Abstract
Background
Residents of plain communities constitute an underserved minority population that is not reliably captured in contemporary surgical outcomes research. We hypothesized that plain communities (PC) patients would have higher postoperative complication rates than a general surgical population.
Methods
A retrospective review of 30-day postoperative outcomes for PC patients compared to a majority (non-PC) matched patient population from September 2014 to March 2020 was performed. The primary outcome measure was any complication within 30 days of surgery.
Results
270 PC patients were matched with 493 non-PC patients. The 30-day complication rate was higher for the PC group (6.3% vs 3.7%, P = .09), though not statistically significant. There was significantly lower utilization of preventive care services, and postoperative follow-up among PC patients.
Discussion
Although our regional PC surgical patient population utilized preventive and postoperative health care services less than the non-PC population, there was no statistically significant difference in overall 30-day postoperative morbidity or mortality.
Keywords
Key Takeaways
There were no statistically significant differences in overall postoperative outcomes between PC patients and other surgical patients. PC patients had significantly lower utilization of preventive care services and rates of postoperative follow-up.
Introduction
Quality health care research demonstrates disparities in underserved and underrepresented populations. Attention to and research about postoperative outcomes in minority populations have notably increased since 2003 when the Institute of Medicine published its landmark study Unequal Treatment. This spurred the nation to address glaring disparities in health care. 1 The U.S. Department of Health & Human Services published a progress report—updated most recently in 2015—to address various strategies by which to continue tackling this issue. 2 However, some minority populations have remained outside the reach of contemporary health quality research. One of these populations is the Plain Communities (PC) population, which includes multiple sub-groups, such as Amish, German Baptist, and Old Order Mennonite. Briefly, these groups can trace their religious roots to a sub-sect of the Anabaptist movement in Europe in the 1500’s, some of which immigrated to North America in the 1600’s and 1700’s owing to ongoing persecution for their religious beliefs. 3
As of 2020, the number of PC people in the United States was estimated to be 350875. 4 The majority of these people are concentrated in settlements across the Midwest, notably in Ohio and Indiana, with a large number in Pennsylvania as well. 5 In Wisconsin, an estimated 22235 PC people comprise 62 unique settlements as of 2020. 4 The PC population grew by more than 37% from 2010 to 2019. 6 For comparison, during the same time frame, the overall growth of the U.S. population was 6.1%, and the fastest-growing populations during that time include Asian Americans (27.8% growth), Hispanic Americans (19.4%), and Native Hawaiian or other Pacific Islander (19.0%). 7
Despite ongoing population growth, the PC population has been difficult to characterize in health care literature. This is for a variety of reasons. Their culture requires an indelible “line in the sand” that necessarily separates their beliefs and lifestyle from the non-PC person to preserve traditions and a particular lifestyle. 8 They typically live in close-knit, rural, agrarian communities that are family- and church-centric. 9 Relatedly, they do not often seek out modern medicine but often explore culturally specific traditional treatments and alternative health care options closer to home prior to seeking contemporary medical care. Research has shown that many PC people utilize alternative health care treatment such as dietary/herbal supplements, medical advice from publications, and advice or treatment from local healers in their community. 10 Additionally, there are barriers to seeking health care, including a lack of transportation and a lack of commercial insurance, opting instead to pay using community or Church-provided communal funds. 11 Despite the lack of modern health care utilization, limited research has shown that life expectancy is not significantly different from that of the general population, and they have lower rates of cancer and cardiovascular diseases. 12 The limited genetic inflow to their communities, however, does predispose them to a higher prevalence of certain genetic disorders. 13
A lack of health care tracking or databases dedicated to monitoring PC health care outcomes contributes to the lack of study regarding health care outcomes. There is a dearth of literature on the PC population across the country, but published research includes attempts to study the PC culture, health care needs, and utilization of resources.14,15 Given their genetic proclivity for certain disorders, attempts have been made to improve newborn screening and create an accurate genetic database.16,17 There are a few targeted studies addressing obstetric outcomes and pediatric trauma,18,19 but there are no published studies in the current literature to our knowledge that address postoperative outcomes of PC people in the United States. The objective of this study is to identify differences in postoperative outcomes or care between the PC population and a general surgical population at a single independent academic institution in western Wisconsin.
