Abstract
Objective
The objective of this study was to describe adherence to behavioral and pelvic floor muscle training in women undergoing vaginal reconstructive surgery for organ prolapse and to examine whether adherence was associated with 24-month outcomes.
Methods
Participants were women ≥18 years of age, with vaginal bulge and stress urinary incontinence symptoms, planning to undergo vaginal reconstructive surgery for stages 2 to 4 vaginal or uterine prolapse. They were randomized to either sacrospinous ligament fixation or uterosacral ligament suspension and to perioperative behavioral and pelvic floor muscle training or usual care. Measurements included anatomic failure, pelvic floor muscle strength, participant-reported symptoms, and perceived improvement. Analyses compared women with lower versus higher adherence.
Results
Forty-eight percent of women performed pelvic floor muscle exercises (PFMEs) daily at the 4- to 6-week visit. Only 33% performed the prescribed number of muscle contractions. At 8 weeks, 37% performed PFMEs daily, and 28% performed the prescribed number of contractions. No significant relationships were found between adherence and 24-month outcomes.
Conclusion
Adherence to a behavioral intervention was low following vaginal reconstructive surgery for pelvic organ prolapse. The degree of adherence to perioperative training did not appear to influence 24-month outcomes in women undergoing vaginal prolapse surgery.
Impact
This study contributes to the understanding of participant adherence to PFMEs and the impact that participant adherence has on outcomes at 2, 4 to 6, 8, and 12 weeks and 24 months postoperatively. It is important to educate women to follow up with their therapist or physician to report new or unresolved pelvic symptoms.
Keywords: Behavior Therapy, Pelvic Floor Disorders, Women’s Health
Introduction
Pelvic floor disorders, including pelvic organ prolapse (POP) and urinary incontinence (UI), affect approximately one-fourth of women in the United States.1
Pelvic floor muscle training (PFMT) is a nonsurgical treatment for both POP and the most common forms of UI (stress, urgency, and mixed).2 A Cochrane review update (2018) reported that women with stress urinary incontinence (SUI) who received PFMT were 8 times more likely to report cure compared with inactive or no treatment.3 In addition, PFMT was found effective in preventing and treating 1 to 3 stage POP. These women obtained a significant decrease in POP symptom score at 12 months compared with those who did not receive training.4,5
Data on the role of perioperative PFMT in women receiving POP reconstructive surgery are inconsistent. Li et al. showed that women who received PFMT gained greater improvement in prolapse symptom score and POP stages compared with controls. They also had greater improvement in muscle strength and endurance.6 However, investigators in the Eunice Kennedy Shriver National Institutes of Child Health and Human Development (NICHD) funded Pelvic Floor Disorders Network (PFDN) found that women randomized to receive perioperative behavioral with pelvic floor muscle training (BPMT) prior to vaginal reconstructive surgery for apical prolapse did not have greater improvements compared with controls with regards to UI symptoms, prolapse symptoms, or anatomic success at either the 2- or 5-year outcomes.7,8
It is widely accepted that the effectiveness of any behavioral therapy or exercise program depends on a “patient’s behavior matching agreed recommendations/instructions from the prescriber.”9 The World Health Organization defines adherence as “the degree to which the person’s behavior corresponds with the agreed upon recommendations from a health care provider.”10 Although high adherence appears to be a key factor for success, long-term adherence to PFMT (1–15 years) is highly variable, ranging from 10 to 70%.11 Participant-related factors, such as forgetting or boredom, appear to play a significant role in adherence to behavioral treatment programs.9,12 Less is known about PFMT adherence, determinants, strategies, barriers, or impact on outcomes in the context of other treatments, such as pelvic reconstructive surgery.
The primary aim of this planned supplemental analysis was to describe adherence to a perioperative BPMT intervention in the “Operations and Pelvic Muscle Training in the Management of Apical Support” (OPTIMAL) surgical trial.13 A secondary aim was to assess whether adherence was associated with better anatomic and participant-reported outcomes.
Methods
Design and Participants
This was a planned supplemental analysis of the PFDN OPTIMAL trial, which was conducted at 9 sites between January 2008 and May 2013.7 All sites received Institutional Review Board approval and obtained informed consent for study participation. The OPTIMAL study was funded by NICHD and the NIH Office of Research on Women’s Health.
Participants were women ≥18 years of age reporting vaginal bulge and stress UI symptoms and planning to receive vaginal reconstructive surgery for stage 2 to 4 vaginal or uterine prolapse. They were randomized in a 2 × 2 factorial design to sacrospinous ligament fixation or uterosacral ligament suspension and to perioperative BPMT or usual care. The methods have been previously published.13 This analysis included only the 186 women randomized to the BPMT intervention.
Interventions
The perioperative BPMT intervention was delivered by physical therapists, registered nurses, and certified registered nurse practitioners at each site. Because experience varied between practitioners, centralized training was designed to teach interventionists the basic components of the BPMT. The interventionist training included teaching behavioral principles and strategies and specific techniques for optimizing adherence. These included individualizing treatment to the participant, understanding the gradual and progressive nature of the learning curve, using positive reinforcement, and shaping, inoculation against failure, facilitating realistic expectations, guidance in self-management, such as integrating new behaviors into daily life, techniques for remembering to do exercises and use strategies, and setting contingencies for completing exercises. Each interventionist completed a performance-based certification to ensure they could perform the basic skills involved in the protocol.
The intervention included 1 preoperative visit at 2 to 4 weeks prior to surgery and 4 postoperative visits at 2, 4 to 6, 8, and 12 weeks. At the preoperative visit, participants’ ability to correctly perform a PFM contraction was assessed by the interventionist using the Brink scale (The Brink scale has been shown to have good test–retest, r = .51–.65, and good interrater reliability, r = .52–.74).14 Vaginal palpation was used to teach pelvic floor muscle contraction and relaxation. The participants were taught how to control pelvic floor muscles selectively, while keeping other muscles relaxed. The interventionist determined the participant’s starting muscle contraction duration for the home exercise program based on the participant’s performance during the visit. At a maximum, a participant was instructed to hold the contraction for 3 seconds. The prescribed relaxation duration between contractions was equal to the contraction duration (for example, a 3-second contraction is followed by a 3-second relaxation) except in the case of a participant with very poor muscle function. When that occurred, the participant was told to relax their muscles for twice the contraction duration (for example, 2-second contraction followed by a 4-second relaxation). Participants were instructed to perform 45 pelvic floor muscle contractions every day, divided into 3 sessions of 15 contractions each, to be conducted while supine, sitting, and standing. The home exercise instructions included a graded progression of the contraction duration during the preoperative period. For example, if a participant had 3 weeks before surgery and began with a 3-second contraction duration, they were advised to increase the duration of the contraction by 1 to 2 seconds per week as tolerated. In this scenario, the participant could potentially work up to a 7-second contraction before surgery.
