Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
10 Causality and Opportunities for Intervention One of the major issues under consideration by this subcommittee is the relationship between gestational weight gain and a variety of maternal and child health outcomes. In particular, its interest focuses on gestational weight gain as an etiologic determinant, i.e., a cause, of these maternal and child outcomes. An understanding of cause is often a prerequisite for effecting change (improving maternal and child health). As mentioned in Chapters 2 and 4, however, low gestational weight gain might also be of potential value as a noncausal marker of risk for adverse pregnancy outcome. A change in pregnancy outcome therefore depends on the ability to alter causal determinants. Since clinicians and public health practitioners cannot directly affect gestational weight gain (without surgically removing or adding tissue to the pregnant woman), factors that can be modified must be considered. The two major modifiable factors are energy intake and expenditure (as they have an impact on energy balance). But these can be modified only indirectly, such as by offering nutritional advice or supplementation, either in the context of regular prenatal care or as part of a special program. It may also be possible to change maternal attitudes that have an impact on energy intake, either by apprising the community of recent research findings and expert opinion or through more formal avenues of health education. In identifying modifiable factors and the likely impact of such modi- fications on maternal and child health, the entire causal pathway depicted in Figure 2-3 must be considered, beginning with attitudes, counseling, and 222
CAUSALITY AND OPPORI~UNIlIES FOR INTERVENTION 223 supplementation and working through actual increases or decreases in en- ergy intake; subsequent changes in gestational weight gain; the short-term maternal and fetal/child outcomes of pregnancy; and finally, longer-term maternal and child health. In the following section, these causal links are discussed in greater detail. THE CAUSAL PATHS Most of the evidence examined by the subcommittee concerns the possible causal relationship between gestational weight gain and a variety of short-term health outcomes for the mother (mortality, complications of pregnancy, lactation performance, and postpartum obesity) and the fetus and infant (mortality, fetal growth, gestational duration, spontaneous abor- tion, and congenital anomalies). As discussed in Chapter 2 and as shown in Figure 2-3, however, these limited causal paths require both proximal and distal extensions. They need to be extended proximally because neither clinical and public health interventions nor changes in maternal attitudes have a direct impact on gestational weight gain. Consideration must be given, at least briefly, to causal paths from maternal attitudes, nutritional counseling, and energy supplementation to energy intake, and from energy intake to gestational weight gain. The paths must be extended distally to examine the relationship between the short-term outcomes enumerated above and the longer-term maternal and child health outcomes that may be of greater importance. For the mother, the main long-term outcome of interest is obesity. For the child, such outcomes include all aspects of survival, morbidity, growth, and performance that could be affected by changes in fetal growth or gestational duration. Very few reports directly link the proximal factors to the important long-term health outcomes at the end of the causal pathway. Exceptions include publications by Naeye and Chez (1981), Singer et al. (1968), and l~vris and Read (1982), which link childhood cognitive development to maternal weight gain. Moreover, there is no strong evidence that long- term outcomes are affected directly, i.e., independently of shorter-term outcomes. In the remainder of this chapter, therefore, discussion is limited to examination of the evidence involving each of the individual causal links. Although most of the evidence and most of the deliberations of the sub- committee focus on links between gestational weight gain and maternal and child outcomes, the entire causal pathway summarized in Figure 2-3 needs to be kept in mind when formulating recommendations and contemplating possible clinical or public health interventions. In examining the evidence for causality in these individual links, the subcommittee based its inferences on standard epidemiologic crite- ria. These criteria include the strength, biologic gradient (dose-response
