascular risk factors do not fully explain the risk of cardiovascular disease after pre-eclampsia. The gap between estimated and observed odds ratios may be explained by an additive risk of cardiovascular disease by pre-eclampsia. Furthermore, lifestyle interventions after pre-eclampsia seem to be effective in decreasing cardiovascular risk. Future research is needed to overcome the numerous assumptions we had to make in our calculations.

Having a Pap smear, quality of life before and after cervical screening: a questionnaire study

OBJECTIVE

To assess the health-related quality of life (HRQoL) impact of cervical cancer screening in women with normal test results.

DESIGN

Questionnaire study.

SETTING

Maastricht, the Netherlands.

POPULATION

A cohort of 789 women were followed from screening invitation until after the receipt of screening results. A female age-matched reference group (n=567) was included.

METHODS

Questionnaires were sent to the home address of the women before screening, after screening, and again with the screening results.

MAIN OUTCOME MEASURES

Generic HRQoL (SF-12, EQ-5D), generic anxiety (STAI-6), screen-specific anxiety (PCQ), and potential symptoms and feelings related to the smear-taking procedure.

RESULTS

A total of 60% of screening participants completed questionnaire 1(n=924): 803 of these women granted permission to access their files; 789 of these 803 women had normal test results (Pap 1), and were included in the analyses. Generic HRQoL (SF-12, EQ-5D) and anxiety (STAI-6) scores were similar in the study and reference groups. Before screening, after screening, and also after the receipt of test results, screening participants reported less screen-specific anxiety (PCQ, P<0.001) than the reference group (n=567), with differences indicating clinical relevance. 19% of screening participants were bothered by feelings of shame, pain, inconvenience, or nervousness during smear taking, and 8 and 5% of women experienced lower abdominal pain, vaginal bleeding, discharge, or urinary problems for 2-3 and 4-7 days, respectively, following the Pap smear.

CONCLUSION

The reduced levels of screen-specific anxiety in screening participants, possibly indicating reassurance, are worthwhile addressing in more depth. We conclude that although considerable numbers of women reported unpleasant effects, there were no adverse HRQoL consequences of cervical screening in women with normal test results.

Advance notification letters increase adherence in colorectal cancer screening: a population-based randomized trial

OBJECTIVE

The population benefit of screening depends not only on the effectiveness of the test, but also on adherence, which, for colorectal cancer (CRC) screening remains low. An advance notification letter may increase adherence, however, no population-based randomized trials have been conducted to provide evidence of this.

METHOD

In 2008, a representative sample of the Dutch population (aged 50-74 years) was randomized. All 2493 invitees in group A were sent an advance notification letter, followed two weeks later by a standard invitation. The 2507 invitees in group B only received the standard invitation. Non-respondents in both groups were sent a reminder 6 weeks after the invitation.

RESULTS

The advance notification letters resulted in a significantly higher adherence (64.4% versus 61.1%, p-value 0.019). Multivariate logistic regression analysis showed no significant interactions between group and age, sex, or socio-economic status. Cost analysis showed that the incremental cost per additional detected advanced neoplasia due to sending an advance notification letter was € 957.

CONCLUSION

This population-based randomized trial demonstrates that sending an advance notification letter significantly increases adherence by 3.3%. The incremental cost per additional detected advanced neoplasia is acceptable. We therefore recommend that such letters are incorporated within the standard CRC-screening invitation process.

What determines individuals' preferences for colorectal cancer screening programmes? A discrete choice experiment

INTRODUCTION

In many countries uptake of colorectal cancer (CRC) screening remains low.

AIM

To assess how procedural characteristics of CRC screening programmes determine preferences for participation and how individuals weigh these against the perceived benefits from participation in CRC screening.

METHODS

A discrete choice experiment was conducted among subjects in the age group of 50-75 years, including both screening-naïve subjects and participants of a CRC screening programme. Subjects were asked on their preferences for aspects of CRC screening programmes using scenarios based on pain, risk of complications, screening location, preparation, duration of procedure, screening interval and risk reduction of CRC-related death.

RESULTS

The response was 31% (156/500) for screening-naïve and 57% (124/210) for CRC screening participants. All aspects proved to significantly influence the respondents' preferences. For both groups combined, respondents required an additional relative risk reduction of CRC-related death by a screening programme of 1% for every additional 10 min of duration, 5% in order to expose themselves to a small risk of complications, 10% to accept mild pain, 10% to undergo preparation with an enema, 12% to use 0.75l of oral preparation combined with 12h fasting and 32% to use an extensive bowel preparation. Screening intervals shorter than 10 years were significantly preferred to a 10-year screening interval.

CONCLUSION

This study shows that especially type of bowel preparation, risk reduction of CRC related death and length of screening interval influence CRC screening preferences. Furthermore, improving awareness on CRC mortality reduction by CRC screening may increase uptake.

Preferences for colorectal cancer screening strategies: a discrete choice experiment

BACKGROUND

Guidelines underline the role of individual preferences in the selection of a screening test, as insufficient evidence is available to recommend one screening test over another. We conducted a study to determine the preferences of individuals and to predict uptake for colorectal cancer (CRC) screening programmes using various screening tests.

METHODS

A discrete choice experiment (DCE) questionnaire was distributed among naive subjects, yet to be screened, and previously screened subjects, aged 50-75 years. Subjects were asked to choose between scenarios on the basis of faecal occult blood test (FOBT), flexible sigmoidoscopy (FS), total colonoscopy (TC) with various test-specific screening intervals and mortality reductions, and no screening (opt-out).

