social coercion for weaning

SOCIALCOERCIONFORWEANING

Janice M. Morse, RN, Phq and MARGARET J. HARRISON, RN, MSN

ABSTRACT

Research investigating the duration of breastfeeding is generally based on the assumption that if lactation is established, then the mother will breastfeed successfully. Longi- tudinal, qualitative research suggests, however, that it is the attitude of others towards the breastfeeding mother that determines the duration of breastfeeding. When the infant is a newborn, breast-feeding is encouraged. As the infant develops, the support of others towards the nursing mother is withdrawn and the mother is encouraged to wean. Further, the doula not only facilitates breastfeeding, but also, when the infant is perceived to be “old enough,” facilitates weaning.

In current breastfeeding research, there is an assumption that the method of infant feeding and the duration of breastfeeding are decided by the mother. Based on this assumption, the following premises give direction to most breastfeeding research and educational programs. First, when given adequate information on the advantages of breastfeeding (and the disadvantages of not breastfeeding), the mother will make a rational choice for the method that will be most advanta- geous for her infant (ie, breastfeeding). Second, once lactation is established, the mother will continue nursing. Therefore, conditions in the immediate postpartum period are made to facilitate breastfeeding. Ef- forts are made to maximize ma- temal-infant attachment, to minimize breast complications, and to provide a supportive environment to ease the breastfeeding process and the establishment of lactation. l Finally,

Address correspondence to: Dr J. Morse, 3rd Floor, Clinical Sciences Building, Univer- sity of Alberta, Edmonton, Alberta T6G 2G3, Canada.

so that lactation will succeed, care is taken to ensure that the infant’s interest and ability to suckle is maxi- mized. Supplements that may inter- fere with the infant’s appetite are not given, and the infant is breastfed “on demand.“2B Inherent in this paradigm is the notion that if breastfeeding is initially a positive experience, then mothers will continue to successfully breastfeed. “Success” is usually measured by the medical profession in time, which is often de- fined as maintaining breastfeeding for at least 3 months and ideally for 6 months.*

Research investigating the duration of breastfeeding is consistent with this general paradigm. For in- stance, most research examining the rapid attrition in breastfeeding rates have focussed on the perinatal pe- nod, statistically correlating factors from this period with the duration of breastfeeding. We are not denying the importance of these factors for the initiation of breastfeeding. In- stead, we argue that events in the perinatal period may have a very weak association with the mainte- nance of breastfeeding for 3,6,9,12

Journal of Nurse-Midwifery ??Vol. 32, No. 4, July/August 1987

Copyright 8 1987 by the American College of Nurse-Midwives

months or longer. We suggest that the nursing experience is a dynamic, open relationship occurring within a social context, and it is the attitude of others towards breastfeeding that modifies the mother’s choice of how the infant is fed, where the infant is fed, and for how long breast-feeding is maintained. In this article, we will critique present research on the duration of breastfeeding, discuss the development of a theory of social coercion for weaning, and suggest fur-, ther research to test this theory.

IMPORTANCE OF THE PROBLEM

In spite of the health advantages of breastfeeding for the mother and the infant, many mothers choose not to breastfeed. Extensive educational programs on the benefits of breastfeeding have been only partially successful, with the number of mothers breastfeeding their infants increasing over the past 15 years to 62Y~~s~ However, there is a rapid decline in the percentage of women breastfeeding after discharge from hospital. By the fourth month postpartum, only 44%, and by the sixth month

205

0091-2182/87/$03.50

31% of mothers continue to breastfeed.7 The current recommendation by the American Pediatric Associa- tion* is that breastfeeding be continued for 6 months. The rapid decline in breastfeeding rates is of suffi- cient concern that in the Public Health Services objectives for the United States in 1980, 1 of 17 nutri- tional objectives listed was to in- crease the incidence of breastfeeding to 75% by 1990.6,qJo

CURRENTRESEARCH

Researchers have sought to delin- eate factors that correlate with the duration of breastfeeding so the characteristics of the “successful” breastfeeding mother may be under- stood and the variables that inhibit or facilitate breastfeeding may be identified. The variables can be placed within three separate timeframes: the prenatal period, the time of birth, and the immediate postpartum period.

