Article Text
PDF
Abstract
At the start of the COVID-19 pandemic, the Jim Pattison Diabetes and Pregnancy (JP DAP) clinic quickly switched from in-person to virtual care for patients with gestational diabetes (GDM) to reduce the risk of viral transmission. Poor glycaemic control in pregnancies increases the risk of maternal–fetal complications and thus women with GDM require education, frequent follow-up and treatment to reduce these risks. Delays in care could potentially result in increased maternal–fetal complications. We conducted a prospective, single-centre quality improvement (QI) study of women with GDM who attended the JP DAP clinic and delivered between 1 September 2019 and 31 March 2021. 2123 singleton pregnancies between 1 September 2019 and 31 March 2021 with GDM were analysed for this study. The time of referral to see the endocrinologist was lower than baseline in the first wave but rose significantly in the second wave. No-shows for appointments increased in the first wave but were lower than baseline after the implementation of time slots. There was no special cause variation for maternal–fetal complications pre pandemic, first wave or during the second wave. A patient satisfaction survey reported that 93% of respondents strongly agreed or agreed with the statement ‘I was satisfied with the care provided to me over the telephone appointments’. The GDM education package, online educational videos in Hindi and English and the glucometer smartphone application helped to maintain the time of referral to first endocrinologist appointment in the first wave and therefore were considered an effective substitute for in-person education. Despite the delays in care seen in the second wave, there was no increase in maternal-fetal complications. Our clinic plans to continue using virtual tools for the foreseeable future.
- Continuous quality improvement
- COVID-19
- Diabetes mellitus
- PDSA
- Quality improvement
Data availability statement
Data are available upon reasonable request. The deidentified data is available upon reasonable request to JE, the data analyst for this project.
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
WHAT IS ALREADY KNOWN ON THIS TOPIC
This is the first quality improvement (QI) study that we are aware of studying the effect of switching to virtual care for patients with gestational diabetes (GDM) during the COVID-19 pandemic.
WHAT THIS STUDY ADDS
This QI study demonstrates that the use of virtual tools in GDM helped to maintain the waiting time to see an Endocrinologist and that there was no detrimental effect on maternal–fetal complications after switching to virtual care.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
Virtual tools can be useful in the management of pregnancies with GDM and were well received by patients. In our clinic, we plan to continue these virtual tools in combination with in-person care for the foreseeable future.
Introduction
Problem description
The COVID-19 pandemic has resulted in major impacts and disruptions to healthcare systems, fundamentally changing how healthcare providers deliver care to their patients in an effective and safe manner. Many areas of medicine have transitioned to offering virtual care to reduce viral transmission. This includes the Diabetes and Pregnancy Clinic based out of Jim Pattison Outpatient Care and Surgical Centre (JP DAP) in Surrey, British Columbia, Canada.
Endocrinologists, nurses and dietitians provide this care and work collaboratively with obstetricians and maternal–fetal medicine in preventing maternal–fetal complications related to poor glycaemic control in pregnancy. This involves numerous factors such as timely access to the screening and diagnosis of gestational diabetes (GDM), appropriate triaging of referrals, conducting initial assessments and follow-ups, as well as the commencement and teaching of any necessary treatments, including glucose monitoring and insulin self-administration. The province of British Columbia has the highest rate of GDM in Canada, affecting an average of 73.7 per 1000 live births.1
When the pandemic was declared in March 2020, we were concerned that the COVID-19 pandemic and associated restrictions could substantially reduce access for pregnant patients to the laboratory, and their physicians in the setting of an overwhelmed healthcare system. To decrease the risk of viral transmission, physicians and healthcare institutions switched to virtual care. In our local healthcare delivery system, virtual care was not widely used before March 2020. As a result, patients and care providers were required to navigate the multiple challenges of the sudden switch to virtual care.
There could be many potential indirect effects on maternal–fetal outcomes from the COVID-19 pandemic, including changes in healthcare-seeking behaviours, delays in diagnosis, disruption and reduced access to maternity services.2 3
The purpose of this project is to assess how the transition to virtual care since March 2020 has impacted the outcomes around GDM, with our aim of identifying and minimising any delays to care for women with GDM. With the joint efforts between maternal–fetal medicine, pediatrics, and endocrinology, we conducted a quality improvement (QI) project by comparing and serially monitoring several measures before and after the transition to virtual care for the DAP clinic.