RESULTS

In total, 489 out of 1498 (33%) screening-naïve subjects (52% male; mean age+/-s.d. 61+/-7 years) and 545 out of 769 (71%) previously screened subjects (52% male; mean age+/-s.d. 61+/-6 years) returned the questionnaire. The type of screening test, screening interval, and risk reduction of CRC-related mortality influenced subjects' preferences (all P<0.05). Screening-naive and previously screened subjects equally preferred 5-yearly FS and 10-yearly TC (P=0.24; P=0.11), but favoured both strategies to annual FOBT screening (all P-values <0.001) if, based on the literature, realistic risk reduction of CRC-related mortality was applied. Screening-naive and previously screened subjects were willing to undergo a 10-yearly TC instead of a 5-yearly FS to obtain an additional risk reduction of CRC-related mortality of 45% (P<0.001).

CONCLUSION

These data provide insight into the extent by which interval and risk reduction of CRC-related mortality affect preferences for CRC screening tests. Assuming realistic test characteristics, subjects in the target population preferred endoscopic screening over FOBT screening, partly, due to the more favourable risk reduction of CRC-related mortality by endoscopy screening. Increasing the knowledge of potential screenees regarding risk reduction by different screening strategies is, therefore, warranted to prevent unrealistic expectations and to optimise informed choice.

Preferences for colorectal cancer screening strategies: a discrete choice experiment

Background:

Guidelines underline the role of individual preferences in the selection of a screening test, as insufficient evidence is available to recommend one screening test over another. We conducted a study to determine the preferences of individuals and to predict uptake for colorectal cancer (CRC) screening programmes using various screening tests.

Methods:

A discrete choice experiment (DCE) questionnaire was distributed among naive subjects, yet to be screened, and previously screened subjects, aged 50–75 years. Subjects were asked to choose between scenarios on the basis of faecal occult blood test (FOBT), flexible sigmoidoscopy (FS), total colonoscopy (TC) with various test-specific screening intervals and mortality reductions, and no screening (opt-out).

Results:

In total, 489 out of 1498 (33%) screening-naïve subjects (52% male; mean age±s.d. 61±7 years) and 545 out of 769 (71%) previously screened subjects (52% male; mean age±s.d. 61±6 years) returned the questionnaire. The type of screening test, screening interval, and risk reduction of CRC-related mortality influenced subjects’ preferences (all P<0.05). Screening-naive and previously screened subjects equally preferred 5-yearly FS and 10-yearly TC (P=0.24; P=0.11), but favoured both strategies to annual FOBT screening (all P-values <0.001) if, based on the literature, realistic risk reduction of CRC-related mortality was applied. Screening-naive and previously screened subjects were willing to undergo a 10-yearly TC instead of a 5-yearly FS to obtain an additional risk reduction of CRC-related mortality of 45% (P<0.001).

Conclusion:

These data provide insight into the extent by which interval and risk reduction of CRC-related mortality affect preferences for CRC screening tests. Assuming realistic test characteristics, subjects in the target population preferred endoscopic screening over FOBT screening, partly, due to the more favourable risk reduction of CRC-related mortality by endoscopy screening. Increasing the knowledge of potential screenees regarding risk reduction by different screening strategies is, therefore, warranted to prevent unrealistic expectations and to optimise informed choice.

Screening for colorectal cancer: randomised trial comparing guaiac-based and immunochemical faecal occult blood testing and flexible sigmoidoscopy

BACKGROUND

Screening for colorectal cancer (CRC) is widely accepted, but there is no consensus on the preferred strategy. We conducted a randomised trial comparing participation and detection rates (DR) per screenee of guaiac-based faecal occult blood test (gFOBT), immunochemical FOBT (FIT), and flexible sigmoidoscopy (FS) for CRC screening.

METHODS

A representative sample of the Dutch population (n = 15 011), aged 50-74 years, was 1:1:1 randomised prior to invitation to one of the three screening strategies. Colonoscopy was indicated for screenees with a positive gFOBT or FIT, and for those in whom FS revealed a polyp with a diameter > or = 10 mm; adenoma with > or = 25% villous component or high grade dysplasia; serrated adenoma; > or = 3 adenomas; > or = 20 hyperplastic polyps; or CRC.

RESULTS

The participation rate was 49.5% (95% confidence interval (CI) 48.1 to 50.9%) for gFOBT, 61.5% (CI, 60.1 to 62.9%) for FIT and 32.4% (CI, 31.1 to 33.7%) for FS screening. gFOBT was positive in 2.8%, FIT in 4.8% and FS in 10.2%. The DR of advanced neoplasia was significantly higher in the FIT (2.4%; OR, 2.0; CI, 1.3 to 3.1) and the FS arm (8.0%; OR, 7.0; CI, 4.6 to 10.7) than the gFOBT arm (1.1%). FS demonstrated a higher diagnostic yield of advanced neoplasia per 100 invitees (2.4; CI, 2.0 to 2.8) than gFOBT (0.6; CI, 0.4 to 0.8) or FIT (1.5; CI, 1.2 to 1.9) screening.

CONCLUSION

This randomised population-based CRC-screening trial demonstrated superior participation and detection rates for FIT compared to gFOBT screening. FIT screening should therefore be strongly preferred over gFOBT screening. FS screening demonstrated a higher diagnostic yield per 100 invitees than both FOBTs.

A prognostic model for amputation in critical lower limb ischemia

In a (negative) multicenter randomized trial on management for inoperable critical lower limb ischemia, comparing spinal cord stimulation and best medical treatment, a number of pre-defined factors were analyzed for prognostic value. We included a radiological arterial disease score, modified from the SVS/ISCVS runoff score. The purpose of this analysis was to evaluate clinical factors and commonly used circulatory measurements for prognostic modeling in patients with critical lower limb ischemia. We determined the incidence of amputation and its relation to various pre-defined risk factors. A total of 120 patients with critical limb ischemia were included in the study. The integrity of circulation in the affected limb was evaluated on five levels: suprainguinal, infrainguinal, popliteal, infrapopliteal and pedal. A total radiological arterial disease score was calculated from 1 (full integrity of circulation) to 20 (maximally compromised state). We used Cox regression analysis to quantify prognostic effects and differential treatment (predictive) effects. Major amputation occurred in 33% of the patients at 6 months and in 51% at 2 years. The presence of ischemic skin lesions and the radiological arterial disease score were independent prognostic factors for amputation. Patients with ulcerations or gangrene had a higher amputation risk (hazard ratio 2.38, p = 0.018 and 2.30, p = 0.036 respectively) as well as patients with a higher radiological arterial disease score (hazard ratio 1.17 per increment, p = 0.003). We did not observe significant interactions between prognostic factors and the effect of spinal cord stimulation. In conclusion, in patients with critical lower limb ischemia, the presence of ischemic skin lesions and the described radiological arterial disease score can be used to estimate amputation risk.

