Reporting systems, reporting rates and completeness of data reported from primary healthcare to a Swedish quality register – The National Diabetes Register

https://doi.org/10.1016/j.ijmedinf.2011.06.006Get rights and content

Abstract

Objective

The aims of this paper were to study the reporting rate and completeness of data reported from primary healthcare centres (PHCCs) in Sweden to the Swedish National Diabetes Register (NDR), with a special attention on the relation between these measures and the reporting system used by the PHCCs.

Method

A national survey conducted in Swedish primary healthcare covering the year 2006. A questionnaire was used to collect data from 523 PHCCs. Data on 87,099 adult diabetic patients attending these PHCCs and reported to the NDR were obtained from the register. In Sweden, participation in the NDR is voluntary. The data were reported through the Internet, either online using a web-based system or by direct transmission. The main outcome measures were reporting rate and completeness of reported data.

Results

Of the 523 PHCCs, almost two-thirds had reported <75% of their diabetic patients to the NDR. The lowest reporting rate was found among the largest PHCCs, while the highest was found among small PHCCs (p < 0.001). Reasons given for not reporting data to the NDR were lack of time and lack of personnel resources. Altogether, 73.1% of the PHCCs reported data to the NDR online using a web-based system, 20.5% used direct transmission and 6.3% used both systems. The PHCCs that reported data through direct transmission systems reported almost 70% of their diabetic patients to the NDR, while PHCCs using web-based systems reported 54% of their diabetic patients to the NDR. Adjusted for other factors, using direct transmission increased the reporting rate by 13.0 percentage points. However, the web-based system contributed to a higher completeness of data than the direct transmission system.

Conclusions

A direct transmission system facilitates a high reporting rate to the register at the expense of lower completeness of the reported data.

Highlights

► It is shown that Swedish primary healthcare centres in general reported data to the National Diabetes Register (NDR) using a web-based system. ► Direct transmission system increased the number of patients reported to the NDR. ► Using a web-based system contributed to higher completeness of data than using direct transmission.

Introduction

Sweden has about 70 national quality registries, which mostly are web-based, i.e., data are entered into the registries using web-based questionnaires. The definition of a Swedish quality register is: “A national quality registry contains individualised data concerning patient problems, medical interventions, and outcomes after treatment; within all healthcare production. It is annually monitored and approved for financial support by an Executive Committee” [1]. One of these 70 quality registries is the Swedish National Diabetes Register (NDR). This quality register was implemented in 1996 as a response to the St. Vincent Declaration's requirement in 1989 to assess the quality in diabetes care [2].

The NDR allows primary healthcare centres (PHCCs) to measure quality based on the indicators listed in the national guidelines for diabetes care and make national comparisons [3], [4], [5]. The PHCCs can follow their quality work regularly and analyse their data through immediate access to their own results in haemoglobin A1c (HbA1c), body mass index, lipids, blood pressure, preventive treatment, various risk factors, diabetes complications and physical activity [6], and with national benchmarks always being updated. The analyses of the data at the PHCCs can be performed using a web-based interface or a stand-alone program, depending on the reporting system used. This allows the PHCCs to continuously monitor changes or improvements made at their PHCC [7].

Since 2003 PHCCs have had the opportunity to be supported in their quality work in diabetes care by participating in the NDR quality work project (NDR-IQ), consisting of education for the PHCC teams and follow-up of changes in quality indicators [7]. This kind of quality improvement project has been found to influence the effectiveness of feedback [8]. Further, the NDR produces an annual report for each PHCC with comparative national statistics as well as an annual report for the national survey [7]. All county councils have appointed co-ordinators that can order regional statistics for their respective county council. The NDR invites these co-ordinators to an annual meeting to inform them and allow them to discuss news, problems and the latest results.

In 2006, ten years after the implementation of the NDR, almost 80% of PHCCs reported data on their adult patients with diabetes to the NDR, but only one-third of patients were reported from the PHCCs [9]. The fact that two-thirds of patients with diabetes were not reported to the register gave rise to the question: What factors facilitate the reporting of patients to a diabetes quality register such as the NDR? One hypothesis was that the reporting system used by the PHCCs for reporting data to the NDR may be a key factor for answering this question.

Section snippets

Aims

The aims of this paper were to study the reporting rate and completeness of data reported from PHCCs in Sweden to the Swedish National Diabetes Register (NDR), with a special attention on the relation between these measures and the reporting system used by the PHCCs.

Material and methods

This paper reports on part of a larger study – the Swedish national survey of the quality and organization of diabetes care in primary healthcare (Swed-QOP) [10]. The data are based on samples from Swedish PHCCs in 2006. A questionnaire (the Swed-QOP questionnaire) was used to collect data from the PHCCs. The patients’ data were obtained from the NDR.

