Evaluation of 17 years of MERIN (Meningitis and Encephalitis register in Lower Saxony, Germany) surveillance system: participants acceptability survey, completeness and timeliness of data

Monitoring cases of acute flaccid paralysis (AFP) among children under 15 years of age is considered gold standard for polio surveillance. It is used in many countries to demonstrate the presence or absence of poliovirus circulation and is one element in the certification process of a country towards polio-free status [1, 2]. However, countries that have been polio free for many years or even decades are challenged to maintain the surveillance’s sensitivity criteria of detecting at least one non-polio AFP case per 100 000 children under the age of 15 years per year, required by the Polio Eradication Initiative [1, 3, 4].

The national AFP surveillance in Germany was established in 1998 as a way of demonstrating polio-free status. However, the results of the surveillance proved to be insufficient according to AFP sensitivity criteria [5]. In Germany only 50 to 60% of the expected AFP cases were reported from the clinics (approx. 60 to 70 of 116 AFP cases) [6]. Therefore, an alternative approach was necessary to increase acceptability of the surveillance among clinics and to fulfil polio surveillance requirements.

In 2003 Lower Saxony was the first federal state in Germany to implement an alternative to AFP surveillance system focussing on patients with suspected aseptic meningitis, encephalitis or acute flaccid paralysis – the Meningitis and Encephalitis Register in Lower Saxony (MERIN) hosted by the Public Health Agency of Lower Saxony (NLGA) [6, 7]. In 2010 MERIN became the model for the federal enterovirus surveillance system, which replaced AFP surveillance to demonstrate polio free status of Germany. Hospitals from the federal state of Bremen started joining MERIN in 2011. MERIN offers laboratory diagnostic testing for hospitalized patients with suspected aseptic meningitis or encephalitis. Participating hospitals, mostly paediatric, neurological and internal medicine clinics in Lower Saxony and Bremen, send specimens such as viral swabs, serum samples and cerebrospinal fluid to the virological laboratory of NLGA and receive laboratory diagnostics free-of charge. The medical reports are delivered by post mail or fax. Routine testing is performed for enteroviruses (EV), including polioviruses, but also for e.g. herpes simplex, varicella-zoster, measles, mumps and tick-borne encephalitis viruses via direct pathogen detection (PCR and cell culture) as well as serological analyses. This broad diagnostic coverage was chosen to enhance engagement of clinicians and their participation in the MERIN and thereby increase its acceptability. During the last 10 years (2013–2022), on average 1800 samples from over 700 patients were sent annually by 39 different hospitals. The majority (79%) of the patients were younger than 15 years. More than a half (56%) of meningitis and encephalitis cases were caused by EV. Polioviruses were not detected [8]. Aside from being a stand-alone surveillance system in two federal states, MERIN is also the major contributor to the national German enterovirus surveillance [5].

Evaluation of surveillance systems is essential to improve their performance and should be conducted on a regular basis. The complexity and the different settings in which they are implemented requires surveillance systems to be targeted and context relevant [9, 10]. Since MERIN is based on voluntary contributions of the participating hospitals, our evaluation focusses on acceptability among participants.

This evaluation aims to better understand the surveillance system and its challenges, views and perceptions of the hospitals currently involved in MERIN, with the outlook to identify areas for improvement and formulate and apply specific recommendations.

Seven hospitals were excluded as they had sent less than one sample during the last two years prior to the survey, resulting in 32 actively participating hospitals (29 from Lower Saxony and 3 from Bremen). We received questionnaires from 21 of the 32 hospitals invited to participate. This corresponds to a response rate of 66%. All 21 hospitals were from Lower Saxony. Six hospitals contributed more than one questionnaire. Of these six, one hospital contributed four questionnaires (three from the pediatric ward and one from the neurological ward), one hospital contributed three questionnaires from the same ward and the other four hospitals contributed two questionnaires each from the same ward. A total of 30 questionnaires were used for the final evaluation, of which 25 came from pediatric and 5 from neurological departments, 27 (93%) were answered by chief or senior physicians (Table 2).

Figure 1 summarizes the results of participants’ satisfaction with communication, timeliness of sample processing, workload, data collection forms and delivery of medical reports. Most of the respondents rated communication with the NLGA positively and were very satisfied with reachability (96%, 25/26) and competence (100%, 26/26) but also with friendliness (100%, 26/26) of the NLGA personnel.

The processing time for direct pathogen detection and for serological pathogen detection was found satisfactory by 81% (22/27) and 89% (24/27) of participants respectively.

