Efficacy of a novel patient-focused intervention aimed at increasing adherence to guideline-based preventive measures in asplenic patients: the PrePSS trial

Asplenia or hyposplenic states are associated with a specific immunodeficiency rendering patients susceptible to invasive infections, the most severe manifestations referred to as post-splenectomy sepsis (PSS) [1‐4]. Even if treated promptly and aggressively, morbidity and mortality remain high. However, most PSS episodes are preventable [1, 5]. An array of preventive measures, including vaccinations and some form of antibiotic prophylaxis, has been found to be effective in reducing mortality and has been adopted in many national and international guidelines [6, 7]. Adherence, however, is low with vaccination coverage among asplenic patients remaining unsatisfactory and patients oftentimes lacking adequate education regarding their medical condition [8]. This is especially unfortunate because well-educated patients have been shown to exhibit lower mortality rates from PSS [9].

Some researchers have evaluated and reported measures to strengthen guideline adherence such as spleen registries and dedicated outpatient services. Although some produced encouraging results, these implementations either depended heavily on local infrastructure or had shortcomings in the inter-sectorial communication (e. g. continuation of vaccinations post-hospitalization) [10, 11]. Therefore, widespread adoption seems unlikely. We hypothesized that adherence to recommended prevention measures can substantially be improved by a novel, health action process approach (HAPA)-informed intervention targeting both patients and their physicians. Several studies propose the HAPA theory as a theoretical framework for understanding health behavior in general [12, 13] and vaccination behavior in particular [14‐16]. Here, we present data from the PrePSS study, a two-armed historical control group intervention study, which aimed to assess the impact of the developed intervention on guideline adherence in asplenic patients.

A total of 247 patients were included in the study, 209 of which we were able to include in our analyses (Fig. 1). In total, 106 were included in the intervention group and 113 served as a historical control group. Although our study was not randomized due to ethical concerns, baseline characteristics were quite balanced even without weighting (Table 1). Main cause for splenectomy in both groups was an abdominal tumor (65.1% and 59.3%, respectively), of which the majority was malignant. Patients in the intervention group reported slightly higher rates of comorbidities and concurrent immunosuppression. The median time between splenectomy and study inclusion was 437 days (IQR 350–597) in the control group and 111 days (IQR 91–139) in the intervention group. After applying overlap weights, baseline covariates were exactly balanced and 102.8 and 109.3 patients were analyzed in the intervention and control group, respectively. Patients were 58.6 years old and a slightly higher prevalence of female gender was noted (55.8%). We recorded one PSS event in the control group and no events in the intervention group.

At six months after the intervention, patients of the intervention group showed a mean increase of 3.96 (95% CI 3.68–4.24) points on the PrePSS scale [range 0–10], indicating a major improvement in guideline adherence compared to the historical control group (Fig. 2). Mean PrePSS scores were roughly doubled with 3.73 and 7.70 in control and intervention group, respectively. As a sensitivity analysis, we repeated the calculations using IPT weighting. Results were virtually unchanged with a mean increase of 4.02 (95% CI 3.77–4.26) points on the PrePSS scale.

There were some marked differences between the individual items of the PrePSS score (Fig. 3 and Table 2). While most patients in the control group did receive at least one anti-pneumococcal vaccination, fewer than half of these patients were given a medical emergency card and fewer than 10% were in possession of a stand-by antibiotic. After the intervention, almost all patients (97.9%) carried their emergency card at all times and nearly two-thirds had a stand-by antibiotic readily available. An increase in complete meningococcal vaccinations was noted with 48.7% percent being fully vaccinated compared to 7.5% in the control group. Complete sequential vaccination against pneumococci did not see an equally noticeable gain, with only 42.2% of patients having completed this course after the intervention. However, this is almost double the rate of the control group and almost three quarter (73.0% vs. 41.7% in the control group) of the patients did in fact receive both anti-pneumococcal vaccines, albeit not in the recommended order. Overall, 21.4 patients in the intervention group achieved the full PrePSS score of 10 points, while no patient in the control group reached this level.

Furthermore, we explored the association between the degree of physician participation in the intervention and the PrePSS score outcome. Many physicians did not receive the complete physician-focused intervention, primarily due to lack of time on the part of the physician. In these cases, a condensed version of the intervention focusing on the patient at hand was presented. Some physicians could not be reached at all. We graded the level of the received intervention (no contact, 13.8 patients; short contact, 64.0 patients; full intervention, 25.0 patients) and explored its relationship with the PrePSS score outcome (Fig. S1). Even patients, whose physician could not be reached at all, exhibited only marginally lower PrePSS-scores compared to the patients, where the physician received the full intervention, although a slight trend was noticeable.

Using a prospective controlled, two-armed historical control group design, we were able to demonstrate that our novel telephone-based HAPA intervention targeting both patient and physician led to considerably improved adherence to recommended preventive measures in asplenic patients. Educating and motivating the patient was identified as the crucial factor of the intervention since even patients, whose physician did not receive the intervention, scored only marginally lower than those, whose physician had received some form of intervention.

