Cumulative incidence rate of medical consultation for fecundity problems--analysis of a prevalent cohort using competing risks.

STUDY QUESTION: What is the incidence of medical consultation for fecundity problems in the French population, taking into account pregnancy occurrence and resumption of contraceptive use?. SUMMARY ANSWER: Considering the occurrence of a pregnancy and resumption of use of contraception as competing risks, the cumulative incidence rate of medical consultation for fecundity problems was 9.0% [95% confidence interval (CI): 6.5%; 11.9%] after 12 months of unprotected intercourse and 12.2% [95% CI: 9.6%; 15.3%] after 24 months. WHAT IS KNOWN ALREADY: Estimates of the prevalence of medical consultation due to involuntary infertility among couples who have sought a pregnancy for more than 12 months range from 25 to 50%. Most of the studies however are limited by retrospective data collection, without considering the duration of time since the beginning of the period of unprotected intercourse (PUI) and without considering medical consultation for fecundity problems as a competing risk. STUDY DESIGN, SIZE, DURATION: This study is based on the Observatory of Fecundity in France survey, a population-based probability survey designed to estimate the frequency of involuntary infertility on a nationwide basis and to explore the associations with environmental factors. Women answered two telephone questionnaires, the first at the time of enrolment in 2007, the second at follow-up 1 year later. The current analysis was performed among a subsample of 6577 women recruited before or during a PUI and followed-up for 1 year. PARTICIPANTS/MATERIALS, SETTING, METHODS: The study sample comprised 940 women aged 18-44 years who had a PUI between the time of enrolment and the 1-year follow-up, and who had not consulted a physician for fecundity problems for the current PUI prior to enrolment. Women reported all the medical consultations they had because of difficulties becoming pregnant during the current PUI. The date of each consultation was carefully assessed. In France, women can consult a gynaecologist directly without referral by their general practitioner. The occurrence of a pregnancy and resumption of contraceptive use were considered as informative censoring events, using a competing risk model. MAIN RESULTS AND THE ROLE OF CHANCE: Using the competing risk survival model, the cumulative incidence rate of first consultation was 9.0% [95% CI: 6.5%; 11.9%] 12 months after the start of the PUI and 12.2% [95% CI: 9.6%; 15.3%] after 24 months. The Kaplan-Meier method, which does not take competing risks into account, yielded substantially higher estimates: 26.0% [95% CI: 18.8%; 32.5%] at 12 months and 56.8% [95% CI: 44.2%; 66.6%] at 24 months. Among the 219 women who had attempted to become pregnant for at least 12 months, cumulative incidences of first medical consultations were 28.2% [18.7-38.9%] 24 months after the start of the PUI, and 31.2% [21.3- 42.4%] after 36 months. The rates were higher among nulliparous but non-nulligravid women, followed by nulligravid women, as compared with parous women. Age was not strongly related to the occurrence of medical consultation. LIMITATIONS, REASONS FOR CAUTION: The main limitation of this study is the number of women lost to follow-up (29.7%). In addition, results regarding the absence of an age effect should be taken with caution as few women in our study were aged over 35 years. Although such an attrition rate is commonly observed in prospective studies in the general population, it could have induced a selection bias that may have led to an underestimation of the rates of medical consultation. Sensitivity analyses, using the inverse probability weighting method suggest that our results are unlikely to be biased. WIDER IMPLICATIONS OF THE FINDINGS: This study reveals frequencies of medical consultation for fertility problems, which, after considering competing events such as pregnancy in a relevant statistical model, are lower than generally reported in the literature. The results also indicate the existence of a difference between the potential need and the actual use of medical care for fecundity problems. This suggests a need for studies to look for factors other than medical recommendations that may play a role in the patterns of medical seeking behaviours for fecundity problems, such as women's reproductive history, socio-economic characteristics or accessibility to infertility services. STUDY FUNDING/COMPETING INTEREST(S): The study was funded by grants from ANR (French Agency for Research, SEST call on Environmental and Occupational Health), ANSES (French Agency for Food, environmental and Occupational Health Safety, EST call on Environmental and Occupational Health), InVS (French Institute for Public Health Surveillance). The team of Environmental Epidemiology applied to Fecundity and Reproduction has been funded by an AVENIR grant from Inserm (2007). Authors declare no conflict of interest.

Trimethoprim use in early pregnancy and the risk of miscarriage: a register-based nationwide cohort study.