What is the evidence on efficacy of spinal cord stimulation in (subgroups of) patients with critical limb ischemia?

The use of spinal cord stimulation (SCS) has been advocated for the management of ischemic pain and the prevention of amputations in patients with inoperable critical limb ischemia (CLI), although data on benefit are conflicting. Several reports described apparently differential treatment effects in subgroups. The purpose of this study was to analyze the data on the efficacy of SCS and to clarify preselection issues. Five randomized trials have been performed with a total number of 332 patients. Primary outcome measures were mortality and limb survival. In the largest multicenter randomized trial (n = 120), which compared SCS treatment and best medical treatment alone in patients with inoperable CLI, we determined the incidence of amputation and its relation to various predefined risk factors. We used Kaplan-Meier and Cox regression analyses to quantify prognostic effects and differential treatment effects. Meta-analysis yielded a relative risk for amputation of 0.79 and a risk difference of -0.07 (p = 0.15). The risk factor analysis clearly showed that patients with ischemic skin lesions (ulcerations or gangrene) had a worse prognosis (i.e., higher risk of amputation) (relative risk 2.30, p = 0.01). We did not observe significant interactions between this prognostic factor (or any other) and the effect of SCS. The analysis did not indicate a subgroup of patients who might specifically be helped by SCS. Meta-analysis including all randomized data shows insufficient evidence for higher efficacy of SCS treatment compared with best medical treatment alone. Although some factors provide prognostic information as to the risk of amputation in patients with CLI, there are no data supporting a more favorable treatment effect in any group.

[Insufficient basis for the inclusion of Human papillomavirus vaccination in the National Immunisation Programme in The Netherlands]

The Dutch Minister of Health, Welfare and Sportintends to implement Human papillomavirus (HPV) vaccination for 12-year-old girls and catch-up vaccination for 13- 16-year-old girls, as part of the National Immunisation Programme from September 2009 onwards. However, due to a well-organised screening programme, cervical cancer is not an important public health problem in the Netherlands any more, which limits the possible impact of HPV vaccination. Vaccine trials thus far have involved a relatively small number ofparticipants with limited follow-up, so the efficacy of the vaccine in preventing cervical cancer is not yet known. There are no data on frequency and severity of possible adverse events and the vaccine has not yet been tested in the intention-to-vaccinate group of 12-year-old girls. Even when we assume that HPV16/18-related cervical cancer is prevented on a lifelong basis, the cost-effectiveness ratio of HPV vaccination is estimated not to be favourable. In conclusion, HPV vaccination does not seem to be urgent in the Netherlands. Therefore we advise studying the safety in 12-year-old girls first while at the same time waiting for the longer follow-up results of ongoing trials.

MUSIDH, multiple use of simulated demographic histories, a novel method to reduce computation time in microsimulation models of infectious diseases

Microsimulation of infectious diseases requires simulation of many life histories of interacting individuals. In particular, relatively rare infections such as leprosy need to be studied in very large populations. Computation time increases disproportionally with the size of the simulated population. We present a novel method, MUSIDH, an acronym for multiple use of simulated demographic histories, to reduce computation time. Demographic history refers to the processes of birth, death and all other demographic events that should be unrelated to the natural course of an infection, thus non-fatal infections. MUSIDH attaches a fixed number of infection histories to each demographic history, and these infection histories interact as if being the infection history of separate individuals. With two examples, mumps and leprosy, we show that the method can give a factor 50 reduction in computation time at the cost of a small loss in precision. The largest reductions are obtained for rare infections with complex demographic histories.

[Intra-uterine insemination with controlled ovarian hyperstimulation compared to an expectant management in couples with unexplained subfertility and an intermediate prognosis: a randomised study]

OBJECTIVE

Intrauterine insemination (IUI) with controlled ovarian hyperstimulation (COH) is commonly used as treatment of first choice in couples with unexplained subfertility. This treatment should only be applied when there is a realistic increase in chance of pregnancy, particularly because it carries the increased risk of multiple pregnancies. We evaluated the effectiveness of IUI with COH relative to expectant management in couples with unexplained subfertility and an intermediate prognosis of a spontaneous ongoing pregnancy.

DESIGN

Multicentre randomised clinical study.

METHOD

253 couples with unexplained subfertility and a probability of a spontaneous ongoing pregnancy of 30% to 40% within 12 months, were randomly assigned to IUI with COH for 6 months or expectant management for 6 months. The primary endpoint of our study was ongoing pregnancy within 6 months. Analysis was carried out according to the intention to treat principle. This study was registered with the Dutch Trial Register and has the International Standard Randomised Clinical Trial number ISRCTN72675518.

RESULTS

Of the 253 couples included, 127 couples were allocated to IUI with COH and 126 to expectant management. In the intervention group, 42 women (33%) conceived, of which 29 pregnancies were ongoing (23%). In the expectant management group, 40 women (32%) conceived, of which 34 pregnancies were ongoing (27%) (relative risk: 0.85; 95% CI: 0.63-1.1). In the expectant management group one twin pregnancy occurred and in the intervention group one woman conceived twins and one a triplet.

