DOCUMENT RESUME

ED 061 069 SE 013 416

TITLE Studies in Family Planning, Volume 3 Number 2.INSTITUTION Population Council, New York, N.Y.PUB DATE Feb 72NOTE 12p.AVAILABLE FROM The Population Council, 245 Park Avenue, New York,

New York 10017 (Free)

EDRS PRICEDESCRIPTORS

ABSTRACT

MF-$0.65 HC-$3.29*Behavioral Science Research; Contraception;Demography; *Family Planning; *Foreign Countries;Human Services; *Programs; *RePorts; Social Sciences;Statistical Data

A new design for government family planning pro,Tramsis proposed in "Family Planning Programs: An Economic Approach,,, theprincipal article in this monthly publication of The PopulationCouncil. The design is intended primarily for low-income countriesthat seek large and rapid reductions in fertility. Thirteen elementsof the proposed system of incentives and services are listed, withparticular emphasis on monetary payments to women who avoid births orpregnancies. Provision of services; client and service unit personnelincentives; follow-up care; training, education, and communication;research, evaluation, and reporting; and cost-benefit estimates aredescribed. Also included in the paper are two additional reports:"Village Midwives in Malaysia" and "Report of the Swedish AbortionCommittee." The first presents the results of an informalquestionnaire administered to village midwives in Malaysia recruitedfor a training program to provide health services in remote areas.The second is a brief review of the Swedish position on legalizedinduced abortion over the past three decades and a summary of therecommendations made by the 1965 Committee on Abortion. (BL)

Studies inamily Plannin

Cr" Volume 3, Number 2

-74

LIJ Family Planning Programs: An Economic Approach

U.S. CFPARTNIENT OF HEALTH.EDUCATION & WELFAREOFFICE OF EDUCATION

THIS DOCUMENT HAS BEEN REPRO-DUCED EXACTLY AS RECEIVED FROMTHE PERSON OR ORGANIZATION ORIG.INATING IT. POINTS OF VIEW OR OPIN-IONS STATED DO NOT NECESSARILYREPRESENT OFFICIAL OFFICE OF EDU-CATION POSITION OR POLICY.

A PUBLICATION OF

THEPOPULATIONCOUNCIL

by ISMAIL SIRAGELDIN and SAMUEL HOPKINS

This paper proposes a new design for government family planning programs. Thedesign is intended particularly for low-income countries that seek large and rapidreductions in fertility. The authors, Ismail Sirageldin, Ph.D., and Samuel B.Hopkins, J.D., M.P.H., were formerly chief advisor and associate advisor,respectively, at the West Pakistan Research and Evaluation Center, Lahore,Pakistan. This center conducts training, research, and evaluation for the PakistanFamily Planning Program. The authors are now in the Department of PopulationDynamics, Johns Hopkins University, School of Hygiene and Public Health.Dr. Sirageldin, an economist, is also on the faculty of the Department of PoliticalEconomy at the same university.

This paper has evolved in large part from the authors' experience in Pakistan.The authors express their appreciation to all their Pakistani colleagues who con-tributed to this experience. Thanks also go to the many persons who gave theauthors comments on earlier drafts of this paper, beginning in April 1969. Specialthanks for comments that influenced the redrafting go to: Bernard Berelson,Parker Mauldin, and Lee Bean of the Population Council ; Tont Croley and LyleSaunders of the Ford Foundation; Paul Harper, Rowland Rider, John Kantner,and Melvin Zelnik of Johns Hopkins University; Sultan Hashmi of ECAFE,Bangkok ; Quentin Lindsey of the University of North Carolina; and Robert Bushand Willard Boynton, of the United States Agency for International Development.

The authors assume full responsibility for views expressed in this paper.

February 1972

"A journey of a thousand miles mustbegin with a single step.--an old Chineseproverb.

". . . It may take another journey of athousand miles to begin that first step in theright direction."--an advisor in populationplanning.

There is general agreement among eco-nomic and social planners that humanfertility must be reduced. About 81 per-cent of the people in the developing worldlive in nations whose governments havepolicies favorable to family planning or atleast provide funds that go into familyplanning efforts (Nortman, 1971). Also,in many developed nations contraceptivesand often other family planning servicesare accessible to much of the population.However, the present overall decline infertility falls far short of what is necessary.Furthermore, the authors feel that thefuture fertility decline will also fall farshort, unless better systems for contracep-tive services, including incentives, are de-

TABLE OFCONTENTS

Family Planning Programs: An Economic ApproachIsmail Sirageldin and Samuel Hopkins

Village Midwives in MalaysiaJ. Y. Peng, Nor Laily bte A. Bakar, and Artffin Bin Marzuki

Report of the Swedish Abortion CommitteeChristopher Tietze

"PERMISSION TO REPRODUCE THIS COPY-RIGHTED MATERIAL HAS BEEN GRANTED

(g) THE POPULATION COUNCIL, INC- 1972 " Hobart EllisPopulation Council

TO ERIC AND ORGANIZATIONS OPERATINGUNDER AGREEMENTS WITH THE US. OFFICEOF EDUCATION. FURTHER REPRODUCTIONOUTSIDE THE ERIC SYSTEM REOUIRES PER-MISSION OF THE COPYRIGHT OWNER."

17

25

28

17

velopod. Systems vary among differentfamily planning programs, and new sys-tems are being continually proposed. Thispaper proposes mill another system, andwe feel that a pilot trial will prove it effec-tive. (For a comprehensive review of pro-posals made before 1969 of new ways toreduce fertility, see Berelson [1969a;1969b; and 1969c1. For subsequent pro-posals, see Ridker [1969; 1971], Finnigan(1972j, Kangas [19701, and Pohlman[19711.)

Summary of the EconomicApproachThe important elements of the pro-

posed system arc listed below. Some ofthese elements have been proposed beforeby others. In particular, monetary pay-ments to womel: who avoid births orpregnancies have been proposed by Enke(1960), Balfour (1962), Leasure (1967),Spengler (1967), Mauldin (1967), Ehrlich(1968), Simon (1968), and Ridker (1969;1971). One basic difference between thepayments proposed in their reports andthe payments proposed in this paper isthat, in general, the payments proposedhere begin sooner after enrollment andare more frequent thereafter. In its detailsand its combination of elements this pro-posal is relatively unique.

1. Cash incentive payments are madeto eligible women who enron in theproposed incentive scheme and whoavoid having births, regardless of themethods they use, within the limits ofsocial and political Lezeptability. Thesewomen are called clients.

2. The cash payments to clients en-courage more use and more effectiveuse of traditional as well as moderncontraceptives.

3. Eligible women are women whoare married and less than age 40 andwho have one or more living children.There is no limit on how many childrena woman can have before her first en-rollment.

4. The avoidance of births is deter-mined by periodic superficial examina-tions for pregnancy. The examinationsdo not attempt to detect pregnancies ofless than four or five months' duration.Insofar as a client can establish that anydetected T _gnancy does not end in alive birth, she remains enrolled andeligible for cash payments.

S. The schedule of cash payments isthe same for all clients.

6. Payments to clients begin fourmonths after enrollment, and are madeevery four or more months thereafter.

7. Payments to clients increase insize with length of enrol;.atent.

Plani

Planningand

decisionmaking

Key.Inputs

=.). OutputsTwo-way flow of informationand administrative control.

FIGURE 1. Schematic Presentation of the System

_ irdikns-a-orntloi

Research.evaluation

andreporting

Oual tycontrol

andadminis-uation

Incentiveunits

(enrollmentand

pregnancytests)

1

rTrairdng,

educationend

rnuni-cation

;

Governmentservice units

Privateclinics andphysicians;indigenous

medicalpractitioners;

stores

8. Payments to clients provide momyto pay for family planning and maternaland child health services.

9. During the first four months ofenrollment, clients and clients' husbandsreceive free family planning services ingovernment clinics. Otherwise, the gov-ernment services are provided to clientsand nonclients at fixed prices. Theprices of services and client incentivesare set at levels that require the averageclient to use only a small fraction ofher incentive payments to purchasesuch services as IUD follow-up, con-ventional contraceptives, or pills.

