7 Family planning as a service

Some eighty-five countries in the developingworld, representing about 95 percent of its popula-tion, now provide some form of public support tofamily planning programs. Tremendous progresshas been made in improving couples' access toinformation and services. But in all countries morecould be done. Nearly all programs still fail toreach most rural people; even in the towns andcities the quality of services is often poor and dis-continuation rates high. In many countries thepotential of the private sector to provide familyplanning services has hardly been tapped; in oth-ers the gap in services provided privately can befilled only by enlarging public programs. Twenty-seven countries have yet to introduce family plan-fling programs. Almost half of these are in Africa,where incomes are the lowest in the world, popu-lation growth is the highest, and the potential ben-efits from family planning may be greatest.

The benefits of family planning, moreover,do not depend on the existence of demographicobjectives.

Family planning improves the health of moth-ers and children. Both infant and maternal mortal-ity in developing countries could be substantiallyreduced if pregnancies were spaced at least twoyears apart, and if pregnancies among teenagersand women over forty were prevented (see Box7.1). Couples with access to family planning serv-ices can prevent unwanted pregnancies that mightotherwise result in poorly performed abortionsand the risk of serious, even fatal, complications.Family planning services were recognized as oneof eight essential components of primary healthcare by the International Conference on PrimaryHealth Care in Alma-Ata in 1978.

Family planning makes responsible parent-hood easier. Parents can have the number of chil-dren for whom they know they can provide ade-quate food, health care, and education.

Family planning enlarges the choices availableto people, a central purpose of economic and socialdevelopment. This is particularly true for women,

who are often caught in a vicious circle in whichtoo many children mean too few opportunities forother kinds of activity, and vice versa. By enablingwomen to control their fertility, family planningfrees them to become better educated and toincrease their own and their children's contribu-tion to development.

Family planning offers the greatest potentialbenefits for the poorest people, whose mortalityand fertility rates are usually the highest of anygroup.

For all these reasons, programs to support familyplanning deserve a central role in the social andeconomic strategies of governments throughoutthe developing world. Properly designed, pro-grams need not be particularly expensive. But lackof finance is one of the reasons family planning is

TABLE 7.1

Percentage of currently married women aged 15 to49 using contraception, by region and for selectedcountries

Note: Numbers are based on recent surveys, except for Indiaand Indonesia, which are based on recent program statistics.

Average weighted by population for all countries in regionwith recent surveys.

Ever-married women aged 15 to 50.Source: World Development Indicators, Table 20.

127

Region and country Total Urban Rural

Sub-Saharan Africa (6) a

Ivory Coast (1980-81) 3 4 2Kenya (1977-78) b 7 12 6

Middle East and North Africa (22)

Egypt (1980) 24 40 12Syria (1978) 20 34 5

East Asia (65)

Philippines (1978) 36 47 31Thailand (1981) 57 64 55

Latin America and Caribbean (4)aColombia (1980) 49 54 37Mexico (1979) 39 51 27

South Asia (25)

Bangladesh (1983-84) 19 36 17Sri Lanka (1982) 55 57 54

Box 7.1 Family planning for health

Early and frequent childbearing contrib-utes substantially to illness and death ofinfants, young children, and mothers indeveloping countries. Family planningprograms can tackle these problemsthrough four main mechanisms:

Lengthening the interval betweenpregnancies (child spacing) The intervalbetween pregnancies is an importantdeterminant of survival for both the new-born baby and his or her older sibling.Infants and children at highest risk ofdeath are those born less than two yearsapart (see first chart). This relation holdseven when allowance is made for birthorder, mother's age, mother's education,urban or rural residence, and the sex ofthe child.

There are two main explanations forthe link between mortality and spacing.The first is that the youngest and nextyoungest child must compete for theresources of the family and for the atten-tion of the mother. When a womanbecomes pregnant again soon after giv-ing birth, the young child may be prema-turely weaned, increasing the risk thathe or she will suffer from malnutrition,gastrointestinal infection, diarrhea, andother illnesses, Second, a rapid succes-sion of pregnancy, breastfeeding, andthen another pregnancy weakens themother and is linked to low birth weight

Infant mortality per 1,000 live births, Peru

Spacing interval (years)

<2

2-3

4-6

7+

Mother's age

8]

90]

102

138

in the newborn baby. One study oftwenty-five developing countries sug-gested that, if births were spaced two tosix years apart, infant mortality woulddecline by an average of 10 percent, and

child mortality would decline by 16 per-cent. In Pakistan infant mortality (cur-rently 140 per thousand) would fall by 30percent if all birth intervals of less thanthirty-six months could be lengthened tothirty-six to forty-seven months.

Preventing births for women undertwenty and over thirty four years of age.In these age groups, women whobecome pregnant carry a grelter risk ofillness and death, both for themselvesand their children. Infant and maternalmortality are highest among teenagemothers. In Pakistan, for example,babies born to teenage mothers have a 50percent greater chance of dying than dothose whose mothers are aged twenty totwentvnine; in Peru the chance is 15percent greater. There were 860 maternaldeaths per 100,000 live births amongteenage mothers in Matlab Thana,Bangladesh in the mid-1970s, comparedwith 450 for women aged twenty totwenty-nine (see second chart). Part ofthe explanation for these contrasts is thatteenage mothers may not he physicallymature enough br a safe pregnancy; inaddition, most of their births are firstbirths, which often carry a higher risk ofinfant and maternal death. As for moth-ers over thirty-five years old, their babiesrun an increased risk of congenitaldefects such as Down's syndrome, cleft

being neglected in some countries and is makingonly slow progress in others. Aid donors have amajor contribution to make in ensuring that familyplanning programs receive the money they need tobe effective.

The use of contraception

Surveys of married women of reproductive age(fifteen to forty-nine) show wide variations in con-traceptive use among developing regions (seeTable 7.1). In East Asia nearly two-thirds of themarried women in that age group use contracep-tion; in China, Hong Kong, and Singapore theproportion is 70 percent or more, as high as in theUnited States and western Europe. Latin Americahas reached about 40 percent, whereas the propor-tion in the Middle East and South Asia is only

128

about 25 percent. Contraceptive use is lowest insub-Saharan Africa, at less than 10 percent of mar-ried women, and this estimate excludes manycountries in which use is negligible but data areunavailable.

Contraceptive use varies widely within countriesas well. In most, a higher proportion of urban thanrural couples use contraception; the distinction isparticularly stark in Syria, where 34 percent ofurban, but only 5 percent of rural, women wereusing contraception in 1978. In the Ivory Coast,Kenya, and Mexico, contraceptive use in ruralareas is roughly half the rate in urban areas, and inEgypt it is less than a third. Regional differencesare also great: in Indonesia, contraceptive useranged from 53 percent of couples on the islands ofJava and Bali to only 16 percent in some of theouter islands in 1983. In Maharashtra and Gujarat

2-3 87

4+ 55

Birth order

0 50 1011 150

Source Rutsten, 1982.

palate, and heart disorders. Infant andmaternal mortality also increase formothers in their thirties and forties.

Because most births are already towomen in the twenty to thirty-four agegroup, confining all births to that groupwould have only a modest effect on over-all infant and child mortality rates. Forexample, both rates would decline byonly 2 to 6 percent in Indonesia, Paki-stan, and the Philippines. The effect on

Maternal mortality per 100,000 live births,Matlab, Bangladesh

Birth order

2-3 254

4-5 456

558

8+

Mother's age

e 20

20-29 450

30-39 580

Chen and others, 1974.

741

860

maternal mortality is potentially greater.A study in the mid-1970s estimated thatmaternal mortality would be reduced by24 percent in the Philippines, by 23 per-cent in Colombia and France, and by 19percent in Mexico, Thailand, Venezuela,and the United States. The effect onmaternal and child illness has not beenestimated, but it would certainly begreater than on mortality.

Allowing couples to have fewer chil-dren. Depending on the country, therisks of infant and maternal mortalityincrease rapidly after the third, fourth, orfifth child. In El Salvador, for example,infant mortality for fifth and later chil-dren is more than twice the level for thesecond and third child. In Matlab Thana,Bangladesh, maternal mortality is about250 per 100,000 live births for the secondand third births but is about 450 perlO()U00 for the fourth and fifth births.These relations hold even when allow-ance is made for differences in the age ofmothers. If all births of fourth and laterchildren were prevented, infant mortal-ity would decline by between 5 and 11percent in Indonesia, Pakistan, the Phil-ippines, and Sri Lanka.

Preventing illness and death result-ing trom unsafe abortion. Abortion isextremely safe when performed in thefirst three months of pregnancy by

trained personnel in sanitary conditions.But in most developing countries theprocedure is illegal, and therefore morelikely to be self-induced or performedunhygienically by untrained people.Such abortions carry with them a highincidence of complications, such asincomplete abortion, pelvic hemorrhage,lacerations of the cervix, perforation ofthe uterus, and tetanus. These complica-tions may require hospitalization andmay damage the mother's fertility; in theworst cases, they can kill her.

Because abortion is illegal in manycountries, the number of women affectedis difficult to estimate. In twenty-fourcountries during 1970-78 complicationsof abortion were cited as a cause ofbetween 6 and 46 percent of all regis-tered, maternity-related deaths. Scat-tered evidence from Africa suggests thathospital admissions for complicationsafter induced abortion are increasing,and that a disproportionate number ofadmissions are teenagers. The Interna-tional Planned Parenthood Federation(IPPF) estimated in the late 1970s that84,000 women die annually from compli-cations of abortion in sixty five developing countries. Provision of safe, effective,and convenient contraception could pre-vent many unwanted pregnancies thatire aborted

states of India, 35 percent of couples were usingcontraception in 1981-82, compared with only 11percent of couples in the states of Uttar Pradeshand Jammu and Kashmir.

Among countries for which more than one sur-vey estimate is available, contraceptive use hasincreased fastest in East Asia and Latin America(see Figure 7.1). In Thailand, for example, the pro-portion of married women aged fifteen to forty-four using contraception rose from 15 percent in1970, the year the official family planning programwas launched, to 59 percent in 1981. Progress inSouth Asia has been slower, with contraceptiveuse increasing by about 1 percent of couples a yearin Nepal, more quickly in Bangladesh, but not atall in Pakistan. In Egypt and Kenya contraceptiveuse has remained unchanged, despite longstand-ing public programs.

These survey-based estimates may underesti-mate contraceptive use because they do notinclude use among unmarried men or women andsometimes exclude use among couples in informalunions. There may also be underreporting bysome women of the use of contraception by hus-bands, and some respondents may be reluctant toadmit to using contraception themselves. At thesame time, these figures may overstate the numberof people protected by contraception because notall couples using a method are equally protectedfrom the risk of pregnancy. Some are using "effi-cient" contraceptive methods such as sterilization,the pill, the IUD, injectable contraceptives, con-doms, spermicidal foam, and the diaphragm. Butothers are using less effective methods, such asdouche, rhythm, and withdrawal, or are abstain-ing (see Box 7.2). In Peru, 53 percent of those using

129

40-49 670

0 200 400 6t+

FIGURE 7.1

Trends in contraceptive prevalence in 1970-83,selected countries

Percentage of married women aged 15-49 using contraception

60

50

40

30

20

10

0

1970

Pakistan

Korea

Wife aged fifteen to forty-four.Ever-married women.

Sources; For Egypt, Kenya, and Pakistan, UN 1983; for others, CPS andWFS data.

Bangladesh

Thailand

Colombia

Mexico

Egypt

Nepal

1975 1980 1983

rhythm or withdrawal had an unwanted preg-nancy within three years after a birth, comparedwith only 29 percent of women who used the pill,the IUD, or injectable contraceptives. The 1978Philippines Fertility Survey found that 36 percentof married women of reproductive age used somemethod, but only 16 percent used an efficientmethod. In contrast, in the Dominican Republic in1975, contraceptive use was 32 percent for allmethods and 26 percent for efficient methods.

Contraception is not the only method of birthcontrol. Induced abortion is widespread, evenwhere it is illegal. There may be as many as 30million to 50 million induced abortions performedannually worldwide; this wide range is due touncertainty about the number of illegal abortions.Illegal abortion carries with it a high risk of compli-cations and death and can affect future fertility. Inmany developing countries abortion is illegalunder any circumstances or is permitted only tosave the life of the mother; China and India aremajor exceptions. Elsewhere legal abortion is animportant method of birth controlin Cuba,Japan, Korea, the USSR, and eastern Europe.Legal abortion rates per thousand women of child-bearing age range from 11 in Canada (1981) and 25in the United States (1980) to 84 in Japan (1975) and

130

88 in Romania (1979). Resort to both legal and ille-gal abortion often results from lack of informationabout, and access to, safe and effective contracep-tive methods.

