Systematic Reviews to Inform Practice, July/August 2023

Efforts to effectively address and improve perinatal mental health have been an ongoing challenge, and that challenge has only grown since the start of the COVID-19 pandemic. Social distancing, fear, and general uncertainty exacerbated the challenges felt by those already attending to a variety of health care needs, including pregnancy.1 As the nation witnessed rising rates of depression, anxiety, and other mental health diagnoses during the pandemic, health care providers have had to reassess priorities regarding screening for and guiding patients in managing mental health symptoms. Loneliness, a subjective experience of isolation different from the concept of being physically alone, is a key component of mental health that has received increasing attention.2 US Surgeon General Dr. Vivek Murthy recently released an advisory recognizing loneliness as a public health crisis and highlighting the urgent need for a national effort to increase social connections.3 The advisory reported that loneliness causes a 60% increased risk of premature death by exacerbating a variety of physical and mental health conditions. Dr. Murthy called for loneliness to receive the same attention as other public health issues such as tobacco and substance abuse, obesity, and heart health.3 The importance of addressing loneliness during and after pregnancy is no different. Applying an intersectional lens to account for the differences in loneliness experienced in various sociocultural contexts, Adlington et al conducted a qualitative metasynthesis aimed at describing experiences of loneliness during the perinatal period as well as factors that improve and worsen loneliness.4 The authors used 4 databases to compile 27 qualitative studies, conducted from 1992 to 2021, that provided perinatal data from 537 women for up to 2 years after birth that were directly relevant to the study aims. All data were from countries defined as high or upper-middle income: United Kingdom (10), United States (6), Canada (5), Australia (3), Sweden (1), Norway (1), and China (1). Of note, 11 of the studies specifically examined the experiences of women from marginalized populations, which included immigrants (4), ethnic minorities (6), and teenage mothers (1). The majority of the studies provided data for the postnatal period (78%) or the entire perinatal period (15%); only 7% provided data restricted to the antenatal period. The authors reported 3 metathemes to describe the experiences of loneliness and perinatal depression across the samples of participants: (1) self-isolation and hiding depressive symptoms due to societal stigma, (2) a sense of emotional disconnection associated with perinatal depression, and (3) a mismatch between expected and actual support associated with loneliness. Participants cited fear of judgment about parenting and made efforts to hide their symptoms from others, delaying diagnosis and treatment. This fear included concerns about possible involvement of social services and losing custody of their children. They also reported feeling disconnected from other mothers, their infants, and themselves as they adjusted to changes in their identity caused by motherhood. Although isolation and a lack of practical support was commonly reported in the perinatal period, some participants reported difficulty connecting with support even when it was available because of their depression. Metathemes for factors that improve loneliness included validation from trusted health care professionals, peer support from other women with perinatal depression, and practical and emotional support from family. Lack of professional support, groups, and facilities in addition to conflict and separation from partner, family, and community were reported as the 2 metathemes reflecting factors that worsen perinatal loneliness. Loneliness was also worsened by sociocultural factors related to membership in a marginalized group, where stigma can contribute to preexisting disparities in perinatal depression. The authors described a rigorous protocol for this metasynthesis and abided by qualitative research standards for study selection and assessing the quality of research reports. They acknowledged that lack of social support is often a proxy for loneliness in some cases and, although social support is closely related to and overlaps with loneliness, they are different concepts that have their own unique attributes. Nevertheless, this metasynthesis demonstrated that loneliness is a common theme in the experiences of those struggling with perinatal depression. The implications of these findings targeting loneliness as a key factor in mental health are far-reaching and are consistent with previous studies on loneliness.5 Other studies demonstrating the impacts of parental loneliness on parenting and child mental health