Article Text
PDF
Abstract
Objective We compared processes of antepartum, intrapartum and postpartum care and obstetrical outcomes between physicians and non-physicians.
Design This is a population-based retrospective matched cohort study.
Setting The study was conducted in Ontario, Canada.
Participants Physicians and non-physicians residing in high-income urban areas from 1 April 2009 to 26 November 2018 were included. Physicians were matched to non-physicians on maternal age, calendar year, parity, conception by assisted reproductive technology and singleton versus multifetal gestation. We compared processes of antepartum, intrapartum and postpartum care between physicians and non-physicians.
Outcome measures The primary outcome was mode of delivery (caesarean section, C-section vs vaginal delivery). Secondary outcomes included obstetrical anal sphincter injury among those experiencing vaginal birth and differences in urgent healthcare contacts (maternal and neonatal) during the postpartum period.
Results 7327 physicians were matched 1:5 to 36 185 non-physicians and were well balanced except for comorbidities (physicians had fewer comorbidities). Physicians had more antenatal ultrasounds and invasive prenatal testing, received labour anaesthesia more often and were more often delivered by their own care provider. In adjusted analyses, physicians and non-physicians had a similar risk of C-section (aRR 0.97, 95% CI 0.93 to 1.00, p=0.07). There was no difference in neonatal urgent care contacts; non-physicians had a higher risk of maternal urgent postpartum care (adjusted relative risk [aRR] 1.22, 95% CI 1.08 to 1.37, p<0.0001).
Conclusions Physicians and non-physicians of similar age and with similar pregnancy characteristics had a comparable rate of C-section, which may be related to a lack of cost drivers for C-section in Ontario.
- Physicians
- Observational Study
- Pregnancy
- Cesarean Section
Data availability statement
All data relevant to the study are included in the article or uploaded as online supplemental information. The dataset from this study is held securely in coded form at ICES. While legal data sharing agreements between ICES and data providers (eg, healthcare organisations and government) prohibit ICES from making the dataset publicly available, access may be granted to those who meet prespecified criteria for confidential access, available at www.ices.on.ca/DAS. The full dataset creation plan and underlying analytic code are available from the authors upon request, understanding that the computer programs may rely on coding templates or macros that are unique to ICES and are therefore either inaccessible or may require modification.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
STRENGTHS AND LIMITATIONS OF THIS STUDY
Population-based study of health outcomes for all physicians in Ontario experiencing an in-hospital obstetrical delivery.
Ontario has a universal healthcare system, enabling comprehensive evaluation of all antenatal visits, obstetrical interventions and outcomes.
We did not have occupation information for non-physicians.
We were unable to explore differences in the use of elective versus urgent C-section among physicians versus non-physicians.
Introduction
Physicians may experience different processes of obstetrical care and outcomes compared with non-physicians by virtue of their knowledge of and access to the healthcare system.1 2 Physicians are more knowledgeable with respect to obstetrical processes of care, and they may request specific interventions that may influence their outcomes. Superior access to the healthcare system might also lead to more unnecessary intervention; for example, additional ultrasound surveillance may result in findings that lack clinical significance but lead to further intervention.3 There is scant literature examining whether physicians choose the same obstetrical treatment for their patients that they would choose for themselves; one US study found significantly lower rates of caesarean delivery among physicians compared with non-physicians, although this may be influenced by financial incentives within the US private system.1
Obstetrical processes of care and outcomes among physicians versus non-physicians within a publicly funded healthcare system have not yet been examined. Residents of Ontario, Canada, access healthcare through a publicly funded system, under which they are covered for all essential care, including obstetrical care. Physician payments for obstetrical deliveries are established in the Ontario Health Insurance Plan (OHIP) Schedule of Benefits, and payments for vaginal deliveries and caesarean sections are comparable, mitigating the effect of financial incentives on obstetrical management decisions.
The objective of this study was to determine whether physicians in Ontario’s publicly funded healthcare system experience different processes of obstetrical care and obstetrical outcomes compared with high-income non-physicians. The primary outcome of interest was mode of delivery (caesarean section vs vaginal delivery) between physicians and non-physicians. Secondary outcomes of interest included differences in other antepartum, intrapartum and postpartum processes of care and obstetrical anal sphincter injury (OASIS) as well as maternal and neonatal postpartum emergency room visits and hospitalisations.
