Summary This report reviews some of the more recent epidemiologic and experimental findings on the relationship of tobacco smoking to lung cancer, and discusses some criticisms directed against the conclusion that tobacco smoking, especially cigarettes, has a causal role in the increase in broncho-genic carcinoma. The magnitude of the excess lung-cancer risk among cigarette smokers is so great that the results can not be interpreted as arising from an indirect association of cigarette smoking with some other agent or characteristic, since this hypothetical agent would have to be at least as strongly associated with lung cancer as cigarette use; no such agent has been found or suggested. The consistency of all the epidemiologic and experimental evidence also supports the conclusion of a causal relationship with cigarette smoking, while there are serious inconsistencies in reconciling the evidence with other hypotheses which have been advanced. Unquestionably there are areas where more research is necessary, and, of course, no single cause accounts for all lung cancer. The information already available, however, is sufficient for planning and activating public health measures. – J. Nat. Cancer Inst. 22:173–203, 1959.

Background Increased autism prevalence rates have generated considerable concern. However, the contribution of changes in diagnostic practices to increased prevalence rates has not been thoroughly examined. Debates over the role of diagnostic substitution also continue. California has been an important test case in these controversies. The objective of this study was to determine the extent to which the increased prevalence of autism in California has been driven by changes in diagnostic practices, diagnostic substitution and diagnostic accretion.

Methods Retrospective case record examination of 7003 patients born before 1987 with autism who were enrolled with the California Department of Developmental Services between 1992 and 2005 was carried out. Of principal interest were 631 patients with a sole diagnosis of mental retardation (MR) who subsequently acquired a diagnosis of autism. The outcome of interest was the probability of acquiring a diagnosis of autism as a result of changes in diagnostic practices was calculated. The probability of diagnostic change is then used to model the proportion of the autism caseload arising from changing diagnostic practices.

Results The odds of a patient acquiring an autism diagnosis were elevated in periods in which the practices for diagnosing autism changed. The odds of change in years in which diagnostic practices changed were 1.68 [95% confidence interval (CI) 1.11–2.54], 1.55 (95% CI 1.03–2.34), 1.58 (95% CI 1.05–2.39), 1.82 (95% CI 1.23–2.7) and 1.61 (95% CI 1.09–2.39). Using the probability of change between 1992 and 2005 to generalize to the population with autism, it is estimated that 26.4% (95% CI 16.25–36.48) of the increased autism caseload in California is uniquely associated with diagnostic change through a single pathway—individuals previously diagnosed with MR.

Conclusion Changes in practices for diagnosing autism have had a substantial effect on autism caseloads, accounting for one-quarter of the observed increase in prevalence in California between 1992 and 2005.

Background It is unclear whether the increase in autism over the past two decades is a real increase or due to changes in diagnosis and ascertainment of autism spectrum disorders (ASDs), which include autism, Asperger syndrome and pervasive developmental disorder not otherwise specified (PDD-NOS). The aim of this study was to examine the trends in ASD over time in Western Australia (WA) and the possible effects and contribution of changes in diagnostic criteria, age at diagnosis, eligibility for service provision based on ASD diagnoses and changes in diagnostic practices.

Methods A population-based study was conducted among the cohort of children born in WA between 1983 and 1999 and diagnosed with ASD between the age of 2 and 8 years up to December 31, 2004. The trend in ASD diagnosis over the study period was assessed by investigating birth cohort and period effects, and examining whether these were modified by age of diagnosis. ASD diagnosis corresponding with changes in diagnostic criteria, funding and service provision over time were also investigated. A subgroup analysis of children aged ≤5 years was also conducted to examine trends in the incidence and age of diagnosis of ASD and intellectual disability (ID) and to investigate the role of changes in diagnostic practices.

Results The overall prevalence of ASD among children born between 1983 and 1999 and diagnosed by age 8 was 30 per 10 000 births with the prevalence of autism comprising 21 per 10 000 births. The prevalence of ASD increased by 11.9% per annum, from 8 cases per 10 000 births in 1983 to 46 cases per 10 000 births in 1999. The annual incidence of ASD, based on newly diagnosed ASD cases in each year from 1985 to 2002, increased over the study period. The increase in incidence of ASD appeared to coincide with changes in diagnostic criteria and availability of funding and services in WA, particularly for children aged <5 years. The age-specific rates of autism and PDD-NOS increased over time and the median age of diagnosis for autism decreased from 4 to 3 years of age throughout the 1990s. For children aged ≤5 years the incidence of ASD diagnosis increased significantly from 1992, with an average annual increase of 22%. Similar findings were found for autism. In the corresponding years the incidence of diagnosis of severe ID fell by 10% per annum and mild–moderate ID increased by 3% per annum.