Prenatal Variables

Several studies have shown that the successful breastfeeding mother has a higher level of educationll-l5 and socioeconomic status.‘*-‘* She is married,13 older,11-16 and less likely to smoke17J8 than her bottle feeding counterpart. These successful

Janice M. Morse is a Professor, Faculty of Nursing and a Research Associate, Department of Anthropology, University of Alberta; an Associate Clinical Nurse Researcher, University of Alberta Hospitals, and an NHRDP Research Scholar. Her research interests are culture and health, clinical nursing research, and women’s health.

Margaret J. Harrison is an Associate Professor, Faculty of Nursing, at the Uniuersity of Alberta and a doctoml candidate in the Department of Family Studies. Her research interest focuses on maternal-child interaction.

206

feeding mothers also attend prenatal classes, 12,17+1q,20 have previously breastfed,l* and demonstrate the desire or motivation to breastfeed.11,13*14 They also have more friends that breastfeed and perceive their husbands to be supportive 11,13,15,16,21

Birth Variables

The mother most likely to succeed at breastfeeding has a normal, sin- gleton birth, a healthy baby,22,23 an opportunity to nurse the infant shortly after delivery,1~24~25 and forms an immediate attachment with the infant.26 Indicators for a less successful breastfeeding experience are a long and difficult labor, anaes- thesia, a cesarean birth,27 and an infant who is preterm, exhausted by labor or is handicapped. 18,27

Postnatal Variables

The majority of the research on successful breastfeeding has focused on the immediate postnatal period. Many authors suggest that breastfeeding will be more successful if the infant is rooming-in with the mother, fed on demand,20*27 and supplements are withheld.lx** Nursing staff should be supportive and knowledgeable about breastfeeding.1J3,28*2q The mother will benefit from a doula, perhaps her husband or her mother.30,31 Breast complications, such as cracked nipples or engorged breasts, will be avoided.32 Because supplements will have been withheld, the infant will be eager to nurse.32

In the hospital, educational programs on the benefits and methods of breastfeeding are correlated with an increased duration of breastfeeding. 33 The provision of mothers with formula samples and/or expo- sure to advertisements for formula are associated with a decreased duration of breastfeeding.s4z35

Research in the subsequent postnatal period is relatively sparse. One of the most frequent reasons

given for weaning is “insufficient milk “13~14,36 but it is unclear in many studies what this term actually means to mothers. This phrase could be used to describe a crying or fussing infant, perceived to be hungry by the mother, as well as a physiological decrease in milk s~pply.‘*,~~,~~ The early introduction of solid foods is associated with a decrease in lactation and early weaning.38 Finally, “convenience”l* or “returning to work”3q is frequently given as a reason for weaning, although studies do not consistently agree on this factor.37,40

CRITIQUE ON CURRENTRESEARCH

A major problem in the research literature is the predominance of retrospective, cross-sectional surveys based on subject recall over a con- siderable period of time.41r42 Re- searchers using these methods have confused causality with associated events43,44 by seeking statistical correlations between past events and subsequent breastfeeding outcomes that are separated by a period of several weeks or months. Researchers have disregarded the effects of his- tory& in a period of rapid and significant role change within the family. Furthermore, many of the researchers have applied univariate statistical methods to a complex, multivariate situation. Such techniques frequently give spurious correlations,46 and this fact alone may account for some of the discrepant results in the literature. In making a plea for longitudinal studies, Simmopoulus and Grave*’ note:

In general, demographic variables are easier to incorporate into question- naires and easier to administer, explain, test and score than psychosocial, and especially, attitudinal variables. Yet the latter are probably closer to true causal determinants of choice in infant feeding practices than demographic variables.

