|Elective Cesarean section to prevent anal incontinence and brachial plexus injuries associated with macrosomia: a decision analysis
|Culligan P J, Myers J A, Goldberg R P, Blackwell L, Gohmann S F, Abell T D
This is a critical abstract of an economic evaluation that meets the criteria for inclusion on NHS EED. Each abstract contains a brief summary of the methods, the results and conclusions followed by a detailed critical assessment on the reliability of the study and the conclusions drawn.
A strategy for the prevention of maternal and infant injuries associated with macrosomia was examined. The strategy involved ultrasound at 39 weeks' gestation, followed by elective Caesarean section (C-section) for those women with estimated foetal weights of greater than or equal to 4,500 g.
The study population comprised a hypothetical cohort of primigravidae.
The setting was a hospital. The economic study was carried out in the USA.
Dates to which data relate
The effectiveness data were derived from studies published between 1973 and 2002. The costs and resource use data appear to have been estimated in 2001, which could have been the price year.
Source of effectiveness data
The effectiveness evidence was derived from a synthesis of published studies and experts' opinions.
A decision tree was constructed to evaluate the cost-effectiveness of the two policies under examination in a hypothetical cohort of 100,000 deliveries. A schematic representation of the model was provided. Diabetics and non-diabetics were considered independently. The major risks and/or complications considered crucial to the analysis were brachial plexus injury to the newborn, maternal anal incontinence, maternal urinary incontinence, emergency hysterectomy, blood transfusion due to postpartum haemorrhage, and maternal mortality. The benefits were estimated for two individuals simultaneously (the mother and the baby).
Outcomes assessed in the review
The outcomes estimated from the literature were the probability estimates associated with specific clinical conditions considered in the model, such as:
brachial plexus injury,
haemorrhage requiring blood transfusion after vaginal delivery,
the rate of peripartum hysterectomy,
anal sphincter disruption,
the prevalence of gestational diabetes,
the prevalence of infants of different weights,
the sensitivity and specificity of ultrasound to detect macrosomia, and
The average age of first-time mothers in the USA, the life expectancy for those women, and the life expectancy of a newborn were also estimated.
Study designs and other criteria for inclusion in the review
The primary studies were identified from a review of the literature, although it was unclear whether it was a systematic review. No information on the design of the primary studies was provided. Data on life expectancy and the average age of first-time mothers in the USA were derived from national statistics.
Sources searched to identify primary studies
MEDLINE was searched from 1980 to identify relevant studies. The bibliographic references of retrieved articles were also checked.
Criteria used to ensure the validity of primary studies
The authors stated that, whenever possible, data from randomised clinical trials were used. Otherwise, less robust studies were used to provide data.
Methods used to judge relevance and validity, and for extracting data
Number of primary studies included
Twenty-four studies provided the evidence used in the decision model.
Methods of combining primary studies
It appears that a narrative method has been used to combine the primary estimates.
