This study provides novel ‘real world’ data on the referral and impact of in-hospital cardiac rehabilitation programs in patients suffering AMI. The study involved more than 13.000 patients admitted to hospitals in the Lazio region of Italy between 2013 and 2015.
The study results showed that IH-CR participation rates are alarmingly low, ranging from 3 to 17% in relation to the groups of the AMI population identified. The highest IH-CR referral rate were observed for STEMI and NSTEMI patients who did not receive PCI during the index hospitalization. Overall, we found that only 1.101 patients (8%) started cardiac rehabilitation program within 30 days from the hospital discharge following an AMI.
As evidenced by the results of this study, IH-CR referral rates have been very low in comparison with other E-B treatments. Previous investigations from several states reported an average referral rate of 30% in Canada, the USA, and the UK, and a little higher at 50% in the rest of Europe . Differences in the organization of health care services and delivery systems between countries and across hospitals may explain, most probably, this variability. Moreover, there are also differences in referral rates between European countries, but little is known, in literature, about the real motivations for underuse of cardiac rehabilitation in different EU countries and hospitals [10, 30]. However, these current low rates of participations in IH-CR programs are wholly inadequate and deprive a large proportion of patients with cardiovascular disease of a safe and effective intervention in reducing disease-related disability, as well as hospital readmission and long-term mortality. There are several reasons for this inadequate use of CR. First of all, the lack of a hospital system that automatically associates the rehabilitation program with each eligible patient, without the intermediation of a physician or health care staff. In fact, previous studies suggested that the most important predictor, is exactly the lack of CR referral at hospital discharge .
Sometimes, it may be that patients do not receive enough information and encouragement to participate in CR by their physician or other health professionals. However, physicians certainly need to take into consideration many factors when giving a CR referral: a patient could be too old or frail to benefit from CR, or conversely, too healthy and may not need this intervention. Last but not least, CR regimes are not standardized across countries regarding duration (3/6 weeks or 12/16 weeks), setting (in-hospital or ambulatory), and main goal of therapy (exercised based or comprehensive recommending). In Italy, there are no top-down standards or guidelines regarding the scope of services in IH-CR. The gold standard of care refers to the guidelines drawn up from the scientific societies. The need for a national clinical register in Italy, which supports the standardization of interventions has been repeatedly advanced by the scientific community . The procedures, the personnel, and modalities necessary for carrying out the activities that can be defined as “rehabilitation intervention”, have been also specified. However, at least in the Lazio region there are no steering documents that make the standard of care in CR mandatory . Moreover, this work increases knowledge and awareness that, even in the most advanced interventional cardiology contexts, where networks hub and spoke for acute myocardial infarction have already been developed for two decades, post-acute care is still very lacking. In fact, consistent with the results of other investigations [26,27,28, 34] in the present study we found that evidence-based pharmacological therapies for secondary prevention after AMI were under-used (overall, 44% and 51% of the patients were deemed adherent to poly-therapy at 6-and 12-month, respectively).