Unmet need

In the surveys from which data on contraceptiveuse have been drawn women were also askedwhether they would like to have more children.Forty to 75 percent of married women of childbear-ing age in East and South Asian countries and inLatin American and Caribbean countries want nomore children. In a few countries women werealso asked whether they wished to delay their nextpregnancy for a year or more. Nineteen percent ofwomen of. childbearing age in Bangladesh andThailand, 25 percent in El Salvador, and 32 percentin Guatemala said yes. In countries where bothquestions have been asked, from 50 to 90 percentof women want either to limit or to space births.

In virtually all countries surveyed, the number ofwomen of childbearing age who want no morechildren exceeds the number using some kind ofcontraception. Some of the women who want nomore children or who wish to delay a pregnancyare not using a method because they are currentlypregnant or because they have been breastfeedingfor less than one year and therefore are affordedsome (but not total) protection. Others are unableto conceive, or their husbands are away. Thesewomen are not "exposed" to the risk of preg-nancy, so they do not need contraception, at leastnot immediately.

The remaining womenthose who would like tospace or to limit births, who are not using contra-ception, and who are exposed to the risk of preg-nancyare said to have "unmet need" for contra-ception. By this definition, 6 to 12 percent ofwomen of childbearing age in Egypt, Kenya, andthe Philippines have unmet need for contraceptionto limit births (see Figure 7.2, low estimate). InBangladesh, Korea, and Peru, where both limitingand spacing questions were asked, 16 to 33 percentof women of childbearing age have unmet need forcontraception. If women who are breastfeedingand those using inefficient methods of contracep-tion are also considered to have unmet need, morethan 40 percent of women in Bangladesh and Peruhave unmet need for limiting and spacing births;22 percent of women in Egypt, 10 percent inKenya, and 29 percent in the Philippines haveunmet need for contraception only to limit births(Figure 7.2, high estimate). Estimates for other

FIGURF 7.2

Contraceptive use and unmet need for contraception, selected countries, 1977-81

Unmet need forlimiting births

0Unmet need forlimiting and spacingbirths

The charts on this page show unmet need for contra-ception among married women aged 15 to 49. In thetop pie and bar chart, unmet need for contraception tolimit births is shown among women who want no morechildren. In the bottom pie and bar chart unmet need isshown for women who want no more children or whowant to postpone a birth. In each case, the low estimateof unmet need is the percentage of women using nocontraception. The high estimate is the percentageusing an inefficient method (withdrawal, rhythm, andthe like), breastfeeding (within a year of a birth), orusing no method.

Estimate ofunmet need

(percent)

High

Estimate of unmet need

Estimate ofunmet need

(percent)

0 10 20 30 40 50 60

Source; Boulier, 1984b.

131

Percentage of married womenaged 15-49 High Low

Egypt(1980) 22.4 12.3

Kenya(1977-78) 9.9 5.9

Philippines(1978) 11.129.0

Percentage of married wontenaged 15-49 -. High Low

Bangladesh(1979) 41.3 32.6

Korea(1979) 26.2 16.1

Peru(1981) 46.2 [6.3

10 20

Key

30 40

Efficientmethod

Inefficientmethod

Breastf ceding(first year)

Nomethod

Box 7.2 Birth planning technology

Several methods of birth control havebeen practiced throughout human his-toryabstinence, abortion, prolongedbreastfeeding, and coitus interruptus(withdrawal)but with uncertain effec-tiveness, and psychological and healthdamage. Contraceptive research in thepast thirty years has made possible amuch greater variety of more effectivemethods. Combined estrogen and pro-gestin oral contraceptives (the 'pill")and various intrauterine devices (HiD)were the first major breakthroughs inthe late 1950s and early 1960s. Since thenother methods have been developed:injectable contraceptives effective for twoto three months; more effective copperand hormone-releasing IUDs; menstrualregulation (vacuum aspiration of theuterus within seven to fourteen days of amissed period); male sterilization; sim-plified female sterilization by laparos-copy and minilaparotomy; low-estrogenpills with fewer side effects; and a pro-gestin-only "minipill." Barrier methods,such as the condom, diaphragm, andspermicides, have also been improved.

In 1980 the most commonly used meth-ods of birth control worldwide were ster-ilization and the pill. Among developedcountries the pill is the most usedmethod, but sterilization has gained inpopularity in the United States and inGreat Britain, where it accounts for abouta quarter of total use among married cou-

pIes of childbearing age. The majorexceptions to this pattern are Spain,Italy, and the Eastern European countries(except Hungary), where withdrawal,rhythm, or abstinence are still the mostprevalent methods.

Among developing countries, steriliza-tion is the most common modernmethod in Bangladesh, El Salvador,India, Korea, Nepal, Pakistan, Panama,Sri Lanka, Thailand, and Tunisia. Thepill is the most favored method in Egypt,Jordan, Syria, much of Latin America,Malaysia, and Indonesia. Injectable con-traceptives are widely used in Jamaica(11 percent of eligible women), Thailand(7 percent), Trinidad and Tobago (5 per-cent), and Mexico (3 percent); thismethod is convenient to use for ruralwomen and, unlike the pill, does notinterfere with lactation. Both the WorldHealth Organization and IPPF haveapproved injectableslegal in more than100 countriesbut greater use in devel-oping countries is partly constrained bythe method's limited availability. TheUnited States, the major contraceptivedonor worldwide, cannot donate inject-ables because US assistance policy pro-hibits supply of drugs not approved fordomestic use.

Despite the greater variety of contra-ceptive methods now available, all haveshortcomings:

Effectiveness. Under the ideal condi-

tions of controlled studies in developedcountries, existing methods can behighly effective in preventing pregnancy:nearly 100 percent for sterilization, thepill, and injectables; 98 percent for theIUD; and as much as 97 percent for thecondom and the diaphragm after oneyear of use. But outside these controlledstudies, some methods can be signifi-cantly less effective owing to incorrect orinconsistent use. In the United States,one in 100 couples using the pill will havea pregnancy within one year, more thantwo couples using the IUD, twelve usingthe condom or diaphragm, and twentyusing rhythm. In the Philippines morethan three women out of 100 using eitherthe IUD or the pill and thirty-three usingrhythm will become pregnant within ayear. The motivation of couples to pre-vent pregnancy is important in the effec-tiveness of contraceptives. Couples whowant no more children are likely to usemethods more effectively than thosewho are spacing births.

Side effects. Physical side effects are amain reason that people switch, or stopusing, contraceptives. For some meth-ods, the long-term health risks of pro-longed use are unknown. Methods suchas the IUD and injectables, which alterbleeding patternsby spotting betweenperiods, increased or decreased flow, oramenorrheamay be culturally unac-ceptable or restrict the activities of users.

Box 7.3 Measuring unmet need for family planning

The concept of "unmet need" used inthis Report is based on two questionsasked of married women in representa-tive nationwide surveys during the pastdecade. In more than forty countrieswomen were asked, "Do you want addi-tional children?" Among women whowere exposed to the risk of pregnancy(that is, they were neither pregnant norinfertile), some said that they did notwant more children. Of them, those whowere not using any contraceptivemethod were defined as having unmetneed for limiting births. In some surveys(in fewer countries), women were alsoasked, "Do you wish to delay pregnancyfor a year or more?" Among women at

risk of pregnancy, some said yes. Ofthem, those who were not using anycontraceptive method were defined ashaving unmet need for spacing births.

Some investigators have suggestedthat responses to such questions aremeaningless or, at best, unreliable. Theyargue that many women in developingcountries are not accustomed to planningtheir families or are uninformed abouthow to affect the number of births theywill eventually have. These criticismsapply most strongly to questions on pre-ferred family size ("Suppose you wererecently married and were able to havejust the number of children you wanted,how many would that be?") and desired

family size ("If you could choose exactlythe number of children you have in yourlife, how many would that be?") Thesequestions contain significant hypotheti-cal components, since women cannotcostlessly choose family size, cannothave fewer children than they alreadyhave, and must imagine alternative lifecycles involving different family sizes.Responses to such questions are not con-sistent even when women are asked thesame question at different dates. Forexample, in Indonesia only 46 percent ofwomen reinterviewed four months afteran initial survey gave an identicalresponse to a question on desired familysize. In a similar study elsewhere, only

132

Worldwide expenditure on reproductiveresearch and contraceptive development,1965-79

Millions of dollars160

140

120 Current dollars!

100

80

60

40

20

0

1965 69 71 73 75 77 79

Source Atkinson and others, 1980.

Constant 1970 dollars

Inconvenience. Barrier methods (con-dom, diaphragm, spermicides) have tobe used each time couples have inter-course. In households in developingcountries, pills and diaphragms are diffi-cult to store and condoms difficult to dis-pose of.

Reversibility. Sterilization is highlyeffective but rarely reversible. Injectables

are completely reversible but delay thereturn to fertility for several months.

Acceptability. To some couples, abor-tion and sterilization are religiously orculturally unacceptable; some mayregard only abstinence or rhythm asacceptable.

Delivery. Sterilization (of both menand women) requires skilled medical orparamedical staff, who are often scarce indeveloping countries. The IUD, inject-ables, and the pill require medicalbackup for treatment of complicationsand side effects. Programs which pro-mote the condom, pill, and spermicidalfoam require a good network of supplypoints.

No single method of contraception isappropriate to the needs of all people noris there one that is completely safe,reversible, effective, and convenient.Nor is such an "ideal" method likely tobe developed in the next twenty years.Family planning programs will have torely on a mix of existing methods and afew new ones whose development isalready well advanced.

Research is being concentrated in twoareas: improving the safety, conven-ience, and life span of existing methods,such as the IUD, pill, injectables, andfemale barrier methods; and developingnew methods, such as a monthly pill toinduce menstruation, long-lasting biode-gradable hormonal implants for women,

nonsurgical chemical sterilization formen and women, a male "pill," and anantipregnancy vaccine for women. Someof these new methodssuch as the hor-monal implant (in the arm), improvedIUDs, the vaginal sponge, cervical cap,and diaphragms which release spermi-cidemay be widely available in the nearfuture. Others, such as new male meth-ods and an antipregnancy vaccine, re-quire much more research and areunlikely to be marketed before the end ofthis century.

Compared with the past few decades,the pace of technological development isslowing. Worldwide funding for contra-ception-related research was $155 millionin 1979, but hcs been declining in realterms since 1972-73 (see chart). About 30percent of the total is spent on contracep-tive development and safety studies; therest goes to training and basic researchon human reproduction. Some 72 per-cent of the total was spent in the UnitedStates. Over 80 percent of the total wasfinanced by the public sector; industry'sshare has shrunk from 32 percent in 1965to less than a tenth. Special testing andregulatory requirements, combined withproduct-liability problems, have length-ened the time between product develop-ment and marketing, increased the costof developing new products, and madethe future profitability of research moreuncertain for private firms.

67 percent of those reinterviewed afterone month gave an identical response toa question on preferred family size.

In contrast, women respond consis-tently over time to a question on whetheror not they want additional children, andtheir replies are reasonably good predic-tors both of contraceptive use and offuture fertility. (The Population DataSupplement gives country-level informa-tion on responses to this question.) Inthe study in which two-thirds ofresponses to a question on preferredfamily size were inconsistent, 90 percentof women were consistent in answeringa question about wanting additional chil-dren. In countries surveyed as part of the

World Fertility Survey, contraceptive useamong women who said they wanted nomore children consistently exceeded useamong women wanting more children.

Obviously, not all women who wantno more children use contraception, andnot all women who want to limit theirfamily size cease bearing children. Sev-eral factors may account for these dis-crepancies between attitudes and behav-ior. First, some women may not considerthe costs of birth control in answeringquestions on family size. If the benefitsof avoiding births are small in relation tothe costs of contraception, women whowant no more children have little motiva-tion to use contraception. Second, the

surveys inquire only about women's, notmen's, attitudes. Where both spouseshave been interviewed, the differencebetween them tends to be small, butthere are differences. Third, growingexperience with children, plus unantici-pated eventsa child dying, illness ofhusband or wifemay lead couples toalter their plans. Fourth, even moderncontraceptive methods can fail, so thatsome women will have more babiesdespite their intention not to increasefamily size.