The content and formal design of the submission form and medical reports were perceived as clearly structured and easy to understand. The delivery of medical reports by fax was satisfactory for 86% (25/27) of the participants, mailing of medical reports per post for almost 80% of them (23/27). A common issue mentioned by several respondents in the free-text fields was digitalisation. Specifically, the need for electronic transmission of the results was emphasized. Participants also indicated they would appreciate receiving regular reports on circulating EV serotypes.

The workload required for participating in MERIN was rated as appropriate or absolutely appropriate by 97% (28/29) of the participants.

Regarding the correlations, workload perceived as appropriate correlated significantly with satisfaction regarding all investigated aspects of communication: reachability (ρ = 0.67, p = 0.001), competence (ρ = 0.71, p = 0.002) and friendliness (ρ = 0.73, p = 0.001).

Regarding the relevance of the MERIN objectives, rapid and reliable diagnostic testing was given the highest priority (93%, 26/28), followed by pathogen spectrum of viral CNS infections (90%, 26/29), currently circulating enterovirus serotypes (86%, 25/29), rapid outbreak management (79%, 23/29) and detection of regional clusters (79%, 23/29). Monitoring of polio-free status in Germany was rated the lowest priority, still 61% (17/29) of the participants considered it an important or a very important objective (Fig. 2).

About half (57%; 17/28) of all participants stated that 100% of all their meningitis/encephalitis patients were included in the MERIN. The median proportion of patients included was 75%. This parameter correlated with the level of satisfaction regarding mailing (ρ = 0.8, p < 0.001) or faxing (ρ = 0.76, p = 0.001) the medical reports (Fig. 3).

The reasons most frequently given for not including all suspected cases of aseptic meningitis/encephalitis in the MERIN were either that the laboratory diagnostics was done by another laboratory (64%; 7/11) and “staff forgot/was unaware” (55%, 6/11). The statement “initial clinical picture inconclusive” was selected four times (36%).

The majority (93%, 26/28) of the participants found the offered scope of laboratory diagnostics sufficient. The analysis of two additional viral pathogens for the differential diagnosis of meningitis/encephalitis (“bocaviruses” and “HHV 6”) was suggested in the open field textbox.

Overall, identified potentials for improvement involved online access of medical records and regular reports on currently circulating enterovirus serotypes.

Complementing the survey results, analysis of data completeness of the variables on the submission form showed consistently high values over the years, ranging between 80 and 99,9% across data fields in 2019 (n = 748). Reporting rate of follow-up forms was 58% (Additional file 2).

The calculated median time between sample collection and arrival at the laboratory was 2 days [IQR 1–3], for EV diagnostics via PCR 1 day [IQR 0–6] and for EV isolation on cell culture 11 days [IQR 10–13]. Participants who were satisfied with the processing time were also satisfied with delivery of medical reports via fax (ρ = 0.73, p = 0.003) (Fig. 3).

Efficiency of any surveillance highly depends on stakeholders’ engagement and participation [10]. Therefore, assessing their willingness to participate is crucial, and can contribute to reduced underreporting, improvement of data quality and also identification of clues for system improvement [10, 12]. As highlighted in a systematic review of the current approaches to evaluate surveillance systems [10], the attribute of acceptability should be viewed as an essential factor for the quality of surveillance. It has also been listed by Center for Disease Control and Prevention as one of the main requirements for proper functioning of the system [9].

Our findings demonstrate that MERIN is a successful model for a voluntary-based surveillance. Our investigation revealed a high level of system acceptability as evidenced by participants’ satisfaction with the system’s objectives, their relationship with the laboratory, the timeliness and modes of processes involved as well as their assigned duties related to MERIN. The high participants’ satisfaction was underlined by findings regarding timeliness and completeness of data (age, gender, zip code).

The study showed that a majority of the respondents were highly satisfied with the communication with the NLGA. All investigated aspects of communication (reachability, competency and friendliness) strongly correlated with the MERIN related workload perceived as appropriate. This finding can be seen as an indicator of the good relationship between the participants and the NLGA laboratory, a prerequisite, in our view, for the success of voluntary surveillance of non-mandatory notifiable diseases.

Regarding timeliness and modes of processes, most survey participants indicated that the data collection forms were clearly structured and easy to understand, participants also appreciated the timeliness and wide range of the provided diagnostic testing. Potentials for improvement were identified particularly with regard to electronic transmission of the results and direct and regular access to the surveillance reports. Moreover, satisfaction with medical reports delivery via post mail and fax highly correlated with the proportion of meningitis/encephalitis patients included in MERIN. This finding may suggest that hospitals that are satisfied with the way of delivery are more likely to include all of their eligible patients to MERIN. As a consequence, providing digital medical reports could increase the number of patients submitted.