10. Preventive maternal and thildhealth (MCH) services are provided ingovernment clinics at low prices foranyone in the population, except thatthese services are free for a client andher children during the first four monthsafter her enrollment.

11. Government clinic personnel re-ceive part of their sales receipts as anincentive. A time lag is introduced,however, between sales receipts and in-centive paymen(s (commissions) to per-sonnel, to allow checks on false report-ing.

12. A system of reporting and evalu-ation provides guidelines for short- andlong-term policy acdons, including esti-

18

2

BirthsTarget prevented

Population (programparticipation)

i11

1

i1

1 Purchases of

--........._21:77::_t servtaicrgeestbYi

rr'''

1

ipopulation

I

ii

1

mates of births prevented and systemcosts.

13. The system increases the marketand demand for family planning serv-ices and supplies in the private sectorby combining cash payments for non-births, regardless of method of birthprevention, with charges for familyplanning services in government clinics.

14. The system allows for "fiscai flex-ibility." The government subsidy of theprogram can be reduced or raisedthrough manipulation a the prices ofcontraceptives and the kyr.] of clients'incentives.

The system is summarized in a flowchart in Figure I. The solid arrows givethe paths of the system's inputs such asincentive payments, family planning serv-ices, training, and educational activities.The broken arrows indicatct the system'soutputs. The outputs are births preventedand sales of family planning services.The flow channels indicated by thin linesshow the interrelations between all partsof the system with respect- to administra-tive research, evaluation, report-ini, and two-way communication.

Our proposal is made with the intentionthat it will first be tested in pilot areas and

that many of the details presented herewilt be modified before implementation ona larger scale. The authors have prepareda detailed proposal for a pilot trial of thissystem in Pakistan. The proposed dura-tion of the trial is four years. The pro-posed target population is 100,030. Theproposal may be obtained on request.

Countries for Which the Systemis Most SuitableOur system is designed for countries

like Pakistan and India that have bothrelatively low per capita incomes andambitious fertility reduction goals. Othersystems should be more efficient in othercountries. Countries with high per capitaincomes can alter certain of their existingsystems, like taxation systems (Barnet,1969), without having to create a new in-centive system, and in high income coun-tries, where most adults pay substantialtaxes, the distinction made in our systembetween -free" and "charged for- govern-ment services is probably unimportant.Countries with very modest goals for fer-tility reduction probably can achieve thesegoals without the 'incentive system pro-posed here, provided that they can sub-stantially increase the availability and at-tractiveness of contraceptive services. Onthe other hand, achieving these modestgoals will not be enough, in most cases, toachieve the necessary reduction in fertility.

Provision of ServicesPREMISES

it is one of the authors' premises thatcharging prices for family planning serv-ices can improve continuation ratesamong family planning acceptors. (Inaddition, charging prices may indirectlyand in the long run improve acceptancerates. However, for increasing acceptancerates the proposed system relies primarilyon client incentives.) A person who de-cides to pay for a service must havethought about it and its utility moreseriously than a person who is getting thesame service free. The cost of not using ormisusing a service is necessarily higher forthose who haye paid for it then for thosewho have received it free, because the costof a purchased item includes the foregoneutility of those goods and services (in-cluding the foregone pleasure of saving)that could have been bought and con-sumed instead. We are assuming that thecost of time spent in acquiring and con-suming these goods and services is thesame whether they are provided free or

not. Also, wc are assuming that no utilityor pleasure is acquired from destroyingfree goods or services.

Another premise is that setting a targetfor new acceptors while services are pro-vided free encourages poor quality ofservice and false reporting.

PLAN FOR SERVICES

In the proposed system, services will beprovided by both the public and privatesectors. In the public sector preventiveMCH services are provided in govern-ment service units (clinics). Family plan-ning services include any method of birthprevention that the government willauthorize, such as IUD insertion, steriliz-ation, and oral and conventional contra-ceptives. The services also cover relatedfollow-up care, including counselling anddispensing of aspirin and vitamins for sideeffects. For treatment of serious sideeffects or for relatively difficult newservices, such as a new method of vasec-tomy, patients are referred to specialservice units or other sources of medicalserviceshospitals and private practi-tioners. MCH services include prenataland postnatal maternal care, immuniza-tions, and health education. (Combiningfamily planning services with healthservices in the same program service unitdoes not imply that a family planning pro-gram should be administered by thehealth department or vice versa. There isa limit, however, to the extent to whichdiversification is advisable and feasible.Provision of additional services will re-quire more resources and personnel. Theadded social cost has to be measuredagainst the added social benefit includingthe added value of any increase in birthsprevented.)

The prices of the various services arefixed by the planning unit and are sub-ject to change. The prices depend on de-mand situations (for example, participa-tion rates relative to anticipated targets),on cost considerations, and on the extentof government subsidy. The price struc-ture also may reflect the planners' prefer-ence fer promoting one or another familyplanning method.

The one exception to charging prices isfor women who have recently enrolled inthe incentive scheme. Clients and theirhusbands and children receive free servicesduring the first four months of the clients'enrollment. Reasons for this are: (1) Theclients have not yet received their firstincentive payment. (2) An initial periodof free i;ei -;:ee encourages service unit per-

19

sonnel to give emphasis to follow-up care.(3) Free service for a itmited time en-courages clients to accept some services atleast on a trial basis.

Each service unit serves a target popu-lation in a limited geographical area. It isproposed that service units have not onlyadequate equipment, facilities, and trainedpersonnel but also accommodations formedical and paramedical staff, especiallyin rural areas where it is difficult to findadequate private acconnnodation withinreasonable traveling distances. Because ofthe shortage of medical personnel in de-veloping nations, one medical doctorsupervises several service units, and para-medical personnel are used extensively.Thus, for the most part, only those MCHservices that can be performed by para-medical personnel are provided in theservice units.*

MCH and family planning services pro-vided by the private sector are also con-sidered part of the system, because thesystem should have important effects onthe demand for and supply of theseservices. The system will provide educa-tion, training, information, and othersupport and concessions that will helpdevelop the private sector's role in pro-viding services.

Client IncentivesPREMISES

It is the authors' premise that smallbut immediate incentive payments or re-wards for nonbirths can create a strongdemand for family planning services. 1m-medi..ate payments should have a greatimpact in a developing society where thetime horizon is short, and regular pay-ments should have a reinforcement effecton client participation.

Another premise is that monetary in-centives for nonbirths, regardless of themethod used, combined with charges forfamily planning services in governmentclinies, will increase the market for andsupply of family planning services in theprivate sector.

The child care incentive might be adminis-tered with the cooperation and assistance ofcertain other welfare programs. The most obvi-ous possibility is the UNICEF program. Butany program that, for example, promotes betternutrition would be suitable. Ruben Rausing hasrecently proposed a nutrition program to theInternational Bank for Reconstruction. Theprogram would use the milk protein and edibleoil surpluses of developed nations, and Mr.Reusing describes some ways that the milk dis-tribution might promote family planning. (Theproposal is dated February 1969 and has notitle.)

THE INCENTIVE UNIT

An inCentiVe Unit is reSpOnSible lot en-rolling Clients in a limited geographicalarca, checking pregnancy status, payingincentives, keeping records, and reportingto the control and planning units (seeFigure 1). Although there are good rea-sons to separate the incentive and serviceunits, we propose an economical adminis-trativ: set-up in which the incentive andservice units are located in the sante build-ing, which has a designation like "FamilyHealth and Welfare Center." Pregnancychecks are made by service unit para-medical staff. The incentive unit staff ad-minister incentive par; Lents and verifythe identity of clients. The main staff ofthe incentive unit are a motivator and anassistant. This administrative pattern hasseveral advantages. It should reduce theoverhead cost. Also, since the client cancome to the same building for MCH andfamily planning services as for incentivepayments, the client can keep her enroll-ment private and can save time. Combin-ing the units may increase the probabilityof collusion between personnel of the twounits, but certain administrative measures,such as relatively frequent shifting of per-sonnel from one location to another,should keep the collusion under control.