Methods
Study Design & Data Sources
We performed a population-based retrospective matched cohort study of reproductive-aged women 20–50 years in the province of Ontario, Canada. The study protocol was published4 at St. Michael’s Hospital. The paper has been written in accordance with the Reporting of studies Conducted using Observational Routinely collected Data guidelines5 (online supplemental table 1). Physicians were identified using data from the College of Physicians and Surgeons of Ontario (CPSO), the sole licencing body for all practising Ontario physicians (including postgraduate trainees). To study important covariates and health outcomes among physicians, we linked CPSO data to the population-based databases housed at ICES (formerly the Institute for Clinical Evaluative Sciences). ICES is an independent, non-profit research institute whose legal status under Ontario’s health information privacy law allows it to collect and analyse healthcare and demographic data, without consent, for health system evaluation and improvement. Deliveries were identified in the ICES MOMBABY dataset, derived from the Canadian Institutes for Health Information Discharge Abstract Database (CIHI DAD), which links inpatient records of delivering mothers and their babies and has been used widely in prior population-based studies.6–8 Demographic information was obtained through the Registered Persons Database (RPDB) and CENSUS datasets to determine area-level income quintile and rurality based on postal code. The Immigration, Refugees and Citizenship Canada Permanent Resident Database was used to ascertain immigration status. The OHIP contains information on physician billings for all Ontario residents, and the ICES Physician Database and Corporate Physician Database contain physician specialty information, allowing ascertainment of antenatal care provider. Assessment of prior healthcare utilisation was determined through CIHI DAD, National Ambulatory Care Reporting System and OHIP. Datasets were linked using unique encoded identifiers and analysed at ICES.
Supplemental material
In Ontario, all medically necessary treatment, including invasive prenatal testing, is covered under OHIP. Since December 2015, the physician fees for one cycle of in vitro fertilisation are covered under OHIP; fertility medications and other assisted reproductive technology (ART) (eg, ovulation induction with intrauterine insemination) are not publicly funded.
Study population, exposure assessment and matching
We included women in Ontario who experienced a live or stillbirth delivered at 20+ weeks from 1 April 2009 to 26 November 2018. The exposure of interest was physician occupation; we included all female physicians licenced by the CPSO between 1 January 1995 and 26 November 2018 residing in urban areas. Non-physician individuals who experienced a delivery were identified from MOMBABY and linked to the RPDB; we restricted to those residing in high-income urban areas to minimise the potential impact of observed differences related to socioeconomic status rather than physician occupation.
Deliveries among physicians were matched to deliveries among non-physicians using 1:5 exact ratio matching without replacement. We matched on maternal age at delivery (±1 year), calendar year of delivery (±1 year), parity (0 vs 1+ prior births), use of ART in the index pregnancy (binary) and singleton versus multifetal gestation.
Demographic characteristics were measured at the time of the delivery and included maternal age (continuous), calendar year of delivery (continuous), immigration status (recent immigrant within 5 years, long-term immigrant>5 years, resident),9 comorbidities (categorical), pre-existing hypertension (binary), hypertensive disorders of pregnancy (binary), pre-existing diabetes (binary), gestational diabetes (binary), previous live births (0, 1, 2+), previous stillbirths (0, 1, 2+), previous miscarriages (binary), multifetal gestation (binary), pregnancy conceived with any type of ART (binary, assessed through MOMBABY), gestational weeks at delivery (<37 weeks, 37–41 weeks, 41+ weeks), severe small for gestational age infant (SGA, defined as <5% for sex and gestational age, which has been found to more reliably capture true pathology than a <10% cut-off in population-based studies)10 and large for gestational age infant (LGA, defined as >10% for sex and gestational age). Comorbidities were categorised into Aggregated Diagnosis Groups (ADGs) (0, 1–5, 6–9, >10) based on healthcare utilisation in 2 years prior to the delivery using the Johns Hopkins ACG System V.10.11
Assessment of processes of care
We evaluated antepartum, intrapartum and postpartum processes of care and compared them between physicians and their matched counterparts. The processes encompassed investigations, procedures or other care decisions that the pregnant individual may choose or decline and which may influence their obstetrical outcome.