Conclusions The rise in incidence of all types of ASDs by year of diagnosis appears to be related to changes in diagnostic and service provision practices in WA. In children aged ≤5 years, diagnosis of severe ID decreased, but mild–moderate ID increased during the study period. A true increase in ASD cannot be ruled out.

Background It is unclear whether the incidence of first episode psychoses is in decline. We had the opportunity to determine whether incidence had changed over a 20-year period in a single setting, and test whether this could be explained by demographic or clinical changes.

Methods The entire population at-risk aged 16–54 in Nottingham over three time periods (1978–80, 1993–95 and 1997–99) were followed up. All participants presenting with an ICD-9/10 first episode psychosis were included. The remainder of the population at-risk formed the denominator. Standardized incidence rates were calculated at each time period with possible change over time assessed via Poisson regression. We studied six outcomes: substance-induced psychoses, schizophrenia, other non-affective psychoses, manic psychoses, depressive psychoses and all psychotic disorders combined.

Results Three hundred and forty-seven participants with a first episode psychosis during 1.2 million person-years of follow-up over three time periods were identified. The incidence of non-affective or affective psychoses had not changed over time following standardization for age, sex and ethnicity. We observed a linear increase in the incidence of substance-induced psychosis, per annum, over time (incidence rate ratios: 1.15; 95% CI 1.05–1.25). This could not be explained by longitudinal changes in the age, sex and ethnic structure of the population at-risk.

Conclusions Our findings suggest psychotic disorders are not in decline, though there has been a change in the syndromal presentation of non-affective disorders, away from schizophrenia towards other non-affective psychoses. The incidence of substance-induced psychosis has increased, consistent with increases in substance toxicity over time, rather than changes in the prevalence or vulnerability to substance misuse. Increased clinical and popular awareness of substance misuse could also not be excluded.

Background The excess risk of fatal and non-fatal cerebrovascular disease in people from low socioeconomic positions is only partially explained by conventional cerebrovascular risk factors. This has led to the suggestion that poor psychosocial work environments provide important additional explanatory power. However, little evidence is available for women.

Methods We examined whether job demands or job control contributed to the socioeconomic gradient in cerebrovascular disease among 48 361 women aged 18–65 years. Job demands, job control and behavioural risk factors were self-reported in 2000–2002; socioeconomic position (as indexed by occupational class) and all of the health measures were obtained from registers. The outcome was recorded hospitalization or death from cerebrovascular disease.

Results During a mean follow-up of 3.4 years, 124 women had a new cerebrovascular disease event. The risk was 2.3 (95% CI 1.3–3.9) times higher among women in low vs high socioeconomic positions. Adjustment for conventional risk factors, such as prevalent hypertension, coronary heart disease, diabetes, smoking, heavy alcohol consumption, physical inactivity and obesity, attenuated this excess risk by 23%. In contrast, adjustment for job demands and job control actually amplified the gradient by 36% suggesting a suppression effect.

Conclusions In this contemporary cohort of employed women, job demands—alone and in combination with job control—suppressed rather than explained socioeconomic differences in cerebrovascular disease.

Background Previous studies have assessed the relative importance of material, psychosocial and behavioural factors in the explanation of relative socio-economic inequalities in mortality, but research into the contribution of biomedical factors has been limited. Our study examines the relative contribution of (i) material, (ii) psychosocial, (iii) behavioural and (iv) biomedical factors in the explanation of relative socio-economic (educational and income) inequalities in mortality.

Methods Cohort study—baseline data from the Norwegian total county population-based HUNT 2 study linked to mortality data (1995/97 to 2003). In this analysis, 18 247 men and 18 278 women aged 24–80 without severe chronic disease at baseline were eligible.