ELIGIBILATY CRITERIA

Eligibility criteria should vary accord-ing to the norms, needs, and experiencewith the system. In general, conditions foraccepting clients into an incentive pro-gram, aside from administrative, political,and economic feasibility, can be designedto enhance client acceptance, discourageintermittent enrollment, and screen outsterile women or women with low ex-pected fertility. The proposed eligibilitycriteria are that a woman must:

I. Be married.2. Have at least one liv ng child.3. Be less than 40 years of age.4. Not have discontinued enrollment

for more than two years or, if thediscontinuation was fbr less thantwo years, not have discontinuedmore than twice. Discontinuation isdefined as a gap of more than fourmonths between examinations thatreveal no pregnancy.

The rnaritat and parity status criteriacould be relaxed depending on prevailingnorms and needs; for example, the maritalstatus condition may be relaxed in asociety where illegitimate births are a

problem that the society is willing to dealwith directly. It should be emphasizedthat restricting clients to women is notdesigned to discourage m, le methods ofbirth control. It is only an administrativeconvenience. Sterilization and contracep-tion could be subsidized to varying de-grees for clients' husbands as well as forclients, depending on the desire of the pro-gram planners to promote one or anothermethod.

Criteria 2 and 3 serve to screen outwomen with low expected fertility. Givencriterion I, the system does not encouragepromiscuity. Criterion 4 allows a womanto have two births (with few exceptions)after her first enrollment in the programwithout losing her eligibility for re-enroll-ment. In a high fertility population thislimitation provides for more restriction onexpected fertility than may appear on firstimpression. Take, for example, the Pakis-tan population of married females, age15-39 with one or more living children.Using POE data* for the period 1962-1965 and PGE assumptions about :m--tility change, about the percent of thepopulation that has living children, andabout the percent of total births that are towomen with living children, it is possibleto estimate that the average number oflive births per female was about four atany time during thc 1962-1965 period. Ithas also been estimated that these femaleswould, on the average, have about fouradditional live births before reaching age40. (These estimates are admittedly veryapproximate, because their calculation re-quired somewhat arbitrary assumptionsto adjust the data used. Accordingly, theestimates are rounded off to the nearestwhole birth.) Thus, if this system wereintroduced into such a population, theimmediate reasonable target would be a25 percent reduction in the average com-pleted number of live births (six insteadof eight), assuming that the clients wererepresentative of the total female popula-tion at risk of pregnancy. Furthermore,the longer the program is in operation,the higher is the proportion Of eligiblewomen who could have enrolled afterhaving only one child. Therefore, thecompleted fertility target in Pakistancould approach three instead of six livebirths per woman.

* PGE (the Population Growth EstimationProject) collected vital events data by registra-tion and survey procedures during 1962-1965 insamplc areas in Pakistan. PGE reported twoestimates for most data. The high estimates wereused in this paper. The source used was WilliamSeltzer (1968).

20

PREGNANCY TESTS

Women will receive only superficialphysical examinations by paramedicalpersonnel to detect pregnancies of four ormore momhs duration. Medical person-nel are called upon only if there is dis-agreement between the client and theparamedical staff. In these cases, clientsmay be asked to return after one or twomonths when the pregnancy, if any, willbe more apparent. (Eor a discussion ofpregnancy tests, see Cabrera [1969].)

CASH INCENTIVES

The expected benefit of keeping awoman from having a birth during a yeardepends on the probability that thewoman will have a live birth during theyear. It is tempting to suggest that thereshould be differential payments based onthe different probabilities. Wc think, how-ever, that the administrative cost of suchselective payments is greater than the ex-pected benefit and that the social cbst, forexample, clients' resentmeat, may also belarge. We propose to pay, at least initially,the same amount of incentives to allclients for the same period of enrollment.However, eV.imates of differential proba-bilities both for the population at largeand for system clients are essential foradequate evaluation and planning.

Incentive payments increase with lengthof enrollment. Payments are scaled togive both frequent rewards for continuingenrollment and penalties for intermittentenrollment. Part of the total annual pay-ment is given in triannual installments,and part is given as an end-of-the-yearbonus. Moreover, the total annual pay-ment is smaller in the first year of con-tinuous enrollment than in the secondyear and smaller in the second year thanin subsequent years. The proposed pay-ments per four-month period of continu-ous enrollment for a country like Pakistanor India are as follows:

PeriodPayment (at end

of period)

onths 1-4) 50.802 (months 5-8) 1.603+ (months 9 plus) 2.00

Amounts are expressed in U.S. dollarsfor illustration only. Payments are madein local currency.

When period payments are supple-mented by end-of-the-year bonuses and

when a client's enrollment is continuous, to give better quality of s rvice and morethe total payments per year are as follows:

Period End of yearYear payments bonus Total°

I 54.40 52.60 57.002 6.00 3.50 9.503 ± 6.00 5.00 I 1,00

a Based on per capita national income corn-parir ons, the value of one dollar in Pakistan or1 ndia is roughly tquivalent to the value of about540 in the USA Thus, the 511.00 annual incen-tive payment suggested in the text is equivalentto the payment of about 5400 in the USA. Thiscomparison was suggested by Richard Reynolds,Ford Foundation. Pakistan.

After any period of discontinuation, ifa woman is still eligible to re-enroll anddoes re-enroll, she must begin again at thebottom of the incentive scale.

It must be emphasized that these pay-ments, although reasonable for a countrylike Pakistan or India, are only a startingpoint for a pilot project designed to esti-mate clients' responses to different typesof incentive payments and different pay-ment scales.

PROCEDURE FOR PAYING INCENTIVES

Payments are made once every fourmonths. As mentioned earlier, a client isasked to come to the incentive unit atleast once every four months for a preg-nancy cheek. At this time she. will receiveher period payment if she is still eligible.The visits every four months are not sofrequent as to be a burden on clients butare still frequent enough to detect preg-nancies by superficial examination.

Alternatives to immediate paymentsand cash payments may be tested. Incountries where post offices function assavings banks and in areas of these coun-tries where post offices are reasonablyaccessible, one alternative would be tomake payments with nontransferablecoupons equivalent to cash only when de-posited in a post office savings accountwhich the client could use only for de-posit of these coupons. Delays in paymentcould then be effected by requiring mini-mum balances before or after with-drawals. This alternative would have theadded value of encouraging saving.

Incentives for Service UnitPersonnel

PREMISE

Another premise of this system is thatincentives for personnel can be designedto improve the quality of services as wellas the quantity. In particular, the premiseis that clinic personnel will be motivated

service if:

I. They have adequate training and facili-ties.

2. They retain part or their clinic's grossreceipts as incentives.

3. These incentives are earned not frominitial service to clients hut mainlyfrom follow-up service.

PLAN r OR INCENTIVES

Service unit personnel receive a portionof their unit's gross earnings from sales offamily planning services. The paymentsare made at the end of the year after theusual accounting and inventory check. Itis proposed that initially the portion be90 percent, subject to adjustment by theplanning unit, depending on field ex-perience.

Foilow-upThe system is uniquely suited to achiev-

ing good follow-up care of women whoadopt contraception. The pregnancychecks of clients at four-month intervalsprovide an ideal opportunity for follow-upcare, and the incentives for service unitpersonnel encourage them to persuadeclients to return for follow-up.

The system also includes follow-up inthe field of at least a sample of the follow-ing types of eligible women to determinetheir reasons for not being enrolled:women who have terminated their enroll-ment by being overdue for a pregnancycheck; women who have previously ob-tained family planning and/or MCI-Iservices at service units but who have notenrolled; and women who have nevereither enrolled or visited service units.

Training, Education, andCommunicationIn-service trAining is often ineffective

because, among other problems, there islack of interest among the trainees. Underour system, personnel should show moreinterest in in-service training. In thissystem personnel incentives must comefrom the client's rather than from govern-ment's pocket; therefore, personnel mustgive better service in order to attract ',hesame number of clients as before.

Part of the function of the service unitpersonnel is to explain the system of in-centives and services to the target popula-tion, as well as to motivate this populationto enroll. Also, incentive unit personnelenumerate households in their designated

21

5

areas, and keep records on every house-hold. Such enumeration helps the plan-ning and execution of motivation andeducation programs.

Research, Evaluation, andReporting

IN P U TS

Data on inputs are continuously gath-ered and include data on expenditures on,and utilization of, such items as incentivepayments, contraceptives, other medicalsupplies, personnel, transport, and adver-tising. The purpose is to obtain a con-tinuous picture of the system's cost bytype of input.