Antepartum care
We evaluated number of antenatal visits, number of antenatal ultrasounds, use of invasive prenatal testing (chorionic villus sampling or amniocentesis) and antenatal external cephalic version (OHIP), a procedure performed late in pregnancy to achieve cephalic presentation, in the event that the fetus is breech.
Intrapartum care
We evaluated the type of care provider attending the delivery, which was ascertained through OHIP billings. If more than one provider was present, both providers were accounted for (eg, if a family physician and obstetrician attended a delivery, the delivery was counted under both providers). We also evaluated whether the individual’s own antenatal care provider attended their delivery, as some providers may deliver their own physician patients as a professional courtesy. We evaluated for differences in induction of labour, using a strict definition of mechanical or pharmacological induction (oral, vaginal or cervical administration of an oxytocic agent). We also evaluated for differences in uptake of episiotomy, OASIS, trial of labour after caesarean (TOLAC) and receipt of labour anaesthesia, defined based on anaesthesia billing codes.
Postpartum care
We evaluated the length of maternal stay in the hospital following delivery. The number of outpatient healthcare visits from discharge to 42 days postpartum was also assessed (maternal and neonatal, emergency department [ED] and office visits). Maternal and neonatal visits were assessed separately based on OHIP billing codes.
Outcome assessment
The primary outcome was mode of delivery (C-section vs operative vaginal delivery vs spontaneous vaginal delivery). Secondary outcomes were differences in other antepartum, intrapartum and postpartum processes of care, OASIS among those with a vaginal delivery and number of urgent healthcare contacts (maternal and neonatal emergency department visits or hospitalisations) during the 42-day postpartum period.
Statistical analysis
We described the distributions of demographic and delivery characteristics of physicians and non-physicians using medians, interquartile ranges and proportions and standardised differences before and after matching. We then evaluated antepartum and intrapartum processes of care within the matched sets to determine if there were differences in these factors between physicians and non-physicians.
To estimate the risk of C-section versus vaginal delivery among physicians versus non-physicians (primary outcome), we applied a modified Poisson regression model under a generalised estimating equation approach to account for correlation between the matched sets. We adjusted for comorbidities (ADGs 0–5, 6–9, >10), diabetes (binary), gestational diabetes (binary), hypertensive disorders of pregnancy (binary), gestational age at delivery (weeks, categorical), SGA and LGA. This model was repeated for the outcome of OASIS among those undergoing vaginal birth. We also evaluated maternal postpartum urgent care (emergency department visits or hospitalisations) using a Poisson regression model, adjusting for the same factors with the addition of mode of delivery. Neonatal care was evaluated using a Poisson regression model, adjusting for SGA, LGA and gestational weeks at delivery.
All statistical tests were two sided, with p<0.05 considered statistically significant and standardised differences>0.10 considered a meaningful difference between groups.12 Complete case analyses were performed as data were rarely missing (use of ART missing in <0.001%). Analyses were performed using SAS V.9.4.
Patient and public involvement
None.
Results
Study population
We identified 7237 deliveries among 4667 physicians and compared them to 177 038 deliveries among 138 074 non-physicians (figure 1). Prior to matching, physicians were older than non-physicians at delivery, with a median age of 34 years (IQR 32–36) in physicians and 32 years (IQR 29–35) in non-physicians (standardised difference (StD) 0.39). Physicians had fewer ADGs, with an overall median of 5 (IQR 3–7) among physicians versus 6 (IQR 4–8, StD=0.40) among non-physicians. Physicians were more likely to be nulliparous compared with non-physicians (48.9% vs 43.2%, StD=0.11) and more likely to have conceived with ART compared with non-physicians (8.7% vs 4.8%, StD=0.16). Physicians and non-physicians had experienced similar rates of prior miscarriage (24.3% vs 25.4%, StD=0.03) (table 1).