The antibiotic trimethoprim acts as a folate antagonist. Since trophoblasts are very sensitive to drugs that interfere with the folic acid cycle and thereby inhibit DNA synthesis, use of trimethoprim during the first trimester could be associated with miscarriage. A nationwide cohort study including all women in Denmark with a registered pregnancy between 1997 and 2005 was conducted. We used nationwide registers to identify all women giving birth, having a record of miscarriage or induced abortion. Data on exposure to trimethoprim were obtained from the National Prescription Register. Cox proportional hazard regression analysis with exposure to trimethoprim as a time-dependent variable was used to estimate the risk of miscarriage. The adjusted hazard ratio of having a miscarriage after exposure to trimethoprim in the first trimester compared to non-exposure was 2∙04 (95% confidence interval 1∙43-2∙91). Our results indicate that trimethoprim exposure in the first trimester is associated with a doubling of the hazard of miscarriage.

Estimation of the frequency of involuntary infertility on a nation-wide basis.

BACKGROUND: Assessing couple fecundity on a nation-wide basis without excluding couples who eventually remain infertile is challenging. Our aim was to describe couple fecundity (in terms of frequency of involuntary infertility) among the general population living in France. METHODS: We used a current-duration design. A random sample of 64 262 households was selected in 2007-2008, allowing us to identify 15 810 women aged 18-44 years. Eligible women (n= 1089) were those having regular sexual intercourse with a male partner, not using any method of contraception and not having delivered in the previous 3 months. These women reported information on the current duration of unprotected intercourse (CDUI, the time elapsed between the start of the period of unprotected intercourse and the time of inclusion in the study). The CDUI distribution was used to estimate the frequency of involuntary infertility, using a newly developed statistical technique that does not require couples to be followed up until the end of the period of unprotected intercourse. RESULTS: CDUI was defined for 867 women. An estimated 46% of couples had no detected pregnancy conceived during the first 6 months of unprotected intercourse [95% confidence interval (CI), 36-56%]. The proportions of couples with no detected pregnancy within 12 and 24 months were 24% (19-30%) and 11% (8-14%), respectively. CONCLUSIONS: These results constitute one of the few descriptions of the fecundity of a nation-wide representative sample of couples from the general population, not limited to couples who eventually conceived or to those resorting to medical help.

Estimation of dynamic treatment strategies for maintenance therapy of children with acute lymphoblastic leukaemia: an application of history-adjusted marginal structural models.

Childhood acute lymphoblastic leukaemia is treated with long-term intensive chemotherapy. During the latter part of the treatment, the maintenance therapy, the patients receive oral doses of two cytostatics. The doses are tailored to blood counts measured on a weekly basis, and the treatment is therefore highly dynamic. In 1992-1996, the Nordic Society of Paediatric Haematology and Oncology (NOPHO) conducted a randomised study (NOPHO-ALL-92) to investigate the effect of a new and more sophisticated dynamic treatment strategy. Unexpectedly, the new strategy worsened the outcome for the girls, whereas there were no treatment differences for the boys. There are as yet no general guidelines for optimising the treatment. On basis of the data from this study, our goal is to formulate an alternative dosing strategy. We use recently developed methods proposed by van der Laan et al. to obtain statistical models that may be used in the guidance of how the physicians should assign the doses to the patients to obtain the target of the treatment. We present a possible strategy and discuss the reliability of this strategy. The implementation is complicated, and we touch upon the limitations of the methods in relation to the formulation of alternative dosing strategies for the maintenance therapy.

The Danish Alzheimer intervention study: rationale, study design and baseline characteristics of the cohort.

BACKGROUND: There is a lack of appropriately designed trials investigating the efficacy of psychosocial interventions for patients with mild dementia and their family caregivers. This paper reports the rationale and design of the Danish Alzheimer Disease Intervention Study and baseline characteristics of the cohort. METHODS: The study was a 1-year multicentre randomized controlled rater-blinded trial with randomization to follow-up and a multifaceted semitailored intervention programme or to follow-up only (with extension of follow-up to 3 years). The intervention included a counselling programme, teaching courses, written information and logbooks. The outcomes included clinical efficacy parameters, patient satisfaction and health economic consequences. RESULTS: A total of 330 patients and their 330 caregivers were included during a period of 18 months. The majority (65.2 %) of the caregivers were spouses. At inclusion the mean age of the patients and caregivers was 76.2 and 66.0 years, respectively. CONCLUSION: The study will explore the added value of a multifaceted intervention programme and contribute to the design of future interventions for patients with mild dementia and their caregivers.

Excess use of coercive measures in psychiatry among migrants compared with native Danes.

