Strategies for Improving Influenza Vaccination Rates in Patients with Chronic Renal Disease
Results from two randomized controlled trials and a prospective interventional study
; ; ; ; ; ; ; ; ; ; ; ;
Background: The influenza vaccination rate among older and chronically ill patients in Germany has declined in the past decade in spite of vaccination campaigns.
Methods: The influenza vaccination rate among persons with chronic renal disease was studied with the aid of billing data from various Associations of Statutory Health Insurance Physicians (Kassenärztliche Vereinigungen, ASHIPs) in Germany. It was tested in a randomized controlled trial whether a written vaccination appeal, sent by physicians to patients, led to an increase in the vaccination rate. It was tested in a further such trial whether the vaccination rate among patients with renal disease could be improved by an appeal for vaccination that was sent by the ASHIPs to the treating nephrologists. Finally, it was also tested in a prospective interventional study whether the vaccination rate could be improved by an appeal for vaccination sent by a health-insurance carrier directly to the patients.
Results: In 2012–2017, the vaccination rate among persons with chronically impaired renal function ranged from 41.1% to 46.9%; it ranged from 31.7% to 33.7% in kidney transplant recipients and from 42.7% to 44.7% in dialysis patients. An appeal for vaccination that was sent from physicians to patients raised the vaccination rate by 8.3% in the intervention group compared to the control group (p = 0.03; number needed to treat [NNT]: 13). On the other hand, an appeal for vaccination that was sent to the nephrologists lowered the vaccination rate by 0.8% in the intervention group compared to the control group. Finally, an appeal for vaccination that was sent by the health-insurance fund to the patients raised the vaccination rate by 3.2% (p<0.001; NNT: 32).
Conclusion: Fewer than half of all patients with chronic renal failure in Germany are vaccinated against influenza. The vaccination rate was found to be increased only after an appeal for vaccination that was sent directly to the patients. A letter sent to the treating physicians had no positive effect at all.
In the winter of 2017/2018, there were 324 416 cases of laboratory-confirmed influenza infection among residents of Germany (1). Around 20% of these persons—principally the chronically ill and those over 60 years of age—required inpatient treatment (1). These figures underline the importance of vaccinating risk groups against seasonal influenza each year, as recommended by the German Standing Vaccination Committee and numerous professional societies (2–4). The value of this recommendation was shown by the results of a recently published benefit analysis in the USA: the authors concluded that vaccination against influenza avoided more than 100 000 admissions to hospital and circa 8000 deaths in the influenza season 2017/2018 (5).
Nevertheless, the rate of vaccination of at-risk patients against influenza in Germany has decreased sharply in recent years. According to an analysis of billing data from the German Association of Statutory Health Insurance Physicians (ASHIP), the national average vaccination rate for persons over 60 was 49% in the season 2006/2007, but only 34.8% in 2016/2017 (6, 7). The situation is similar for the chronically ill, for whom the vaccination rate decreased from 46.2% in the season 2010/2011 to 40.4% in 2013/2014 (8, 9).
This negative trend is worrying, particularly in light of the numerous campaigns to increase public awareness of the risk and acceptance of vaccination in recent years. These campaigns focused predominantly on direct persuasion of high-risk patients as to the importance of vaccination. Measures were also taken to encourage primary care physicians to recommend vaccination (10).
These figures clearly illustrate the urgency of effective preventive measures to avoid future epidemics of influenza. In the study presented here, we explored potential strategies specifically for application in patients with chronic renal disease, a group for whom vaccination against influenza is clearly indicated because of the particularly high risk associated with their pronounced multimorbidity (3, 11). After initial analysis of the current vaccination rates on the basis of billing data, we posed the following questions in three independent studies:
1. Does a recommendation in writing directly from physician to patient have a positive effect on the vaccination rate in patients with chronic renal disease?
2. Can a written appeal from a regional ASHIP persuade office-based nephrologists to increase their rate of vaccination of chronically ill patients?