Flowchart of physicians and non-physicians included in the study. GA, gestational age.
Characteristics of physicians and non-physicians at the time of delivery before and after matching
After matching, physicians and non-physicians were well balanced in their characteristics, except for comorbidities (median ADGs 5 (IQR 3–7) for physicians and 6 (IQR 4–8) among non-physicians, StD=0.40). Physicians were less likely to deliver postdates; only 7.5% of physicians delivered in or beyond their 41st week of pregnancy compared with 11.5% of non-physicians (StD=0.11). Compared with physicians, non-physicians were more likely to have an LGA fetus (4.4% vs 2.4%, StD=0.11).
Processes of care
Antepartum care
The median number of antepartum visits was similar between physicians and non-physicians (10 visits (IQR 9–12) vs 11 visits (IQR 8–12), StD=0.08). The median number of ultrasounds during pregnancy was higher among physicians (6 (IQR 4–8) vs 5 (IQR 3–8), StD=0.12). Physicians were more likely to undergo invasive prenatal diagnostic testing (6.3% vs 3.9%, StD=0.11) (table 2).
Antepartum and intrapartum processes of care for matched physicians and non-physicians
Intrapartum care
Physicians and non-physicians had a similar frequency of attendance by an obstetrician at their delivery (85.0% and 84.8%, respectively, StD=0.03). Physicians were more likely to be delivered by the obstetrician or family physicians who followed them in their pregnancy compared with non-physicians (physicians: n=2537, 35.1%; non-physicians: n=9772, 27.0%). Non-physicians were more likely to have a provider other than an obstetrician or family physician attend their delivery (9.5% of non-physicians vs 5.4% of physicians, StD=0.16). There were no differences in induction of labour, use of episiotomy or TOLAC between physicians and non-physicians. Physicians were more likely to have labour anaesthesia (87.2% vs 80.0%, StD=0.20) (table 2).
Mode of delivery
The estimated relative risk of C-section was slightly lower for physicians (29.9% vs 32.2%, RR 0.93, 95% CI 0.89 to 0.86, p<0.0001). After adjusting for comorbidities, diabetes (pre-existing and gestational), hypertension (pre-existing and hypertensive disorders of pregnancy), gestational age, SGA and LGA, there was no difference in the risk of C-section (aRR 0.97, 95% CI 0.91 to 1.00, p=0.07). Conversely, the risk of spontaneous vaginal delivery was slightly higher among physicians in unadjusted analyses (58.6% vs 56.6%, RR 1.03, 95% CI 1.01 to 1.06, p=0.0013), but after adjustment, there was no difference (aRR 1.01, 95% CI 0.99 to 1.03, p=0.25). The risk of operative vaginal delivery was similar between physicians and non-physicians before and after adjustment (11.6% vs 11.2%, aRR 1.03, 95% CI −0.96 to 1.11, p=0.36); among those undergoing operative vaginal delivery, 172 (20.5%) of physicians and 953 (23.6%) of non-physicians had a forceps-assisted vaginal delivery; the remainder had vacuum-assisted deliveries. There was no difference in the risk of OASIS; the aRR was not calculable due to the low number of individuals experiencing the outcome.
Postpartum
Following delivery, there were no differences in length of stay or outpatient maternal and neonatal visits between physicians and non-physicians (table 2). In adjusted analyses, considering maternal comorbidities and mode of delivery, non-physicians had a higher risk of maternal urgent postpartum care (ED or hospitalisation in the first 6 weeks postpartum) than physicians (aRR 1.22, 95% CI 1.08 to 1.37, p<0.0001). Children born to physicians and non-physicians had a similar risk of ED visits and hospitalisation, after adjustment for gestational weeks of delivery and growth abnormalities (aRR 0.81, 95% CI 0.62 to 1.06, p=0.13) (table 3). Although the direction of the effect is lower, these differences were not significant.