Available knowledge
Optimising glycaemic control in pregnancy involves numerous factors such as timely access to the screening and diagnosis of GDM, appropriate triaging of referrals, conducting initial assessments and follow-ups, glucose monitoring and insulin self-administration if necessary.4
Endocrinologists, nurses and dietitians work collaboratively with obstetricians and maternal–fetal medicine to prevent maternal–fetal complications secondary to poor glycaemic control in pregnancies complicated with GDM.
Poor glycaemic control is known to increase the risk of maternal–fetal complications in GDM such as large for gestational age (LGA) infant, neonatal hypoglycaemia, jaundice, need for c-section and neonatal intensive care unit (NICU) admission.4 5
The treatment for GDM has been shown to decrease these complications.6 The treatment starts with counselling on self-monitoring, diet and activity. Some patients, despite optimal diet and lifestyle, still require insulin in pregnancy.4
Rationale
If the JP DAP clinic had to be switched from in-person to virtual care because of COVID-19 precautions, then there should be no delay in care for women with GDM so that they can achieve optimal glycaemic control in their pregnancy, and therefore no increase in maternal–fetal complications.
In response, our clinic made several changes to the clinic structure and flow to switch to virtual care to ensure we maintained effective patient care and followed COVID-19 precautions.
Delays in care could delay the optimisation of glycaemic control and thus increase the risk of maternal–fetal complications in pregnancies with GDM.
Specific aims
There should be no delay in the care for women with GDM attending the JP DAP clinic during the first year of the COVID-19 pandemic.
Methods
Context
This was a prospective, single-centre QI study of women with GDM who attended the JP DAP clinic in Surrey, British Columbia. This clinic provides care to pregnant women with GDM, pre-diabetes, type 2 diabetes and type 1 diabetes. This clinic is associated with Surrey Memorial Hospital, the primary delivery site for the patients attending the clinic. The JP DAP clinic is in the Fraser Health Authority, which is the largest health authority in British Columbia. Surrey is the second largest city in British Columbia with a population of over 568 000 and growing. Surrey is a multicultural city and a third of the population is South Asian.7 Hindi and Punjabi are the predominant languages in the South Asian community.
In this QI study, we analysed deliveries of patients attending the JP DAP clinic from 1 September 2019 to 31 March 2021 in Fraser Health. Pregnancies with pre-existing type 1 diabetes, type 2 diabetes or pre-diabetes, multiple pregnancies and deliveries outside of Fraser Health were excluded.
The clinical team consists of six endocrinologists, nurses, dieticians, clerical staff and administration. The QI team consisted of endocrinologists, the endocrinology fellow (Dr Tong), administration and data analyst/QI consultant.
The clinical team met every 3 months. In between meetings, the clinical team would discuss over email changes made to the clinic structure and flow. The QI team communicated frequently during the data collection phase. The QI team collaborated with maternal–fetal medicine and pediatrics before the project began and later as needed.
Prior to the pandemic, GDM patients were booked in person for classes led by nurses and dieticians (‘clinicians’) for education on the aetiology and management of GDM. At the next visit, the patient would see either a nurse or dietician and then have their consult with the endocrinologist. All consults and follow-ups were done in person except for phone follow-ups which were booked as needed.
The COVID-19 pandemic was declared a public health emergency by the WHO and Canada in March 2020. The British Columbia Public Health Officer recommended all regulated health professionals switch to minimal in-person patient visits to prevent the spread of the virus. Based on these and directives from the Fraser Health Authority, we switched to primarily virtual care on 17 March 2020.
Interventions
Given the urgent change to virtual care, the clinic had to make multiples changes including limiting in-person appointments except for high-risk patients such as those with pre-existing diabetes and new insulin starts, implementing Fraser Health COVID-19 safety protocols, mandating a limit of a maximum of 32 patients a day and no longer accepting referrals for the preconceptual consults.