highlight the need for early assessment and intervention.2,6 Health care providers who screen for perinatal depression should develop more effective ways of assessing for loneliness as a contributing factor and avoid equating patients’ reports of adequate physical social support with a low risk for loneliness. The UCLA Loneliness Scale, Version 3, has been validated for use during the perinatal period and is an effective clinical tool for assessing social and emotional components of loneliness.2 Assessments that focus on patients’ perceptions of loneliness separately from the presence of a supportive social network would assist health care providers in offering the validation needed to help alleviate loneliness and connecting patients with resources to increase their social connectedness.7 Hormonal contraceptives are the most commonly used forms of contraception in the United States and the United Kingdom.1-3 Historically, most hormonal contraception was delivered as combined estrogen and progestin oral contraceptive pills, but in recent years, the availability and use of a variety of long-acting reversible progestin-only contraceptive methods have increased.4,5 In the United Kingdom, the use of progestin-only oral contraceptives has also increased.3 In the United States, data are tracked by the Centers for Disease Control and Prevention based on the mode of delivery—oral, implant, intrauterine device (IUD)—rather than the form of hormones delivered, so it is more difficult to identify trends in progestin-only oral contraceptives.1 A substantial body of evidence has documented an association between combined hormonal contraception use and increased breast cancer risk,6-8 although less evidence exists about the influence of progestin-only methods. Therefore, Fitzpatrick et al decided to look at the association between progestin-only forms of contraception and breast cancer rates.9 The authors used a novel methodology that combined a nested case-control study and a meta-analysis. By using these 2 methods, they were able to review a large sample of patients in the United Kingdom who developed premenopausal invasive breast cancer and then identify if they had been exposed to hormonal contraception prior to diagnosis. Whereas the nested case-control arm of the study explored the association between invasive breast cancer and a variety of types of contraception, the meta-analysis solely examined the relationship between progestin-only forms of contraception and invasive breast cancer. The nested case-control arm of the study used the Clinical Practice Research Datalink (CRPD), which is the centralized, anonymized database for all health care records associated with a general practitioner in the National Health Service in the United Kingdom. The CRPD contains the health information of approximately 11 million people, with about 7% of the United Kingdom's population actively enrolled in the system. The researchers were able to use this database to identify patients who developed invasive breast cancer before age 50 between 1996 to 2017. The researchers then matched the patient with breast cancer to 2 patients without breast cancer who served as controls. The patients were matched by date of diagnosis, year of birth, general practitioner, and duration of time that health care records were available. The health care records were then analyzed to identify if any of the participants had been prescribed contraception and how recently they had received that prescription. Types of contraception that were evaluated included oral combined contraceptive, oral progestin-only contraceptive, injectable progestin, progestin implant, progestin IUD, combined contraceptive ring, combined contraceptive patch, and copper IUD. In total, this arm of the study reviewed the records of 9498 participants with diagnosed invasive breast cancer and 18,171 participants in the control group. When comparing patients with no hormonal contraception exposure and patients with at least one prescription for hormonal contraception, there was a statistically significantly increased odds of breast cancer (adjusted odds ratio [OR], 1.25; 95% CI, 1.18-1.33, P < .001). The OR for breast cancer was increased for all types of hormonal contraceptive, including progestin-only forms of contraception: oral combined (OR, 1.23; 95% CI, 1.14-1.32, P < .001), oral progestin-only (OR, 1.26; 95% CI, 1.16-1.37, P < .001), injectable progestin (OR, 1.25; 95% CI, 1.07-1.45, P = .004), and progestin IUD (OR, 1.32; 95% CI, 1.17-1.49, P < .001). There was a small number of participants who used implanted progestins, leading to an OR of breast cancer that was not statistically significant (OR, 1.22; 95% CI, 0.93-1.59, P = .2). In the patients with invasive breast cancer, the mean time from exposure to diagnosis was 3.1 years. The increased risk of breast cancer declined over time from last