224 NUTRITIONAL STATUS AND WEIGHT GAIN effect), lack of bias, statistical significance, specificity, consistency, and bi- ologic plausibility and coherence of the association between exposure (the putative cause) and outcome (DHEW, 1964; Hill 1965; Susser 1973, 1988~. THE EVIDENCE Do Nutritional Counseling, Energy Supplementation, and Maternal Attitudes Affect Energy Intake? Remarkably few studies have been conducted to assess the efficacy of nutritional counseling (without supplementation) on energy intake by pregnant women either within or outside the context of regular prenatal care. In early studies carried out at the Montreal Diet Dispensary, for example, counseling was combined with some degree of energy supplemen- tation, and inappropriate comparisons were made between women who either were referred for or requested those nutrition services and those who did not (Higgins, 1976; Primrose and Higgins, 1971~. Thus, the rather large increases in net energy intake (and birth weight) attributed to use of the nutrition services are likely to be overestimates. Rush (1981) attempted to pair-match Montreal Diet Dispensary participants and nonparticipants according to potential confounding factors. The increases in energy intake and birth weight were more modest than those reported by Higgins (1976) and Primrose and Higgins (1971), but they were still in the expected direc- tion. Higgins et al. (1989) found an increase in infant birth weight among mothers who participated in the program in the second but not the first of two pregnancies, but they provided no data on differences in energy intake. The Special Supplemental Food Program for Women, Infants, and Children (WIC) in the United States also includes a mixture of nutritional counseling and supplementation. Pregnant women are given vouchers that can be used to obtain highly nutritious foods; however, there is no guarantee that the women will consume these foods and not share them with others. Several evaluations of WIC, including the largest and most recent one by Rush et al. (1988), indicate that WIC participants increase their energy intake and have higher gestational weight gains as a result of participation in the program (Edozien et al., 1979; Endres et al., 1981; Metcoff et al., 1985; Rush et al., 1988~. The net increase in energy intake is modest, however, probably on the order of 100 to 150 kcaVday. Supplementation trials indicate that women who are given energy supplements during pregnancy increase their total energy intakes (Lechtig et al., 1975; Mora et al., 1979; Prentice et al., 1983; Rush et al., 1980), but in most studies, the magnitude of the increased intake ranged from 100 to 250 kcal/day-less than the energy content of the supplement provided. Larger
CAUSALITY AND OPPORTUNITIES FOR INTERVENTION 225 increments have been achieved in certain undernourished populations in developing countries, e.g., in The Gambia (Prentice et al., 1983~. The subcommittee found no formal research to determine the effect of maternal attitudes on energy intake. These attitudes change slowly in response to many factors, including the reporting (and mass media dis- semination) of new research findings, recommendations by national and international bodies or individual experts, health education (including that provided at schools and public prenatal classes), practices of health care providers, and word of mouth. Such factors are difficult to measure ob- jectively. Moreover, concurrent controls are often impossible to obtain, since entire countries tend to be exposed to the messages simultaneously. Before-and-after comparisons may therefore be the only feasible way to assess the influence of these factors on energy intake. Unfortunately, small changes in energy intake are also difficult to measure (see Chapter 7~. Thus, any attempt to assess changes in energy intake as a consequence of (i.e., caused by) changes in maternal attitudes would be difficult, to say the least. Nonetheless, as discussed in Chapter 3, authorities have increased their recommended energy intakes and target weight gains for pregnant women over the past 30 to 40 years, and these changes have been accompanied by corresponding increases in gestational weight gain. It is difficult to ascribe recent weight gain changes to anything other than increased energy-intake during pregnancy (see discussion below). Although increased energy intake may be attributable, in part, to individual nutritional counseling by obste- tricians, nurse-midwives, dietitians, and other health care professionals, it seems reasonable to infer that gradual changes in public awareness and maternal attitudes have also played a role. Does Energy Intake Affect Gestational Weight Gain? Everyday experience and carefully controlled experimental studies (Sims et al., 1968) have demonstrated that people who consume excess energy gain weight. There is no reason to believe that pregnant women are an exception to this general rule. Many of the clinical and epidemiologic studies linking energy intake to gestational weight gain have found rather modest, and often nonsignificant, correlations between the two, in part because most women with high energy intakes also have high energy ex- penditures and, in part, because of the difficulty of accurately measuring the change in energy intake. Problems in measurement of energy intake have probably led to underestimates of the true correlation between maternal energy intake and gestational weight gain. The clearest evidence favoring the link between energy intake and gestational weight gain comes from studies of human famine and from