All patients were instructed to pick up a GDM education package that contained the glucose logbook, education material, glucometer and prescription for glucometer strips. Patients were booked for televisits for intake and review of the educational material with a clinician, as in-person group classes were no longer held due to COVID-19 precautions.
In-person consultation with an endocrinologist was booked if the patient was considered high risk for needing insulin during pregnancy (fasting blood glucose of >5.5 mmol/L on the oral glucose tolerance test, fasting glucose average>5.5 mmol/L at the intake visit or a fetal abdominal circumference greater than the 90th percentile) so that insulin teaching, if indicated, could be started as soon as possible.
Virtual care was conducted over the telephone, as access to video appointments was not initially available at our institution. The staff and patients were comfortable with the televisits, so the clinical team decided not to switch to video conferencing.
Follow-up visits were booked virtually unless booked otherwise by the endocrinologists. The endocrinologists called the patients who were not on insulin therapy (ie, ‘diet controlled’), while the clinicians called the patients on insulin and passed on the chart to the endocrinologist who would call the patient afterward for follow-up. This change in practice helped to alleviate the burden of work on the clinicians who were adapting to the new system.
During the first month of virtual care, the clinicians conducting the intake visits found that some patients were having difficulties with the education reading material. Some did not read the educational material and/or needed more time for teaching than a virtual visit could provide. In April 2020, the English (https://www.youtube.com/watch?v=BHSHhahP15A_) and the Hindi GDM education video (https://www.youtube.com/watch?v=Q9ai6DmLa0U&feature=youtu.be) became available for patients to review as an additional resource.
The endocrinologists and clinicians found that the virtual visits were taking longer than expected. One reason was that clinicians found obtaining glucose readings verbally over the telephone to be inefficient. The clinic was already using the OneTouch Verio glucometer, which has a smartphone application that synchronises blood glucose readings via Bluetooth from the patient’s glucometer metre to a secured cloud. Thus, the clinicians and endocrinologists could review the glucometer data from the One Touch website (https://www.onetouchreveal.ca/login) remotely and prior to the virtual visit. The use of this application was approved for use in the clinic by Fraser Health in May 2020.
There was a high frequency of ‘no-shows’ or patients who did not answer the phone or call back for their televisits. In response, in August 2020, we allocated time slots for televisits (ie, patients were booked between 830 and 1000 hours, 1000 and 1200 hours or 1300 and 1500 hours) and tracked the no-show rate and patient satisfaction data.
Study of the interventions
We used the Institute of Healthcare Improvement QI methodology and developed a driver diagram to help guide the tests of change in the improvement and used the measures below to study the effect of the interventions on our aim.
Data was obtained from Meditech and the British Columbia Perinatal database for all measures, with the exception of patient satisfaction. To gather patient satisfaction feedback, a randomised selection process was employed, and 20 patients per week received an emailed survey from 8 October 2020 to 7 January 2021. The survey was administered using SurveyMonkey, with language options available in English, Punjabi, Mandarin and Arabic. Responses were kept anonymous and not linked to patient identification. It is worth noting that no user satisfaction data was available prior to the pandemic.
Measures
Outcome measures
The time of referral to first endocrinologist appointment.
Process measures
The number of no-shows for appointments for mothers who delivered.
The number of patients started on insulin in pregnancy.
Balancing measures
The percentage of pregnancies complicated with GDM with one or more of the following maternal–fetal complication(s): LGA infants, neonatal hypoglycaemia, neonatal hyperbilirubinaemia, NICU admission and c-section.
Patient satisfaction.
The outcome measure of the time of referral to first endocrinologist appointment was chosen for two reasons. Prior to the pandemic, the clinic was using this as a measure of waiting time. The target time of referral to first endocrinologist appointment is 7–10 days for the clinic. Secondly, long delays in care due to the pandemic could result in prolonged fetal exposure to uncontrolled maternal glycaemic control which would subsequently increase the risk of maternal–fetal complications. GDM is usually screened between 24 and 28 weeks of gestational age, therefore most patients only have 12 to 16 weeks to obtain optimal glycaemic control in pregnancy, and thus, assessments completed in a timely manner are crucial.