CONCLUSION

A substantial beneficial effect of IUI with COH in couples with unexplained subfertility and an intermediate prognosis can be excluded. Expectant management for a period of 6 months therefore appears justified in these couples.

Population screening for colorectal cancer: faeces, endoscopes or X-rays?

Colorectal carcinoma (CRC) is a common cancer and the second most common cause of death. The therapeutic costs for this disease will continue to rise due to an increasing incidence and the introduction of new chemotherapeutic modalities. Colorectal carcinoma is preceded by precursor lesions, which can be used as a target for early detection and therapy. Biennial population screening with faecal occult blood tests (FOBT) lowers CRC mortality with 14-18%. Five year screening with flexible sigmoidoscopy is a cost-effective alternative, which yields a higher preventive effect when similar participation rates are achieved. Screening colonoscopy has the advantage of examination of the complete colon but disadvantages are the high participant burden and the higher demand for endoscopic personnel and endoscopy units. Future screening modalities like faecal DNA markers and CT colonography are promising but need further improvement. In Europe, faecal occult blood testing and flexible sigmoidoscopy are currently the most suitable screening modalities for colorectal cancer screening.

Predicting ongoing pregnancy chances after IVF and ICSI: a national prospective study

BACKGROUND

The Dutch IVF guideline suggests triage of patients for IVF based on diagnostic category, duration of infertility and female age. There is no evidence for the effectiveness of these criteria. We evaluated the predictive value of patient characteristics that are used in the Dutch IVF guideline and developed a model that predicts the IVF ongoing pregnancy chance within 12 months.

METHODS

In a national prospective cohort study, pregnancy chances after IVF and ICSI treatment were assessed. Couples eligible for IVF or ICSI were followed during 12 months, using the databases of 11 IVF centres and 20 transport IVF clinics. Kaplan-Meier analysis was performed to estimate the cumulative probability of an ongoing pregnancy, and Cox regression was used for assessing the effects of predictors of pregnancy.

RESULTS

4928 couples starting IVF/ICSI treatment were prospectively followed. On average, couples had 1.8 cycles in 12 months for both IVF and ICSI. The 1-year probability of ongoing pregnancy was 44.8% (95% CI 42.1-47.5%). ICSI for severe oligospermia had a significantly higher ongoing pregnancy rate than IVF indicated treatments, with a multivariate Hazard ratio (HR) of 1.22 (95% CI 1.07-1.39). The success rates were comparable for all diagnostic categories of IVF. The highest success rate is at age 30, with a slight decline towards younger women and women up to 35 and a sharp drop after 35. Primary subfertility with a HR of 0.90 (95% CI 0.83-0.99) and duration of subfertility with a HR of 0.97 (95% CI 0.95-0.99) per year significantly affected the pregnancy chance.

CONCLUSIONS

The most important predictors of the pregnancy chance after IVF and ICSI are women's age and ICSI. The diagnostic category is of no consequence. Duration of subfertility and pregnancy history are of limited prognostic value.

[The consequences of postponing pregnancy]

The postponement of childbearing is determined by societal factors and is related to the fact that it is often difficult for women to combine an education, a job or a career with having children and taking care of a family. Especially gynaecologists are increasingly confronted with women who undergo the medical consequences of such postponement. Postponing the first pregnancy is accompanied by an increased risk of unwanted infertility. If women do succeed in becoming pregnant later in life, there is an increased risk of complications during pregnancy and delivery. The child runs a greater risk of chromosomal aberrations and of mental and physical handicaps related to increased numbers of premature births and fertility treatments. All these problems begin to increase after age 30, but especially after age 35. Finally, the risk of breast cancer is also increased if a woman delays the birth of her first child or remains childless.

[The 1996 revision of the Dutch cervical cancer screening programme: increased coverage, fewer repeat smears and less opportunistic screening]

OBJECTIVE

Comparison of the indicators of effectiveness and efficiency of the Dutch national cervical cancerscreening programme in 2003 and 1994, the last year before implementation of important changes in the medical and organisational guidelines.

DESIGN

Descriptive.

METHOD

Data on all Pap smears made in 1994 and 2003 were retrieved from the Pathologic Anatomical National Automated Archive (PALGA), together with the matching cytological and histological follow-up until April 2004. In order to calculate the 5-year coverage, the number of women that had had a smear taken was placed in the numerator and divided by the number of women that had been invited for the screening programme during those 5 years.

RESULTS

The 5-year coverage in the age range 30-64 years increased from 69 in 1994 to 77% in 2003. The percentage of smears resulting in a recommendation for a repeat smear decreased from 10 to 2. The percentage of timely compliance with recommendations for a repeat smear increased from 47 to 86, while that of smears with an immediate referral recommendation remained the same (about go). There was a sharp decrease in screening outside of the target-age range and screening with too short an interval. As a consequence, despite the higher coverage, the total number of smears decreased.

CONCLUSION

The changes in the Dutch cervical cancerscreening programme in 1996 with regard to participation, the number of and compliance with recommendations for repeat smears, and screening activity outside of the target group were accompanied by significant improvements in agreement with the goals of the revision. The potential consequences for the effectiveness of the screening programme (reduction of cervical cancer mortality) will become apparent in the years to come.

Female sex workers and unsafe sex in urban and rural Nyanza, Kenya: regular partners may contribute more to HIV transmission than clients

OBJECTIVES

To compare the sexual behaviour of female sex workers in urban and rural areas in Nyanza province in Kenya, and to compare their unsafe sex with clients and with regular partners.

METHODS

In a cross-sectional study among 64 sex workers (32/32 in urban/rural areas), sex workers kept a sexual diary for 14 days after being interviewed face-to-face.