Regarding surveillance objectives, participating hospitals valued rapid and reliable diagnostics highest, while monitoring polio free status in Germany was rated the lowest. Prompt laboratory diagnosis of an EV infection may reduce antibiotic or antiviral drug usage, limit unnecessary and costly investigations, shorten the length of patient hospitalization and minimize the risk of complications [14‐16]. Our findings showed that participants’ needs and expectations relating to system objectives differ from those of the surveillance provider (NLGA). This was demonstrated by an almost reverse pattern of prioritization of the surveillance objectives between the two partners. The acknowledgment of the importance and consideration of the participant’s needs seems to be a critical step towards enhancing the effectiveness of surveillance efforts. Relatedly, understanding the relevance of stakeholder’s perceptions to ensure proper functioning of the surveillance system has been highlighted in previous research [10, 17, 18].

The participants’ input on surveillance system performance does not only foster its improvement, their engagement can also make it more acceptable. The insights obtained from the survey participants contributed to enhancing the MERIN with regular reports on circulating EV serotypes. Likewise, updating the current website was initiated. Participating hospitals were each provided with a written report, summarizing the overall survey results and implemented improvement measures. Lack of electronic reporting was identified as a considerable challenge. In order to retain participants and to expand the system, efforts to simplify access to laboratory results digitally should continue. Over the years, the diagnostic offer has been constantly advanced and adapted in order to accommodate new diagnostic or technological advances. Further improvements will include implementation of molecular typing of EV via next-generation sequencing.

A limitation of our study is that the hospitals that stopped participating in MERIN were not contacted. Identifying the reasons for discontinuing participation and understanding the potential barriers could help identify factors that could enhance hospitals’ involvement in the MERIN. Although a relatively high response rate [19] among the participating hospitals was found in our study, the reasons for non-participation were not further explored. In particular, identifying the reasons for non-participation among the three hospitals in Bremen might be of importance. Lacking representation of this federal state might have introduced non-response bias. In addition, the respondents consisted mainly of experienced medical personnel. This group might not be involved in daily work activities and might have a different view-point than other hospital personnel. Our study could also suffer from social desirability bias since the laboratory diagnostics are offered free of charge. However, it has been shown that social desirability bias is less likely to occur with online surveys than other data collection methods like face-to-face interviews or telephone surveys [20]. Another limitation of our study, common for cross-sectional surveys, is that it offers a snapshot taken at a point in time and does not allow for monitoring of changes. The latter will be addressed by repeated surveys in regular intervals in the future. Elucidating the hospitals’ motivations to leave MERIN retrospectively is challenged by the difficulty of finding staff that was entrusted with the decision at the time. The questionnaire in the current study was specifically designed for active participants, however, more research is needed including follow-up studies of why certain hospitals discontinued MERIN participation as well as non-respondents to the survey to better understand challenges and barriers for participation.

In spite of the listed limitations, our study clearly demonstrates that MERIN is a highly accepted surveillance system. This approach can be transferred to similar projects and can be used for implementation in other settings or to surveillance systems of other non-mandatory reported diseases based on voluntary stakeholder contribution. EV surveillance differs substantially between individual European countries but typically focuses on enteroviruses detected from patients with severe infections presenting with neurological symptoms [21]. Since the notification of EV is non-mandatory in most countries in the EU, including Germany, their surveillance relies on voluntary reporting of clinicians, hospitals or laboratories [5, 21]. Efforts to harmonize and collate surveillance data on the European level have recently been established through the European Non-Poliovirus Enterovirus Network (ENPEN) [21‐23].

MERIN was evaluated as a highly acceptable surveillance system. Identified areas of improvement contributed to enhancing the system with regular reports on circulating EV serotypes and efforts to simplify access to laboratory results. Our results emphasise the importance of recognition and consideration of participant motivations and expectations, provision of high-quality services and regular and effective communication between the partners. With the obtained knowledge, we aim to increase surveillance structure, raise awareness about MERIN among eligible hospitals in Lower Saxony and Bremen and increase participation rate which in consequence could contribute to increase performance of the system. This approach can be easily applied to other settings or to surveillance systems of other non-mandatory notifiable diseases that wish to strengthen the acceptability of their surveillance systems.