Transjugular intrahepatic portosystemic shunt: midterm clinical and angiographic follow-up.

PURPOSE: To determine the impact of Child class and indication for transjugular intrahepatic portosystemic shunt (TIPS) placement on patient survival and reintervention rate. MATERIALS AND METHODS: Metal stents were used to successfully create single shunts in 63 patients during a 3-year period. Angioplasty and repeated stent placement were used to maintain shunt patency, and patients were followed up clinically and angiographically. Statistical analysis of survival and patency was performed with Kaplan-Meier product-limit survival functions. A Karnofsky performance status score was derived for each follow-up encounter. RESULTS: Early and late mortalities varied with Child class and procedure indications. Thirty-day mortality was 19% overall (12 of 63 patients) and was 33% for Child class C patients (10 of 30 patients). Thirty-day mortality was 31% (four of 13 patients) for patients with ascites and 16% (eight of 50 patients) for those with bleeding. Reintervention was required in 20 of 33 patients and was not predictable on the basis of Child class. CONCLUSION: Child class and indication for procedures are significant predictors of survival but not of the need for reintervention. Ninety-day survivors had uniformly good performance status.

Endoscopic balloon dilation for ulcer-induced gastric outlet obstruction.

OBJECTIVE: To determine the safety and efficacy of endoscopic balloon dilation for ulcer-induced gastric outlet strictures. METHODS: Review of the cumulative experience from two institutions involving 30 patients who underwent endoscopic balloon dilation for peptic ulcer-induced gastric outlet obstruction. Follow-up was by standardized interview at a mean of 15 months (range 4-28 months). RESULTS: Symptoms occurred for an average of 6 months before dilation and included weight loss, epigastric pain, nausea, vomiting, early satiety, bloating, and anorexia. Gastric outlet strictures had a median diameter of 6 mm (range, 0-10 mm). Ten (33%) patients had active ulcers. Six-millimeter to 18-mm (median 15-mm) balloons were inflated a median of 2 times (range 1-4 times) for a median of 60 s (range 30-180 s). Fifty-one procedures (1.7/patient) were performed; 20 (67%) patients had one treatment and 10 (33%) had multiple treatments. Twenty-four (80%) patients achieved sustained symptom relief, 17 of 20 having a single procedure and 7 of 10 required multiple sessions. Dilation failed in 4 (13%) patients with long duodenal strictures. Two (6.7%) patients dilated to 18 mm suffered perforation. Both recovered uneventfully after surgery. CONCLUSIONS: Endoscopic balloon dilation is safe and effective for most patients with ulcer-induced gastric outlet obstruction.

Effects of 1,25-dihydroxyvitamin D3 on proliferation and differentiation of Caco-2 cells.

The effect of 1,25-dihydroxyvitamin D3 [1,25-(OH)2D3], or calcitriol, on the proliferation and differentiation of Caco-2 cells was studied. Vitamin D receptor mRNA was detected in both pre- and postconfluent cells, and its abundance was unchanged with time and in response to calcitriol. 1,25-(OH)2D3-binding activity increased during differentiation, but there was no difference in binding between 1,25-(OH)2D3-treated and control cells. 1,25-(OH)2D3 caused a dose-dependent reduction in proliferation, as assessed by [3H]thymidine incorporation and DNA content. 1,25-(OH)2D3 significantly enhanced the normal rise in alkaline phosphatase activity during differentiation and increased alkaline phosphatase mRNA abundance. In contrast, 1,25-(OH)2D3 inhibited the normal rise in sucrase-isomaltase activity and the corresponding mRNA level, although the inhibition occurred after the initial period of cell differentiation (> 10 days postplating). Morphological analysis demonstrated that by day 12 postplating, 1,25-(OH)2D3 increased the mean dome diameter and microvillus length and density. Although 1,25-(OH)2D3 decreases the proliferation of Caco-2 cells and enhances certain parameters of differentiation, not all brush-border hydrolases respond in a similar fashion, making it necessary to interpret with caution their individual use as markers of differentiation.

Caco-2 cell transfection by rat intestinal alkaline phosphatase cDNA increases surfactant-like particles.