OUTPUTS

The primary purpose of the output re-se-rch, evaluation, and reporting is theestimation of how many births the systemprevents. Births prevented are the finaloutput. There is also, however, assessmentof what may be called intermediate out-put. This output includes the number ofclients, the proportion of all eligiblewomen who are clients, and the rates ofcontinuation of enrollment and contra-ceptive use. The assessment of this outputwill include analysis of the numbers, pro-portions, and continuation rates hy demo-graphic and other characteristics of thewomen and analysis of how these outputvariables change over time. This assess-ment of intermectiate output may indicate,for example, that the system is not effec-tively reaching, enrolling, and retainingwomen of low age and parity. As a result,there may be experimentation with newtechniques of education, publicity, andmotivation. At the same time there maybe revision of the incentive design. Womenunder age 30 with one or two childrenmight be offered double the monetary in-centives offered to other clients.

Intermediate output data are gatheredby enumeration of the target populationand by the incentive uni7 reporting pro-cedure. The incentive unit reports eachinitial and follow-up visit on a simplerecord form. (An initial visit is either afirst enrollment or a re-enrollment after aperiod of nonparticipation, which re-quires, as noted earlier, that the clientbegin again at the bottom of the incentivescale. More details about the reportingsystem, specific to Pakistan, are given inthe authors' project proposal for a pilottrial in Pakistan.) For each initial visit therecord form reports the usual backgrounddata, including age, parity, open interval,and method that the woman is using or

plans to use to prevent pregnancy. Foreach repeat visit, the form reports themonths of continuous participation andthe method, if any, being used. These dataare sent to the research arld evaluationunit.

The estimation of final output (birthsprevented) requires fertility rates for thetarget population, in addition to the aboveintermediate output data. Fertility rateswill be estimated by surveys and/orsample registration. These methods havebeen tried in Pakistan with some success(Naseem, 1971; Mosley, 1968).

The estimate of births prevented duringa given period is related to incentives paidand other costs incurred during a periodof the same duration but ending ninemonths earlier. COSI per birth preventedis calculated both for total clients and forclient subgroups.

In addition, as noted earlier, attentionis given to the eligible women who ter-minate enrollment or who have neverenrolled. Sample surveys of these womenevaluate reasons for nonenrollment.

Cost-Benefit EstimatesNEED FOR ESTIMATES

Is the system proposed in this papermore economical (in terms of cost benefitcriteria) than other family planning sys-tems and other alternative investments?Could a typical developing country affordsuch a system both in the short and thelong run? What are the social costs andbenefits of introducing such a system, andare they quantifiable and measurable?These are some of the questions that needto be at least partially answered before theproposed system can be recommended forimplementation on a large scale. Many ofthese questions, however, cannot beanswered a priori, Rather, pilot trials arenecessary. (For a recent review of the liter-ature relevant to cost-benefit analysis, offamily planning programs, see Robinsonand Horlacher [1971]); for estimates of thecost-effectiveness of six national familyplanning programs, see Department ofEconomics, Pennsylvania State University[1969].)

PRELIMINARY COST-BENEFIT ESTIMATESFOR DETERMINING A FEASIBLE LEVEL

FOR CLIENT INCENTIVES

The following estimates are for Pakis-tan.

Cost: We estimate that the incentivesproposed earlier (pp. 20-21) and the othercost elements of our system will besufficient to prevent many births but yet

inoderwe enough to keep the cost of pre-venting each birth in the range of US$85to $110 on the average.

Benefit: Using the increase in percapita GNP as the measure of benefit, aconservative estimate of thc benefit de-rived from preventing each birth is $180on the average.* (This estimate is for amagnitude of fertility reduction in therange of 20-50 percent. Also, note that$180 is the present value of the benefit.The benefit is spread out over futureyears in the form of higher per capitaGNP, unlike the cost which is incurredimmediately. The future benefits must bediscounted to arrive at an estimate ofbenefit that can be meaningfully com-pared to the estimate of cost [Robinson,1971]).

Rate of return: Comparing the aboveestimates of benefit and cost permits thecalculation of a rate of return on themoney invested in the proposed system,If $85 must be invested per birth pre-vented, the net benefit is $95, that is,$180 -S85, and thc ratc of return is

about i 12 percent, that is, (S95/585) X100. Similarly, if $110 must be investedper birth prevented, the net benefit is $70,and the rate of return is about 64 percent.These rates of return are clearly higherthan those from alternative investmentsdesigned to increase GNP. (Alternativereturns are probably less than 30 percent.)

Feasible levels for client incentives:Given the difference between the rates ofreturn from the proposed system andfrom alternative investments, we concludethat the proposed level of client incen-tives, as well as other system costs, arewell below the maximum we could justify.This permits experimentation with rais-ing, as well as lowering, the incentives andother inputs.

Other benefits: The above estimate ofbenefit does not include several kinds ofbenefits which the system is likely to pro-ducefor example, improvements in ma-ternal and child health, in child develop-

* Several persons have published estimates ofthe value of preventing a birth under particulareconomic conditions. The estimate of $180 forPakistan is about twice the current Pakistan percapita income. This estimate is suggested by therecent worl of Stephen Enke and Richard Zind(1969), usil.g a dynamic demographic-economicmodel of a typical less-developed country. An-other economist, however, gives evidence thatthe value of a birth prevented in Pakistan maybe much higher: in a study of India, Simmons(1971) felt the best estimate of the current mar-ginal value of a birth prevented during a 20 per-cent to 30 percent reduction in fertility was Rs.7800 or about 14 times India's per capita incomein 1967-1968.

rnent, and in other wel are variables re-lated to birth spacing, limitation, andsmaller family size.

Asswnplions behind the cost estimates:The cost estimates for the proposedscheme are not minimum estimates. Theyprovide for some wastage and inefficiency.For example, they assume that a sig-nificant portion of the women who enrollin the incentive payment scheme will notactually try to prevent pregnancies. Theestimates also allow for higher costs forsupplies, personnel, and transportationper unit of target population than therearc in the present Pakistan program. Onthe other hand, it is assumed that thesystem will operate at a higher percent ofcapacity than the present Pakistan pro-gram.

DiscussionPROBLEMS OF MEASUREMENT

Many of the difficulties of evaluatingfamily planning programs have been welldocumented. (For a recent, comprehensivearticle on measuring fertility change, seeSeltzer [1970].) These difficulties should bereduced somewhat by certain elements ofour system such as the frequent (everyfour months) checks oa the status of whatshould be a very large portion of allpersons practicing family planning in theprogram area.

Accurate estimation of the return onincentive payments is quite complex.Introduction of incentive payments into asystem changes the relationships amongall other inputs and their net productivi-ties. Furthermore, the net value productof clients' incentives is affected by pastand present levels of utilization of allother inputs. (Net value product and netproductivity mean the net addition to thesystem's output [births prevented] as a re-sult of increasing a specified input, such asincentives, and none other.)

In order to meaningfully assess thetotal return from investment in the sys-tem, the different inputs and the outputsmust be defined and measured unambigu-ously with reliability and continuity. Forexample, it is necessary to prorate the costof those inputs, such as medical personnel,that are used jointly by the proposed sys-tem and by other operations like govern-ment curative health services. Attentionmust also be given to fixed overhead costsand to foreign aid, both technical andfinancial, as inputs to the program.

Rigorous estimation of births pre-vented is complex and involves many un-certainties, especially in developing na-

tions. Aceura a: estimates require a varietyof data that arc difficult to obtain and ifobtained, are usually unreliable (Seltzer,1970; Potter, 1969; Bean and Seltzer,1968; Ross, Stephan, and Watson, 1969;Mauldin, 1967; Byung and Isbister, 1966).William Seltzer (1970) has even suggestedthat the uncertainties are so great that theestimates should not, be made at all. It isthe authors' opinion, however, that in theproposed system the estimates of birthsprevented are justified. The uncertaintiesin the estimates can he fairly reflected bystating estimates or both births preventedand rates of return in terms of ranges.Furthermore, estimates of births pre-vented will (or should) be necessary toobtain and maintain government supportof the system, because estimating the rateof return from the system is necessary tocompare the value of the system to thevalue of competing development pro-grams. Although the estimate of the rateof return may be very uncertain, it shouldbe remembered that the uncertainties inestimating the rates of return for manyother developmental programs are equallygreat. Finally, some creditable estimatesof births prevented have been made re-cently (Chang, 1969; Haynes, 1969; Leeand Isbister, 1966).

