The present study was designed as a national cross-sectional online survey study, which is part of a larger research initiative (17, 19, 40) aimed at exploring the long-term impacts of SPT on HRQoL, employment, and healthcare interactions.

In general, individuals aged 20–49 in Sweden report similar HRQoL (41), while notable declines in physical health are observed in older age groups, particularly among those aged 80 and above. Significant age-related differences are also evident in mental health; however, older age groups, excluding the oldest (those aged 80 and above), report higher energy and emotional well-being. Additionally, individuals with a university education exhibit significantly higher HRQoL compared to those with only mandatory education. Employees have the highest HRQoL, whereas individuals on sick leave have the lowest. Finally, men in Sweden exhibit significantly higher HRQoL than women (41).

Life expectancy for women in Sweden is high in global comparison (84.3 years), and the maternal mortality rate is four deaths per 100,000 live births (42). Parental leave of up to 480 days per child, as well as healthcare, is tax-funded (43, 44). Gender differences in health reflect global trends, with men experiencing higher mortality rates, while women demonstrate greater morbidity and utilize healthcare services more frequently (45). Women are predominantly on sick leave for psychiatric and musculoskeletal conditions and exhibit higher sickness rates than men (46). The employment rate for women stands at 72.8%, with a substantial proportion employed in the healthcare and education sectors (47).

In Sweden, sexual and reproductive health services are predominantly delivered within the primary care sector. In contrast, childbirth care is almost exclusively hospital-based, with home births and freestanding birth centers representing exceptionally uncommon alternatives. Midwives autonomously manage care during normal pregnancies and uncomplicated births, whereas obstetricians are consulted in complicated cases (48, 49). As late as 2025, national guidelines for pelvic floor dysfunction were implemented in Sweden (50).

The study involved adult women (18 years or older) proficient in understanding and speaking Swedish who had experienced third- or fourth-degree perineal lacerations at least 18 months before the data collection, irrespective of vaginal birth mode. Participants were selected without regard to the existence or seriousness of self-reported SPT-related health problems, as well as whether they had undergone reconstructive surgery or not. Conversely, women were not eligible if they were still on full-time parental leave after SPT or pregnant and were excluded if they experienced a stillbirth or perinatal death of the baby within 28 days postpartum at the time of the SPT.

A digital poster was disseminated through various online channels focusing on perineal health and pelvic floor functioning, including social media platforms such as Facebook, Instagram, and TikTok, as well as blogs and interest groups. Additionally, our department's social media account on Facebook contributed to this effort. The poster outlined the research project and linked to the project's homepage, which provided detailed written information on the project, contact information for the research team, and a link to the Research Electronic Data Capture (REDCap)-based (51, 52) questionnaire for participation.

An online questionnaire was developed in Swedish. Study data were collected and managed using REDCap electronic data capture tools hosted at Umeå University (51, 52). REDCap is a secure, web-based software platform designed to support data capture for research studies, providing 1) an intuitive interface for validated data capture; 2) audit trails for tracking data manipulation and export procedures; 3) automated export procedures for seamless data downloads to common statistical packages; and 4) procedures for data integration and interoperability with external sources (51, 52). The questionnaire employed for data collection in this study, comprising eight sections and a total of 130 items, was designed to require approximately 30 min for completion (Figure 1). The initial sections gathered demographic data, lifestyle choices, and reproductive health information. The latter sections employed five validated instruments assessing HRQoL, symptoms, and bother from the pelvic floor, sexual functioning, and work ability. More specifically, HRQoL was assessed using a combination of instruments that evaluate HRQoL, such as the RAND 36-Item Health Survey 1.0 (53–55) (included in this paper), as well as condition-specific HRQoL instruments (not reported here). The questionnaire underwent pilot testing with two women facing SPT-related health problems and eight healthcare professionals at Umeå University, resulting in minor adjustments to the language and programming of the questionnaire. Public involvement in the research project played a crucial role in both study design regarding the content of the questionnaire and assistance during social media outreach for data collection. Participants were required to confirm their eligibility and provide written informed consent before accessing the questionnaire. Access was revoked for those who did not meet these criteria. Data collection occurred from November 20, 2023, to March 15, 2024, with two reminders on the same online channels issued after 2–3 weeks and an additional 4 weeks, respectively.