3. Does a direct appeal from a health insurance fund to its members lead to an increase in the vaccination rate among those with chronic renal disease?
First, the billing data of three regional ASHIPs (North Rhine, Schleswig-Holstein, and Thuringia) were analyzed to ascertain the influenza vaccination rates among patients with chronic renal disease. To this end, data from about 12 million statutory health insurees were analyzed.
To answer the questions described above, we then carried out three separate studies:
1. In a patient-centered randomized controlled intervention study, all patients who had undergone renal transplantation at the Kiel campus of the University Hospital of Schleswig-Holstein (UKSH) up to spring 2016 (n = 460) were randomized into two groups. Following determination of the patients’ vaccination status and attitude to vaccination against influenza in an initial survey, we conducted a second survey to analyze the effect of an appeal from a physician on the vaccination rate and the patients’ attitudes. The appeal came from physicians working in the renal transplantation clinic of the UKSH.
2. In a physician-centered randomized controlled intervention study, the office-based nephrologists of the North Rhine ASHIP (n = 141) were randomized into two groups. The intervention group were called upon by the ASHIP to recommend vaccination, and the effect was ascertained by evaluating the vaccination billing data for the following year.
3. In an insuree-centered prospective intervention study, the IKK Südwest health insurance fund sent a vaccination appeal directly to those of their members with chronic renal disease (n = 617). The effect of the intervention was established by analyzing the numbers of vaccinations billed.
Figures 1–3 are flow charts of the three intervention studies. The methods used are described in detail in the eMethods.
Influenza vaccination rates in the observation period 2012–2017
During the observation period the influenza vaccination rate for patients with chronic mpairment of renal function ranged from a minimum of 41.1% to a maximum of 46.9%. The vaccination rate for renal transplantation patients varied between 31.7% and 33.7%, that for dialysis patients between 42.7% and 44.7%. As well as these fluctuations over time, there were distinct differences among the ASHIP regions (Table, eTable 1). The groups differed considerably in number of physician–patient contacts. In Schleswig-Holstein, for example, the average of contacts with the patient’s primary care physician or nephrologist was 12.6 for those with chronically impaired renal function, 12.1 for renal transplantation patients, and 35.5 for patients dependent on dialysis. In 79.5% of patients with chronically impaired renal function, influenza vaccination was additionally indicated on grounds of age (>60 years). This also applied to 43.1% of patients who had undergone renal transplantation and to 66.4% of dialysis patients (eTable 2).
Findings of the patient-centered intervention study
The patients’ characteristics (age, sex, type of donation, and time since transplantation) did not differ significantly between the first and second surveys or between the control and intervention groups (eTable 3). In the intention to treat (ITT) analysis, the influenza vaccination rate was 37.4% in the intervention group and 29.1% in the control group. Compared with the first survey (vaccination rate 35.2%), the rate rose by 2.2% in the intervention group and fell by 6.1% in the control group (p = 0.03), yielding an intervention effect of 8.2% (Figure 4). Therefore 13 intervention letters (= number needed to treat, NNT) had to be sent to achieve one additional vaccination against influenza. The proportion of participants who were of the opinion that an influenza vaccination could have severe adverse effects was 7.8% in the intervention group and 6.5% in the control group, decreasing from the figure in the initial survey (11.1%) by 3.3% and 4.6% respectively (p = 0.29). Influenza was believed to be a dangerous disease by 37% of participants in the intervention group and 36.1% of those in the control group (p = 0.42). In comparison with the initial value (40.7%), these figures represented decreases of 3.7% and 4.6% respectively. The flow chart in Figure 1 shows the numbers of drop-outs after randomization. The results of per protocol analysis can be found in eTable 4.