With respect to components of the sys-tem's rate of return other than births pre-vented, there are indirect costs and bene-fits of the system that should be measured.For example, the new cash flows into theeconomy via the incentive paymentsshould have desirable effects on both themicro and macro levels by increasing themonetized (versus barter) sector of theeconomy. At the same time, there may besome inflationary effects.

PARTITION OF THE SYSTEM

All parts of the proposed system or anyother family planning program have im-portant interrelationships and interac-tions. As a result, the merits of any onepart such as monetary incentives cannotbe viewed in the abstract. Whether mone-tary incentives to women who preventbirths can work depends in this proposalnot only on the amount of the incentivebut on how the rest of the system affectsthe quality of contraceptive services, thepopulation's perception and valuation ofthose services, and the service personnel'sconception of what efforts in their workare in their best interestfor example,emphasis on continuation rates. In short,the authors feel that pilot trials wouldshow that the proposed client monetary

incentives are a more profitable invest-ment as part 01 the proposed system thanas additions to family planning programsas they now exist in countries like Indiaand Pakistan.

Cost. or 'rue Sysl EstOur proposed system may be expensive

relative to some existing family planningprograms. It is not meaningful, however,to compare our system to most existingprograms. The proposed system is neithernecessary nor appropriate for nationswhere the goal of the family planning pro-gram is only a modest reduction in thebirth rate. Other less complicated systemscan probably achieve modest goals at lesscost. On the other hand, these less costlysystems arc incapable of ever achievingmore ambitious goais. The fertility reduc-tion goals in many nations are flat modest.It may be possible to use a simple systemand spend only $20 per birth preventedwhen reducing the birth rate in a givencountry from 50 to 40. But to reduce thebirth rate in the same country from 40 to20 will require a different and perhapsmore complicated system that may cost$100 per birth prevented. Speed in achiev-ing a fertility decline also mcrits a highercost per birth prevented.

When a country's fertility rate has de-clined substantially without the use of in-centives, the cost of using the proposedsystem to prevent additional births will behigh. The cost will be high because: (1)relatively large incentive payments may benecessary to motivate additional womento adopt contraception; and (2) a rela-tively large number of women will have toreceive payments per additional birth pre-vented, because many women who areeligible for payments will have alreadyadopted contraception, and their con-tinued practice will not prevent addi-tional births. A modification of the eligi-bility requirements in the proposed sys-tem, however, could reduce this cost.

It was observed in a recent report onthe national family planning programs ofKorea and Taiwan that: "Both countrieshave become increasingly aware that, tokeep the birth rate going down, they mustinvest more money than before" (Ross,Han, Keeny, and Cernada, 1970).

Summary and RecommendationsIn this paper we have outlined a system

of providing family planning incentivesand related services. Incentives are givento enhance acceptance rates and continua-tion rates. There are both direct and in-

;3

direct incentives for this purpose. Thedirect incentives are monetary paymentsto eligible women (married, under age40, with one or more children) for as longas they avoid having births. Direct incen-tives also include free or subsidized familyplanning and MCH services for a limitedtime period. The indirect incentives arepayments to medical personnel of a por-tion of the prices they charge for providingservices. The prices charged are also in-direct incentives to the clients to value theservices and thereby show high continua-tion rates. Service statistics are designedto give direct measures of system inpufsand outputs during specific time periodsin terms of births prevented and cost perbirth prevented.

Whether on balance the system's frame-work of incentives, checks, responsibili-ties, and chain of command improves onpresent family planning programs shouldbe tested by pilot projects in several coun-tries simultaneously for per iods of at leastthree years. Field experiments should bedesigned to test the behavioral assump-tions as well as the administrative andfinancial feasibility of the system. Thepilot project population might be limitedto families of persons employed in certaininstitutions, such as government, largepublic enterprises, or large private indus-tries, in addition to limiting the populationby geographical area.* Also, some of theother proposals for improving familyplanning programs and this paper's pro-posal are not mutually exclusive. For ex-ample, "family planning bonds" thatserve as old age security (Balfour, 1962;Ridker, 1969) and non-monetary incen-tives like free schooling for children inone- or two-child families might be addedto the proposed system.

ReferencesBalfour, Marshall C. "A Scheme for Reward-

ing Successful Family Planners." Memo-randum. The Population Council, June1962.

Barnett, Larry D. -Population Policy: Pay-ments for Fertility Limitation in the UnitedStates." Social Biology 16, no. 4 (December1969): 239-248.

Bean, Lee L., and William Seltzer. "CoupleYears of Protection and Births Averted: AMethodological Examination." Demogra-phy 5, No. 2 (1968): 947-959.

* The idea of limiting the trial to families ofemployees in certain institutions was suggestedto the authors by Ronald Freedman of theUniversity of Michigan.

Berelson, Bernard. 1969a. -Beyond FamilyPlanning." Studies in Family Planning I,no. 38 (February 1969): 1-14.

1969b. "National Family PlanningPrograms: Where We Stand," Studies inFamily Planning 1, no. 39, supplement(March 1969): 343.

ed. 1969c. Family Planning Pro-grams: An International Survey. New York ;Basic Books.

Bogue, Donald J. 1968. "Progress and Prob-lems of World Fertility Control." Demog.raphy 5, no. 2 (1968): 539-540.

Byung Moo Lee and John Isbister, "The Im-pact of Birth Control Programs on Fer-tility." In Family Planning and PopidationPrograms. Edited by Bernard Berelson andothers. Chicago: University of ChicagoPress, 1966.

Carrera, Hugh A. -Evaluation of PregnancyTests." Anwrican Journal of Obstetrics andGynecology 105, no. 1 (January 1969):32-38,

Chang, M. C., T. H. Liu, and L. P. Chow."Study by Matching of the DemographicImpact of an IUD Program: A PreliminaryReport." The Milbank Memorial FundQuarterly 47, no. 2 (April 1969): 137-157.

Department of Economics, The PennsylvaniaState University. -A Cost-EffectivenessAnalysis of Selected National Family Plan-ning Programs." A Report on Phase 11 ofthe Penn-USA1D Population Project "Cost-Benefit and Cost-Effectiveness Evaluationof Family Planning Programs" Contractno. AID/esd-1884. (nlimeo), UniversityPark, Pennsylvania, December 1969,

Ehrlich, Paul R. The Population Bomb. NewYork: Ballantine Books, 1968 (revised1971).

Enke, Stephen. "The Economics of Govern-ment Payments to Limit Population."Economic Development and Cultural Change8, no. 4, Part 1 (July 1960): 339-348,

Enke, Stephen and Richard Zind. "Effect ofFewer Births on Average Income." Journalof Biosocial Science 1, no. 1 (January 1969):41-56.

Finnigan, Oliver D. "Planning, Starting, andOperating an Educational Incentives Pro-gram." Studies in Family Planning 3, no. 1(January 1972): 1-7.

-Governmental Policy Statements on Popu-lation: An inventory." Reports un Popula-tion/Family Planning. (February 1970):1-20.

Haynes, M. Alfred, George E. Immerwahr,Aleyamma George, and P. S. J. Nayar, "AStudy on the Effectiveness of Sterilizationsin Reducing the Birth Rate." Demography6, no. 1 (February 1969): 1-11.

Kangas, Lenni. W. -Integrated Incentives forFertility Control." Science 169 (3952):1278-1283. September 1970.

Leasure, J. William. -Some Economic Bene-fits of Birth Prevention. Milbank Me-morial Fund Quarterly 45, no. 4 (October1967): 417-425.

Mauldin, W. Parker. 1967a. "Births Avertedin Family Planning Programs," Stndies inFamily Planning 1, no. 33 (August 1967):1-7.

. 1967h, "Prevention of IllegitimateBirths: A Bonus Scheme.-,Memorandum.The Population Council, August 1967.

