Sage Journals: Discover world-class research

Abstract

Background:

Strict monitoring of blood glucose during pregnancy is essential for ensuring optimal maternal and neonatal outcomes. Telemedicine could be a promising solution for supporting diabetes management; however, an updated meta-analysis is warranted. This study assesses the effects of telemedicine solutions for managing gestational and pregestational diabetes.

Methods:

PubMed, EMBASE, Cochrane Library Central Register of Controlled Trials, and CINAHL were searched up to October 14, 2020. All randomized trials assessing the effects of telemedicine in managing diabetes in pregnancy relative to any comparator without the use of telemedicine were included. The primary outcome was infant birth weight. A meta-analysis comparing the mean difference (MD) in birth weight across studies was applied, and subgroup and sensitivity analyses were performed. The revised Cochrane tool was applied to assess the risk of bias, and the certainty of evidence was evaluated using the Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) approach.

Results:

From a total of 18 studies, ten (totaling 899 participants) were used to calculate the effect on infant birth weight. The results nonsignificantly favored the control (MD of 19.34 g; [95% confidence interval, CI −47.8; 86.47]), with moderate effect certainty. Heterogeneity was moderate (I² = 37.39%). Statistically significant secondary outcomes included differences in two-hour glucose tolerance postpartum (gestational diabetes; two studies: standardized mean difference 9.62 mg/dL [95% CI: 1.95; 17.28]) that favored the control (GRADE level, very low) and risk of shoulder dystocia (four studies: log odds −1.34 [95% CI: −2.61; −0.08]) that favored telemedicine (GRADE, low).

Conclusions:

No evidence was found to support telemedicine as an alternative to usual care when considering maternal and fetal outcomes. However, further research is needed, including economic evaluations.

Keywords

gestational diabetes mellitus pregnancy self-monitoring telehealth glycemic control systematic review

Introduction

Diabetes during pregnancy is a major health problem worldwide. It is estimated that preexisting type 1 diabetes mellitus (T1DM), type 2 diabetes mellitus (T2DM), or gestational diabetes mellitus (GDM) are present in approximately 16% of all pregnancies.^1,2 If not sufficiently controlled, diabetes during pregnancy is associated with neonatal and maternal complications,³ such as neonatal hypoglycemia, perinatal mortality, polyhydramnios, shoulder dystocia, stillbirths, large for gestational age (LGA), and increased cesarean section risk.^4-7 Furthermore, both preexisting diabetes (T1DM and T2DM) and GDM have been shown to increase fetal macrosomia risk (birth weight >4000 g).^7-9 Fetal macrosomia is associated with serious adverse maternal and neonatal adverse outcomes, including emergency cesarean section, severe postpartum hemorrhage, obstetric anal sphincter injury, shoulder dystocia, obstetric brachial plexus injury, birth fractures, and hypoxic-ischemic encephalopathy.^10,11 Thus, interventions to manage diabetes during pregnancy to reduce the risk of fetal macrosomia are essential.^7,10 Another reason to manage diabetes during pregnancy is to minimize the risk of developing T2DM later in life.^11-13

Current management of preexisting T2DM and GDM typically includes nutrition therapy, exercise, and close glucose monitoring. Management of preexisting T1DM includes frequent titration of insulin to avoid hypoglycemia and hyperglycemia.^13-15 However, some women with preexisting diabetes or GDM do not meet the recommendations and target levels of glycemic control in diabetic pregnancy. Poor glycemic control in pregnancy is not well characterized, although the prevalence of women with poor glycemic control in developing countries has been reported to be 6% to 58% in South Africa,¹⁶ Sudan,¹⁷ Nigeria,¹⁸ and Pakistan.¹⁹ Inadequate glycemic control seems to be related to barriers to adherence,^20,21 which include lack of health information, teaching sessions, consultations, long waiting times, less tailored physical activity assessments, not knowing where to access appropriate information on the Internet, and unavailability of social media or support groups for women with GDM.²¹ The barriers related to poor glycemic control call for new solutions in the management of GDM to avoid maternal and neonatal complications as well as to prevent pressure on maternal and diabetic health services.

Telemedicine has shown promise in regard to supporting people with diabetes in their disease management.^22,23 Telemedicine is a broad term that includes telecommunication and information technology in the delivery of health care services. These services may include education, consultation, monitoring, and counseling tasks at a distance.^23,24 Telemedicine services can be provided using various constellations, including text messaging, data transmission (eg, blood glucose and blood pressure), or video consultations. Individualized feedback from health care professionals is a key element.^25,26 One may argue that telemedicine is an obvious solution for the management of diabetes in general since diabetes is mainly managed outside a hospital setting. In addition, telemedicine may provide value to patients who are unable to travel to health care facilities due to long distances and/or disabilities.²⁷

Previous research has shown that telemedicine is a promising method for improving glycemic control in people with diabetes in general.^28-31 Such potential benefits of telemedicine may also extend to people with preexisting diabetes and GDM.³² However, the approach still needs to be sufficiently evaluated with regard to its effectiveness in terms of patient-related outcomes within the diabetes context. In this context, a comparison and review of the effectiveness of different types of telemedicine interventions within preexisting diabetes in pregnancy and GDM is highly relevant. Other reviews have been performed within the field of preexisting diabetes and GDM.^32-36 However, a more comprehensive review is needed because the existing reviews are limited to a specific diabetes type^34,36 or narrow definitions of telemedicine.^32,33 Moreover, due to the rapid development of telemedicine solutions, new reviews and meta-analyses should be performed on a regular basis to embrace studies published since previous reviews. Hence, this systematic review and meta-analysis aimed to evaluate the effectiveness of telemedicine solutions relative to any comparator without the use of telemedicine with regard to diabetes-related outcomes (maternal, fetal, and delivery) for patients with gestational or pregestational diabetes.

