The Decline in Maternal Mortality in Sweden: The Role of Community Midwifery ~ Ulf Högberg, MD, PhD (part 6)

by faithgibson on December 1, 2012

Link back to part 7 of “Story I Hate To Tell & Nobody Should Ever Have to Hear” 

August 2004, Vol 94, No. 8 | American Journal of Public Health 1312-1320

© 2004 American Public Health Association PUBLIC HEALTH THEN AND NOW

The author is with Obstetrics & Gynecology, Department of Clinical Science, and Epidemiology, Department of Public Health and Clinical Medicine, University of Umeå, Umeå, Sweden.

Correspondence: Requests for reprints should be sent to Ulf Högberg, MD, PhD, Obstetrics & Gynecology, Department of Clinical Science, University of Umeå, S-901 87 Umeå, Sweden (e-mail:


The maternal mortality rate in Sweden in the early 20th centurywas one third that in the United States. This rate was recognizedby American visitors as an achievement of Swedish maternitycare, in which highly competent midwives attend home deliveries.The 19th century decline in maternal mortality was largely causedby improvements in obstetric care, but was also helped alongby the national health strategy of giving midwives and doctorscomplementary roles in maternity care, as well as equal involvementin setting public health policy.

THE DECLINE OF MATERNAL mortality in Western countries afterthe 1930s is believed to be associated mainly with the emergenceof modern obstetric care, while it has been proposed that publichealth policy, poverty, and the malnutrition associated withpoverty were of relatively minor importance.1 But the maternalmortality pattern before the emergence of modern medical technologywas not uniform in all Western countries.

In The Netherlands,Norway, and Sweden, low maternal mortality rates were reportedby the early 20th century and were believed to be a result ofan extensive collaboration between physicians and highly competent,locally available midwives.2

From 1900 through 1904, Swedenhad an annual maternal mortality of 230 per 100 000 live births,while the rate for England and Wales was 440 per 100 000. Forthe year 1900, the United States reported 520 to 850 maternaldeaths per 100 000 live births.3 This very high maternal mortalityrate, especially if compared with the lower rates achieved inseveral less prosperous European countries, caused some Americanobstetricians to express concern.

The 20th century decline in maternal mortality, seen in allWestern countries, was made possible by the emergence of modernmedicine. However, the contribution of the mobilization of humanresources should not be underestimated, nor should key developmentsin public health policy.

Joseph B. DeLee, commemorated as a titan of 20th-centuryobstetrics, studied maternity services in Europe before he establishedthe Chicago Lying-In Hospital and Dispensary in 1895. His aimwas to provide delivery assistance to poor women by also offeringthem the option of having a safe and inexpensive home delivery.4

George W. Kosmak5 visited Scandinavia in 1926 and was reportedto have been very impressed with the medical systems in placethere. In an address to the American Medical Association, Kosmaktalked about the good results obtained in a carefully supervisedsystem of midwife instruction and practice. He stated,

To begin with, the midwife in Scandinavia is not regarded aspariah. . . . One sees, therefore, in the training schools formidwives, bright, healthy looking, intelligent young women ofthe type from whom our best class of trained nurses would berecruited in this country, who are proud of being associatedwith an important community work, and whose profession is recognizedby medical men as an important factor in the art of obstetrics,with which they have no quarrel.

He concluded, “The results of this midwife training are evidentlyexcellent because the mortality rates of these countries areremarkably low and likewise, the morbidity following childbirth.”5

What, then, was the history of this system that turned out tobe a good example for the United States before the emergenceof modern medicine in the 1930s? The aim of this review is todepict the Swedish intervention against maternal mortality inthe 18th and 19th centuries and the decline in maternal mortalityin the Western countries in the 20th century.

