Abstract
Study design
A retrospective cross-sectional study.
Objective
To identify who prescribes outpatient antibiotics among a primary care spinal cord injury (SCI) cohort.
Setting
ICES databases in Ontario, Canada.
Methods
A cohort of individuals with SCI were retrospectively identified using a tested-algorithm and chart reviews in a primary care electronic medical records database. The cohort was linked to a drug dispensing database to obtain outpatient antibiotic prescribing information, and prescriber details were obtained from a physician database.
Results
Final cohort included three hundred and twenty individuals with SCI. The average annual number of antibiotic courses dispensed for the SCI cohort was 2.0 ± 6.2. For dispensed antibiotics, 58.9% were prescribed by rostered-primary care practice physicians, compared to 17.9% by emergency and non-rostered primary care physicians, 17.4% by specialists and 6.1% by non-physician prescribers. Those who lived in urban areas and rural areas, compared to those who lived in suburban areas, were more likely to receive antibiotics from emergency and non-rostered primary care physicians than from rostered-primary care practice physicians.
Conclusion
Although individuals with SCI received outpatient antibiotic prescriptions from multiple sources, physicians from an individual’s rostered-primary care practice were the main antibiotic prescribers. As such, interventions to optimize antibiotics use in the SCI population should target the primary care practice.
Subject terms: Health services, Spinal cord diseases
Introduction
Individuals with spinal cord injuries (SCI) may be frequent antibiotic users due to recurrent infections, and as a result may have high rates of antimicrobial resistance (AMR) bacteria [1–8]. Since individuals with SCI often exhibit multidrug resistant bacteria, they are more likely to experience ineffective treatment or require antibiotic prescription revisions after susceptibility test results [3, 9]. They may also experience negative side effects, such as allergic reactions or diarrhea, which require follow-up care [10, 11]. Therefore, individuals with SCI should have adequate continuity of care to ensure appropriate follow-up after antibiotic prescriptions.
Primary care providers (PCPs) are intended to be the main point of contact for disease prevention, illness diagnoses, and medical treatment, as well as to coordinate care between other healthcare providers, such as specialists for individuals with SCI [12]. Several studies examined the type of care provided by PCPs and specialists for the SCI population [13–16]. A study investigating preventive health-care services for three SCI cohorts in Canada, United States (US) and United Kingdom (UK) reported 93% had a PCP, and 63% had an SCI specialist to manage their care [13]. They also reported individuals with SCI were more likely to visit their PCP for new issues, previous medical conditions, providing urine specimens, bowel/bladder problems or annual physicals than a SCI specialist [13]. However, another study, which surveyed PCPs, found they had varying levels of comfort in managing SCI-related conditions, and approximately half of these PCPs reported their patients with SCI were co-managed by specialists [14]. Similarly, a study surveying healthcare providers of SCI patients reported PCPs felt a lack of knowledge about optimal SCI care in primary care [15]. Physicians felt the care of individuals with SCI should be a shared between PCPs, specialists, and other non-physician healthcare professionals [15]. In terms of frequency of visits, the study conducted in the SCI cohort in Canada, US and UK found almost half of individuals had visited a specialist within the last year, and on average individuals visited 1.5 different specialists annually [13]. Another study in Canada found on average patients had 5.4 annual visits to their PCP, and 2.5 annual visits to another healthcare professional, which includes specialists [14]. Although it is clear specialists and PCPs are both important for managing care, there is no published research on their role in antibiotic prescribing.