In the instructions for the postop period, participants were allowed to progress to 10 seconds. The interventionists instructed the participants to increase the duration and/or number of pelvic floor muscle contractions based on the participant’s symptoms (maximum contraction duration is 10 seconds), her adherence, barriers to exercise, and performance on the evaluation of her exercise technique.
At the postoperative visits, participants were screened for UI symptoms (using the Optimal Encounter Form, which included a complete symptom checklist), as well as urinary urgency, urinary frequency, voiding dysfunction, and defecatory dysfunction. At each visit, participants were also taught bladder and bowel strategies to protect their surgical repair by reducing intraabdominal pressure on the pelvic floor and to prevent stress and urgency UI during daily life. These strategies included (1) the Knack or Stress Strategy,15 active PFM contraction when coughing, sneezing, or lifting heavy objects; (2) an urge suppression strategy to reduce urgency and prevent urgency UI16; (3) responding to the gastrocolic reflex17 to defecate, and (4) healthy toileting techniques, including posture, relaxation of PFMs to initiate the urine stream and facilitate the passage of stool, taking time to void, and avoiding straining.18 At the final 12-week postoperative visit, a maintenance program of 15 PFM contractions per day at the maximum achieved contraction duration was prescribed. After this visit and during the 24-month follow-up period, adherence was encouraged using flyers mailed to participants on a quarterly basis.
Outcome Assessment
Assessment of Adherence
At each postoperative visit (2, 4–6, and 8 weeks), participants completed an author-designed adherence questionnaire that queried the number of days/week exercises were performed; the number and duration of PFM contractions performed in a typical day; and adherence to 2 of the behavioral strategies (stress and urge suppression strategies). Conceptualization and development of the adherence questionnaire were based on expert opinion. No psychometric testing was conducted. Questions were developed by investigators to assess the following information:
Characteristics of the exercises the participant was doing each day and week (number of repetitions, number of contractions, length of time for each contraction, etc.).
Whether the participant was relaxing her pelvic floor and abdominal muscles when she emptied her bladder, and whether or not she relaxed her pelvic floor muscles and avoided straining during bowel movements.
Whether the participant performed the stress or urge strategy (if appropriate).
Barriers the participants encountered in doing exercises and using strategies.
The interventionist reviewed the adherence questionnaire, discussed any problem areas that were identified, and problem-solved with the participant to identify reasonable solutions to improve adherence.
Additional questions assessed potential barriers to exercise including difficulty finding time to exercise, unsure if exercises were being performed correctly, trouble remembering to exercise, exercises not seen as helpful, no need to exercise due to improvement, and no need to exercise due to surgery. At the 2-week postoperative visit, subjects were asked if fear of pain and fear that exercises would harm their surgical repair were barriers to exercise. These barriers were not subsequently assessed, as this fear was anticipated to decrease over time. Subjects were asked if the exercises caused pain at all subsequent visits. When barriers to PFM exercise were identified, the interventionist provided advice and assisted the participant to identify individualized solutions to each barrier. Such assistance could include developing a plan to exercise at specific times of the day, link the exercises to other established behaviors (eg, brushing teeth or stopping at a red light when driving to and from work), and/or place a reminder note on the refrigerator, computer, or bathroom mirror to do their exercises if remembering their exercises was a challenge.
Outcomes Assessed at 24 Months
Pelvic floor outcomes included: (1) anatomic failure (defined by 1 of the following: descent of the vaginal apex more than one-third into the vaginal canal, anterior or posterior vaginal wall descent beyond the hymen, or retreatment for prolapse); (2) the individual apical descent and anterior prolapse components of anatomic failure; and (3) PFM strength using the Brink grading system.19 The PFM strength grading using the Brink scale was performed by a trained examiner masked to the intervention.
Self-reported participant outcome measures included: (1) the Participant Global Impression of Improvement (in bladder symptoms) (PGII) scale20; (2) the 3 subscales of the Pelvic Floor Distress Inventory (PFDI), the Urinary Distress Inventory (UDI), the Pelvic Organ Prolapse Distress Inventory (POPDI), and the Colorectal-Anal Distress Inventory (CRADI)21; (3) the Pelvic Organ Prolapse/Urinary Incontinence Sexual Questionnaire (PISQ-12)22 for sexually active women; and (4) the Incontinence Severity Index (ISI).23 For the UDI, POPDI, CRADI, PISQ, ISI, and the Brink scales, the 24-month outcome score was evaluated on its own and by calculating change from baseline. The PFDI, which includes UDI, POPDI and CRADI, is a validated, condition-specific instrument that assesses pelvic symptoms and the associated degree of bother. These 3 scales assess lower urinary (UDI), lower gastrointestinal (CRADI), and prolapse symptoms (POPDI).
Data Analysis
Participants’ demographic and clinical characteristics, adherence characteristics, and barriers to exercise were summarized as number and percent for categorical variables and as median [minimum, maximum] for continuous variables.
A latent class analysis (LCA) 24 was employed to identify groups (classes) of subjects with similar response patterns to questions about adherence to the exercise regimen and behavioral strategies collected during each of the follow-up visits (at 2, 4–6, and 8 weeks).
Because the LCA required categorical variables as input, we combined the response options for the number of days exercises were performed in the past week as “0 to 2 days,” “3 to 6 days,” or “7 days.” We categorized the number of contractions completed on a typical day as “0 to 14,” “15 to 44,” or “45 or more.” Response options for the stress and urge behavioral strategies were: “Yes,” “No,” and “Didn’t need to use.”