Findings of the physician-centered intervention study
The characteristics of the office-based nephrologists did not differ between the intervention and control groups after randomization (eTable 5). There were no study drop-outs following randomization. The increase in influenza vaccination rate was lower in the intervention group than in the control group, at 0.3% (2015/2016: 39.7%, 2016/2017: 40.0%) and 1.1% (2015/2016: 42.2%, 2016/2017: 43.3%) respectively. The intervention effect was therefore –0.8% (Figure 4). Restricting analysis to patients who were vaccinated by their nephrologist, the vaccination rate fell by 1.0% in the intervention group (2015/2016: 9.6%, 2016/2017: 8.6%) and rose by 1.0% in the control group (2015/2016: 11.9%, 2016/2017: 12.9%).
Findings of the insuree-centered intervention study
In the ITT analysis, the rate of vaccination against influenza climbed from 36.3% (2015/2016) to 39.5% (2016/2017), an increase of 3.2% (NNT 32, p<0.001) (Figure 4). Including only those who lived to the end of the vaccination season (per protocol analysis), the vaccination rate rose by 3.8% from 36.3% (2015/2016) to 40.1% (2016/2017) (NNT 27; p<0.001). None of the insurees used the hotline that had been set up to answer any queries. To uncover any general changes in influenza vaccination behavior in the same vaccination season, we investigated the vaccination rate among insurees with chronic pulmonary disease (n = 79 411) and found a decrease from 13.3% to 13.2%.
The study presented here is the first to analyze the vaccination rate for chronically ill patients in Germany by means of billing data rather than telephone interviews. We found that less than half of our study population of patients with renal disease were vaccinated against influenza over the course of the observation period. Interestingly, the vaccination rate for patients with chronic impairment of renal function was only slightly lower than that for patients dependent on dialysis, although the latter had almost three times as many contacts with physicians per quarter. The vaccination rate is thus associated neither with the number of physician–patient contacts nor with the severity of the chronic illness. Rather, comparison of patients with chronically impaired renal function and those who had undergone renal transplantation shows that the vaccination rate tends to be determined by the proportion of patients for whom vaccination is also indicated on grounds of age. In daily practice, therefore, awareness of the importance of vaccination against influenza for the chronically ill seems lower than that of its significance in the elderly. This had already been indicated by previous investigations showing higher vaccination rates in chronically ill patients over 60 than in their younger counterparts (9).
In the first of the three approaches presented here, we evaluated whether an appeal in writing from a physician has the potential to increase the vaccination rate in patients with renal transplants or positively influence their attitude to influenza vaccination. The described intervention led to an increase of 8.3 percentage points in the vaccination rate compared with the control group. Thirteen intervention letters and circa €10 in printing and postage costs were thus required to achieve one additional vaccination. This finding points in the same direction as several studies from other countries that also demonstrated a positive effect of a vaccination reminder from the treating physician (12–14). Assuming, as we did, that the selected intervention has no negative effect on the vaccination rate among the patients contacted (one-sided test, p = 0.03), the intervention exerted a significant influence on the vaccination rate. However, in interpreting the effect of the intervention one must take into account the wide confidence interval (Figure 4), which is largely determined by the sample size. For instance, it is quite feasible that a closer relationship between physician and patient than was the case in our study might positively influence the intervention effect.
With regard to the secondary endpoint of the study, the intervention had no positive effect on the patients’ assessment of the risk of adverse effects or the severity of influenza as a disease. The increase in the vaccination rate of the intervention group was therefore not due to any modification of the patients’ attitudes. This is a striking finding, because previous studies have presented convincing data to show that insufficiently or inaccurately informed patients have often not been vaccinated (9). Therefore many campaigns—like ours—are informative in nature. Our results show, however, that it is obviously not necessary to change patients’ attitudes in order to increase the vaccination rate. It seems that thinking about vaccination, or just being reminded about it, is more important for the patients.
In the second approach we explored whether an appeal in writing from the regional ASHIP can influence the vaccination behavior of office-based physicians and thus indirectly help to increase the vaccination rate. This intervention showed no positive effect on the patients’ vaccination rate. Since consistently correct coding and billing of the vaccinations given can be expected on financial grounds, this is unlikely to be a false-negative result. It is more likely that the information provided in our study was not conveyed with enough urgency to attract attention in daily practice, or that the appeal was not followed because of hitherto unknown barriers to implementation.