OBJECTIVE: To investigate differences in risk of compulsory admission and other coercive measures in psychiatric emergencies among refugees and immigrants compared with that among native Danes. METHOD: A register-based retrospective cohort design. All refugees (n = 29 174) and immigrants (n = 33 287) who received residence permission in Denmark from 1.1.1993 to 31.12.1999 were included and matched 1 : 4 on age and sex with native Danes. Civil registration numbers were cross-linked to the Danish Psychiatric Central Register and the Registry of Coercive Measures in Psychiatric Treatment. RESULTS: Refugees (RR = 1.82; 95%CI: 1.45; 2.29) and immigrants (RR = 1.14; 95%CI: 0.83; 1.56) experienced higher rates of compulsory admissions than did native Danes. This was most striking for refugee men (RR = 2.00; 95%CI: 1.53; 2.61) and immigrant women (RR = 1.73; 95%CI: 1.45; 2.60). Moreover, refugees and immigrants experienced higher frequencies of other coercive measures during hospitalisation compared with native Danes. CONCLUSION: Coercive measures in psychiatry are more likely to be experienced by migrants than by native Danes.

The role of biological fertility in predicting family size.

BACKGROUND: It is plausible that a couple's ability to achieve the desired number of children is limited by biological fertility, especially if childbearing is postponed. Family size has declined and semen quality may have deteriorated in much of Europe, although studies have found an increase rather than a decrease in couple fertility. METHODS: Using four high-quality European datasets, we took the reported time to pregnancy (TTP) as the predictor variable; births reported as following contraceptive failure were an additional category. The outcome variable was final or near-final family size. Potential confounders were maternal age when unprotected sex began prior to the first birth, and maternal smoking. Desired family size was available in only one of the datasets. RESULTS: Couples with a TTP of at least 12 months tended to have smaller families, with odds ratios for the risk of not having a second child approximately 1.8, and for the risk of not having a third child approximately 1.6. Below 12 months no association was observed. Findings were generally consistent across datasets. There was also a more than 2-fold risk of not achieving the desired family size if TTP was 12 months or more for the first child. CONCLUSIONS: Within the limits of the available data quality, family size appears to be predicted by biological fertility, even after adjustment for maternal age, if the woman was at least 20 years old when the couple's first attempt at conception started. The contribution of behavioural factors to this result also needs to be investigated.

Risk of mental disorders in refugees and native Danes: a register-based retrospective cohort study.

BACKGROUND: Refugees are vulnerable to mental health disorders because of migration trauma. However, register-based prevalence studies are missing. AIMS: To investigate the risk of mental disorders among refugees compared with that among native Danes. METHOD: Refugees (n = 29,139), who received residence permission in Denmark from 1.1.1993 to 31.12.1999 were matched 1:4 on age and sex with native Danes (n = 116,556). Civil registration numbers were linked to the Danish Psychiatric Central Register to obtain data on ICD-10 diagnosis upon discharge for all first-time psychiatric hospital contacts for refugees (n = 2,120) and native Danes (n = 5,044) between 1.1.1994 and 31.12.2003. Treated prevalence was then calculated using a Poisson regression model. RESULTS: Refugee men (RR = 2.02; 95%CI = 1.75-2.34) and refugee women (RR = 1.49; 95%CI = 1.29-1.72) had higher overall risks of having a first-time psychiatric contact for mental disorders than did native Danes; specific risks of psychotic, affective and neurotic disorders were even higher. The results were most striking for refugee men, and for refugees from the former Yugoslavia, Iraq and the Middle East. CONCLUSIONS: Refugees have high rates of various mental disorders. Healthcare services should target refugees' mental health from arrival in the receiving country.

Cancer incidence among 1st generation migrants compared to native Danes--a retrospective cohort study.

STUDY AIMS: To investigate the incidence of cancer among 1st generation migrants compared to native Danes, including time trends in the risk of cancer among migrants. METHODS: A retrospective cohort study design. Migrants were matched 1:4 on age and sex with a Danish born reference population. The cohort was linked to the Danish Cancer Register and cancer cases among migrants (n=537) and native Danes (n=2829) were identified. RESULTS: The overall cancer incidence did not differ significantly between migrants from East Europe and native Danes; whereas migrants from the Middle East and North Africa had a significantly lower incidence. All migrants had a significantly lower incidence of breast and colorectal cancer but male migrants from East Europe had a significantly higher incidence of lung cancer. CONCLUSIONS: The overall cancer incidence among migrants was lower compared to native Danes. The time trends of the study are interesting and a relevant topic for further research.