226 NUTRITIONAL STATUS AND WEIGHT GAIN supplementation trials. It is clear from the Dutch famine study that reduced energy intake leads to reduced gestational weight gain (Stein et al., 1975; Susser and Stein, 1982~. It is equally clear from supplementation trials in both developed (Rush et al., 1980) and developing (More et al., 1979) countries that increased energy intake leads to larger maternal weight gains. Does Gestational Weight Gain Affect Short-Term MaternaVChild Health? Virtually all epidemiologic studies with adequate sample sizes have demonstrated an association between gestational weight gain and fetal growth. Many supplementation trials and observational studies have been based on rigorous epidemiologic methods to minimize most sources of bias; however, reverse causality (temporal precedence) has not received adequate attention. For example, the amount of weight that a woman gains during a pregnancy is influenced not only by the deposition of fat, protein, and glycogen stores potentially available for provision of nutrients to the fetus but also by large increases in body water. It is well known that expanded plasma volume and modest degrees of dependent edema are associated with favorable pregnancy outcomes (Campbell and MacGillivray, 1975; Duffus et al., 1971; Hytten and Thomson, 1976; Naeye, 1981a,b; Thomson et al., 1967~; but in the absence of data establishing that increases in body water precede increases in fetal growth rate, they may just as likely be an effect, rather than a cause, of a well-functioning fetoplacental unit. A specific example of reverse causality is provided by the evidence that twin births are associated with higher net maternal weight gains. Thus, it seems clear that the fetus and placenta can affect maternal weight gain apart from the weight they contribute. Furthermore, the amount of weight gained during pregnancy is inDu- enced by the weights of the fetus, placenta, and amniotic fluid. Nonetheless, studies that examined net (i.e., maternal) weight gain have shown smaller, but consistent, effects on fetal growth (see Chapter 8~. The strongest evidence that gestational weight gain can affect fetal growth is provided by supplementation trials (Lechtig et al., 1975; Mora et al., 1979, Prentice et al., 1983; Rush et al., 1980) and studies of human famine (Stein et al., 1975; Susser and Stein, 1982~. In both of these settings, changes in energy intake result in changes in both gestational weight gain and gestational age-adjusted birth weight. Both nutritional and nonnutritional (i.e., expansion of body water) factors may be involved in producing the effect of gestational weight gain on fetal growth, as suggested by the trial of Campbell and MacGillivray (1975) that included use of energy restriction and diuretics. The causal link is further strengthened by coincidental temporal trends over the last 20 to 30 years: increasing gestational weight gains and mean birth weights and decreasing rates of intrauterine growth retardation.
CAUSALITY AND OPPORTUNITIES FOR INlERVENTION 227 However, the magnitude of the causal effect size (e.g., increase in birth weight) is modest. The data indicate, for example, that in women with a normal prepregnancy weight for height, birth weight increases 20 g, on average, for a change of 1 kg in gestational weight gain (Kramer, 1987~. Thus, a 5-kg (11-lb) difference in gestational weight gain is associated with an average difference in birth weight of only 100 g. The actual change in birth weight for an individual woman who increases or decreases her gestational weight gain by this amount is likely to vary considerably around this average. Moreover, the 20-g/kg effect is probably an overestimate, since the analyses in most studies are based on total rather than on net weight gain (see Chapter 4~. When based on net weight gain, the effect appears to be reduced by one-third to about 13 g of birth weight per 1 kg (2.2 lb) of gestational weight gain (Kramer et al., 1989~. Given this modest effect size and the large variability in weight gains as- sociated with optimal fetal growth, even within groups with similar prepreg- nancy weights for height, knowledge of an individual mother's weight gain does not substantially enhance her (or her physician's) ability to predict the growth of her fetus. Nor is the evidence bearing on this causal link particularly useful in the diagnostic setting. The low birth weight of an infant born to a mother who gained only 5 kg (11 lb) cannot, therefore, be confidently attributed to her low gestational weight gain, nor can the high birth weight of a baby whose mother gained 25 kg (55 lb) be attributed to the high gestational weight gain. Thus, despite the strong evidence that gestational weight gain can affect fetal growth, the chance that it will (or did) in an individual case may be far