War, Humanitarian Crises, Population Displacement, and Fertility A Review of Evidence (2004) / Chapter Skim
Currently Skimming:
Contents of Report
Contents of Report
Pages 1-34
The Chapter Skim interface presents what we've algorithmically identified as the most significant single chunk of text within every page in the chapter.
Select key terms on the right to highlight them within pages of the chapter.
From page 1... ...
Although these are entirely appropriate priorities, little attention has been paid to the consequences for fertility, the major determinant of medium-term population dynamics, of humanitarian crises. The number of studies of fertility in refugee or displaced person populations has been very limited (Hynes et al., 2002)
|
From page 2... ...
. Whereas Davis and Blake were interested in cultural factors, of interest here is the impact of humanitarian crises and population displacement, but the principle is the same: change in the intermediate variables is necessary and sufficient for fertility change, so any effect of crisis must work through one or more of these variables.
|
From page 3... ...
I will thus retain the broader range of intermediate variables proposed by Davis and Blake, in order to be sure that abnormal circumstances are covered. The key characteristics of intermediate variables are that they are both necessary and sufficient for fertility change, and that the direction of change in fertility resulting from a change in an intermediate variable is unambiguous.
|
From page 4... ...
, effects on fertility can be expected to be largely involuntary, except perhaps for changes in the incidence of induced abortion. In a population with a substantial level of fertility control, in contrast, changes in use of contraception in response to changes in fertility preferences (concerning both timing and ultimate family size)
|
From page 5... ...
However, registration of births in many developing countries is not complete, and a humanitarian crisis is likely to have an additional adverse effect on an existing registration system. Thus civil registration data are not generally useful sources of data about the fertility effects of severe crises (although they may be useful for studying the effects of economic fluctuations on fertility)
|
From page 6... ...
The time-truncated birth history has been found to underestimate fertility greatly, and birth histories truncated by some number of births are likely to be affected by errors of date recording that are also likely to reduce recent fertility.
|
From page 7... ...
, and fertility trends from birth histories have been found often to exaggerate fertility declines, as a result of reporting births further in the past than they actually occurred. That said, full-birth histories are not much more time-intensive in terms of field work than, say, a truncated history focusing on the last three births, but they allow much more flexibility in analysis and provide a much stronger basis for internal consistency checks to identify possible errors.
|
From page 8... ...
Failure to take account of age distributional factors can give rise to highly misleading comparisons. We obsess about exposure time in part because we obsess about age (and thus need to get exposure to risk in different age ranges correct)
|
From page 9... ...
In normal populations, the GFR is quite insensitive to typical variations in age distributions, but displaced populations are likely to be abnormal, and the GFR needs to be examined carefully. For comparisons between populations, indirect standardization using the full age distribution is preferable to the GFR, and if any age of mother detail at all is available for births, direct standardization (as described, for example, in Preston et al., 2001)
|
From page 10... ...
The monthly data on pregnancy outcomes in Matlab, collected by regular household visits by field workers, represent perhaps the only prospective data source concerning monthly miscarriages, stillbirths, and live births during a famine in the developing world. The pattern of miscarriages in comparison to births does not seem to provide strong evidence for changes in overall pregnancy loss.
|
From page 11... ...
Miscarriages, presumably including induced abortions in these data, fall well below expected values as early as March 1974 and remain low until September 1975. These data certainly give no support to the idea that miscarriages might have increased as a result of the famine, and it is not until September 1975 that they give any support to the idea that induced abortions might have increased (but the fact that miscarriages increased to "normal" levels by September 1975, 9 months before births recovered, may suggest some induced abortion in the later stages of the famine)
|
From page 12... ...
. However, the ratio of induced abortions to live births doubled during the war, from one abortion per pregnancy taken to term prior to the war to at least two during the war.
|
From page 13... ...
The paper states that more than 60 percent of the women in Sakaeo were under age 25, but fertility measures were not age-standardized because only two age intervals, 15-24 and 25-44, were available. Given the very different age distributions, even a two-age group standardization would have been helpful.
|
From page 14... ...
. Here, however, we are interested in the fertility effects of more catastrophic changes: famines, civil wars, and mass population displacements.
|
From page 15... ...
An indirect method was needed for the Beirut study because information on exposure time, specifically the numbers of women by age and religion, was not available. The method uses stable or quasi-stable population relationships to infer the underlying age distribution and derive an estimate of the total fertility rate (TFR)
|
From page 16... ...
effects, and it is limited to live births, so provides no indication of Process 3 effects on fertility. Checks on data quality showed no evidence of heaping on ages 0 and 5, and infant mortality rates calculated from the birth histories are close to other estimates for Ethiopia.
|
From page 17... ...
The data in this case are from a truncated birth history, collecting data on the dates of the three most recent births for each woman as well as information on fertility preferences. Also using discrete time logistic regression, birth probabilities were modeled on socioeconomic and biodemographic factors, areas of residence (categorized as the national capital area of Luanda, areas more affected by conflict, and areas less affected by conflict)
|
From page 18... ...