Investigation of differences between primary studies
Results of the review
Some of the estimates used in the model were as follows:
anal incontinence following anal sphincter disruption was 0.23 (range: 0.03 - 0.38);
maternal mortality following vaginal delivery was 0.00004 (range: 0.00002 - 0.00005);
maternal mortality following elective C-section was 0.000028 (range: 0.00002 - 0.00005);
maternal mortality following non-elective C-section was 0.0003 (range: 0.0001 - 0.001);
the rate of episiotomy among primigravidae was 0.41 (range: 0.25 - 0.50);
peripartum hysterectomy following vaginal delivery was 0.0002 (range: 0.0001 - 0.01);
peripartum hysterectomy following elective C-section was 0.007 (range: 0.001 - 0.01);
peripartum hysterectomy following non-elective C-section was 0.007 (range: 0.001 - 0.01);
the test for gestational diabetes had a sensitivity of 0.99 (range: 0.90 - 0.9950) and a specificity of 0.66 (range: 0.50 - 0.80);
ultrasound had a sensitivity of 0.57 (range: 0.50 - 0.95) to detect macrosomia among gestational diabetics and a specificity of 0.94 (range: 0.75 - 0.95);
ultrasound had a sensitivity of 0.59 (range: 0.50 - 0.95) to detect macrosomia among non-diabetics and a specificity of 0.91 (range: 0.75 - 0.95);
the overall prevalence of gestational diabetes was 0.03 (range: 0.01 - 0.09);
the prevalence of infants weighing at least 4,500 g among gestational diabetics was 0.06 (range: 0.02 - 0.10);
the prevalence of infants weighing 4,000 - 4,499 g among gestational diabetics was 0.17 (range: 0.05 - 0.25);
the prevalence of infants weighing less than 4,000 g among gestational diabetics was 0.77 (range: 0.50 - 0.90);
the prevalence of infants weighing at least 4,500 g among non-diabetics was 0.015 (range: 0.005 - 0.03);
the prevalence of infants weighing 4,000 - 4,499 g among non-diabetics was 0.082 (range: 0.03 - 0.10); and
the prevalence of infants weighing less than 4,000 g among non-diabetics was 0.9 (range: 0.75 - 0.99).
The average age of first-time mothers in the USA was 24.3 years. The life expectancy for those women was an additional 55.4 years. The life expectancy of a newborn (weighted average of males and females) was 76.85 years.
Methods used to derive estimates of effectiveness
A panel of seven experts was contacted to estimate the utility values used to calculate quality-adjusted life-years (QALYs).
Estimates of effectiveness and key assumptions
The estimated utility values were:
uncomplicated vaginal delivery and healthy child, 1.00 (range: 0.90 - 1.00);
vaginal delivery including 1st or 2nd degree episiotomy that heals normally and healthy child, 0.995 (range: 0.90 - 1.00);
brachial plexus injury that resolves within 2 months, 0.99 (range: 0.90 - 1.00);
haemorrhage requiring blood transfusion, 0.96 (range: 0.90 - 1.00);
third-fourth anal sphincter disruption that heals well (asymptomatic), 0.85 (range: 0.75 - 0.95);
peripartum hysterectomy and healthy child, 0.71 (range: 0.61 - 0.81);
urinary incontinence, 0.70 (range: 0.60 - 0.80);
permanent brachial plexus injury (mild to moderate), 0.60 (range: 0.50 - 0.70);
anal incontinence and healthy child, 0.50 (range: 0.40 - 0.65);
anal incontinence, peripartum hysterectomy and healthy child, 0.49 (range: 0.35 - 0.65);
anal incontinence, urinary incontinence and healthy child, 0.48 (range: 0.35 - 0.65);
anal incontinence and permanent brachial plexus injury (mild to moderate), 0.46 (range: 0.30 - 0.60);
permanent brachial plexus injury (severe) and uncomplicated delivery, 0.45 (range: 0.30 - 0.60);
permanent brachial plexus injury (severe) and anal incontinence, 0.35 (range: 0.20 - 0.60);
permanent brachial plexus injury (severe), anal incontinence and urinary incontinence, 0.34 (range: 0.20 - 0.60);
permanent brachial plexus injury (severe), anal incontinence, urinary incontinence, peripartum hysterectomy and blood transfusion, 0.30 (range: 0.20 - 0.60);
maternal death, 0.03 (range: 0.00 - 0.10);
maternal death and brachial plexus injury that resolves within 2 months, 0.0297 (range: 0.00 - 0.10);
maternal death and permanent brachial plexus injury (mild to moderate), 0.018 (range: 0.00 - 0.10);
maternal death and permanent brachial plexus injury (severe), 0.0 (range: 0.0 - 0.10).
Measure of benefits used in the economic analysis
The summary benefit measure used was the QALYs. An annual discount rate of 3% was applied to expected survival. Other model outputs were the cases of maternal anal incontinence and the cases of permanent neonatal brachial plexus injury avoided in comparison with standard care.