Beyond an individual’s designated PCP and specialists, it is possible that individuals with SCI access other primary care services, such as walk-in clinics and emergency departments for antibiotics. A study conducted in Canada in a cohort of 559 individuals with traumatic SCI (TSCI) found within one-year after their initial injury they had an average of 11.6 visits to a PCP, 20.1 visits to a specialist and also 1.1 visits to the emergency department [16]. A study conducted in individuals with TSCI found, approximately half of the 4,403 emergency department visits over six-years were used as an inappropriate substitute for primary care [17]. Another study conducted in emergency departments in the US from 2002 to 2007 in veterans with SCI found 22.8% of their visits resulted in antibiotics prescritpions [18]. While emergency departments, walk-in clinics or other urgent care centers do not require patients to make appointments, there is also no pre-existing relationship between the patient and the healthcare provider. As such, these services may disrupt the continuity of care patients receive. Previous research indicates individuals with better continuity of care have lower rates of hospitalization and mortality, as well higher rates of preventive services and patient satisfaction [19–24]. There is no published research on whether individuals with SCI use emergency departments or other primary care services to receive antibiotics prescriptions despite being rostered to a PCP. Understanding who prescribes outpatient antibiotics for the SCI population is important, as it can aid with the creation of antibiotic consumption optimization strategies for the SCI population.
The purpose of this paper is to identify who prescribes dispensed antibiotics among a rostered primary care SCI cohort. The specific objectives are (1) to describe types and amount of antibiotics dispensed, by prescriber and recipient, and (2) to determine patient factors associated with receiving antibiotics from other primary care or emergency department services.
Methods
This is a retrospective cross-sectional study to identify prescribers of dispensed antibiotics for a primary care SCI cohort using linked health administrative databases at ICES (formerly known as the Institute for Clinical Evaluative Sciences), in Ontario, Canada from January 1, 2013 to December 31, 2015. ICES is an independent, non-profit research institute, funded by an annual grant from the Ontario Ministry of Health (MOH) and the Ministry of Long-Term Care (MLTC), whose legal status under Ontario’s health information privacy law allows it to collect and analyze health care and demographic data, without consent, for health system evaluation and improvement. The use of the data in this project is authorized under section 45 of Ontario’s Personal Health Information Protection Act (PHIPA) and does not require any additional review by a Research Ethics Board.
Cohort development
A previous study investigating primary care antibiotic prescribing in the SCI population identified a cohort of 432 individuals with SCI from the Electronic Medical Records Primary Care (EMRPC) database (previously known as EMRald) at ICES using a validated and tested-algorithm and chart reviews to verify the cases [25]. The original cohort of 432 individuals with a SCI from this previous study was linked to the Ontario Drug Benefit (ODB) database to obtain information on dispensed antibiotics which contains prescription drug claims received under the Ontario Drug Benefit program [26]. As such only 320 individuals eligible for ODB comprised the cohort for this current study. If an individual was eligible for ODB and filled at least one drug prescription, they could be identified in the ODB database. The ODB database also lists the prescribing healthcare providers’ identification number for any prescribed antibiotics. The EMRPC database allows for the identification of patients’ rostering-PCP. By rostering to a PCP, patients agree to obtain medical care only from their PCP, unless in an emergency situation, and PCPs agree to provide comprehensive care for their rostered-patients [27]. Essentially rostering is when a patient formally registers with a PCP and/or clinic with the intention they mainly receive primary care from their rostering-PCP. All individuals with a confirmed SCI in the EMRPC database, who could be linked to the ODB database were included in this study. As such, patients were identified first using a tested-algorithm and chart reviews, then the cohort was linked to the ODB database to obtain their full antibiotic dispensing information, and then prescribers of these antibiotics were identified.
