From concept to measurement: operationalizing WHO’s definition of unsafe abortion

The World Health Organization (WHO) defines unsafe abortion as a procedure for terminating a pregnancy performed by persons lacking the necessary skills or in an environment not in conformity with minimal medical standards, or both. This definition embodies concepts first outlined in a 1992 WHO Technical Consultation.1 Although widely used, it is inconsistently interpreted. In this editorial we discuss its correct interpretation and operationalization. WHO’s definition of unsafe abortion was conceptualized within the framework of emerging guidelines on the management of the complications of induced abortion and was intended to be interpreted within that context. This linkage to technical guidelines is crucial for its correct interpretation. Nothing in the definition predetermines who should be considered a “safe” abortion provider or what the appropriate skills or standards for performing abortions should be. Such things are not static; they evolve in line with evidence-based WHO recommendations. For example, WHO guidelines now recommend mifepristone and misoprostol – or misoprostol alone if mifepristone is not available – and vacuum aspiration in lieu of the sharp curettage used formerly. They now consider induced abortions provided at the primary care level or by non-physician health-care providers as safe.2 The guidelines on task shifting that are being developed are expected to clarify who can safely provide an abortion under current standards. To ensure that “unsafe abortion” is correctly interpreted, we recommend always providing an explanatory note along with the definition, as follows: “The persons, skills and medical standards considered safe in the provision of abortion are different for medical and surgical abortion and also depend on the duration of the pregnancy. What is considered ‘safe’ should be interpreted in line with current WHO technical and policy guidance.” Although unsafe abortions are, by definition, risky, safety cannot be dichotomized because risk runs along a continuum. Risk is lowest if an evidence-based method is used to terminate an early pregnancy in a health facility;3 it is highest if a dangerous method, such as the use of caustic substances orally or vaginally or the insertion of sticks into the uterus, is employed clandestinely to terminate an advanced pregnancy. There is a spectrum of risk between these two extremes. Along that spectrum, for example, lie cases of self-administration of misoprostol or the use of outdated procedures, such as sharp curettage, by skilled health-care providers. The immediate determinants of the risks of an induced abortion, such as the termination method used and gestational age, are influenced, in turn, by underlying social determinants: i.e. the legal context, the availability of safe abortion services, the level of stigma surrounding abortion, the degree of women’s access to information on abortion, and a woman’s age and socioeconomic status. The legal context and the level of safety are closely intertwined, but the association is context-specific. For example, where restrictive laws are liberally interpreted, women can receive safe care in certain contexts; conversely, where liberal laws are poorly implemented, women sometimes abort with delay and under unsafe conditions. Thus, illegal abortion is not synonymous with unsafe abortion, as indicated by the original definition: “…legality or illegality of services, however, may not be the defining factor of their safety … the safety of abortion must be considered within both the legal and legally restricted contexts.”1 Rates of induced abortion are difficult to measure because of frequent underreporting or misclassification in surveys, hospital records and health statistics.4 In light of this, WHO has historically used a pragmatic operational construct that measures safety in terms of only one dimension – legality – in developing its regional and global estimates of rates of unsafe abortion.4,5 However, the widespread informal use of misoprostol has added a layer of complexity to the concept of “safety”. As a result, it has become essential to apply a multi-dimensional risk continuum to measure abortion safety. The adverse outcomes associated with unsafe abortion need to be measured as well. Since deaths resulting from unsafe abortion have decreased in recent years,4 perhaps because of safer methods, the focus should now be broadened from mortality to morbidity as well. A multi-dimensional assessment of the safety of induced abortions, as described, makes estimation more difficult, but the more nuanced measures involved could generate more innovative research and improve the data collected locally and nationally. Assessing the safety of induced abortion does not suffice, however. In the longer term, global consensus will be needed on the broader indicators used to assess the provision of safe abortion in line with WHO guidance – i.e. indicators capturing access, equity, quality of care and linkages to post-abortion contraception.