Historical Setting: Sweden

The history of maternity care in Sweden should be interpretedin light of the involvement of the state in public health. Oneimportant part of the emergence of the Swedish national statein the 16th century was the creation of the Lutheran State Church. In the 17th century, the Swedish clergy created an informationsystem that included all individuals in their parishes olderthan 6 to 7 years. By the middle of the 18th century, this registrationincluded the entire population. The information system was based on the annual catechetical examination of every household, where the clergy examined knowledge of the catechism as well as the reading ability of all household members. To this “church book,”other types of records were linked: records of in- and outmigrations,births and baptisms, bans and marriages, and deaths and burials.The Office of the Registrar General (Tabellverkskommissionen),founded in 1749, compiled national statistics from the ecclesiasticalregistry. National vital statistics were therefore availablein Sweden before they were available in any other European country.

The profession of physician was legalized in 1663 with the foundationof the Collegium Medicum. In the 17th and 18th centuries, many Swedish academics obtained their postdoctoral training from universities in Germany, France, Italy, England, and The Netherlands. By the beginning of the 18th century, Sweden had declined as a major power in northern Europe. Inside Sweden, the power of the Swedish parliament was enhanced; a so-called “Time of Freedom” was introduced that coincided with the Age of Enlightenment.There began an era of scientific blossoming.

The two professors of medicine at Uppsala University, Carl von Linné (1707–1778)and Nils Rosén von Rosenstein (1706–1773), and the head of the Collegium Medicum, Abraham Bäck (1713–1795),were the initiators and promoters of health care and public health within the Commission of Health (Sundhetskommissionen)from 1737 to 1766. They presented programs for primary healthcare and preventive measures for communicable diseases and published pamphlets on health education, nutrition, and hygiene. From the start, the public health program had an equity perspectiveby reaching out to the poor rural population and making healthcare accessible to them. The policy fit in with the prevailing political ideology of the time, mercantilism, which definedthe wealth of the nation by the number of its citizens.6 The military need of the nation has also been proposed as an argumentfor investment in mothers’ and children’s health.7

The first national statistics on maternal mortality were presented in 1751, revealing a rate of almost 900 maternal deaths per 100, 000 live births. In the same year, the Commission of Health stated, “Out of 651 women dying in childbirth, at least 400 could have been saved if only there had been enough midwives.”

This became the starting point for the Swedish authorities to campaign for improvements in obstetric care, mainly by improving training for physicians and midwives and implementing a system of surveillance {i.e.accountability to healthcare officials} of midwives, both at the county and national level. What they did not know at the time was that it would take 150 years to achieve their goal.6


The professionalization of birth assistance in Sweden beganin the early 18th century. Pioneering this was Johan von Hoorn (1662–1724), who trained in obstetrics at the Hotel DieuHospital in Paris before returning to Sweden. In 1697, von Hoorn published a textbook titled The Well-Trained Swedish Midwife (Den Swenska wäl-öfwade Jord Gumman) intended for use by both midwives and the public.

In 1711, the Collegium Medicum announced a decree of authorization for midwives that requireda 2-year training period with an experienced midwife, followed by an examination given by the Collegium Medicum. In 1715, vonHoorn published a textbook for midwifery training with Soranus,the famous Roman gynecologist (50–129 AD), as a sourceof inspiration; in it, he stressed the importance of surveillance of the delivery by internal examination—that is, the non-interventionist approach emphasizing patience and waiting. He also described the mouth-to-mouth resuscitation method for reviving an apparently dead newborn. Soon the need for licensed midwives became apparent and the Collegium Medicum urged Sweden’s parliament to push for a national midwifery school. However, it was not until the end of the century that such a school was started.

In 1757, the Collegium Medicum’s proposal for a national training program for midwives covering all parishes was finally approved. Each parish was expected to pay for its students’allowance in Stockholm. The first professor in obstetrics was appointed in 1761, and the first lying-in hospital, AllmännaBarnbördshuset in Stockholm, was founded in 1775.

The founding of Stockholm’s Karolinska Institute in 1810 led to a further improvement of obstetric care at a national level. A new government decree stated in 1819 that every parish was required to employ a licensed midwife, and that the parishes were also responsible for variola (smallpox) vaccinations. The midwife’s formal education was extended to 6 months, and the government paid allowances for 12 students each year. This meant that instead of limiting the training program to the women sent by the parishes, the profession was opened up to all interested women.

