Prognosis Scenario

You are a nurse caring for a 28-year old woman who has just had a D & C following a spontaneous miscarriage. She was 10 weeks pregnant and this was her first pregnancy. In a team meeting, one of your colleagues vaguely recalls seeing an article about grief after miscarriage and you decide to track it down to determine whether your patient is at risk of severe or prolonged grief.

You formulate the question, in healthy women who have recently had a miscarriage, what is the usual grieving process and are any factors associated with longer than normal grieving?

Searching terms and evidence source:
Both a MEDLINE search and a search in Best Evidence produced the same citation. Search terms in MEDLINE were “grief” as a subject heading and text word; “abortion” as a subject term or “pregnancy loss” as a text word; and “risk” as an index term or text word. The Best Evidence search used just the terms “grief” and “pregnancy loss”.

Read the article and decide:

Is the evidence from this study valid?
If valid, is this evidence important?
If valid and important, can you apply this evidence in caring for your patient?

Completed Prognosis Worksheet for Evidence-Based Nursing

Citation

Janssen HG, Cuisinier MC, de Graauw KP, Hoogduin KA. A prospective study of risk factors predicting grief intensity following pregnancy loss. Arch Gen Psychiatry. 1997;54:56-61.

Are the results of this prognosis study valid?

Was a defined, representative sample of patients assembled at a common (usually early) point in the course of their disease?: Yes. Data on baseline variables for 227 women who had a miscarriage were collected before the miscarriage. These women had volunteered to be part of a long-term study about how they coped with normal pregnancy, delivery, and complications. Baseline data included information on reproductive history, factors around this pregnancy, sociodemographic data, and personal history.
Was patient follow-up sufficiently long and complete?: Yes. 94% of all women who had a miscarriage after a singleton pregnancy completed all questionnaires (at baseline before the miscarriage, just after the miscarriage, and 6, 12, and 18 months after the miscarriage).
Were objective outcome criteria applied in a “blind” fashion?: No, but outcomes were measured in many ways using several standardised and validated forms.
If subgroups with different prognoses are identified, was there adjustment for important prognostic factors?: Repeated measure analyses factored in differences in baseline characteristics.
Was there validation in an independent group (“test-set”) of patients?: No.

Are the valid results of this prognosis study important?

How likely are the outcomes over time?: Data were not presented this way. However, grief intensity was predicted by length of pregnancy before the loss, time since the loss, preloss neuroticism, preloss psychiatric symptoms, and absence of living children. Childless and older women showed more intense grief. Grief intensity, active grief, difficulty with coping, and despair decreased with time.
How precise are the prognostic estimates?: No confidence intervals provided.

If you want to calculate a Confidence Interval around the measure of Prognosis:

Clinical Measure	Standard Error (SE)	Typical calculation of CI
Proportion (as in the rate of some prognostic event, etc) where: the number of patients = n the proportion of these patients who experience the event = p	$$sqrt{frac{ptimes(1-p)}{n}}\\$$ where p is proportion and n is number of patients	If p = 24/60 = 0.4 (or 40%) and n=60 $$ sqrt{frac{0.4times(1-0.4)}{60}} = 0.063 text{ (or 6.3%)}\\$$ 95% CI is 40% ± 1.96 × 6.3% or 27.6% to 52.4%
n from your evidence: ________ p from your evidence: ________	$$sqrt{frac{ptimes(1-p)}{n}}\\$$ where p is proportion and n is number of patients	Your calculation: SE = ____________ 95% CI: ____________

Clinical Measure

Standard Error (SE)

Typical calculation of CI

Proportion (as in the rate of some prognostic event, etc) where:

the number of patients = n

the proportion of these patients who experience the event = p

$$sqrt{frac{ptimes(1-p)}{n}}\\$$
where p is proportion and n is number of patients

If p = 24/60 = 0.4 (or 40%) and n=60
$$ sqrt{frac{0.4times(1-0.4)}{60}}
= 0.063 text{ (or 6.3%)}\\$$
95% CI is 40% ± 1.96 × 6.3% or 27.6% to 52.4%

n from your evidence: ________

p from your evidence: ________

$$sqrt{frac{ptimes(1-p)}{n}}\\$$
where p is proportion and n is number of patients

Your calculation:
SE = ____________

95% CI: ____________

Can you apply this valid, important evidence about prognosis in caring for your patient?

Were the study patients similar to your own?: Yes, on many factors, especially age. The biggest difference was nationality. There is, however, no reason why women from a developed countries would differ in ways that would make the research irrelevant.
Will this evidence make a clinically important impact on your conclusions about what to offer or tell your patient?: Yes, helpful to know that grief is more intense in women who have no other children, and that grief lessens with time.

Additional Notes

–

Grief after pregnancy loss – predicted by length of pregnancy, neuroticisim, psychiatric symptoms,and absence of other children

Clinical Bottom Line

Patients with pre-pregnancy neuroticism or psychiatric symptoms, and without other children are at increased risk of a more intense grief reaction.

Citation

Janssen HG, Cuisinier MC, de Graauw KP, Hoogduin KA. A prospective study of risk factors predicting grief intensity following pregnancy loss.
Arch Gen Psychiatry. 1997;54:56-61.

Clinical Question

In healthy women who have had a miscarriage, what is the usual grieving process and are any factors associated with longer than normal grieving?

Search Terms

Both MEDLINE and Best Evidence retrieve the same citation. MEDLINE search terms were “grief” as an index term AND (“abortion” as an index term OR “pregnancy loss” as a text phrase) AND “risk” as an index term or text word). The Best Evidence search terms were “grief” and “pregnancy loss”.

The Study

The Study Patients: 227 women (mean age 29 years) who had volunteered for a study on coping with normal pregnancy, delivery, and complications and who reported a miscarriage. 91% of the losses occurred at < 20 weeks of pregnancy. 97% of the women were married or in stable relationships, 32% had no other children, and 41% had a previous pregnancy loss. Follow-up at 18 months after miscarriage was 94%.
Prognostic Factors: Risk factors were assessed using the Dutch version of the Symptom Checklist-90 (psychiatric symptoms) and the Dutch Personality Questionnaire (neuroticism, low self-esteem, social inadequacy, general inadequacy, and aggrievedness). Information was also collected on quality of partner relationships, education, employment, religious background, social support, feelings about the pregnancy, pregnancy and conception variables, family demographics, and physical symptoms.
The Outcome: Grief and its categories (active grief, difficulty coping, and despair) measured by the Perinatal Grief Scale immediately after the miscarriage and at 6, 12, and 18 months.

Study Feature	Yes	No
Well-defined sample at uniform (early) stage of illness?	x
Follow-up long enough?	x
Follow-up complete?	x
Blind and objective outcome criteria?		x
Adjustment for other prognostic factors?	x
Validation in an independent “test-set” of patients?		x

The Evidence

Multivariate analysis showed that grief intensity was higher for women who had been pregnant longer (p< 0.001), had pre-loss neurotic personalities (p< 0.001), had pre-loss psychiatric symptoms (p = 0.02), and did not have other living children (p = 0.01). Grief intensity, active grief, difficulty coping, and despair decreased with time (p< 0.001 for all 4 comparisons).

Comments

Bottom line is that all factors except previous pregnancy loss predicted grief intensity on univariate analysis.

Appraised By

K. Ann McKibbon, MLS.