Intrapartum and postpartum outcomes for matched physicians and non-physicians
Discussion
In this population-based study of over 200 000 individuals, there were few differences in processes of obstetrical care for physicians and non-physicians. Physicians were significantly older at the time of delivery, were more likely to be nulliparous, had fewer comorbidities and were significantly more likely to have conceived with ART. After matching to ensure balance between groups, physicians were more likely to undergo prenatal ultrasounds and invasive prenatal diagnostic testing, receive labour anaesthesia and be delivered by their own care provider but were less likely to have postpartum maternal urgent care visits than non-physicians. There were no differences in mode of delivery or other any other obstetrical outcomes.
Physician access to the healthcare system may lead to different obstetrical outcomes through attendance of their own antenatal care provider at delivery and access to additional visits and diagnostic imaging through pregnancy. In this study, physicians were more likely than non-physicians to be delivered by their own antenatal care obstetrician provider. Outcomes associated with delivery by one’s antenatal care provider compared with the on-call obstetrician have been evaluated in one single-centre study, and only marginal effects on caesarean delivery and third-degree and fourth-degree lacerations were seen, favouring delivery by the antenatal care provider.13 Conversely, increased access may result in poorer outcomes; as an example, additional ultrasound surveillance may result in findings that lack clinical significance but lead to further intervention.3 Although physicians in this study experienced higher rates of invasive prenatal diagnostic testing and more ultrasounds compared with non-physicians, their processes of care and outcomes were largely similar. Notably, rates of caesarean delivery were similar between physicians and non-physicians. Johnson and Rehavi1 evaluated differences in C-section rates between physicians and non-physician parents with at least one college degree in California and Texas (where the baseline C-section rates are 33.2% and 35.3%, respectively) and found that US physicians were 10% less likely to have a caesarean delivery compared with their non-physician counterparts. They postulated that physicians were less susceptible to the financial incentives of treating physicians in the US healthcare system.1 In our study, the rates of operative delivery (forceps- and vacuum-assisted delivery) between physicians and non-physicians were similar. Our findings suggest that in Ontario’s public healthcare system, where healthcare providers are paid equally regardless of mode of delivery, non-physicians receive the same standard of intrapartum obstetrical care as physicians.
A prior Canadian study examined labour and delivery experiences of physicians through qualitative interviews.14 They found that physicians’ high level of knowledge regarding obstetrics was a ‘double-edged sword’, as they exhibited a higher level of understanding but also more anticipatory anxiety based on prior clinical experiences. Many of the interviewees felt that physicians may have worse obstetrical outcomes due to increased monitoring and intervention or lower tolerance for risk. Our study, inclusive of all physicians delivering in Ontario, is reassuring—in spite of the higher use of some medical interventions among physicians, such as labour anaesthesia, there were no differences in mode of delivery. Our prior work using these data has also demonstrated that maternal and perinatal outcomes for physicians are similar to those for non-physicians, after adjusting for maternal age and other important factors.15
Both the risk of miscarriage and infertility among physicians have been raised as concerns in recent literature.16–18 Physicians report higher rates of miscarriage than would be expected for the general population17 19; we were unable to evaluate rates of miscarriage among all physicians, but in this cohort of physicians experiencing an obstetrical delivery, they had similar rates of prior miscarriage compared with their non-physician counterparts. The higher rates of ART use among physicians in this study might be explained by superior access to healthcare resources or by a truly higher risk of infertility among physicians. The increased use of ART in our unmatched cohort should be noted; although the mechanism cannot be elucidated using these data, it may corroborate prior literature, where one in four woman physicians self-reported a diagnosis of infertility.16 Further investigation is needed to determine whether physicians’ increased use of these services is due to increased rates of infertility or increased access to care.
Our finding of fewer urgent postpartum visits among physicians suggests that physicians may have other means of access to care outside of emergency services in the postpartum period compared with non-physicians. Recent Canadian literature has identified limited access to follow-up with obstetrical providers in the 6-week postpartum period as a significant gap.20 21 In our matched cohort, 5.6% of physicians and 7.8% of non-physicians accessed emergency services during the postpartum period; providing options for timely care outside of the emergency department is an urgent priority for postpartum individuals in Ontario.
