What is a study quality assurance system and why was it required in the NIDIAG persistent fever syndrome study?

The quality assurance system of a clinical study usually includes the development and implementation of SOPs along with staff training and internal and external quality control (QC) activities [8]. SOPs are powerful tools to enhance the quality of clinical research activities through harmonization and standardization. By providing detailed instructions and guidance to operators performing a given task, SOPs enable researchers to achieve uniformity in the performance of this task across different sites and for the total duration of the study. SOPs also help ensure that the study is implemented according to what is set forth in the protocol and that measures to control for bias are in place. This in turn ensures that robust, accurate, and reproducible study data are generated, and that reliable conclusions are drawn from studies. Well-designed SOPs not only warrant quality and integrity of study data, but they also leave a permanent record of the methods employed and help provide public assurance that ethical standards are met and patients’ safety is safeguarded. Last but not least, SOPs help ensure compliance with regulations and international standards. According to GCP guidelines, sponsors are responsible for implementing and maintaining appropriate quality assurance and QC systems, including SOPs for all research-related activities. Many clinical studies develop their own set of SOPs, and some research groups have made them public [9–11].

Training is another essential pillar of any study quality assurance system. Training of staff involved in clinical research is not only an ethical mandate, but it is also key to a successful study. Although often seen as a one-time activity that takes place before study start, training is actually an ongoing pursuit. Staff turnover, protocol amendments, changes in disease diagnosis and management recommendations, or detection of protocol deviations all prompt additional or refresh trainings. As far as possible, training should be tailored to the specific objectives of the study and should be articulated around the study SOPs. Whenever possible, dummy runs should be organized as part of the initial study training to verify that everything works in order before the first patient is included.

QC encompasses a set of internal and external activities that are crucial to verify that the research is being conducted in accordance with initial plans and with applicable standards and regulatory requirements. Internal QC is continuous and concurrent with the conduct of the clinical study and is carried out by the research staff. Internal QC includes, for example, the definition of reference standards and calibrators for laboratory measurements and the evaluation of day-to-day conformity with these standards. External QC, on the other hand, is conducted by independent organizations and is often linked to accreditation processes. It usually includes monitoring, which takes place at regular, predefined intervals, and external audits.

The six study sites participating in the NIDIAG persistent fever syndrome study showed a high heterogeneity with regards to epidemiological context (i.e., prevalence of target conditions), diagnostic-treatment algorithms in use, existing disease control activities, laboratory capacity and health staff knowledge, and experience in clinical research. Hence, harmonizing study activities throughout study sites was paramount. In addition, the high number of laboratory procedures, some of which were introduced for the purpose of the study in local laboratories, necessitated detailed guidance, extensive training and coaching of laboratory staff, and a robust QC framework.

How was the NIDIAG persistent fever syndrome study conducted: how were patients recruited and followed-up, and what diagnostic work-up was performed?

The NIDIAG persistent fever syndrome study targeted all patients presenting with more than one week of fever at one of the study sites during the study period. In all sites, patients aged five years and above were included, except in Cambodia, where only adults were included. Patients with an existing laboratory-confirmed diagnosis (including diseases not targeted by NIDIAG) and patients in need of immediate intensive care due to shock or respiratory distress were excluded from the study. Study investigators assessed inclusion and exclusion criteria through history taking, review of existing medical documents, and rapid clinical assessment.

After providing informed consent, patients underwent a full clinical assessment as well as blood and urine sampling, on which a standard, country-specific panel of ancillary, RDT, and reference laboratory testing was performed (see Table 1). The health care provider remained free to order other optional diagnostic procedures (e.g., chest X-ray, abdominal ultrasound). Patients were followed-up daily during their hospital stay or at each outpatient visit for those who were not hospitalized. A single systematic follow-up visit was performed at four weeks post discharge to assess clinical response to treatment and to collect additional serum samples. Additional follow-up visits were scheduled at the discretion of the care provider.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Diagnostic plan for the NIDIAG persistent fever syndrome study.

https://doi.org/10.1371/journal.pntd.0004749.t001

All new or persistent clinical symptoms and signs were recorded, together with new therapies and treatment modifications. After completing the first-line diagnostic work-up, patients for whom no confirmed or presumptive diagnosis could be established underwent further testing, including chest X-ray, abdominal ultrasound, sputum collection, lymph node puncture, bone marrow aspiration, and lumbar puncture (if indicated by the appearance of a new clinical feature or nonresponse to initial treatment). Results of this second-line diagnostic work-up were also recorded.