Patients were started on insulin if their capillary blood glucose values were consistently above target despite adequate dietary and lifestyle management. Target blood glucose readings at Fraser Health for GDM are fasting<5.3 mmol/L, 1 hour after meals<7.8 mmol/L and 2 hour after meals<6.7 mmol/L.
The number of no-shows for appointment, commonly referred to as ‘no-shows’, for mothers who delivered was a significant concern. When patients miss their scheduled appointment, it necessitates rescheduling which not only occupies a slot that could have been used by other patients but also leads to delays in providing care.
The number of patients started on insulin in pregnancy was chosen as a process measure because it is a marker of poor glycaemic control. Delays in care and lockdown measures could increase the risk of poor glycaemic control requiring insulin treatment.
All the individual components of the composite maternal–fetal outcome measure are known complications of poorly controlled GDM.5 LGA was defined as per growth charts based on local British Columbia data for all newborns born between 1981 and 2000 and was used as a reference to categorise newborns as LGA and was defined as being greater than the 90th percentile based on sex. Neonatal hypoglycaemia was defined as a plasma blood glucose<2.6 mmol/L in the infant’s first 48 hours of life. Neonatal hyperbilirubinaemia and c-section were coded in the chart. Primary and secondary c-section rates were included, as it was difficult for the perinatal database to differentiate between them.
Even though reducing GDM complications has the utmost importance, it was not a feasible outcome measure in our study due to our system’s capacity to capture GDM outcomes (ie, mothers would have to deliver first, data needed to be coded after discharge and some metrics had to be manually reviewed, etc), and it would interfere with our ability to quickly respond with changes during the evolving pandemic. Nonetheless, we were still able to capture this as a balancing measure.
Analysis
Data were obtained from scanned charts, Meditech electronic medical record and the British Columbia Perinatal Database and were analysed by our QI data analyst/consultant using Excel and SQC pack V.7.
We used Shewhart charts to compare data from the baseline prior to the pandemic (1 September 2019 to 16 March 2020) and the intervention period (17 March 2020 to 31 March 2021). These charts were analysed using standard rules to detect special causes of variation in the system.
The patient satisfaction survey results were analysed using Pareto charts.
This study follows the SQUIRE V.2.0 publications for reporting.8
Results
In total, 2123 singleton pregnancies with GDM between 1 September 2019 and 31 March 2021 were reviewed for this study.
From September 2019 to March 2020 (baseline), there was an average of 159 referrals per month with a compared with an average of 136 referrals per month from April 2020 to October 2020, earlier in the pandemic. However, in 2021, the mean number of referrals started to increase back to the baseline.
Figure 1 shows the time of referral to first endocrinologist appointment from August 2019 to the end of April 2021. Prior to the pandemic, patients had to wait 18 days on average to see an endocrinologist, exceeding the clinic’s target of 7–10 days. To improve waiting times, an extra nurse was added on Wednesdays and Fridays and another clinic on Fridays with an endocrinologist was added resulting in a decrease in the average wait time from 18 to 15 days. Following the first wave of COVID-19, which occurred between 16 March 2020 and 18 October 2020, additional changes were tested, including switching to virtual appointments, introducing educational videos, implementing a glucometer smartphone application and modifying the appointment scheduling process. These changes led to a reduction in the mean wait time from 15 to 12 days. However, this improvement was not sustained, as the wait time significantly increased at the onset of the second wave, peaking at 27 days before subsequently decreasing to an average of 14 days by the end of the second wave.
Wait time to first Endocrinologist appointment.
Figure 2 shows the number of no-shows for appointments for mothers who delivered. Prior to the pandemic, there was a mean of 5 no-shows weekly which rose to mean of 16 no-shows in the first wave of COVID-19. By implementing the time spots for appointments, the number of no-shows decreased substantially to mean of 2 no-shows a week, which is even lower than the prepandemic baseline.
Number of "no-shows" for their virtual or in-person appointment
Figure 3 presents the percentage of births with at least one maternal–fetal complication, including conditions such as LGA infant, c-section, NICU admission, neonatal hypoglycaemia and neonatal hyperbilirubinaemia. There was no evidence of any notable increase in maternal–fetal complications during the first year of the pandemic.