RESULTS

Most sex workers were separated/divorced and had one or two regular partners, who were mostly married to someone else. Sex workers in Kisumu town were younger, had started sex work at an earlier age, and had more clients in the past 14 days than rural women (6.6 vs. 2.4). Both groups had an equal number of sex contacts with regular partners (4.7). With clients, condom use was fairly frequent (75%) but with regular partners, it was rather infrequent (<40%). For both urban and rural areas, the mean number of sex acts in which no condom was used was greater for regular partners (3.2 and 2.8 respectively) than for clients (1.9 and 1.0 respectively).

CONCLUSIONS

Sex workers in urban and rural areas of Nyanza province practise more unsafe sex with regular partners than with clients. Interventions for sex workers should also focus on condom use in regular partnerships.

Effects of an individualized multimedia computer program for health education in patients with a recent minor stroke or transient ischemic attack - a randomized controlled trial

BACKGROUND

Transient ischemic attack (TIA) and stroke patients often show a striking lack of knowledge about their disease. We developed a computer program that provided health education fitting the educational level, risk profile and symptoms of patients and evaluated it in a randomized controlled trial.

METHODS

Transient ischemic attack or minor stroke patients were allocated to health education by a physician (n = 32) or to a combination of education by a physician and the computer program (n = 33). Knowledge was tested by means of a questionnaire at 1 and 12 weeks after inclusion. The maximum possible score was 71 points.

RESULTS

Overall knowledge was low - the mean score was 43.6 at 1 week and 42.0 points at 12 weeks for both the groups. The intervention group had slightly better scores at 1 week after using the computer program, 45.4 vs 41.5 (P = 0.09), with the difference increasing to 4.3 points after (post-hoc) adjustment for age and level of education (P = 0.06). After 12 weeks, the score in the intervention group dropped significantly to 42.0 points (P = 0.05), and was no longer different from the standard group.

CONCLUSION

This study did not show a lasting effect of health education by an individualized computer program on the knowledge of TIA and minor stroke patients.

Sequence variations in the MBL gene and their relationship to pulmonary tuberculosis in the Chinese Han population

SETTING

The mannan binding lectin (MBL) gene is thought to play a role in human innate immune response to tuberculosis (TB) infection.

OBJECTIVE

To investigate the possible association between MBL sequence variants and TB infection in the Chinese Han population.

DESIGN

A total of 152 male pulmonary tuberculosis (PTB) patients and 293 healthy male subjects were recruited. Six MBL single nucleotide polymorphisms (SNPs) (A/B, A/C, A/D, H/L, Y/X and P/Q) were genotyped and haplotyped using the combined analysis of polymerase chain reaction using sequence-specific primers (PCR-SSP) and the PCR-sequence specific oligonucleotide probe (PCR-SSOP) assay. The genotype and haplotype frequencies were compared between TB cases and controls using an unconditional logistic regression model.

RESULTS

Neither the genotypes nor the haplotypes of the five loci were significantly associated with the disease when considered individually. After the haplotypes were regrouped, however, the XB haplotype group coding for diminished MBL levels was present at a significantly higher frequency in the patients compared with the YA group (OR 1.57, 95% CI 1.02-2.41, P < 0.05).

CONCLUSION

No convincing evidence of association between MBL sequence variants and PTB was observed individually, although the low-producing XB haplotype group may serve as a minor risk factor for PTB infection in the male Chinese Han population.

Pregnancy is predictable: a large-scale prospective external validation of the prediction of spontaneous pregnancy in subfertile couples*

BACKGROUND

Prediction models for spontaneous pregnancy may be useful tools to select subfertile couples that have good fertility prospects and should therefore be counselled for expectant management. We assessed the accuracy of a recently published prediction model for spontaneous pregnancy in a large prospective validation study.

METHODS

In 38 centres, we studied a consecutive cohort of subfertile couples, referred for an infertility work-up. Patients had a regular menstrual cycle, patent tubes and a total motile sperm count (TMC) >3 x 10(6). After the infertility work-up had been completed, we used a prediction model to calculate the chance of a spontaneous ongoing pregnancy (www.freya.nl/probability.php). The primary end-point was time until the occurrence of a spontaneous ongoing pregnancy within 1 year. The performance of the pregnancy prediction model was assessed with calibration, which is the comparison of predicted and observed ongoing pregnancy rates for groups of patients and discrimination.

RESULTS

We included 3021 couples of whom 543 (18%) had a spontaneous ongoing pregnancy, 57 (2%) a non-successful pregnancy, 1316 (44%) started treatment, 825 (27%) neither started treatment nor became pregnant and 280 (9%) were lost to follow-up. Calibration of the prediction model was almost perfect. In the 977 couples (32%) with a calculated probability between 30 and 40%, the observed cumulative pregnancy rate at 12 months was 30%, and in 611 couples (20%) with a probability of >or=40%, this was 46%. The discriminative capacity was similar to the one in which the model was developed (c-statistic 0.59).

CONCLUSIONS

As the chance of a spontaneous ongoing pregnancy among subfertile couples can be accurately calculated, this prediction model can be used as an essential tool for clinical decision-making and in counselling patients. The use of the prediction model may help to prevent unnecessary treatment.

Do clinical prediction models improve concordance of treatment decisions in reproductive medicine?

OBJECTIVE

To assess whether the use of clinical prediction models improves concordance between gynaecologists with respect to treatment decisions in reproductive medicine.

DESIGN

We constructed 16 vignettes of subfertile couples by varying fertility history, postcoital test, sperm motility, follicle-stimulating hormone level and Chlamydia antibody titre.

SETTING

Thirty-five gynaecologists estimated three probabilities, i.e. the 1-year probability of spontaneous pregnancy, the pregnancy chance after intrauterine insemination (IUI) and the pregnancy chance after in vitro fertilisation (IVF). Subsequently they proposed therapeutic regimens for these 16 fictional couples, i.e. expectant management, IUI or IVF. Three months later, the participant gynaecologists again had to propose therapeutic regimes for the same 16 fictional cases but this time accompanied by pregnancy chances obtained from prediction models: predictions on spontaneous pregnancy, IUI and IVF.