Results No socio-economic inequalities in mortality among women were found. In men, educational- and income-related inequalities in mortality were found with a relative risk for the lowest educational group of 1.67 (1.29–2.15) and the lowest income quartile of 2.03 (1.57–2.70). Together, the four explanatory factors reduced the relative risk of mortality of the lowest educational group to 1.18 (0.90–1.55) and the relative risk of mortality in the lowest income quartile was attenuated to 1.17 (0.83–1.63). Known biomedical factors contributed least to both educational and income inequalities in mortality.

Conclusions Material factors were the most important in explaining income inequalities in mortality amongst men, whereas psychosocial and behavioural factors were the most important in explaining educational inequalities. This suggests that improving the material, psychosocial and behavioural circumstances of men might bring more substantial reductions in relative socio-economic inequalities in mortality.

Background Tuberculosis is known to have socio-economic determinants at individual and at area levels, but it is not known whether they are independent, whether they interact and their relative contributions to the burden of tuberculosis.

Methods A case–control study was conducted in Recife, Brazil, to investigate individual and area social determinants of tuberculosis, to explore the relationship between determinants at the two levels and to calculate their relative contribution to the burden of tuberculosis. It included 1452 cases of tuberculosis diagnosed by the tuberculosis services and 5808 controls selected at random from questionnaires completed for the demographic census. Exhaustive information on social factors was collected from cases, using the questionnaire used in the census. Socio-economic information for areas was downloaded from the census. Multilevel logistic regression investigated individual and area effects.

Results There was a marked and independent influence of social variables on the risk of tuberculosis, both at individual and area levels. At individual level, being aged ≥20, being male, being illiterate, not working in the previous 7 days and possessing few goods, all increased the risk of tuberculosis. At area level, living in an area with many illiterate people and where few households own a computer also increased this risk; individual and area levels did not appear to interact. Twice as many cases were attributable to social variables at individual level than at area level.

Conclusions Although individual characteristics are the main contributor to the risk of tuberculosis, contextual characteristics make a substantial independent contribution.

Background The notion that chronic stress contributes to health inequalities by socio-economic status (SES) through physiological wear and tear has received widespread attention. This article reviews the literature testing associations between SES and cortisol, an important biomarker of stress, as well as the summary index of allostatic load (AL).

Methods A search of all published literature on the PubMed and ISI Web of Knowledge literature search engines was conducted using broad search terms. The authors reviewed abstracts and selected articles that met the inclusion criteria. A total of 26 published studies were included in the review.

Results Overall, SES was not consistently related to cortisol. Although several studies found an association between lower SES and higher levels of cortisol, many found no association, with some finding the opposite relationship. Lower SES was more consistently related to a blunted pattern of diurnal cortisol secretion, but whether this corresponded to higher or lower overall cortisol exposure varied by study. Approaches to collecting and analysing cortisol varied widely, likely contributing to inconsistent results. Lower SES was more consistently related to higher levels of AL, but primarily via the cardiovascular and metabolic components of AL rather than the neuroendocrine markers.

Conclusions Current empirical evidence linking SES to cortisol and AL is weak. Future work should standardize approaches to measuring SES, chronic stress and cortisol to better understand these relationships.

Background More years of schooling is generally associated with better health. However, this association may be confounded by unobserved common prior causes such as inherited ability, personality such as patience, or early family circumstances. The twin fixed-effect approach can potentially address this problem by cancelling these factors between twin pairs. The purpose of this study is to identify the causal effects of education on health and health behaviours using a twin fixed-effect approach.

Methods We used twin data from the National Survey of Midlife Development in the United States, 1995–1996. The study population included 302 male [55.6% monozygotic (MZ) and 44.4% dizygotic (DZ)] and 387 female twin pairs (47.3% MZ and 52.7% DZ). A range of health outcomes [perceived global, physical and mental health, body mass index (BMI), waist circumference, waist–hip ratio, number of depressive symptoms] and health behaviours (smoking and physical activity) were examined among twin pairs who were discordant on years of schooling.

Results Among MZ twins, more years of education was associated with better perceived global health. For all other health outcomes/behaviours, the point estimates of the effect of education in the fixed-effect analyses suggested a weak protective association. Among DZ male twins, each additional year of schooling lowered the prevalence of smoking by 32% [odds ratio (OR): 0.68, 95% confidence interval (CI): 0.48–0.97] in the fixed-effect analysis.