from clear (Kramer, 1988; Lane, 1984~. With regard to the effects of gestational weight gain on gestational duration, the evidence for causality is considerably weaker than that for fetal growth. Once again, reverse causality is a key issue and one that has not been adequately considered in most epidemiologic studies, especially those in which gestational weight gain has been based on total weight gain, rather than on early weight gain or rate of weight gain. Obviously, a pregnancy that ends prematurely will be associated with a smaller total weight gain, because the mother will have had less total time in which to gain weight. Birth weight can be measured without appreciable random or system- atic error. By contrast, as mentioned above, gestational age is difficult to estimate and errors in such estimates, particularly at the extremes of ma- turity (Kramer et al., 1988), are likely to lead to misclassification of some growth-retarded babies as preterm and preterm babies as growth retarded. Nonetheless, evidence from many of the better epidemiologic studies does suggest that a low rate of gestational weight gain can increase the risk of preterm delivery (Abrams et al., 1989; Berkowitz, 1981; Hediger et al., 1989; Miller and Merritt, 1979; Scholl et al., 1989; van den Berg and Oechsli,
228 NUTRITIONAL STATUS AND WEIGHT GAIN 1984~. The evidence is not unanimous, however (Kleinman, 1990), and there is no clear biologic mechanism whereby changes in gestational weight gain would lead to earlier or later labor and delivery. Furthermore, there has been no clear trend toward increasing gestational duration paralleling the increase in gestational weight gain in the United States during the past two or three decades. For all these reasons, the subcommittee finds that the evidence for an important causal impact on gestational duration is . . Inconclusive. There are fewer data concerning the links between gestational weight gain and the other short-term maternal and child outcomes shown in Figure 2-3. Data from the Collaborative Perinatal Project (Naeye, 1979) and the 1980 National Fetal Mortality Survey Duffel, 1986) indicate an increased perinatal mortality rate among infants born to women with low gestational weight gains, especially those with low prepregnangy weights for height. The evidence does not suggest that gestational weight gain affects congenital anomalies, spontaneous abortion, maternal mortality, or the volume or composition of human milk during lactation. Evidence from the large body of literature concerning the effects of gestational weight gain on fetal growth indicates that large gestational weight gains do increase the risk of high-birth-weight infants (Ounsted and Scott, 1981; Scholl et al., 1988; Udall et al., 1978), which can lead to fetopelvic disproportion and, secondarily, to increased risks of midforceps delivery, cesarean delivery, shoulder dystocia, meconium aspiration, clavic ular fracture, brachial plexus injury, and neonatal asphyxia (Acker et al., 1985; Boyd et al., 1983; Koff and Potter, 1939; Modanlou et al., 1980; Sandmire and O'Halloin, 1988~. But the magnitude of the erect of gesta tional weight gain on these outcomes appears to be small. In particular, very little of the recent marked increase in the rate at which cesarean deliveries are performed can be attributed to larger maternal weight gains. Finally, published evidence indicates that women who have large weight gains during pregnancy tend to remain somewhat heavier in the immediate postpartum period (Billewicz and Thomson, 1970; Greene et al., 1988~. Do Short-Term Maternal and Child Health Effects Lead to Longer-Term Effects? Most of the questions about the long-term effects of gestational weight gain on maternal or child health concern postpartum maternal obesity and the survival, morbidity, growth, and performance of the offspring. Some maternal weight added during pregnancy may be retained permanently, particularly if weight is retained following each of several pregnancies (see Chapter 8~. Further research in this area is clearly required. Obesity has well-documented adverse health consequences, including hypertension,
CAUSALITY AND OPPORTUNITIES FOR INTERVENTION 229 non-insulin-dependent diabetes mellitus, gallbladder disease, osteoarthritis, and increased overall mortality (Hoffmans et al., 1988; Mann, 1974a,b; NIH, 1985; Van Itallie, 1979~. Therefore, any tendency toward excessive fat retention must be regarded as undesirable. Fortunately, most women retain only an average of approximately 1 kg (2.2 lb) per pregnancy, although this figure might underestimate weight retentions associated with the large weight gains observed in recent years. Most of the issues regarding the effects on the child focus on the prognostic Implications of intrauterine growth retardation (IUGR) or high birth weight. There is fairly convincing evidence that IUGR infants have increased morbidity in the newborn period, including increased risks of polycythemia, hypoglycemia, hypocalcemia, and