133

TABLE 7.2

Percentage of married women aged 15 to 49 practicing efficient contraception among those who want nomore children

134

countries are shown in the Population Data Sup-plement, Table 3.

These high and low calculations of unmet needprovide rough estimates, given existing prefer-ences for family size, of the potential for additionalcontraceptive use. Some analysts, however, havequestioned the validity of estimates based on theresponses of married women to survey questions(see Box 7.3). Others have noted that even womenwho are pregnant may have had unmet need inthe past that resulted in an unplanned pregnancy,and that such women may shortly be in needagain. Nor do these surveys include unmet needamong unmarried people. Clearly, use of contra-ception depends not only on accessibility and cost,but also on how intensely a couple wishes to avoida birth. This factor is difficult to measure in sur-veys. Whether unmet need can ever be completelysatisfied is debatable. But in the United States,where contraception is widely available, unmetneed for limiting births was estimated at only 4 to 8

Note: Efficient contraception includes male and female sterilization, pills, IUD, injectables, diaphragm, and condoms. Women who arepregnant or infecurmd are excluded from this table.

Not available.a. Family planning index is interpolated from 1972 and 1982 data to year shown. See notes for Table 6 of the Population DataSupplement.Sources: CPS and WFS data; Lapham and Mauldin, 1984.

percent of married women of childbearing age in1976.

The concept of unmet need is not static. Unmetneed may decline as more people have access tocontraception or as the nature of services changes.It may increase as people want fewer children, oras the better availability of services raises interestin regulating fertility faster than new services canmeet new need. Many women who say they wantmore children might be potential users of servicesif given the chance to plan their births. To someextent family planning programs do more thansimply satisfy unmet need; they actually generateand then fill such need. In this sense "demand"for contraceptive services is not easily measured; itis partly a function of their supply.

In most countries women in rural areas and withless education are less likely to want to stop child-bearing than are urban and more educatedwomen. But of the former, those who do want tostop are less likely than their urban and educated

Country and familyplanning index"

Residence and education

Urban Rural

Noeducation

Seven years'educationor more

Noeducation

Seven years'education

or more

Strong or very strongKorea, Rep. of, 1979 58 62 61 63Colombia, 1980 35 69 30 52

ModerateMalaysia, 1974 37 27 49Thailand, 1981 72 83 53 58

Philippines, 1978 22 45 15 38

Tunisia, 1979 50 60 45Bangladesh, 1979 21 55 16 37Mexico, 1978 40 71 17 53

Weak or very weakNepal, 1981 40 71 15 45

Egypt, Arab Rep., 1980 53 72 24 70Ecuador, 1979 17 60 6 58Pakistan, 1975 17 35 6 17

Venezuela, 1976 53 67 18 54

Kenya, 1977-78 13 44 12 32Honduras, 1981 53 58 15 49Ghana, 1979-80 11 30 8 25

counterparts to be practicing contraception. Gov-ernment plays a central role in narrowing thesegaps, especially between urban and rural areas(see Table 7.2). In Colombia and Korea, whichhave strong family planning programs, ruralwomen who want no more children are as likely asurban women to be practicing contraception. InKenya, Nepal, and Pakistan, which have weakerprograms, the contrast between rural and urbanareas is much greater.

Reasons for not using contraception

Couples who wish to plan their families face cer-tain costsfinancial, psychological, medical, andtime-related costs. If these exceed the net costs ofadditional children, couples will not regulate theirfertility, even if, ideally, they would prefer to post-pone or to prevent a pregnancy. To individuals,the costs of contraception include:

Informationthe effort to find out where con-traceptive methods can be obtained and how theyare properly used. In Kenya 58 percent of marriedwomen aged fifteen to forty-nine who are exposedto the risk of pregnancy do not know where theycan obtain a modern method of contraception; inMexico the figure is 47 percent.

Travel and waiting timethe money and timeneeded to go to and from a shop or clinic and toobtain family planning services. Average waitingtimes are as high as three hours in hospitals andfamily planning clinics in El Salvador. Family plan-ning programs in Bangladesh, India, and SriLanka compensate sterilization clients for theirtransport costs and lost wages.

Purchasethe financial cost of either contra-ceptive supplies (condoms, pills, injections) orservices (sterilization, IUD insertion and periodiccheckups, menstrual regulation, and abortion).Most public family planning programs providesupplies and services free of charge or at highlysubsidized rates. Purchase costs from private sup-pliers and practitioners may be substantiallyhigher.

Side effects and health risksthe unpleasantand sometimes medically serious symptoms thatsome women experience while practicing contra-ception. Users of the pill may gain weight or feelill. The IUD may cause excessive menstrual bleed-ing, persistent spotting, and painful cramps. Inaddition, in some countries women are forbiddenfor religious or cultural reasons from cooking dur-ing their menstrual periods; spotting and heaviermenstrual flow caused by the IUD can further

restrict their activities. Some methods increase therisk of developing serious health problems; higherrisk of pelvic inflammatory disease among IUDusers and of cardiovascular disease among users ofthe pill have been reported. (These risks, however,are small compared with those associated withpregnancy and childbirth.)

Social disapprovalthe private nature of fam-ily planning and the difficulty of discussing it withproviders of services or even with spouses. Familyplanning may violate personal beliefs, create mari-tal disharmony, or be socially, culturally, or reli-giously unacceptable.

Surveys of contraceptive use in ten countriesasked married women not practicing contraceptionwhy they were not doing so. Unless they wantedanother child or were pregnant, their reasonsincluded lack of knowledge of a source or methodof contraception, medical side effects of methods,religious beliefs, opposition from husbands, andfinancial costs. In Nepal lack of knowledge of asource was the main reason. In Honduras, Mexico,and Thailand half of the women who did not prac-tice contraception but were exposed to the risk ofpregnancy either knew of no source of contracep-tion or feared side effects. In Bangladesh, Barba-dos, and Nepal as much as a quarter to a third ofall married women were not using contraceptionfor these reasons. Contraceptive prevalence clearlycould be increased by better information and serv-icesdirected to men as well as to women.

Discontinuation rates tell a similar story. Accord-ing to surveys in thirty-three countries, as many as30 percent of married women of childbearing agehave used contraception in the past but are nolonger doing so (see Table 7.3). When contracep-tion is being used to space births, some discontinu-ation is normal. But many who discontinue contra-ceptive use do not want more children. As thesecond column of Table 7.3 shows, as many as 10percent of all married women are discontinuerswho want no more children and are at risk of get-ting pregnant. In Barbados, Guyana, Jamaica,Korea, and Pakistan, the proportion exceeds one-third (column 3). Follow-up surveys of womenwho have accepted contraception typically findthat much discontinuaton is due to medical sideeffects. In a follow-up survey in the Philippines,for example, this reason was cited by 66 percent ofthose who stopped using the pill and 43 percent ofthose who stopped using the IUD. Reducing dis-continuation among women who want no morechildren could increase contraceptive use by atleast one-fifth in eight countries (column 4).

135

TABLE 7.3

Discontinuation of contraception, recent surveys

Supplying family planning services

Family planning programs have evolved in variousways, but a typical pattern begins with servicesbeing provided only by private family planningassociations and a few concerned doctors andnurses. These groups gradually show that familyplanning is feasible and acceptable and start press-

136

Note: Figures in columns 1 and 2 were rounded after computing columns 3 and 4.(.) Less than half of 1 percent.a. Not pregnant or infecund.Source: Ainsworth, 1984.

ing for government support. Once persuaded,governments typically provide family planningthrough the public health system. But becausehealth care is often underfinanced and concen-trated in urban areas, and because family planningcompetes with other medical priorities, the qualityof services is uneven and available to only a smallproportion of people. Eventually programs are

Percentage of married women aged 15 to 49 Percentage of all Discontinuers whodiscontinuers who are are exposed and want

exposed and want no more children (2),no more children as a percentage(2 divided by 1) of current users

Used contraceptionbut are not

current users("discontinuers")

Discontinued use,exposed' andwant no more

childrenCountry (1) (2) (3) (4)

Sub-Saharan AfricaCameroon (1978) 6 (.) 1 2

Ghana (1979-80) 30 2 7 23

Kenya (1977-78) 25 2 7 24

Lesotho (1977) 18 2 10 33

Sudan (1979) 8 I 8 15

Middle East and North AfricaEgypt (1980) 17 4 25 18

Jordan (1976) 22 3 12 10

Syria (1978) 14 1 9 6

Tunisia (1978) 15 3 18 9

South AsiaBangladesh (1979) 9 3 30 21

Nepal (1981) 2 (.) 8 2

Pakistan (1975) 5 2 43 42

Sri Lanka (1975) 14 4 29 12

East AsiaIndonesia (1976) 12 2 16 7

Korea, Rep. of (1979) 24 8 33 16

Philippines (1978) 23 4 19 12

Thailand (1981) 21 5 26 9

Latin America and CaribbeanBarbados (1981) 28 10 36 21

Colombia (1980) 20 4 22 9

Costa Rica (1980) 23 3 14 5

Dominican Republic (1975) 18 3 17 8

Ecuador (1979) 20 4 20 12

Guyana (1975) 22 8 34 22

Haiti (1977) 17 3 15 14

Honduras (1981) 15 2 10 6

Jamaica (1975-76) 26 9 36 24

Mexico (1978) 15 3 20 8

Panama (1976) 21 4 17 7

Peru (1981) 20 3 14 7

Paraguay (1979) 21 3 12 7

Trinidad and Tobago (1977) 27 8 28 14

Venezuela (1977) 20 4 19 8

extended to the countryside, often by paramedicaland semiskilled staff with backup support fromhealth centers. More attention is paid to increasingthe range of contraceptive methods, providing fol-low-up services to clients, and working with com-munity leaders to encourage local support. Com-mercial organizations are also encouraged toprovide family planning. Private associations aredelegated major responsibilities within thenational program for certain services or targetgroups and continue to test new ways of providingservices.

Public family planning programs are now at dif-ferent stages of development in different regions.

East Asia. Governments have a longstandingcommitment to reduce population growth. Theyhave been extremely successful in improvingaccess to family planning services and in wideningthe range of contraceptive methods available.Large numbers of field workers have beenrecruited to provide family planning, and some-times basic health care, in villages in China, Indo-nesia, and Thailand. Contraceptive use hasincreased dramatically during the past decade.

South Asia. Official commitment to reduce fer-tility is strong, but results have been mixed. Con-traceptive use is highest in Sri Lanka and severalstates in southern India, and is lowest in Nepaland Pakistan. The demand for contraception is stillconstrained by high infant mortality and by a pref-erence for large families. At the same time, recentsurveys have revealed substantial unmet need forboth limiting and spacing births. Most programshave yet to achieve the rural spread found in EastAsia and have tended to emphasize sterilization.Other methods have been largely suppliedthrough subsidized commercial outlets.

Latin America and the Caribbean. At first, wide-spread demand for family planning was metlargely by private doctors, pharmacies, and non-profit organizations, primarily in urban areas.Government support was weak, in part because ofopposition from some religious authorities. The1970s saw a growing interest on the part of govern-ments and a greater tolerance by religious authori-ties. Most governments now support family plan-fling services for health and humanitarianpurposes; Barbados, Colombia, the DominicanRepublic, El Salvador, Guatemala, Haiti, Jamaica,Mexico, and Trinidad and Tobago do so to reducefertility as well. In rural areas, access to services isstill inadequate in most countries.

Middle East and North Africa. Some countriesin North AfricaEgypt, Morocco, and Tunisia, for

examplehave long-established programs toreduce fertility. About half the countries in theMiddle East provide family planning to improvechild spacing and to promote health; only Turkey'sprogram seeks to reduce fertility. In a few MiddleEastern countries, contraception is illegal. In oth-ers, cultural practices often confine women to theirhouseholds, which makes it difficult for them toseek out family planning services. Programs thatinclude home visits by family planning workersare not well developed.

Sub-Saha ran Africa. Of forty-one governmentsfor which data are available, only nine have demo-graphic objectives. Most governments that sup-port family planning do so for health reasons, andtwelve countries still provide no official backing forfamily planning. Where services exist, they areprovided through health care systems that haveonly limited coverage, particularly in rural areas.Throughout Africa couples want large families,and infant mortality is high. There is somedemand for family planning but it is poorly met byexisting programs. As traditional ways of childspacing (prolonged breastfeeding and sexual absti-nence) erode, the demand for modern contracep-tion increases. Private organizations have helpedto demonstrate that demand and to press for gov-ernment support.