Demographic information, such as sex, date of birth, and primary postal code for the SCI study cohort was identified in the Registered Persons Database (RPDB), which contains this information about anyone who has ever received an Ontario Health Insurance Plan (OHIP) card [26]. The Rurality Index for Ontario (RIO), a widely used scale to determine rurality based on an individual’s postal codes and on an area’s population factors (e.g. size, number of hospital, and number of physicians) and access to healthcare services, was used to classify urban, suburban and rural areas for this study [28]. A RIO score of less than 10 for urban, 10–39 for suburban, and 40 and above for rural was used [28]. Clinical and injury characteristics for this SCI cohort were abstracted via chart review, which included injury duration, type of SCI injury (TSCI vs. non-traumatic SCI (NTSCI)), catheter use, and type of impairments (sensorimotor, sensory only, bladder and bowel) [29]. Catheter use included both indwelling catheters and intermittent catheterization anytime as indicated in the patient chart during the study period. Sensorimotor impairments included loss of fine motor movement, weakness, use of mobility aids, functional loss of arms or legs, paralysis, spasticity, paraplegia, tetraplegia, and quadriplegia as indicated in the patient chart [30]. Sensory only impairments was defined as pain, burning sensations, tingling, loss of sensation, and numbness, without any motor impairements [30]. Bladder impairments was defined as incontinence, urinary retention, bladder urgency, neurogenic bladder, and spastic bladder [30]. Bowel impairments was defined as the need for a bowel routine, bowel incontinence, loss of sensation in bowel area, and neurogenic bowel [30].
Antibiotic dispensing information
All antibiotic dispensing information for this study was obtained in the ODB database for the SCI study cohort. The ODB database contains all medication claims funded through the ODB program, including antibiotics [31]. It contains information on the type of drug dispensed, number of days supplied, the cost of the drug, and the dispensed date of the drug [31]. The ODB database is over 99% accurate for correctly identifying dispensed prescriptions compared to manual chart abstraction [31]. The antibiotics dispensed in ODB were identified using the DRUGLIST file, which contains an exhaustive list of Drug Identification Numbers (DIN) used in Canada from 1990 until present day. Any antibiotic dispensed in ODB from January 1, 2013, to December 31, 2015, among the SCI study cohort was identified and variables collected included antibiotic name, quantity, days of supply, and date of prescription. For antibiotics dispensed, average annual antibiotic days of therapy (DOT) and total antibiotic DOT by prescriber type was determined. DOT was calculated by adding together antibiotic prescription days of supply for a given individual patient, or for an antibiotic prescriber type (described below). Antibiotics were sorted into these classes: (1) fluoroquinolones, (2) nitrofurantoin, (3) cephalosporins, (4) penicillin, (5) sulfonamides, trimethoprim & combinations, (6) macrolides, (7) tetracycline, (8) clindamycin, and (9) other.
Prescriber information
Dispensed antibiotics were linked to the prescribing physician in the ODB database using an encoded physician billing number. General demographic information about these prescribing physicians was obtained from the ICES Physician Database (IPDB). The IPDB contains information regarding physician date of birth, sex, main practice speciality, years of practice and Canadian Medical School Graduate (CMG) or International Medical Graduate (IMG) [26].
For this study, the physician’s main practice speciality was used to group prescribers. PCPs from the EMRPC database were either a patient’s rostering-PCP or another physician practicing in the same clinic as the rostering-PCP. All PCPs in the EMRPC database practice in a model of care that rosters patients to a specific PCP. In these care models, colleagues can cover each other and see each other’s patients without the rostering-PCP being financially penalized. Therefore, the rostering-PCP and other physicians working in the same EMRPC primary care clinic comprised the “rostered-primary care practice physicians” category.
If a physician’s speciality was primary care or general practice, but they practiced outside of the rostered primary care practice, they were labeled as non-rostered PCP. If a physician’s speciality was emergency medicine, they were labeled as emergency physicians. Together, the non-rostering PCP and emergency physicians comprised the “emergency and non-rostered primary care physicians” category.
All other physician specialities were grouped under “specialists”. Other antibiotic prescribers such as nurse practitioners or dentists, who do not have a physician billing number, and as such cannot be linked in the IPDB to obtain demographic information, were grouped as “non-physicians”.
Data analysis
Means, standard deviations and proportions were used to describe the demographic, injury and clinical characteristics for the SCI cohort and physicians. Difference between groups in the SCI cohort were calculated using chi-square, T-test, and Mann–Whitney U test.
