This article is more than 5 years old. Information may no longer be current.
In-hospital mortality in AF more prevalent in rural hospitals
Patients who were admitted to rural hospitals for atrial fibrillation had an increased risk for in-hospital mortality compared with those who were admitted to urban hospitals, according to a study published in HeartRhythm.
“The identification of health care disparities is of utmost importance at this time to improve the overall care that is delivered in our health care system,” Wesley T. O’Neal, MD, MPH, fellow in the division of cardiology at Emory University School of Medicine, said in a press release. “Our research shows that urban-rural differences exist regarding the risk of hospital mortality among patients who are admitted for AF.”
AF hospitalizations
Researchers analyzed data from 248,731 patients (mean age, 69 years; 48% women) from the National Inpatient Sample database who were hospitalized for a primary diagnosis of AF between 2012 and 2014. Patients were excluded if they were transferred to another acute care hospital when they were discharged.
Data such as sex, age, race and insurance status were reviewed. Hospitals were categorized as urban if they were in a metropolitan area with at least one urbanized area of at least 50,000 people. Rural hospitals had at least one urban cluster with a population between 10,000 and 50,000 people or were not considered metropolitan or micropolitan.
The main outcome of interest was in-hospital mortality, defined as any cause of death during hospitalization. Common secondary diagnoses were also reviewed, including hypertension, HF, diabetes, hyperlipidemia and acute kidney injury.
Most patients were admitted to urban hospitals (88%) compared with rural hospitals (12%).
Patients admitted to rural hospitals had a 17% increased risk for death vs. those admitted to urban hospitals (OR = 1.17; 95% CI, 1.04-1.32) in a multivariable model for different patient characteristics. A propensity score-matched cohort and subgroup analyses for race, sex and region had similar results.
Patients in the urban hospital group were more likely to have secondary diagnoses of hyperlipidemia (4.5% vs. 3.3%; P < .001), HF (16.7% vs. 16.1%; P = .048) and acute kidney injury (3% vs. 2.2%) compared with the rural hospital group. The reverse was true for a secondary diagnosis of hypertension (6.9% vs. 7.9%; P < .001).
Study implications
“These data are of interest to policymakers with the aim of reducing mortality in rural regions,” O’Neal and colleagues wrote. “Residents from rural regions are more likely to rely on public sources of health insurance, and variation in quality of cardiovascular care and access to medical care exist. This would result in patients with AF who are more likely to present to a hospital for AF care rather than to an outpatient clinic. Therefore, reductions in in-hospital mortality of AF possibly would be observed with improvements in access to care in rural regions, as they would be less likely to rely on hospital care for AF.”
“We believe it is fair to state that their study raises more important clinical and epidemiologic questions than it provides answers,” Thomas Deering, MD, FHRS, cardiac electrophysiologist at Piedmont Heart Institute in Atlanta, and Ashish A. Bhimani, MD, FHRS, clinical cardiac electrophysiologist at Piedmont Heart Institute, wrote in a related editorial. “Although rural AF patients had a higher mortality rate than urban AF patients, information about the many operative factors (eg, associated comorbidities and their severity, access to care, patient lifestyle decisions, patient compliance, physician adherence to diagnostic and therapeutic guideline recommendations), which may have contributed to producing the observed outcomes, remains unknown. Accordingly, claims-based analyses such as these should be viewed as hypothesis-generating instead of categorical in nature.” – by Darlene Dobkowski
Disclosures: The authors, Deering and Bhimani report no relevant financial disclosures.
Perspective
Back to Top
When considering studies that advance the scientific knowledge base, it is very important for us to realize that there are revolutionary studies which change the paradigm of how we treat patients and other studies which raise very important clinical questions while suggesting areas for additional research. This study falls into the latter group.
The findings of this study do have some implications relevant to clinical practice. The results presented in this study suggest that the mortality rate for patients hospitalized with AF in rural settings is higher than that observed among patients hospitalized in urban areas. This possibility should cause any clinician taking care of patients in a rural setting to consider his or her practice patterns and what steps they might take to mitigate higher mortality rates. For example, “Do I know what the guidelines tell me to do? Am I practicing by the guidelines? Do I know when to refer my patient to a more sophisticated center?” While these approaches ring true in any environment, the presented data suggesting that rural outcomes are inferior, should make practitioners in that environment pay particular attention to these issues.
This study does not provide definitive answers. Accordingly, while acknowledging the concerns raised by these researchers, it is important to state that we need more information and that we need to put the findings of this study into context.
Whenever one observes differences in clinical outcomes based upon geography, in addition to addressing care delivery differences that may have had an impact upon those results, one should also consider differences in the clinical characteristics of the associated patients and the capabilities of the settings in which that care is delivered, since high-end treatment is not typically available in rural arenas. Accordingly, all smaller rural health care institutions should have a positive referral relationship with an institution that is capable of providing high-end care for patients with AF. While this consideration was not assessed in the associated manuscript, this health care need is an important option.
When considering limitations associated with this study it is important to note that this is a claims-based study. While claims-based studies usually include a larger number of patients than those included in a prospective randomized trial, they have a number of limitations. Because of the associated methodology, claims-based study analyses typically raise more questions than they provide definitive answers. Given the nature of these analyses, it is difficult to take into account how the insufficient information about the type and the duration of AF, all of the existing external factors, comorbidities, other nonclinical conditions and treatment options might have influenced the outcomes.
To confirm or refute the presented data, we need to undertake additional studies. A more limited “fact-finding” approach could focus upon three options. A registry-based study has the capacity to assess outcomes in a defined patient population. Unfortunately, the ability to do such among rural patients is rather limited.
Another alternative is to match patients based upon similar demographics. This too is difficult.
Ideally a large, prospective, well-designed randomized trial is the best benchmark. Unfortunately, these types of trials are expensive and difficult to orchestrate. Accordingly, it is oftentimes easier to perform one of the former studies than the latter. Those studies can then provide more information to help to refine further scientific investigation. When data obtained from these less than perfect approaches supports the proposed postulate, this alignment can lead health care teams to initiate process-improvement approaches while data, which does not align, creates the substrate for further investigation.
In summary, we need to perform more research to analyze matched, at-risk populations to determine effective health policies. Apple-to-apple comparisons are better than apple-to-orange comparisons. Frequently, when scientists review a study like this, they have a tendency to find fault with the methodological limitations while failing to consider the health policy implications of the study if confirmed. This study suggests but does not prove that there are geographic differences in the delivery and outcomes of care.
This should create a clarion call to the medical community to ask the question, “How can we do better?” In the interim, every institution and every clinician should look to see if she/he is doing the best that that they can.
While acknowledging the potential concerns raised by this study, we should take a deep breath and acknowledge that the mortality difference between the rural and the urban locations was very small, 1.3% vs. 1%. Accordingly, we cannot say definitively that this constitutes a clinically meaningful difference. However, if we look at the results from a broader prospective, we should view the findings as a focal point to determine if this issue is a real problem on which we need to focus.