POPULATION

Thirty-five gynaecologists working in academic and nonacademic hospitals in the Netherlands.

METHODS

Setting section. Main outcome measures The concordance between gynaecologists of probability estimates, expressed as interclass correlation coefficient (ICC) and the concordance between gynaecologists of treatment decisions, analysed by calculating Cohen's kappa (kappa).

RESULTS

The gynaecologists differed widely in estimating pregnancy chances (ICC: 0.34). Furthermore, there was a huge variation in the proposed therapeutic regimens (kappa: 0.21). The treatment decisions made by gynaecologists were consistent with the ranking of their probability estimates. When prediction models were used, the concordance (kappa) for treatment decisions increased from 0.21 to 0.38. The number of gynaecologists counselling for expectant management increased from 39 to 51%, whereas counselling for IVF dropped from 23 to 14%.

CONCLUSION

Gynaecologists differed widely in their estimation of prognosis in 16 fictional cases of subfertile couples. Their therapeutic regimens showed likewise huge variation. After confrontation with prediction models in the same 16 fictional cases, the proposed therapeutic regimens showed only slightly better concordance. Therefore a simple introduction of validated prediction models is insufficient to introduce concordant management between doctors.

Spinal cord stimulation is not cost-effective for non-surgical management of critical limb ischaemia

OBJECTIVE

To quantify the costs of treatment in critical limb ischaemia (CLI) and to compare costs and effectiveness of two treatment strategies: spinal cord stimulation (SCS) and best medical treatment.

METHODS

One hundred and twenty patients with CLI not suitable for vascular reconstruction were randomised to either SCS in addition to best medical treatment or best medical treatment alone. Primary outcomes were mortality, amputation and cost. Cost analysis was based on resources used by patients for 2 years after randomisation. Both medical and non-medical costs were included.

RESULTS

Patient and limb survival were similar in the two treatment groups. Costs of in-hospital-stay and institutional rehabilitation constituted the predominant part (+/-70%) of the total costs of medical care in CLI. Cost of SCS-implantation and complications (7950 euro per patient) exceeded by far cost due to amputation procedures (410 euro per patient). The total costs of treatment were 36,600 euro per patient over 2 years for the SCS-group vs. 28,700 euro for best medical treatment alone (28% higher for SCS-group, p=0.009).

CONCLUSIONS

Total costs of treatment in CLI are high. Major components are hospital and rehabilitation costs. In contrast to recent reviews, there were no long-term benefits of SCS-treatment. Therefore, cost-effectiveness is reduced to cost-minimisation and SCS-treatment is considerably more expensive than best medical treatment.

Comparison of different treatment strategies in IVF with cumulative live birth over a given period of time as the primary end-point: methodological considerations on a randomized controlled non-inferiority trial

BACKGROUND

We discuss methodological considerations related to a study in IVF, which compares the effectiveness, health economics and patient discomfort of two treatment strategies that differ in both ovarian stimulation and embryo transfer policies.

METHODS

This was a randomized controlled clinical trial in two large Dutch IVF centres. The tested treatment strategies are: mild ovarian stimulation [including gonadotrophin-releasing hormone (GnRH) antagonist co-treatment] together with the transfer of one embryo, versus conventional stimulation (with GnRH agonist long protocol co-treatment) and the transfer of two embryos. Outcome measures are: (i) pregnancies resulting in term live birth; (ii) total costs per term live birth; and (iii) patient stress/discomfort per started IVF treatment, over a 12 month period. Power considerations for this study were an overall cumulative live birth rate of 45% for the conventional treatment strategy, with non-inferiority of the mild treatment strategy defined as a live birth rate no more than 12.5% lower compared with the conventional study arm. For a power of 80% and alpha of 0.05, 400 subjects are required.

RESULTS

As planned, from February 2002 until February 2004, 410 patients were enrolled.

CONCLUSIONS

This effectiveness study applies an integrated medical, health economics and psychological approach with term live birth over a given period of time after starting IVF as the end-point. Complete and timely patient enrolment vindicates many of the design decisions.

Use of stimulated serum estradiol measurements for the prediction of hyperresponse to ovarian stimulation in in vitro fertilization (IVF)

PURPOSE

In ovarian stimulation an exaggerated ovarian response is often seen and is related to medical complications, such as ovarian hyperstimulation syndrome (OHSS), and increased patient discomfort. If it were possible to identify hyperresponders at an early stage of the stimulation phase, adaptation of the stimulation protocol would become feasible to minimize potential complications. Therefore, we studied the usefulness of measuring stimulated serum estradiol (E2) levels in predicting ovarian hyperresponse.

METHODS

A total of 109 patients undergoing their first IVF treatment cycle using a long protocol with GnRH agonist was prospectively included. The E2 level was evaluated on day 3 and 5 of the stimulation phase. Two outcome measures were defined. The first was ovarian hyperresponse (collection of > or = 15 oocytes at retrieval and/or peak E2 > 10000 pmol/L, or cancellation due to > or = 30 follicles growing and/or peak E2 > 15000 pmol/L, or OHSS developed). The second outcome measure comprised a subgroup representing the more severe hyperresponders. named extreme-response (cancellation or OHSS developed).

RESULTS

The data of 108 patients were analyzed. The predictive accuracy of E2 measured on stimulation day 3 towards ovarian hyperresponse was clearly lower than that of E2 measured on stimulation day 5 (area under the receiver operating characteristic curve (ROCAUC) 0.75 and 0.81, respectively). For extreme-response the predictive accuracy of E2 measured on stimulation day 3 or 5 was comparable (ROCAUC 0.81 and 0.82, respectively). For both outcome measures the stimulated E2 tests yielded only acceptable specificity with moderate sensitivity at higher cutoff levels. Prediction of extreme-response seemed slightly more effective due to a lower error rate.