Methods

Literature Search

This systematic review and meta-analysis followed the guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-analyses statement.³⁷ A search protocol has been published elsewhere (PROSPERO number CRD42020123565).³⁸ The literature search included database searches in PubMed, EMBASE, the Cochrane Library Central Register of Controlled Trials, and CINAHL. The search histories covering T1DM, T2DM, and gestational diabetes are presented in Supplemental Appendix 1. All studies published prior to October 14, 2020 were considered. Two review authors (S.H.L. and L.B.) performed the database searches in collaboration with a research librarian. Various search terms (different synonyms, near-synonyms, spellings, and acronyms) were included and combined to perform an exhaustive search. The main search terms were “telemedicine,” “diabetes,” and “randomized controlled trials.” The pregnancy component was included in the selection process (screening of titles/abstracts and full text). Different search functions, such as thesauruses, Boolean operators, truncation, abstracts/titles/keywords, phrases, free text, and advanced searches, were applied to focus and structure the search. Additional searches included citation searches in SCOPUS and Web of Science, manual searches of reference lists of relevant systematic reviews, and searches of other relevant references.

Study Selection

We considered randomized controlled trials published as peer-reviewed full-text articles that compared telemedicine solutions with any comparator. Both pregestational and gestational diabetes (GDM) were considered for inclusion. There were no restrictions on publication date, although we only included articles published in English, Danish, Norwegian, and Swedish. Four reviewers (S.H.L., S.H., L.B., and J.D.A.) independently screened all titles and abstracts for eligibility and read the remaining full-text versions. Any disagreement was resolved by discussion among the four reviewers or by including other reviewers (P.V., F.W.U., P.H.S., or O.K.H.) in the decision-making process.

Outcomes

All diabetes-related outcomes (maternal, fetal, and delivery) for patients with gestational or pregestational diabetes were considered in the review. The primary outcome was neonatal birth weight. Secondary outcomes were as follows:

Neonatal outcomes: stillbirth, jaundice, LGA, hypoglycemia, admissions to neonatal intensive care, respiratory distress, and shoulder dystocia.

Maternal outcomes: hypoglycemic episodes, two-hour oral glucose intolerance (postpartum in women with GDM), fasting blood glucose, glycated hemoglobin (HbA1c%), and pregnancy-induced hypertension/preeclampsia.

Delivery outcomes: gestational age, cesarean section, labor-induced delivery, preterm delivery, premature membrane rupture, and placental abruption.

Outcomes were included if at least two of the included studies reported the same outcome.

Data Extraction

Four authors (S.H.L., S.H., P.H.S., J.D.A.) utilized a standardized worksheet in Microsoft Excel 2016 and resolved potential disagreements through discussion. Extracted data included trial characteristics (author, publication year, country, and sample size), average age, body mass index (BMI), and mean and standard deviation for outcomes in each treatment group. The following telemedicine characteristics were recorded: purpose (ie, general glycemic improvements, dietary support, exercise, weight loss, diabetes education, and prevention of hypoglycemia or hyperglycemic events), applied technology (eg, glucometers or weights), and setting (ie, primary care, hospital, specialized diabetes clinic, university, community, or cross-sectional). Data were transformed into means and standard deviations using traditional methods.^39,40

Data Synthesis

Analyses were performed in Stata 16 (StataCorp. 2019. Stata Statistical Software: Release 16. College Station, TX: StataCorp LLC). Continuous outcomes were pooled using the mean difference (MD) (or standardized mean difference [SMD] when different scales could not be easily transformed) across studies using fixed- or random-effects meta-analysis depending on the amount of heterogeneity (random-effects model if I² ≥50%). Binary outcomes were pooled based on the log odds ratio. Post hoc subgroup analysis for the primary outcome of all extracted study characteristics was applied, and sensitivity analyses investigated the impact of excluding study outliers on the results. Publication bias was assessed using funnel plot inspection and Egger’s test.

Assessment of Risk of Bias

Two reviewers (J.D.A. and F.W.U.) independently assessed the risk of bias for each included study. The revised Cochrane risk-of-bias tool was used.⁴¹

Assessment of the Certainty of Evidence

Following the handbook for the Grading of Recommendations, Assessment, Development, and Evaluation (GRADE),⁴² the overall judgment of the risk of bias,⁴³ indirectness,⁴⁴ imprecision,⁴⁵ inconsistency,⁴⁶ and risk of publication bias⁴⁷ was used to create a summary of findings table in GRADEpro GDT 2015 (McMaster University, Hamilton, Ontario),^48,49 which was used to rank the quality of evidence.

Results

The selection process for studies is shown in Figure 1. Originally, 16 155 studies were found (11 062 after removal of duplicates). After screening, 1125 full-text articles were evaluated. Eventually, 18 articles were included.

Figure 1.

Selection of studies.

Table 1 presents the study characteristics. One study was published prior to 2000 (6%), two were published in the 2000s (11%), and 15 were published in the 2010s (83%). Six studies were from Europe (33%), five were from North America (28%), four were from Australia/New Zealand (22%), and three were from Asia/Middle East (17%). The study size averaged 88 participants per study and ranged from 19 to 238. The mean age was 32 years (range 26-35 years), the mean baseline BMI was 29.8 kg/m² (range 25.4-34.1 kg/m²), and the average baseline HbA1c% was 6.3% (range 5.0%-8.8%).

Table 1.

Study Characteristics.

Study	Publication year	Country	Duration	Sample size	Mean age (years)	Baseline HbA1c%	Baseline BMI (kg/m²)
Downs et al⁵⁰	2017	United States	Week 32	43	31.5	N/A	28.8
Durnwald et al⁵¹	2016	United States	To birth	101	32.0	N/A	31.5
Fallucca et al⁵²	1996	Italy	Week 37,6	19	28.5	6.3	N/A
Given et al⁵³	2015	Ireland	To birth	50	31.7	N/A	33.1
Homko et al⁵⁴	2007	United States	To birth	63	29.5	N/A	33.0
Homko et al⁵⁵	2012	United States	To birth	80	30.2	N/A	34.1
Mackillop et al⁵⁶	2018	United Kingdom	To birth	203	33.4	5.4	31.4
Miremberg et al⁵⁷	2018	Israel	To birth	126	31.9	5.2	27.1
Perez-Ferre et al⁵⁸	2010	Spain	To birth	100	33.8	5.1	28.5
Rasekaba et al⁵⁹	2018	Australia	To birth	95	32.0	N/A	N/A
Ładyzyński et al⁶⁰	2007	Poland	Six months post partum	32	26.0	8.0	N/A
Al-Olfi et al⁶¹	2018	Saudi Arabia	Six weeks post partum	57	32.5	N/A	30.6
Borgen et al⁶²	2019	Norway	To birth	238	N/A	N/A	N/A
Carolan-Olah and Sayakhot⁶³	2019	Australia	To birth	110	31.7	8.8	29.6
Cheung et al⁶⁴	2019	Australia	To birth	60	34.0	N/A	28.4
Guo et al⁶⁵	2019	Canada	To birth	124	30.9	6.0	25.7
Jelsma et al⁶⁶	2018	Australia	Six months post partum	59	35.4	N/A	30.7
Sung et al⁶⁷	2019	South Korea	To birth	21	33.4	5.4	25.4

Abbreviations: BMI, body mass index; N/A, not available.