A total of 224 patients were contacted, and 66 patients completed the patient satisfaction survey (29.4% response rate). Only one survey was done in Punjabi, one survey was done in Mandarin and no surveys were done in Arabic. Overall, 89% of respondents strongly agreed or agreed with the statement ‘clinic services were available to me when I needed them’. Overall, 93% of respondents strongly agreed or agreed with the statement ‘I was satisfied with the care provided to me over the telephone appointments’. Overall, 85% of respondents strongly agreed or agreed with the statement ‘If I had to come in for an in-person appointment, I would feel safe doing so’. About 85% of the respondents watched the online educational video prior to their first appointment at the clinic. Overall, 86% of respondents found the education material useful for first visit preparation. Overall, 93% of the respondents read the Diabetes and Pregnancy handbook prior to their first appointment at the clinic. Altogether, 75% of the respondents knew about the OneTouch Reveal mobile application that can be used to share their blood glucose results with the healthcare team online. There was no satisfaction data available before the pandemic.
Percentage of births with any event(s): large for gestational age, c-section, neonatal intensive care unit, hypoglycaemia and/or neonatal hyperbilirubinaemia.
Discussion
Summary
The GDM education package, online GDM educational videos in Hindi and English and the glucometer smartphone application helped to maintain a similar time of referral to first endocrinologist appointment during the first wave and therefore were considered an effective substitute for in-person education. Patients were satisfied with these virtual tools and the virtual care they were provided in the first year of the pandemic based on the patient satisfaction survey results. The no-show rate was even lower than before the pandemic with the implementation of time slots.
However, in the second wave, the time of the referral to the first endocrinologist appointment and the number of insulin starts started to increase. Despite this, there was no difference in maternal–fetal outcomes prior to the pandemic and in the first year of the pandemic.
To our knowledge, this is the first published QI study about the transition to virtual care for GDM during the pandemic.
A meta-analysis of randomised control trials studying the effect of digital health solutions in GDM showed better glycaemic control and a reduction in the risk of macrosomia and c-section.9 Another meta-analysis of randomised control trials studying the effect of telehealth versus standard care showed an improvement in glycaemic control and a reduction in the rates of c- section, pregnancy-induced hypertension or pre-eclampsia, premature rupture of membranes, premature birth, neonatal asphyxia and polyhydramnios.10
The sudden switch to virtual care presented challenges for the clinical team. The team addressed these challenges using plan–do–study–act cycles as discussed above. As a result, the clinical team was able to adjust processes and implement changes to improve waiting times and other metrics. In addition, it helped to foster a QI culture which continues to this day.
Streamlining and optimisation of this data collection with semiregular QI review can provide valuable information including ongoing monitoring of the system, feedback to the healthcare teams involved and prompt need for further changes even after the pandemic.
Interpretation
The time of referral to first endocrinologist appointment was decreased in the first wave. Reduced access to primary and obstetrical care in the first wave likely resulted in fewer referrals and screening for GDM. There was also significant anxiety about the risk of COVID-19 transmission for patients attending in-person care during the first wave which could have also decreased the volume of referrals. Fewer referrals would thus result in quicker waiting times to see an endocrinologist.
Earlier in the pandemic, the Society of Obstetricians and Gynaecologists of Canada and Diabetes Canada posted a joint statement recommending alternative, less stringent testing for GDM with an A1c and random glucose to ease the burden on the healthcare system in the first wave instead of the standard testing with a 75 gm oral glucose tolerance test.11 However, in British Columbia, we continued the standard screening protocols due to the lower pooled sensitivity of A1c and random glucose (25%) and the presumed low prevalence of COVID-19 at that time.12 Our clinic reviewed this issue and recommended continuing the standard screening protocols as there were minimal disruptions to laboratory testing.
If the less stringent criteria had been used, then fewer cases of GDM would have been diagnosed, so potentially there could have been a lower rate of GDM related maternal–fetal complications. However, the less stringent criteria would have captured patients at very high risk of complications, so it is hard to quantify the true risk that missing milder GDM cases could have had.