LCA combined the 3 measures of adherence (number of days performed pelvic floor muscle exercises [PFMEs] in the past week, number of contractions in a typical day, and use of stress and urge strategies) to form separate levels of adherence. The optimal number of classes was assessed using the Bayesian Information Criterion and entropy, which indicates how well participants can be differentiated between classes.25,26 LCA models were run using the poLCA package in R statistical software (Suppl. Table).27
Categorical outcomes at 24 months were compared between adherence classes using unadjusted logistic regression and using models adjusted for preoperative POPQ stage (2 or 3–4). Continuous outcomes at 24 months, and changes from baseline to 24 months, were compared between groups using Student t-tests and general linear models adjusted for baseline score and preoperative POPQ stage. The exception was that the model for 24-month PISQ-12 score was not adjusted for baseline score so that all women who were sexually active at 24 months (who may or may not have been sexually active preoperatively) could be included. The normality of the continuous outcome distributions was not formally assessed; however, the analysis methods used are relatively robust to violations of the normality assumption, particularly for samples as large as ours, and we consider the results to be meaningful despite possible violations of the assumption.
Role of the Funding Source
The Eunice Kennedy Shriver National Institute of Child Health and Human Development Project Scientist and Medical Officer for the PFDN, Dr. Donna Mazloomdoost played a role in the concept and design of the study, as well as data analysis. In addition, Dr. Mazloomdoost played a role in the preparation, review, and approval of the manuscript.
Results
Demographic characteristics of participants are presented in Table 1. Women were predominantly white (83%) and non-Hispanic (80%); the majority were postmenopausal (66%).
Table 1.
Demographic and Clinical Characteristics of Participants (N = 186)a
| Demographic and Clinical Characteristics | No. of Participants (%) |
|---|---|
| Age, y, mean (SD) | 57.5 (10.9) |
| Race, n (%) White Black/African American Asian American Indian/Alaskan Other |
154 (82.8) 15 (8.1) 1 (0.5) 1 (0.5) 15 (8.1) |
| Ethnicity, n (%) Hispanic Non-Hispanic |
38 (20.4) 148 (79.6) |
| BMI, Median [minimum, maximum] (N = 185) | 29 [19.3, 49.8] |
| Menstrual status, n (%) Premenopausal Postmenopausal Not sure |
50 (26.9) 123 (66.1) 13 (7.0) |
| Prior hysterectomy, n (%) | 45 (24.2) |
| Prior stress incontinence surgery, n (%) | 6 (3.2) |
| Prior pelvic organ prolapse surgery, n (%) | 9 (4.8) |
| Vaginal deliveries, median [minimum, maximum] | 3.0 [0.0, 15.0] |
| Cesarean deliveries, median [minimum, maximum] | 0.0 [0.0, 4.0] |
| Preoperative POP-Q stage, n (%) Stage 2 Stage 3 Stage 4 |
72 (38.7%) 105 (56.5%) 9 (4.8%) |
BMI = body mass index.
Adherence characteristics of the 186 participants randomized to BPMT are displayed in Table 2. Overall adherence was highest at the 4- to 6-week postoperative visit and tapered off at 8 weeks. The number of missing responses increased with follow-up time, ranging from 9 to 11% at 2 weeks and increasing to 26 to 30% by the 8-week follow-up visit.
Table 2.
Adherence Characteristics of Participants (N = 186)a
| Characteristic | Categoryb | 2 weeks | 4–6 weeks | 8 weeks |
|---|---|---|---|---|
| No. of days performed PFMEs in past week, n (%) | 0–2 3–6 7 Missing |
35 (18.8) 69 (37.1) 64 (34.4) 18 (9.7) |
12 (6.5) 62 (33.3) 89 (47.8) 23 (12.4) |
5 (2.7) 57 (30.6) 69 (37.1) 55 (29.6) |
| No. of contractions in a typical day, n (%) | 0–14 | 74 (39.8) | 32 (17.2) | 21 (11.3) |
| 15–44 | 63 (33.9) | 66 (35.5) | 58 (31.2) | |
| ≥45 | 29 (15.6) | 62 (33.3) | 52 (28.0) | |
| Missing | 20 (10.8) | 26 (14.0) | 55 (29.6) | |
| Used stress strategy, n (%) | Yes | 58 (31.2) | 98 (52.7) | 91 (48.9) |
| No | 64 (34.4) | 42 (22.6) | 25 (13.4) | |
| Did not need to use | 46 (24.7) | 25 (13.4) | 22 (11.8) | |
| Missing | 18 (9.7) | 21 (11.3) | 48 (25.8) | |
| Used urge suppression strategy, n (%) | Yes | 71 (38.2) | 113 (60.8) | 98 (52.7) |
| No | 54 (29.0) | 30 (16.1) | 16 (8.6) | |
| Did not need to use | 44 (23.7) | 23 (12.4) | 24 (12.9) | |
| Missing | 17 (9.1) | 20 (10.8) | 48 (25.8) | |
| Used voiding strategies, n (%) | Yes No Missing |
156 (83.9) 12 (6.5) 18 (9.7) |
158 (84.9) 8 (4.3) 20 (10.8) |
136 (73.1) 2 (1.1) 48 (25.8) |
| Used defecation strategies, n (%) | Yes No Missing |
146 (78.5) 23 (12.4) 17 (9.1) |
146 (78.5) 20 (10.8) 20 (10.8) |
117 (62.9) 20 (10.8) 49 (26.3) |
a PFMEs = pelvic floor muscle exercises.
b Categorical variables are summarized as number and percent.
Regarding the number of days the participant performed PFMEs, 34% reported performing PFMEs every day at the 2 week follow-up, which increased to 48% at the 4 to 6-week follow-up visit. This was reduced to 37% at 8 weeks. Nineteen percent of participants reported performing PFMEs less than 3 days at 2 weeks, which was reduced to 7% at 4 to 6 weeks and 3% at 8 weeks. For the number of contractions performed on a typical day, 16% of participants reported doing the recommended 45 or more exercises per day at the 2-week visit, which increased to 33% at 4 to 6 weeks and decreased slightly to 28% at 8 weeks. Participants indicated a similar pattern of increasing use of both the stress incontinence and urge suppression strategies. The number of participants reporting no need to use the strategy decreased, from approximately 25 (Stress Strategy) and 24% (Urge Strategy) to 13% (Stress Strategy) and 12% (Urge Strategy), between the 2-week visit and the 4- to 6-week visit, then remained stable at the 8-week visit. When comparing those who reported using the strategy to those who reported “not needing it,” there was no difference in the numbers of contractions.