There is not much evidence of intentional fertility regulation on a scale large enough to affect population aggregates. In terms of data type, the authors state that they would have preferred complete birth histories but claim that recorded dates of birth and death for the three most recently born children are "generally realistic." It is not clear whether this is a statement about their own data set or about the developing world in general, but if it was meant to be the latter, it is far from correct and should not go unchallenged.
|
From page 19... ...
A REVIEW OF EVIDENCE 19 250 200 Month yb 150 ths Bir 100 50 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 Year FIGURE 3(a) Births reported in Rwanda DHS by month, 1988 to 1999.
|
From page 20... ...
is correct. Once again, the substantive conclusion is that the effect of displacement on fertility was not very great, and the methodological conclusion is that retrospective birth histories, even when complete histories, are uncertain sources of information about past fertility fluctuations, because of typical patterns of misdating of events.
|
From page 21... ...
In a regression analysis of camp CBR on a range of indicators -- age of camp, season, distance to border or conflict zone, time to a referral hospital, obstetric referral rate, percentage of births in the health facility, supplemental food for pregnant women, and staffing (traditional birth attendants, community health workers, and local health staff, separately) -- only the age of the camp (negatively for camps in existence 7 or more years)
|
From page 22... ...
Jews born in Europe and Christian Arab Israelis have had a replacement level of fertility, whereas Palestinians in the Gaza Strip have had a TFR of nearly eight. Jewish fertility has shown a pattern of convergence among different place of origin groups since 1948, whereas the fertility of Palestinians has diverged since 1970, with fertility remaining very high in Gaza, declining moderately in the West Bank, and declining rapidly (although not to Jewish levels)
|
From page 23... ...
There is little evidence about the fertility effects of return (although the Rwanda DHS data in Figure 3 do not suggest a major effect)
|
From page 24... ...
There is little evidence to support a strong effect on spontaneous abortion, and evidence concerning induced abortion is largely anecdotal. Where short-term effects appear to be large (such as in the Sakaeo refugee camp in Thailand in 1979-1980)
|
From page 25... ...
Sample surveys can provide useful fertility and reproductive health indicators from surprisingly small samples of no more than 3,000 households. A full-birth history is recommended: either time-limited or birthlimited truncation is likely to result in underestimates of recent fertility, as well as providing no basis for evaluating data quality.
|
From page 26... ...
1969 Disciplining demographic data. In Proceedings of the IUSSP International Population Conference, London.
|
From page 27... ...
Spiegel 2002 Reproductive health indicators and outcomes among refugee and internally displaced persons in postemergency phase camps. Journal of the American Medical Association 288(5)
|
From page 28... ...
Wulf, D., ed. 1994 Refugee Women and Reproductive Health Care: Reassessing Priorities.
|
From page 30... ...
30 97 1999 106 141 148 149 153 137 143 154 147 184 122 92 82 80 96 1998 118 111 110 128 152 130 116 116 1997 123 133 124 162 110 129 176 128 111 111 117 123 1996 165 126 120 150 156 171 173 136 129 120 126 116 83 72 69 97 1995 126 129 154 156 144 116 112 104 77 87 1994 113 157 138 232 151 164 194 159 129 116 Births of 97 97 76 87 1993 106 136 132 143 125 115 143 111 umber N 87 95 63 85 1992 130 122 107 155 129 136 135 159 . Year 92 98 67 97 85 69 61 64 1991 101 121 115 101 and 3 onth to 75 92 1 1990 102 115 133 138 133 133 127 123 125 119 M by Figures 84 88 90 92 93 69 62 64 Births 1989 100 113 109 116 2000: 95 89 84 52 65 nderlying 1988 100 142 101 115 123 108 105 DHS U Birth Rwanda umbers of y N y CE: onth arch ay ctober ovember ecember Raw M Januar Februar M April M June July August September O N D SOUR
|
From page 31... ...
31 Live Births 973 813 739 801 886 1,188 1,009 1,042 1,146 1,328 1,352 1,111 ths Still- bir 45 37 38 36 31 48 40 36 49 63 59 47 . iscarr 78 76 91 94 94 84 71 75 75 48 M 123 105 Live Births 614 577 638 501 668 752 898 1,013 1,281 1,367 1,487 1,375 ths Still- bir 26 22 27 29 29 29 36 39 64 57 55 56 .
|
From page 32... ...
; the measurement of child mortality (with particular emphasis on tracking national trends and linking them to other changes) ; the exploration of links between demographic parameters and economic crisis; the impact of policy and programs on demographic change; the role of gender preferences on child health behaviors and fertility; the demography of Sub-Saharan Africa; the role of development, particularly child mortality change, on fertility decline; and the measurement of demographic parameters for populations undergoing complex emergencies.
|
From page 33... ...
Research Ethics in Complex Humanitarian Emergencies: Summary of a Workshop (2002) Demographic Assessment Techniques in Complex Humanitarian Emergencies: Summary of a Workshop (2002)
|
Key Terms
This material may be derived from roughly machine-read images, and so is provided only to facilitate research.
More
information on Chapter Skim is available.