An annual discount rate of 3% was applied since long-term costs were estimated. The unit costs were not presented separately from the quantities of resources used. The health services included in the economic evaluation were repair of any episiotomy or laceration, 39-week ultrasound, brachial plexus injury (spontaneous resolution), anal incontinence, uncomplicated vaginal delivery, complicated vaginal delivery, uncomplicated elective C-section, complicated elective C-section, uncomplicated non-elective C-section, complicated non-elective C-Section, permanent brachial plexus injury (mild, moderate, or severe), and urinary incontinence. The components of these cost categories of costs, which also included multiple aspects of care (i.e. complicated delivery), were listed. The authors stated that a societal perspective was adopted, but it appears that only those costs relevant to the third-party payer have been included in the economic evaluation. Resource use and costs were estimated from Medicare reimbursement rates and other data derived from the geographical area of Louisville (KY). The price year could have been 2001.
Statistical analysis of costs
The costs were treated deterministically.
The indirect costs were not considered in the economic evaluation.
A one-way sensitivity analysis was carried out to examine the robustness of the model benefits and costs. Al model inputs were varied in the analysis. The ranges of values were based on data derived from the literature or authors or experts' opinions. A sub-group analysis which separated diabetic and non-diabetic women was also performed.
Estimated benefits used in the economic analysis
For every 100,000 deliveries, the policy of near-term ultrasound followed by elective C-sections for foetuses believed to weigh at least 4,500 g resulted in 185.7 fewer cases of maternal anal incontinence and 16.6 fewer cases of permanent neonatal brachial plexus injury than the current standard care. Thus, one case of anal incontinence would be prevented for every 539 elective C-sections performed, and one permanent brachial plexus injury would be prevented for every 6,024 elective C-sections performed.
The expected quality of life (for the mother/newborn dyad) of the elective C-section policy was higher than that for the current standard of care (0.923 versus 0.917). The standard care approach resulted in 53.2 QALYs and the 4,500 g screening policy resulted in 53.6 QALYs.
The new policy provided 0.33 additional QALYs among diabetic women and 0.37 QALYs among non-diabetic women.
For every 100,000 deliveries, standard care cost $850,581,000 and the elective C-section cost $847,370,000. Thus, the new policy resulted in cost-savings of $3,211,000 for every 100,000 deliveries.
The new policy would save $6,375,000 per 100,000 deliveries among diabetic women and $3,113,000 per 100,000 deliveries among non-diabetic women.
Synthesis of costs and benefits
The costs and benefits were not synthesised because the selective use of elective C-section resulted in a higher quality of life per delivery at lower monetary costs (dominant). Similar results were obtained in the sub-groups of diabetic and non-diabetic women.
The sensitivity analysis showed that, in general, the model results (both costs and benefits) were robust to variations in the model inputs. Some changes in the estimated costs were obtained if the probability of urinary incontinence resulting from a vaginal delivery decreased to 0.217 (0.245 in the base-case), or if the probability of urinary incontinence resulting from an elective C-section was 0.072 (0.05 in the base-case); the monetary costs would be equal for the two strategies.
The cost estimates were also sensitive to variations in the costs associated with urinary incontinence, uncomplicated vaginal delivery, uncomplicated elective C-section and uncomplicated non-elective C-section.
With urinary incontinence removed from the model entirely, the elective C-section policy resulted in an incremental gain of 0.112 QALYs per delivery and incremental costs of $21,432,000 more per 100,000 deliveries than the standard care policy. Thus, the incremental cost per QALY would have been $1,921.
A policy whereby all primigravidae patients would be offered an ultrasound at 39 weeks' gestation, followed by elective C-section for any estimated foetal weights of at least 4,500 g, was cost-effective in comparison with actual standard care in the USA.
CRD COMMENTARY - Selection of comparators
The choice of the comparator, spontaneous labour followed by either vaginal delivery or C-section as indicated, was appropriate as it reflected standard care for primigravidae. You should decide whether this is a valid comparator in your own setting.