Potential barriers to adherence are shown in Table 3. The most common problems reported by participants at 4 to 8 weeks were difficulty remembering to do exercises (43% of participants), finding time (28%), and uncertainty regarding performing the exercises correctly (28%). Fourteen percent of subjects believed that surgery replaced the need to do their exercises.
Table 3.
Barriers to Exercise at 4–8 Weeks Postsurgery (N = 186)
| Barrier | Response | n (%) |
|---|---|---|
| Hard to find time to exercise | Yes | 52 (28.0) |
| No | 113 (60.8) | |
| Uncertain | 5 (2.7) | |
| Missing | 16 (8.6) | |
| Unsure if doing exercises correctly | Yes | 52 (28.0) |
| No | 118 (63.4) | |
| Missing | 16 (8.6) | |
| Exercises caused pain | Yes | 18 (9.7) |
| No | 148 (79.6) | |
| Uncertain | 4 (2.2) | |
| Missing | 16 (8.6) | |
| Trouble remembering | Yes | 79 (42.5) |
| No | 91 (48.9) | |
| Missing | 16 (8.6) | |
| Exercises not helping | Yes | 28 (15.1) |
| No | 124 (66.7) | |
| Uncertain | 18 (9.7) | |
| Missing | 16 (8.6) | |
| So much better, did not need exercises | Yes | 20 (10.8) |
| No | 149 (80.1) | |
| Uncertain | 1 (0.5) | |
| Missing | 16 (8.6) | |
| Surgery replaced need for exercises | Yes | 26 (14.0) |
| No | 139 (74.7) | |
| Uncertain | 5 (2.7) | |
| Missing | 16 (8.6) | |
| Other | Yes | 31 (16.7) |
| No | 86 (46.2) | |
| Uncertain | 2 (1.1) | |
| Missing | 67 (36.0) |
LCA identified 2 classes of adherence, descriptively labeled Higher Adherence and Lower Adherence. Table 4 shows the comparison of categorical outcomes across the Higher and Lower Adherence classes at 24 months. The Lower Adherence class shows slightly higher rates of poorer outcomes for all categorical outcomes, but none of the differences were statistically significant. Similarly, the proportion of participants who reported that they were “much better” or “very much better” on the Patient Global Impression of Improvement scale did not differ significantly between the classes.
Table 4.
Comparison of Categorical Outcomes Between Higher and Lower Adherence Classes at 24 Months
| Outcome | Higher Adherence (N = 142) |
Lower Adherencea (N = 44) |
Unadjusted Odds Ratio (95% CI) P Value |
Adjusted Odds Ratio (95% CI) P Value |
|---|---|---|---|---|
| Anatomic failure | 28/119 (23.5%) | 9/34 (26%) | 0.85 (0.36 to 2.04) .72 |
0.9 (0.37 to 2.19) .81 |
| Apical descent more than 1/3 into the vaginal canal | 15/116 (12.9%) | 6/33 (18%) | 0.67 (0.24 to 1.89) .45 |
0.68 (0.24 to 1.93) .47 |
| Anterior prolapse | 13/116 (11.2%) | 5/33 (15%) | 0.71 (0.23 to 2.15) .54 |
0.74 (0.24 to 2.29) .60 |
| Patient Global Impression of Improvement (“very much better” or “much better” vs other categories) | 81/120 (67.5%) | 22/34 (65%) | 1.13 (0.51 to 2.52) .76 |
1.12 (0.5 to 2.51) .79 |
Lower adherence used as reference group.
Comparisons of continuous outcomes at 24 months between Higher and Lower Adherence classes are shown in Table 5. For the subscales of the PFDI, participants in the Lower Adherence class compared to the Higher Adherence class show slightly higher, but not statistically different, levels of distress on the UDI, POPDI, and CRADI scales, with between-group differences of 12 to 17 points (P > .05). No significant differences in ISI or Brink score were found between the adherence classes. PISQ-12 scores among sexually active women were better in the Higher Adherence group compared with the Lower Adherence group at 24 months after adjusting for baseline prolapse stage (Tab. 5). No significant differences were found between the adherence classes in the change scores from baseline to 24 months for any of the continuous outcome measures (Tab. 6).
Table 5.
Comparison of Continuous Outcomes Between Higher and Lower Adherence Classes at 24 Months
| Outcome | Higher Unadjusted Mean (SD) (N = 142) |
Lower Unadjusted Mean (SD) (N = 44) |
Unadjusted Difference (95% CI) P Value |
Higher Adjusted Mean (SE) |
Lower Adjusted Mean (SE) |
Adjusted Difference (95% CI) P Value |
|---|---|---|---|---|---|---|
| Urinary Distress Inventory | 39.3 (48.4) | 51.2 (63.1) | −11.9 (−35.5 to 11.6) .31 |
30.2 (9.5) | 41.5 (11.5) | −11.3 (−31.7 to 9.1) .27 |
| Pelvic Organ Prolapse Distress Inventory | 44.1 (55.9) | 61.3 (75.9) | −17.2 (−45.4 to 11) .22 |
36.5 (10.6) | 48.4 (12.7) | −11.9 (−34.6 to 10.7) .30 |
| Colorectal-Anal Distress Inventory | 49.7 (58.1) | 63.9 (79.2) | −14.2 (−43.6 to 15.2) .34 |
41.9 (11) | 54.8 (13.3) | −12.9 (−36.4 to 10.7) .28 |
| Pelvic Organ Prolapse/Urinary Incontinence Sexual Questionnaire | 36.7 (6.1) | 32.8 (10.1) | 3.8 (−1.1 to 8.8) .12 |
36.6 (0.9) | 32.8 (1.6) | 3.8 (0.1 to 7.6) .04 |
| Incontinence Severity Index | 2.7 (3.4) | 2.7 (2.9) | 0 (−1.3 to 1.2) .98 |
2 (0.6) | 1.9 (0.7) | 0.1 (−1.2 to 1.4) .93 |
| Pelvic Muscle Strength (Brink Score) | 8.2 (2.1) | 8 (2.0) | 0.2 (−0.6 to 1.0) .57 |
8 (0.3) | 8.1 (0.4) | −0.1 (−0.9 to 0.6) .70 |
Table 6.