One of the most important vaccination campaigns of recent years in Germany is the “Let’s Beat Flu to It” (“Wir kommen der Grippe zuvor”) campaign, jointly run by the Federal Centre for Health Education and the Robert Koch Institute since 2006 (10). Repeatedly, more than 50 000 primary care physicians have been sent letters urging them to vaccinate their patients against influenza. The effect has not yet been scientifically evaluated; however, our findings in the study presented here suggest that such a campaign does not result in an increased vaccination rate (10, 15). On the basis of our data it would seem appropriate to question the utility of this kind of approach.
In the third approach we tested whether a vaccination reminder from a health insurance fund has an effect comparable to a reminder from a physician. We found that this intervention did indeed significantly increase the vaccination rate of patients with chronic renal disease, but not to the same extent as our patient-centered intervention. Overall, 32 letters had to be sent to achieve one additional vaccination against influenza. This result is in the same order of magnitude as the effect of a centralized reminder system recently tested in Colorado, USA (16). In that study the intervention raised the rate of vaccination against influenza by 3.4% (NNT: 30). Several studies have indicated that influenza vaccination is a cost-effective prophylactic measure, particularly in high-risk patients (17–19). It is estimated that US$ 73 (circa € 65) in direct health-related costs can be saved by each vaccination (20). To what extent these figures are transferable to the German situation is unclear, but in view of the overall cost of an influenza vaccination (circa € 22) the expense generated by our intervention (circa € 25 for printing and postage to achieve one additional vaccination) does not seem excessively high.
Our study suggests that physicians could raise vaccination rates among their patients by adding written prompts to direct verbal reminders. To achieve a nationwide increase in vaccination rates, however, it would seem more pragmatic to establish vaccination reminder systems at the level of the health insurance funds. Such an approach has several potential advantages: An insurance fund covers far more patients than a physician’s office, facilitating widespread implementation. Moreover, redundant establishment of reminder systems would be ruled out. Furthermore, the principle could be extended, as and when required, to other indications, e.g., to combat low rates of vaccination against measles or HPV (21, 22). In the form tested here, the insurance fund approach offers scope for improvement: The wording and timing of the communications could be better adapted to the requirements of the target group, and insurees not vaccinated in the previous year could be prioritized. On grounds of data protection and costs, a key role could be played by digital reminder systems in which the insurees control the data. Any such system should be contained in the electronic medical record.
This approach would, for the first time, largely individualize the dissemination of information about vaccination. Rather than sending the same standard communications to everyone, information material could be tailored to the needs of the respective target group.
- Direct comparison of our different studies is limited by the fact that the patient collectives are not uniform (kidney transplant patients versus patients with chronic renal disease).
- Because patients with other chronic illnesses might react differently to the same interventions, the findings cannot be generalized with any certainty.
- Despite the similarity of the patient characteristics between the initial and follow-up surveys of the patient-centered intervention study, and despite the high response rate, the anonymization that was carried out means there is no guarantee that the same patients participated in both surveys.
- On data protection grounds, the ASHIP did not carry out any person-related data evaluation. The samples in this intervention study are therefore not strictly paired, so the possibility that the results are affected by patients who changed physician cannot be ruled out. However, the size of the sample means that a meaningful effect is very unlikely.
- With regard to the control group of the insuree-centered intervention study (insured persons with chronic pulmonary disease), it cannot be absolutely ruled out that factors external to the study (e.g., vaccination campaigns) affected only the behavior of patients with chronic renal disease.
We are particularly grateful to Sarah Jacobskötter for her painstaking editing of the German manuscript. We also thank Florian Ernst (IKK Südwest) and Dr. Heidrun Thaiss (Federal Centre for Health Education) for their valuable support in carrying out this research.