Reporting rate and completeness of reported data to the NDR

Of the 523 PHCCs, 70 (13.4%) had reported less than 25% of their patients with diabetes to the NDR, 111 (21.2%) had reported 25–49.9%, 184 (35.2%) had reported 50–74.9%, and 158 (30.2%) had reported ≥75%. Altogether, 73.1% of the PHCCs reported data to the NDR online using a web-based system, 20.5% used direct transmission and 6.3% used both systems.

Statistically significant differences in reporting rate to the NDR were found between PHCCs with different list sizes (p < 0.001) and between PHCCs

Discussion

More than two-thirds of the 523 PHCCs reported <75% of their patients to the NDR. It should be noted, however, that 144 of the 684 PHCCs (21.1%) that responded to the Swed-QOP questionnaire had not reported any patients to the NDR. Further, the reporting rate to the NDR among the 237 of 921 PHCCs (25.7%) who did not respond to the Swed-QOP questionnaire was presumably lower than among those who responded. The overall reporting rate to the NDR was thus probably considerably lower. The fact that

Authors’ contributions

Eva Thors Adolfsson and Andreas Rosenblad were involved in the conception and design of the study. Eva Thors Adolfsson has collected the data. Andreas Rosenblad did the statistical analyses. Both authors have been involved in the interpretation of the data and drafting of the article. Both authors approve the submission.

Conflict of interests

Neither author has any conflicts of interest to declare.

Summary points

What was already known on the topic?

  • Sweden has about 70 quality registries. Most of them are web-based.

  • New web-based quality registries are continuously being introduced in the Swedish primary healthcare.

  • High completeness is necessary to be able to draw conclusions from data reported to quality registries.

What this study added to our knowledge?

  • Primary healthcare centres in general reported data to the Swedish National

Acknowledgement

The study was supported by grants from the Medical Faculty, Uppsala University.

References (20)

There are more references available in the full text version of this article.

Cited by (15)

  • Associations between quality of work features in primary health care and glycaemic control in people with Type 2 diabetes mellitus: A nationwide survey

    2019, Primary Care Diabetes
    Citation Excerpt :

    Clinical data are reported at least once a year from medical check-ups by general practitioners (GPs) or registered nurses (RNs). Individual clinical data are reported continuously either online or by electronic transmissions from medical charts [22]. Each patient provides informed consent [21].

  • Low fructosamine and mortality – A long term follow-up of 215,011 non-diabetic subjects in the Swedish AMORIS study

    2016, Nutrition, Metabolism and Cardiovascular Diseases
    Citation Excerpt :

    Diabetes at baseline was defined as meeting at least one of the three following criteria: 1) a diabetes diagnosis prior to baseline in the Swedish National Diabetes Register (NDR) [20,21]. In the NDR, the estimated coverage of subjects with diabetes is around 80% [20,21]; 2) a fasting glucose value ≥7.0 mmol/L or, an HbA1c ≥ 6.5% at the baseline examination; 3) a hospital discharge diagnosis of diabetes prior to the baseline examination recorded in the National Patient Register. A non-diabetic subject at baseline fulfilled none of these criteria.

  • Assessing the quality of routine data for the prevention of mother-to-child transmission of HIV: An analytical observational study in two health districts with high HIV prevalence in South Africa

    2016, International Journal of Medical Informatics
    Citation Excerpt :

    Training is not adequately provided for clinic staff involved in the data collection processes [23–26], who often have very limited data quality checking skills and do not understand the importance of collecting data [27]. Other reasons for poor quality data include design/structural issues relating to inappropriate data collection tools and procedures [13,28], poor recording due to inadequate skills [13,24–27,29], inadequate resources [30] leading to poor capturing and reporting of data [24,25,31] errors in processing data [32], too many data elements [33], staff attrition [13,25,29], and lack of use of already generated data [24], which may hinder constructive feedback to data producers. Feedback is a crucial part of the supervisory process, and is critical for enhancing data quality, especially when audits are done [22,25,32].

  • Validity of reported data in the Swedish Cardiac Arrest Register in selected parts in Sweden

    2013, Resuscitation
    Citation Excerpt :

    Previous experience, based on 45% of the Swedish population, indicates that between 15% and 30% of OHCA cases are not reported to the register by the EMS crews.1 Moreover, other data show that a high level of individual completeness is essential before conclusions can be drawn.17 Some studies describe the quandary created by not having a knowledge of the OHCA incidence.6,18

View all citing articles on Scopus
View full text