A data analysis plan was outlined prior to initiation of data analysis and discussed with a statistician. Descriptive statistics are expressed as numbers (n) and their associated percentages. A missing data analysis revealed randomness in the missing data and a non-monotone pattern with 1.1% of values missing in total (Supplementary Data S1). Hence, missing data was addressed by applying a fully conditional specification multiple imputation (Markov chain Monte Carlo method) with five imputations (57).

The RAND-36 was scored by following the official scoring instructions provided by RAND Health Care® (58), as shown in Supplementary Figure S1. Unanswered items were excluded from this calculation, so the scale scores reflect the average of only those items that the respondent completed (54). The PCS score for physical health (56) was calculated by summing the subscale scores for physical functioning, role limitations due to physical health problems, pain, and general health perceptions, and then dividing by four. Similarly, the MCS score for mental health was derived by adding the scores for energy/fatigue, social functioning, role limitations due to emotional health problems, and emotional well-being, and dividing by four (56). A one-sample t-test was performed to evaluate whether the mean score of PCS and MCS differed from the reference population's mean PCS of 73.625 (SD = 29.45) and mean MCS of 71.7 (SD = 27.15). Data on Women’s mean RAND-36 scores in Sweden published by Ohlsson-Nevo et al. (41) were used as the reference population. Further, the raw mean RAND-36 scores in our sample were standardized by calculating T-scores, with a mean of 50 and a standard deviation of 10, to visualize the difference between the mean RAND-36 scores in our sample and those in the general population in Sweden (41). Floor and ceiling effects of the RAND-36 were also examined.

A multiple linear regression analysis was performed on the imputed dataset to identify predictors of PCS and MCS after SPT (59). The two models were constructed in a single block, with theoretically and clinically a priori identified sociodemographic, obstetric, and occupational predictors (Table 1). The models display adjusted unstandardized coefficients (B) with 95% confidence intervals. After fitting the models, assumptions in linear regression, i.e., linearity, homoscedasticity, independent and normally distributed errors and multicollinearity, were confirmed. We could not detect any multicollinearity between predictors. Further, as outliers and influential cases in the dataset were identified, a sensitivity analysis was conducted to test the robustness of the findings, where cases with high residuals and high leverage were excluded and models were adjusted for the same predictors. A dropout analysis (Supplementary Table S1) was ultimately performed, comparing participants who provided only background data with those included in the linear regression models, applying independent-samples t-test on continuous data and the Chi-Squared test or Fisher's Exact Test on categorical data (59).

Data were analyzed using IBM SPSS Statistics version 29, with statistical significance determined at p < 0.05. The paper complies with the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) guidelines for cross-sectional studies (60) (Supplementary Table S2). The authors utilized Copilot® and Grammarly® for language editing of the manuscript, subsequently reviewing and refining the content language wise, and thus, assume full responsibility for the publication's accuracy.

The research project received ethical approval from the Swedish Ethical Review Authority on August 28, 2023 (Dnr: 2023-04018-01) and adheres to the Declaration of Helsinki (61). Participation was voluntary, allowing individuals to withdraw at any time without needing to provide a reason. Participants gave written informed consent digitally before accessing the questionnaire, which included options to skip intimate questions. The questionnaire was anonymous, with no email or IP addresses collected, and no incentives were provided. Furthermore, the funding organizations were not involved in any aspect of the study's design, data management, analysis, or the preparation and submission of the manuscript.

The link to the questionnaire was accessed by a total of 282 individuals, with 255 individuals proceeding beyond the consent form. The inclusion procedure for this study is shown in Figure 2. The final sample consisted of 221 participants.

Table 2 provides an overview of the demographics of the total study population (n = 221), all identifying themselves as women. Most included women were born in Sweden (93.2%), had university-level education (81.9%), and cohabited with their partner (87.8%). Most women were employed (86.0%) or self-employed (4.1%), worked full-time (68.8%) and overall, the study population reported good work ability. Regarding obstetric properties, most women were multiparous (76.0%), and approximately one-third of the women sustained a fourth-degree perineal laceration (34.4%). On average, women had been living with SPT for 10.0 (SD 8.2; median 8.0, Q1−Q3: 4.0–13.0) years, and one third had pursued reconstructive surgery (30.3%). Based on self-reported symptom bother, the women were distributed evenly across three groups: mild (30.3%–31.7%), moderate (32.6%–33.9%), and severe symptom bother (34.8%–36.2%). The dropout analysis (Supplementary Table S1) revealed no significant differences in characteristics between women who dropped out and those who completed the questionnaire, except for the highest completed educational level. University-level education was overrepresented among included participants, and primary/secondary education was overrepresented among excluded participants.