Mosley, Wily H., A. K. M. Alauddin Chowd-hury, K. M. Ashraful Aziz, M. ShafiqulIslam, M. Fahimuddin. "DemographicStudies in Rural East Pakistan. Pakistan-Seato Cholera Research Laboratory. June1968.

Naseem, M., I. Farooqui, and G. M. Farooq,"Final Report of the Population GrowthEstimation Experiment: 1962-1965."Dacca: Pakistan Institute of DevelopmentEconomics, July 1971.

Nortman, Dorothy. "Population and FamilyPlanning Programs: A Factbook." Reportson Population1Family Planning, no. 7

(June 1971).

Pohlman, E. Incentives and Compensations inBirth Planning. Chapel Hill: CarolinaPopulation Center, 1971.

Potter, Robert. "Estimating Births Averted ina Family Planning Program." In Fertilityand Faintly Planning: A World View.Edited by 5, J. Berhman and others. AnnArbor: University of Michigan Press, 1969,pages 413-434.

Ridker, Ronald G. 1969. "Synopsis of aProposal for a Family Planning Bond."Studies in Family Plaiming 1, no. 43 (June1969): 11-16.

. 1971. "Savings Accounts for Famil%Planning, Au Illustration front the TeaEstates of India.- Studies in Family Plan-ning 2, no. 7 (July 1971): 150-152.

Robinson, Warren C. and David E. Hor-lacher. -Population Growth and EconomicWelfare." Reports on Population/ FatnryPlanning, no. 6 (February 1971).

Ross, John A., Frederick F. Stephan. andWalter B. Watson. A Handbook for ServiceStatistics in Family Planning Programs.New York: The Population Council. 1969.

Ross, John A., Dae Woo Han, S. M. Keeny,and George Cernada. "Korea/Taiwan1969: Report on the National FamilyPlanning Programs." Studies in FamilyMaiming I, no. 54 (June 1970): 15.

Seltzer, William. 1968. "Benehwork Data forPakistan: A Review of Summary EstimatesDerived from the PGE Experiment." Re-search Report No. 66. Karachi: PakistanInstitute of Development Economics.

. 1970, "Measurement of Accomplish-ment: The Evaluation of Family PlanningEfrorts. Studies in Family Planning 1, no.53 (May 1970); 9-16.

Simmons, George B. The Indian Investment inFamily Planning. New York: The Popula-tion Council, 1971.

Simon, Julian "Money Incentives to ReduceBirth Rates in Low-Income Countries:A Proposal to Determine the Effect Ex-perimentally. Manuscript. Chicago: Uni-versity of Illinois, 1968,

Spengler, Joseph J. -Agricultural Develop-ment Is Not Enough," Manuscript pre-pared for Conference on World PopulationProblems, Indiana University, May 1967,pages 29-30,

Those interested in obtaining the pro-posal for a pilot trial of this system inPakistan may write to the authors at;

The Johns Hopkins UniversitySchool of Hygiene and Public Health615 North Wolfe StreetBaltimore, Maryland 21205U.S.A.

Village Midwive, in Ivialaysia

by J. Y. PENG, Noa LAILY BTE A. BAKAR, and ARIFFIN BIN MARZUKI

This paper presents the results ofan Mfornull questionnaire administered to villagemidwives in Malaysia recruited ,for a training program. Dr. Peng, M.D., D.P.H.,is an assistant professor of Population Planning, Center for Population P17-oing,School of Public Health, University of Michigan, Ann Arbor, Michigan, L ..). A.;Dr. Nor Lady, M.B.B.S., is chief of the Training Division and Dr. An/fin,M.B.B.S., F,R.C.O.G,, is director general of the National Family Planning Boardin Kuala Lumpur, Malaysia.

This study was undertaken as part of the continued collaboration between theNational Family Planning Board of the Government of Malaysia and the Center

jOr Population Planning of the University of Michigan. Thanks are due to thestaff of the National Family Planning Board and the Center for Population Plan-ning for help in collecting and processing data.

Most Malaysian officials in health andfamily planning agree that kampongbidans, the traditional village midwives,should be used to promote maternal andchild health and family planning in ruralcommunities. Although these women haveno formal training, they are active andinfluential in their communities and theirparticipation in these programs wouldlessen the burden on the rural health staffthat will arise from the integration offamily planning services into rural healthactivities. But important questions arise.Are kampong bidans interested in pro-moting family planning? What functionscan they perform? How can thcy be paidfor participating in maternal and childhealth and family planning services?

To determine their background andtheir interest in participating in familyplanning programs, we interviewed 292bidans using a standard questionnaire.The sample consisted of bidans whowere about to attend a three-week trainingprogram in health and family planningoffered by the Malaysia National FamilyPlanning Program and their responsesmay not be representative of the attitudesof all midwives in Malaysia. Nonetheless,the findings were encouraging: The bidansexpressed positive attitudes toward pro-viding family planning services for mar-ried women. Almost none expressedanxiety that government family planningservices might interfere with their jobs.Most expressed willingness to participatein the government program.

BackgroundROLE OF MIDWIVES IN ASIA

IR Asian countries, particularly in ruralareas, village midwives with no formaltraining play an important role in attend-

ing deliveries. In several countries, effortsare being made to integrate these mid-wives in family planning programs. Thegovernments of India and Pakistan havedeveloped incentive systems to enlist theparticipation of midwives in maternal andchild health and family planning pro-grams on a large scale (1-5). In Thailand,village midwives were used to recruit IUDacceptors under an incentive scheme inthe Potharam project, in the pioneeringstage of the Thai family planning pro-gram. By 1968 about 16,000 midwives hadbeen trained for this program withUNICEF assistance (6). In the Philip-pines, in another program with UNICEFsupport, traditional midwives were trainedto assist in provision of maternal andchild health services. In Indonesia, plansto enlist midwives' participation in familyplanning are underway.

The number of traditional midwives,their social and demographic character-istics, their methods, and their attitudestoward family planning are areas in whichresearch is needed to determine how mid-wives can best contribute to family plan-ning services. A number of countries haveconducted studies along these lines,among which are: a rather intensive studyon dais in West Pakistan by Gardezi andInayatullah (1); a study in East Pakistanby Croley, et a/. (7); and a study in Indiaby Bhandari covering 50 dais in Najafgarhand Ujwa (8).

ROLE OF MIDWIVES IN MA AYSIA

There are an estimated 3,000 traditionalvillage midwives or kampong bidans inrural Malaysia. (Kampong means villageand bidan means midwife.) According toa report by the Maternal and ChildHealth Committee of the National Health

25

9

Council (9) about 174,000 births or 57percent of all deliveries in 1964 wereunder medical supervision, e.g., throughgovernment hospitals, health center, non-government nursing homes, and clinics atrubber estates and tin mines. The remain-ing 43 percent of all deliveries were at-tended by private midwives almost all ofwhom were kampong bidans. Accordingto the West Malaysian Family Surveyconducted in 1966-67, 39 percent of ruraland 31 percent of all respondents' lastlive births were attended by kampongbidans (10). Although the Ministry ofHealth is producing trained midwives andit is expected that in most areas kampongbidans will be replaced by trained mid-wives in about 20 to 30 years, currentlythe bidans play an important role in rela-tion to maternal and child health. Anissue for the government at present is

how to change the kampong bidans' rolefrom attending deliveries in a traditionalway to assisting in providing family plan-ning services and to referring mothcrs torural health centers for modern maternaland child health services.

Training ProgramThe Malaysia National Family Plan-

ning Program needs to expand its servicesto rural areas. The first step in this ex-pansion will be to integrate family plan-ning services with maternal and childhealth activities in rural health units (nowunder the Ministry of Health). The secondstep will be to utilize kampong bidans torecruit acceptors and to distribute contra-ceptives in remote areas where ruralhealth facilities do not exist. Toward thisaim, the National Family Planning Boardstarted training kampong bidans in Janu-ary 1969 with UNICEF support. In eachstate bidans were selected for the trainingby the nursing sister, the highest adminis-trative person in nursing at the statelevel. Candidates were selected fromamong those traditional midwives whowere known by the State Health Depart-ment and who said that they could comefor training. While the process was notrandom selection, it is believed that therespondents were representative of theethnic composition and economic con-dition of their respective states. In gen-eral, more active and senior bidans werechosen. They were trained at the generalhospital at each state capital for three

weeks. Training consisted of one week :0the maternity ward, one week of instruc-tion on family planning, and one week ofpractical training at the local healthcenter. Between January 1969 and Decem-ber 1970, 363 bidans were trained.