Table 2 presents an overview of the telemedicine characteristics. Eight studies (44%) were conducted in a hospital setting, seven were conducted in a specialized diabetes clinic (39%), and three were conducted at a university (17%). Two studies performed daily measurements (11%), three performed biweekly measurements (17%), five performed weekly measurements (28%), and three had a tailored approach (17%). The purpose of the intervention components was glycemic improvement (14 studies, 78%), dietary support (16 studies, 89%), exercise support (12 studies, 67%), education (eight studies, 44%), prevention of glycemic events (hypoglycemia or hyperglycemia, ten studies, 56%), and weight control (five studies, 28%). Seven studies used a web portal (61%), seven studies used a telephone (61%), seven studies applied an application (61%), six studies used Short Message System (SMS) technology (33%), and three studies used e-mail (17%). The applied peripherals were glucometers (14 studies, 78%), pedometers (four studies, 22%), blood pressure monitors (one study, 6%), and weights (one study, 6%).

Table 2.

Telemedicine Characteristics.

Study	Year	Setting	Frequency of contact	Included technology									Purpose of intervention (components)
Study	Year	Setting	Frequency of contact	SMS	Weight	Glucometer	Telephone	Web site	App	Blood pressure monitor	Pedometer	E-mail	Glycemic improvement	Dietary support	Exercise support	Weight control	Prevention of glycemic events	Educational
Downs et al⁵⁰	2017	University	Biweekly	—	—	—	Yes	—	—	—	Yes	—	—	Yes	Yes	Yes	—	Yes
Durnwald et al⁵¹	2016	Hospital	Weekly	—	—	Yes	Yes	—	—	—	Yes	—	Yes	Yes	Yes	Yes	—	Yes
Fallucca et al⁵²	1996	Specialized	Weekly	—	—	Yes	—	—	—	—	—	—	Yes	Yes	Yes	—	Yes	—
Given et al⁵³	2015	Specialized	Weekly	—	Yes	Yes	Yes	Yes	—	Yes	—	Yes	Yes	—	—	Yes	Yes	—
Homko et al⁵⁴	2007	University	N/A	—	—	Yes	—	Yes	—	—	—	—	Yes	Yes	Yes	—	Yes	Yes
Homko et al⁵⁵	2012	University	Weekly	—	—	Yes	Yes	Yes	—	—	—	—	Yes	Yes	Yes	—	Yes	—
Mackillop et al⁵⁶	2018	Hospital	N/A	Yes	—	Yes	—	Yes	Yes	—	—	—	Yes	Yes	—	—	Yes	—
Miremberg et al⁵⁷	2018	Specialized	Daily	—	—	Yes	—	—	Yes	—	—	Yes	Yes	Yes	—	—	—	—
Perez-Ferre et al⁵⁸	2010	Hospital	N/A	Yes	—	Yes	Yes	Yes	—	—	—	—	Yes	Yes	—	—	Yes	—
Rasekaba et al⁵⁹	2018	Specialized	As needed	—	—	Yes	—	Yes	—	—	—	—	Yes	Yes	—	—	Yes	—
Ładyzyński et al⁶⁰	2007	Specialized	As needed	—	—	Yes	Yes	—	—	—	—	—	Yes	Yes	Yes	—	Yes	—
Al-Olfi et al⁶¹	2018	Specialized	Weekly	Yes	—	Yes	—	—	Yes	—	—	—	Yes	Yes	Yes	Yes	Yes	—
Borgen et al⁶²	2019	Specialized	N/A	—	—	Yes	—	—	Yes	—	—	—	Yes	Yes	Yes	—	—	Yes
Carolan-Olah and Sayakhot⁶³	2019	Hospital	N/A	—	—	—	—	Yes	—	—	—	—	—	—	—	—	—	—
Cheung et al⁶⁴	2019	Hospital	As needed	Yes	—	—	—	—	Yes	—	—	—	—	Yes	Yes	—	—	Yes
Guo et al⁶⁵	2019	Hospital	Daily	—	—	Yes	—	—	Yes	—	—	—	Yes	Yes	Yes	Yes	Yes	Yes
Jelsma et al⁶⁶	2018	Hospital	Biweekly	Yes	—	—	Yes	—	—	—	Yes	Yes	—	Yes	Yes	—	—	Yes
Sung et al⁶⁷	2019	Hospital	Biweekly	Yes	—	Yes	—	—	Yes	—	Yes	—	Yes	Yes	Yes	—	—	Yes

Abbreviation: SMS, Short Message System; N/A, not available.

Supplemental Appendix 2 contains the risk of bias assessment for individual studies. Overall, there was a low risk of bias in 45.5% of the included studies, a moderate risk in 36.4% of the included studies, and a high risk in 18.2% of the included studies (Figure 2). Studies with a high risk of bias had a significant amount of missing outcome data at follow-up (18.2% of studies) without appropriately accounting for the potential impact of missing data or deviations from the intended interventions (9.1%).

Figure 2.

Risk of bias assessment.

