Gilchrist - Revision history

Barb: /* Low grade under the JBI Critical Appraisal Checklist for Case Series */

2025-09-19T23:48:55Z

Low grade under the JBI Critical Appraisal Checklist for Case Series

← Older revision		Revision as of 18:48, 19 September 2025
Line 220:		Line 220:

	=== Low grade under the [https://jbi.global/sites/default/files/2021-10/Checklist_for_Case_Series.docx JBI Critical Appraisal Checklist for Case Series] ===		=== Low grade under the [https://jbi.global/sites/default/files/2021-10/Checklist_for_Case_Series.docx JBI Critical Appraisal Checklist for Case Series] ===
	When ~~an appropriate~~ checklist ~~is used to evaluate Gilchrist as a case series study~~, ~~it becomes increasingly clear that it cannot be rated as a high quality study~~.		When assessed under the '''Joanna Briggs Institute (JBI) Critical Appraisal Checklist for Case Series,''' Gilchrist et al. (1995) performs poorly across most domains. The checklist highlights the study’s lack of representativeness, incomplete reporting, and methodological weaknesses.
	{\| class="wikitable"		{\| class="wikitable"
	\|'''JBI Critical Appraisal Item'''		\|'''JBI Critical Appraisal Item'''
Line 236:		Line 236:
	\|3. Were valid methods used for identification of the condition for all participants?		\|3. Were valid methods used for identification of the condition for all participants?
	\|Unclear		\|Unclear
	\|~~Diagnosis/measurement tools for psychiatric morbidity were valid and appropriate (~~ICD-8 ~~diagnoses~~ were ~~reported~~ by ~~GP's~~ not psychiatrists)		\|ICD-8 is a valid classification system, but assessments were made by GPs, not psychiatrists; risk of diagnostic misclassification acknowledged by authors.
	\|-		\|-
	\|4. Did the case series have consecutive inclusion of participants?		\|4. Did the case series have consecutive inclusion of participants?
	\|No		\|No
	\|~~Convenience~~ GP ~~selection precludes~~ consecutive inclusion.		\|Recruitment depended on voluntary GP reporting; consecutive inclusion cannot be established
	\|-		\|-
	\|5. Did the case series have complete inclusion of participants?		\|5. Did the case series have complete inclusion of participants?
	\|No		\|No
	\|~~Sample limited~~ to ~~selected~~ GPs; ~~incomplete coverage~~ of ~~all eligible cases~~.		\|Limited to patients of participating GPs; no attempt to capture a random sample representative of national population.
	\|-		\|-
	\|6. Was there clear reporting of the demographics of the participants in the study?		\|6. Was there clear reporting of the demographics of the participants in the study?
	\|Unclear		\|Unclear
	\|~~Age~~, parity, ~~and other sociodemographic characteristics~~ were not reported ~~but were used~~ in ~~calculating adjusted RR~~.		\|Basic demographics (age, parity, smoking) were used for adjustment but not fully reported in detail.
	\|-		\|-
	\|7. Was there clear reporting of clinical information of the participants?		\|7. Was there clear reporting of clinical information of the participants?
	\|No		\|No
	\|Prior and subsequent abortion history not reported~~, which may have impacted mental health and adulterated control~~ groups		\|Prior and subsequent abortion history not reported resulting in contamination of the reference groups.
	\|-		\|-
	\|8. Were the outcomes or follow-up results of cases clearly reported?		\|8. Were the outcomes or follow-up results of cases clearly reported?
	\|Unclear		\|Unclear
	\|~~Reported follow-up~~ outcomes ~~for~~ psychiatric morbidity ~~over the study period.~~		\|Some outcomes (new psychiatric morbidity) reported, but critical endpoints (timing of events, mortality) and any evaluation of worsening of prior psychiatric morbidity omitted
	\|-		\|-
	\|9. Was there clear reporting of the presenting site(s) / clinic(s) demographic information?		\|9. Was there clear reporting of the presenting site(s) / clinic(s) demographic information?
Line 264:		Line 264:
	\|10. Was statistical analysis appropriate?		\|10. Was statistical analysis appropriate?
	\|Unclear		\|Unclear
	\|Statistical ~~analysis suitable~~ for ~~descriptive~~ comparisons,. ~~But sample size too small for selfharm analysis~~		\|Statistical approach adequate for crude comparisons, but variance estimation, handling of clustering, and small subgroup analyses were methodologically weak.
	\|}		\|}

Barb: /* Low grade under the JBI Critical Appraisal Checklist for Case Series */

2025-09-19T20:20:37Z

Low grade under the JBI Critical Appraisal Checklist for Case Series

← Older revision		Revision as of 15:20, 19 September 2025
Line 220:		Line 220:

	=== Low grade under the [https://jbi.global/sites/default/files/2021-10/Checklist_for_Case_Series.docx JBI Critical Appraisal Checklist for Case Series] ===		=== Low grade under the [https://jbi.global/sites/default/files/2021-10/Checklist_for_Case_Series.docx JBI Critical Appraisal Checklist for Case Series] ===
			When an appropriate checklist is used to evaluate Gilchrist as a case series study, it becomes increasingly clear that it cannot be rated as a high quality study.
	{\| class="wikitable"		{\| class="wikitable"
	\|Item		\|'''JBI Critical Appraisal Item'''
	\|Response (Yes/No/Unclear)		\|'''Response (Yes/No/Unclear)'''
	\|Notes		\|'''Notes informing the Response'''
	\|-		\|-
	\|1. Were there clear criteria for inclusion in the case series?		\|1. Were there clear criteria for inclusion in the case series?

Barb: /* A Middling Newcastle-Ottawa Scale for Cohort Studies */

2025-09-19T20:17:25Z

A Middling Newcastle-Ottawa Scale for Cohort Studies

← Older revision		Revision as of 15:17, 19 September 2025
Line 163:		Line 163:
	In the Case-Control version of the NOS, it is clear that Gilchrist's sample of women, chosen by a group of volunteer general practitioners, is not random nor does it include all eligible cases of women. So it is no representative of all cases. In addition, while women who decided against abortion or were refused abortions, were treated as control groups, NOS requires that "If cases are first occurrence of outcome, then it must explicitly state that controls have no history of this outcome. If cases have new (not necessarily first) occurrence of outcome, then controls with previous occurrences of outcome of interest should not be excluded." But Gilchrist does not control for abortions that may have occurred before or after the index pregnancy event upon which the 1509 volunteer GP's selected and place women into one of the three groups. This means there were at least some women in the two control groups who had prior and/or subsequent abortions.		In the Case-Control version of the NOS, it is clear that Gilchrist's sample of women, chosen by a group of volunteer general practitioners, is not random nor does it include all eligible cases of women. So it is no representative of all cases. In addition, while women who decided against abortion or were refused abortions, were treated as control groups, NOS requires that "If cases are first occurrence of outcome, then it must explicitly state that controls have no history of this outcome. If cases have new (not necessarily first) occurrence of outcome, then controls with previous occurrences of outcome of interest should not be excluded." But Gilchrist does not control for abortions that may have occurred before or after the index pregnancy event upon which the 1509 volunteer GP's selected and place women into one of the three groups. This means there were at least some women in the two control groups who had prior and/or subsequent abortions.

	In the Cohort version NOS, the selection criteria is poor. It is not a representative sample since it relied upon both on volunteer group of GP's and only those women who agreed to have their information shared.		In the Cohort version NOS, the selection criteria is poor. It is not a representative sample since it relied upon both on volunteer group of GP's and only those women who agreed to have their information shared. As shown in the table below, as judged by this quality index, Gilchrist has a score of 5 out of a possible 9 points.
	{\| class="wikitable"		{\| class="wikitable"
	\|Author		\|Author

Barb: /* Grading of Gilchrist Study */

2025-09-19T20:15:33Z

Grading of Gilchrist Study

← Older revision		Revision as of 15:15, 19 September 2025
Line 156:		Line 156:
	== Grading of Gilchrist Study ==		== Grading of Gilchrist Study ==

	=== A ~~Weal~~ Newcastle-Ottawa Scale for Cohort Studies ===		=== A Middling Newcastle-Ottawa Scale for Cohort Studies ===
	[https://www.ohri.ca/programs/clinical_epidemiology/oxford.asp The Newcastle-Ottawa Scale (NOS)] is commonly used to grade the quality of studies. It is useful for identifying where Gilchrist's study falls short.		[https://www.ohri.ca/programs/clinical_epidemiology/oxford.asp The Newcastle-Ottawa Scale (NOS)] is commonly used to grade the quality of studies. It is useful for identifying where Gilchrist's study falls short.