birth asphyxia (Arora et al., 1987; Kramer et al., 1989; Ounsted et al., 1988; Usher, 1970~. The evidence also indicates that IUGR leads to small but persistent effects in stature, brain growth, and neurocognitive performance (as reviewed by Teberg et al., 1988), although control for confounding postnatal influences has not always been adequate in previous studies. In addition, a small but definite risk of fetal and infant mortality seems to be attributable to IUGR (Arora et al., 1987; Haas et al., 1987; Koops et al., 1982; Kramer et al., 1989; Usher, 1970~. High-birth-weight infants, on the other hand, tend to be taller and heavier throughout childhood and have an increased risk of obesity (Binkin et al., 1988; Fisch et al., 1975; Kramer et al., 1985; Ounsted et al., 1982~. If there is confirmation of several recent studies suggesting that in- creased maternal weight gain can increase gestational duration and that women with low weight gain are at increased risk for preterm delivery, the promotion of gestational weight gain might have important long-term bene- fits for child health. Preterm infants, especially those born before 34 weeks of gestation, are at greatly increased risk for perinatal and infant mortal- ity, as well as rather severe and persistent pulmonary and neurocognitive sequelae. The causal impact of preterm birth on these adverse outcomes is extremely well established and of large magnitude. Thus, even if insuf- ficient gestational weight gain increases the risk of preterm birth only to a small degree, the negative effects on child health may be large. These considerations once again underline the need for further research on the link between maternal weight gain and gestational duration. IMPLICATIONS The evidence suggests that an across-the-board increase in gestational weight gain among U.S. women would have both beneficial and adverse effects on maternal and child health. Fetal and infant mortality would be reduced, as would the incidence of IUGR (and its short-term and longer
230 NUTRITIONAL STATUS AND WEIGHT GAIN term sequelae). Such an increase in weight gain would also result in an increased risk of maternal obesity (and its secondary health sequelae) for women with very large weight gains and an increased incidence of high- birth-weight infants, which has some undeniable consequences both for mothers and infants. A formal decision analysis, in which probabilities and utilities (values) are assigned to each potential outcome, might assist in balancing these risks and benefits. Essential components of such an analysis include the probability of each beneficial and adverse outcome occurring with and with- out nutritional counseling, energy supplementation, or changes in maternal attitudes and the value that mothers, children, and society place on each of those outcomes. Even then, it is important to consider who would be helped and who would be harmed, as well as whose values should be con- sidered. For example, if increased energy intakes lead to an elevated risk of maternal obesity but better overall outcomes in the infants, how should risks to the mother be balanced against benefits to the child? And who will make the decision? The analysis could be refined by taking into account the fact that the risks and benefits are likely to differ according to the mother's prepregnancy weight for height, existing energy intake, and other factors that affect the outcomes under consideration. It may well be possible to maximize benefits and minimize risks by focusing educational efforts, individual nutritional counseling, and energy supplementation on women who are undernourished or who have other risk factors. Increased energy intakes by such women can be expected to have a substantial impact on intrauterine growth and its longer-term child health sequelae without incurring the appreciable risks of maternal obesity and high birth weight. Finally, even if the overall benefits of increased energy intake exceed the risks, and even if the above technical and philosophical objections can be overcome, society will need to consider the costs of these interventions. Cost-benefit and cost-effectiveness analytic techniques could be used to de- cide whether the net benefit is worth the expense, or whether the required resources can be more productively channeled in other directions. Public media campaigns, individual nutritional counseling, and energy supplemen- tation all require financial resources. Even if those resources were kept within the maternal and child health sector, the benefit:cost ratio for the various interventions should be compared with that for public health and clinical interventions to convince mothers to stop smoking during pregnancy and with that for family planning and contraceptive services (particularly for adolescents). Although such a balancing of benefits, risks, and costs should play an important role in public health policy in this domain, these considerations extend considerably beyond the subcommittee's mandate.