Educational intervention toward preventive home visitors reduced functional decline in community-living older women.

OBJECTIVE: To investigate whether immediate effects of a 3-year educational intervention in primary health care were confirmed 18 months after the end of the intervention. STUDY DESIGN AND SETTING: A controlled 3-year intervention study in 34 Danish municipalities with randomization and intervention at municipality level. The 17 intervention municipality visitors received regular education, and GPs were introduced to a short assessment program. The effect was measured at the individual level by questions about functional ability at the end of the intervention period and 1(1/2) years later; 4,060 older adults living in the municipalities participated. We adopt the approach introduced by Dufouil et al. (2004) and treat dropouts due to death differently from dropouts from other reasons. RESULTS: Educational intervention to primary care professionals was associated with better functional ability in surviving women at the end of the intervention (odds ratio [OR]: 1.24, 95% confidence interval [CI]=1.07-1.45), from the end of the intervention until 1(1/2) years later (OR: 1.21, 95% CI=1.03-1.44) and during the total study period (OR: 1.22, 95% CI=1.06-1.42). No effects were seen in men. CONCLUSION: The effect of a brief, feasible educational intervention for primary care professionals is sustained in women 1(1/2) years after the end of the intervention.

The association between early menopause and risk of ischaemic heart disease: influence of Hormone Therapy.

UNLABELLED: Randomised clinical trials find no protection against development of ischaemic heart disease by use of Hormone Therapy (HT) after the age of 50 years. Observational studies suggest that early menopause is a risk factor for ischaemic heart disease. Yet, a clinical very relevant question is whether HT reduces this risk associated with early menopause. OBJECTIVE: To analyse whether early menopause based on various causes are independent risk factors for ischaemic heart disease, and to investigate whether the risks are modified by use of HT. METHODS: In a prospective cohort study questionnaires were mailed to Danish female nurses above 44 years of age in 1993. Information on menopause, use of HT and lifestyle was obtained. In total 19,898 (86%) nurses fulfilled the questionnaire, among them 10.533 were postmenopausal with definable menopausal age, free of previous ischaemic heart disease, stroke or cancer. Through individual linkage to national register incident cases of ischaemic heart disease were identified until end of 1998. RESULTS: Menopause below both age 40 and 45 was associated with an increased risk of ischaemic heart disease, seeming most pronounced for women who had an early ovariectomy but also among spontaneous menopausal women. Generally HT did not reduce the risk except for the early-ovariectomised women, where no increased risk of ischaemic heart disease for HT users was found. CONCLUSION: We found an increased risk of ischaemic heart disease associated with early removal of the ovaries that might be reduced with HT. The present study need confirmation from other studies but suggests that early ovariectomised women could benefit from HT.

Prognostic value of steroid hormone receptors: multivariate analysis of systemically untreated patients with node negative primary breast cancer.

The value of estrogen and progesterone receptor (ER and PgR, respectively) determinations in predicting the recurrence-free survival (RFS) has been evaluated in a group of 807 node negative breast cancer patients. All of these patients are enrolled in the Danish Breast Cancer Cooperative Group (DBCG) 77-1a and 82-a protocols for low risk patients, and none of them have received systemic adjuvant therapy. At a median observation time of 50 months and in an evaluation of the total patient population as an entity, ER+ patients had only a marginally significant (P = 0.07) longer RFS than ER- patients while PgR+ patients experienced a significant advantage (P = 0.02). Among patients subgrouped according to menopausal status, both ER and PgR statuses were found to be significant prognostic factors for predicting RFS in the premenopausal women (less than 50 years) but not in peri- or postmenopausal women. Using Cox's multivariate analysis, nuclear pleomorphy was found to be the only significant prognostic variable, while the value of PgR status as a prognostic factor approached significance (P = 0.065). Although knowledge of ER status did not significantly improve distinction between patients with good and poor prognoses in the relatively small subgroup of premenopausal patients (n = 120) when PgR status was known, ER+PgR- patients have a lower risk of recurrence or death than ER-PgR- patients. Using a log-likelihood model, significant and distinct cut-off limits for the definition of receptor positivity were found for premenopausal patients: these were 5 fmol/mg cytosol protein for ER and 10 fmol/mg cytosol protein for PgR. These cut-off levels may reflect the ability of the ligand binding assay method used to discriminate between tissues with and without receptor proteins. Qualitative assessment of receptor status was as valuable as quantitative expression of receptor concentrations in predicting the RFS of the natural course of the disease among node negative premenopausal patients.