use of hormonal contraceptives but remained similar among the groups of participants at each specified time, regardless of the methods of hormonal contraceptives used. The meta-analysis arm of the study was restricted to progestin-only contraceptives, as the authors felt that there was already substantial evidence of the role of combined hormonal contraceptives in the risk of breast cancer in the literature. They identified one previous meta-analysis and 11 other studies for inclusion, for a total of an additional 2454 participants who were diagnosed with breast cancer and over 46,000 participants who had never used hormonal contraception included in the meta-analysis. The included studies were conducted in the United States, Norway, Denmark, Sweden, South Africa, Denmark, and Germany. Six were retrospective studies, and 5 were prospective studies. The previously published meta-analysis included studies published prior to 1996, which was well before the other included studies, so cases could not have been double-counted. When combining these participants with those from the nested case-control arm, the authors found a statistically significant increased risk of breast cancer with all 4 forms of progestin-only contraception: oral progestin-only (relative risk [RR], 1.29; 95% CI, 1.21- 1.37), injected progestin (RR, 1.18; 95% CI, 1.07-1.30), implanted progestin (RR, 1.29; 95% CI, 1.08- 1.51), and progestin-releasing IUDs (RR, 1.21; 95% CI, 1.14-1.28). When the results of the nested case-control arm were combined with the meta-analysis to examine the influence of time from last use, the relative risk of breast cancer was found to diminish over time: current users had the highest relative risk (RR, 1.27; 95% CI, 1.21-1.33); followed by users in the past 1 to 4 years (RR, 1.16; 95% CI, 1.11-1.22) and users in the last 5 to 9 years (RR, 1.08; 95% CI, 1.04-1.13). There was no excess risk of breast cancer 10 years or more after discontinuing oral combined contraceptives or progestin-only contraceptives. Fitzgerald et al also calculated the absolute risk of breast cancer in people who used oral combined or progestin-only contraceptives. Given the complexity of the study and the numerous findings presented, the absolute risk may be the most useful to clinicians. The 15-year excess absolute risk of breast cancer in individuals aged 16 to 20 years is approximately 8 per 100,000 users. In individuals aged 25 to 29, the excess absolute risk is approximately 61 per 100,000 hormonal contraceptive users, and in those aged 35 to 39, the excess absolute risk is approximately 265 per 100,000 hormonal contraceptive users. For some patients, this increased risk may be a cause for concern, and in others, the benefits of using hormonal contraceptives may outweigh the risks. Thoughtful and careful counseling about the increased risk of breast cancer associated with the use of hormonal contraception, particularly for patients in their 30s and older, should be presented to all patients considering these methods of contraception. Sleep disordered breathing is a term encompassing a spectrum from loud snoring to obstructive sleep apnea (OSA). Because of the physiologic changes of pregnancy, the percentage of people with sleep disordered breathing increases with gestational age.1 Snoring among pregnant people is estimated at 7% to 11%, increasing to 16% to 35% by the third trimester.2-4 The prevalence of OSA is estimated around 8% in the first trimester and 20% in the third trimester with higher rates for those for pregnant people with obesity, the largest risk factor of OSA.5,6 Age is another risk factor for OSA. OSA is associated with inflammatory changes, which are thought to be the genesis for increased risk of hypertension, diabetes, and other chronic diseases associated with OSA in middle and older ages. These same inflammatory changes and inflammatory biomarkers associated with OSA are also those associated with gestational hypertension and preeclampsia, gestational diabetes, and potentially inflammatory pathways leading to premature birth.1 Given rising rates of obesity globally and the differences in rates between loud snoring in pregnancy and diagnosed OSA, OSA in pregnancy is a largely underrecognized problem. Although the first steps for treating OSA among pregnant people with obesity are the lifestyle modifications of limiting gestational weight gain, diet, and exercise, these do not appear to be sufficient to improve pregnancy outcomes.7 Continuous positive airway pressure (CPAP) is recognized as the best nonsurgical treatment for OSA in the general population. This systematic review aimed to synthesize research on the use of CPAP to treat OSA in the pregnant population. The authors searched 4 databases, including ClinicalTrials.gov, for any type of study that examined the use of CPAP to treat OSA in pregnancy. Studies were limited to