Effect on Infant Birth Weight

The forest plot in Figure 3 reveals that the MD across studies was 19.34 g (95% confidence interval [CI]: −47.80; 86.47). Although statistically insignificant, the results favored alternatives to telemedicine (the control alternative). Heterogeneity was moderate (I² = 37.39%). Subgroups of studies with an average baseline BMI ≤30 kg/m² achieved significantly higher effects with telemedicine than studies with average BMI values >30 kg/m² (MD: −106.43 g [95% CI: −216.32; 3.47] vs MD: 94.22 g [95% CI: 9.42; 179.02]; see Supplemental Appendix 3). Furthermore, studies with no glycemic prevention component or no Web site component reported better outcomes on birth weight than studies including a glycemic or Web site component (MD: −110 g [95% CI: 226.45; 5.42] vs MD: 84.85 g [95% CI: 2.5; 167.19]; also, Supplemental Appendix 2). Removing the two largest differences in effect (Cheung et al and Given et al) did not alter the conclusion (MD: 22.12 [95% CI: −49.23; 93.48]; Supplemental Appendix 4).

Figure 3.

Results from meta-analysis on infant birth weight.

Certainty of Findings for Infant Birth Weight

The GRADE profile and reasons for downgrading outcomes are presented in Table 3. The pooled estimate of the effect of telemedicine on infant birth weight was assessed as moderate due to serious problems with imprecision. No downgrades were performed due to risk of bias (a low proportion of studies with a high risk of bias, a low risk of bias among studies with high meta-analysis weights, and a nonsignificant relations between the risk of bias score and study weights or treatment effect) and indirectness (all studies fulfilled broad inclusion criteria). Evidence was not downgraded due to inconsistency. Although point estimates varied considerably, the test for equal treatment effects in studies was insignificant and heterogeneity was moderate. Publication bias was not strongly suspected (the funnel plot in Figure 4 is rather symmetrical), and Egger’s test was insignificant (P = .65). Industry sponsorship was undeclared or unknown in four of ten studies, although significant relationships were not observed among the presence of industry sponsorship, study size, or treatment effect. The evidence was not upgraded.

Table 3.

GRADE Profile.

Certainty assessment							No. of patients		Effect		Certainty	Importance
No. of studies	Study design	Risk of bias	Inconsistency	Indirectness	Imprecision	Other considerations	Telemedicine	Usual care	Relative (95% CI)	Absolute (95% CI)	Certainty	Importance
Infant birth weight
10	Randomized trials	Not serious	Not serious	Not serious	Serious^a	None	471	428	—	MD 19.34 g higher (47.8 lower to 86.47 higher)	⨁⨁⨁◯Moderate
Hypoglycemic events (mother)
2	Randomized trials	Serious^b	Not serious	Not serious	Very serious^c	None	26	25	—	SMD 0.34 SD lower(0.9 lower to 0.22 higher)	⨁◯◯◯Very low
HbA1c%
5	Randomized trials	Not serious	Not serious	Not serious	Very serious^a	None	148	140	—	MD 0.14 lower(0.51 lower to 0.23 higher)	⨁⨁◯◯Low
Fasting blood glucose
4	Randomized trials	Serious^d	Not serious	Not serious	Serious^a	None	134	131	—	MD 0.12 mg/dL lower(3.13 lower to 2.89 higher)	⨁⨁◯◯Low
Two-hour glucose tolerance
2	Randomized trials	Not serious	Not serious	Serious^e	Very serious^a	None	164	175	—	MD 9.62 mg/dL higher(1.95 higher to 17.28 higher)	⨁◯◯◯Very low
Weight gain during pregnancy (mother)
2	Randomized trials	Not serious	Not serious	Not serious	Very serious^a	None	150	154	—	SMD 0 SD(0.22 lower to 0.23 higher)	⨁⨁◯◯Low
Pregnancy-induced hypertension/preeclampsia
7	Randomized trials	Not serious	Not serious	Not serious	Very serious^a	None	29/352 (8.2%)	26/350 (7.4%)	Log odds 0.21 (−0.37 to 0.79)	— per 1000 (from — to —)	⨁⨁◯◯Low
Gestational age at delivery
11	Randomized trials	Not serious	Not serious	Not serious	Serious^a	None	504	478	—	MD 0.05 weeks lower(0.24 lower to 0.14 higher)	⨁⨁⨁◯Moderate
Induced labor
3	Randomized trials	Not serious	Not serious	Not serious	Very serious^a	None	104/221 (47.1%)	110/233 (47.2%)	Log odds 0.00(−0.37 to 0.38)	— per 1000(from — to —)	⨁⨁◯◯Low
Cesarean delivery
11	Randomized trials	Not serious	Serious^f	Not serious	Serious^a	None	191/595 (32.1%)	209/575 (36.3%)	Log odds −0.11(−0.50 to 0.27)	— per 1000(from — to —)	⨁⨁◯◯Low
Preterm delivery
3	Randomized trials	Serious^d	Not serious	Not serious	Very serious^a	None	10/117 (8.5%)	12/111 (10.8%)	Log odds −0.38(−1.31 to 0.55)	— per 1000(from — to —)	⨁◯◯◯Very low
Premature membrane rupture
2	Randomized trials	Serious^d	Serious^g	Not serious	Very serious^a	None	3/68 (4.4%)	3/65 (4.6%)	Log odds 0.00(−3.13 to 3.13)	— per 1000(from — to —)	⨁◯◯◯Very low
Placental abruption
2	Randomized trials	Serious^d	Not serious	Not serious	Very serious^a	None	1/81 (1.2%)	1/73 (1.4%)	Log odds −0.14(−2.14 to 1.87)	— per 1000(from — to —)	⨁◯◯◯Very low
Stillbirth
2	Randomized trials	Not serious	Not serious	Not serious	Very serious^a	Publication bias strongly suspected^h	0/81 (0.0%)	1/86 (1.2%)	Log odds −0.58(−3.02 to 1.86)	— per 1000(from — to —)	⨁◯◯◯Very low
Neonatal hypoglycemia
7	Randomized trials	Not serious	Not serious	Not serious	Serious^a	None	51/358 (14.2%)	44/354 (12.4%)	Log odds 0.19(−0.27 to 0.64)	— per 1000(from — to —)	⨁⨁⨁◯Moderate
Jaundice
5	Randomized trials	Not serious	Not serious	Not serious	Very serious^a	None	26/238 (10.9%)	26/240 (10.8%)	Log odds 0.14(−0.47 to 0.76)	— per 1000(from — to —)	⨁⨁◯◯Low
Shoulder dystocia
4	Randomized trials	Not serious	Not serious	Not serious	Very serious^a	None	1/234 (0.4%)	10/236 (4.2%)	Log odds −1.34(−2.61 to −0.08)	— per 1000(from — to —)	⨁⨁◯◯Low
Infant respiratory distress
3	Randomized trials	Serious^d	Not serious	Not serious	Very serious^a	None	8/89 (9.0%)	12/89 (13.5%)	Log odds −0.53(−1.49 to 0.44)	— per 1000(from — to —)	⨁◯◯◯Very low
Neonates large for gestational age
5	Randomized trials	Serious^d	Not serious	Not serious	Very serious^a	None	30/189 (15.9%)	21/180 (11.7%)	Log odds 0.34(−0.26 to 0.94)	— per 1000(from — to —)	⨁◯◯◯Very low
Admissions to neonatal intensive care
7	Randomized trials	Not serious	Not serious	Not serious	Serious^a	None	57/423 (13.5%)	61/403 (15.1%)	Log odds −0.12(−0.53 to 0.28)	— per 1000(from — to —)	⨁⨁⨁◯Moderate