What SOPs were developed for the NIDIAG persistent fever syndrome study?

A set of 46 SOPs were developed specifically for the persistent fever syndrome study (see S1 Table and S1 Text). Those included clinical SOPs (4), laboratory SOPs (29), data management SOPs (2), and QC SOPs (11). Besides these, a few additional SOPs had to be developed for and adapted to local circumstances (e.g., SOP on study samples transport or SOP on inoculating and growing bacterial cultures). Those SOPs were meant to be used by different target groups: clinicians and health care workers, laboratory staff, data entry clerks and data managers, and staff involved in QC activities. SOPs were made available in both French and English, depending on whether they were meant for French- or English-speaking staff. All SOPs used in the persistent fever syndrome study can be found in annex, except site-specific SOPs and SOPs in French, which are available upon request.

All NIDIAG SOPs were developed based on a template and following a SOP for developing SOP to ensure consistency in among all different SOPs developed. A specific numbering system was used, and each SOP received a unique identification number. The tasks for SOP development were shared in the research group: principal investigators, laboratory referents, and QC experts contributed as first authors for specific SOPs. Each draft SOP was then reviewed by a second author who was familiar with the procedure described. Input from study site teams was sought through specific training workshops and piloting of the SOPs. Finally, all SOPs and their annexes were reviewed and approved by the GC(L)P expert who was responsible for ensuring consistency with other documentations and maintaining version control. All approved SOPs were published as PDF files on the NIDIAG website and were made available for study sites to download. Study personnel were informed of SOPs revisions, and additional training was organized when necessary.

What challenges did study sites meet in conducting the study and complying with the SOPs?

Between January 2013 and October 2014, a total of 3,409 patients were screened at the 6 study sites (427 in Cambodia, 856 in Sudan, 646 in Nepal, and 1,480 in DRC), of whom 1,927 were included in the NIDIAG persistent fever syndrome study (378 in Cambodia, 670 in Sudan, 578 in Nepal, and 301 in DRC). Eighty-six (4.5%) patients died, and 454 (23.5%) were lost to follow-up (LFU) during the study. After the last patient’s last visit, a short survey was conducted among research staff about the challenges faced during the study. The outcomes of this survey are shown in Tables 2 and 3. Most challenges highlighted by the research teams pertained to logistical and organizational aspects; however, compliance with quality standards and adherence to SOPs was also listed by laboratory staff. This perceived difficulty could be related to the high number of SOPs implemented or to the relative inexperience of some teams in conducting GC(L)P-compliant studies. Recruitment and retention of study participants was also among the main challenges faced by study sites. This is in line with published research, which shows that recruitment and retention of research participants is among the top challenges of clinical research today. According to a recent report, 89% of 150 clinical studies assessed met the recruitment target, but time for doing so was generally double what was forecasted in the study protocol [12].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Challenges encountered by the laboratory staff during the NIDIAG persistent fever syndrome study.

https://doi.org/10.1371/journal.pntd.0004749.t002

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Challenges encountered by the clinical staff during the NIDIAG persistent fever syndrome study.

https://doi.org/10.1371/journal.pntd.0004749.t003

QC activities conducted during the NIDIAG persistent fever syndrome study did not identify significant deviations from the protocol and SOPs, and compliance was found very good overall, suggesting that most end users found the SOPs valuable and convenient to use. In conclusion, although labor-intensive, the development of SOPs for the febrile syndrome study proved extremely useful to ensure consistency between the different study sites and to warrant the quality of the data generated.

Supporting Information