Poor knowledge and misunderstandings: perinatal data validity and work place change in midwifery

Data accuracy in the Victorian Perinatal Data Collection: Results of a validation study of 2011 data

Objective:

Public health data sets such as the Victorian Perinatal Data Collection (VPDC) provide an important source for health planning, monitoring, policy, research and reporting purposes. Data quality is paramount, requiring periodic assessment of data accuracy. This article describes the conduct and findings of a validation study of data on births in 2011 extracted from the VPDC.

Method:

Data from a random sample of one percent of births in Victoria in 2011 were extracted from original medical records at the birth hospital and compared with data held in the VPDC. Accuracy was determined for 93 variables. Sensitivity, specificity, positive predictive value and negative predictive value were calculated for dichotomous items.

Results:

Accuracy of 17 data items was 99% or more, the majority being neonatal and intrapartum items, and 95% or more for 46 items. Episodes of care with the highest proportion of items with accuracy of 95% or more were neonatal and postnatal items at 80 and 64%, respectively. Accuracy was below 80% for nine items introduced in 2009. Agreement between medical records and VPDC data ranged from 48% to 100%, the exception being two highly inaccurate smoking-related items. Reasons for discrepancies between VPDC data and medical records included miscoding, missing and inconsistent information.

Conclusion:

This study found high levels of accuracy for data reported to the VPDC for births in 2011; however, some data items introduced in 2009 and not previously validated were less accurate. Data may be used with confidence overall and with awareness of limitations for some new items.

Perinatal data collection: current practice in the Australian nursing and midwifery healthcare context

The collection of perinatal data within Queensland, Australia, has traditionally been achieved via a paper form completed by midwives after each birth. Recently, with an increase in the use of e-health systems in healthcare, perinatal data collection has migrated to an online system. It is suggested that this move from paper to an ehealth platform has resulted in improvement to error rates, completion levels, timeliness of data transfer from healthcare institutions to the perinatal data collection and subsequent publication of data items. Worldwide, perinatal data are collected utilising a variety of methods, but essentially data are used for similar purposes: to monitor outcome patterns within obstetrics and midwifery. This paper discusses current practice in relation to perinatal data collection worldwide and within Australia, with a specific focus on Queensland, highlights relevant issues for midwives, and points to the need for further research into the efficient use of an e-health platform for perinatal data collection.

Precancerous changes in the cervix and risk of subsequent preterm birth

OBJECTIVE

The aims of this study were (i) to examine whether women referred for assessment of precancerous changes in the cervix had higher rates of preterm birth compared with those in the general population and (ii) to compare preterm birth rates for treated and untreated women adjusting for possible confounding factors.

DESIGN

Retrospective cohort design.

SETTING

Teaching hospital.

POPULATION

All women referred to the Royal Women's Hospital, Melbourne (1982-2000), who subsequently had a birth recorded on the Victorian Perinatal Data Collection system (n = 5548).

METHODS

Record linkage of hospital dysplasia clinic records and population-based birth records.

MAIN OUTCOME MEASURES

Total preterm delivery (<37 weeks of gestation) and subtypes.

RESULTS

Both treated and untreated women were at a significantly increased risk for preterm birth compared with those in the general population: treated--standardised prevalence ratio (SPR) 2.0, 95% CI 1.8-2.3 and untreated--SPR 1.5, 95% CI 1.4-1.7. Within the cohort, the treated women were significantly more likely to give birth preterm (adjusted OR 1.23, 95% CI 1.01-1.51). An increased risk of preterm birth was also associated with a history of induced or spontaneous abortions, illicit drug use during pregnancy or a major maternal medical condition. Cone biopsy, loop electrosurgical excision procedure and diathermy were associated with preterm birth. After adjusting for possible confounding factors, only diathermy remained significant (adjusted OR 1.72, 95% CI 1.36-2.17). Women treated using laser ablation were not at an increased risk for preterm birth (adjusted OR 1.1, 95% CI 0.8-1.4).

CONCLUSIONS

Diagnosis of precancerous changes in the cervix (regardless of the treatment) was associated with an increased risk of preterm birth. Consideration should be given to the preferential use of ablative treatments.

Validation of the Victorian Perinatal Morbidity Statistics form: new items, pre-coded text and free text

This study assessed the quality of information on the form used by the Victorian Perinatal Data Collection Unit. Information relating to four new and one modified data items, and to pre-coded and free text regions relating to maternal morbidity, was compared with that in the medical record. The new data items were documented correctly on 95.5% of forms, as were 67% of pre-coded conditions. Between 53% and 75% of free text items were correctly recorded. The results highlight the need for information and education sessions for midwives and other staff in order to collect and maintain high quality data.

The Perinatal Database of the Netherlands

OBJECTIVE

In the Netherlands, the Perinatal Database of the Netherlands (Landelijke Verloskunderegistratie, LVR) was set up in 1982 for secondary care obstetric departments on a voluntary participation basis, its main goal being quality monitoring. At the outset of the database, 70% of Dutch obstetric departments participated immediately. This percentage has now increased to almost 100%. How the LVR was set up, its aims, participation, general function and some aspects of the reliability are described in this article.

RESULTS

Assessment of the reliability of the data on a number of points found that the frequency of various types of nearly all errors has decreased during the existence of the LVR, with the exception of a complete registration of the first-week mortality.

CONCLUSIONS

The quality of the data has improved over the years. More reliable figures for the first-week mortality would be available if the LVR data were linked to those of the Neonatal Database of the Netherlands (Landelijk Neonatale Registratie, LNR), so that the condition of each child is known up to and including 28 days after birth. Improvements in the reliability of other items can be attained by implementing more checks during the data input stage and by making the data more accessible to the gynaecologists, in order to further increase motivation to record the data.

The accuracy of eclampsia cases reported to the Victorian Inpatient Minimum Database and the Perinatal Data Collection Unit

The aim of this study was to compare the quality of the reporting and coding of eclampsia in two routine data collections: the Victorian Inpatient Minimum Database (VIMD) and the Perinatal Data Collection Unit (PDCU). The validity of cases in the two data sets was confirmed by reference to the original medical record data. Only 12 cases were the same in both data sets (i.e., 35.3% agreement). There were an additional 51 cases that were reported to either one or the other of the data sets and, of these, only 15 (i.e., 29%) were confirmed as eclampsia. The overall number of cases confirmed for both systems in 1995 was 27, or 0.4 per 1000 confinements. Reasons for these discrepancies were investigated and three basic problems identified: quality of documentation in the medical record, coding errors, and use of data from computer-generated forms. Neither the VIMD nor the PDCU was regarded as having sufficiently accurate data for adequate reporting of maternal morbidity. By combining the information from both databases a better estimate of incidence can be obtained, but improved reporting and coding is essential for accurate assessment of this condition.