The management of family planning programs

Perhaps more than any other social programs,family planning programs can be effective only tothe extent that they meet the needs of individuals,both for better information about the benefits ofcontrolling fertility and for better services to facili-tate doing so. At the same time family planningprograms, like all public programs, operate withincertain constraints: the availability of manpowerand finance, the capacity for training and supervi-sion, and the transport and communications infra-structure. Medical backup is necessary to deliversome contraceptive methods. The challenge forfamily planning managers is to address individualneeds within the confines of these constraints, andin the longer term to ease such limitations.

The personal nature of family planning serviceshas several important implications for designingand managing programs. First, programs must beable to accommodate local and individual needsand a variety of users. Potential clients includemen and women; those who are married and un-married; those of different social, economic, cul-tural, or religious backgrounds; and those who

137

may be delaying a first pregnancy, spacingbetween children, or preventing additional preg-nancies. Staff must be discreet, sensitive to theindividual needs of clients, and familiar with localcustoms and beliefs. This requirement has beenaddressed in several ways: by selecting staff fromlocal communities, by training staff in the environ-ment in which they will work, and by making spe-cial efforts to hire female workers. Special serviceshave also been targeted for specific client groups:adolescents, women who have just given birth,and mothers with young children.

Second, programs must encourage clients notonly to accept a method of contraception but alsoto use it effectively and continuously. In societiesin which people marry young, couples who arespacing and limiting births may have to use contra-ception for twenty years. Prolonged, effective useis easier if information and support regarding side

effects are assured, resupply is convenient, andthe opportunity to switch methods is available.Medical backup and referral is critical, as is thecapacity to follow up on clients. Managers needinformation not only on new acceptors but also oncontinuing users, dropouts, and nonparticipants(see Box 7.4). Indonesia is one country with aneffective monitoring system, including acceptorrecords, quarterly follow-up surveys of acceptors,and periodic sample surveys of households inwhich information on fertility and contraceptiveuse is collected.

Third, because information about the benefits offamily planning and of small families may not bewidespread, programs must create an awarenessof services and their benefits, as well as spreadinformation about the proper use of methods.Information and education activities are necessaryboth within and outside the system for delivering

Box 7.4 Management information systems for improved service delivery

The arrangements for providing familyplanning services in many countries areplagued by lack of reliable informationon which to base management decisions.Requirements for data collection areimposed on overburdened staff andsupervised by medical or other technicalpersonnel untrained to make use of theinformation. Much time is spent collect-ing information that is never used.

A management information system(MIS) is any system which organizes thecollection and interpretation of dataneeded by managers to make decisions.The rural health supervisor reviewing aworkers records to assess performance,and the health minister reviewing infor-mation on hiring and deployment of staffare both using an MIS. For a family plan-ning program, an MIS could includeinformation on target group size andcharacteristics, new and continuingacceptor rates and characteristics, num-bers and types of follow-up visits, birthrates, staffing patterns, and availabilityof supplies. These data allow managersto make decisions based on up-to-dateand reliable information that is collectedas a matter of routine.

Studies in two states in India, Karna-taka and Uttar Pradesh, in the mid-1970sshowed that fieldworkers providing

health and family planning services werespending as much as 60 percent of theirtime on activities not directly related todelivering their services. Keeping rec-ords and attending meetings were themost common extraneous activities. Atotal of forty-six registers were main-tained by five types of fieldworkers,relating to a range of subjects (familyplanning, maternal and child health,immunization, malaria control) and withconsiderable overlap of the data theyrecorded. An assistant nurse-midwifealone maintained twenty-two recordsand prepared twelve reports a month.The information was not used by super-visors and managers, nor did workersreceive any systematic feedback on theirperformance compared with others.There was little incentive to maintaingood records and to report regularly andon time.

Following a review of the system,recordkeeping and reporting werestreamlined. The number of registerskept by fieldworkers was reduced fromforty-six to six: a register of eligible cou-ples and children, a maternal and childhealth register, a report on blood smearsfor malaria, a birth and death register, astock and issue register, and a diary ofdaily activities. The various separate

reports forwarded to program managerswere replaced by a single monthly reportby each fieldworker, a single report byeach supervisor, and a single report fromeach primary health center. Family plan-fling staff were told immediately howthey were measuring up to predeter-mined targets. To encourage competi-tiofl, feedback reports from the district tothe primary health centers also rankedcenters on the basis of ten indicators,such as the number of immunizationsand the number of sterilizations as a per-centage of annual targets.

In three districts in the state of AndhraPradesh where this system was intro-duced, the time spent on recordkeepingand reporting has been reduced consid-erably. An assistant nurse-midwife, forexample, now spends only about half anhour a day with the new system com-pared with two hours before. Reports arecomplete and are submitted on time (inother districts reporting is about threemonths behind schedule), and managersare responding better to local needs.Steps to expand the system for statewideuse are now being taken in AndhraPradesh, and the government of India isrecommending that all states adopt thenew MIS.

138

services. Program staff recruit potential clients andoffer information on proper use of methods. Themass media can be used to inform people of thebenefits of small families and how to obtain contra-ceptive methods. Instruction on human reproduc-tion, family planning, responsible parenthood,and problems of rapid population growth as partof school curricula can inform young people beforethey marry; such instruction can also be offeredthrough nonformal education, such as adult liter-acy programs. These efforts complement othereconomic and social policies, discussed in Chapter6, to create demand for smaller families.

Because of the need for medical services for pro-vision or follow-up of many contraceptive meth-ods, most family planning programs are linked tothe public health system. The nature of these linksvaries among countries and has often changed. Insome programs, family planning workers provideservices through clinics administered by the minis-try of health, but are responsible to some otherbody. In Pakistan primary responsibility for familyplanning lies with the Population Welfare Divisionof the Ministry of Planning and Development,using the division's own specialized facilities andworkers. Elsewhere family planning is directlyadministered by the ministry of health, through aspecial department of family planning (as inEgypt) or as part of preventive or maternal andchild health services (as in Botswana, Kenya, andMalawi). Staff may specialize in family planning(that is, as "single-purpose" workers), as in Kenya,Pakistan, and Indonesia; or they may be responsi-ble for general health or maternal and child healthservices in addition to family planning (that is, as"multipurpose" workers), as in Bangladesh, Bot-swana, and India.

There have been obvious advantages in integrat-ing health and family planning in the delivery ofservices. The health benefits for mothers and chil-dren of spacing and limiting births clearly establishfamily planning as a valuable component of mater-nal and child health services. For both services themain target groupmarried women of childbear-ing ageis the same. Joint delivery can reduce unitcosts, and in countries where family planning iscontroversial, integrated services make the pro-gram more acceptable.

But integrated services also present difficulties.Health ministries are often understaffed andunderfunded; they cannot always mobilize thepolitical and administrative wherewithal to imple-ment an effective family planning program. Heavydemands for health care may eclipse the provision

of family planning services, and medical staff maygive priority to curative rather than preventiveservices. Multipurpose workers who are over-loaded with responsibilities will do none of theirtasks well. If an integrated delivery systememploys single-purpose workers, friction mayarise over differences in training, seniority, sala-ries, and promotion. For example, in addition totheir salaries, family planning workers have some-times received incentive payments based on thenumber of acceptors they recruit, whereas healthworkers receive only salaries. In Kenya familyhealth field educators (with family planningresponsibilities) were paid more than the enrolledcommunity nurses to whom they were to report.These personnel issues can seriously affect workermorale and performance.

Although family planning programs need somelink with health systems, family planning servicesneed not be confined to them. When services areprovided through a maternal and child health pro-gram, important client groups may be overlooked:men, adolescents, unmarried women, and non-pregnant women. Ministries of health may bepoorly equipped to organize social marketingschemes (for subsidized commercial distribution ofcontraceptives, discussed below), to develop massmedia programs, or to coordinate public, private,nongovernmental, and commercial activities.Some of these responsibilities are often delegated,for example, to information or education minis-tries. Many programs have boards within or out-side a ministry to coordinate the wide range offamily planning activities. In Mexico the semi-autonomous Coordinacion General del ProgramaNacional de Planificacion Familiar monitors andcoordinates all family planning activities; it islocated within the Ministry of Health but has directaccess to the president and works closely with theNational Population Council (CONAPO), a sepa-rate body responsible for population policy. InIndonesia the National Family Planning Coordi-nating Board (BKKBN) is an autonomous bodythat collects data, produces information and edu-cation programs, coordinates activities, and has itsown fieldworkers who promote family planning,refer clients, and set up community distributionpoints. In some countries these family planningboards are also responsible for overall populationpolicya role discussed more fully in Chapter 8.

In conclusion, there is no simple formula for thebest organization of family planning programs.Programs that differ widely in structure can beequally successful. Workers in India deliver both

139

family planning and maternal and child healthservices and are under the general guidance of theDivision of Family Welfare within the central Min-istry of Health and Family Welfare. Indonesia pro-vides family planning as part of maternal and childhealth services within the health system, but alsouses single-purpose fieldworkers responsible tothe BKKBN. The Chinese program relies on jointpersonnel in the health system but has a separatepolicymaking body for family planning and overallpopulation policy. No matter how service deliveryis organized, all programs need some healthbackup.

Other significant factors in the success of pro-grams are the degree of political commitment andthe overall administrative capacity of governmentto coordinate the deployment, training, supervi-sion, and availability of staff. These influence theeffectiveness of three program strategies forexpanding contraceptive use: increasing access toservices, improving service quality, and ensuringsocial acceptability.

Increasing access

Perhaps the greatest achievement of family plan-fling programs in the past decade has been tomake information and services more accessible tothose who need them. In twenty-three of twenty-nine developing countries in which surveys havetaken place, more than 80 percent of marriedwomen are aware of at least one effective methodof contraception. In urban areas of almost all ofthirty-six countries examined by the World FertilitySurvey (WFS), family planning methods are avail-able within an hour's travel from home. In CostaRica and Thailand most people in rural areas arealso less than an hour away from services. Further-more, most public programs provide services freeof charge or at heavily subsidized rates.

But there are still many countries and areas inwhich information and travel costs are majorobstacles to satisfying the unmet need of clients.According to household surveys in Guatemala andPiaui State, Brazil, 15 percent of married women ofchildbearing age said that they would like to usecontraceptives but did not know where to getthem. In Nepal half of married women do notknow of a method of contraception; about 15 per-cent know of a method but not of an outlet. InHonduras about a quarter of women are unawareof either method or outlet. Of those women inrural areas who know where to obtain contracep-tives, 32 percent in Colombia, 42 percent in Hon-

140

duras, and 62 percent in Nepal live more than anhour away from the source of supply. These barri-erslack of information and distanceare particu-larly high in sub-Saharan Africa: more than halfthe eligible women in Senegal and Sudan areunaware of modern contraceptive methods, and inmost African countries contraceptives are availableonly in urban areas.

To reach the rural areas, family planning pro-grams have placed special emphasis on extendingthe work of health centers into communities andhouseholds through the use of fieldworkers andother outreach staff. Access has also been in-creased in many countries by encouraging the pri-vate sector to provide family planning services.

EXTENDING PUBLIC SERVICES THROUGH OUT-

REACH." Until a decade ago almost all public familyplanning programs provided services from cen-tersusually clinicsand relied heavily on medi-cal staff. Because health services were not wellestablished in rural areas and medical staff werescarce, access to family planning as well as to med-ical care was limited.

Today many large family planning programshave succeeded in using their health centers as aspringboard for taking services and supplies intothe villages.

Paramedical workers have been trained toprovide many methods formerly provided only byphysicians. In Thailand, for example, auxiliarymidwives insert IUDs and administer injectablecontraceptives. Elsewhere nonmedical workersdistribute the pill; they receive careful training onscreening for contraindications, proper use, howto deal with side effects, and referral procedures.

Staff based in clinics have been supplementedwith fieldworkers who provide a link betweenthe clinic and the community (see Box 7.5).Fieldworkers periodically visit homes and outlyingcommunities to refer clients to service outlets; todistribute nonclinical methods such as the pill,condom, and spermicidal foam; and to reassureusers. In some cases fieldworkers also superviselocal volunteers.

Official outlets have been increased by orga-nizing local supply depots for nonclinical meth-ods. Such local outlets in Mexico and Indonesiaassist the work of field staff and reduce costs toclients.