Abbreviations: GRADE, Grading of Recommendations, Assessment, Development, and Evaluation; CI, confidence interval; MD, mean difference; SMD, standardized mean difference; HbA1c, glycated hemoglobin A1c.

Recommendation would be different depending on whether or not the lower or the upper boundary of the confidence interval is the true underlying effect. Furthermore, evidence fails to meet the optimal information criteria.

One of the two studies had a high risk of bias.

Only two small studies, broad confidence intervals, not fulfilling optimal information size.

Studies with high weights in meta-analysis have serious or moderate problems with bias.

Assessed up to three months post partum.

Point estimates vary, test for similar effect size statistically significant, heterogeneity large.

Point estimates vary, heterogeneity large.

Studies assigned dominant weight in meta-analysis in favor of telemedicine is industry sponsored.

Figure 4.

Funnel plot.

Secondary Outcomes

Meta-analyses were conducted on maternal (Supplemental Appendix 5), delivery (Supplemental Appendix 6), and neonatal outcomes (Supplemental Appendix 7). Associated GRADE levels are presented in Table 3. For maternal outcomes, only the difference (SMD: 9.62 mg/dL [95% CI: 1.95; 17.28]) in two-hour glucose tolerance was statistically significant (5% level) and favored the comparator (GRADE level: very low). Nonsignificant differences favoring telemedicine were hypoglycemic events during pregnancy (GRADE: very low), HbA1c% (GRADE: low), and fasting blood glucose (GRADE: low). The nonsignificant risk of pregnancy-induced hypertension or preeclampsia favored the control group (GRADE: low). For neonatal outcomes, the risk of shoulder dystocia significantly favored telemedicine (log odds −1.34 [95% CI: −2.61; −0.08], GRADE level: low). The nonsignificant risks that favored telemedicine were risk of stillbirth (GRADE: very low), infant respiratory distress (GRADE: very low), and admission to neonatal intensive care (GRADE: moderate). Outcomes favoring the control were risk of jaundice (GRADE: low), neonatal hypoglycemia (GRADE: moderate), and receiving LGA (GRADE: very low). No differences in delivery outcomes were statistically significant, although outcomes favoring telemedicine included the risk of placental abruption (GRADE: very low) and preterm and cesarean delivery (GRADE: very low and low, respectively), while the outcome that favored the control was differences in gestational age at delivery (GRADE: moderate).

Discussion

The present systematic review and meta-analysis aimed to evaluate the effectiveness of telemedicine solutions versus any comparator without the use of telemedicine with regard to diabetes-related maternal and fetal outcomes for patients with gestational or pregestational diabetes. Eighteen studies were included in the final sample. With regard to the primary outcome (infant birth weight), the results nonsignificantly favored the control. For maternal outcomes, only the difference in two-hour glucose tolerance was statistically significant (favoring control). For neonatal outcomes, only the risk of shoulder dystocia was statistically significant (favoring telemedicine). The remaining outcomes were all insignificant and varied based on whether they favored telemedicine or control. Moreover, the GRADE value was low or very low for most of the outcomes.

A strength of the present review is that it was based on a very comprehensive literature search, although there are also several limitations to this study. First, the synthesis is based on a small number of studies in general and very few studies for some secondary outcomes. Second, the choice of unadjusted infant birth weight could ideally have been adjusted for both gestational age at delivery and fetal gender simultaneously to nuance the results. However, gestational age at delivery is underreported in the included studies (included in 9 of 18 studies for both groups) and fetal gender is lacking, making it hard to conduct a meaningful meta-regression. Third, despite a comprehensive search, certain articles may not have been identified because only studies published in English or Scandinavian languages were considered.

The findings of the present review are in line with the findings from a previous systematic review and meta-analysis by Rasekaba et al,³⁴ who found similar glycemic control and cesarean section rates when comparing telemedicine for GDM to usual care. Similarly, a systematic review by Dennis and Kingston comparing telephone support to usual care found no significant improvement in the number of preterm births. In contrast, Dennis and Kingston found a statistically significant difference in the number of low-birth weight infants in favor of telephone support.³³ A systematic review and meta-analysis by Ming et al³⁵ found a modest significant improvement in HbA1c% in favor of telemedicine when comparing telemedicine for GDM to usual care. However, no significant differences were found in the mean glucose levels, fasting glucose, two-hour postprandial blood glucose, postprandial hypertension or preeclampsia rates, cesarean sections rates, birth weight, macrosomia, LGA, neonatal intensive care unit admissions, or neonatal hypoglycemia.³⁵ Hence, our findings mirror the findings of Ming et al, who did not find sufficient evidence to support the superiority of telemedicine for GDM over usual care.³⁵

One may assume that the lack of evidence supporting telemedicine for diabetes in pregnancy may partly be attributable to the fact that pregnant women are followed closely by relevant health care professionals throughout pregnancy.⁶⁸ Hence, adding telemedicine may not contribute to improved diabetes care since women may already be closely monitored as part of the usual care scenario.

