Identifying data for the empirical assessment of law (IDEAL): a realist approach to research gaps on the health effects of abortion law

Reproductive rights have been the focus of United Nations consensus documents, a priority for agencies like the WHO, and the subject of judgments issued by national and international courts. Human rights approaches have galvanised abortion law reform across numerous countries, but human rights analysis is not designed to empirically assess how legal provisions regulating abortion shape the actual delivery of abortion services and outcomes. Reliable empirical measurement of the health and social effects of abortion regulation is vital input for policymakers and public health guidance for abortion policy and practice, but research focused explicitly on assessing the health effects of abortion law and policy is limited at the global level. This paper describes a method for Identifying Data for the Empirical Assessment of Law (IDEAL), to assess potential health effects of abortion regulations. The approach was applied to six critical legal interventions: mandatory waiting periods, third-party authorisation, gestational limits, criminalisation, provider restrictions and conscientious objection. The IDEAL process allowed researchers to link legal interventions and processes that have not been investigated fully in empirical research to processes and outcomes that have been more thoroughly studied. To the extent these links are both transparent and plausible, using IDEAL to make them explicit allows both researchers and policy stakeholders to make better informed assessments and guidance related to abortion law. The IDEAL method also identifies gaps in scientific research. Given the importance of law to public health generally, the utility of IDEAL is not limited to abortion law.


I.
Parental Involvement in Minor's Abortion 3 II.
Spousal Consent for Abortion 6 III.
Mandatory Waiting Period Requirement 14 VI.
Legally Prohibited Abortion 34

Introduction to the Data Supplement
This supplemental document provides detailed results from our study using the IDEAL process (Identifying Data for the Empirical Assessment of Law) investigating six critical legal interventions for abortion: mandatory waiting periods, third-party authorization, 1 gestational limits, criminalization, provider restrictions, and conscientious objection. This supplement contains the full set of causal models developed during the study, along with tables that summarize the causal pathways and provide examples of relevant non-legal studies identified through the research process. The purpose and methods of the IDEAL process and a sample of the results are further described in the published paper.
The IDEAL study was an exploratory test of the method and developed in connection with the revision of WHO Safe Abortion: Technical and Policy Guidance for Health Systems. The WHO process for developing guidelines includes a rigorous literature review. IDEAL was developed and tested to help reviewers identify and draw on existing data to explore questions of health effects of abortion law, by identifying potentially important legal questions and pointing to examples of studies that addressed them. The study itself did not aim to identify all relevant research, or to select or classify examples based on rigor. References to specific research studies in this supplement are exemplary, rather than exhaustive or critical.
Finally, we note that the causal models we created do not explicitly include travel to a more permissive jurisdiction as a response to legal restrictions in a pregnant person's home country, or province. This alternative for accessing services may arise anywhere legal restrictions hamper local abortion access, and the phenomenon has been studied in many different legal and regional contexts. 2 Travel to visit an abortion provider, including in another jurisdiction, can be a source of higher costs and delayed healthcare, which are common outcomes covered in the causal models presented. services, as well as logistical barriers. Some minors may be unable to complete the process, shifting to the parental involvement or non-compliance pathways. Should the alternative procedure not lead to a lawful abortion, the minor may give birth, obtain a legally prohibited abortion or shift to the parental involvement pathway. 3 Pathway D (red) represents the minor's non-compliance with the parental involvement law, leading to an unintended birth or a legally prohibited abortion. The minor may alternatively choose to travel to another jurisdiction without legal restrictions to obtain an abortion.
All pathways could result in delayed health care and increased costs associated with the law (see Models IX and X). Table I reports research we identified through the IDEAL process that provides findings on these processes and/or outcomes. Pathway D (red) represents the pregnant person's non-compliance with the spousal involvement law, leading to an unintended birth or a legally prohibited abortion. Mediating factors on a pregnant person's decision to involve their parent or not can depend on characteristics of the spousal relationship, financial ability to seek services, and/or abortion stigma.
All pathways could result in delayed health care and increased costs associated with the law. Table II reports research we identified through the IDEAL process that provides findings on these processes and/or outcomes. Each of these pathways may also contribute to increases in delayed care or costs. Travel to another jurisdiction to avoid the rule (not shown) can also increase costs or cause delay. Table III reports research we identified through the IDEAL process that provides findings on these processes and/or outcomes. Pathway D (green) depicts the pathway of a pregnant individual seeking an abortion within the gestational age limit set by law and obtaining a lawful abortion.
Pathways B and C may contribute to delayed care or increased costs even for those who obtain a legal abortion. Travel to another jurisdiction to avoid the rule (not shown) can also increase costs or cause delay. Table IV reports research we identified through the IDEAL process that provides findings on these processes and/or outcomes. abortion services, which triggers the waiting period. The time required for this process will determine how close the individual is to a gestational age limit.
Pathway (red) depicts the pathway of a pregnant individual who is unable to return for care at the expiration of the waiting period requirement due to logistical barriers such as finances and childcare, and accessibility of clinics or providers. It also captures the case of compliance with the waiting period and related delays leading to the pregnant individual reaching a gestational limit. As a result of inability to comply, the individual may have an unwanted or unintended childbirth, or seek an abortion outside legal parameters. (Not shown: the individual who reaches the gestational limit may qualify for an exception and be able to secure a legal abortion; see Model IV.) Pathway C (green) depicts the pathway of a pregnant individual who is able to return for service after the waiting period requirement and obtain a lawful abortion. This includes individuals who qualify for an exemption, such as medical emergency. Compliance with the law may contribute to increased costs and delay in obtaining care.
Both of these pathways may also contribute to increases in delayed care or costs. Travel to another jurisdiction to avoid the rule (not shown) can also increase costs or cause delay. Table V reports research we identified through the IDEAL process that provides findings on these processes and/or outcomes.