CONCLUSIONS

There is a significant predictive association between E2 levels measured on stimulation day 3 and 5 and both ovarian hyperresponse and extreme-response in IVF. However, the clinical value of stimulated E2 levels for the prediction of hyperresponse is low because of the modest sensitivity and the high false positive rate. For the prediction of extreme-response the clinical value of stimulated E2 levels is moderate.

The effect of treatment on health-related quality of life in patients with hypertension and renal artery stenosis

The quality of life in patients with hypertension is considered to be impaired mainly by side effects of antihypertensive drug therapy. Since balloon angioplasty for renal artery stenosis has a medication-sparing effect, it may lead to an improvement in quality of life. The objective of the study is to compare the effect of antihypertensive drug therapy and balloon angioplasty on quality of life in patients with hypertension and renal artery stenosis. We compared the quality of life in 56 patients randomised to balloon angioplasty to that in 50 patients randomised to antihypertensive drug therapy after 3 and 12 months of follow-up. Quality of life was measured using a questionnaire on physical symptoms associated with hypertension and antihypertensive drugs, and two generic health questionnaires (MOS Survey and EuroQol instrument). After follow-up, the patients who underwent angioplasty used less antihypertensive drugs than the patients who were treated with antihypertensive drugs only (mean+/-s.d., 1.9+/-0.9 vs 2.5+/-1.0 drugs after 3 months, P=0.002). They reported similar physical complaints, however, and a similar quality of life. The results after 12 months of follow-up were the same. In conclusion, for patients with hypertension and renal artery stenosis, the decrease in antihypertensive medication after intervention is too small to lead to a detectable improvement in quality of life.

Equally valid models gave divergent predictions for mortality in acute myocardial infarction patients in a comparison of logistic [corrected] regression models

OBJECTIVE

Models that predict mortality after acute myocardial infarction (AMI) contain different predictors and are based on different populations. We studied the agreement and validity of predictions for individual patients.

STUDY DESIGN AND SETTING

We compared predictions from five predictive logistic regression models for short-term mortality after AMI. Three models were developed previously, and two models were developed in the GUSTO-I data, where all five models were applied (n =40,830, 7.0% 30-day mortality). Agreement was studied with weighted kappa statistics of categorized predictions. Validity was assessed by comparing observed frequencies with predictions (indicating calibration) and by the area under the receiver operating characteristic curve (AUC), indicating discriminative ability.

RESULTS

The predictions from the five models varied considerably for individual patients, with low agreement between most (kappa <0.6). Risk predictions from the three previously developed models were on average too high, which could be corrected by re-calibration of the model intercept. The AUC ranged from 0.76-0.78 and increased to 0.78-0.79 with re-estimated regression coefficients that were optimal for the GUSTO-I patients. The two more detailed GUSTO-I based models performed better (AUC approximately 0.82).

CONCLUSION

Models with different predictors may have a similar validity while the agreement between predictions for individual patients is poor. The main concerns in the applicability of predictive models for AMI should relate to the selected predictors and average calibration.

Dynamic decision analysis to determine optimal treatment duration in chronic hepatitis C

BACKGROUND

Current guidelines for stopping treatment of chronic hepatitis C are based on hepatitis C ribonucleic acid measurements at 12 and 24 weeks.

AIM

To explore an alternative approach for making individualized recommendations about treatment duration, based on simple alanine aminotransferase tests and on cost-per-cure.

METHODS

We analysed individual patient data from 13 randomized, controlled trials with interferon alone or combined with ribavirin. Using multiple logistic regression, we built a model that estimated the probability of sustained virological response for treatment durations of 24 and 48 weeks. Decisions to prolong treatment were based on an increase in probability of sustained virological response. If the increase was 10%, the cost-per-cure became decisive with a limit of 50,000.

RESULTS

Noncirrhotics with genotype 2 or 3 did not benefit when treatment was continued beyond 24 weeks. Sustained virological response rates in cirrhotic patients increased by 14-47% if treatment was continued up to 48 weeks. In noncirrhotic genotype 1 or 4 patients who had elevated alanine aminotransferase levels at week 4, the probability of sustained virological response increased by <10% if treatment was continued up to 48 weeks; the cost-per-cure for these patients would exceed 50,000.

CONCLUSION

The dynamics of alanine aminotransferase levels and cost-per-cure provides a useful alternative to determine the duration of therapy in chronic hepatitis C.

Determinants of the impact of sexually transmitted infection treatment on prevention of HIV infection: a synthesis of evidence from the Mwanza, Rakai, and Masaka intervention trials

Community-randomized trials in Mwanza, Tanzania, and Rakai and Masaka, Uganda, suggested that population characteristics were an important determinant of the impact of sexually transmitted infection (STI) treatment interventions on incidence of human immunodeficiency virus (HIV) infection. We performed simulation modeling of HIV and STI transmission, which confirmed that the low trial impact in Rakai and Masaka could be explained by low prevalences of curable STI resulting from lower-risk sexual behavior in Uganda. The mature HIV epidemics in Uganda, with most HIV transmission occurring outside core groups with high STI rates, also contributed to the low impact on HIV incidence. Simulated impact on HIV was much greater in Mwanza, although the observed impact was larger than predicted from STI reductions, suggesting that random error also may have played some role. Of proposed alternative explanations, increasing herpetic ulceration due to HIV-related immunosuppression contributed little to the diminishing impact of antibiotic treatment during the Ugandan epidemics. The strategy of STI treatment also was unimportant, since syndromic treatment and annual mass treatment showed similar effectiveness in simulations of each trial population. In conclusion, lower-risk behavior and the mature HIV epidemic explain the limited impact of STI treatment on HIV incidence in Uganda in the 1990s. In populations with high-risk sexual behavior and high STI rates, STIs treatment interventions may contribute substantially to prevention of HIV infection.