Telemedicine studies often differ in terms of technology, intervention, organizational setup, and so on and are thus difficult to compare. However, for the present review, the heterogeneity was moderate (I² = 37%), which is fairly low when compared with other reviews within the field of telemedicine.^23,69 The included studies varied greatly in terms of duration, follow-up time, intervention components, and technologies used (Tables 1 and 2), which may explain the variations in the results of the meta-analysis. Another explanation for the variations could be the heterogeneity following diabetes subtypes. Thus, the differences in disease management, pathophysiology, and progress between preexisting diabetes and GDM may also be a possible explanation for the high variance reported in the study results.^11,70,71

The focus of this review was solely on the effectiveness of telemedicine solutions for diabetes-related maternal, fetal, and delivery outcomes. Thus, the impact on personal resources and economics was not considered. However, such perspectives are highly relevant considering the growing diabetic population,^1,12 lack of health care resources,^72,73 and consequences of the COVID-19 pandemic.⁷⁴ With similar maternal, fetal, and delivery outcomes between telemedicine and usual care, the question remains as to whether telemedicine solutions may offer resource savings and cost advantages over usual care. Such a finding would be consistent with speculations in previous studies, where telemedicine has been stated to improve efficiency⁵⁵ and lead to net future cost savings in patients with gestational diabetes.³⁴ Moreover, a Danish study found an estimated net cost savings of approximately $9000 to $11 000 per 200 consultations per year when evaluating a telemedicine service for patients with T1DM and T2DM, which was mainly due to lower transportation costs.⁷⁵ Such perspectives would be highly relevant in future studies.

Conclusions

In conclusion, the present systematic review and meta-analysis did not find sufficient evidence to support telemedicine as an alternative to usual care when considering maternal and fetal outcomes for patients with pregestational or gestational diabetes. However, studies on telemedicine solutions for pregnant women with diabetes as well as economic evaluations in the field are still limited and should be prioritized in future research.

Supplemental Material

sj-docx-1-dst-10.1177_19322968221094626 – Supplemental material for Effectiveness of Telemedicine in Managing Diabetes in Pregnancy: A Systematic Review and Meta-Analysis

Supplemental material, sj-docx-1-dst-10.1177_19322968221094626 for Effectiveness of Telemedicine in Managing Diabetes in Pregnancy: A Systematic Review and Meta-Analysis by Sisse H. Laursen, Lise Boel, Flemming W. Udsen, Pernille H. Secher, Jonas D. Andersen, Peter Vestergaard, Ole K. Hejlesen and Stine Hangaard in Journal of Diabetes Science and Technology

Footnotes

Acknowledgements

The authors would like to thank the Research Librarian Connie Skrubbeltrang, who assisted in the literature search.

Abbreviations

GRADE, The Grading of Recommendations, Assessment, Development, and Evaluation; GDM, gestational diabetes mellitus; SMD, standardized mean difference; T1DM, type 1 diabetes mellitus; T2DM, type 2 diabetes mellitus; PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analyses; CENTRAL, the Cochrane Library Central Register of Controlled Trials; RCT, randomized controlled trial; BMI, body mass index; MD, mean difference; RoB tool, revised Cochrane risk-of-bias tool; HbA1c, glycated hemoglobin A1c.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This is an independent article commissioned and jointly funded by the Steno Diabetes Center North Denmark and Aalborg University.

ORCID iDs

Sisse H. Laursen

Flemming W. Udsen

Jonas D. Andersen

Stine Hangaard

Data Sharing

No additional data are available.

Supplemental Material

Supplemental material for this article is available online.

References

Cho

Shaw

Karuranga

, et al. IDF Diabetes Atlas: global estimates of diabetes prevalence for 2017 and projections for 2045. Diabetes Res Clin Pract. 2018;138:271-281.

Iversen

DeVoe

. Addressing hyperglycemia in pregnancy: the impact on maternal health and beyond. Diabetes Res Clin Pract. 2018;145:15-16.

Bonora

DeFronzo

. Diabetes Complications, Comorbidities and Related Disorders. Cham, Switzerland: Springer; 2018.

Macintosh

MCM

Fleming

Bailey

, et al. Perinatal mortality and congenital anomalies in babies of women with type 1 or type 2 diabetes in England, Wales, and Northern Ireland: population based study. Br Med J. 2006;333:177-180.

Wahabi

. Maternal and perinatal outcomes of pregnancies complicated with pregestational and gestational diabetes mellitus in Saudi Arabia. J Diabetes Metab. 2014;5:7.

Buhary

Almohareb

Aljohani

, et al. Glycemic control and pregnancy outcomes in patients with diabetes in pregnancy: a retrospective study. Indian J Endochrinology Metab. 2016;20:481-490.

Kong

Nilsson

IAK

Gissler

Lavebratt

. Associations of maternal diabetes and body mass index with offspring birth weight and prematurity. JAMA Pediatr. 2019;173:371-378.

Kim

Sharma

Sappenfield

Wilson

Salihu

. Association of maternal body mass index, excessive weight gain, and gestational diabetes mellitus with large-for-gestational-age births. Obs Gynecol. 2015;123:737-744.

Yanga

Wang

, et al. The association of gestational diabetes mellitus with fetal birth weight. J Diabetes Complications. 2018;32:635-642.

10.

Beta

Khan

Fiolna

Khalil

Ramadan

Akolekar

. Maternal and neonatal complications of fetal macrosomia : cohort study. Ultrasound Obs Gynecol. 2019;54:319-325.

11.

Ringholm

Damm

Mathiesen

. Improving pregnancy outcomes in women with diabetes mellitus: modern management. Nat Rev Endocrinol. 2019;15:406-416.

12.

Zhu

Zhang

. Prevalence of gestational diabetes and risk of progression to type 2 diabetes: a global perspective. Curr Diab Rep. 2016;16(1):1-11.

13.

American Diabetes Association. 14. Management of diabetes in pregnancy: standards of medical care in diabetes—2019. Diabetes Care. 2019;42(suppl 1):165-172.

14.

Ballas

Moore

Ramos

. Management of diabetes in pregnancy. Curr Diab Rep. 2012;12:33-42.

15.

Tion

ADA

. 13. Management of diabetes in pregnancy: standards of medical care in diabetesd—2018. Diabetes Care. 2018;41:137-143.

16.

Daponte

Guidozzi

Moisuc

Marineanu

. Management of diabetic pregnant patients in a tertiary center in the developing world. Int J Gynecol Obstet. 1999;64:141-146.