CAUSALITY AND OPPORTUNITIES FOR INTERVENTION REFERENCES 231 Abrams, B., V. Newman, T. Key, and J. Parker. 1989. Maternal weight gain and preterm delivery. Obstet. Gynecol. 74:577-583. Aeker, D.B., B.P. Saehs, and E.A. Friedman. 1985. Risk factors for shoulder dystocia. Obstet. Gyneeol. 66:762-768. Arora, N.K., V.K. Paul, and M. Singh. 1987. Morbidity and mortality in term infants with intrauterine growth retardation. J. Crop. Pediatr. 33:186-189. Berkowitz, G.S. 1981. An epidemiologic study of preterm delivery. Am. J. Epidemiol. 113:81-92. Billewiez, W.Z., and A M. Thomson. 1970. Body weight in parous women. Br. J. Prev. Soc. Med. 24:97-104. Binkin, N.J., R. Yip, L. Fleshood, and FL. Drawbridge. 1988. Birth weight and childhood growth. Pediatrics 82:828-834. Boyd, M.E., R.H. Usher, and F.H. McLean. 1983. Fetal maerosomia: prediction, risks, proposed management. Obstet. Gynecol. 61:715-72Z Campbell, D.M., and I. MaeGillivray. 1975. The effect of a low calorie diet or a thiazide diuretic on the incidence of pre-eelampsia and on birth-weight. Br. J. Obstet. Gynaecol. 82:572-577. DHEW (Department of Health, Education, and Welfare). 1964. Smoking and Health: Report of the Advisory Committee to the Surgeon General of the Public Health Service. PHS Publ. No. 1103. Public Health Service, U.S. Department of Health, Education, and Welfare. U.S. Government Printing Office, Washington, D.C. 387 pp. Duffus, G.M., I. MaeGillivray, and K.J. Dennis. 1971. Ibe relationship between baby weight and changes in maternal weight, total body water, plasma volume, electrolytes and proteins, and urinary oestriol excretion. J. Obstet. Gynaecol. Br. Commonw. 78:97-104. Edozien, J.C., B.R. Switzer, and R.B. Bryan. 1979. Medical evaluation of the Special Supplemental Food Program for Women, Infants, and Children. Am. J. Clin. Nutr. 32:677-692. Endres, J.M., M. Sawieki, and J.A. Casper. 1981. Dietary assessment of pregnant women in a supplemental food program. J. Am. Diet. Assoe. 79 121-126. Fiseh, R.O., M.K. Bilek, and R. Ulstrom. 1975. Obesity and leanness at birth and their relationship to body habitue in later childhood. Pediatrics 56:521-528. Greene, G.W. , H. Smieiklas-Wright, T. O. Scholl, and RJ. Karp. 1988. Postpartum weight change: how much of the weight gained in pregnancy will be lost after delivery. Obstet. Gyneeol. 71:701-707. Haas, J.D., H. Baleazar, and L. Caulfield. 1987. Variation in early neonatal mortality for different types of fetal growth retardation. Am. J. Phys. Anthropol. 73:467473. Hediger, M.L., TO. Scholl, D.H. Belsly, I.G. Anees, and R.W. Salmon. 1989. Patterns of weight gain in adolescent pregnancy: effects on birth weight and preterm delivery. Obstet. Gyneeol. 74:6-12. Higgins, A.C. 1976. Nutritional status and the outcome of pregnancy. J. Can. Diet. Assoe. 37:17-35. Higgins, A. C., J.E. Moxley, P. B. Pencharz, D. Mikolainis, and S. . Dubois. 1989. Impact of the Higgins Nutrition InteIvention Program on birth weight a within-mother analysis. J. Am. Diet. Assoc. 89:1097-1103. Hill, A.B. 1965. We environment and disease: association or causation? Proe. R. Soe. Med. 58:295-300. Hoffmans, M.D., D. Kromhout, and C. de Lezenne Coulander. 1988. The impact of body mass index of 78,612 18-year old Dutch men on 32-year mortality from all causes. J. Clin. Epidemiol. 41:749-756. Hytten, F.E., and A.M. Thomson. 1976. Weight gain in pregnant. Pp. 179-187 in M.D. Lindheimer, NI. Katz, and F.P. Zuspan, eds. Hypertension in Pregnancy. John Wiley & Sons, New York.