Methods
This study is a retrospective case-matched cohort design to analyze postoperative follow-up and outcomes for PC patients compared to a matched majority (white, non-PC) patient population. The patient population was extracted from a multispecialty, integrated health system that covers 22 counties over a tristate area including Wisconsin, Minnesota, and Iowa. PC patients typically utilize a community financial system and were identified by payor status through our medical center’s billing system information over a 5-year period, from January 2014 through December 2019. PC status was verified by review of specific religious affiliation, geographic location (zip code, city/state that correspond to locations where PC reside), and alignment with common PC family surnames. A total of 19 unique plain communities were represented in this dataset. Demographic data was extracted from our health system’s integrated electronic health record (EHR) system. These patients were matched with non-PC surgical patients up to a 1:2 ratio for comparison. Matching was performed based on age (±5 years), sex, date of operation (±90 days), American Society of Anesthesiologists (ASA) Physical Status Classification, primary surgical service line, and emergency admission status. Body mass index (BMI) and home address were additional demographic variables collected for all patients. The research database was evaluated for duplicate encounters among both PC and comparison groups, which were eliminated prior to statistical analysis. This research project was approved by the Gundersen Clinic Ltd. Human Subjects Committee/Institutional Review Board.
Thirty-day postoperative outcomes, rates of preventive care utilization, and postoperative follow-up rates were evaluated by querying and reviewing ICD-10 codes and encounters by department extrapolated from the EHR. The primary outcome measure was a composite measure of any complication within 30 days after surgery. Secondary outcomes included 30-day rates of mortality, myocardial infarction, pneumonia, sepsis, urinary tract infection, renal failure, pulmonary embolism, and deep venous thrombosis. The use of preventive care services was evaluated by analyzing cholesterol screening, colonoscopy, and mammography rates within the year prior to surgery. Lastly, the rate of attendance at postoperative follow-up visits was assessed. Data were reviewed and verified by our research team prior to statistical analysis.
Descriptive statistics are reported as counts and percentages and as means and standard deviations. Study group clinicodemographic factors and patient outcomes were compared using x2, Fisher’s exact, and rank-sum tests. A threshold P value <.05 was set to define statistical significance. All analyses were performed using SAS 9.4 (SAS Foundation, Cary, NC).
Results
Descriptive characteristics of Plain Community and non-Plain Community patients.
Abbreviations: PC, Plain Community; ASA Class, American Society of Anesthesiologists Physical Status Classification; BMI, body mass index; SD, standard deviation.
aData are presented as no. (%) unless otherwise noted.
bn = 215.
cn = 422.
30-Day Postoperative Complications in Plain Community versus non-Plain Community Patients.
Abbreviations: PC, Plain Community; UTI, urinary tract infection; AMI, acute myocardial infarction; PE, pulmonary embolism; DVT, deep vein thrombosis.
There was lower utilization of select preventive care services in the year prior to surgery among the PC versus non-PC population for all surrogate measures (Figure 1). Subgroup analysis for only general surgery patients also showed lower rates of cholesterol screening (3.1% vs 23.4%, P < .0001), mammography (2.2% vs 28.9%, P = .002), and colonoscopy (2.1% vs 8.8%, P = .04). This trend was also reflected among PC versus non-PC patients greater than 65 years of age for cholesterol (36.8% vs 73.0%, P = .02) and mammography screening (0% vs 63.2%, P = .003). Use of select preventive care services in 365 days prior to surgery in PC versus non-PC patients.
The proportion of patients who attended at least one postoperative follow-up appointment was significantly lower for PC patients than for their non-PC counterparts (Figure 2). Subgroup analysis by age and surgical specialty also reflected this trend with PC patients <15 years being statistically less likely to attend a follow up appointment (33/69, 47.8%) compared to non-PC patients of that age group (90/119, 75.6%, P = .0001). Attendance at postoperative follow-up visit in PC versus non-PC patients.
Discussion
The goal of this study was to evaluate postoperative outcomes of our previously understudied local PC population in comparison with a matched non-PC population. We were also interested in the use of preventative and postoperative services by the PC population. The analysis revealed that although PC patients had lower rates of select preventive care services and were less likely to attend postoperative follow-up visits, postoperative morbidity and mortality rates were similar to those of the non-PC population.