All of the data on participants in the patient-centered intervention study can be provided to interested researchers in anonymized form. The primary data from the physician-centered and insuree-centered intervention studies cannot be made available. All applicants should enclose a methodologically sound proposal for analysis of the data concerned and guarantee that the data will not be used for any other purpose. Applications should be sent to Dr. med. Kevin Schulte, Medizinische Klinik IV, Rosalind-Franklin-Str. 12, 24105 Kiel, Germany. The data will be available from 3 months to 5 years after publication of this article.
Conflict of interest statement
The authors declare that no conflict of interest exists.
Manuscript received on 15 December 2018, revised version accepted on 4 April 2019
Translated from the original German by David Roseveare
Dr. med. Kevin Schulte
Klinik für Nieren- und Hochdruckkrankheiten
24105 Kiel, Germany
Cite this as:
Schulte K, Schierke H, Tamayo M, Hager L, Engehausen R, Raspe M, Hübner RH, Schlieper G, Borzikowsky C, Urbschat A, Auerswald S, Kunzendorf U, Feldkamp T: Strategies for improving influenza vaccination rates in patients with chronic renal disease—results from two randomized controlled trials and a prospective interventional study.
Dtsch Arztebl Int 2019; 116: 413–9. DOI: 10.3238/arztebl.2019.0413
eMethods, eTables, eSupplement:
Department of Strategic Data Analysis and Health Policy North Rhine–Westphalia, North Rhine Association of Statutory Health Insurance Physicians, Düsseldorf: Miguel Tamayo
IKK Südwest, Saarbrücken: Dr. Lutz Hager, Roland Engehausen
Medical Department, Division of Infectiology and Pneumology, Faculty of Medicine, Charite Berlin: Dr. Matthias Raspe, PD Dr. Ralf-Harto Hübner
Center for Renal, Hypertensive, and Metabolic Diseases, Hanover: PD Dr. Georg Schlieper
Institute of Medical Informatics and Statistics, University of Kiel: Dr. Christoph Borzikowsky
Division of Information Technology—Analysis and Development, Schleswig-Holstein Association of Statutory Health Insurance Physicians, Bad Segeberg: Andreas Urbschat
Thuringia Association of Statutory Health Insurance Physicians, Weimar: Sven Auerswald
|1.||Influenza Wochenbericht Kalenderwoche 14/2018, AGI-Studiengruppe, Robert-Koch Institut. https://influenza.rki.de/Wochenberichte.aspx (last accessed on 10 June 2018).|
|2.||Deutsche Gesellschaft für Pneumologie und Beatmungsmedizin e. V., Deutsche Atemwegsliga e. V., Österreichische Gesellschaft für Pneumologie: Leitlinie zur Diagnostik und Therapie von Patienten mit chronisch obstruktiver Bronchitis und Lungenemphysem (COPD). https://pneumologie.de/fileadmin/user_upload/2018-COPD_Pneumologie.pdf (last accessed on 1 August 2018).|
|3.||KDIGO-Guidelines for management of chronic kidney disease. www.kdigo.org/guidelines/ckd-evaluation-and-management (last accessed on 16 March 2019).|
|4.||Empfehlungen der Ständigen Impfkommission (STIKO). Epidemiologisches Bulletin 34, 2017. www.rki.de/DE/Content/Infekt/EpidBull/Archiv/2017/Ausgaben/34_17.pdf?__blob=publicationFile (last accessed on 16 March 2019).|
|5.||Rolfes MA, Flannery B, Chung J, et al.: Effects of influenza vaccination in the United States during the 2017–2018 influenza season. Clin Infect Dis 2019 (ahead of print).|
|6.||Reuss AM, Walter D, Feig M, et al.: Influenza vaccination coverage in the 2004/05, 2005/06, and 2006/07 seasons: a secondary data analysis based on billing data of the German associations of statutory health insurance physicians. Dtsch Arztebl Int 2010; 107: 845–50 VOLLTEXT|