The rat enterocyte produces a particle with surfactant-like properties (including a whorled appearance, enrichment for dipalmitoyl phosphatidylcholine, and ability to lower surface tension) that also is enriched for intestinal alkaline phosphatase. Human Caco-2 cells grown on polycarbonate filters were utilized to study the secretion of these particles and exhibited whorls and strands of unilamellar membranes, particularly concentrated at the apical pole or near junctional complexes. Concentrated culture medium from these cells separated on continuous NaBr gradients revealed a fraction at density = 1.07 g/l enriched for phosphatidylcholine and intestinal alkaline phosphatase. This fraction contained membranous sheets containing alkaline phosphatase, detected by immunolocalization. Phosphatidylcholine comprised 54% of phospholipid in this fraction, compared with 20% in brush borders. When Caco-2 cells were transfected with cDNA encoding rat intestinal alkaline phosphatase, cellular phosphatase activity increased twofold, but activity in the medium increased 14-fold to > 200 (average 32)-fold. Ultrastructurally, compared with mock-transfected cells or cells transfected with human placental alkaline phosphatase, transfection with rat intestinal alkaline phosphatase cDNA led to intracellular and extracellular accumulation of surfactant-like particles. We conclude that surfactant-like particles are produced by Caco-2 cells, and their production can be enhanced by transfection with a cDNA encoding a protein known to be associated with such particles.

Differential regulation of mRNAs encoding for rat intestinal alkaline phosphatase.

A cDNA probe encoding the entire structural region of the 62-kDa rat intestinal alkaline phosphatase from amino acid residues 1 to 531 detected multiple mRNA species (3.0, 2.7, and 2.2 kb) in rat intestinal RNA. The 3.0-kb species was most evident in duodenum but could be easily detected in jejunum using a 48-mer oligonucleotide encoding amino acid residues 492-508. This 48-mer oligonucleotide bound preferentially to the 3.0-kb mRNA, suggesting that the 2.7-kb mRNA differed in this region. To determine whether each of the mRNAs encoding rat intestinal alkaline phosphatase responded coordinately to physiological stimuli, the full-length cDNA and the 48-mer oligonucleotide were used as probes for the 2.7- and 2.2-kb and the 3.0-kb mRNAs, respectively. Intestinal mRNA concentration was measured by Northern blot analysis in acute (single feed, 17 kcal) and chronic (3 wk, 30% fat diet) fat feeding and in rachitic rats after 1,25-dihydroxyvitamin D3 therapy. There was a large increase (8- to 25-fold) in the 3.0-kb mRNA 7 h after acute fat feeding, with a much smaller increase (1.4- to 5.0-fold) in the 2.7- and 2.2-kb species. The peak in 3.0-kb mRNA accumulation correlated in time with the maximal activity of serum phosphatase activity after acute fat feeding (4- to 5-fold increase). In contrast, there was a much smaller increase in all mRNAs and in tissue and serum enzyme activity after chronic fat feeding.(ABSTRACT TRUNCATED AT 250 WORDS)

Properties of rat and mouse beta-glucuronidase mRNA and cDNA, including evidence for sequence polymorphism and genetic regulation of mRNA levels.

cDNA clones containing partial sequences for beta-glucuronidase (beta G) were constructed from rat preputial gland RNA and identified by their ability to selectively hybridize beta G mRNA. One such rat clone was used to isolate several cross-hybridizing clones from a mouse-cDNA library prepared from kidney RNA from androgen-treated animals. Together, the set of mouse clones spans about 2.0 kb of the 2.6-kb beta G mRNA. Using these cDNA clones as probes, a genomic polymorphism for DNA restriction fragment size was found that proved to be genetically linked to the beta G gene complex. A fragment of beta G cDNA was subcloned into a vector carrying an SP6 polymerase promoter to provide a template for the in vitro synthesis of single-stranded RNA complementary to beta G mRNA. This provided an extremely sensitive probe for the assay of beta G mRNA sequences. Using either nick-translated cDNA or transcribed RNA as a hybridization probe, we found that mouse beta G RNA levels are strongly induced by testosterone, and that induction by testosterone is pituitary-dependent. During the lag period preceding induction, during the induction period itself, and during deinduction following removal of testosterone, beta G mRNA levels paralleled rates of beta G synthesis previously measured by in vivo pulse-labelling experiments. Genetic variation in the extent of induction affected either the level of beta G mRNA or its efficiency of translation depending on the strain of mice tested.

The public health effects of legal abortion in the United States.