Comparison of Change in Continuous Outcomes Between Higher and Lower Adherence Classes From Baseline to 24 Months
| Outcome | Higher Unadjusted Mean (SD) (N = 142) |
Lower Unadjusted Mean (SD) (N = 44) |
Unadjusted Difference (95% CI) P Value |
Higher Adjusted Mean (SE) |
Lower Adjusted Mean (SE) |
Adjusted Difference (95% CI) P Value |
|---|---|---|---|---|---|---|
| Urinary Distress Inventory change from baseline | −85.3 (68.6) | −85.8 (68.8) | 0.5 (−26.7 to 27.6) .97 |
−97.8 (9.5) | −86.5 (11.5) | −11.3 (−31.7 to 9.1) .27 |
| Pelvic Organ Prolapse Distress Inventory change from baseline | −78.5 (71.4) | −77.3 (74.0) | −1.2 (−29.7 to 27.3) .93 |
−90.1 (10.6) | −78.2 (12.7) | −11.9 (−34.6 to 10.7) .30 |
| Colorectal-Anal Distress Inventory change from baseline | −57.8 (81.3) | −43.6 (81.9) | −14.2 (−46.4 to 18) .38 |
−65.8 (11) | −53 (13.3) | −12.9 (−36.4 to 10.7) .28 |
| Pelvic Organ Prolapse/Urinary Incontinence Sexual Questionnaire change from baseline | 5.4 (6.3) | 3.0 (10.2) | 2.4 (−3.5 to 8.3) .40 |
5.7 (0.9) | 2.2 (1.6) | 3.4 (−0.1 to 7) .06 |
| Incontinence Severity Index change from baseline | −2.6 (4.1) | −3.1 (4.2) | 0.4 (−1.2 to 2.1) .59 |
−3.5 (0.6) | −3.5 (0.7) | 0.1 (−1.2 to 1.4) .93 |
| Pelvic Muscle Strength (Brink Score) change from baseline | 0.1 (1.9) | 0.4 (2.1) | −0.3 (−1.1 to 0.4) .38 |
0 (0.3) | 0.1 (0.4) | −0.1 (−0.9 to 0.6) .70 |
Discussion
The primary aim of this supplemental analysis was to describe adherence to a behavioral intervention that included PFMT and behavioral strategies to reduce UI, voiding dysfunction and poor defecatory habits among women receiving vaginal reconstructive surgery for POP. Adherence to the components of BPMT was low despite interventionists’ efforts to address barriers to exercises and use of incontinence strategies. It is also noteworthy that the adherence rates in this study were not especially high compared with other studies in the literature. At 4 to 6 weeks postoperative, only 48% of women performed PFMEs daily, and 33% performed the prescribed 45 PFM contractions. Similarly, 49 and 53% of women were using the stress strategy (Knack) and urge suppression strategy respectively at the 8-week follow-up visit. Between the 4- to 6-week and 8-week visits, 11 and 5% fewer women were exercising daily and performing 45 PFM contractions on a typical day, respectively. The low rates of adherence to this behavioral intervention might have been influenced by the exercise barriers reported by participants.
Braekken et al. (2010) studied women with prolapse stages I, II, and III. In the PFMT Group, women participated in 89% of the prescribed home exercises and 86% of the physical therapist training sessions. Ten percent of women in the control group reported that they had performed more PFMT than they did before baseline.28 Fine et al. (2007) studied primiparous women who had learned about PFM exercises. Of those, 68% performed the exercises after delivery and 67% performed exercises at least daily, 26% performed exercises less than once daily, but greater than once per week, and 19% performed exercises less that once per week.29 Borello-France et al. (2013) studied women with SUI. Of those, 86% exercised at least 5 days a week, with 80% performing at least 30 contractions when they exercised. At 3, 6, and 12 months post randomization, 95, 88, and 80% of women, respectively, indicated that they were still performing PFM exercises.30 Chiarelli et al. (2002) studied women who had assisted delivery (forceps or ventouse). In the Intervention group, approximately 84% of women performed the pelvic floor exercises 3 times per week or more.31
Regarding the correlation of participant adherence to anatomic and/or participant-reported outcomes, no statistically significant relationship was found between BPMT adherence and reported outcomes 24 months postoperatively. Similarly, Hung et al. found that improvements in UI were influenced by symptom severity and PFM strength, not exercise adherence, based on a 4-month home PFM strengthening exercise program.32 Felicissimo et al. also found no difference in PFM strength, urine leakage, or quality-of-life outcomes in women with SUI following 8 weeks of either unsupervised or intensively supervised PFM training.33
It is feasible that surgical correction was so robust as to overshadow both the clinical effects of BPMT as well as the more subtle effect of adherence. Furthermore, the results cannot be generalized to situations in which BPMT is used alone or with concomitant nonsurgical interventions. Pain is a suspected factor in performing PFM exercises after vaginal reconstructive surgery for POP, particularly early in the postoperative period. However, only 9.7% indicated that pain actually interfered with their exercise. Nonetheless, women may have been hesitant to contract their PFMs quickly and forcefully (“the Knack”) in this postoperative period, decreasing the effectiveness to prevent urine leakage. Furthermore, 14% considered BPMT unnecessary because of their recent POP surgery. With the behavioral program and surgery being implemented concurrently, women may not have been able to discern the relative effects of the interventions and may have put more faith in the surgical correction, thereby lessening their motivation and fidelity to a more labor-intensive therapy. Additionally, despite educational strategies being implemented, subjects most frequently endorsed recall as a barrier to perform exercises. Interestingly, prior work from the PFDN demonstrated that women with SUI or stress-predominant mixed UI randomized to intravaginal continence pessary, behavioral intervention, or both found 80% of women still performed PFM exercises at 12 months postrandomization, with 61% exercising at least 3 days per week. Furthermore, ≥87% continued using the “Knack” to prevent SUI after 3 months of the intervention.30 This is encouraging that perhaps in the correct environment, subjects may have enhanced adherence to behavioral programs.