232 NUTRITIONAL STATUS AND WEIGHT GAIN Kleinman, J.C. 1990. Maternal Weight Gain During Pregnancy Determinants and Conse- quences. NCHS Working Paper Series No. 33. National Center for Health Statistics, Public Health Service, U.S. Department of Health and Human Services, Hyattsville, Md. 24 pp. Koff, ~K, and E.L Potter. 1939. The complications associated with excessive development of the human fetus. Am. J. Obstet. Gynecol. 38:412-423. Koops, B.L~, LO. Morgan, and F.C. Battaglia. 1982. Neonatal mortality risk in relation to birth weight and gestational age: update. J. Pediatr. 101:969-977. Kramer, M.S. 198~7. Determinants of low birth weight: methodological assessment and meta-analysis. Bull. W.H.O. 65:663-737. Kramer, M.S. 1988. Causality. Pp. 255-269 in Clinical Epidemiology and Biostatistics: A Primer for Clinical Investigators and Decision-Makem. Springer-Verlag, Berlin. Kramer, M.S ., R.G. Barr, D. G. Leduc, C. BoisJoly, and I.B. Pless. 1985. Infant determinants of childhood weight and adiposity. J. Pediatr. 107:104-107. Kramer, M.S., F.H. McLean, M.E. Boyd, and R.H. Usher. 1988. The validity of gestational age estimation by menstrual dating in term, preterm, and postterm gestations. J. Am. Med. Assoc. 260:330~3308. Kramer, M.S., F.H. McLean, M. Olivier, D.M. W'llis, and R.H. Usher. 1989. Body proportionality and head and length 'sparing' in growth-retarded neonates: a critical reappraisal. Pediatrics 84:717-723. Lane, D.A. 1984. A probabilist's view of causality assessment. Drug Inf. J. 18:323-330. Lechtig, ^, J.P. Habicht, H. Delgado, R.E. Klein, C. Yarbrough, and R. Martorell. 1975. Effect of food supplementation during pregnancy on birthweight. Pediatrics 56:508-520. Mann, G.V. 1974a. The influence of obesity on health (first of two parts). N. Engl. J. Med. 291:178-185. Mann, G.V. 1974b. The influence of obesity on health (second of two parts). N. Engl. J. Med. 291:226-232. Metcoff, J., P. Costiloe, W.M. Crosby, S. Dutta, H.H. Sandstead, D. Milne, C.E. Bodwell, and S.H. Majors. 1985. Effect of food supplementation (WIC) during pregnancy on birth weight. Am. J. Clin. Nutr. 41:933-947. Miller, H.C., and T.A. Merritt. 1979. Fetal Growth in Humans. Year Book Medical Publishers, Chicago. 180 pp. Modanlou, H.D., W.L Dorchester, A. Thorosian, and R.K Freeman. 1980. Macrosomia maternal, fetal, and neonatal implications. Obstet. Gynecol. 55:420~24. Mora, J.O., B. de Paredes, M. Wagner, L~ de Navarro, J. Suescum, N. Christiansen, and M.G. Herrera. 1979. Nutritional supplementation and the outcome of pregnancy. I. Birth weight. Am. J. Clin. Nutr. 32:455-462. Naeye, R.L~ 1979. Weight gain and the outcome of pregnan~y. Am. J. Obstet. Gynecol. 135:3-9. Naeye' R.L~ 1981a. Maternal blood pressure and fetal growth. Am. J. Obstet. Gynecol. 141:780-787. Naeye, R.L. 1981b. Nutritional/nonnutritional interactions that affect the outcome of pregnanc~r. Am. J. Clin. Nutr. 34:727-731. Naeye, R.L, and R.A. Chez. 1981. Effects of maternal acetonuria and low pregnancy weight gain on children's psychomotor development. Am. J. Obstet. Gynecol. 139:189-193. NIH (National Institutes of Health). 1985. Health implications of obesity: National Institutes of Health Consensus Development Conference Statement. Ann. Intern. Med. 103:1073-1077. Ounsted, M., and A. Scott. 1981. Associations between maternal weight, height, weight- for-height, weight-gain and birth weight. Pp. 113-129 in J. Dobbing, ed. Maternal Nutrition in Pregnancy: Eating for Two? Academic Press, London. Ounsted, M., V. Moar, and ~ Scott. 1982. Growth in the first four years. II. Diversity within groups of small-for-date and large-for-date babies. Early Hum. Dev. 7:29-39. Ounsted, M., V.A. Moar, and A. Scott. 1988. Neurological development of small-for- gestational age babies during the first year of life. Early Hum. Dev. 16:163-172. Prentice, A.M., R.G. Whitehead, M. Watkinson, W.H. Lamb, and TJ. Cole. 1983. Prenatal dietary supplementation of African women and birth-weight. Lancet 1:489~92.