Barb: /* Grading of Gilchrist Study */

2025-09-19T20:15:03Z

Grading of Gilchrist Study

@@ Line 155: / Line 155: @@
 == Grading of Gilchrist Study ==
 [https://www.ohri.ca/programs/clinical_epidemiology/oxford.asp The Newcastle-Ottawa Scale (NOS)] is commonly used to grade the quality of studies. It is useful for identifying where Gilchrist's study falls short.
 In the Case-Control version of the NOS, it is clear that Gilchrist's sample of women, chosen by a group of volunteer general practitioners, is not random nor does it include all eligible cases of women.  So it is no representative of all cases.  In addition, while women who decided against abortion or were refused abortions, were treated as control groups, NOS requires that "If cases are first occurrence of outcome, then it must explicitly state that controls have no history of this outcome. If cases have new (not necessarily first) occurrence of outcome, then controls with previous occurrences of outcome of interest should not be excluded."  But Gilchrist does not control for abortions that may have occurred before or after the index pregnancy event upon which the 1509 volunteer GP's selected and place women into one of the three groups.  This means there were at least some women in the two control groups who had prior and/or subsequent abortions.
 In the Cohort version NOS, the selection criteria is poor.  It is not a representative sample since it relied upon both on volunteer group of GP's and only those women who agreed to have their information shared.
-==== Gilchrist is actually a case series, not a cohort study ====
+=== Gilchrist is actually a case series, not a cohort study ===
 Although Gilchrist et al. enrolled a non-random, convenience sample of women chosen by a volunteer group GPs who asked a convenience sample of women if they would "agree" to allow their family doctor to provide data to the research team.  The GP's ''after'' the women had already sought an abortion and/or from a sample of women they deemed to have not planned their pregnancies at least three months before conceiving.

Barb: /* Gilchrist is actually a case series, not a cohort study */

2025-09-19T20:03:10Z

Gilchrist is actually a case series, not a cohort study

@@ Line 185: / Line 185: @@
 * ''Presence of comparison groups is misleading:'' Even if the paper compares women who terminated vs continued, that alone doesn’t make it a cohort — the sampling frame and denominator definition do. Dekkers explicitly notes that a comparison group *doesn’t* define a cohort; sampling method does.
 * ''Terminology slippage in clinical journals.'' Words like “prospective consecutive case series” or “cohort” are used loosely.
 == Notes that may require further investigation ==

Barb: /* Table of Claims versus Problems Issues */

2025-09-18T19:19:13Z

Table of Claims versus Problems Issues

@@ Line 68: / Line 68: @@
 #Women who have miscarriages are known to have higher rates of subsequent psychological distress compared to women who deliver health children. By including women who miscarry with women who carried to term, the study fails to provide a comparison between rates of psychological illness for women who carry to term--which is of course their intent. While miscarriage is an unavoidable risk, the choice women face is between trying to carry to term and having an induced abortion. Therefore, it seems that the comparison between psychological risks of abortion and carrying to term would be relevant to both women and physicians--excluding the risks of psychiatric distress that may follow a miscarriage. While all measures are relevant, the failure to distinguish between successful delivery and miscarriages in this study may have obscured a relative risk of abortion compared to delivery.
 #Gilchrist et al. (1995) used outcome-based, convenience sampling (women identified after making a pregnancy decision via selected general practitioners), which prevents estimation of absolute risk in an exposed population; under the criteria of [https://citeseerx.ist.psu.edu/document?repid=rep1&type=pdf&doi=9805a181224c88d888803d906e6c967e2e107829 Dekkers et al. (2012)] this design is more appropriately classified as a case series rather than a cohort study.
 == Table of Claims versus Problems Issues ==
@@ Line 105: / Line 144: @@
 |-
 | '''Applicable to general abortion–mental health debates'''
-| UK’s dual‑physician screening and risk‑benefit analysis likely filter out highest‑risk women. Findings are not generalizable to contexts (e.g., US) without such safeguards.
+| UK’s dual‑physician screening and risk‑benefit analysis likely filter out highest‑risk women. Findings are not generalizable to contexts (e.g., US) where doctors' risks assessments are mandatory.
 |-
 | '''Large sample size is a strength'''

Barb: /* Grading of Gilchrist Study */

2025-09-18T19:12:46Z

Grading of Gilchrist Study

@@ Line 69: / Line 69: @@
 #Gilchrist et al. (1995) used outcome-based, convenience sampling (women identified after making a pregnancy decision via selected general practitioners), which prevents estimation of absolute risk in an exposed population; under the criteria of [https://citeseerx.ist.psu.edu/document?repid=rep1&type=pdf&doi=9805a181224c88d888803d906e6c967e2e107829 Dekkers et al. (2012)] this design is more appropriately classified as a case series rather than a cohort study.
-== Grading of Gilchrist Study ==
+== Table of Claims versus Problems Issues ==
 [https://www.ohri.ca/programs/clinical_epidemiology/oxford.asp The Newcastle-Ottawa Scale (NOS)] is commonly used to grade the quality of studies. It is useful for identifying where Gilchrist's study falls short.