Conclusion The widely reported associations between schooling and health outcomes/behaviours may not reflect causal relationships in every instance. Although low statistical power may explain some of the null associations, our twin fixed-effect analyses suggest that at least some cases of the education/health relationship reflect confounding by unobserved third variables.

Objective To examine the effects, by income group, of targeted food taxes and subsidies on nutrition, health and expenditure in the UK.

Methods A model based on consumption data and demand elasticity was constructed to predict the effects of four food taxation-subsidy regimens. Resulting changes in demand, expenditure, nutrition, cardiovascular disease (CVD) and cancer mortality were estimated.

Data Expenditure data were taken from the Expenditure and Food Survey; estimates of price elasticities of demand for food were taken from a report based on the National Food Survey 1988–2000. Estimates of effect on CVD and cancer mortality of changing fat, salt, fruit and vegetable intake were taken from previous meta-analyses.

Results (i) Taxing principal sources of dietary saturated fat is unlikely to reduce cardiovascular disease (CVD) or cancer mortality. (ii) Taxing ‘less healthy’ foods (defined by the WXYfm nutrient profiling model) could increase CVD and cancer deaths by 35–1300 yearly. (iii) Taxing ‘less healthy’ foods and subsidising fruits and vegetables by 17.5% could avert up to 2900 CVD and cancer deaths yearly. (iv) Taxing ‘less healthy’ foods and using all tax revenue to subsidize fruits and vegetables could avert up to 6400 CVD and cancer deaths yearly. Few obesity-related CVD deaths are averted by any of the regimens. All four regimens would be economically regressive and positive health effects will not necessarily be greater in lower-income groups where the need for dietary improvement is higher.

Conclusions A targeted food tax combined with the appropriate subsidy on fruits and vegetables could reduce deaths from CVD and cancer.

Background Self-inflicted injury in youth has increased in many Western countries during recent decades. Education is the most influential societal determinant of living conditions in young people after early childhood. This study tested the hypothesis that school performance predicts self-inflicted injury.

Methods A national cohort of 447 929 children born during 1973–77 was followed prospectively in the National Patient Discharge Register from the end of their ninth and last year of compulsory school until 2001. Multivariate Cox analyses of proportional hazards were used to test hypotheses regarding grades in ninth grade as predictors of hospital admission due to self-inflicted injury.

Results The risk of hospital admission because of self-inflicted injury increased steeply in a step-wise manner with decreasing grade point average. Hazard ratios were 6.2 (95% confidence interval 5.5–7.0) in those with the lowest level of grade point average compared with the highest. The risks were similar for women and men. Adjustment for potential socio-economic confounders in a multivariate proportional hazards regression analysis attenuated this strong gradient only marginally.

Conclusion School performance is a strong factor for predicting future mental ill-health as expressed by self-inflicted injury.

Background Children are at high risk of exposure to environmental tobacco smoke and biofuel smoke at home in developing countries. This study examines whether exposure to cigarette and biofuel smoke is associated with height-for-age of children (0–59 months) in seven developing countries.

Methods The data are from Demographic and Health Surveys conducted in Cambodia, Dominican Republic, Haiti, Jordan, Moldova, Namibia and Nepal between 2005 and 2007. The respondents were women (15–49 years) and their children in seven countries (n = 28 439), and men (15–59 years) from four countries. The outcome is a physical measurement of child height-for-age in standard deviation units.

Results Multilevel regression analysis showed that the country of residence modified the association between maternal smoking and child height-for-age. Exposure to maternal smoking was associated negatively with child height-for-age in Cambodia, Namibia and Nepal, whereas it was not in other countries. Multilevel regression analysis revealed that biofuel smoke exposure was associated with a decrease in child height-for-age [b = –0.13, 95% confidence interval (CI) = –0.19 to –0.07, P < 0.001]. No interaction was found between country of residence and biofuel smoke exposure. Multinomial logistic regression results showed that biofuel smoke exposure was associated with both stunting [odds ratio (OR) = 1.25, 95% CI = 1.08–1.44, P = 0.002) and severe stunting (OR = 1.27, 95% CI = 1.02–1.59, P = 0.04), after controlling for covariates.

Conclusion The findings suggest that exposure to maternal smoking and biofuel smoke may contribute to growth deficiencies in young children. Programmes are needed to ensure smoke-free home environments for children.