PIP: Beneficial effects of the progressive liberalization of abortion laws in the US since 1967 are presented. The total number of pregnancy-related deaths averted by legal abortions appears to be on the order of 1500. The number of life-threatening, but not fatal, complications averted probably reached several 10s of thousands. Another beneficial effect of the liberalization of abortion laws has been that women with medical contraindications to continued pregnancy, especially poor and minority women, now can receive therapeutic abortions. A 3rd positive effect of abortion legalization has been the possibility of preventing the birth of infants with major physical or mental defects. In many cases fetal defects can now be detected with certainity or near-certainty by a variety of methods. At present, the number of abortions performed on the basis of prenatal diagnosis is quite small--in 1982 it may have been on the order of 1500 or 1/10 of 1% of all US legal abortions that year.^ieng

Legal abortion and public health.

PIP: Over 15 million abortions have been performed in the US since the process of abortion legalization began in 1967. Consequences of legalization have included a marked reduction of pregnancy-related mortality and the prevention in many cases of the birth of infants with major physical or mental defects. Prenatal diagnosis, backed up by selective abortion, has made procreation a possibility for many couples who might otherwise avoid childbearing. However, the number of abortions performed on the basis of prenatal diagnosis remains small, comprising only about .01% of all legal abortions. In recent months, the pro-choice movement in the US has been handed 2 important victories: the US Supreme Court reaffirmed the 1973 decision legalizing abortion and the US Senate defeated a constitutional amendment intended to reverse this decision. As a result of these victories, contributions to pro-choice groups have declined. Continued vigilance is needed to protect these victories.^ieng

Impact of vacuum aspiration abortion on future childbearing: a review.

Ever since induced abortion was legalized in the United States, there has been a running controversy over whether induced abortion affects subsequent childbearing; for example, it has been claimed that women who terminate a pregnancy are at a greater risk of miscarrying a subsequent pregnancy or of having a low-birth-weight baby. Ten studies of the later impact of first-trimester induced abortion by vacuum aspiration (the dominant method in the United States) are examined here; they find that compared with women who carry their first pregnancy to term, women whose first pregnancy ends in induced abortion have no greater risk of bearing low-birth-weight babies, delivering prematurely or suffering spontaneous abortions in subsequent pregnancies. However, these studies also show that induced abortion of a woman's first pregnancy does not have the protective effect on her first live birth that carrying a first birth to term has on later deliveries. In addition, some evidence from other studies links dilatation and curettage (D&C) procedures with later infertility, but most studies have found no such association. No definite conclusions can be reached about the impact of multiple induced abortions, since the results of 13 different epidemiologic studies are almost evenly divided between those that show no effect and those reporting related reproductive problems.

PIP: Ever since induced abortion was legalized in the U.S. there has been a running controversy over whether induced abortion affects subsequent childbearing; for example, it has been claimed that women who terminate a pregnancy are at a greater risk of miscarrying a subsequent pregnancy or of having a low birthweight baby. 10 studies of the later impact of 1st trimester induced abortion by vacuum aspiration (the dominant method in the U.S.) are examined; compared with women who carry their 1st pregnancy to term, women whose 1st pregnancy ends in induced abortion have no greater risk of bearing low birthweight babies, delivering prematurely or suffering spontaneous abortions in subsequent pregnancies. However, these studies also show that induced abortion of a woman's 1st pregnancy is without the protective effect on her 1st live birth that carrying a 1st birth to term has on later deliveries. In addition, some evidence from other studies links dilatation and curettage procedures with later infertility, but most studies have found no such association. No definite conclusions can be reached about the impact of multiple induced abortions, since the results of 13 different epidemiologic studies are almost evenly divided between those that show no effect and those reporting related reproductive problems.

Repeat abortion in the United States: new insights.

The increasing numbers and proportions of repeat abortions in the United States, including higher order repeat abortions, can be adequately explained by the increasing numbers of women at risk and the heterogeneity of the population in regard to the need for and utilization of abortion services. The same factors, rather than a deterioration of contraceptive practice following abortion, explain the higher levels of repeat abortion rates compared with first abortion rates.

Race-specific patterns of abortion use by American teenagers.

