The politics of location: when and where knowledge is generated

Psychology research interest in abortion emerged in the 1970s, with most attention being paid to the issue in the 1990s. Zero articles appeared in the 1960s, 15.7% of the dataset in the 1970s, 22.9% in the 1980s, 30.7% in the 1990s, 24.1% in the 2000s and 6.6% in 2010–2015.

Knowledge production is dominated by research conducted in North America, with over half of the articles being generated in this context (table 1). All articles generated in sub-Saharan Africa come from South Africa.

Table 2 shows that knowledge about abortion in the 1970s was predominantly produced in North America (17 articles). The dominance of North America in knowledge production continued into the 1990s (35 articles in that decade). Knowledge production from Europe and sub-Saharan Africa (South Africa) increased proportionally to other regions in the 2000s compared with previous decades.

The politics of representation: type of research and issues studied

Empirical quantitative methods dominate (69.3%). Over the full timespan, only 9% of articles use qualitative methods, 9.6% are reviews, 9.6% are commentaries, 1.2% mixed methods and 1.2% theoretical.

Two topics dominate knowledge production: the psychological consequences of abortion (41%) and attitudes towards abortion (28.9%). Abortion decision making (10.2%), abortion services (6%), the factors associated with abortion(4.8%), the experience of abortion (4.2%), the incidence of abortion (3%) and unsafe abortion (0.6%) received scant attention over the years.

Linking the politics of location and of representation

A cross-tabulation of decade by type of research showed that empirical quantitative articles dominated throughout the decades (lowest at 63.2% in the 1980s and 76.9% in the 1970s). Qualitative articles only started to appear in the 1980s (5.3% of articles), with 27.3% of articles appearing in the last 6 years being qualitative.

In table 3 results of the cross-tabulation of decade and main topic are presented. The proportion of articles dedicated to the psychological consequences of abortion has steadily increased from the 1970s to the 2000s, with the trend continuing in the last 6 years. Over the decades, the proportion of articles exploring attitudes to abortion has varied from about a quarter to a third (minimum eight and maximum 15 articles in a single decade). The proportion of articles concentrating on the experience of abortion has steadily increased, a trend that may be associated with the increase in articles using qualitative methodologies.

A cross-tabulation of region of first author affiliation and type of research shows a dominance of quantitative research written by authors from North America with only 2.1% of articles being qualitative. In contrast, 53.9% and 23.1% of the articles by European-based first authors were quantitative and qualitative, respectively, while in South Africa the trend is reversed (27.3% and 45.5%, respectively).

In table 4A cross-tabulation of regions in which more than 10 articles over the period under review were produced (understood through first author affiliation) and main topic is presented. Similar levels of interest in the psychological consequences of abortion are noted across the three main regions. Outside of this main issue, European authors appear to have concentrated on attitudes to abortion, while North American writers have tackled a range of topics. The only article tackling unsafe abortion emanated from South Africa.

Representations of the psychological consequences of abortion in the 1970s and 2000s

In this section we outline how the dominant topic of research as revealed in the above analysis was approached in the 1970s and 2000s. This snapshot provides a brief insight into developments in constructions of this topic. Given space constraints, this analysis is necessarily truncated.

Research on the psychological consequences of abortion appearing in the 1970s, all of which emanated from the United States in the decade in which abortion was legalised, for the most part indicates that there are no or few negative outcomes. Gordon6 and Shusterman7 conducted literature reviews of studies concerning the psychological outcomes of abortion. After critiquing the methodologies of some studies (lack of adequate control group) and pointing to the falsity of comparing studies conducted in different cultural and legal contexts, Gordon concluded that "legal abortion has few serious psychological sequelae" (p.45). Shusterman’s review, which included studies on the medical and psychological consequences of abortion, reached a similar conclusion: "the psychological consequences of abortion on request appear to be mostly benign" (p. 79). Adler8 suggested, however, that methodologically studies are unlikely to pick up women who experience negative consequences.

Authors sought to locate the suggestion of inevitable psychological consequences following abortion within reproductive and gendered context. Thus, Gordon6 postulated that any "inroad on a person’s body" (p.39), including birth, has some psychological component. Shusterman7 indicated that emotional responses are related to the attitudes of medical staff to women presenting for abortion. Fingerer9 argued that anticipation of depression is associated with "society’s acceptance of the 'motherhood myth', that is that woman’s highest function is to be a good mother" (p.223–4). Martin10 investigated the influence, inter alia, of women’s relationship with the male responsible for the pregnancy, parents, siblings, and peers, religious and moral convictions, and medical care on psychological responses to abortion.

Methodological critiques were accompanied by debates about the best way to measure psychological sequelae. Robbins11 argued that there was no relation between objective measures, such as the Minnesota Multiphasic Personality Inventory, and women’s self-report, concluding that "affectivity reported by some women after induced abortion may have little impact on overall psychological functioning" (p.994). Gayton et al 12 suggested that while "the majority of women who receive abortions seldom manifest serious psychiatric sequalae" (p.649), a screening instrument such as the Mini-Mult was necessary for high-risk cases.

By the 2000s the notion of abortion potentially constituting a form of trauma had taken firm root. Much research was conducted to explore the factors associated with post-traumatic symptomatology following abortion,13–15 including "recognition of the life of the fetus, attachment" (pp.14, 41). The term post-abortion syndrome was taken up as a real clinical diagnosis in the description of therapeutic cases.16

The range of consequences researched expanded considerably to include not only anxiety and depression, but also difficulties maintaining committed relationships, sexual dysfunction,17 quality of child-caring and children’s development,18 substance use,19 attachment styles,20 grief,21 self-efficacy,22 suicidal behaviour,23 shame, embarrassment, guilt, and moral concerns.24 Some research homed in on young women or minors specifically,23 25 and reviews of the evidence were conducted.26–28

There was, however, also dispute concerning the use of the term post-abortion syndrome.29 Researchers pointed to the methodological weaknesses in studies that found negative psychological consequences.28 Others argued that there were a diversity of responses,26 and that reactions are located in social context,30 including a history of violence and abuse.31

The entrenchment of traumatology talk and the increase in the kinds of negative consequences envisaged and researched meant that significant labour was necessary to undermine such claims. To this effect, a major study was commissioned by the American Psychological Association.28 The researchers pointed to many methodological flaws in studies. They concluded that the most rigorous research showed that "the relative risk of mental health problems among adult women who have a single, legal, first-trimester abortion of an unwanted pregnancy is no greater than the risk among women who deliver an unwanted pregnancy" (p.863).