Validity of estimate of measure of effectiveness
The effectiveness evidence came from experts' opinions as well as from a synthesis of published studies. It was unclear whether a systematic review of the literature was undertaken to identify primary studies, as a single source was searched. Further, the design and characteristics of the studies providing the data were not described. Information on the health care providers involved in the assessment of utility weights was provided. However, a description of the approach used to reach consensus was not given. Owing to uncertainty in the model inputs, plausible ranges of values were defined and all inputs were varied in the sensitivity analysis. The authors stated that the assumptions made in the model were biased against the proposed policy. If more realistic assumptions had been made, the model would have further favoured the proposed policy.
Validity of estimate of measure of benefit
The summary benefit measure was appropriate because it measures the impact of the intervention on two dimensions of care, survival and quality of life. QALYs are comparable with the benefits of other health care interventions. Discounting was performed, as US guidelines recommend. Other model outputs, which could be more interesting for service providers, were also reported.
Validity of estimate of costs
Although the authors had stated that a societal perspective was adopted, only direct medical costs were included in the analysis. The inclusion of indirect (patient) costs would have been interesting. A detailed breakdown of the items was provided, but the unit costs and the quantities of resources used were not given. This limits the possibility of replicating the analysis in comparable contexts. The source of the data was reported. Reimbursement rates were used to estimate the costs. Some information on the price year was given, which aids reflation exercises in other settings. No statistical analyses of the costs were performed. However, all economic estimates were varied in the sensitivity analysis. The authors stated that the model tended to underestimate the cost-savings associated with the proposed policy.
The authors compared their findings with those from a published study and provided a possible justification for the differences observed. The issue of the generalisability of the study results to other settings was implicitly addressed in the sensitivity analysis, which enhances the external validity of the analysis. The authors stated that because of the complexity and size of their decision tree, other important but primarily transient conditions (e.g. endometritis or transient tachypnoea of the newborn) were not considered. It was also noted that the model focused on primigravidae and the proposed policy might not be appropriate for women planning large families.
Implications of the study
The study results supported the proposed policy whereby all primigravidae patients would be offered an ultrasound at 39 weeks' gestation, followed by elective C-section for any estimated foetal weights greater than or equal to 4,500 g. After considering the implications of adopting the new policy (including the extra time and effort that will be required of obstetric practitioners during the informed consent process), the authors advocated the addition of "delivery mode counselling" codes to the Medicare physician fee schedule, with reimbursement rates that accurately reflect the resultant efforts of obstetric practitioners.
Culligan P J, Myers J A, Goldberg R P, Blackwell L, Gohmann S F, Abell T D. Elective Cesarean section to prevent anal incontinence and brachial plexus injuries associated with macrosomia: a decision analysis. International Urogynecology Journal 2005; 16(1): 19-28
Other publications of related interest
Rouse DJ, Owen J, Goldenberg RL, et al. The effectiveness and costs of elective Cesarean delivery for fetal macrosomia diagnosed by ultrasound. JAMA 1996;276:1480-6.
Gilliam M, Rosenberg D, Davis F, et al. The likelihood of placenta previa with greater number of Cesarean deliveries and higher parity. Obstetrics and Gynecology 2002;99:976-80.
Bewley S, Cockburn J. The unethics of 'request' Caesarean section. British Journal of Obstetrics and Gynaecology 2002;109:593-6.
Sand PK, Grobman W. Scholarly debate: prophylactic benefits of elective Cesarean delivery/broad-based conversion to elective Cesarean delivery is not justified. Female Patient 2002;27:18-28.
Subject indexing assigned by NLM
Adult; Birth Weight; Brachial Plexus Neuropathies /etiology /prevention & Cesarean Section /economics /utilization; Cost Savings; Cost-Benefit Analysis; Decision Trees; Elective Surgical Procedures; Fecal Incontinence /economics /etiology /prevention & Female; Fetal Macrosomia; Health Policy; Humans; Pregnancy; Quality of Life; Treatment Outcome; control; control
Date bibliographic record published
Date abstract record published