There are several limitations to our study. First, measurement of adherence was based on self-report, which may be influenced by reporting bias. Though used in previous trials, the adherence questionnaire was not validated, and there is a continued need for validated adherence outcome measures.13,14 Second, adherence was measured at each visit, rather than in real time, making it subject to recall bias. Measuring adherence and barriers to adherence at each postoperative visit enabled interventionists to follow up on prior recommendations and provide advice on any new or emerging barriers. However, data on the interventionists’ recommendations were not collected, and we did not examine changes in barriers over time. Third, although we conducted centralized training and certification, interventionists had varying degrees of prior behavioral experience, and there is the potential for variation in implementation of the interventions across sites. Fourth, because study participants were volunteers in a clinical trial, they may not have been representative of the larger population of women receiving surgery for POP. It is possible that study participants were more motivated to adhere to the behavioral intervention compared with women who did not enroll in the trial, and such selection bias could impact results. Finally, results could have been affected by missing data, which occurred in up to 30% of participants at individual time points.
Study strengths include the use of data obtained from a randomized clinical trial conducted in a large sample of women from diverse participating clinical sites. The intervention was comprehensive in terms of having multiple components based on evidence from previous behavioral and physical therapist trials. It included a particular emphasis on adherence, including assessment at each visit and specific attention to addressing barriers. Interventionist training was comprehensive and standardized across sites based on centralized training. The interventionist training included teaching behavioral principles and strategies, and specific techniques for optimizing adherence. These included individualizing treatment to the participant, understanding the gradual and progressive nature of the learning curve, using positive reinforcement, and shaping, inoculation against failure, facilitating realistic expectations, guidance in self-management such as integrating new behaviors into daily life, techniques for remembering to do exercises and use strategies, and setting contingencies for completing exercises. This enhances the study generalizability. Lastly, outcomes were measured using validated questionnaires.
Conclusions
The results of this study indicate that adherence to a program of perioperative behavioral therapy with PFM training was low in women receiving vaginal reconstructive surgery for prolapse, despite efforts to encourage participants and to address barriers. Furthermore, we were unable to demonstrate that degree of adherence was related to clinical outcomes for women subsequent to vaginal reconstructive surgery for POP.
Supplementary Material
Acknowledgments
Pelvic Floor Disorders Network Contributors
Brown University, Providence, Rhode Island: Nicole Korbly.
Cleveland Clinic Foundation, Cleveland, Ohio: Matthew D. Barber, Anna Frick, Betsy O’Dougherty, Katie Propst, Ly Pung, Beri M. Ridgeway, Cheryl Williams.
Duke University Medical Center, Durham, North Carolina: J. Eric Jelovsek, Cindy L. Amundsen, Ingrid Harm-Ernandes, Mary Raynor, Nazema Y. Siddiqui, Anthony G. Visco, Alison C. Weidner, Jennifer M. Wu.
Kaiser Permanente Bellflower: Sharon Jakus-Waldman.
Kaiser Permanente San Diego: Gouri Diwadkar, Keisha Y. Dyer, Lynn M. Hall, Linda M. Mackinnon, Shawn A. Menefee, John N. Nguyen, Jasmine Tan-Kim, Gisselle Zazueta-Damian.
Loyola University, Chicago, Illinois: Linda Brubaker, Elizabeth Mueller, Mary Tulke.
Magee-Women’s Hospital, Pittsburgh, Pennsylvania: Diane Borello-France.
Oregon Health & Science University, Portland, Oregon: Sara Cichowski.
RTI International: Marie Gantz, Lauren Klein Warren, Daryl Matthews, Scott Grey, Amanda Shaffer, Tamara T. Terry, Jutta Thornberry, Dennis Wallace, Ryan E. Whitworth, Kevin A. Wilson.
Vanderbilt University: Katherine Hartmann.
University of Alabama at Birmingham, Department of Medicine: Alicia Ballard, Julie Burge, Kathryn L. Burgio, Kathy Carter, Patricia S. Goode, Alayne D. Markland, Lisa S. Pair, Candace Parker-Autry, Holly E. Richter, R. Edward Varner, Tracey S. Wilson.
University of California at San Diego: Michael E. Albo, Cara Grimes, Emily S. Lukacz, Charles W. Nager.
University of Michigan, Ann Arbor, Michigan: Yang Wang Casher, Yeh-Hsin Chen, Donna DiFranco, Bev Marchant, Cathie Spino, John T. Wei.
University of Pennsylvania, Philadelphia, Pennsylvania: Diane Newman.
University of Utah: Jan Baker, Yvonne Hsu, Maria Masters, Amy Orr.
University of Texas Southwestern: Shanna Atnip, Sunil Balgobin, Elva Kelly Moore, David Rahn, Joseph Schaffer, Clifford Wai.
Northwest Physician Group, Amarillo, Texas: Susan F. Meikle.
Birmingham/Atlanta Geriatric Research, Education, and Clinical Center at the Birmingham VA Health Care System, Birmingham, Alabama: Kathryn L. Burgio, Alayne D. Markland.
Uniformed Services University of Health Sciences, Washington, DC-Baltimore Area: Scott Grey.
Contributor Information
Diane Borello-France, Department of OB/GYN, Magee-Womens Hospital, Department of Physical Therapy, Duquesne University, Pittsburgh, Pennsylvania, USA.
Diane K Newman, Division of Urology, Penn Center for Continence and Pelvic Health, University of Pennsylvania, Philadelphia, Pennsylvania, USA.
Alayne D Markland, Department of Medicine, Birmingham/Atlanta Geriatric Research, Education, and Clinical Center at the Birmingham VA Health Care System, University of Alabama at Birmingham, Birmingham, Alabama, USA.
Katie Propst, Department of Obstetrics & Gynecology, Cleveland Clinic Foundation, Cleveland, Ohio, USA.
J Eric Jelovsek, Department of Obstetrics & Gynecology, Duke University, Durham, North Carolina, USA.
Sara Cichowski, Department of Obstetrics & Gynecology, Oregon Health & Science University, Portland, Oregon, USA.
Marie G Gantz, Department of Biostatistics and Epidemiology, RTI International, Triangle Park, North Carolina, USA.
Sunil Balgobin, Department of Obstetrics & Gynecology, University of Texas Southwestern, Dallas, Texas, USA.
Sharon Jakus-Waldman, Department of Obstetrics, Gynecology and Urogynecology, Kaiser Permanente, Downey, California, USA.
Nicole Korbly, Department of Obstetrics & Gynecology, Brown University, Providence, Rhode Island, USA.