the English language, with risk of bias assessed using Cochrane criteria appropriate to study type and the overall study quality assessed using the GRADE (Grading of Recommendations, Assessment, Development, and Evaluations) criteria. The outcomes examined were effects of OSA on blood pressure, preeclampsia, birth weight, and prematurity as well as adherence to use. Of the 7 studies that met the study criteria, 3 were randomized controlled trials (RCTs), one was quasiexperimental, 2 were overnight observational studies among those with preeclampsia, and one was a prospective cohort study with no control group. Studies were conducted in 4 different countries: Brazil, Thailand, the United States, and Serbia had one each, and 3 were from Australia. Samples sizes were relatively small, ranging from 10 to 110. With the exception of one RCT, most studies were at high risk of bias for the majority of criteria. There were either not enough studies assessing an outcome or the heterogeneity among studies was too great to conduct a meta-analysis, so results were simply synthesized. Among the 4 studies examining blood pressure, 2 assessed the effect of CPAP on blood pressure during a single night's treatment in the third trimester, and 2 assessed treatment with CPAP in early pregnancy among those with hypertension. The single-night studies showed that CPAP given to people with preeclampsia lowered mean arterial pressure 3 mmHg or 10 mmHg for both systolic and diastolic pressure.8,9 The studies in which CPAP was started in early pregnancy showed that those using CPAP had lower blood pressure throughout gestation and needed no or lower antihypertensive medication dose increase compared with those not using CPAP. Three studies examined the effect of CPAP on the incidence of preeclampsia, but the participants’ characteristics and the timing and length of interventions varied. An RCT randomized 36 participants with gestational diabetes mellitus (GDM) to either immediate CPAP or a 2-week waitlist control in late second or third trimester.10 There was no difference in preeclampsia rates, although insulin secretion improved with CPAP adherence. In a 3-arm Serbian trial with 110 participants, CPAP was given for 4 weeks starting at 24 to 28 weeks’ gestation.11 Those with OSA treated with CPAP had significantly lower levels of preeclampsia than those not treated (8% vs 24%) or a 67% decreased risk for preeclampsia (RR, 0.33; 95% CI, 0.11-0.97). The prospective study had no comparison group to assess preeclampsia incidence.12 Neonatal outcomes of birth weight and premature birth for the most part showed little difference between those treated with CPAP and those who were not. The exception was for the Serbian trial, which found significantly higher birth weight of over 200 g with CPAP use and lower rates of preterm labor, although the criterion of preterm labor was not defined and actual participant numbers were missing from the published study.11 Adherence to the use of CPAP is an important issue. Among the 4 studies measuring adherence, good adherence was defined as using CPAP for 4 hours per night for 70% of nights. In 3 out of 4 studies, CPAP adherence was good, and in the RCT among those with GDM, adherence was good among those with more symptoms. Overall, the GRADE rating of evidence on this topic was low or very low for all outcomes due to small sample sizes, indirectness of evidence, and potential publication bias. The authors note that there are 2 large RCTs that have closed or will be closing soon that should soon be reporting results and will provide more robust evidence related to the efficacy of CPAP on improving pregnancy outcomes. The implications for clinical practice—that is, who to screen, refer for OSA assessment, and treat with CPAP—are uncertain given the evidence. Most of the pregnant persons who were tested for OSA or were given CPAP without specific OSA testing either had chronic hypertension or were at high risk for preeclampsia or GDM with a focus on those with obesity. Only one national guideline mentions screening pregnant people for OSA: the American College of Obstetricians and Gynecologists practice bulletin on obesity in pregnancy.13 It is reasonable that midwives might ask about the presence of loud snoring early in pregnancy and again in the late second trimester, especially among those with obesity. Unfortunately, common OSA screening tools like the Epworth Sleepiness Scale and STOP-BANG perform poorly in pregnancy.14 The Facco 4-variable model consists of summing [(age + body mass index) + 15 if snoring (≥3 times/week) + 15 (if chronic hypertension)].15 A cutoff score of 75 or greater on the Facco tool has some of the best psychometrics among different screening tools,14,15 and a high score can be used for referral to a specialist for further OSA testing.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.