Causal Process
Intermediary or Primary Outcome(s)

Selection of Relevant Studies
The impact of mandatory waiting period laws is mediated by other individual and contextual factors influencing the gestational point at which the pregnant individual presents for an abortion (Pathway A) Pregnant individual presents to a provider for an abortion  Pathway B (green) depicts the path to lawful abortion for those who are able to obtain abortion services.
Pathway C (orange) depicts the results for individuals unable to access lawful abortion as a result of the decrease in abortion providers. The lack of available abortion providers may result in abortion obtained outside legal parameters or unintended childbirth.
Each of these pathways may also contribute to increases in delayed care or costs. Travel to another jurisdiction to access services (not shown) can also increase costs or cause delay. Pathway C (orange) depicts lack of access to abortion within legal parameters, leading to unintended childbirth or seeking a legally prohibited abortion.
Each of these pathways may also contribute to increases in delayed care or costs. Travel to another jurisdiction to access services (not shown) can also increase costs or cause delay.
Table VII reports research we identified through the IDEAL process that provides findings on these processes and/or outcomes. Pathway C (orange) depicts possible causal chains arising from an objecting medical provider or institution. Legal provisions may or may not require the objector to provide a referral to a willing abortion provider. Pregnant individuals who eventually find a non-objecting provider may suffer emotional harm and stigmatization, experience delay in obtaining an abortion, and/or incur higher costs due to the need to travel or manage other logistical challenges. 11 Pathway D (red) depicts a pregnant individual who is unable to find a willing provider, particularly in an environment with restricted abortion access. The pathway may result from the general lack of willing providers due to conscientious objection, or inability to find a willing provider at all or within applicable gestational limits. Pregnant individuals unable to access lawful abortion services in time may be faced with unintended childbirth or obtaining abortion outside legal parameters, with delay and cost compounding the effects of provider refusal. 12 These may have negative health and socioeconomic outcomes, as well as health systems costs.
Travel to another jurisdiction to access services (not shown) can also increase costs or cause delay.
Table VIII reports research we identified through the IDEAL process that provides findings on these processes and/or outcomes.

Causal Process
Intermediary or Primary Outcome(s)

Selection of Relevant Studies
Conscientious objection law, abortion stigma, and provider training infrastructure interact (Pathway A) Availability and accessibility of providers     This pathway describes a pregnant individual who does not qualify for a legal abortion but obtains an abortion less safe or least safe. Fear of abortion stigma may influence an individual's decision to obtain unsafe abortion and deter them from seeking care for complications. Lack of health services and infrastructure may also factor into unsafe abortions. Abortion complications and maternal morbidity or mortality that result from less-safe or least-safe abortions can lead to poorer health and socioeconomic outcomes, as well as increased costs. A less or least safe may also result I criminal prosecution. 14 Delayed care and increased results may arise in both pathways.
Table XII reports research we identified through the IDEAL process that provides findings on these processes and/or outcomes.

Causal Process
Intermediary or Primary Outcome(s)

Selection of Relevant Studies
Health system characteristics including the availability of abortion information and abortion stigma interact with law prohibiting abortion to afford access to a safe abortion outside legal parameters (Pathway A) Pregnancy is terminated through a safe abortion