Integral part of all PSS prevention is the implementation of a vaccination plan. Due to different study designs and methodological approaches, investigators report a wide range of vaccine coverage in asplenic patients [5, 8, 22‐24]. Most of the studies reporting high vaccine coverages report data from a “one-stop-shop” approach with clinicians administering vaccines during the hospital stay after splenectomy. This strategy, although effective initially, might prove challenging in the long term, because patients generally rely on their physician for booster shots and continuation of initiated vaccination schedules. Accordingly, studies with a more population-based approach reveal that vaccination coverage in asplenic patients in general is still unsatisfactory [8]. Where data regarding stand-by antibiotics are available, rates tend to be low, which is in good agreement with our findings [22, 25].

Several attempts have been made to improve guideline adherence. One popular approach is to establish an outpatient service in order to follow up asplenic patients and provide them with the necessary education and, if needed, vaccinations [10, 26]. Although effective and conceivably producing a sustained response, this approach is resource-intensive and by design regionally limited. Improvements in vaccination coverages have also been reported for implementations of local registries [11, 27]. However, these strategies depend on activities of local research groups; therefore, outreach is limited and sustainability often difficult to accomplish. Our intervention aimed to improve on some shortcomings of other approaches. Firstly, our intervention addresses the need for patient education. Although patients’ disease knowledge has long been recognized as crucial in disease prevention in general [28, 29] and specifically in prevention of PSS [9], surveys find that many patients did either not receive disease-specific education at all or have poor recollection of the delivered educational contents [30]. In our cohort, too, most patients stated that they had not been made aware of the long-term implications of splenectomy and were surprised by our initiative for the purpose of further education. We tried to address this issue by providing easy to understand information concerning asplenia in our telephonic intervention. Importantly, we integrated HAPA-informed elements, which were designed to involve the patients and achieve a sustained effect by treating the patients not as mere recipients of information but as actors in the process of their healthcare planning. It should be emphasized that our theory-based intervention takes into account both educational and volitional aspects to ensure that asplenic patients are empowered to apply prevention measures in the long term.

Secondly, our intervention is easily distributable, resource-saving when compared to face-to-face contacts and not confined to any specific setting in terms of both place and time. It can be understood as a telehealth preventive intervention, well applicable for example in underserved rural areas, and does not need dedicated structures but instead leverages existing ones, e. g. by involving the physician of the patient. By making our manual publicly available, we hope to open up a wide range of possibilities for implementation both in the inpatient and outpatient setting.

Thirdly, our intervention sought to bridge the intersectorial divide by involving the physicians in the prevention process. Intensifying the collaboration between clinicians and physicians was shown to be essential in order to improve the quality of care for asplenic patients [31]. However, we found that many physicians could not fully participate in the intervention as planned. In particular, with the added workload of the COVID-19 pandemic, most of them simply could not spare the time to participate in our study. Interestingly, though the level of asplenia-specific education, the physician received on our part was not heavily associated with PrePSS score outcomes. This further underlines the considerable potential of patient-focused interventions aiming for education and empowerment of patients with regard to their health or disease management. The role of the physician remains pivotal, however, since in Germany 85–90% of all vaccinations are carried out by the physician [32]. The physician’s contribution is especially essential in cases where patients are less well suited for our intervention due to, e.g., old age or comorbidities.

Finally, a unique feature of our design lies in the important role it assigns to the health insurance provider. Our approach includes an easy to perform and inexpensive screening procedure that enables health insurance companies to actively take part in delivering preventive strategies in a focused and patient-specific way. Leveraging the existing infrastructure of the health insurance provider not only facilitated inclusion of patients but also may prove pivotal in further implementation and widespread roll-out of our intervention. Given its aptitude for printed distribution, part of the intervention could conceivably be automated and sent out to both patients and their physicians upon screening and identification via OPS codes.

There are some limitations to our study. It is possible that our results are influenced by a responder bias, since first contact with patients was made via mail from the health insurance provider and we could only include those that responded to this letter. This might have led to an overrepresentation of patients, who were already motivated to obtain additional asplenia-related information. Our study was performed in cooperation with a singular health insurance provider. A potential barrier to widespread implementation could lie in the highly diverse landscape of the German health insurance system. While in theory all insurance companies should have access to the required data, the ability to implement the described intervention could conceivably vary depending on the type of insurance company (e.g., state funded vs. privately owned) and on the region in question. Additionally, due to the non-randomized design of our study, we cannot rule out confounding. Although we used a well-designed statistical model to account for this, unmeasured confounding might still be present. Due to the short follow-up time, we cannot speak to the long-term success of our intervention, yet we are currently performing a follow-up concerning primary and secondary outcomes three years after the intervention.

In conclusion, asplenic patients who had received a HAPA-based, telephonic intervention exhibited a significantly higher adherence to guideline-based preventive measures compared to a historical control group. Widespread adoption of this pragmatic intervention could improve patient care and ultimately lead to reduced morbidity and mortality from PSS. Moreover, our strategy of health insurance provider-initiated identification of at-risk patients followed by a patient-focused intervention may serve as a blueprint for a wide range of other preventive efforts leading to patient empowerment and ultimately to better adherence to standards of care.