Onset of the release of spermatozoa (spermarche) in boys in relation to age, testicular growth, pubic hair, and height.

The onset of production of spermatozoa (spermarche) is the basis for achievement of reproductive capacity in men. We collected 24-h urine samples every 3 months in a 7-yr longitudinal study of 40 normal boys initially aged 8.6-11.7 yr. After centrifugation, the urine was analyzed for the presence of spermatozoa by microscopic examination, and spermarche was estimated on the basis of age at first observed spermaturia. The results were corrected for the intermittent occurrence of spermatozoa in the urine after first observed spermaturia and the fact that the urine samples were collected quarterly. In addition, physical examination, including determination of testicular size by orchidometer measurement, pubic hair distribution (Tanner stage), and height, was carried out every 6 months. Spermarche occurred at a median age of 13.4 yr (range, 11.7-15.3 yr), at a time when testicular size was 4.7-19.6 ml (median, 11.5 ml), and pubic hair distribution was 1-5 (median, 2.5). In most boys, spermarche preceded the age of peak height velocity (median, 13.8 yr; range, 12.2-15.2 yr); at the time of spermarche, median peak height growth velocity was 9.9 cm/yr (range, 7.5-13.4 cm/yr), and median height was 160.4 cm (range, 151.7-175.9 cm). We conclude that spermarche is an early pubertal event and that a wide variation in testicular size and secondary sex characteristics is found at that time. In particular, spermarche may occur when little or no pubic hair has developed, and the testes have grown only slightly.

Epidemiology of myasthenia gravis in Denmark. A longitudinal and comprehensive population survey.

The incidence of myasthenia gravis (MG) was found to be constant in calendar time. The mean annual incidence rate was 4.4 per million population. Age- and sex-specific incidence rates disclosed a bimodal appearance for both sexes, with a peak age at onset located in the early-onset group and another peak for late onset of MG. Early onset of MG appeared 10 years later for male individuals than for female individuals, whereas the peak for late onset of MG was located at the same age for both sexes. It is suggested that the separation between early onset and late onset of MG should be at the age of 50 years for both sexes, rather than at 35 to 40 years as accepted in most studies. The prevalence of MG has increased in time. On January 1, 1988, the point prevalence rate was 77 per million population (female subjects, 96, and male subjects, 57). This reflects an improvement in prognosis despite the fact that life expectancy was found to be significantly lower for MG patients than that of the sex- and age-matched population. The factors causing increased mortality were found to be operative throughout the duration of the disease. Maximum severity of disease was reached within 2 years from onset in 78% of the cases, and more than 50% of all MG-related deaths occurred during the same period. In the course of MG, 70% of all patients experience generalized muscular weakness, and 30% to 40% also suffer from respiratory problems.

Evidence for increasing incidence of abnormalities of the human testis: a review.

Recent reports have suggested that the incidence of genitourinary abnormalities in human males has increased during the past 50 years, including congenital abnormalities such as cryptorchidism and hypospadia, which seem to be occurring more commonly. Also, the incidence of testicular cancer has increased 3- to 4-fold since the 1940s. This increase seems to be worldwide including countries with a very high frequency of testicular neoplasia as well as those in which this cancer is rather uncommon. It has also been postulated that semen quality has been decreasing for the last half century. A recent study showed that the average sperm density has decreased significantly from 113 million/mL in 1940 to 66 million/mL in 1990. The mean seminal volume has also declined, indicating that the decrease in the total sperm count is even more pronounced than the fall in sperm density would indicate. The remarkable increase in frequency of testicular abnormalities over a relatively short period of time may be due to environmental rather than genetic factors. There is an epidemiological link between the occurrence of different testicular abnormalities. Therefore, common prenatally acting etiological factors with adverse effects on the fetal male gonad might be suspected. However, postnatal influences may also have a deleterious effect on male fertility. From the reproductive point of view, an increased impact on the human male gonad is of concern.

Declining semen quality and increasing incidence of testicular cancer: is there a common cause?

Male reproduction has been given little attention in science and in medical practice. However, a recent metaanalysis on semen quality, which clearly pointed to a decrease over the past 50 years, has been repeatedly quoted. Three recent reports have found that semen quality has declined among candidate semen donors during the past 20 years. The evidence of decline in the quality of semen is not the only indicator that the human testis is at risk. During the past 50 years, cancer of the testis has also become more common. This is a disorder of young men, and it is associated with a high rate of other abnormalities of the testis including undescended testis and poor semen quality. Furthermore, the incidence of both hypospadias and undescended testis has been reported to be rising in the general population. We believe that the evidence of declining semen quality should be seen in the light of these trends in other reproductive disorders of men. However, the etiology is unknown. A recent hypothesis that links the trends in the health of the male reproductive system to xenoestrogens in the environment is discussed.