Donna Mazloomdoost, The Eunice Kennedy Shriver National Institute of Child Health and Human Development, Bethesda, Maryland, USA.
Kathryn L Burgio, Department of Medicine, Birmingham/Atlanta Geriatric Research, Education, and Clinical Center at the Birmingham VA Health Care System, University of Alabama at Birmingham, Birmingham, Alabama, USA.
for The NICHD Pelvic Floor Disorders Network:
Nicole Korbly, Matthew D Barber, Anna Frick, Betsy O’Dougherty, Katie Propst, Ly Pung, Beri M Ridgeway, Cheryl Williams, J Eric Jelovsek, Cindy L Amundsen, Ingrid Harm-Ernandes, Mary Raynor, Nazema Y Siddiqui, Anthony G Visco, Alison C Weidner, Jennifer M Wu, Sharon Jakus-Waldman, Gouri Diwadkar, Keisha Y Dyer, Lynn M Hall, Linda M Mackinnon, Shawn A Menefee, John N Nguyen, Jasmine Tan-Kim, Gisselle Zazueta-Damian, Linda Brubaker, Elizabeth Mueller, Mary Tulke, Diane Borello-France, Sara Cichowski, Marie Gantz, Lauren Klein Warren, Daryl Matthews, Scott Grey, Amanda Shaffer, Tamara T Terry, Jutta Thornberry, Dennis Wallace, Ryan E Whitworth, Kevin A Wilson, Katherine Hartmann, Alicia Ballard, Julie Burge, Kathryn L Burgio, Kathy Carter, Patricia S Goode, Alayne D Markland, Lisa S Pair, Candace Parker-Autry, Holly E Richter, R Edward Varner, Tracey S Wilson, Michael E Albo, Cara Grimes, Emily S Lukacz, Charles W Nager, Yang Wang Casher, Yeh-Hsin Chen, Donna DiFranco, Bev Marchant, Cathie Spino, John T Wei, Diane Newman, Jan Baker, Yvonne Hsu, Maria Masters, Amy Orr, Shanna Atnip, Sunil Balgobin, Elva Kelly Moore, David Rahn, Joseph Schaffer, Clifford Wai, Susan F Meikle, Kathryn L Burgio, Alayne D Markland, and Scott Grey
Author Contributions
Concept/idea/research design: D. Borello-France, D.K. Newman, A.D. Markland, J.E. Jelovsek, S. Cichowski, D. Mazloomdoost, K.L. Burgio
Writing: D. Borello-France, D.K. Newman, A.D. Markland, K. Propst, J.E. Jelovsek, S. Cichowski, M.G. Gantz, S. Balgobin, N. Korbly, D. Mazloomdoost, K.L. Burgio
Data collection: S. Balgobin
Data analysis: D.K. Newman, A.D. Markland, M.G. Gantz, D. Mazloomdoost
Project management: D. Borello-France, K.L. Burgio
Fund procurement: M.G. Gantz, K.L. Burgio
Providing participants: A.D. Markland, S. Balgobin
Providing institutional liaisons: D.K. Newman, A.D. Markland
Consultation (including review of manuscript before submitting): D.K. Newman, A.D. Markland, J.E. Jelovsek, M.G. Gantz, S. Jakus-Waldman, K.L. Burgio
Funding
This study was supported by grants from the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) and the National Institutes of Health (NIH) Office of Research on Women’s Health (U01 HD41249, U10 HD41250, U10 HD41261, U10 HD41267, U10 HD54136, U10 HD54214, U10 HD54215, U01 HD069031, and U10 HD54241). D. Mazloomdoost was affiliated with the funding source for this study.
Ethics Approval
Institutional review boards at each network site provided protocol approval, and each participant provided informed consent to participate.
Data Availability
The OPTIMAL public use dataset is available through the Eunice Kennedy Shriver National Institute of Child Health and Human Development Data and Specimen Hub (DASH) https://dash.nichd.nih.gov/.
Clinical Trial Registration
This trial was registered at ClinicalTrials.gov (NCT00597935).
Disclosures
The authors completed the ICMJE Form for Disclosure of Potential Conflicts of Interest and reported no conflicts of interest.
References
- 1. Wu JM, Vaughan CP, Goode PS, et al. Prevalence and trends of symptomatic pelvic floor disorders in U.S. women. Obstet Gynecol. 2014;123:141–148. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Dumoulin C, Alewijnse D, Bo K, et al. Pelvic floor muscle training adherence: tools, measurements, and strategies – 2011 ICS state of the science seminar research paper II of IV. Neurourol Urodynam. 2015;34:615–621. [DOI] [PubMed] [Google Scholar]
- 3. Dumoulin C, Cacciari LP, Hay-Smith EJC. Pelvic floor muscle training versus no treatment, or inactive control treatments, for urinary incontinence in women. Cochrane Database Syst Rev. 2018;10:CD005654. 10.1002/14651858.CD005654.pub4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Wu YM, Welk B. Revisiting current treatment options for stress urinary incontinence and pelvic organ prolapse: a contemporary literature review. Res Rep Urol. 2019;11:179–188. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Hagen S, Glazener C, McClurg D, et al. Pelvic floor muscle training for secondary prevention of pelvic organ prolapse (PREVPROL): a multicentre randomised controlled trial. Lancet. 2017;389:393–402 Epub 2016 Dec 21. PMID: 28010994. [DOI] [PubMed] [Google Scholar]
- 6. Li C, Gong Y, Wang B. The efficacy of pelvic floor muscle training for pelvic organ prolapse: a systematic review and meta-analysis. Int Urogynecol J. 2016;27:981–992. [DOI] [PubMed] [Google Scholar]
- 7. Barber MD, Brubaker L, Burgio KL, et al. Comparison of 2 transvaginal surgical approaches and perioperative behavioral therapy for apical vaginal prolapse: the OPTIMAL randomized trial. JAMA. 2014;311:1023–1034. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Jelovsek JE, Barber MD, Brubaker L, et al. Effect of uterosacral ligament suspension vs sacrospinous ligament fixation with or without perioperative behavioral therapy for pelvic organ vaginal prolapse on surgical outcomes and prolapse symptoms at 5 years in the OPTIMAL randomized clinical trial. JAMA. 2018;319:1554–1565. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Dumoulin C, Hay-Smith J, Frawley H, et al. 2014 consensus statement on improving pelvic floor muscle training adherence: international continence society 2011 state-of-the-science seminar. Neurourol Urodyn. 2015;34:600–605. [DOI] [PubMed] [Google Scholar]