For the past 2 decades, most of the research has been deductive.


Variables identified in the literature or from previous research have been tested repeatedly in different settings and with different populations. Be- cause many of the researchers have no practical experience for ad- dressing the problem, the underlying assumptions have rarely been examined or challenged. It is time to question the present paradigm and, using inductive methods, develop alternate explanations on the duration of breastfeeding.

SOCIAL COERCION FOR WEANING

In 1983, we conducted a longitudinal study of 30 breastfeeding mothers.47 We were investigating minimal breastfeeding (ie, the man- agement of lactation when mothers breastfeed once or twice a day, without expressing). As current advice to nursing mothers is that the breasts must be expressed if she cannot nurse, we were interested to know how long and for what reasons mothers minimally breastfed. A longitudinal study was planned to determine the rational for minimal breastfeeding and how mothers learn about this style of breastfeeding. A convenience sample of mothers who were nursing once or twice a day was obtained by advertising in the public health im- munization clinics. The demographic characteristics of the sample follow. Mothers ranged in age from 21 to 31 years (mean age, 28.5 years). All mothers were upper-middle class and well educated 25 mothers (ie, 83%) were educated beyond high school, 4 completed high school, and 1 did not complete high school. All but two mothers were married. Twelve mothers had breastfed previously (nine had breastfed once, two had breastfed twice, and one mother had breastfed three times), and 18 mothers had not breastfed previously. The number of people residing in each household, including the infant, ranged from 2 (ie, mother-infant dyads) to 7 (one

family), with most families (n = 16) consisting of 3 members.

The partners’ (n = 28) ages ranged from 24 to 42 years (mean age, 31.1 years). Twenty (ie, 71%) were educated beyond high school, seven completed high school, and one did not complete high school. The infants in the study consisted of 15 males and 15 females. Birth weights ranged from 3.09 to 9 lb (mean weight, 7.83 lb).

Semistructured, taperecorded tele- phone interviews were conducted with these mothers each month until the infant was weaned, or for 12 months. Altogether, 88 interviews were conducted, transcribed and an- alyzed. It is important to note that this was a select sample: all mothers were, by present criteria, successful at breastfeeding, having continued to breastfeed at least 6 months. Ap- proximately one-third of the sample were employed outside the home, one-third were weaning their infants slowly, and one-third were comfort nursing older infants. Data used in this discussion were derived from one part of the questionnaire related to the support for breastfeeding: who the mothers perceived as the most supportive and as least supportive of their breastfeeding, and of mothers’ perceived attitudes of others, including their physician, nurses, family, friends and acquaintances, towards their breastfeeding.a

Unexpectedly, mothers’ responses to these questions varied over time. Responses of each mother about their husband or “doula,” parents and parents-in-law, physician and nurse, and friends and colleagues in the workplace were sorted into three general categories: supportive/en- couraging, noncommittal/silent, or discouraging. These responses were then placed on a time-line according to the infant’s age.

Definite trends were immediately apparent. All mothers, without exception, reported in the immediate postpartum period that “every one was supportive of breastfeeding.” However, when the infant was about

Journal of Nurse-Midwifery ??Vol. 32, No. 4. July/August 1987

6 to 8 months of age, mothers reported that their friends became silent and “didn’t say anything” about breastfeeding. Later, when the infant was 9 to 10 months of age, mothers reported that their friends began to make comments such as “Are you still nursing?’ These comments resulted in a behavior change for the nursing couple. Mothers withdrew from those friends who were most negative and sought friendship from mothers who were also “still nursing,” or they actively sought contacts in the La Leche League. Mothers who continued to nurse avoided listening to those who were negative and told them to “mind their own business.” They deliber- ately kept secret the information that they were “still nursing” so that others could not comment. Mothers no longer nursed in public nor per- mitted the child to ty to nurse in public, and they developed a special word (one that no-one else would understand) to refer to nursing. If the child asked to nurse, he/she would be told to wait until they got home. One mother reported:

People are making negative comments-for example in the baby room at church. They say, ‘He’s get- ting too old! It’s bad for your health!’ I start feeling guilty and anxious.