The professor of obstetrics at the time, Pehr Gustaf Cederschiöld(1782–1848), pushed hard to increase the competence of midwives. By 1829, health reform brought new regulations authorizing midwives, after an extended training period, to use forceps, sharp hooks, and perforators, in addition to their ability to perform manual removal of the placenta and extraction in breech presentation.

This reform was opposed by contemporary internationalmedical societies7 but was motivated by the long tradition of community midwives who assisted at home deliveries. The widely scattered rural Swedish population made it a necessity for midwives to be capable of acting in emergencies when physicians could not be reached. Cederschiöld argued that the reform would strengthen the authority and acceptance of the midwife in the parishes.8 Cederschiöld then wrote the textbooks Manualfor Midwives (Handbok för Barnmorskor) and Guide to Instrumental Obstetrics (Utkast till Handbok i den Instrumentala Förlossningskonsten) in support of his ambition to increase the competence of midwives.

By the government decree of 1819, midwives were required to ensure that every newborn child had his or her own bed to prevents uffocation, although little observance of this rule was reported.9In the mid-19th century, the authorities added more regulations for midwives. It was decided that their duties should not belimited only to childbirth, but should also include subsequent care of the infant. Consequently, education in basic neonatal care at the midwifery school was improved, with an emphasis on warmth, neonatal resuscitation with tactile stimuli for asphyctic children, daily care of the umbilicus, and early breastfeeding. Many mothers fed their newborns cows’ milk, and doctors and midwives began informing young mothers and mothers-to-be about the benefits of breastfeeding. This strategy soon had the desired effect, and infant mortality was reduced by 20%.10

The antiseptic technique was introduced in the lying-in hospitals during the late 1870s and, by law, to midwives in rural districtsin 1881. Also, the Credé prophylaxis to prevent neonatal blennorrhea became one of the midwife’s duties.


The professionalization of birth attendance was NOT a smooth process. Historian Christina Romlid describes the antagonism,struggles, and conflicts that arose between the medical profession and traditional birth attendants until the late 19th century.8In the Swedish parliament, the peasantry protested against the midwife regulation of 1777. This rule contained a “quackery paragraph” that banned traditional birth attendants, whom the peasants viewed as experienced and skilled, not as dangerous and harmful as stated in the regulation.

Subsequently, the Crown withdrew the paragraph and reinstated the right of district medical officers and licensed midwives to train women locally.The paragraph was reinstated in 1819 in a milder form, allowing traditional birth attendants when a licensed midwife could not attend or arrive in time. However, during the 19th century, several traditional birth attendants were prosecuted and found guilty of unauthorized help during childbirth.8 Not until the late 19th century did professional midwifery become fully established and legitimized in the rural areas of Sweden.

Whereas during the 18th century midwives were recruited from among farming families, by the 19th century the profession of midwife had become a legitimate occupation for women from all walks of life, and it carried as much weight and respect as that of primary school teacher.11 Consequently, the community midwife became a central figure and was often the only person representing health care at the parish level. Over time, any technical constraints were overcome and there was good social representation among midwives, thus ensuring a successful implementation of obstetric services within the specific cultural context of rural Sweden.

The professionalization of birth assistance can be interpreted from a gender theory perspective as a successive subordination of women consequent to the appearance of male obstetricians. Birthing is a natural event, yet female traditional birth attendants were pushed aside with the medicalization of childbirth. The American and British experience of conflicts between doctorsand midwives is a recurrent theme, and Swedish historians have reported parallels in Sweden, although more in Stockholm than in the rural areas.8,12 However, studies addressing the professionalization of Swedish midwives in relation to the theories of sociology, modernity, gender, and the evolution toward scientifically based obstetric care have found few conflicts between doctors and midwives.11 There was a gender division in the professionalization process; however, since doctors and midwives were disseminators of the same discourse and worked toward the same goal, theycomplemented rather than competed against each other, unlike in the US urban setting.11