InterviewMETHOD

Information was obtained by interviewwith the kampong bidans at the time theycame to report for training. This reportdeals with results of interviews conductedbetween January 1969 and December1970. Two groups were not interviewedand the interview forms of one group werelost in the mail on the way to the head-quarters so that 292 completed inter-views were received.

The questionnaire form we designedand used contained questions coveringsocial and demographic characteristics ofrespondents, the history, nature, andscope of their services, and questions onthe bidans' attitudes toward family plan-ning and their willingness to participatein a family planning program. The inter-views were conducted by nursing sistersor staff nurses of the National FamilyPlanning Board.

Because interviews were conducted indifferent locations and over a period oftime, it was not possible to maintain closesupervision of the interviewers for theduration of the study. Some respondentswere not asked all the questions on theoriginal questionnaire, and we have notedin the text the number of respondents foreach question.

RESULTS

Aspects of the general background ofthe bidans, by state, ai ,. shown in Table I.All hut four of the 292 bidans inter-viewed were Malay. Their mean age was47.3, ranging from 20 (Johore) to 71(Perlis). About 84 percent (244) of thebidans were over 40. About 73 percent(213) were currently married, 18 percent(52) widowed, 8 percent (23) divorced orseparated, and 1 percent (4) never married.

Eighty percent (233) had had no school-ing, 18 percent (53) had only one to fiveyears education, and 1 percent (4) hadsix to seven years education.

Of the husbands of the currently mar-ried, 73 percent (157) were seasonal orunskilled workers. Fifteen percent (32)were unemployed or not working, and11 percent were semi-skilled and skilledworkers, including a few office workersand uniformed personnel.

Of all the bidans interviewed, 43 per-cent (125) had practiced less than 10years, 32 percent (94) between 10 and 20years, 21 percent (62) between 20 and 30years, and 4 percent (II) more than 30years.

When asked who taught them to de-liver babies, 27 percent (80) said theirgrandmother; 22 percent (64) theirmother; 8 percent (23) an aunt; 5 percent(14) a friend; 2 percent (6) mothers-in-law ; 6 percent (18) a combination ofmother, grandmother, aunt, in-laws andrelatives; and 30 percent (86) othersources.

Two questions were asked about thenumber of deliveries attended by kampongbidans; "How many deliveries have youactually attended during the last month?"and "How many deliveries have you at-tended during the past year?" On theaverage, each kampong bidan reportedattending three deliveries during the past

TABLE

month and 26 deliveries during the pastyear (Table 1). Those kampong bidanswith more than ten years of practice hadthe higher average number of deliveriesattended (Table 2).

When asked "How much do you chargefor attending each delivery?", the aver=age charge reported by the bidans was5.4 Malay dollars (about USS100) perdelivery (see Table 1). Charges variedfrom one to 25 Malayan dollars (US$8.00). Fifty-three bidans said that theyreceived money but did not cite theamount. No information was ascertainedfrom 27 bidans. In a breakdown by state,the amount of money received for eachdelivery was relatively high in Malacca,Penang, Selangor, Johore, and Kedah. Instates on the east coast such as Kelantan,Trengganu, and Pahang the amount wasrelatively low. This is consistent with thefact that the states on the east coast areeconomically disadvantaged.

1, Mean Age, and Characteristics of Roles of Bidans by State

NumberState interviewed

Meanage

Mean number of deliveries Average amountreceived per

delivery"Last month Past year

Johore 13 47.7 5.5 33.4 8.0Kedah 36 48.3 2.4 23.5 7.9bKelantan 49 45.7 3.7 28.4 2.8Ma lacca 7 44.4 2.3 54.8 15.7Neg.

Sembilan 18 44.7 1.6 7.211 7.1bPahang 56 44.5 3.4 15.7 5.6bPenang 42.0 2.3 19.1 12.3bPerak 19 48.0 4.9 55.8 6.61"

Perlis 14 54.9 2.2 25.1b N.A.

Selangor 4 45.8 7.5 75.7 10.5Trenggan u 59 50.8 3.4 26.0 3.8

Total 292 47.3 3.3 26.1 5.4

"In Malay dollars. US$1.00=M$3.00.b Figures exclude 53 bidans who said they received money but did not cite the amount, and

27 bidans from whom no information on payment was ascertained.N,A.=:no answer.

TABLE 2. Mean Number of Deliveries Attended Last Month and Past Year by

Years of Practice

Years ofpractice

Numberinterviewed

Mean number of deliveries

Last month Past year

Less than 5 40 2.1 14.35-9 85 3.2 22.3

10-14 42 4.0 30.915-19 52 3.0 29.6020-24 41 3.4 23.7"25-29 21 4.7 36.830-34 3.4 28.635+ 4 12.0 87.2

Total 292 3.4 26.2

°Figure excludes one to two c

26

7, nto

s from whom the information was not ascertained.

The bidans were asked the question:"Do you get anything other than cash foryour services ?Chickensclothingo (her/specify--". For 72 bidans no informationwas ascertained. Of those whose responsescould be ascertained, 79 (36 percent) saidthey did not receive gifts; 21 (9 percent)received clothes; 4 (2 percent) receivedchicken; 50 (23 percent) received a combi-nation of chicken, clothing, and otheritems; 66 (30 percent) received items otherthan chickens and clothes.

Information was obtained from 267bidans on the question: "What do you doto the mother other than conducting de-liveries? Abortion-- Giving herbs--Massage-- Other-- specify". Ofthese respondents 134 (50 percent) saidthat they performed massages, and 81(30 percent) performed massages com-bined with other services. Thirty-three(12 percent) said they did not do any-thing other than midwifery, and 11 (7percent) provided services other thanmassage, e.g., prescribing herbs, and per-forming abortions. Six bidans mentionedabortion alone or combined with otherservices. The information was not ascer-tained from 25 bidans.

When the. bidans were asked about thestage of pregnancy at which they first hadcontact with the mother, 185 (67 percent)said between the seventh and ninth monthof pregnancy; 67 (24 percent) between thefourth and sixth month; and 19 (7 per-cent) at or before the third month ofpregnancy. Five bidans answered thatthey started at an indefinite month, andinformation was not ascertained for 16.

One hundred and seventeen bidans (40percent) said that the postpartum carelasted less than one week, and another40 percent (118) said that they lookedafter postpartum mothers from one totwo weeks. About 18 percent (53) saidthat they had contact with postpartummothers between two and seven weeks.Information on postpartum care was notascertained for four bidans.

One hundred and ninety-eight bidanswere asked the question: "Do you ap-prove or disapprove of providing marriedwomen with family planning services?Approve-- Disapprove--." Only twodisapproved. The following informationwas also obtained from these 198 bidans:

One hundred forty-nine (75 per-cent) knew that the government was pro-viding family planning services.

One hundred twenty-three bidans (62percent) had been approached by women

about family planning information ser-vices during the last three months.

One hundred and seventy-four (99percent) said that they were not worriedthat the government's family planningservices would affect their job in conduct-ing deliveries.

One hundred eighty-eight (95 per-cent) thought that they could help to pro-mote the government's program by re-cruiting patients and distributing contra-ceptives. Ninety-seven of these women (49percent) said that they could participatein the program without being paid and96 preferred to receive a commission fordistribution of contraceptives.

Although it may be argued that bidansresponded in such a positive way underthe special circumstances, it is worth-while to note that almost all the bidansquestioned were willing to participate inthe government program of providingfamily planning services to marriedwomen. By comparison, a high propor-tion of Pakistan's dais showed an un-willingness to provide information re-garding family planning during the firstphase analysis (1).

CommentGiven the apparent willingness of

kampong bidans in Malaysia to promotematernal and child health and familyplanning services, important questions re-main: (I) how kampong bidans can beutilized; and (2) how kampong bidanscan be paid for their participation.