- 10. Burkhart PV, Sabaté E. Adherence to long-term therapies: evidence for action. J Nurs Scholarsh. 2003;35:207. PMID: 14562485. 10.1111/j.1547-5069.2003.tb00001.x. [DOI] [PubMed] [Google Scholar]
- 11. Bø K, Hilde G. Does it work in the long term? A systematic review on pelvic floor muscle training for female stress urinary incontinence. Neurourol Urodyn. 2013;32:215–223. 10.1002/nau.22292. Epub 2012 Jul 27. [DOI] [PubMed] [Google Scholar]
- 12. Venegas M, Carrasco B, Casas-Cordero R. Factors influencing long-term adherence to pelvic floor exercises in women with urinary incontinence. Neurourol Urodyn. 2018;37:1120–1127. [DOI] [PubMed] [Google Scholar]
- 13. Barber MD, Brubaker L, Menefee S, et al. Pelvic floor disorders network. Operations and pelvic muscle training in the management of apical support loss (OPTIMAL) trial: design and methods. Contemp Clin Trials. 2009;30:178–189. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Brink C, Wells T, Sampselle C, Taillie ER, Mayer R. A digital test for pelvic muscle strength in women with urinary incontinence. Nurs Res. 1994;43:352–356. [PubMed] [Google Scholar]
- 15. Burgio KL, Locher JL, Goode PS, et al. Behavioral versus drug treatment for urge incontinence in older women: a randomized controlled trial. JAMA. 1998;280:1995–2000. [DOI] [PubMed] [Google Scholar]
- 16. Wyman JF, Burgio KL, Newman DK. Practical aspects of lifestyle modification and behavioural interventions in the treatment of overactive bladder and urgency urinary incontinence. Int J Clin Pract. 2009;63:1177–1191. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Duthie HL. Colonic response to eating. Gastroenterology. 1978;75:527–529. [PubMed] [Google Scholar]
- 18. Devreese AM, Nuyens G, Staes F, Vereecken RL, de Weerdt W, Stappaerts K. Do posture and straining influence urinary-flow parameters in normal women? Neurourol Urodyn. 2000;19:3–8. [DOI] [PubMed] [Google Scholar]
- 19. Brink CA, Sampselle CM, Wells TJ, Diokno AC, Gillis GL. A digital test for pelvic muscle strength in older women with urinary incontinence. Nurs Res. 1989;38:196–199. [PubMed] [Google Scholar]
- 20. Yalcin I, Bump RC. Validation of two global impression questionnaires for incontinence. Am J Obstet Gynecol. 2003;189:98–101. [DOI] [PubMed] [Google Scholar]
- 21. Barber MD, Kuchibhatla MN, Pieper CF, Bump RC. Psychometric evaluation of 2 comprehensive condition-specific quality of life instruments for women with pelvic floor disorders. Am J Obstet Gynecol. 2001;185:1388–1395. [DOI] [PubMed] [Google Scholar]
- 22. Rogers RG, Coates KW, Kammerer-Doak D, Khalsa S, Qualls C. A short form of the pelvic organ prolapse/urinary incontinence sexual questionnaire (PISQ-12). Int Urogynecol J Pelvic Floor Dysfunct. 2003;14:164–168. [DOI] [PubMed] [Google Scholar]
- 23. Sandvik H, Hunskaar S, Seim A, Hermstad R, Vanvik A, Bratt H. Validation of a severity index in female urinary incontinence and its implementation in an epidemiological survey. J Epidemiol Community Health. 1993;47:497–499. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Collins LM, Lanza ST. Latent Class and Latent Transition Analysis: With Applications in the Social, Behavioral, and Health Sciences. New York: Wiley; 2010. 10.1002/9780470567333. [DOI] [Google Scholar]
- 25. Kim SY. Determining the number of latent classes in single- and multi-phase growth mixture models. Struct Equ Modeling. 2014;21:263–279. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Celeux G, Soromenho G. An entropy criterion for assessing the number of clusters in a mixture model. J Classif. 1996;13:195–212. [Google Scholar]
- 27. Linzer DA, Lewis JB. poLCA: an R package for polytomous variable latent class analysis. J Stat Software. 2011;42:1–29. 10.18637/jss.v042.i10. [DOI] [Google Scholar]
- 28. Brækken IH, Majida M, Ellström Engh M, Bø K. Can pelvic floor muscle training reverse pelvic organ prolapse and reduce prolapse symptoms? An assessor-blinded, randomized, controlled trial. Am J Obstet and Gynecol. 2010;203:170.e1–170.e7. [DOI] [PubMed] [Google Scholar]
- 29. Fine P, Burgio K, Borello-France D, et al. Teaching and practicing of pelvic floor muscle exercises in primiparous women during pregnancy and the postpartum period. Am J Obstet Gynecol. 2007;197:107.e1–107.e5. [DOI] [PubMed] [Google Scholar]
- 30. Borello-France D, Burgio K, Goode P, et al. Adherence to behavioral interventions for stress incontinence: rates, barriers, and predictors. Phys Ther. 2013;93:757–773. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Chiarelli P, Cockburn J. Promoting urinary continence in women after delivery: randomised controlled trial. BMJ. 2002;324:1241. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Hung HC, Chih SY, Lin HH, Tsauo JY. Exercise adherence to pelvic floor muscle strengthening is not a significant predictor of symptom reduction for women with urinary incontinence. Arch Phys Med Rehabil. 2012;93:1795–1800. [DOI] [PubMed] [Google Scholar]
- 33. Felicissimo MF, Carneiro MM, Saleme CS, Pinto TZ, da Fonseca AM, da Silva-Filho AL. Intensive supervised versus unsupervised pelvic floor muscle training for the treatment of stress urinary incontinence: a randomized comparative trial. Int Urogyneol J Pelvic Floor Dysfunct. 2010;21:835–840. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The OPTIMAL public use dataset is available through the Eunice Kennedy Shriver National Institute of Child Health and Human Development Data and Specimen Hub (DASH) https://dash.nichd.nih.gov/.