Colorectal cancer risk after colonoscopic polypectomy: a population-based study and literature search

Adenoma patients are considered to be at an elevated risk for colorectal cancer, even after their adenomas have been removed. The aim of this study was to estimate the colorectal cancer risk after colonoscopic polypectomy compared with age- and gender-matched general population controls. Colorectal cancer incidence was studied in 553 consecutive patients without cancer whose adenomas were colonoscopically removed in the endoscopy department of a general hospital. The colorectal cancer relative risk in these patients was 0.9 (0.3-2.0). A literature search was performed to identify all published studies on relative colorectal cancer risk after polypectomy. The relative risk estimates in seven other studies ranged from 0.2 (0.1-0.6) to 1.3 (0.6-2.3). The difference can, be explained partially by the inclusion or exclusion of patients with large sessile polyps and other factors. Our review shows that colorectal cancer risk after colonoscopic polypectomy does not exceed the risk in the general population.

Towards less confusing terminology in reproductive medicine: a proposal

The use of the term "infertility" and related terms in reproductive medicine is reviewed. Current terminology is found to be ambiguous, confusing and misleading. We recommend that the fertility investigation report of a couple should consist of statements concerning description, diagnosis and prognosis. The description concerns the duration of non-pregnancy before consulting the clinician. A system for prognostic grading is proposed. The fertility investigation report forms the basis for further action, including the possibility of waiting with treatment in case of almost normal or only slightly reduced fertility. The use of the terms infertility, subfertility and fecundity is not necessary, and it is recommended to avoid them.

Expected poor responders on the basis of an antral follicle count do not benefit from a higher starting dose of gonadotrophins in IVF treatment: a randomized controlled trial

BACKGROUND

The aim of this study was to evaluate the effect of doubling the starting dose of gonadotrophins on the ovarian response in IVF patients with a low antral follicle count (AFC).

METHODS

Fifty-two patients with an AFC of <5 follicles of 2-5 mm diameter before starting their first IVF cycle participated in this randomized controlled trial. They were randomized by opening a sealed envelope, receiving either 150 IU (group I, n = 26) or 300 IU (group II, n = 26) of rFSH as a starting dose. The main outcome measures of the study were number of oocytes, poor response (<4 oocytes at retrieval or cancellation due to insufficient follicle growth) and ongoing pregnancy (12 weeks of gestation).

RESULTS

The groups were comparable regarding patient characteristics and outcome of the IVF treatment. The median number of oocytes collected was 3 for both groups (P = 0.79). The difference in the mean number of oocytes was 0.3 oocytes in favour of group I (P=0.69). Sixty-five per cent of the patients in group I experienced a poor response and 62% in group II. The ongoing pregnancy rate was 8% in group I and 4% in group II (P = 0.55).

CONCLUSIONS

Expected poor response patients, defined as patients with an AFC <5, are likely not to benefit from a higher starting dose of gonadotrophins in IVF.

Meta-analysis of age-prevalence patterns in lymphatic filariasis: no decline in microfilaraemia prevalence in older age groups as predicted by models with acquired immunity

The role of acquired immunity in lymphatic filariasis is uncertain. Assuming that immunity against new infections develops gradually with accumulated experience of infection, models predict a decline in prevalence after teenage or early adulthood. A strong indication for acquired immunity was found in longitudinal data from Pondicherry, India, where Mf prevalence was highest around the age of 20 and declined thereafter. We reviewed published studies from India and Subsaharan Africa to investigate whether their age-prevalence patterns support the models with acquired immunity. By comparing prevalence levels in 2 adult age groups we tested whether prevalence declined at older age. For India, comparison of age groups 20–39 and 40+ revealed a significant decline in only 6 out of 53 sites, whereas a significant increase occurred more often (10 sites). Comparison of older age groups provided no indication that a decline would start at a later age. Results from Africa were even more striking, with many more significant increases than declines, irrespective of the age groups compared. The occurrence of a decline was not related to the overall Mf prevalence and seems to be a chance finding. We conclude that there is no evidence of a general age-prevalence pattern that would correspond to the acquired immunity models. The Pondicherry study is an exceptional situation that may have guided us in the wrong direction.

Comparison of outcomes after aortic valve replacement with a mechanical valve or a bioprosthesis using microsimulation

BACKGROUND

Mechanical valves and bioprostheses are widely used for aortic valve replacement. Though previous randomised studies indicate that there is no important difference in outcome after implantation with either type of valve, knowledge of outcomes after aortic valve replacement is incomplete.

OBJECTIVE

To predict age and sex specific outcomes of patients after aortic valve replacement with bileaflet mechanical valves and stented porcine bioprostheses, and to provide evidence based support for the choice of prosthesis.

METHODS

Meta-analysis of published results of primary aortic valve replacement with bileaflet mechanical prostheses (nine reports, 4274 patients, and 25,726 patient-years) and stented porcine bioprostheses (13 reports, 9007 patients, and 54,151 patient-years) was used to estimate the annual risks of postoperative valve related events and their outcomes. These estimates were entered into a microsimulation model, which was employed to calculate age and sex specific outcomes after aortic valve replacement.

RESULTS

Life expectancy (LE) and event-free life expectancy (EFLE) for a 65 year old man after implantation with a mechanical valve or a bioprosthesis were 10.4 and 10.7 years and 7.7 and 8.4 years, respectively. The lifetime risk of at least one valve related event for a mechanical valve was 48%, and for a bioprosthesis, 44%. For LE and EFLE, the age crossover point between the two valve types was 59 and 60 years, respectively.

CONCLUSIONS

Meta-analysis based microsimulation provides insight into the long term outcome after aortic valve replacement and suggests that the currently recommended age threshold for implanting a bioprosthesis could be lowered further.