Background Exposure to indoor air pollution due to open burning of biomass fuel is common in low- and middle-income countries. Previous studies linked this exposure to an increased risk of respiratory illness, low birth weight (LBW) and other disorders. We assessed the association between exposure to biomass fuel sources and second-hand tobacco smoke (SHTS) in the home and adverse health outcomes in early infancy in a population in rural south India.

Methods A population-based cohort of newborns was followed from birth through 6 months. Household characteristics were assessed during an enrolment interview including the primary type of cooking fuel and smoking behaviour of household residents. Follow-up visits for morbidity were carried out every 2 weeks after delivery. Infants were discharged at 6 months when anthropometric measurements were collected.

Results 11 728 live-born infants were enrolled and followed, of whom 92.3% resided in households that used wood and/or dung as a primary source of fuel. Exposure to biomass fuel was associated with an adjusted 49% increased risk of LBW, a 34% increased incidence of respiratory illness and a 21% increased risk of 6-month infant mortality. Exposed infants also had 45 and 30% increased risks of underweight and stunting at 6 months. SHTS exposure was also associated with these adverse health outcomes except for attained growth.

Conclusions Open burning of biomass fuel in the home is associated with significant health risks to the newborn child and young infant. Community-based trials are needed to clarify causal connections and identify effective approaches to reduce this burden of illnesses.

Background The P-value approach has been employed to prioritizing genome-wide association (GWA) scan signals, with a genome-wide significance defined by a prior P-value threshold, although this is not ideal. A rationale put forward is that the association signals rather should be expected to give less support for single nucleotide polymorphisms (SNPs) that are rare (with associated low-power tests) than for common SNPs with equivalent P-values, unless investigators believe, a priori, that rare causative variants contribute to the disease and have more pronounced effects.

Methods Using data from a GWA scan for type 2 diabetes (1924 cases, 2938 controls, 393 453 SNPs), we compared P-values with four alternative signal measures: likelihood ratio (LR), Bayes factor (BF; with a specified prior distribution for true effects), ‘frequentist factor’ (FF; reflecting the ratio between estimated—post-data— ‘power’ and P-value) and probability of pronounced effect size (PrPES).

Results The 19 common SNPs [minor allele frequency (MAF) among the controls >29%] yielding strong P-value signals (P < 5 x 10^–7) were also top ranked by the other approaches. There was a strong similarity between the P-values, LR and BF signals, in terms of ranking SNPs. In contrast, FF and PrPES signals down-weighted rare SNPs (control MAF <10%) with low P-values.

Conclusions For prioritization of signals that do not achieve compelling levels of evidence for association, the main driving force behind observed differences between the various association signals appears to be SNP MAF. The statistical power afforded by follow-up samples for establishing replication should be taken into account when tailoring the signal selection strategy.

Background The human paraoxonase (PON1) protein detoxifies certain organophosphates, and the PON1 Q192R polymorphism (rs662) affects PON1 activity. Groups with higher dose exposure to organophosphate sheep dips or first Gulf War nerve toxins reported poorer health if they had 192R, and these associations have been used to exemplify Mendelian randomization analysis. However, a reported association of 192R with depression in a population-based study of older women recently cast doubt on the specificity of the higher dose findings. We aimed to examine associations between the PON1 Q192R polymorphism and self-reported poor health and depression in two independent population-based studies.

Methods We used logistic regression models to examine the associations in men and women aged 60–79 years from the English Longitudinal Study of Ageing (ELSA, n = 3158) and InCHIANTI (n = 761) population studies. Outcomes included the Center for Epidemiologic Studies Depression (CES-D) scale, self-rated general health status and (in ELSA only) diagnoses of depression.

Results The PON1 Q192R polymorphism was not associated with self-reported poor health {meta-analysis: odds ratio (OR) = 1.01 [confidence interval (CI) 0.91–1.13], P = 0.80} or depressive symptoms in either study or in meta-analyses [CES-D: OR = 1.01 (CI 0.87–1.17), P = 0.90]. There was also no association with histories of diagnosed depression in ELSA [OR = 1.03 (CI 0.82–1.30), P = 0.80].

Conclusions We found no evidence of an association between the PON1 Q192R polymorphism and poor general or mental health in two independent population-based studies. Neither the claimed Q192R association with depression in the general population nor its theoretical implications were supported.