We compared obstetrical deliveries among all Ontario physicians versus non-physicians to explore processes of care and outcomes through a unique data linkage. This analysis included all practising physicians in Ontario; the data linkage allowed for a comprehensive assessment of outcomes. Still, there are limitations. We did not have occupational information on non-physicians. To create a fair comparator group, we compared physicians to non-physicians residing in high-income areas, as they would have fewer financial barriers to ART and fewer sociodemographic risk factors that may drive obstetrical outcomes. We were unable to explore the specific method of ART with the available data, as this variable included any method of ART. More physicians had pregnancies conceived with the use of ART compared with non-physicians, but we could not determine whether these differences were due to ART access or increased rates of infertility among physicians. Similarly, we could not determine if the increased rate of invasive testing among physicians was related to screening abnormalities or physician choice. We were unable to assess for differences in the use of elective versus urgent C-section due to a high proportion of patients missing this variable. Other important covariates, such as prepregnancy body mass index, were not available in these data.
Conclusions
Physicians and non-physicians delivering in Ontario experience similar obstetrical processes of care and outcomes, with a few notable exceptions, including increased antenatal ultrasounds, invasive prenatal testing, receipt of labour analgesia and delivery by own care provider. We found no difference in the rate of C-section among physicians and non-physicians, which may be related to the lack of cost drivers for C-section compared with vaginal birth in Ontario.
Data availability statement
All data relevant to the study are included in the article or uploaded as online supplemental information. The dataset from this study is held securely in coded form at ICES. While legal data sharing agreements between ICES and data providers (eg, healthcare organisations and government) prohibit ICES from making the dataset publicly available, access may be granted to those who meet prespecified criteria for confidential access, available at www.ices.on.ca/DAS. The full dataset creation plan and underlying analytic code are available from the authors upon request, understanding that the computer programs may rely on coding templates or macros that are unique to ICES and are therefore either inaccessible or may require modification.
Ethics statements
Patient consent for publication
Ethics approval
The study protocol was approved by the research ethics board at St. Michael’s Hospital (Toronto, Ontario, #18-248).
Acknowledgments
The authors thank Dr Peter Tanuseputro, Dr Manish Sood and Emily Rhodes at the Ottawa Hospital Research Institute for their assistance with data acquisition and the College of Physicians and Surgeons of Ontario for providing these data. This document used data adapted from the Statistics Canada Postal CodeOM Conversion File, which is based on data licensed from Canada Post Corporation, and/or data adapted from the Ontario Ministry of Health Postal Code Conversion File, which contains data copied under licence from Canada Post Corporation and Statistics Canada. Parts of this material are based on data and/or information compiled and provided by Canadian Institutes for Health Information and the Ontario Ministry of Health. Parts or whole of this material are based on data and/or information compiled and provided by Immigration, Refugees and Citizenship Canada (IRCC) current to September 2020. However, the analyses, conclusions, opinions and statements expressed in the material are those of the author(s) and not necessarily those of IRCC.
References
Footnotes
X @mccusimano
Contributors All authors contributed to the design of this study. ANS, MCC and NNB obtained ethics approval. ANS and NNB participated in data acquisition. ANS, RS, MCC, EM and NNB developed the analytic plan. EM and RS performed the statistical analysis. ANS wrote the first draft and created all tables and figures. All authors critically revised the manuscript and approved the final version submitted and agree to be accountable for all aspects of the work. ANS is responsible for the overall content as guarantor.
Funding This work was supported by the Physicians’ Services Incorporated (PSI) New Investigator Grant (grant #19-12, ANS and NNB) and by ICES, which is funded by an annual grant from the Ontario Ministry of Health (MOH) and the Ministry of Long-Term Care (MLTC). The analyses, conclusions, opinions and statements expressed here are solely those of the authors and do not reflect those of the funding or data sources; no endorsement by ICES or the MOHLTC is intended or should be inferred . The funder had no role in conducting the research or writing this manuscript.
Competing interests ANS receives salary support from the St. Michael's Hospital Chair in Women’s Health (Dr Sari Kives) and the University of Toronto Department of Obstetrics & Gynaecology Merit Award. Other authors declare no competing interests .
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.