Development and Evaluation of a Prediction Model for Ascertaining Rheumatic Heart Disease Status in Administrative Data.

BACKGROUND: Previous research has raised substantial concerns regarding the validity of the International Statistical Classification of Diseases and Related Health Problems (ICD) codes (ICD-10 I05-I09) for rheumatic heart disease (RHD) due to likely misclassification of non-rheumatic valvular disease (non-rheumatic VHD) as RHD. There is currently no validated, quantitative approach for reliable case ascertainment of RHD in administrative hospital data. METHODS: A comprehensive dataset of validated Australian RHD cases was compiled and linked to inpatient hospital records with an RHD ICD code (2000-2018, n=7555). A prediction model was developed based on a generalized linear mixed model structure considering an extensive range of demographic and clinical variables. It was validated internally using randomly selected cross-validation samples and externally. Conditional optimal probability cutpoints were calculated, maximising discrimination separately for high-risk versus low-risk populations. RESULTS: The proposed model reduced the false-positive rate (FPR) from acute rheumatic fever (ARF) cases misclassified as RHD from 0.59 to 0.27; similarly for non-rheumatic VHD from 0.77 to 0.22. Overall, the model achieved strong discriminant capacity (AUC: 0.93) and maintained a similar robust performance during external validation (AUC: 0.88). It can also be used when only basic demographic and diagnosis data are available. CONCLUSION: This paper is the first to show that not only misclassification of non-rheumatic VHD but also of ARF as RHD yields substantial FPRs. Both sources of bias can be successfully addressed with the proposed model which provides an effective solution for reliable RHD case ascertainment from hospital data for epidemiological disease monitoring and policy evaluation.

Stillbirth risk across pregnancy by size for gestational age in Western Cape Province, South Africa: Application of the fetuses-at-risk approach using perinatal audit data.

BACKGROUND: There is little published work on the risk of stillbirth across pregnancy for small-for-gestational-age (SGA) and large-for-gestational (LGA) pregnancies in low-resource settings. OBJECTIVES: To compare stillbirth risk across pregnancy between SGA and appropriate-for-gestational-age (AGA) pregnancies in Western Cape Province, South Africa (SA). METHODS: A retrospective audit of perinatal mortality data using data from the SA Perinatal Problem Identification Program was conducted. All audited stillbirths with information on size for gestational age (N=677) in the Western Cape between October 2013 and August 2015 were included in the study. The Western Cape has antenatal care (ANC) appointments at booking and at 20, 26, 32, 34, 36, 38 and 41 (if required) weeks' gestation. A fetuses-at-risk approach was adopted to examine stillbirth risk (28 - 42 weeks' gestation, ≥1 000 g) across gestation by size for gestational age (SGA <10th centile Theron growth curves, LGA >90th centile). Stillbirth risk was compared between SGA/LGA and AGA pregnancies. RESULTS: SGA pregnancies were at an increased risk of stillbirth compared with AGA pregnancies between 30 and 40 weeks' gestation, with the relative risk (RR) ranging from 3.5 (95% confidence interval (CI) 1.6 - 7.6) at 30 weeks' gestation to 15.3 (95% CI 8.8 - 26.4) at 33 weeks' gestation (p<0.001). The risk for LGA babies increased by at least 3.5-fold in the later stages of pregnancy (from 37 weeks) (p<0.001). At 38  weeks, the greatest increased risk was seen for LGA pregnancies (RR 6.6, 95% CI 3.1 - 14.2; p<0.001). CONCLUSIONS: There is an increased risk of stillbirth for SGA pregnancies, specifically between 33 and 40 weeks' gestation, despite fortnightly ANC visits during this time. LGA pregnancies are at an increased risk of stillbirth after 37 weeks' gestation. This high-risk period highlights potential issues with the detection of fetuses at risk of stillbirth even when ANC is frequent.

INTERGROWTH-21st v. local South African growth standards (Theron-Thompson) for identification of small-for-gestational-age fetuses in stillbirths: A closer look at variation across pregnancy.

BACKGROUND: Global growth standards for fetuses were recently developed (INTERGROWTH-21st). It has been advocated that professional bodies should adopt these global standards. OBJECTIVES: To compare the ability of INTERGROWTH-21st with local standards (Theron-Thompson) to identify small-for-gestational-age (SGA) fetuses in stillbirths in the South African (SA) setting. METHODS: Stillbirths across SA were investigated (>500 g, 28 - 40 weeks) between October 2013 and December 2016 (N=14 776). The study applied the INTERGROWTH-21st standards to classify stillbirths as <10th centile (SGA) compared with Theron-Thompson growth charts, across pregnancy overall and at specific gestational ages. RESULTS: The prevalence of SGA was estimated at 32.2% and 31.1% by INTERGROWTH-21st and Theron-Thompson, respectively. INTERGROWTH-21st captured 13.8% more stillbirths as SGA in the earlier gestations (28 - 30 weeks, p<0.001), but 4.0% (n=315) fewer between 33 and 38 weeks (p<0.001). Observed agreement and the Kappa coefficient were lower at earlier gestations and at 34 - 36 weeks. CONCLUSIONS: Our findings demonstrated differences in the proportion of stillbirths considered SGA at each gestational age between the INTERGROWTH-21st and the local SA standard, which have not been considered previously by other studies.