Between 1972 and 1978, as legal abortion became more widely available nationally, abortion rates (abortions per 1,000 women) and ratios (abortions per 1,000 live births) increased for all American teenagers; from 1972 to 1975, the rates and ratios for teenagers for Black and other races increased faster than those for White teenagers. For all seven years, abortion rates were higher for teenagers of Black and other races than for white teenagers. This reflected both higher proportions of sexually active teenagers of Black and other races and a greater risk of pregnancy in these teenagers compared with White teenagers. Race-specific differences in legal abortion ratios narrowed during the seven-year interval, as did differences in alternative outcomes of teenage premarital pregnancies (term births, illegal abortions).

Mannose-specific endocytosis receptor of alveolar macrophages: demonstration of two functionally distinct intracellular pools of receptor and their roles in receptor recycling.

Receptor-mediated endocytosis of rat preputial beta-glucuronidase and the glycoconjugate mannose-BSA by rat alveolar macrophages is inhibited by chloroquine and ammonium chloride. We have previously reported that these drugs cause a loss of cell surface binding activity and that they do not inhibit internalization of receptor ligand complexes when incubated with cells at 37 degrees C. In this report we more clearly delineate the intracellular site of weak base inhibition of receptor recycling and the mechanism of that inhibition. From our analysis of the kinetics of ligand transport we conclude that there are two functionally distinct intracellular pools of receptor. One of these, the cycling pool, is not sensitive to the presence of weak bases, and receptor-ligand complexes return from this pool to the cell surface intact. The second pool is responsible for the time-dependent intracellular delivery of ligand to acid vesicles, which is inhibited by weak bases. Chloroquine and ammonium chloride appear to inhibit the dissociation of receptor-ligand complexed in this second pool and thereby the production of free receptors for the continuation of receptor-mediated endocytosis. We examine the internalization and binding of ligand in normal and paraformaldehyde-treated cells and find that these are strongly affected by pH. In particular, the dissociation rate of receptor ligand complexes is enhanced greater than 7.5 fold by lowering the medium pH from 7 to 6. From these results we propose that weak bases raise the pH of acid intracellular compartments, slowing the rate of receptor-ligand dissociation and thereby reducing the cellular pool of free receptors available for further uptake of ligand. In addition, we demonstrate that receptor-ligand complexes cannot return to the cell surface from the amine-sensitive (acid) intracellular pool that led us to call this the nonreleasable pool. This final observation indicates that receptor movements through these two pools are functionally distinct processes.

Abortion services in the United States, 1979 and 1980.

PIP: Reports results of the 7th national survey by the Alan Guttmacher Institute of all known providers of abortion in the US. There were 1.55 million abortions in 1980, about 1/4 of all pregnancies and 1/2 of unintended ones. About 3% of women of childbearing age obtained an abortion. The rate increased by 2% from 1979 to 1980, compared to an increase of 4% over the previous 2 years. As the increases have been smaller each year, the rate may stabilize in 1982 at 1980 levels. Despite the continuing controversy, geographic availability improved somewhat, although 78% of all counties had no provider and 59 of the 305 metropolitan areas had no facilities reporting abortions. Almost 70% of nonmetropolitan women live in counties with no facilities reporting abortions--only 4.5% of abortions take place in nonmetropolitan areas. Other factors restricting availability of abortion include lack of Medicaid funding in most states, parental notification and consent requirements (by law in 7 states), period of gestation at which abortions may be performed (30% of providers will only perform abortions up to 10 weeks' gestation; only 21% will perform them after 14 weeks), and the way in which hospitals are organized to provide services. 3/4 of all abortions were performed in 459 clinics and hospitals which reported 1000 or more abortions each. The proportion performed in nonhospital facilities has risen steadily since legalization: nearly 80% were performed in these facilities in 1980 compared with less than 1/2 in 1973. Improvements and simplification in abortion techniques have contributed to this trend. Only 55% of hospital abortion providers perform any outpatient abortions, and in most cases women are unable to arrange for a hospital abortion without a referral, further restricting access to abortion services in many cases.^ieng

The effects of induced abortion on subsequent reproduction.