Male reproductive health and environmental xenoestrogens.

Male reproductive health has deteriorated in many countries during the last few decades. In the 1990s, declining semen quality has been reported from Belgium, Denmark, France, and Great Britain. The incidence of testicular cancer has increased during the same time incidences of hypospadias and cryptorchidism also appear to be increasing. Similar reproductive problems occur in many wildlife species. There are marked geographic differences in the prevalence of male reproductive disorders. While the reasons for these differences are currently unknown, both clinical and laboratory research suggest that the adverse changes may be inter-related and have a common origin in fetal life or childhood. Exposure of the male fetus to supranormal levels of estrogens, such as diethlylstilbestrol, can result in the above-mentioned reproductive defects. The growing number of reports demonstrating that common environmental contaminants and natural factors possess estrogenic activity presents the working hypothesis that the adverse trends in male reproductive health may be, at least in part, associated with exposure to estrogenic or other hormonally active (e.g., antiandrogenic) environmental chemicals during fetal and childhood development. An extensive research program is needed to understand the extent of the problem, its underlying etiology, and the development of a strategy for prevention and intervention.

Statistical methods for the analysis and presentation of the results of bone marrow transplants. Part I: unadjusted analysis.

In this paper, we describe modern statistical methods for presentation of the results of studies of bone marrow transplantation. We focus here on 'univariate' or unadjusted techniques to describe the outcomes of such studies. In another paper we will discuss multivariate methods. We discuss the type of data one may have available to make inference about outcomes. We explain the differences between the Kaplan-Meier estimator of the survival function and the cumulative incidence curve, how these curves should be interpreted and when each is the appropriate summary statistic. We discuss the weighted log rank statistic and show how different weights can be used to put emphasis on detecting differences between groups in different time periods. We also present a simple estimate of current leukemia-free survival which is useful in summarizing post-transplant events.

Statistical methods for the analysis and presentation of the results of bone marrow transplants. Part 2: Regression modeling.

In this paper, we address methods of multivariate regression. We discuss the value of regression compared to matched pairs analysis, methods of coding variables, basic concepts of the Cox model and interpretation of results of the Cox model. We present methods of handling variables whose effect changes with time. We present methods to check the assumptions of the Cox regression. Finally, and perhaps most importantly, we provide suggestions for presenting the results in clear and thorough tables and graphs.

Graft-versus-leukaemia activity associated with CMV-seropositive donor, post-transplant CMV infection, young donor age and chronic graft-versus-host disease in bone marrow allograft recipients. The Nordic Bone Marrow Transplantation Group.

Risk factors for post-transplant relapse were analysed retrospectively in 163 patients treated with allogenic bone marrow transplantation for acute myeloid leukaemia (AML), acute lymphoblastic leukaemia (ALL) or lymphoblastic lymphoma in first to fourth remission or during relapse. Multifactorial analysis was performed according to Cox with fixed pretransplant covariates and post-transplant cytomegalovirus (CMV) infection and graft-versus-host (GVHD) as time-dependent covariates. Advanced stage of leukemia at the time of transplantation was an important risk factor for subsequent relapse. Furthermore, the study confirmed a graft-versus-leukaemia (GVL) activity associated with chronic GVHD, including de novo chronic GVHD (intensity factor 0.08, p = 0.004). In a model excluding chronic GVHD, female donor-to-male recipient (a risk factor for GVHD), was associated with decreased post-transplant relapse risk (intensity factor 0.3, p = 0.008), suggesting that an allo-reaction against a minor transplantation antigen (Hy) may mediate antileukaemic activity. A decrease of the relapse risk by a factor 0.18 was observed in recipients with AML as well as ALL when the donor was CMV seropositive (p = 0.0002). This effect was restricted to patients who had laboratory evidence of post-transplant CMV infection. When CMV infection occurred and donor was seropositive the relapse risk was reduced by a factor 0.035. The effect was not mediated through an increased occurrence of grade 2-4 acute or chronic GVHD and could not be explained by a statistical bias due to censoring of patients who died in remission. Rather, donor CMV immunity was associated with GVHD independent GVL activity.(ABSTRACT TRUNCATED AT 250 WORDS)