As is true for many culturally diverse groups, working with the PC population in a contemporary medical setting requires sensitivity and adaptation of what may be considered a “typical” patient approach. Traditionally, English is a second language, and formal education is completed through an 8th-grade level. Health literacy has not been studied in-depth in this population. The only available study on this topic was conducted among Ohio Appalachia Amish communities, which found that more than half of Amish participants had limited or marginal health literacy. 20 Perhaps because they often minimize contact with the health care system, fewer typical data or health care variables are gathered from them than from a non-PC patient. For example, some of the PC patients in our dataset did not have a documented BMI. Similarly, identification of postoperative complications may be lower than expected because events such as a stitch abscess, superficial wound dehiscence, and other minor complaints may be handled by a local healer instead of returning to the primary health care system.
Our regional PC surgical patient population is less likely to attend postoperative follow-up appointments. Despite this, there is not an overall statistically significant difference in 30-day morbidity or mortality between the groups. These findings may be due partly to the relatively low number of PC patients captured in this dataset or, as noted above, to alternate follow-up methods. Accordingly, the finding of a lower rate of postoperative follow-up for our PC patients is unsurprising. Often for more routine procedures (eg, uncomplicated appendectomy, cholecystectomy, or hernia repair), PC patients will follow up with a local health care provider outside of our system or with a healer instead of making the often lengthy trip back to the primary health care system. This is highly dependent on the relationship between the PC community and local healers with the regional health care system. With development of good relationships, PC community members are often comfortable with local follow-up and a provision to call or return should they have any unanticipated difficulties during the postoperative course.
We did note a statistically significant higher number of PC patients that underwent emergent, urgent, or trauma surgery. Anecdotally, the higher rates of emergent and urgent surgeries is unsurprising, as often we anticipate that these patients may present later or with further disease progression after having tried home remedies and/or alternatives therapies prior to presentation.
According to the results of this study, preventive health care/screening procedures are not as well attended by the PC population. As with postoperative care, this may be in part due to PC patients pursuing screening studies at local or alternate health care sites outside of our system. Despite the lower use of preventive services, limited literature has demonstrated a similar life expectancy between PC and non-PC persons and lower cancer rates among the PC population. Based on the costs associated with transportation and services for preventive measures, these findings beg the question of whether routine cancer screening should be applied to the PC population in the same way as it is to the non-PC population.
Our study has several limitations. Part of the difficulty in studying the PC population is their relative isolation from contemporary medicine and medical facilities due to mostly rurally located domestic locations. The PC population will typically approach illness in what may be considered a stepwise fashion. Home remedies and local healers are appealed to prior to contact with a medical facility. To our knowledge, a database or other health care tracking or analysis has not been created to evaluate postoperative complications in this population. The relatively low number of patients included in this study limits its power and applicability. Additionally, we suspect that the PC population has a better baseline level of health given known lower levels of cardiovascular disease and cancer; this may inherently decrease risks of some postoperative complications and subsequently decrease the likelihood of finding a significant difference in outcomes even if they do experience more health care disparity. Expanding the capture of these patients for a better powered study would allow us to draw more reliable conclusions regarding their use of contemporary health care services. Further iterations and expansion of this research will examine financial and additional health care utilization data to answer this and additional questions regarding this growing minority population. Optimally, more research, as well as development and nurturing of community partnerships with PC will help determine better ways to appropriately engage and continue improving efficient and meaningful care for the PC population.
The PC population is a unique minority that has not yet been well represented in health care quality and outcomes research despite a rapidly growing population, particularly in the Midwest. This growth makes awareness of their cultural intricacies more important than ever for frontline health care workers. We found that the majority of postoperative outcomes for PC patients do not differ significantly from that of the non-PC postoperative patient, despite less frequent follow-up appointments. Additionally, they do not use preventive care services as often as non-PC patients. We suspect that these findings are likely reflective of PC cultural tendencies to limit contact with modern health care systems if not necessary; however, more research is needed to delineate the precise reasons for these trends.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.