The advantages of outreach are considerable:fieldworkers take less time and money to trainthan do medical professionals; health staff canspend more time on health care than they other-

Box 7.5 Family planning fieldworkers

Outreach systems using fieldworkershave been a key to success in effectivenational family planning programsovercoming the relative inaccessibility ofphysicians and lowering the costs of con-traceptive use by bringing servicesdirectly to beneficiaries. Experience indifferent countries illustrates a diversityof approaches to the training, duties, andcoverage of fieldworkers.

india. Family planning services aredelivered by male and female multipur-pose workers. Female workers providepre- and post-natal services to mothers,spread family planning information, dis-tribute condoms, and deliver babies. Thegovernment has recently sanctioned thedistribution of oral contraceptives byfemale workers; workers are trained toscreen clients for contraindications, andeach acceptor must be examined by adoctor within three months. Male work-ers concentrate mainly on environmentalsanitation but also provide family plan-ning information and distribute con-doms. Between them they are expectedto cover a population of 5,000 (3,000 inremote hilly and tribal areas), althoughin many parts of India, this coverage hasnot yet been achieved.

Indonesia. On the islands of Java and

Bali, there is about one family planningfieldworker to every 2,000 eligible cou-ples. The fieldworkers, who are normallysecondary school graduates, recruit newacceptors, provide door-to-door sup-plies, and provide the managerial linkbetween health clinics and part-timelocal volunteers who run village and sub-

village contraceptive resupply centers.Financing constraints have precludedreliance on paid fieldworkers in recentextensions of the program into the otherislands.

Kenya. The privately run Chogoriahospital project in the Meru district hasused volunteer workers selected by local

health committees to provide pills, con-doms, and other basic health services tovillages. Volunteers are supervised bypaid workers attached to local healthcenters. Contraceptive prevalence hasreached 28 percent, compared with a rateof 7 percent in the rest of the country.

Korea. Full-time paid family plan-ning fieldworkersnurses, midwives,and nurse aidesare assigned to healthsubcenters from which they spend atleast fifteen days each month makinghome visits and organizing group meet-ings to recruit eligible couples. They alsodistribute condoms and pills and referIUD and sterilization clients to desig-nated family planning clinics. Coverageaverages one fieldworker per 2,600 mar-ried women of reproductive age nation-wide but is greater in rural areas (one per1,200 couples), than in urban (one per6,900 couples) because of greater dis-tances in rural areas.

Mexico. The national program pro-'ides outreach services through four dif-ferent government agencies. The Secre-tariat of Health and Welfare trainsmultipurpose fieldworkers who concen-trate mainly on family planning. Theyare local volunteers who receive smallincentive payments. The Social SecurityInstitute runs a program to reach isolatedareas by training traditional midwivesand other local volunteers to provideinformation and supplies in exchange fora modest payment. The Secretariat ofAgrarian Reform and the National Sys-tem for Integrated Family Developmentalso provide services through outreachworkers.

Pakistan. In 1981 the governmentreorganized its program to include a sys-tem of fieldworkers and community vol-unteers. The earlier system was basedentirely on paid fieldworkers, whichproved costly and ineffective. The newprogram uses locally recruited male and

female volunteersincluding satisfiedclients, barbers, and teachersto informcouples about available services, teachthem the advantages of family planning,encourage breastfeeding and childspac-ing, distribute nonclinical methods, andrefer clients to family welfare centers forother methods. The volunteers aretrained and supervised by one male andone female worker at the family welfarecenter.

Philippines. About 3,000 outreachworkersone to every 2,000 eligible cou-pleswork as full-time governmentemployees. Each worker recruits, trains,and supervises about sixteen communityvolunteers who provide information tocouples, supply condoms and pills tocurrent users, and make referrals to gov-ernment health clinics. Some 50,000 vol-unteers serve almost three-quarters ofthe nation's eligible couples. The futureof the outreach program is uncertainbecause external funding will terminatein 1985 and local governments have notbeen able to absorb the cost of thefieldworkers' salaries as rapidly asexpected.

Thailand. Until recently, the Thainational program has been clinic based.Now multipurpose village health volun-teersserving nearly half of the nation'svillageshave been trained to providefamily planning information and areauthorized to resupply pill and condomacceptors. They also serve as referralagents for a mobile sterilization service.

Zimbabwe. The Child Spacing andFamily Planning Council, a parastatalunder the Ministry of Health, providesmany of the services and has about 300full-time, single-purpose outreach work-ers who supply oral contraceptives torural couples through regular home vis-its. A new project will train another 500to 600 fieldworkers by 1987.

wise would; and community-based fieldworkersare often most aware of local needs. But the exten-sive use of fieldworkers requires regular, support-ive supervision. They must be trained well at theoutset and must receive periodic refresher courses

to maintain the quality of services. They shouldconcentrate on a few main tasks; additional res-ponsibilities must be introduced only gradually.Fieldworkers also require a good medical backupand referral system so that any side effects that

141

clients may develop can be promptly treated.Finally, supervisors and fieldworkers must travelfrequently, and contraceptive supplies must bemade available in an increasing number of remoteoutlets. Money for transport is often first to besacrificed when budgets are cut, yet the wholestrategy depends on extensive travel and goodlogistics.

ENCOURAGING PRIVATE SUPPLIERS. Another way in

which governments have increased access to fam-ily planning services is by encouraging wider pri-vate involvement. This strategy makes fewerdemands on scarce public funds and on adminis-trative capacity. Policies include subsidizingcommercial distribution of contraceptives, coordi-nating with and encouraging private nongovern-mental organizations (NGOs), and removing legaland other barriers to private and commercial provi-sion of contraception.

Subsidized provision of contraception throughcommercial outletsoften called social market-inghas been tried with some success in at leastthirty countries. Social marketing programs useexisting commercial distribution systems and retailoutlets to sell, without prescription, contraceptivesthat are provided free or at low cost by govern-ments or external donors. The first social market-ing scheme was in India, selling subsidized"Nirodh" condoms. Almost all countries withsuch schemes sell condoms, and at least seventeenare known to sell oral contraceptives, sometimesseveral brands. Spermicides, in the form of sup-positories, creams, pressurized foam, and foamingtablets are also commonly sold. Until recently,social marketing schemes have been limited tomethods that do not require clinical services fordistribution. But Egypt now sells subsidized IUDsthrough private doctors and pharmacies. And inBangladesh there are plans to test-market inject-able contraceptives through social marketingarrangements.

Social marketing makes family planning sup-plies more easily accessible by increasing the num-ber and variety of outlets through which they canbe obtained: pharmacies, groceries, bazaars, streethawkers, and vending machines. In Sri Lanka some6,000 commercial outlets sell subsidized condomsand pillsmore than five times the number of gov-ernment family planning outlets. In the late 1970ssocial marketing schemes accounted for more than10 percent of total contraceptive use in Jamaica,Colombia, Thailand, and Sri Lanka. In Bangladeshthe social marketing program supplied about one-

142

quarter of couples who used contraception in 1983:it accounted for 67 percent of total condom use, 12percent of oral contraceptive use, and 70 percent ofspermicide use. In 1981 about half of all pill usersand 80 percent of condom users in Sri Lankaobtained supplies from the social marketing pro-gram.

Reliance on commercial distributors does not liftall the burden off the public sector, however. Thepublic sector still has to provide advertising, pro-motion, contraceptive supplies, distribution, andmedical backup. Some training is necessary forcommercial suppliers to dispense oral contracep-tives and to advise clients how to use them prop-erly, as has been done in Jamaica, Korea, Nepal,and Thailand. Failing that, some system of referralor prescriptions must be developed.

Although government subsidies to the commer-cial sector are usually provided for contraceptivesupplies only, some governments also subsidizeIUD insertion, abortion, and sterilization by pri-vate physicians. In Korea more than 2,300 physi-cians have been trained and authorized by the gov-ernment to provide family planning services. Thegovernment pays the entire cost of sterilization,but the cost of IUD insertion is sharedtwo-thirdsby the government, one-third by the client. Theinvolvement of private physicians has been a cru-cial factor in the success of the Korean program,although in 1978 about 60 percent of rural town-ships still had no authorized physician.

Access to services has also been increased by col-laborative efforts between government and NGOs.This collaboration has taken many forms: subsidi-zation of or grants to NGO services, coordinationof NGO and government services to assure maxi-mum coverage and allocation of responsibilty forcritical functions or services in certain regions toNGOs. In Bangladesh and Indonesia, for example,government services are allocated to rural areas,leaving NGOs to provide a large share of urbanservices. Since 1973 the Brazilian Family PlanningAssociation (BEMFAM) has worked with the gov-ernments of several states in Brazil to establishcommunity-based programs for low-incomegroups in the Northeast. The private nonprofitprogram in Thailand acts as an extension of thegovernment's rural health service and recruitslocal distributors to promote family planning andsell subsidized contraceptives donated by the gov-ernment and international agencies. By mid-1978there were some 10,000 distributors covering one-quarter of the 600 districts in Thailand. In Kenya in1980, NGOs were operating 374 out of 1,204 rural

health facilities. But less than 1 percent of theNGO facilities offered daily family planning serv-ices, and only 7 percent offered part-time services.A new project is creating family planning servicedelivery points in at least thirty of the NGO facili-ties. In addition, both government and NGO rep-resentatives will sit on a National Council on Pop-ulation and Development that will coordinatenational efforts in population information, educa-tion, and communications.

Governments have also removed legal and regu-latory obstacles that restrict commercial distribu-tion. In Egypt the sale of oral contraceptivesthrough private pharmacies does not require aphysician's prescription, although their provisionthrough government clinics serving rural areasdoes. Several countriesincluding China, Mexico,Morocco, the Philippines, and Thailandallowpills to be distributed in facilities other than phar-macies or health centers. Other options for stimu-lating the private sector include removal of importtariffs on contraceptive supplies (Korea recentlyeliminated a 40 percent tariff on raw materials fordomestically produced contraceptives); active gov-ernment promotion of condoms, spermicides, andpills that can be easily supplied through commer-cial outlets; and training of private pharmacistsand physicians who frequently have little knowl-edge of modern family planning methods.

Improving quality

The quality of family planning services matters inall phases of program development. In the earlystages services are new, and contraception stilllacks social legitimacy. Once programs are wellestablished and accessible, quality counts becauseother costs of family planningsuch as physicalside effectshave replaced access as the factor lim-iting the success of the program. Three ingredientsof qualitythe mix of contraceptive methods, theinformation and choice provided, and programfollow-uphave contributed much to programsuccess (see Box 7.6).

THE METHOD MIX OF PROGRAMS. The number andcharacteristics of available contraceptive methodsaffect the ability and willingness of clients to prac-tice birth control. Additional options are likely toincrease acceptance, permit switching, and reducediscontinuation rates.

Some women have medical conditions thatrule out certain methods. Oral contraceptivesshould not be prescribed for women who are over

forty years old, who smoke and are over thirty-fiveyears old, who are breastfeeding, or who have ahistory of stroke, thromboembolism, cancer, liverdamage, or heart attack. The IUD is undesirablefor women with pelvic infection or a history ofectopic pregnancy. Some women cannot be prop-erly fitted with diaphragms.

If the side effects of one method cannot betolerated, the availability of other methodsimproves the chance that couples will switchrather than stop using contraception altogether.For example, in Matlab Thana, Bangladesh, 36 per-cent of women had switched methods within six-teen to eighteen months after initial acceptance.And a study in the United States showed that mar-ried white women aged twenty-five to thirty-ninehad used an average of more than two methods;more than a third of those aged twenty-five totwenty-nine had used three or more.

Couples' preferences are influenced by theirfertility goalspostponing a first birth, spacingbetween children, or limiting family size. Womenusing the pill tend to be younger and to have hadfewer births than those protected by sterilization;many of the former are spacing births, while thelatter have completed their families.

Some methods of fertility control may be reli-giously or culturally unacceptable. Two-fifths ofthe world's countries, comprising 28 percent of itspopulation, either prohibit abortion completely orpermit it only to save the life of the mother. Forreligious reasons, sterilization is illegal in severalcountries. When couples regard periodic absti-nence as the only acceptable form of birth control,programs should provide information on propertiming of abstinence, although this method carrieshigher risks of unwanted pregnancy.