|7.||Aktuelles aus der KV-Impfsurveillance – Impfquoten ausgewählter Schutzimpfungen. Epidemiologisches Bulletin 1, 2018. www.rki.de/DE/Content/Infekt/EpidBull/Archiv/2018/Ausgaben/01_18.html (last accessed on 16 March 2019).|
|8.||Bodeker B, Remschmidt C, Muters S, Wichmann O: [Influenza, tetanus, and pertussis vaccination coverage among adults in Germany]. Bundesgesundheitsblatt Gesundheitsforschung Gesundheitsschutz 2015; 58: 174–81 CrossRef MEDLINE|
|9.||Bodeker B, Remschmidt C, Schmich P, Wichmann O: Why are older adults and individuals with underlying chronic diseases in Germany not vaccinated against flu? A population-based study. BMC Public Health 2015; 15: 618 CrossRef MEDLINE PubMed Central|
|10.||Bundeszentrale für gesundheitliche Aufklärung, Robert-Koch Institut: „Wir kommen der Grippe zuvor“. www.bfr.bund.de/cm/343/influenza_impfkampagne_wir_kommen_der_grippe_zuvor.pdf (last accessed on 16 March 2019).|
|11.||Tonelli M, Wiebe N, Manns BJ, et al.: Comparison of the complexity of patients seen by different medical subspecialists in a universal health care system. JAMA Netw Open 2018; 1: e184852 CrossRef MEDLINEPubMed Central|
|12.||Anderson KK, Sebaldt RJ, Lohfeld L, et al.: Patient views on reminder letters for influenza vaccinations in an older primary care patient population: a mixed methods study. Can J Public Health 2008; 99: 133–6.|
|13.||Briss PA, Rodewald LE, Hinman AR, et al.: Reviews of evidence regarding interventions to improve vaccination coverage in children, adolescents, and adults. The Task Force on Community Preventive Services. Am J Prev Med 2000; 18 (1 Suppl): 97–140.|
|14.||Thomas RE, Lorenzetti DL: Interventions to increase influenza vaccination rates of those 60 years and older in the community. Cochrane Database Syst Rev 2018; 5: CD005188 CrossRef PubMed Central|
|15.||Wortberg S, Walter D, Knesebeck M, Reiter S: [Physicians as key communicators of the influenza vaccination for the elderly, patients with chronic conditions, and health care workers. Results of a nationwide survey in the context of the national influenza vaccination campaign]. Bundesgesundheitsblatt Gesundheitsforschung Gesundheitsschutz 2009; 52: 945–52 CrossRef MEDLINE|
|16.||Hurley LP, Beaty B, Lockhart S, et al.: RCT of centralized vaccine reminder/recall for adults. Am J Prev Med 2018; 55: 231–9 CrossRef MEDLINE|
|17.||Scuffham PA, West PA: Economic evaluation of strategies for the control and management of influenza in Europe. Vaccine 2002; 20: 2562–78 CrossRef|
|18.||Rychlik R, Heinen-Kammerer T, Rusche H, Piercy J, Scuffham P, Zollner Y: [Cost-effectiveness of prophylaxis and treatment of influenza]. Dtsch Med Wochenschr 2003; 128: 2267–70 CrossRef MEDLINE|
|19.||Nichol KL, Margolis KL, Wuorenma J, von Sternberg T: The efficacy and cost effectiveness of vaccination against influenza among elderly persons living in the community. N Engl J Med 1994; 331: 778–84 CrossRef MEDLINE|
|20.||Nichol KL, Wuorenma J, von Sternberg T: Benefits of influenza vaccination for low-, intermediate-, and high-risk senior citizens. Arch Intern Med 1998; 158: 1769–76 CrossRef|
|21.||Storr C, Sanftenberg L, Schelling J, et al.: Measles status-barriers to vaccination and strategies for overcoming them. Dtsch Arztebl Int 2018; 115: 723–30 VOLLTEXT|
|22.||Rieck T, Feig M, Delerè Y, et al.: Utilization of administrative data to assess the association of an adolscent health check-up with human papillomavirus vaccine uptake in Germany. Vaccine 2014; 32: 5564–9 CrossRef MEDLINE|