When these infants were approximately 10 months of age, the mother’s parents and parents-in-law (who were previously supportive) became silent about the breastfeeding and withdrew support. Then, as the infant approached 12 months, the mother or mother-in-law would suggest that continuing breastfeeding was “draining her (the mother’s) body,” and perhaps if the child were weaned, he could stay overnight with them. Thus, the grandparents actively encouraged weaning:

My mother and mother-in-law feel I should quit at 12 months. They were quite vocal about this! My mother-in- law feels I will get cancer from the prolonged tugging at the breast.

Comments from the grandparents

207

would also be directed at the toddler: “They tell her she’s too big for that, now.” If the mother chose to continue breastfeeding, this fact was withheld by the mother from her parents and parents-in-law. One mother who continued to breastfeed finally weaned because her parents- in-law were coming to visit for the Christmas holidays; the continued breastfeeding could no longer be concealed with houseguests present.

blouse. My neighbor just looked at me and said Are you still nursing him? It was so obvious that I couldn’t say no! (Infant 16 months)

This study, consistent with other North American studies,21*30,31 showed the main source of support (ie, the doula) was the husband. Early in the study, mothers all reported that their husbands were most supportive and felt that “breast was best.” However, when the infant was approximately 12 months of age, mothers reported that their husbands became quiet about their breastfeeding: “He doesn’t say anything.” Mothers described their husbands’ embarrassment if, for example, the infant pulled at the mother’s blouse in public in an at- tempt to get at the breast. If the mother continued to breastfeed, the husband would consistently try to coerce his wife into weaning by making comments such as: “What’s a big boy like that doing there!” After the husband withdrew his support, all mothers weaned within a few weeks, with the exception of one mother who continued to nurse without telling her husband.

Importantly, mothers who continued to nurse their infants beyond the social norm did so in secret:

I feel pressured sometimes. We always have to sit in the front of the church; we’re always late. When we’re at church for an hour and a half, that’s all that Carl can think of because he’s sit- ting there and he’s on my lap and so he’s pulling and arching his back. I give him to my husband and I feel a little uncomfortable! . I nursed him in the front row when he was younger, but not now! (Infant 16 months)

From these data, it is apparent that the duration of breastfeeding is in- fluenced by social and cultural norms and not solely determined by the mother. The course of breastfeeding is a dynamic process, and although the variables examined by other researchers may be associated with the duration of breastfeeding, social support for the nursing mother is perhaps the most significant vari- able. The changing attitudes of others over time are shown in Fig- ure 1.

“old enough to wean.” Variations in the cultural or subcultural norms about when the infant is “old enough to wean” may explain the significant differences in the duration of breastfeeding that researchers have found correlated with variables such as socioeconomic class, ethnic@, and educational levels. Fur- thermore, through research examining patterns of weaning, indications of when the infant is “old enough to wean” may be obtained and thus provide insights into the social norms. Developmental milestones, such as when the infant begins to sit, to develop teeth, to talk or to walk, were used by mothers in this study as indicators that weaning should occur.

SUGGESTIONS FOR FURTHERRESEARCH

RaphaePO and others21p31 have suggested that lactation cannot be established without a doula; we suggest that the role of the doula is also to facilitate weaning and to assist with severance when, as determined by the cultural norm, the infant is

In this study, all mothers were atyp- ical in that they were successfully breastfeeding, and some of the mothers, those practicing prolonged breastfeeding (ie, over 12 months), may be considered extraordinary. Thus, the ages of the infants in each of the stages of coercion must be re- garded as preliminary indicators only. Further research on larger, rep- resentative samples must be conducted to obtain a clearer description

Active suppon for breastfeeding

-Friends -Parents -Partners

Now that she’s older I don’t talk about it-except to moms who have also breastfed.