Due to sheer lack of alternatives, early familyplanning programs offered only a limited range ofcontraceptive methods. In the late 1950s and early1960s, the Indian program had to rely on rhythm,the diaphragm, and the condom. Today, mostnational programs offer a wider variety of meth-ods, although the number available at any givenoutlet is often fewer than that implied by officialstatements. Some governments still promote a sin-gle method because such an approach is easier toadminister or because certain methods, such assterilization and the IUD, are viewed as more"effective" and require less follow-up over thelong run than do other methods. For example,India, Korea, and Sri Lanka continue to emphasizesterilization. Until recently, Indonesia had almost

143

Box 7.6 The impact of service quality: Matlab Thana, BangladeshMatlab Thana is an administrative divi-sion of 280,000 people in a rural area ofBangladesh. Its population density is2,000 people per square mile. Transport

is difficultmostly by boatand incomesare low. Fishing and farming are themain activities.

Between 1975 and 1981 the Interna-tional Centre for Diarrhoeal DiseaseResearch, Bangladesh conducted twoexperiments in Matlab Thana to measure

the effect that availability, access, andquality of family planning services hadon contraceptive use. Before 1975 family

planning services were based in a gov-ernment-run center in Matlab town. Asmall staff provided a conventional range

of contraceptives and HiD insertions but,

with the exception of two brief house-to-

house campaigns conducted nationally,made little attempt to reach out to thevillagers. Throughout Bangladesh, unmet

need for contraception clearly existed.A national survey in 1968 showedthat 55 percent of rural married womenwanted no more children and that

13 percent would consider using contra-ception, but that only 1.9 percent werecurrently using a method.

The Contraceptive Distribution Pro-

,'rn,,:nIe (CDP). The first of two experi-ments, from 1975 to 1978, tested theeffect of house-to-house distribution oforal contraceptives and, one year later, of

condoms. Female workers were given six

half-days of training on the proper use of

the condom and the pill, adverse symp-

Comparison of the cumulative contraceptive acceptance and user rates in the first 18-24months, for the simple household Contraceptive Distribution Project (CDP) and theFamily PlanningHealth Services Project (FPHSP)

Percentage of eligible women

40

CDP

, Acceptors30

20

10

0

0 3 6 9 12 15 18 21

Oct. 75 Months

Users

toms, expected side effects, and simpletreatments for them. These workers were

mostly elderly, widowed, and illiteratewomen, with almost no personal experi-ence of contraceptives. Beginning inOctober 1975, they visited each house-hold in the project area of 150 villages.During a 5-10-minute visit, women weretold about the benefits of spacing andlimiting births, proper use of the pill, andpossible side effects. Those who wereinterested were given a six-month sup-ply of pills. For thirty months, workerswere responsible for continuing to

20/

10

0

24 3 6 9 12 15 18

Oct. 77 Months

Source: Reprinted with the permission of the Population Council from Shushum Bhatia and others. "TheMatlab Family Planning-Health Services Project," Studies in Family Planning 11, no. 6 Oune 1980): 210.

recruit acceptors, resupplying users, andadvising on side effects.

The impact of the CDP was great butshortlived. Contraceptive use in the proj-ect area jumped from 1.1 to 17.9 percent

in three months, but declined to 11 per-cent after two years. During the sameperiod, the rate of contraceptive use Out-

side the project area increased from 2.9to 3.8 percent. After a year, 34 percent of

married women in the project area hadaccepted contraception, but only 42 per-cent of these women were continuing to

use it (see chart). Some ten to fourteen

exclusively promoted the pill but now is givingmore emphasis to the IUD.

Supply constraints also limit the availability ofdifferent methods. Most contraceptives areimported and are often provided free or cheaply bydonors; China, India, and Korea, which producemost of their own contraceptives, are major excep-tions. Heavy reliance on one donor can causeproblems, since some donors can supply only cer-tain types of contraceptives. The United StatesAgency for International Development (USAID) islegally prevented from financing abortion trainingor services and does not finance Depo-Provera, aninjectable contraceptive, because it has not been

144

approved for use in the United States. Because ofthe limited number of donors that supply inject-able contraceptives, Thailand almost exhausted itssupplies in 1982, raising the prospect that manyclients would have to switch methods or discon-tinue altogether. Difficulties can also arise if donorschange suppliers, since the hormonal makeup oforal contraceptives varies from one manufacturerto another. Other factors restricting method mixinclude shortages of trained staff to perform sterili-zations, poor transport and logistics for timelyresupply, and the great distances that clients musttravel to obtain some methods.

To improve the method mix of programs, male

Percentage of eligible women

40 . Acceptors

FPHSP /Users

30

months into the program, fertility haddeclined by 11 to 17 percent, but thiseffect lasted only one year. The project'slimited impact was attributed to poormanagement of side effects, inadequatetraining of staff, insufficient informationprovided to clients, the narrow range ofcontraceptive methods (which discour-aged method switching), and too littlesupervision.

The Family Planning-Health Services Proj-

ect (FPHSP). In October 1977 a secondexperiment also tested house-to-housedistribution of contraception, but withmuch better quality of services. Femalevillage workers were recruited locallyand received seven weeks of preservicetraining and weekly in-service trainingsessions. They were literate, marriedwith children, had contraceptive experi-ence, and came from respected families.Eighty workersone per 1,000 peoplereceived technical supervision and medi-cal backup from four clinics staffed byqualified women paramedics, and admin-istrative supervision from a male seniorhealth assistant.

The FPHSP provided comprehensiveservices for the special needs of each cur-rent and prospective client. The methodsoffered included not only pills and con-doms, but foam tablets and injectables.In addition, women were referred to cen-ters where tubectomy, IUD insertion,and menstrual regulation could be per-formed, and where their husbands couldget vasectomies. All households were

visited once a fortnight, regardless ofwhether couples were using contracep-tives. Side effects were managedthrough reassurance, frequent method-switching, and medical referral for treat-ment. Workers also offered aspirin, vita-mins, and iron tablets, thereby gainingaccess to households that had previouslyrejected family planning.

In the first three months contraceptiveuse in the pro)ect area rose from 7 to 21percent. Unlike the trend in the CDP,however, the rate continued to climbslowly to 34 percent. Continuation rateswere dramatically improved: after a year,39 percent of eligible women in theFPHSP had accepted contraception and81 percent of these women were continu-ing to use it (see chart). During the firsttwo project years, fertility declined by 22to 25 percent compared with villages out-side the project area. After a three-yearplateau at 34 percent, contraceptive usebegan to rise and now stands at 41 per-centalmost exclusively modern meth-ods. The injectable, Depo-Provera,accounts for almost half of contraceptiveuse. In the rest of the country in 1983,modern methods account for only 14 per-cent of contraceptive use. Tetanus, tox-oid, and oral rehydration therapy havebeen added to the Matlab project's ser-vice package hut were apparently notresponsible for increased contraceptiveprevalence.

Replicahility. The FPHSP has beenhighly effective in increasing contracep-

tive use in field conditions typical of ruralBangladesh. But it may be hard to repli-cate on a larger scale because the FPHSPwas able to draw on extra resourcesunavailable to the national family plan-ning program. For example, althoughfieldworkers in the project receive sala-ries equivalent to workers in the nationalprogram, their supervisors' salaries aremuch higher. The project also used costlyspeedboats to move supervisors andresearch staff around the area. And man-agement was decentralized to an extentrarely found in national programs. Themanagerial and organizational structurethat guaranteed close, supportive super-vision, worker accountability, continu-ous training, good recordkeeping, andcontinuous feedback to workers shouldtake much of the credit for the project'ssuccess.

The government of Bangladesh andthe International Centre are nowembarking on an extension project totransfer some of the management tech-niques of the Matlab project to govern-ment health and family planning work-ers in several thanas in North Bengal andto measure the impact of these changeson fertility, mortality, and contraceptiveuse. The project will make minimalchanges in the existing program struc-ture and there will be no special inputsother than for training, organization-building, and research.

and female sterilization and IUDs can be mademore readily available through mobile facilities(such as sterilization vans in Thailand) or periodic"camps" (such as vasectomy and tubectomycamps in India and IUD "safaris" in Indonesia).Careful attention must be paid to providing follow-up services in the case of complications, however.Paramedical workers can be trained to provide theIUD and injectable contraceptives in clinics andeven in homes. Referral procedures can bestrengthened so that clients are informed about allmethods available from public, private, and com-mercial sources. Private suppliers can be encour-aged to offer contraceptive methods that are in

short supply or that cannot be offered by the offi-cial program. Finally, governments can sponsorlocal research on the effectiveness, side effects,and acceptability of methods that might be intro-duced into the national program.

INFORMED CHOICE. Although family planningworkers may know more about the advantagesand disadvantages of each method, clients are bestequipped to choose what suits themprovidedthey have information on effectiveness, sideeffects, reversibility, and proper use. In the earlystages of the Indian and Pakistani programs, theside effects of the IUD were not fully explained, a

145

medical examination was not always conductedbefore insertion, and there was little in the way oftreatment or referral for side effects. For yearsafterward, IUDs were shunned. With a fullerexplanation of side effects and greater care paid toscreening and medical backup, the IUD is nowregaining popularity. When private pharmacies inColombia provided their customers with pam-phlets explaining effectiveness, proper use, andside effects, sales of contraceptives increased.

Virtually all family planning programs providesome information to clients about methods, butfully informed choice is still only an ideal in manycountries. Family planning workers still tend todoubt the ability of couples to use effectively meth-ods such as the condom and pill, thereby discour-aging their use. Staff may also fail to mentionmethods of which they disapprove, such as thepill, abortion, or sterilization. When incentives areoffered to staff for recruiting acceptors of somemethods but not of others, the information pro-vided to clients may be biased. Sometimes clientsare given inaccurate or incomplete informationbecause family planning staff are themselves notproperly informed about methods and their sideeffects. A survey of the Dominican Republic,Kenya, and the Philippines by the United NationsFund for Population Activities (UNFPA) in themid-1970s found that workers felt that their train-ing in methods had been inadequate. A study inIndia, Korea, the Philippines, and Turkey demon-strated the strong influence of providers of serv-ices on clients' choice of method: clients given athorough explanation of all available contraceptivemethods chose a very different mix of methodsthan did those prior to the study, who had notbeen given this information.

From the manager's viewpoint, what are the crit-ical requirements for better information? First isappropriate training. Workers must be trained toexplain properly the methods available to clientsand to encourage them to participate in the choice.Informal explanation works better than formal pre-sentations that use technical or anatomical terms.As new contraceptive methods are included inprograms, staff must receive prompt training. Sec-ond is more and better supervision of workers toensure that they are not holding back informationon methods because of their own prejudices orbecause they are receiving financial incentives forencouraging some but not all methods. The incen-tive structure might also be altered by offeringfinancial or other awards (such as educationalopportunities or additional training) to the worker

146

who attracts and retains the most clients for a vari-ety of different methods.

FOLLOWING UP ACCEPTORS. In their early stages,family planning programs devoted much time torecruiting new clients. It is now obvious that sus-tained use cannot be assumedfollow-up supportis needed. Follow-up support includes medicalbackup and referral for side effects; encouragingclients to change contraceptive methods if their ini-tial choice has caused problems or if their needshave changed; reassuring them that they are usingcontraceptives properly; and reminding them ofthe benefits.

Follow-up is most important in the first fewmonths after acceptance, since this is when sideeffects are first experienced, when clients are learn-ing to use methods properly, and when they needreassurance in the face of social disapproval. Astudy in Calabar State, Nigeria, found that 11 per-cent of pill acceptors never took even the firstmonth's allotment of pills, and only 53 percentwere using the pill three months later. A lack ofconcern with follow-up is believed to be the majorcontributory cause of the low continuation ratesamong IUD and pill users in Korea. According to asurvey of contraceptive acceptors, only 24 percentwere followed up at home or returned to healthcenters for consultation on side effects. Korea'sprogram sets targets for the number of acceptors,but none for follow-up work.

Follow-up cannot be left to clients, who are likelyto return to the family planning center only if theyare living close by or if they experience severe sideeffects that they cannot correct even by abandon-ing contraception. Follow-up is best provided byfieldworkers and by community-based services.But in areas in which family planning is stillregarded with suspicion, some clients would liketo be spared the embarrassment of a follow-upvisit from a family planning worker. Some pro-grams have managed this by having fieldworkersdeliver health services as well.

With or without an extensive field network, fam-ily planning programs can improve follow-up.

They can change policies that encourage staffto recruit new acceptors but not to follow up onthem. Targets and incentives can be offered to staffon the basis of the number of current users of con-traception or of the number of checkups, ratherthan only on the basis of the number of new accep-tors. Training must also emphasize follow-upprocedures.