A clustered logistic regression was conducted to see if any patient factors (age, sex, duration of injury, rurality, bladder impairment, bowel impairment, sensorimotor impairment, NTSCI vs. TSCI, prophylaxis use, and catheter use) were associated with receiving a dispensed antibiotic from rostered-primary care practice physicians, compared to emergency and non-rostered primary care physicians. Clustering was done at the individual level to account for multiple antibiotics dispensed to the same individual. An assessment of multicollinearity was done to ensure factors exhibiting multicollinearity (p ≤ 0.05) were not modeled together. Multiple iterations of the regression analyses were conducted to identify key factors to include in the final analyses.
Results
Of the 432 individuals with SCI, 320 were identified in the ODB database, with those found in ODB to be older and have a longer duration of injury. Of the 320 individuals identified in ODB, 185 (57.8%) received at least one dispensed antibiotic between Jan 1, 2013, and Dec 31, 2015. See Table 1 for descriptive characteristics of those who were successfully linked to ODB. Those who received one or more antibiotics, compared to those who received no antibiotics, were significantly older and more likely to use catheters. A total of 1930 antibiotics and 24,024 DOT were dispensed. The average individual with SCI in the ODB database received 2.0 (SD: 6.2, Q1 = 0, Median = 0.3, Q3 = 1.7) antibiotics annually. See Table 2 for more detailed information regarding annual number of antibiotics prescribed and DOT.
Table 1.
Demographic, injury and clinical characteristics of SCI cohort clustered by received zero versus one or more dispensed for study duration (Jan 1, 2013–Dec 31, 2015).
| No antibiotics dispensed (n = 135) | One or more antibiotics dispensed (n = 185) | Full ODB SCI cohort (n = 320) | |
|---|---|---|---|
| Characteristic | % | % | % |
| Age at end of 2013* | |||
| Under 40 | 13.3 | 17.8 | 15.9 |
| 40–60 | 60.7 | 28.1 | 41.9 |
| Over 60 | 25.9 | 54.1 | 42.2 |
| Sex | |||
| Female | 40.7 | 41.6 | 41.3 |
| Male | 59.3 | 58.4 | 58.8 |
| Rurality | |||
| Urban | 65.2 | 61.6 | 63.1 |
| Suburban | 17.8 | 24.3 | 21.6 |
| Rural | 17.0 | 14.1 | 15.3 |
| Type of Injury | |||
| NTSCI | 65.9 | 62.7 | 64.1 |
| TSCI | 34.1 | 37.3 | 35.9 |
| Duration of Injury as of January 1, 2013 | |||
| Less than 1 year | 24.4 | 18.4 | 20.9 |
| 1–10 years | 32.6 | 29.2 | 30.6 |
| 11–20 years | 11.9 | 17.3 | 15.0 |
| Over 20 years | 31.1 | 35.1 | 33.4 |
| Catheter User* | |||
| Yes | 60.7 | 52.4 | 46.9 |
| No | 39.3 | 47.6 | 53.1 |
| Impairments | |||
| Sensorimotor | 87.4 | 88.1 | 87.8 |
| Sensory Only | 12.6 | 11.9 | 12.2 |
| Bladder | 65.9 | 70.8 | 68.8 |
| Bowel | 30.4 | 35.1 | 33.1 |
*Indicates significant (p < 0.05) difference between individuals who received no dispensed antibiotics versus individuals who had one or more antibiotic dispensed.
**To avoid small cells from being re-calculated (less than six), missing data for rurality was listed as urban, and missing data for duration of injury as over 20 years. Unclassifiable cases of SCI were listed under the TSCI category.
Table 2.