Sometimes, however, breastfeeding could not be concealed, and this resulted in awkward situations for the mother:

Level of Breastfeeding perceived tolerated support

Breastfeeding ignored

0 :::::: Weaning ‘i:j:I facilitated

My neighbor was really shocked that I’m still nursing. We were grocery shopping yesterday and it was around Kelley’s usual nap time and he began to fuss and started pulling at my

Active encouragement for weaning

FIGURE 1.

2 4 6 8 10 12 14 Age of infant (months)

Response of others towards the breastfeeding couple by age of the infant.

208 Journal of Nurse-Midwifery ??Vol. 32, No. 4, July/August 1987

I of cult&al and social norms. It is pos- sible that, in a population that pre- dominantly bottle feeds their infants, the social and cultural forces against breastfeeding are so strong that breastfeeding is socially prohibited. In populations where mothers wean their infants very quickly (ie, almost as soon as the mother is discharged from hospital), the stages we have described may occur much earlier.

To test the relationship between the attitudes of the larger society on infant feeding and the duration of breastfeeding, a large randomized survey of the city is planned. A rep- resentative sample of residents of all ages will be shown pictures of infants at various developmental stages and asked how these infants should be fed. If there is a correlation between the identified social norms for the duration of breastfeeding and the mothers’ actual weaning patterns, then some additional confirmation for the theoy of social coercion will be obtained.

It is interesting to observe the amount of advice that a new mother receives. Advice on infant care is provided to mothers even by strangers who would not dare to censor or to advise the mother on any other topic. Somehow, with infants, everyone is an expert. It would be interesting to examine why others dissuade the mother from breastfeeding “too long” and why the coercion to wean is an emotive social issue.

CONCLUSION

As stated, the present thrust of research on the duration of breastfeeding focuses on the mother. It is assumed that breastfeeding is mother’s “choice,” and the responsible choice for good mothering is to breastfeed. Current trends in health education are targetted at the mother so that she will make the right choice. To quote Dr. Everett Koop, United States Surgeon Gen- eral: “Persuading more women to breastfeed their infants is a major

U.S. public health goal.“lO(italics added).

Considering our research, we are suggesting that it is naive to focus en- tirely on mothers and to attribute the lack of breastfeeding or premature weaning as a problem caused by ir- responsible women not choosing “the best” for their infant. Our research suggests that if a lack of breastfeeding is a “problem,” then it is a general problem of all citizens. Breastfeeding does not occur in iso- lation, either apart from the family context or separate from the community at large. As such, health education programs must be aimed at all sectors of the population: to grandparents and schoolchildren, to the married and the unmarried, to males as well as to females. Furthermore, at present, we doubt if women really have a freedom of choice regarding infant feeding methods. In some postnatal units, mothers are coerced into breastfeeding, and this may, in part, explain the very rapid change to bottle feeding l3 after discharge. Health professionals may give mothers mixed messages regarding infant feeding,2g,4g verbally sup- porting breastfeeding but, at the same time, giving very clear non- verbal messages to bottle feedss*35 with supplements23 or advising bottle feeding when minor difficulties arise.28 Social coercion for weaning, the mourning for the loss that mothers sometimes feel when weaning occurs, and the phenom- enon of closet nursingso suggest that the method and duration of infant feeding is not mother’s choice but a response to cultural norms.

This work was funded by the University of Al- berta and, in part, by an NHRDP Research Scholar Award to Dr. J. Morse. Thii article is develowd from an earlier chapter, “What De- term&s the Duration of Br&tfeeding,” in the book, The Anthropology of Childbirth in America, edited bv K. Michaelson, and now being published by Bergin & Garvey, S. Hadleu. Mass. The article. “Minimal Breast- feeding,” in JOGNN, 15,333-338,1986, by the same authors also was used as source ma- terial.