Our findings showed a differential impact of cardiac rehabilitation program on polytherapy adherence both at 6-and 12-month follow-up assessment. The magnitude of this impact depending on the different groups of AMI patients considered, which are related to IH-CR participation. Principally, there would seem to be a strong different impact between patients who underwent or not PCI, regardless of the AMI diagnosis, during the index hospitalization. In fact, during the 6-month follow-up, after adjusting for potential confounders, STEMI-NO-PCI and NSTEMI-NO-PCI patients who started IH-CR program were, 63% and 85% respectively, more likely to be adherent as compared with non-participants. Interestingly, polytherapy adherence was maintained for STEMI-NO-PCI participant patients (63%), and even increased for NSTEMI-NO-PCI patients at 12-month follow-up, suggesting that patients who could not undergo revascularization by coronary angioplasty may have been more often referred to IH-CR than those who received coronary angioplasty. Likewise, the rehabilitation intervention results in a greater adherence to drug therapy. Among NSTEMI-NO-PCI patients who participated in IH-CR were more than twice as likely to be adherent to E-B drugs (OR 2.13; 95% CI 1.74–2.60; p value: < 0.001). Of note, estimates were adjusted for all variables identified as potential confounders such as age, gender, and 18 comorbidities listed in the Additional file 3: Selection of comorbidities from hospital discharge records. By contrast, the effect of IH-CR intervention on adherence was not observed in the other two subgroups of AMI patients (STEMI and NSTEMI) who underwent PCI during the index event. This result may be partially due to the fact that AMI patients who underwent a primary percutaneous coronary intervention have already been inserted in specific care pathways. In fact, these two groups of patients, regardless of IH-CR participation status, showed the highest adherence, for both 6-and 12-month follow-up (52% and 60% for STEMI; 49% and 56% for NSTEMI patients). Therefore, the beneficial effect of IH-CR treatment on medication adherence could be attenuated by the fact that AMI patients who underwent revascularization had been more carefully monitored and made aware of the long-term benefits generated by a continuous and persistent drug treatment.
The evidence of greater effect of IH-CR on the adherence to long-term evidence-based drugs in more complex groups, i.e., in patients who have had an acute coronary event but not revascularized, reinforces the concept that rehabilitation intervention is extremely relevant from a prognostic point of view. Non-revascularized patients with multiple comorbidities are recognized as those with the highest residual cardiovascular risk. Numerous clinical, epidemiological and intervention studies confirm this finding. In Italy, data from the National Outcomes Program (NOP) of 2016 show a reduction in mortality at 30 days from admission for acute myocardial infarction (AMI), from 10.4% in 2010 to 9.0% in 2015. An Italian retrospective study based on the administrative database of hospital discharge forms (HDF) examined over one million patients admitted for AMI from 2001 to 2011 in all Italian hospitals, evaluating the mortality and hospital readmission rates at 30 days and at 1 year and re-hospitalization for all causes. While the mortality rate at the index event declined the rate of fatal readmission at 1 year increased from 4.75 to 5.28% (p = 0.0019) . It thus emerges that, while the in-hospital management of ACS has shown great progress in terms of its diagnostic-therapeutic efficacy, post-discharge care has not had a positive impact on the post-discharge prognosis of ACS. This is in part attributable to the inadequacy and poor application of appropriate cardiac care pathways for the post-discharge period, based on different care needs modelled according to the individual patient’s level of risk. A high residual risk of MACCE recurrence can be detected both by clinical factors, such as diabetes mellitus, renal failure, peripheral artery disease, a history of angina or previous AMI, and by anatomical/surgical factors, such as the presence of multivessel disease, especially if treated with incomplete revascularization, or no revascularization at all. Elderly age is an associated high-risk factor included in the above comorbidities. A recent Italian study, based on HDF administrative data, analysed the risk of thrombosis in patients admitted in the years 2009 and 2010. At multivariate analysis, most of the factors defining high thrombotic risk were independent predictors of 5-year mortality. The effect of “thrombotic risk” on mortality resulted to be time-dependent with a hazard ratio (HR) that strongly increased in the first two years of follow-up, then slowly levelled out in the following years to reach a plateau around the 5th year. This finding confirms not only the importance of targeting intensive prevention therapeutic strategies in the period immediately following ACS but points to the need for long-term secondary prevention programs in subgroups of high-risk patients . Intervention studies have confirmed this epidemiological trend as regards dual antiplatelet therapy [36,37,38], LDL target [39, 40] and even low-dose anticoagulation in the more recent COMPASS trial (Cardiovascular Outcomes for People Using Anticoagulation Strategies) . In this context, therefore, adherence to evidence-based therapy, as well as implementation of new secondary prevention pharmacological strategies in patients at very high residual risk plays a fundamental role. Unfortunately, despite the evidence, as also evidenced by the results of this research, the adherence to pharmacological therapies after an acute event is still far from optimal.