Average annual number of dispensed antibiotics and DOT for the SCI cohort.
| Characteristic | SCI cohort (n = 320) | |
|---|---|---|
| Average Annual Number of Antibiotic Prescriptions | 2.0 ± 6.2 | |
| Percent Breakdown | n | % |
| None | 135 | 42.2 |
| <3 | 79 | 24.7 |
| Between 3 and 6 | 41 | 12.8 |
| >6 | 65 | 20.3 |
| Average Annual Days of Therapy (DOT) for Antibiotics | 25.0 ± 69.0 | |
| Percent Breakdown | n | % |
| None | 135 | 42.2 |
| 1–30 | 135 | 42.2 |
| 31–120 | 33 | 10.3 |
| >120 | 17 | 5.3 |
Antibiotics were prescribed by 512 unique physicians: however, 118 of these physicians were missing demographic and specialty information from the IPDB. See Table 3 for the physician demographic information. Rostered-primary care practice physicians prescribed 58.9% of dispensed antibiotics, while emergency and non-rostered primary care physicians prescribed 17.9%, specialists prescribed 17.4% and non-physicians prescribed 6.1%. See Table 4 for number of antibiotics and DOT for dispensed antibiotics grouped by prescriber.
Table 3.
SCI cohort antibiotic prescriber physician demographic information.
| Physicians in rostered-primary care practice (n = 133) | Emergency & non-rostered primary care physicians (n = 220) | Specialist (n = 159) | Total (n = 512) | |
|---|---|---|---|---|
| Characteristic | % | % | % | % |
| Age at end of 2013 | ||||
| 39 & under | 38.3 | 51.4 | 20.1 | 38.3 |
| 40–59 | 48.9 | 37.7 | 28.3 | 37.7 |
| 60+ | 12.8 | 10.9 | 5.7 | 9.8 |
| Missing | – | – | 45.9 | 14.3 |
| Sex | ||||
| Female | 45.1 | 37.3 | 17.6 | 33.2 |
| Male | 54.9 | 62.7 | 36.5 | 52.5 |
| Missing | – | – | 45.9 | 14.3 |
| Years of Practise at end of 2013 | ||||
| <11 | 24.8 | 41.4 | 7.6 | 26.6 |
| 11–25 | 45.9 | 30.5 | 27.0 | 33.4 |
| >25 | 29.3 | 28.2 | 19.5 | 25.8 |
| Missing | – | – | 45.9 | 14.3 |
| Training Location | ||||
| Canada | 90.2 | 71.8 | 43.4 | * |
| International | 9.8 | 17.7 | 6.9 | * |
| Missing | * | 10.5 | 49.7 | * |
*To avoid small cells from being re-calculated (less than six), missing data for training location for “Physicians in Rostered-Primary Care Practice” was listed as Canada.
Table 4.
Number of dispensed antibiotics and DOT based on Prescriber Speciality for the SCI cohort.
| Prescriber speciality | Antibiotics dispensed (n = 1930) |
|---|---|
| % | |
| Physicians in Rostered-Primary Care Practice | 58.9 |
| Rostered-PCP | 53.2 |
| Other PCP within rostered primary care practice | 5.6 |
| Emergency & Non-Rostered Primary Care Physicians | 18.0 |
| Non-rostered PCP | 14.5 |
| Emergency Medicine Speciality | 3.5 |
| Specialist | 17.4 |
| Urology | 3.2 |
| Infectious Diseases | 1.3 |
| Physical Medicine and Rehabilitation | 0.7 |
| Nephrology | 2.0 |
| Surgery (general, plastic, neurology, orthopedic) | 0.9 |
| Obstetrician-Gynecologists | 0.5 |
| Other Speciality | 2.7 |
| Unknown Speciality | 6.1 |
| Non-physician | 6.1 |
Nitrofurantoin was the top antibiotic class prescribed for the SCI cohort at 21.7%, followed by fluoroquinolones at 18.4%, and cephalosporins at 16.6%. Table 5 outlines the antibiotic class by prescriber type. The most common antibiotic class prescribed by rostered-primary care practice physicians was nitrofurantoin. Emergency and non-rostered primary care physicians, as well as specialists prescribed fluoroquinolones most frequently. Non-physicians most frequently prescribed penicillin.
Table 5.