17.

Mirghani

Saeed

. A simplified management of diabetic pregnant women. Saudi Med J. 2000;21:335-339.

18.

Ozumba

Obi

Oli

. Diabetes mellitus in pregnancy in an African population. Int J Gynecol Obstet. 2004;84:114-119.

19.

Ahkter

Qureshi

Rahim

, et al. Diabetes in pregnancy in Pakistani women: prevalence and complications in an indigenous south Asian community. Diabet Med. 1996;13:189-191.

20.

Mukona

Munjanja

Zvinavashe

Stray-Pederson

. Barriers of adherence and possible solutions to nonadherence to antidiabetic therapy in women with diabetes in pregnancy: patients’ perspective. J Diabetes Res. 2017;1-10.

21.

Martis

Brown

McAra-Couper

Crowther

. Enablers and barriers for women with gestational diabetes mellitus to achieve optimal glycaemic control—a qualitative study using the theoretical domains framework. BMC Pregnancy Childbirth. 2018;18:1-22.

22.

Kaufman

Salahi

. Using digital health technology to prevent and treat diabetes. Diabetes Technol Ther. 2017;19(suppl 1):59-73.

23.

Faruque

Wiebe

Ehteshami-Afshar

, et al. Effect of telemedicine on glycated hemoglobin in diabetes: a systematic review and meta-analysis of randomized trials. CMAJ. 2017;189(9):341-364.

24.

Aberer

Hochfellner

Mader

. Application of telemedicine in diabetes care: the time is now. Diabetes Ther. 2021;12:629-639.

25.

Crico

Renzi

Graf

, et al. mHealth and telemedicine apps : in search of a common regulation. Ecancermedicalscience. 2018;12:853.

26.

Mahar

Rosencrance

Rasmussen

. Telemedicine: past, present, and future. Cleve Clin J Med. 2018;85(12):938-942.

27.

Shah

Tariq

Phillips

Davison

Hoare

Hasan

. Health care for all: effective, community supported, healthcare with innovative use of telemedicine technology. J Pharm Policy Pr. 2018;11:10-12.

28.

Hanlon

Daines

Campbell

McKinstry

Weller

Pinnock

. Telehealth interventions to support self-management of long-term conditions: a systematic metareview of diabetes, heart failure, asthma, chronic obstructive pulmonary disease, and cancer. J Med Internet Res. 2017;19(5):e172.

29.

Chung

JWY

. Telehealth for diabetes self-management in primary healthcare: a systematic review and meta-analysis. J Telemed Telecare. 2018;24:356-364.

30.

Borries

Dunbar

Bhukhen

, et al. The impact of telemedicine on patient self-management processes and clinical outcomes for patients with Types I or II Diabetes Mellitus in the United States: a scoping review. Diabetes Metab Syndr Clin Res Rev. 2019;13:1353-1357.

31.

Von Storch

Graaf

Wunderlich

Rietz

Polidori

Woopen

. Telemedicine-assisted self-management program for type 2 diabetes patients. Diabetes Technol Ther. 2019;21:514-521.

32.

Sushko

Menezes

Strachan

Butt

Sherifali

. Self-management education among women with pre-existing diabetes in pregnancy: a scoping review. Int J Nurs Stud. 2021;117:103883.

33.

Dennis

Kingston

. A systematic review of telephone support for women during pregnancy and the early postpartum period. J Obstet Gynecol Neonatal Nurs. 2008;37:301-314.

34.

Rasekaba

Furler

Blackberry

Tacey

Gray

Lim

. Telemedicine interventions for gestational diabetes mellitus: a systematic review and meta-analysis. Diabetes Res Clin Pract. 2015;110:1-9.

35.

Ming

Mackillop

Farmer

, et al. Telemedicine technologies for diabetes in pregnancy: a systematic review and meta-analysis. J Med Internet Res. 2016;18:1-12.

36.

Xie

Dai

Qin

Yang

. Effectiveness of telemedicine for pregnant women with gestational diabetes mellitus: an updated meta-analysis of 32 randomized controlled trials with trial sequential analysis. BMC Pregnancy Childbirth. 2020;20:1-14.

37.

Moher

Liberati

Tetzlaff

Altman

, The Prisma Group. Preferred Reporting Items for Systematic Reviews and Meta-Analyses: the PRISMA Statement. PLoS Med. 2009;6(7):e1000097.

38.

Laursen

Hangaard

Udsen

Vestergaard

Hejlesen

. Effectiveness of telemedicine solutions for the management of patients with diabetes: protocol for a systematic review and meta-analysis. JMIR Res Protoc. 2020;9:1-7.

39.

Wan

Wang

Liu

Tong

. Estimating the sample mean and standard deviation from the sample size, median, range and/or interquartile range. BMC Med Res Methodol. 2014;14:1-13.

40.

Hozo

Djulbegovic

Hozo

. Estimating the mean and variance from the median, range, and the size of a sample. BMC Med Res Methodol. 2005;5:1-10.

41.

Sterne

JAC

Savović

Page

, et al. RoB 2: a revised tool for assessing risk of bias in randomised trials. BMJ. 2019;366:1-8.

42.

Schünemann

Brozek

Guyatt

Oxman

. Handbook for grading the quality of evidence and the strength of recommendations using the GRADE approach. 2013.

43.

Guyatt

Oxman

Vist

, et al. GRADE guidelines: 4. Rating the quality of evidence—study limitations (risk of bias). J Clin Epidemiol. 2011;64:407-415.

44.

Guyatt

Oxman

Kunz

, et al. GRADE guidelines: 8. Rating the quality of evidence—indirectness. J Clin Epidemiol. 2011;64:1303-1310.

45.

Guyatt

Oxman

Kunz

, et al. GRADE guidelines 6. Rating the quality of evidence—imprecision. J Clin Epidemiol. 2011;64:1283-1293.

46.

Guyatt

Oxman

Kunz

, et al. GRADE guidelines: 7. Rating the quality of evidence—inconsistency. J Clin Epidemiol. 2011;64:1294-1302.

47.

Guyatt

Oxman

Montori

, et al. GRADE guidelines: 5. Rating the quality of evidence—publication bias. J Clin Epidemiol. 2011;64:1277-1282.