Where the burden of follow-up rests on cli-ents, programs can experiment with various waysof encouraging clients to seek appointments. Forexample, financial incentives might be offered toclients who return for a follow-up visit within aspecified period of time, just as South Asian pro-grams offer compensation to acceptors of steriliza-tion for the costs of transport, food, and work timelost. The media can also be used to reassure accep-tors about side effects and to encourage them toreturn for checkups.

The quality of follow-up can be monitored byperiodic sample surveys of acceptors.

Ensuring social acceptability

To be successful, family planning programs musthave the support of the clients and communitiesthey serve. But in communities in which modernfamily planning has never been provided, theremay be little evident demand because potential cli-ents are not aware of the benefits of the service, ofsmaller families, or of longer child spacing inter-vals. Services introduced by an "outside" agencywith few local links and little appreciation of localcustoms and needs may not be readily accepted.The absence of links to the local community can bea weakness for family planning in particular,because it is a personal matter and may conflictwith social norms that favor high fertility.

Private family planning associations and NGOshave led in experimenting with new ways toinvolve clients and communities. Their strategieshave included consultation with local leaders,training local people as paid or volunteer workers,consulting and training traditional midwives andhealers, establishing local management or reviewcommittees, encouraging local contributions ofmoney and labor, and organizing groups of familyplanning acceptors to reinforce effective use and toengage in other community development projects.

In communities where there is no apparentdemand for family planning, it can be introducedjointly with services in greater demand. The Hon-duras Family Planning Association includes aplanned parenthood theme in its community-based adult literacy program. In Awutu, Ghana,family planning is promoted for child spacing aspart of a maternal and child health project. Familyplanning is provided with agricultural extension toa population of 100,000 in Allahabad (in the stateof Uttar Pradesh, India) and as part of the nation-wide Integrated Rural Development Project inPakistan. It has been offered through the resettle-

ment schemes of the Federal Land DevelopmentAuthority in Malaysia and through women's ruralcredit cooperatives and vocational training inBangladesh. Profamilia, the private family plan-ning organization in Colombia, extended its serv-ices to the countryside through the National Fede-ration of Coffee Growers. In China, India, and thePhilippines family planning services are organizedin factories. Both the Indonesian and Chinese pro-grams have used strong political organizations,which extend into rural areas, to provide manyeconomic and social services, including familyplanning.

Private family planning associations are wellsuited to implement these approaches: they aresmall, decentralized, well staffed, highly moti-vated; have greater control over service quality;and are less confined by the bureaucratic con-straints of government. But many of theseapproaches have also been tried on a larger scale.For example, the Planned Parenthood Federationof Korea pioneered the highly successful mothers'club program. At first these clubs served assources of contraceptives, of reassurance for accep-tors, and of information on the benefits of familyplanning. They now have merged with theSaemaul Women's Association and are alsoinvolved in agricultural cooperativs and commu-nity construction projects. Mothers' clubs havealso been used by programs in Indonesia andBangladesh. The national program in Indonesiahas successfully involved village headmen, reli-gious leaders, and local volunteers on the islandsof Java and Bali, where more than two-thirds ofIndonesia's population lives. In the Philippinessome outlets for contraceptives are organized andrun by local volunteers.

Where communities and clients are involved,they are less likely to see family planning as beingimposed by outsiders. Use of traditional midwivesand volunteers, and local contributions in cash orin kind also reduce the cost of services. But thesestrategies require certain managerial qualities notalways found in larger public programs: decentral-ized decisionmaking; technical and organizationalexpertise to support local organizations, volun-teers, and clients; skilled managers and field-workers who can identify local leaders, stimulatecommunity activities, supervise volunteers, andreconcile local needs with program capabilities;and, sometimes, workers who are technically com-petent in more than one field. Finally, social accep-tance of family planning takes time and is a contin-uous process. There is no benchmark for

147

measuring social acceptability, or easy formula forensuring it.

Financing family planning

Public family planning programs, like programs ineducation and health, are heavily subsidized, andservices are often offered free of charge. Althoughthe private sector makes a significant contributionto providing services in some countries, publicfinance will continue to be critical, especially inlow-income countries and in backward regions,where contraceptive demand is limited and healthservices are weak.

Public spending

China and Indiathe two most populous coun-tries in the developing world, with approximatelyhalf its populationspent roughly $1.00 and $0.30per capita, respectively, on population programs in1980. In most of three dozen developing countriesfor which rough estimates are available, spendingfell within this range (see Table 7.4). If other devel-oping countries with programs were spendingequivalent amounts, the total spent on populationactivities in all developing countries in 1980 musthave been about $2 billion.

Practically all spending on population in China,and close to 80 percent of the total in India, isfinanced from domestic resources. For all otherdeveloping countries combined, government andforeign donors each contribute about 50 percent.The government share tends to rise the longer aprogram has been in existence. Three out of fourcountries with programs less than five years oldwere contributing less than 10 percent of the costsof their programs, in contrast to an average of 54percent among twenty-seven countries with pro-grams at least ten years old. Nepal is one of therare exceptions: the share of domestic governmentfinancing fell from 80 percent of its spending onpopulation in 1975 to 40 percent in 1980.

Even among well-established programs there iswide variation in government spending. Domesticbudgetary outlays in 1980 are estimated to havebeen $0.42 per capita in Sri Lanka, about $0.71 percapita in Korea, and $1.45 per capita in Costa Rica.But these estimates probably understate the truegovernment contribution. The cost of health work-ers, whose functions often include family plan-ning, is not always imputed to the population pro-gramnor are contributions by local government.

The estimate for China of $1 per capita includes

148

the amount spent by its formal layers of govern-mentcentral, provincial, prefectural, andcountyon providing contraceptive supplies freeto users;. reimbursing service fees for sterilization,abortion, and IUD insertion; and providing train-ing and information on family planning. Thesecosts amount to $213 million annually, about $0.21per capita. In addition, the rural collective systemfinances the family planning staff at the communeor brigade level (at an estimated cost of $0.34 percapita) and pays incentives, in the form of foodsupplements and reimbursement of travel costs, toholders of one-child certificates ($0.25 per capita)and to individuals undergoing sterilization ($0.15per capita). Finally, additional time is spent bybarefoot doctors on family planning work (thoughnot much: in Shandong Province they allocate anaverage of 1.5 percent of their time to family plan-ning, valued at approximately $3 million). Healthworkers and midwives probably spend more timeon family planning. Adding all these contributionstogether produces a figure for family planningexpenditure in China of nearly $1 per capita.

Although governments finance a large share oftheir population programs, the amounts spent arestill trivialboth in absolute terms and in relationto other government outlays (see Box 7.7). InChina the state budget for the family planning pro-gram absorbs only 0.4 percent of total currentspending, compared with 5.2 percent for healthand 13.1 percent for education. In India andMauritius spending on family planning in 1981accounted for only 0.5 percent of total governmentexpenditure. The figures are even lower in Korea(0.2 percent) and in Malaysia (less than 0.1percent).

Foreign donors spent an estimated $491 millionfor population programs in developing countriesin 1981; about two-thirds of this amount was forfamily planning and related programs. In realterms, population assistance grew at almost 6 per-cent a year during the 1970s but fell 3 percent in1980 and 6 percent in 1981. The prospects forincreased assistance are not good: UNFPA, a majorchannel for population assistance, expects itsspending to rise by barely 1 percent over the nextfour years. Population assistance from donors isdiscussed further in the next chapter.

Private spending

Important constraints limit the growth of privatesuppliers of family planning, especially in ruralareas. The most severe constraint is the need for

TABLE 7.4

Note: Expenditure includes funding from domestic and foreign sources on population activities, including (but not limited to) familyplanning services.a. Contraceptive prevalence rate unavailable or close to zero.Source: Bulatao, 1984a.

149

Public expenditure on population programs, selected countries, 1980

Totalpublic

expenditure

Per ca pitapublic

expenditure

Expenditureper current

contraceptiveuser

Region and country (millions of dollars) (dollars) (dollars)

Sub-Saharan AfricaGhana 2.8 0.24 16Kenya 11.8 0.71 68Liberia 2.3 1.22 aMauritius 1.7 1.81 24Sierra Leone 1.5 0.44 aSwaziland 1.8 2.89 aTanzania 3.3 0.18 aZaire 1.8 0.06 aZimbabwe (1978) 1.9 0.27 13

Middle East and North AfricaEgypt, Arab Rep. 34.1 0.81 22Iran, Islamic Rep. (1976) 50.6 1.30 38Jordan 2.5 0.78 21Morocco 13.3 0.66 aTunisia 8.3 1.31 32

South AsiaBangladesh 45.1 0.51 26India 226.9 0.34 10Nepal 10.6 0.72 69Pakistan 24.5 0.30 33Sri Lanka 6.2 0.42 7

East AsiaChina 979.6 1.00 10Hong Kong 2.0 0.40 3Indonesia 86.2 0.59 11

Korea, Rep. of 27.1 0.71 9

Malaysia 16.4 1.18 19Philippines 37.6 0.78 11

Singapore 1.8 0.74 7Thailand 28.1 0.60 7

Latin America and CaribbeanBolivia (1977) 0.1 0.03 a

Brazil 10.6 0.09 aColombia 8.1 0.31 4Costa Rica 3.3 1.45 15

Dominican Rep. 3.8 0.70 11Ecuador 6.3 0.75 15El Salvador 8.1 1.77 35Guatemala 9.3 1.28 47Haiti 3.9 0.77 27Honduras 3.0 0.81 20Jamacia 4.8 2.19 27Mexico 61.3 0.88 15

Panama 4.4 2.42 26Paraguay 2.1 0.69 13

Peru 5.3 0.32 5

150

Box 7.7 Military versus social expenditure

Military spending is not easy to measure.What estimates there are indicate thatglobal military expenditure in constant1982 dollars has risen from $300 billion. to

more than $600 billion in the past twentyyears. The amount spent in the develop-ing world quadrupled from $30 billion tomore than $138 billion. In 1981 devel-oped countries spent more than 4.9 per-cent of GNP on defense, and about 0.3percent of GNP on aid to deelopingcountries. In 1980 the United Statesspent 5.6 percent of its GNP on defense,almost $170 billion, and 0.28 percent($8.2 billion) on aid. In developing coun-tries almost as much is spent on defenseas on education and health combined(see chart).

Expenditures on defense, education, andhealth as a percentage of GNP, 1980

Sourcn Sivard, 1983.

medical backup for providing contraceptives.Although prescription regulations have been liber-alized in nineteen countries, making condoms andoral contraceptives available through nonclinicalsuppliers such as pharmacies, the demand forIUDs, male and female sterilization, and abortioncan be met only by trained health workers. Veryfew of them work in the private sector. In addition,the cost of providing family planning services ishigh in rural and marginal urban areas, whereasthe ability to pay for commercial services is low.Private suppliers cannot appeal to the nationalinterest the way governments can to stimulatedemand for contraception and cannot use commu-nity institutions and pressures to spread familyplanning. Finally, the development of private sup-ply is often inhibited by a combination of govern-ment policies, including price controls; prohibitionof, or tariffs on, the import of contraceptives; andrestrictions on certain kinds of family planning ser-vices, especially sterilization and abortion in Mus-lim and Catholic countries.

Despite these constraints, private suppliers pro-vided more than 20 percent of all family planningservices in more than two-thirds of the countriesstudied in recent surveys (see Table 7.5). In somecountries private suppliers play a major or evendominant role, especially among urban con-sumers. In Korea 42 percent of all contraceptiveusers are supplied by pharmacies or physicians; in

Defense

Education

Health

iA/orld Des eloped Developing

Average government per capita expenditureon defense, education, and health, 1980

World Developed Developing

the state of Sao Paulo, Brazil (a country with somestate, but no central government, programs), theproportion is as high as 63 percent.

Private spending on family planning services asa consequence equals about a fourth of publicspending on these services in the developingworld. In some regions private spending isgreater: in Latin America it may in fact be slightlyabove public spending.

In urban areas some commercial suppliers maybe displaced by publicly subsidized contracep-tives: half of the initial users of an official programof oral contraception in Piaui State, Brazil, in 1979had shifted over from the private sector. Yet pri-vate suppliers do benefit from family planningadvertising financed by the government. It is prob-ably not a coincidence that they flourish in severalcountries, such as Korea, Mexico, and Thailand,where government strongly supports birth con-trol.