The delayed time to see an endocrinologist observed in the second wave may be likely due to factors such as improved access to the healthcare system as COVID-19 precautions were more established and the overwhelming demand on the medical system as the numbers of COVID-19 cases and hospitalisations rose in the second wave.
In particular, the lockdown measures in the first wave may have contributed to the delays in seeing an endocrinologist and the increased frequency of insulin starts in the second wave. The patients who delivered in the second wave conceived or were early in their pregnancy around the time of the first COVID-19 lockdown in March 2020. Restrictions in British Columbia included an order for working from home, a switch to virtual healthcare and the closing of gyms and other activities. It was a stressful time for many, particularly pregnant women who were not able to access prenatal care in person and thus referrals to the JP DAP clinic may have been delayed. Stress is known to increase blood glucose. In addition, patients may have been not eating as healthy before and were less active, due to the lockdown all of which could cause weight gain and thus increase insulin resistance and lead to more cases of GDM and poor glycaemic control.
Limitations
Unfortunately, the survey response rate fell below our expectations. In hindsight, it appears that we would have achieved a higher response rate had we proactively reached out to patients after their initial visit with the endocrinologist. This approach would have enabled us to determine the preferred language for the survey.
Our findings show that virtual care worked well when COVID-19 precautions were mandated but it lacked face-to-face human interaction between patients and clinicians. Once in-person care was recommended by the British Columbia Ministry of Health and Fraser Health Authority in 2021, our clinic switched to a hybrid model of in-person and virtual care. Patients are initially seen in person for their first endocrinologist appointment and then have follow-ups via televisits unless the patient requires insulin. We plan to continue with this hybrid in-person/virtual model of care for the foreseeable future, as patients and staff expressed a preference for keeping some form of virtual care.
The lack of sustainability data beyond April 2021 is due to the limitations of the ethics approval. However, this presents an opportunity for further work in this field.
One limitation of this study is that it was conducted at a single DAP clinic. There are no plans to spread this project as our clinic has now switched to a hybrid model of in-person and virtual care as discussed earlier. In addition, it would be difficult to spread as other DAP clinics have different clinic structures and flow and have already implemented changes to their clinics due to the pandemic. However, we shared our resources including the videos to other clinics in our region.
Conclusion
This QI study showcases the journey that our patients with GDM and staff have made in the transition to virtual care in the first year of the pandemic. Despite the delays in care seen in the second wave and higher amount of insulin starts, there was reassuringly no increase in maternal-fetal complications in the first and second year of the COVID-19 pandemic. If these virtual tools were not implemented at the start of the pandemic, there would have likely been significant delays in care, which could have led to poorer glycaemic control and potentially higher rates of maternal–fetal complications.
Data availability statement
Data are available upon reasonable request. The deidentified data is available upon reasonable request to JE, the data analyst for this project.
Ethics statements
Patient consent for publication
Ethics approval
The Fraser Health Research Ethics Board provided a waiver for this study as it is a quality improvement study. The ARECCI Ethics Guideline for Quality Improvement and Evaluation Projects was completed with a score of 4.
Acknowledgments
We would like to thank and acknowledge the administrators, clerical staff, nurses and dietitians at the Jim Pattison Diabetes and Pregnancy clinic for their support and assistance for this project.
Footnotes
Contributors Guarantor: RK. Conceptualisation and methodology: RK, JT and JE. Data collection: JT, RK and JE. Data analysis and graph preparation: JE. Original draft preparation: RK. Review and editing: RK, JT, JE and SS. Participation in virtual care clinic and survey: RK, JT, SS, NS, AJ, JK and SP.
Funding Surrey Medical Staff Association, no award/grant number. Doctors of BC: Funding for manuscript writing and publication, no award/grant number.
Competing interests RK: funding for manuscript writing and publishing from Doctors of BC. All authors:
Funding for project from Surrey Medical Staff Association. No award/grant number.
Patient and public involvement Patients and/or the public were involved in the design, or conduct, or reporting, or dissemination plans of this research. Refer to the Methods section for further details.
Provenance and peer review Not commissioned; externally peer reviewed.