PIP: This review of the longterm impact of induced abortion on reproductive function includes a comprehensive survey of the published and the preliminary literature. Over 200 anecdotal, descriptive, and observational epidemiologic reports were reviewed, and more than 150 studies in 11 languages from 21 countries that employed 1 of 4 standard research designs were identified. With these investigations as the foundation, the influence of induced abortion on secondary infertility, ectopic pregnancy, spontaneous abortion, shortened gestation, low birth weight, complications of pregnancy or delivery, and infant outcome were evaluated. The data from each of these investigations were systematically reexamined. In cases where the data were incomplete, additional information was requested from the authors. These studies were categorized by such factors as their research design, the type of abortion procedure used, and whether potential confounding variables were considered. To compare the relative power of studies, the 95% confidence interval around the point estimate of relative risk was calculated where possible. Studies that failed to consider the many confounding factors associated with both induced abortion and adverse pregnancy outcome were of limited value because women who obtain induced abortions differ substantially from those who do not. Results from these studies usually indicated a higher risk estimate than results from better designed, controlled investigations. Relying more heavily on the latter, the following were among the findings: 1) secondary infertility, which may be a rare late effect of a complicated abortion, occurs so infrequently that its risk is not significantly elevated, even in studies capable of detecting a 2-fold rise in relative risk; 2) ectopic pregnancy, which also may be a rare late effect of an abortion complicated by infection or trauma, occurs so infrequently that its risk is not significantly elevated in studies capable of detecting a 2-fold rise in relative risk; 3) midtrimester spontaneous abortion is not more frequent for women who have experienced 1 previous abortion than for women who are in their 1st pregnancy; 4) shortened gestation in pregnancies after 1 induced abortion has an incidence similar to that in 1st pregnancies; 5) low birth weight is more frequent in 1st births after abortion performed by dilatation and curettage under general anesthesia, but it is not more frequent in 1st births after abortions performed by other methods, when compared with 1st pregnancy births; 6) 1 induced abortion does not protect a woman from the higher risks of low birth weight and pregnancy complications associated with a 1st birth; 7) repeat induced abortion has not been studied sufficiently to clarify whether the increased risks associated with it in some studies are caused by confounding factors or the procedures performed; and 8) 1st born infants whose mothers had 1 induced abortion are at similar risk of morbidity and infant mortality as other 1st born children.^ieng

Abortion in the United States, 1978-1979.

PIP: The total number of legal abortions in the US continues to rise; it was 1.5 million in 1979. Since 1967 when the 1st abortion reform laws were passed, some 7.4 million women have obtained legal abortions -- about 15% of all US women of reproductive age. Nevertheless, some 641 thousand women in need of abortion services in 1979 were unable to obtain them and went on to have unintended births. Large proportions of these women were poor, very young, rural and members of minority groups. 1 of the major reasons for the large unmet need for abortion services is the continuing inaccessibility of abortion services: 8 out of 10 counties have no facility in which abortions are performed. The scarcity is particularly notable in rural counties: only 5% of abortions are performed outside of metropolitan areas. Although only 8% of women who obtained abortions in 1978 had to cross state lines to do so, a much larger proportion had to travel, often considerable distances, outside their home communities. In addition to geographic inaccessibility, the withdrawal of public funds for abortion services as a result of the Hyde Amendment and its state counterparts has placed new financial obstacles in the way of poor women who want to obtain abortions. Only 1/4 of all abortions are performed in hospitals, where most other surgery takes place. Each year, a smaller proportion of abortions are performed in hospitals and a larger proportion performed in free standing clinics. This trend is largely attributable to the much higher costs of hospital procedures, as well as to the unwillingness of many hospitals to provide abortions: only 18% of public hospitals and 36% of private and voluntary hospitals reported any abortions in 1978. The failure of hospitals -- especially public hospitals -- to offer abortion services has contributed to the inaccessibility of abortion services.^ieng

[The mathematics of repeat abortion: explaining the increase (author's transl)]. / Les regles mathematiques de l'I.V.G. repetee: analyse de sa progression.

PIP: The increase in the proportion of repeated abortion after legalization of induced abortion has been attributed either to a decrease in the practice of contraception, or to the negative influence on morality of a permissive society. These suppositions are not supported by data. The authors present a simple mathematical model to quantify the percentage of repeated abortion for a population of women aged 15-44, on the assumption that the risk of abortion is constant over time, and not dependent on prior abortion experience. Results of the investigation show that: 1) the proportion of repeated abortions increases with time; 2) the final percentage, 30 years after legalization of abortion, increases with the abortion rate; 3) the concentration of abortions within a determined age group does not influence the final percentage; and 4) that the heterogeneity of the population causes a great difference in the rate of first abortions and in that of repeated abortions, and that the first is higher than the second.^ieng

Abortion in the United States, 1977-1978.