Moreover, the present study investigated the long-term effects of IH-CR intervention on all-cause mortality, hospital readmission due to MACCE, and admission to ED at 3-year follow-up. As previously mentioned, the participation in IH-CR had a “protective effect” on the occurrence of these three adverse secondary outcomes, for three out of the four subpopulations identified. However, this protective effect was statistically significant only for NSTEMI-NO-PCI patients. The study results showed that IH-CR participation, was associated with 24% of risk reduction of mortality (IRR 0.76; 95% CI 0.60–0.95; p value: 0.021), 22% of MACCE readmission (IRR 0.78; 95% CI 0.65–0.94; p value: 0.011), and 20% of ED admission (IRR 0.80; 95% CI 0.70–0.91; p value: 0.001). Much of this effect is probably due to the increase in adherence to evidence-based therapies. Our findings are consistent with the results of other investigations that reported a relationship between CR participation and risk reduction for AMI patients [42,43,44]. In addition, meta-analyses suggested that CR participation can reduce all-cause mortality by 15–28% for AMI patients [45, 46], confirming our results. We should be pointed out how NSTEMI-NO-PCI patients had the highest percentage of patients who participated in IH-CR intervention (17%) compared to the other AMI subgroups. Moreover, in absolute terms, patients with a diagnosis of NSTEMI who did not undergo PCI, represented the 50% of all AMI patients that have participated in IH-CR program in this research.
Rather, we found that all these secondary outcomes tended to be higher in the IH-CR group for STEMI-PCI patients. However, based on the factors showed in Table 1, it appears evident a different severity of patient’s condition who participated or not to IH-CR program in this specific subgroup of population. In fact, IH-CR participants were more likely to have a history of heart failure (29% vs. 7%), conduction disorders or arrhythmias (32% vs. 19%), and were more likely to be treated with other operations on heart and pericardium (11% vs. 1%), compared with those who did not participate in IH-CR program. Consequently, all-cause mortality tended to be higher in the IH-CR group (17% vs. 6%).
Notable, the median follow-up time to death for patients who participated in IH-CR was 384 days (IQR 161–688) versus 603 days (IQR 301–822) for those who did not participate. For the above-mentioned reasons, it is reasonable to expect a strong disparity in the characteristics of patients who did or did not participate in IH-CR even for all variables that cannot be measured. Although we evaluated all available potential factors to adjust patient deviation, we could not adjust for imbalance in unmeasured confounders. Therefore, most likely the lack of more detailed data (e.g., ejection fraction, history of smoking, BMI, number of sessions completed, exercise capacity, vital sign, functional status, frailty, or social risk factors) has caused unmeasured confounding resulting in higher incidence rates for secondary outcomes, among STEMI-PCI patients who participated in cardiac rehabilitation compared with non-participants.
Strengths and limitations of the study
A strength of the study is the possibility to integrate different Health Information Systems of the Lazio region of Italy, in order to achieve a well-defined (e.g., chronological, demographical, clinical) healthcare-related patient profile and to involve a large number of patients. The robustness of the applied statistical methodologies, a population study design and a long-term follow-up are the major strengths. A weakness of this study is that the results derived from a single region of Italy, so may not be generalizable to other contexts due to differences in healthcare policies. Moreover, a misclassification of drug utilization may have occurred because in our database the prescribed daily doses were not known, and the defined daily doses were used as the dosage assumption to measure adherence to polytherapy [47, 48]. In addition, we only used data from RAD-R for the definition of study’s exposure (participation in cardiac rehabilitation). This information system collects data about patients referred to CR specialized facilities from hospital acute care centers and may not reflect the experience of other cardiac rehabilitation centers. Therefore, the IH-CR participation proportion may have underestimated. Finally, the lack of more detailed clinical data may have caused unmeasured confounding, despite all available potential confounders were considered to adjust for differences in characteristics of patients.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.