Table 3 depicts the RAND-36 score distribution across the 8 subscales, the health change (single item), and the two composite summary scores for physical and mental health (PCS and MCS). Participants scored the lowest on the subscale energy/fatigue [mean 46.9, SD (Standard deviation) 20.3, CI (Confidence interval) 95% 44.2–49.6] and the highest on physical functioning (mean 82.8, SD 19.8, CI 95% 80.1–85.4). The single-item health change was rather stagnant at 54.5 (SD 21.6; CI 95% 51.6–57.4), suggesting improvement in overall health status during the past year. The mean PCS score was 70.7 (SD 22.1, CI 95% 67.7–73.6), while the mean MCS score was 63.2 (SD 21.4; CI 95% 60.4–66.0). The mean PCS for women in our sample (M = 70.7, SD = 22.1) was significantly lower than in the reference population (41) of women in Sweden (M = 73.63, SD = 29.45), t [df (degrees of freedom) 220] = −1.99, p = 0.047, Cohen's d = 0.13. Similarly, the mean MCS in our sample of women (M = 63.2, SD = 21.4) was significantly lower than in the reference population (41) of women in Sweden (M = 71.7, SD = 27.15), t (df 220) = −5.90, p < 0.001, Cohen's d = 0.40. Additionally, our sample reported worse-than-average RAND-36 scores compared to the reference population (visualized in T-scores) across all RAND-36 subscales and composite summary scores except for physical functioning and pain. The data was predominantly skewed, and the dimensions of role limitations due to physical health and due to emotional problems displayed floor effects. Further, physical functioning, role limitations due to physical health, role limitations due to emotional problems, social functioning, and pain showed ceiling effects.

A strength of the study is that we achieved a comparatively robust sample size despite the potentially sensitive nature of the topic. Moreover, our sample encompassed a spectrum of women reporting no symptom bother after SPT to those experiencing severe symptom bother, strengthening external validity. The use of a widely accepted and thoroughly validated instrument for measuring HRQoL (53–55) further strengthened the construct validity and reliability of this study. The standardization of scores presented as T-scores (74) facilitated comparison to the normative population, further enhancing the external validity of our findings.

However, there are sources of bias that need to be discussed. Our sample was predominantly composed of Swedish-born women with university-level education who cohabited with partners and were engaged in full-time employment, suggesting a considerable degree of socioeconomic stability in their lives, which may have introduced potential selection bias to the findings. Empirical evidence suggests that the presence of a supportive partner and robust social networks significantly contribute to individual well-being among middle-aged women (75). Therefore, research should expand beyond our cohort to include a more diverse range of participants, encompassing various ethnicities, educational backgrounds, and socioeconomic classes, which could facilitate a more in-depth understanding of HRQoL in women with SPT, especially in potentially vulnerable subgroups. Another potential selection bias was that approximately 30% of the women reported fourth-degree perineal laceration, contrasting with official Swedish reports (14, 76) that report an annual incidence of fourth-degree perineal laceration of 5%–10%, but the degree of perineal laceration was not associated with physical or mental health. The included data relies solely on patient-reported outcome measures (PROMs) collected via digital platforms. While PROMs provide valuable insights into subjective experiences, they are prone to information bias, including social desirability and recall bias (77), which may affect reliability. Moreover, the use of social media and patient organizations as recruitment channels may have disproportionately attracted individuals with particularly salient experiences, introducing information overload (78) and impacting the validity of results. To reduce the risk of information bias and strengthen the study's validity and reliability, validated instruments were incorporated into the questionnaire whenever feasible, except for background items. However, future research could benefit from incorporating objective health measures, derived from registers or medical records, alongside PROMs to improve the validity and reliability of results. The wide confidence intervals for some categorical predictors suggest caution in interpreting their associations with the outcome, highlighting the need for more precise measurement tools to improve reliability. Notably, significant ceiling effects were observed in the RAND-36 role limitation subscales, which limited the detection of variability among high-scoring participants−a limitation also reported in previous studies (41, 79). However, the use of composite summary scores in the regression model mitigated this issue, as they did not exhibit floor or ceiling effects. Floor and ceiling effects underscore the potential value of alternative scoring methods for RAND-36, such as item response theory (74), and the use of condition-specific QoL instruments to better capture HRQoL following SPT.