Background The role of maternal anaemia in preterm birth remains poorly defined, and the association between anaemia and preterm birth clinical subtypes remain unclear. We examined if maternal anaemia exposure both within and across trimesters during gestation is associated with preterm birth.

Methods This was a secondary analysis of data from a population-based prospective cohort study in 13 counties of East China (1993–96). All singleton live births delivered at 20–44 weeks to women with at least one haemoglobin measure during pregnancy were included (n = 160 700). Risk of preterm birth (<37 weeks) was examined by clinical subtypes, namely, preterm premature rupture of membranes (PROM), spontaneous preterm labour and medically indicated preterm birth. Haemoglobin changes across trimesters were assessed as proxy of haemo-dilution and haemo-concentration. Multivariable Cox proportional hazards regression models were fitted.

Results Preterm birth rates of preterm birth were 4.1% for anaemic and 5% for non-anaemic women (P < 0.05). Compared with haemoglobin of 11 g/dl (reference), values ≤5 g/dl in the first trimester were associated with increased risk for preterm PROM [hazard ratio (HR) 3.3, 95% confidence interval (CI) 1.4–7.7], whereas low haemoglobin in the third trimester was associated with reduced risk of spontaneous preterm labour. Haemodilution was associated with reduced risk for preterm birth.

Conclusions Anaemia in early pregnancy was found to be associated with increased risk for preterm PROM, whereas exposure in late pregnancy was associated with reduced risk for spontaneous preterm labour.

Background The nature of the association between birth weight and behavioural problems in adolescence is unclear. Recent studies are limited by their capacity to adjust for important obstetric and measurement issues.

Aim To examine the nature of the association between birth weight and a range of behavioural symptoms, including anxiety/depression and social problems, in adolescence.

Methods Data from 4971 mothers and their children participating in the Mater University Study of Pregnancy (MUSP), a prospective, population-based birth cohort, are presented. This study commenced in Brisbane, Australia, in 1981. Mothers and their children were followed up at 3–5 days post-birth, and 6 months, 5 years and 14 years after the initial interview. Internalizing and externalizing behaviour problems, social problems and anxiety/depressive symptoms were assessed using Achenbach's Youth Self Report (YSR) of the Child Behaviour Check List (CBCL).

Results There was no evidence of a linear association between birth weight and behavioural symptoms, when birth weight z-scores were examined as a continuous variable. However, a non-linear association was identified when birth weight z-scores were categorized into quintiles. Children in the lowest and highest quintiles were at higher risk of increased anxiety/depressive and social problems symptoms. After adjustment for potential confounders, birth weight showed a non-linear association with the log odds of social problems {Quintile 1 odds ratio (OR) 1.59 [95% confidence interval (CI) 1.13, 2.23] Quintile 5 OR 1.57 (95% CI 1.12, 2.20)}.

Conclusions Our findings provide further support for an association between birth weight and some adolescent behaviour problems. This association is likely to be non-linear, affecting babies at both the lower and higher ends of the birth weight distribution.

Background Evidence suggests that intrauterine growth restriction followed by rapid post-natal growth is associated with high blood pressure. We assessed the effect of early size and post-natal growth on blood pressure in a population from West Africa, where fetal growth retardation and childhood malnutrition are common.

Methods A total of 1288 Senegalese subjects were followed from infancy to young adulthood (mean age 17.9 years). Adult systolic blood pressure (SBP) was regressed on infant and adult anthropometric characteristics.

Results In unadjusted analyses, infant size was positively associated with adult SBP (1.1 ± 0.3; P = 0.001 for weight; 0.7 ± 0.3; P = 0.04 for length). With adjustment for current size, the regression coefficients for infant size were reversed (–0.2 ± 0.3; P = 0.51 for weight; –0.3 ± 0.3; P = 0.35 for length). SBP increased by 4.1 and 2.9 mmHg for 1 standard deviation (SD) increase in current weight or height, respectively. No interaction between infant size and current size was found in the overall models (P = 0.11 for weight, P = 0.95 for height), but this term interacted with sex for weight effect. A negative interaction was found in males (–0.9 ± 0.4; P = 0.02) but not in females (0.3 ± 0.4; P = 0.46). The association of current weight with SBP was stronger in lighter weight male infants.

Conclusions These findings support the hypothesis that subjects who were small in early life and experienced enhanced post-natal growth have higher levels of SBP, even in low-income settings.