CAUSALITY AND OPPORTUNITIES FOR INTERVENTION 233 Primrose, T., and A. Higgins. 1971. A study in human antepartum nutrition. J. Reprod. Med. 7:257-264. Rush, D. 1981. Nutritional services during pregnancy and birthweight: a retrospective matched pair analysis. Can. Med. Assoc. J. 125:567-576. Rush, D., Z. Stein, and M. Susser. 1980. A randomized controlled trial of prenatal nutritional supplementation in New York City. Pediatrics 65:683~97. Rush, D., N.L~ Sloan, J. Leighton, J.M. Alvir, D.G. Hontitz, W.B. Seaver, G.C Garbowski, S.S. Johnson, R.A Kulka, M. Holt, J.W. Devore, J.T. Lynch, M.B. Woodside, and D.S. Shanklin. 1988. The National WIC Evaluation: evaluation of the Special Supplemental Food Program for Women, Infants, and Children. V. Longitudinal study of pregnant women. Am. J. Clin. Nutr. 48:439~83. Sandmire, H.F., and T.J. O'Halloin. 1988. Shoulder dystocia: its incidence and associated risk factors. Int. J. Gynaecol. Obstet. 26:65-73. Scholl, T.O., M.L~ Hediger, I.G. Ances, and C.E. Cronk. 1988. Growth during early teenage pregnancies. Lancet 1:701-702. Scholl, T.O., M.L. Hedi~er, R.W. Salmon, D.H. Belsky, and I.G. Ances. 1989. Influence of prepregnant body mass and weight gain for gestation on spontaneous preterm delivery and duration of gestation during adolescent pregnancy. Am. J. Hum. Biol. 1:657-664. Sims, E.A.H., R.~. Goldman, CM. Gluck, E.S. Horton, P.C. Kelleher, and D.W. Rowe. 1968. Experimental obesity in man. Trans. Assoc. Am. Physicians 81:153-170. Singer, J.E., M. Westphal, and K. Niswander. 1968. Relationship of weight gain during pregnancy to birth weight and infant growth and development in the fimt year of life: a report from the Collaborative Study of Cerebral Palsy. Obstet. Gynecol. 31:417-423. Stein, ~, M. Susser, G. Saenger, and F. Marolla. 1975. Famine and Human Development: The Dutch Hunger W~nter of 1944-1945. Oxford University Press, New York. 284 pp. Susser, M. 1973. Causal Thinking in the Health Sciences: Concepts and Strategies of Epidemiology. Oxford University Press, New York. 181 pp. Susser, M. 1988. Falsification, verification and causal inference in epidemiology: reconsid- erations in the light of Sir Karl Popper's philosophy. Pp. 33-57 in KJ. Rothman, ed. Causal Inference. Epidemiology Resources, Chestnut Hill, Mass. Susser, M., and Z. Stein. 1982. Third variable analysis: application to causal sequences among nutrient intake, maternal weight, birthweight, placental weight, and gestation. Stat. Med. 1:105-120. Taffel, S.M. 1986. Maternal Weight Gain and the Outcome of Pregnancy: United States, 1980. Vital and Health Statistics, Series 21, No. 44. DHHS Publ. No. (PHS) 86-1922. National Center for Health Statistics, Public Health Se~vice, U.S. Department of Health and Human Services, Hyattsville, Md. 25 pp. Tavris, D.R., and J.A. Read. 1982. Effect of maternal weight gain on fetal, infant, and childhood death and on cognitive development. Obstet. G:rnecol. 60:689~94. Teberg, A.J., FJ. Walther, and I.C. Penal 1988. Mortality, morbidity, and outcome of the small-for-gestational age infant. Semin. Perinatol. 12:84-94. Thomson, A.M., F.E. Hytten, and W.Z. Billewicz. 1967. The epidemiology of oedema during pregnangy. J. Obstet. Gynaecol. Br. Commonw. 74:1-10. Udall, J.N., G.G. Harrison, Y. Vaucher, P.D. Walson, and G. Morrow III. 1978. Interaction of maternal and neonatal obesity. Pediatrics 62:17-21. Usher, R.H. 1970. Clinical and therapeutic aspects of fetal malnutrition. Pediatr. Clin. North Am. 17:169-183. van den Berg, BJ., and F.W. Oechsli. 1984. Prematurity. Pp. 69-85 in M.B. Bracken, ed. Perinatal Epidemiology. Oxford Univemity Press, New York. Van Itallie, T.B. 1979. Obesity: adverse effects on health and longevity. Am. J. Clin. Nutr. 32:2723-2733.