Select antibiotic class prescribed based on Prescriber Speciality for the SCI cohort.
| Physicians in rostered-primary care practice (n = 1136) | Emergency & non-rostered primary care physicians (n = 347) | Specialist (n = 330) | Non-physician (n = 117) | |
|---|---|---|---|---|
| Antibiotic Categories | % | % | % | % |
| Fluoroquinolones | 16.9 | 25.1 | 19.7 | 9.4 |
| Nitrofurantoin | 27.6 | 17.3 | 11.5 | * |
| Cephalosporin | 17.7 | 16.4 | 15.5 | 9.4 |
| Penicillin | 8.6 | 11.0 | 18.5 | 47.9 |
| Sulfonamides, Trimethoprim & combinations | 11.0 | 19.6 | 17.9 | 6.8 |
| Macrolides | 3.1 | 4.9 | 3.9 | * |
| Tetracycline | 13.8 | * | * | * |
| Clindamycin | * | 3.5 | 8.5 | 13.7 |
*Since absolute cell count was less than six or to avoid small cells from being re-calculated information was not included in the table
Only select antibiotic classes were reported.
A clustered logistic regression was used to assess age, sex, and rurality to determine factors associated with receiving a dispensed antibiotic from rostered-primary care practice physicians, compared to emergency and non-rostered primary care physicians, and found rurality to be significant. Our findings showed those who lived in urban areas (RR = 4.2, 95% CI: 1.4–12.4, p = 0.01) and rural areas (RR = 4.1, 95% CI: 1.1–16.0, p = 0.04) were more likely to receive antibiotics from emergency and non-rostered-primary care physicians than from rostered-primary care practice physicians compared to those who live in suburban areas (Table 6).
Table 6.
Patient’s rurality effect on receiving antibiotics from emergency and non-rostered primary care physicians vs. rostered-primary care practice physicians.
| Proportion of antibiotics prescribed | Multivariate clustered logistic regression significant results | ||||
|---|---|---|---|---|---|
| Rurality | Physicians in rostered-primary care practice (%) | Emergency & non-rostered primary care physicians (%) | RR | 95% CI | P value |
| Urban | 47.5 | 22.7 | 4.2 | 1.4–12.4 | 0.01 |
| Rural | 43.5 | 20.5 | 4.1 | 1.1–16.0 | 0.04 |
| Suburban | 78.2 | 10.6 | Ref | ||
For proportions the remaining antibiotics were prescribed by specialists or non-physician prescribers, which were not included in the table.
Discussion
Approximately 60% of individuals with SCI in this study filled at least one outpatient antibiotic prescription. These individuals were older and more likely to be catheter users. On average they received two dispensed antibiotic courses annually. The rostering primary care practice prescribed the most antibiotics, while emergency and other primary care services accounted for the second most antibiotics, followed by specialists, and then non-physician prescribers. Compared to individuals with SCI who resided in suburban areas, those who lived in urban areas and rural areas were approximately four times more likely to receive antibiotics from emergency and non-rostered-primary care physicians.
Individuals with SCI received their antibiotic prescriptions from multiple sources in the community; however, physicians from their rostering primary care practice accounted for the majority of antibiotics prescribed. As such, interventions created to optimize antibiotics use in the SCI population should target primary care physicians. It should also be noted that specialists play an important role in the prescription of outpatient antibiotics in the SCI population as they accounted for 17.4% of antibiotic prescriptions. However, no one speciality accounted for greater than 4.0% of prescribed antibiotics. As such, targeting a particular speciality for intervention may not significantly impact antibiotic prescribing in the SCI population.