Barb: /* Gilchrist is actually a case series, not a cohort study */

2025-09-10T21:05:32Z

Gilchrist is actually a case series, not a cohort study

← Older revision		Revision as of 16:05, 10 September 2025
Line 80:		Line 80:

	==== Gilchrist is actually a case series, not a cohort study ====		==== Gilchrist is actually a case series, not a cohort study ====
	Although Gilchrist et al. enrolled a non-random ~~set~~ of GPs who ~~followed~~ a convenience sample of women ~~who~~ "~~agreed~~" to allow their family doctor to provide data to the research team. The GP's ''after'' the women had already sought an abortion and/or from a sample of women they deemed to have not planned their pregnancies at least three months before conceiving.		Although Gilchrist et al. enrolled a non-random, convenience sample of women chosen by a volunteer group GPs who asked a convenience sample of women if they would "agree" to allow their family doctor to provide data to the research team. The GP's ''after'' the women had already sought an abortion and/or from a sample of women they deemed to have not planned their pregnancies at least three months before conceiving.

	Because the sampling was from GP's who referred for or provided abortions who non-randomly chose who to invite...and only women who agreed to participate were reported upon (with no data on what percentage of women refused to be reported upon) the study sample is clearly not representative of all women at risk of unplanned pregnancies. Because it does not include sampling at clear inception point (prior to pregnancy, or immediately upon learning one was pregnant) the design is best described as a case series rather than a cohort study.		Because the sampling was from GP's who referred for or provided abortions who non-randomly chose who to invite...and only women who agreed to participate were reported upon (with no data on what percentage of women refused to be reported upon) the study sample is clearly not representative of all women at risk of unplanned pregnancies. Because it does not include sampling at clear inception point (prior to pregnancy, or immediately upon learning one was pregnant) the design is best described as a case series rather than a cohort study.

Barb: /* Gilchrist is actually a case series, not a cohort study */

2025-08-11T15:57:51Z

Gilchrist is actually a case series, not a cohort study

← Older revision		Revision as of 10:57, 11 August 2025
Line 80:		Line 80:

	==== Gilchrist is actually a case series, not a cohort study ====		==== Gilchrist is actually a case series, not a cohort study ====
	Although Gilchrist et al. enrolled ~~and~~ followed women~~, they recruited participants~~ after the women had already ~~made their pregnancy decision~~ and from a ~~non-random set~~ of ~~GPs~~. Because the sampling was non-~~random~~ and ~~decision-based~~ the study ~~fails to investigate a well-defined and~~ representative ~~population~~ of women at risk of unplanned pregnancies. Because it does not include sampling at clear inception point (prior to pregnancy, or immediately upon learning one was pregnant) the design is best described as a case series rather than a cohort study.		Although Gilchrist et al. enrolled a non-random set of GPs who followed a convenience sample of women who "agreed" to allow their family doctor to provide data to the research team. The GP's ''after'' the women had already sought an abortion and/or from a sample of women they deemed to have not planned their pregnancies at least three months before conceiving.

			Because the sampling was from GP's who referred for or provided abortions who non-randomly chose who to invite...and only women who agreed to participate were reported upon (with no data on what percentage of women refused to be reported upon) the study sample is clearly not representative of all women at risk of unplanned pregnancies. Because it does not include sampling at clear inception point (prior to pregnancy, or immediately upon learning one was pregnant) the design is best described as a case series rather than a cohort study.

	1) The rule of thumb distinguishing case series from cohort studies: ([https://citeseerx.ist.psu.edu/document?repid=rep1&type=pdf&doi=9805a181224c88d888803d906e6c967e2e107829 Dekkers et a]l.)		1) The rule of thumb distinguishing case series from cohort studies: ([https://citeseerx.ist.psu.edu/document?repid=rep1&type=pdf&doi=9805a181224c88d888803d906e6c967e2e107829 Dekkers et a]l.)

	* Cohort = sampling based on exposure (or a clearly defined inception cohort) wherein participants free of the outcome at baseline, followed over time, so you '''can''' calculate absolute risks or rates.		* Cohort = sampling based on exposure (or a clearly defined inception cohort) wherein participants free of the outcome at baseline, followed over time, so you '''can''' calculate absolute risks or rates. In this case, participants should be identified and followed prior to their becoming pregnant, such as in the example of [https://www.cambridge.org/core/journals/the-british-journal-of-psychiatry/article/abortion-and-mental-health-disorders-evidence-from-a-30year-longitudinal-study/59A90CBF3A58C58B342CBCFFBBFEBD2E Fergusson 2008], a true cohort study which examined mental health effects associated with pregnancy outcomes.
	* Case series = sampling based on the outcome (or outcome+exposure), so you '''cannot''' calculate an absolute risk for the outcome in an exposed population		* Case series = sampling based on the outcome (or outcome+exposure), so you '''cannot''' calculate an absolute risk for the outcome in an exposed population