CRD summary

The review found that women who had an abortion were at higher risk of adverse mental health outcomes such as depression, anxiety, alcohol use, marijuana use and suicidal behaviour. The results of the review should be treated with caution due to concerns with the review methodology and reporting and the uncertain appropriateness of the statistical analysis.

Authors' objectives

To summarise associations between induced abortion and adverse mental health outcomes and estimate the population-attributable risk percentages for various mental health outcomes.

Searching

MEDLINE and PsycINFO were searched for relevant studies published in English between 1995 and 2009. Search terms were not reported.

Study selection

Studies that reported mental health outcomes in women who underwent abortion compared with women who completed a pregnancy were eligible for inclusion if they included at least 100 participants, controlled for confounding factors and reported odds ratios (ORs).

Sample populations were largely from USA; one study was from Australia. The comparison group for half of the studies was "unintended pregnancy delivered". Outcomes assessed were depression, anxiety, alcohol use, marijuana use or suicidal behaviour.

The author did not state how many reviewers performed the study selection.

Assessment of study quality

The author did not state that she assessed validity.

Data extraction

Where more than one effect per outcome was reported for a particular study, a composite odds ratio was used. Where studies had more than one comparison group, priority was given first to the "unintended pregnancy delivered" group, then the "pregnancy delivered" group and finally the "no abortion" group. Where studies stratified their results on number of abortions, results for women who had one abortion were used. Where authors used the same sample and variables in more than one publication only the most recent publication was selected. Where distinct samples from the same data set were published by different authors, both sets of results were included.

The author did not state how many reviewers performed the data extraction.

Methods of synthesis

Random-effects meta-analyses were conducted to calculate pooled odds ratios and associated 95% confidence intervals (CI). Subgroup analyses were calculated based on the type of comparison group and on specific mental health outcomes. The author did not state whether statistical heterogeneity was assessed. Each study could contribute several measures of effect to the overall analysis if several outcomes were reported.

Results of the review

Twenty-two studies (877,181 participants) were included in the review, with 36 comparisons when various outcomes from each study were assessed individually.

Women who had an abortion were more likely to report adverse mental health outcomes compared with women who completed a pregnancy (OR 1.81, 95% CI 1.57 to 2.09). The results were broadly similar when women who had an abortion were compared to women who had a delivery (OR 2.39, 95% CI 1.63 to 3.50), women who had no abortion (OR 1.59, 95% CI 1.36 to 1.86) or women who had an unintended pregnancy (OR 1.55, 95% CI 1.31 to 1.84).

All outcomes were more frequent in women who had an abortion: depression (OR 1.37, 95% CI 1.22 to 1.54), anxiety (OR 1.34, 95% CI 1.12 to 1.60), alcohol use (OR 2.10, 95% CI 1.77 to 2.50), marijuana use (OR 3.50, 95% CI 1.65 to 7.44) and suicide and self-harm (OR 2.55, 95% CI 1.31 to 4.96).

Population attributable fractions were: all outcomes 9.9%, depression 8.5%, anxiety 8.1%, alcohol use 10.7%, marijuana use 26.5%, all suicidal behaviours 20.9% and suicide 34.9%.

Authors' conclusions

There was a moderate to highly increased risk of mental health problems after abortion.

CRD commentary

The review question was clearly stated. The search strategy was not reported so it was not clear whether the search was sufficiently comprehensive. The restriction to published studies in English meant that the results may have been affected by publication and language biases. This was a single-author review and there was no mention of duplicate processes for study selection and data extraction to minimise the chances of error and bias in the review process. It appeared that there was no assessment of validity for the individual studies. Few details of the included studies are given so it was not possible to assess the generalisability of the results. Definitions were not stated for the types of comparison group (women who had a delivery, women who had no abortion and women who had an unintended pregnancy). The author acknowledged a lack of uniformity in control variables, demographic characteristics of the participants, length of follow-up time and variation in outcome measurement.

The overall meta-analysis was not appropriate as several measures of effect from individual studies were included but treated as independent observations. This would not have effected the subgroup analyses based on individual outcomes. No measure of statistical heterogeneity was reported but the forest plot suggested that the results were heterogenous. Therefore, it was unclear whether the pooled results were appropriate. In the subgroup analyses based on comparison groups, the same study was sometimes included in more than one subgroup. Without details of the individual studies it was not possible to assess whether or not this was appropriate. For the calculation of population attributable fractions, the author did not state what values were used for the prevalence of abortion.

The results of the review should be treated with caution due to concerns with the review methodology and reporting, and uncertain appropriateness of the statistical analysis.

Implications of the review for practice and research

Practice: The author did not state any implications for practice.

Research: The author stated that a quantitative review of studies of women who obtained abortions without inclusion of a comparison group was needed and that future studies should explore possible process mechanisms linking abortion to substance misuse and suicidal behaviour.

Funding

None stated.

Bibliographic details

Coleman PK. Abortion and mental health: quantitative synthesis and analysis of research published 1995-2009. British Journal of Psychiatry 2011; 199(3): 180-186. [PubMed: 21881096]

Original Paper URL

http://bjp.rcpsych.org/content/199/3/180.abstract

Indexing Status

Subject indexing assigned by NLM

MeSH

Abortion, Induced /adverse effects /psychology /statistics & numerical data; Adolescent; Adult; Bias (Epidemiology); Evidence-Based Practice; Female; Humans; Mental Disorders /epidemiology; Meta-Analysis as Topic; Odds Ratio; Pregnancy; Pregnancy, Unplanned /psychology; Pregnancy, Unwanted /psychology; Risk Factors; Substance-Related Disorders /epidemiology; Suicide /psychology /statistics & numerical data

AccessionNumber

12011007082

Database entry date

10/10/2012

Record Status

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.