REFERENCES

1. Winikoff B, Laukaran VH, Myer VH, et al: Dynamics of infant feeding: Mothers, professional and the institu- tional context in a large urban hospital. Pediatrics 77~357-365, 1986.

2. Auerbach KG: The role of the nurse in support of breastfeeding. J Adv Nurs, 4263-285, 1979.

3. Houston MJ: Indicators of successful breastfeeding, in Houston MJ (ed): Maternal and Infant Care. London, Churchill-Livingstone, pp 26-48, 1984.

4. Harrison MJ, Morse JM, Prowse M: Successful breastfeeding: The mother’s dilemma. J Adv Nurs 10261-269, 1985.

5. Quant S: Biological and behav- ioural predictions of exclusive breastfeeding duration. Med Anthropol. g(2): 139-151, 1985.

6. Koop CE, Brannon ME: Breast feeding-The community norm. Report of a workshop. Public Health Rep. 99(6):550-557, 1984.

7. McNally E, Hendricks S, Horowits I: A look at breastfeeding trends in Canada (1963-1982). Can J Public Health 76: lOl- 107, 1985.

8. American Academy of Pediatrics: Policy statement based on task force report: The promotion of breastfeeding. Pediatrics 69:654, 1982.

9. Arango JO: Promoting breast feeding: A national perspective. Public Health Rep 99(6):559-564, 1984.

10. Smith SJ: Surgeon general urges more moms to breast-feed babies. Times-Union, Rochester, NY, June 13, 1984.

11. Bloom K, Goldbloom RB, Rob- inson SC, et al: Breast versus formula feeding. Acta Paediatr Stand 300 (Suppl):3-25, 1982.

12. Clark LL, Beal VA: Prevalence and duration of breast-feeding in Mani- toba. Can Med Assoc J 126:1173- 1175,1982.

13. Sacks SH, Brada M, Hill AM, et al: To breast feed or not to breast feed. Practitioner 216183-191, 1976.

14. Sjolin S, Hofvander Y, Hillervik C: Factors related to the early termina- tion of breastfeeding: A retrospective study in Sweden. Acta Paediatr Stand 66505-511, 1977.

15. Switzky LT, Vietze P, Switzky HN: Attitudinal and demographic predictors

209

of breast-feeding and bottle-feeding behavior by mothers of six-week-old infants. Psycho1 Rep 45:3- 14, 1979.

16. Beske EJ, Garvis MS: Important factors in breast-feeding success. MCN 7:174-179, 1982.

17. Florack E, Obermann-de Boer G, Van Kampen-Donker M, et al: Breast- feeding, bottle-feeding and related factors. Acta Paediatr Stand 73:789- 795, 1984.

18. Goodine LA, Fried PA: Infant feeding practices: Pre- and postnatal factors affecting choice of method and the duration of breastfeeding. Can J Public Health 75439-443, 1984.

19. Gulick EE: Informational corre- lates of successful breast-feeding. The American Journal of Maternal Child- Nursing 7:370-375, 1982.

20. Winikoff B, Baer EC: The obstetri- cian’s opportunity: Translating “breast is best” from theory to practice. Am J Ob- stet Gynecol 138: 105- 115, 1980.

21. Baranowski T, Bee DE, Rassin DK, et al: Social support, social influ- ence, ethnic@ and the breast-feeding decision. Sot Sci Med 17(21):1599- 1611, 1983.

22. Beyer LR: When should one dis- courage breast-feeding. Pediatrics 67: 300-302, 1981.

23. Ellis DJ, Hewat RJ: Factors related to breastfeeding duration. Can Fam Phy- sician 30: 1479- 1484, 1984.

24. de Chateau P, Holmberg H, Ja- kobsson K, et al: A study of factors promoting and inhibiting lactation. Develop- mental Medicine and Child Neurology 19:575-584, 1977.

25. Salariya EM, Easton PM, Cater JI: Duration of breast-feeding after early initiation and frequent feeding. Lancet ii:1141-1142,1978.