There were 1.32 million legal abortions in the United States in 1977 and a projected 1.37 million in 1978, an increase of four percent between 1977 and 1978 compared with one of 12 percent between 1976 and 1977. In 1978, 29 percent of pregnant women chose to terminate their pregnancies by abortion. Almost three percent of U.S. women of reproductive age obtained an abortion in 1978. From 1967 through 1978, approximately six million women obtained almost eight million legal abortions; about one in eight U.S. women of reproductive age has had a legal abortion. The number of hospitals reporting that they provided abortion services dropped slightly from 1,695 in 1976 to 1,661 in 1977, but the number of nonhospital abortion clinics increased from 448 to 522, and the number of physicians who reported performing abortions in their offices grew from 424 to 533. Between 1976 and 1977, the average number of abortions per hospital facility decreased from 246 to 237, while the average number per nonhospital provider increased from 875 to 879. The percentage of abortions performed in hospitals declined from 35 in 1976 to 30 in 1977, while the percentage reported by free-standing clinics increased from 61 to 66; the percentage performed in physicians' offices remained at four. Ninety-five percent of abortions in 1977 occurred in metropolitan areas, where 75 percent of the women in need of abortion services live. In 1977, there were identified abortion providers in only 23 percent of U.S. counties. Nine percent (more than 118,000) of the women who obtained abortions in 1977 had to travel to another state for services, and many traveled to other, often distant, counties in their home states. One in three abortions in 1977 were obtained by teenagers, and three in four were obtained by unmarried women. Twenty-eight percent of the women estimated to be in need of abortion services in 1977, and 26 percent in 1978, were unable to obtain them. In FY 1977, before Hyde amendment restrictions on government financing of abortions for poor women, 133,000 of the estimated 427,000 Medicaid-eligible women in need of publicly funded abortion services were unable to obtain them.(ABSTRACT TRUNCATED AT 400 WORDS)

PIP: Results of the Alan Guttmacher Institute's 5th annual survey of abortion services in the U.S. are presented. Data were collected on the number of abortions performed by each provider in 1977 and the first quarter of 1978. In 1977, 1,320,000 abortions were reported, an increase of 12% over 1976. The legal abortion rate rose between 1973 and 1978 from 16.6 to 27.5 per 1000 women of reproductive age. In 1978 there were 406 abortions for every 1000 live births. 1 in every 8 women of reproductive age in the U.S. has had at least 1 legal abortion since 1967. Most abortion services are concentrated in metropolitan areas. Abortion occurrence rates in 1977 ranged from a low of 5.2 per 1000 women in Mississippi to a high of 46.0 in New York. Nationally, 72% of women needing abortions in 1977 and 74% needing them in 1978 were able to obtain them, but 506,000 women in 1977 and 479,000 women in 1978 were unable to obtain services. Women who obtained abortions in 1977 were mainly young, white, unmarried, childless, and undergoing their first abortion. More than 90% of abortions were performed in the first trimester by suction or sharp curettage. Adolescents obtained nearly 1/3 of all abortions, and women 20-24 another 1/3. Abortion rates in 1977 were highest for women 18-19 years old, for nonwhite and unmarried women, and for poor women eligible for Medicaid. In 1978, the number of federally funded Medicaid abortions was about 1% of the number paid for before the Hyde amendment took effect.

[Statistical analysis of abortion data]. / Analysis estadistico de los datos sobre aborto.

PIP: The psychosocial antecedents of induced abortion must be clearly understood before rational discussion of appropriate statistical methodology for analysis of its incidence can be undertaken. The "desire for a child" is a specific concept at the moment when the decision is made to seek an abortion or carry the pregnancy to term. The 2 most important analytical instruments for assessing the incidence of abortion are the abortion rate and the proportion of abortions. Abortion rates are generally calculated as annual rates per 1000 women of reproductive age. Specific rates for age, civil status or other characteristics may also be calculated. The proportion of abortions is generally calculated as the number of abortions per 1000 live births or per 1000 pregnancies. Period rates of abortion are often impossible to obtain from survey data because the samples are too small, leading many investigators to publish their findings in terms of cumulative rates and proportions for groups of women of successive ages. Period rates are necessary to assess the extent of abortion in the present and immediate past, and adequate samples are essential.^ieng