To answer the first question: The bidanscan be asked to bring pregnant women tobe registered at the rural health center forantenatal, natal, and postnatal care. Atthe same time, they can be organized tobring women, especially postpartumwomen, to the health clinic for familyplanning services. It should be decidedwhether the bidans should only recruitacceptors or should distribute contracep-tives. The best approach might be forbidans to bring women to health clinicsfor the initial acceptance so that a staffnurse can screen the women for acceptinga contraceptive method and fill out acomplete acceptance record. The oral isthe most frequently used contraceptive inMalaysia. The bidans could resupply theoral acceptors with contraceptives. Underthese circumstances, bidans could play animportant role in promoting contracep-tive continuation in rural areas by pro-viding women with a continuous motiva-tion through their constant contact. Sincemost of the bidans are illiterate, a special

27

record form should be designed andbidans should be trained to fill out theform to record the resupplying of contra-ceptives and other important information.

The importance of supervision ofkampong bidans by a nurse at the healthclinic cannot be overemphasized. Thesuccess of the utilization of kampongbidans in rural family planning serviceswill depend greatly on this supervision.

The next important question concernsreimbursement of bidans for participatingin family planning services: Is it feasibleand advisable to pay bidans? If so, inwhat form should payment be and whatamount? Bidans can be paid with a flatsalary type paynlent, with an incentivescheme alone, or with a combination ofboth. Salaries and incentives could becombined as follows: Bidans could re-ceive regular salaries for routine activi-ties, visiting individual homes, findingclients, and bringing clients to clinics. Forthese activities an area covering a suitablesize of the population could be designatedfor each bidan. Incentive payments couldbe reserved for the continuation of theuse of contraceptives by clients. As analternative, an incentive scheme could becreated based on a scale of increase ofsalary to bidans according to evaluationof their performance by the nursing staffsupervisor of the rural health unit: Care-ful consideration is important in imple-menting either of these or other pro-posals.

Integration of family planning servicesinto rural health units was started inMalaysia in early 1971. The attempt toutilize traditional midwives for familyplanning services is a new venture in thehope that the rural family planning pro-gram will be accelerated along withmaternal and child health services.Whether or not this effort will be success-ful remains to be seen.

BIBLIOGRAPHYJ. Hassan N. Gargezi and Attiya Inaya-

tullah, The Dai Study, Karachi, The WestPakistan Family Planning Association,1969.

2. United Nations Advisory Mission, AnEvaluation of the Family Planniv Pro-gramme of the Government of India, 1969,paras. 105-109.

3. Family Planning Scheme for Pakistan,Third Five Year Plan Period, 1965-1970,Karachi Ministry of Health, Labour andSocial Welfare, Government of Pakistan,1965.

4. S. A. Jafarey, J. O. Hardee, and A. P.Satterthwaite, Use of Medical-Para-

medical Personnel and Traditional Mid- 6.

wives in the Pakistan Family PlanningProgram," Demography 5, no. 2 (1968). 7.

5. Lee L. Bean, "Pakistan's Population inthe 1970's: Certainties and Uncertain-ties." Paper presented at the meeting ofAssociation of Asian Studies, March 8.

1971, the Population Council.

Report of the Swedishby CHRISTOPHER TIETZE

Following are a brief review of theSwedish positions on legalized inducedabortion over the past three decades anda summary of the recommendationsmade by the 1965 Committee on Abor-tion. Christopher Tietze, M.D., is asso-ciate director of the Biomedical Divisionof the Population Council.

Sweden has been a pioneer in the liber-alization of abortion laws since 1938.Every major change in legislation has beenpreceded by the publication of a compre-hensive report prepared by a commissionor committee, appointed by the govern-ment, after giving careful consideration toall aspects of abortion. The earlier reportsof such bodies, established in 1934, 1941,and 1950, have become classics in thehistory of social legislation.

The Abortion Committee of 1934 rec-ommended legalization of abortion oneugenic, humanitarian, and social grounds.The latter type of indication was not ac-cepted by the Swedish riksdag which in1938 authorized abortion to avert a seri-ous threat to the woman's life or healthdue to disease, physical impairment, orweakness; in cases of pregnancy resultingfrom rape and other sex offenses; or ifeither parent was likely to transmit to theexpected child inheritable mental disease,mental deficiency, or severe physical de-fect. These grounds for the legal termina-tion of pregnancy were essentially thoseproposed in 1962 by the American lawin its Model Penal Code which, in turn,has served as the basis for reform legisla-tion in the United States since 1967.

The Swedish Population Commission of1941 and the Abortion Commission of1950 did not recommend major changes inthe abortion law itself, but devoted them-selves primarily to administrative prob-

Ministry of Public Health. Public Healthin Thailand. Bangkok, Thailand, 1968.H. T. Croley, S. Z. Haider, S. Bigum andH. C. Gustafson. -Characteristics andUtilization of Midwives in a SelectedRural Area of East Pakistan." Demogra-phy 3, no. 2 (1966),V. Bhandari and J. Bhandari. "Indige-nous Dais: A Study of 50 Cases." Family

Abortion Committeelems and to the prevention of abortion byprovision of a variety of social services,by economic benefits to families, and bysex education, including the dissemina-tion of contraceptive information. Theriksdag, however, enacted amendments tothe abortion law in 1946 and 1963, liber-alizing it by degrees without changing itsbasic character.

A new committee was appointed in 1965with a mandate to consider further liber-alization of the abortion law. The reportof this committee, published in September1971 under the title "The Right to Abor-tion," contains 192 pages of text and 12appendices, totaling 371 pages.

The committee recommends that thepresent prohibition of abortion, includingself-abortion, be removed entirely fromthe penal code. However, termination ofpregnancy by a physician outside a hos-pital and all attempts at abortion by laypersons, other than the pregnant womanherself, shall remain prohibited under theproposed abortion statute.

The committee further recommendsthat every woman who is a resident ofSweden shall be entitled to have her preg-nancy terminated (I) if it can be assumedthat her health will be threatened or herstrength seriously reduced by a continua-tion of pregnancy, or (2) if it can be as-sumed that the child to be born wouldsuffer from a severe illness or defect, or (3)if "for another reason it is an unreason-able hardship for her to continue herpregnancy." The broad language of thislast clause is intended tO cover a widevariety of reasons, including multiparity,advanced age, immaturity, economic diffi-culties, various situations of conflict, andother personal reasons. The report makesit clear that the committee feels that awoman's request to have her pregnancyterminated should ordinarily be approvedunless there are significant medical orpsychological contraindications.

28

12

Planning News 11, no, 2, February, 1970.

A Study of the Maternal and child HealthService in the States of Malaya. Maternaland Child Health Committee, NationalHealth Council, Kuala Lumpur, 1965.

10. Report on West Malaysian Family Survey1966-1967. National Family PlanningBoard, Kuala Lumpur, Malaysia, 1968.

The committee, rejecting the notionthat termination of pregnancy should bemade available "on demand," recom-mends that a Swedish woman's right to asurgical termination of pregnancy shall bedetermined by boards to be established atall obstetrical-gynecological departmentsof hospitals. Each board shall consist ofan obstetrician-gynecologist who acts asthe chairman, a psychiatrist, and a laymember. Most applications for abortionsby women who are less than three monthspregnant should be approved by theobstetrician-gynecologist on his own re-sponsibility. Doubtful cases and all casesinvolving second trimester pregnancies areto be decided by a majority of the board.Any rejection of a request for terminationof pregnancy shall be automatically re-ferred to the Central Social Board inStockholm for final decision.

All abortions shall be performed inpublic hospitals and other approvedfacilities at public expense, as early aspossible, and by the least traumatic pro-cedure. The committee also recommendsthat a wide range of existing welfaremeasures be strengthened and new serv-ices, including public family planningservices, be established in order to im-prove the social and economic security ofwomen and to make it possible for bothparents to combine the care of the childwith their obligations in other fields ofife.

THE POPULATION COUNCIL245 Park Avenue, New York, New York 10017

The Population Council is a private nonprofitorganization established in 1952 for scientifictraining and .study in population matters. It en=deavors to advance knowledge in the broad fieldor population by fostering research, training, andtechnical consultation and assistance in the socialand biomedical sciences,

John D. Rockefeller 3rd,Chairman of the Board of Triatens

Bernard Berelson, President