26. Carlsson SG, Larsson K, Schaller

J: Early mother-child contact and nursing. Reprod Nutr Dev 20:881-889, 1980.

27. Reames ES: Opinions of physi- cians and hospitals on early infant- feeding practices. J Am Diet Assoc 85(1):79-80, 1985.

28. Verronen P, Visakorpi JK, Lammi A, et al: Promotion of breast feeding: Ef- fect on neonates of change of feeding routine at a maternity unit. Acta Paediatr Stand 69:279-282, 1980.

29. Ellis DJ, Hewat RJ: Do we support breast-feeding mothers? Midwives Chronicle 97( 1153):45-47, 1984.

30. Raphael D: The tender gift. Engle- wood Cliffs, Prentice-Hall, 1973.

31. Jelliffe DB, Jelliffe EFP: Doulas, confidence, and the science of lactation. J Pediatr 84(3):462-464, 1974.

32. Loughlin HH, Clapp-Channing NE, Gehlbach SH, et al: Early termina- tion of breast-feeding: identifying those at risk. Pediatrics 75(3):508-513, 1985.

33. Grassley J, Davis K: Common concerns of mothers who breast-feed. The American Journal of Maternal Child- Nursing 3:347-351, 1978.

34. Bergevin Y, Dougherty C, Kramer MS: Do infant formula samples shorten the duration of breast-feeding? Lancet i:1148-1151, 1983.

35. Reiff MI, Essock-Vitale SM: Hos- pital influences on early infant-feeding practices. Pediatrics 76(6):872-878, 1985.

36. Gussler J, Briesemeister LH: The insufficient milk syndrome: A biocultural explanation. Med Anthropol 4(2):3-24, 1980.

37. Gunn TR: The incidence of breast feeding and reasons for weaning. NZ Med J 97:360-363, 1984.

38. Sloper K, McKean L, Baum JD: Factors influencing breast feeding. Arch Dis Child 50:165-170, 1975.

39. Ekwo EE, Dusdieker L, Baoth B, et al: Psychosocial factors influencing the duration of breastfeeding by primigra- vidas. Acta Paediatr Stand 73:241-247, 1984.

40. Van Esterik P, Greiner T: Breast- feeding and women’s work: Constraints and opportunities. Stud Fam Plann 12(4):184-197, 1981.

41. Simopoulos AP, Grave GD: Factors associated with the choice and duration of infant-feeding practice. Pedi- atrics 74(Supp1):603-613, 1984.

42. Winikoff B: Issues in the design of breastfeeding research. Stud Fam Plann 12(4):177-184, 1981.

43. Popper KR: The logic of scientific discovery. New York, Harper & Row, pp 88-89, 1968.

44. Puterman ZM The concept of the causal connection. Uppsala, Uppsala Universitet, p 25, 1977.

45. Campbell DT, Stanley JC: Expen- mental and quasi-experimental design for research. Chicago, Rand McNally College Pub1 Co., 1963.

46. Bhattachayya GK, Johnson RA: Statistical Methods and Concepts. New York, John Wiley & Sons, 1977.

47. Morse JM, Harrison MJ, Prowse, M: Minimal Breastfeeding, J Obstet Gy- necol Neonatal Nurs 15(4):333-338, 1986.

48. Morse JM, Harrison MJ, Williams KM: What determines the duration of breastfeeding? in Michaelson K (ed): An- thropology of Childbirth in America. S. Hadley, Mass., Bergin & Garvey, in press.

49. Ellis DJ, Hewat RJ: Do nurses help or hinder others who breastfeed? J Adv Nurs 8:281-288, 1983.

50. Avery JL: Closet nursing: A symptom of intolerance and a forerunner of social change? Journal of Keeping Abreast 2:212-226, 1977.

210 Journal of Nurse-Midwifery ??Vol. 32, No. 4, July/August 1987

social coercion for weaning

Documents