48.

Guyatt

Thorlund

Oxman

, et al. GRADE guidelines: 13. Preparing summary of findings tables and evidence profiles—continuous outcomes. J Clin Epidemiol. 2013;66:173-183.

49.

Guyatt

Oxman

Sultan

, et al. GRADE guidelines: 11. Making an overall rating of confidence in effect estimates for a single outcome and for all outcomes. J Clin Epidemiol. 2013;66:151-157.

50.

Downs

DiNallo

Birch

Paul

Ulbrecht

. Randomized face-to-face vs. home exercise interventions in pregnant women with gestational diabetes. Psychol Sport Exerc. 2017;30:73-81.

51.

Durnwald

Kallan

Allison

, et al. A randomized clinical trial of an intensive behavior education program in gestational diabetes. Am J Perinatol. 2016;33:1145-1151.

52.

Fallucca

Di Biase

Sabbatini

Borrello

Sciullo

Napoli

. Telemedicine in the treatment of diabetic pregnancy. Ann Ist Super Sanita. 1997;33:347-351.

53.

Given

Bunting

O’Kane

Dunne

Coates

. Tele-Mum: a feasibility study for a randomized controlled trial exploring the potential for telemedicine in the diabetes care of those with gestational diabetes. Diabetes Technol Ther. 2015;17:880-888.

54.

Homko

Santamore

Bower

, et al. Use of an internet-based telemedicine system to manage underserved women with gestational diabetes mellitus. Diabetes Technol Ther. 2007;9:297-306.

55.

Homko

Deeb

Rohrbacher

, et al. Impact of a telemedicine system with automated reminders on outcomes in women with gestational diabetes mellitus. Diabetes Technol Ther. 2012;14:624-629.

56.

Mackillop

Hirst

Bartlett

, et al. Comparing the efficacy of a mobile phone-based blood glucose management system with standard clinic care in women with gestational diabetes: randomized controlled trial. JMIR Mhealth Uhealth. 2018;6:e71.

57.

Miremberg

Ben-Ari

Betzer

, et al. The impact of a daily smartphone-based feedback system among women with gestational diabetes on compliance, glycemic control, satisfaction, and pregnancy outcome: a randomized controlled trial. Am J Obstet Gynecol. 2018;218:453e1-453e7.

58.

Perez-Ferre

Galindo

Fernandez

, et al. The outcomes of gestational diabetes mellitus after a telecare approach are not inferior to traditional outpatient clinic visits. Int J Endocrinol. 2010;2010:386941.

59.

Rasekaba

Furler

Young

, et al. Using technology to support care in gestational diabetes mellitus: quantitative outcomes of an exploratory randomised control trial of adjunct telemedicine for gestational diabetes mellitus (TeleGDM). Diabetes Res Clin Pract. 2018;142:276-285.

60.

Ładyzyński

Wójcicki

. Home telecare during intensive insulin treatment—Metabolic control does not improve as much as expected. J Telemed Telecare. 2007;13:44-47.

61.

Al-Olfi

Mosli

Ghramri

Ghazali

. Management of postprandial hyperglycaemia and weight gain in women with gestational diabetes mellitus using a novel telemonitoring system. J Int Med Res. 2019;47(2):754-764.

62.

Borgen

Småstuen

Jacobsen

, et al. Effect of the Pregnant+ smartphone application in women with gestational diabetes mellitus: a randomised controlled trial in Norway. BMJ Open. 2019;9:1-12.

63.

Carolan-Olah

Sayakhot

. A randomized controlled trial of a web-based education intervention for women with gestational diabetes mellitus. Midwifery. 2019;68:39-47.

64.

Cheung

Blumenthal

Smith

, et al. A pilot randomised controlled trial of a text messaging intervention with customisation using linked data from wireless wearable activity monitors to improve risk factors following gestational diabetes. Nutrients. 2019;11(3):590.

65.

Guo

Zhang

Zhou

Chen

. Evaluating the effects of mobile health intervention on weight management, glycemic control and pregnancy outcomes in patients with gestational diabetes mellitus. J Endocrinol Invest. 2019;42(6):709-714.

66.

Jelsma

van Poppel

Smith

Cinnadaio

Bauman

. Changing psychosocial determinants of physical activity and diet in women with a history of gestational diabetes mellitus. Diabetes/ Metab Res Rev. 2018;34(1):e2942-1-e2942-9.

67.

Sung

Lee

Min

Park

. Peripartum management of gestational diabetes using a digital health care service: a pilot, randomized controlled study. Clin Ther. 2019;41(11):2426-2434.

68.

American Diabetes Association. 14. Management of diabetes in pregnancy : standards of medical care in diabetes—2021. Diabetes Care. 2021;44:200-210.

69.

Tchero

Kangambega

Briatte

Brunet-Houdard

Retali

Rusch

. Clinical effectiveness of telemedicine in diabetes mellitus: a meta-analysis of 42 randomized controlled trials. Telemed e-Health. 2019;25:569-583.

70.

Alexopoulos

Blair

Peters

, et al. Management of preexisting diabetes in pregnancy: a Review. JAMA. 2019;321:1811-1819.

71.

Zhang

Zhou

Zhong

Wang

Ding

. Current guidelines on the management of gestational diabetes mellitus : a content analysis and appraisal. 2019;2:1-15.

72.

Blanchfield

. Job stressors and burnout among immigrant nurses caring for the elderly in the United States of America: the role of working environment condition. J of Human Resource & Leadership. 2021;5(2):1-13.

73.

Halter

Boiko

Pelone

, et al. The determinants and consequences of adult nursing staff turnover: a systematic review of systematic reviews. BMC Health Serv Res. 2017;17:824.

74.

Manteghinejad

Javanmard

. Challenges and opportunities of digital health in a post—COVID19 world. J Res Med Sci. 2021;26(1):1-6.

75.

Levin

Madsen

Petersen

Wanscher

Hangaard

. Telemedicine diabetes consultations are cost-effective, and effects on essential diabetes treatment parameters are similar to conventional treatment: 7-year results from the Svendborg telemedicine diabetes project. J Diabetes Sci Technol. 2013;7:587-595.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

4.83 MB