The activities of private suppliers demonstratethat many people in developing countries are will-ing to pay for contraceptive services. Althoughcharges for publicly subsidized services are usuallylow or nonexistent, data for twenty developingcountries show that private sector prices can behigh enough to absorb a significant fraction ofhousehold income. The cost of a year's supply oforal contraceptives averaged $25 in 1980, rangingfrom $5 in Mexico and $6 in Egypt to as much as

Percent

10

Constant 1982 dollars

600

$90 in Nigeria. Across countries, the various formsof contraception cost an average of $20 to $40 ayear.

In the better-off developing countries, the cost ofbuying commercially available contraceptives issmall in relation to average income per capita(although even in those countries the cost may berelatively large for the poor). For example, theretail price of a year's supply of oral contraceptivesin 1979 was equivalent to only 0.3 percent of percapita income in Mexico and to 0.5 percent inBrazil. But in low-income countries the cost can beprohibitiveequivalent to 17 percent of per capitaincome in Bangladesh, for example, and 18 percentin Zaire, or about 3 percent of total income for theaverage household. All these figures understatethe real cost of obtaining family planning services,whether private or public, because people alsohave to pay for the time and travel needed toobtain their contraceptives.

In Korea some 1.2 million users bought contra-ceptives commercially in 1979 at an average annualcost of about $12a total outlay of $15 million,about $0.40 per capita for Korea's entire popula-tion, and roughly equivalent to the $0.42 per capitaspent on the domestic government budget, exclu-sive of foreign donor contributions. In Peru about

TABLE 7.5

Source of contraception among currently married women aged 15 to 44 and their husbands(percentage distribution of current contraceptive users)

300,000 users of commercial sources of contracep-tion spent an average of $30 each in 1981, theirtotal outlay being several times what the govern-ment spent. Private spending on this scalewhichunderstates the total because it excludes accesscostsis not typical of all developing countries,but it shows a widespread willingness to pay forcontraception.

Allocation of public expenditu res

The bulk of public spending on populationalmost 50 percent in seventeen countries reportingdetails of expendituresgoes directly to providingcontraceptive services. Progressively smallershares are taken up by general program adminis-tration, information-education-communicationactivities, research and evaluation, and personneltraining.

With all public spending on family planningtaken into account, expenditure averages about$0.70 per capita across all developing countries.For each contraceptive user, spending is muchhigheraround $21 a year. But most users are inChina and India, where programs spend less peruser, so the weighted average is lower at $11. Add-ing private expenditures could easily double the

a. Applies to rhythm or withdrawal; other may include contraceptives obtained from a friend or in a foreign country.Sources: Morris and others, 1981; Merrick, 1984.

151

Region and countryGovernment

programsOther publicly funded or

subsidized programsPrivatesector

No source orother'

East AsiaKorea, Rep. of (1979) 36 0 42 22Thailand (1978) 37 35 18 10

Latin America and CaribbeanBrazil

Piaui (1979) 59 0 23 18Sao Paulo (1978) 16 0 63 21Bahia (1980) 27 1 48 24Rio Grande do Norte (1980) 57 0 22 21

Colombia (1978) 21 27 33 19Costa Rica (1978) 57 0 28 15El Salvador (1978) 73 8 12 6Guatemala (1978) 44 11 26 18Jamaica (1979) 63 27 7 3Mexico (1978) 42 2 36 20Panama (1979-80) 71 0 19 10Paraguay (1977) 41 8 28 22

North AfricaTunisia

Jendouba (1979) 91 0 5 4

costs per user. Public cost per user varies amongcountries, as Table 7.4 shows, depending on manyfactors, including local salaries and program eff i-ciency and quality.

Cost per user tends to be very high in the firstfew years of a family planning program; it thenfalls sharply as the rate of contraceptive use risesabove 5 percent. At higher rates the cost per usertends to stabilize, or perhaps to rise slightly.Between 1965 and 1980, while contraceptive use inKorea rose from 12 to 30 percent, cost per userfluctuated (with little apparent trend) between $7and $13 (in constant 1982 dollars).

In any country with contraceptive use of at least5 percent, current cost per user is a conservativeguide to costs at higher levels of use. Marginalcosts could rise if new users are in inaccessiblerural areas with high delivery costs, though theycould also fall if services are more intensively used.

Future financial requirements

What would it cost to satisfy the unmet need forlimiting births? Some idea can be obtained byextrapolating levels of unmet needthe propor-tion of women exposed to the risk of pregnancywho want no more childrenin thirty-five devel-oping countries in the mid-1970s to cover thedeveloping world as a whole. That extrapolationsuggests a possible increase in the rate of contra-ceptive use of 13 percentage points. If the publiccost for each additional user were the same, coun-try by country, as the cost per user in 1980, such anincrease would require another $1 billion in publicspending (see Table 7.6).

In the next two decades total spending for familyplanning programs will need to increase becauseof the growing number of women of childbearingage and the increasing proportion of them who arelikely to want modern contraceptives. World Bankprojections indicate that the number of marriedwomen of reproductive age in all developing coun-tries will increase from about 500 million to morethan 700 million between 1980 and 2000. About 40percent of these women used contraception in1980.

The "standard" projections in Chapter 4 implyan average total fertility rate of 3.3 in developingcountries in the year 2000. If it is assumed that thefertility effects of later marriage and of shorterbreastfeeding will largely cancel each other outand that the abortion rate will stay constant,achieving this fertility decline will require anincrease in the rate of contraceptive use to 58 per-

152

cent. For the projections of a "rapid" decline infertility, which imply a total fertility rate of 2.4 in2000, contraceptive use would need to reach 72percent.

How much would this cost? To achieve thestandard decline in fertility, and assuming 1980costs per user, total public spending on populationprograms would need to reach $5.6 billion (in con-stant 1980 dollars) by the year 2000a rise in realterms of 5 percent a year. To ensure the rapiddecline, spending would need to total $7.6 billionby 2000, a rise of 7 percent a year in real terms.

Growth in spending will have to be muchgreater in some regions than in others. Averagereal increases in spending of 2.5 percent a yearwould be enough to meet targets in East Asia as awhole (though not for individual countries), and 5percent would be enough for Latin America andthe Caribbean. In South Asia, the Middle East andNorth Africa, and sub-Saharan Africa, however,population spending would have to grow 8 to 10percent every year to achieve a standard decline infertility, and in sub-Saharan Africa as much as 16percent every year to support a rapid decline.

Because spending on population currently repre-sents less than 1 percent of government budgets,small increases could go a long way toward meet-ing the requirement for higher spending. Thesame is true for external assistance. Only about 1percent of official aid now goes for populationassistance (and only a part of that for family plan-ning). Increasing spending by 50 percent could fill"unmet need" today, but larger increases will beneeded in the future. In many countries therequired increase in public expenditure for familyplanning would be more than offset by reductionsin public expenditure in other sectors. With con-stant enrollment rates, rapid fertility decline wouldgenerate per capita savings in education expendi-ture in the year 2000 of $1.80 in Egypt, $3.30 inKenya, $6.00 in Korea, and $6.60 in Zimbabwe.

Obstacles to program expansion

If the financial resources to expand family plan-ning services were made available, could they beput to good use? Program expansion may be diffi-cult for a variety of reasons, including administra-tive and logistical obstacles, scarcity of personnel,and limited demand. These tend to limit the rate atwhich a good program can be expanded, but notexpansion itself.

The administrative and logistical obstaclesinclude many of the same constraints that hamper

TABLE 7.6

Fertility targets and estimates of population program expenditures, 1980 and 2000

other development programs. For example, a fam-ily planning program requires a system for obtain-ing, storing, and distributing contraceptives. If aprogram attempts to provide a mix of methods,this system can become complicated; it mayrequire more than one distribution network-com-mercial, clinical, and nonclinical. Where overallgovernment administration is weak, roads arepoor, and communications slow, even the best-runprograms will appear inefficient and incapable ofsustained expansion. These limits may not be evi-dent in small pilot projects, but they can becomeimportant when an attempt is made to extend

Note: The fertility targets and per capita expenditure figures are population-weighted means. Because of lack of data on contraceptiveprevalence for many countries, regional estimates include country rates which were estimated based on various social and economicdata.

Includes Republic of South Africa.Includes Afghanistan.

Source: Bulatao, 1984a.

services on a larger, national scale. In areas wherehealth services are scant or nonexistent, a familyplanning program will be extremely difficult toimplement.

The personnel requirements for an extensivefamily planning program are not large in relationto the supply of educated people. Desirable ratiosare about 1 fieldworker to 300 families, and 1supervisor for every 8 fieldworkers. For UpperVolta, a country with extremely low literacy, a pro-gram could be fully staffed at these ratios by abouta tenth of a single year's primary and secondaryschool graduates. The conclusion becomes less

153

Region and scenario Year

Fertility targetsExpenditures

Per capita(constant 1980 dollars)

Total(millions of

constant1980 dollars)

Totalfertility

rate

Contraceptiveprevalence(percent)

All developing countriesCurrent estimate 1980 4.36 39 0.62 2,016With unmet need filled 1980 3.54 52 0.90 2,961Under standard decline 2000 3.30 58 1.14 5,569Under rapid decline 2000 2.32 72 1.66 7,591

Sub-Saharan AfricaaCurrent estimate 1980 6.59 11 0.29 112With unmet need filled 1980 6.03 20 0.76 297Under standard decline 2000 5.81 24 1.07 791Under rapid decline 2000 2.69 73 3.72 2,353

Middle East and North AfricaCurrent estimate 1980 5.70 24 0.66 142With unmet need filled 1980 4.98 35 1.04 222Under standard decline 2000 3.73 59 1.94 726Under rapid decline 2000 2.39 74 2.43 812

South AsiabCurrent estimate 1980 5.22 20 0.35 315With unmet need filled 1980 4.15 38 0.77 688Under standard decline 2000 3.43 51 1.10 1,517Under rapid decline 2000 2.42 67 1.50 1,873

East AsiaCurrent estimate 1980 3.02 61 0.87 1,238With unmet need filled 1980 2.27 72 1.04 1,480Under standard decline 2000 2.28 75 1.09 2,022Under rapid decline 2000 2.16 74 1.08 2,015

Latin America and CaribbeanCurrent estimate 1980 4.28 40 0.59 209With unmet need filled 1980 3.53 51 0.77 274Under standard decline 2000 2.80 63 0.95 513Under rapid decline 2000 2.17 72 1.07 538

sanguine, however, as soon as one takes intoaccount specific requirements for fieldworkers: forinstance, they should be village-based rather thancity-based, belong to the appropriate ethnic, lin-guistic, or caste group, and be favorably disposedto contraception. The Pakistani program has facedrecruitment problems of this sort. In the late 1960sonly a seventh of the midwives assigned asfieldworkers believed in the efficacy of moderncontraceptives. In the early 1970s they werereplaced by a group including many unmarriedwomen from urban areas who did not have theconfidence of the villagers. If finding appropriatefieldworkers in each area is difficult, findinghigher-level supervisors can be even more of aproblem.

Program expansion also depends on the demandfor contraceptive services. A principal task of pro-grams is to generate some of this demand itself,but where initial interest is low or nonexistent thistask can take time.

Taken together, these limits to rapid expansionmight seem to suggest that programs could notmake good use of more money. But such a viewwould be wrong. During the 1970s India, Pakistan,Bangladesh, and. Sri Lanka were spending asmuch as $2.50 per married woman on family plan-fling programs and were still producing contracep-tive users at acceptable costunder $20 each, in

154

some years much less. Despite the unpromisingconditionsper capita GNP between $100 and$300, adult literacy rates as low as 20 percent, andinfant mortality rates as high as 150 per thou-sandspending on family planning was effectiveand economical.

Furthermore, many of the factors that hampereffectiveness can be overcome as a programdevelops. Culturally acceptable solutions toadministrative and personnel problems, and tolimited public interest, take time to develop, as dothe quality improvements discussed above. But inevery part of the world where an effort has beenmade, there has been progress.

Foreign funding has been largely absent in theearly stages of some family planning programs, asit continues to be in China. In other programs ithas played a catalytic rolefor instance, throughstimulating pioneering research of demographicproblems. Local finance, however, eventuallybecomes critical; most of the older, more effectiveprograms in 1980 had 40 percent or less foreignfunding. For one thing, local finance demonstratespolitical commitment to family planning, the sub-ject of the next chapter. Many of the obstacles toexpansion of family planning can be overcomewith sufficient commitment, and most of themcannot be overcome without it.