These complementary roles were facilitated by the conditions of health care in Sweden. As recently as the late 19th century, only 10% of the Swedish population lived in urban areas. Obstetricianswere in office only in the lying-in hospitals of Stockholm and, from 1865, also in Gothenburg and Lund. Otherwise, general practitionersin the counties and towns were the medical counterparts of the midwives assisting at home deliveries. In practice, no system of referral was available in the 19th century. In medical emergencies, the midwife called for the doctor, but this rarely happened. This setting facilitated a more non-interventionist attitude, manifesting fairly low rates of assisted delivery throughout Swedish history and strengthening the midwife in her role as the indisputable birth attendant, in contrast to the more doctor-oriented obstetrics emerging in the United States by the 20th century.11

The Swedish model of maternity services was distinct even from the European perspective. In 1870, the ratio was 3.1 midwives for every doctor for Sweden, while it was 1.4 in Denmark andNorway7 and 1.2 in France.13

Community midwifery was based on a system of very close supervision and retraining. In each county, each midwife was required to report to the county general practitioner. Her report had to be detailed and include the actual record, in diary form, of all deliveries she had attended, with information on the identity of the parturient, complications, the sex of the child, birthweight, and outcome for the mother and child. Also, review courses for midwives were obligatory on a regular basis. A standardized protocol was necessary when midwives used forceps, sharp hooks, or perforators, giving the reasons for the intervention and the outcome. This protocol had to be signed by the county physician and was registered at the National Health Bureau.


In the 17th century, maternal deaths accounted for 10% of allfemale deaths between the ages of 15 and 49 years.14 In womenaged between 20 and 34 years, 40% to 45% of deaths among marriedwomen were caused by complications of pregnancy or delivery.Among married women, 1 of 14 died during childbirth.15

Maternal mortality declined from 900 per 100 000 live birthsto 230 per 100 000 from 1751 to 1900. The general trend towarda decline was interrupted during the years 1850 to 1880, whenthe recorded septic maternal mortality coincided with an increasein total mortality due to communicable diseases. During the19th century, areas of high maternal mortality were not restrictedto the urban environments, where there was a known high deathrate due to puerperal sepsis.14

During the 19th century, the decline in maternal mortality wasfar greater than that in infant mortality, or in mortality dueto tuberculosis. The decline in maternal mortality was especiallypronounced between 1861 and 1900, when the percent reductiondropped from 59% to 24%, while the female mortality reductionleveled out.16

In the 19th century, two thirds of maternal deaths had directobstetrical causes, such as difficult labor, eclampsia, hemorrhage,and sepsis, while one third were indirect obstetric deaths dueto diseases such as pneumonia, tuberculosis, dysentery, heartdisease, and malnutrition.15,17 In the lying-in hospitals, beforeantiseptic techniques became known most maternal deaths werecaused by puerperal sepsis.18 However, the epidemics of puerperalsepsis in the lying-in hospitals did not dramatically alterthe national maternal mortality rates. Between 1775 and 1900,a total of 1720 parturients were recorded to have died frompuerperal sepsis in the lying-in hospitals, which represents2.2% of all maternal deaths during the period. It was duringthe second half of the 19th century, when the national statisticsrecorded puerperal sepsis separately, that the nationwide problembecame obvious. Between 1861 and 1900, 54% of maternal deathswere caused by puerperal sepsis, most of them following homedeliveries. This percentage was even higher for home deliveriesbefore the introduction of antiseptic technique,18 possiblyalso caused by an increased virulence of the dominant strainof streptococcus at the time.1 The diagnosis of puerperal sepsiswas probably not confounded by septic abortions during the 19thcentury.16

The adverse effects of medical technology were predisposing,positive risk factors. Before the introduction of antiseptictechniques, lying-in hospitals were a positive risk factor inthe transmission of puerperal sepsis. As can be seen by extrapolatingfrom the mortality rate of puerperal sepsis between 1881 and1895 (after the introduction of antiseptic techniques), if suchtechniques had been available from 1776 through 1900, the numberof puerperal deaths in lying-in hospitals would have been 119instead of 1720. The difference, 1601 deaths, is a measure ofthe potentially adverse effects that the lying-in hospitalshad on the number of maternal deaths nationwide from 1776 through1900 (n = 76,776).

However, the protective effect of these hospitals as educational centers for midwives and physicians practicingin rural areas has not been considered.16