This study found 18% of antibiotics were prescribed by emergency and non-rostering primary care physicians. Beyond the initial antibiotic prescription, it is important that individuals with SCI have adequate continuity of care to address potential side effects, ineffective treatment, and susceptibility test results. Although it may be unavoidable, the use of emergency and other primary care services for obtaining their antibiotics, instead of their rostering primary care clinic, may disrupt the continuity of care individuals with SCI receive. Previous literature indicates those with higher continuity of care defined as receiving care from the same PCP have better patient outcomes and patient satisfaction [19–24]. A survey conducted in 2007 in the US with 3,436 adults with a PCP found continuity of care improves coordination of care with specialist visits [32]. As suggested from our study, specialists are key prescribers of antibiotics in the SCI population; hence, appropriate coordination is important. Continuity of care may also be extended from a single physician to a clinic or medical group. A study conducted in the US for insured patients from 2005–2009 found those with high medical group continuity, despite being older and having more comorbidities, compared to those with medium or low continuity, had lower inpatient expenditures, emergency services utilization, and total medical costs [33]. As such, if an individual is unable to access their rostering-PCP, it may be better to seek treatment from another physician in the same medical practice to maintain continuity of care, rather than from emergency, walk-in clinic or other urgent care services. If the use of emergency and non-rostering primary care physicians is unavoidable, it is important that individuals with SCI receive adequate follow-up post-antibiotic prescription to preserve continuity of care. Ensuring that the rostered-PCP is notified of the visit and test results may be a simple way to maintain continuity of care.
It was found that those in rural and urban areas compared to suburban areas were significantly more likely to access emergency and non-rostering primary care physicians for their antibiotic prescriptions. Healthcare practice patterns and access to health care services differs across urban, suburban, and rural areas. If an individual is unable to receive care from their primary care practice, urban areas have access to walk-in clinics; however, emergency departments are the only option for rural patients [34]. Previous literature indicated individuals with SCI living in rural areas are more likely to access emergency services as a substitute for primary care services [17]. A qualitative study conducted with rural, suburban, and urban primary care physicians in the US found that distance for rural patients and competing multiple healthcare systems for urban patients were major obstacles for continuity of care [35]. As such, distance and access to their primary health care clinic may be an issue preventing those in rural areas from visiting their PCP. PCP in rural areas not only provide care in primary care clinics, but also deliver care in the hospital or emergency department settings [34, 36]. Hence, PCPs may work in emergency departments in rural areas in addition to their office practice. Frequently rural primary care practices have limited clinic hours as well [34]. Patients in urban settings may have more options, such as walk-in or drop-in clinics, compared to patients in suburban areas, and accessing these services may be more convenient than visiting their PCP [34, 35]. Future research is needed to further understand these differences by rurality and why individuals with SCI in urban and rural areas visit emergency and other primary care services, instead of their rostering-primary care practice, to receive antibiotic prescriptions.
Since this study used administrative data, we are limited by the information available in the databases. We were missing demographic information for 118 specialists; however, we had full demographic information for all primary care physicians. Not all individuals from the original SCI cohort in EMRPC could be linked to the ODB database; however, we found nearly three-quarters could be linked. Additionally, the ODB database does not provide information about indication for antibiotic prescription. As such, based on the available information, the study cannot comment regarding the appropriateness of reasons for receiving antibiotics, and/or the type of antibiotic classes prescribed. The ODB database records dispensed antibiotics; therefore, it is unclear if individuals consumed the dispensed antibiotics they received. Physician main speciality was also used to determine the setting in which an antibiotic was prescribed; however, it is possible a physician was practicing in a different setting than their main speciality at the time of prescription.
This study identified the rostering-primary care practice as a key prescriber for outpatient antibiotics in an outpatient SCI population. More research is needed in different SCI populations to verify findings from this study; however, this paper provides a good starting point for further exploration. To our knowledge this is the first paper to investigate outpatient antibiotic prescribing in an SCI population living in the community. Future research should aim to understand issues with accessing the rostering-primary care practice for antibiotic prescriptions and treatment outcomes as this may have implications for development of interventions on continuity of care.
Acknowledgements
This study was supported by the ICES, which is funded by an annual grant from the Ontario Ministry of Health (MOH) and the Ministry of Long-Term Care (MLTC). We thank IQVIA Solutions Canada Inc. for use of their Drug Information File. Parts of this material are based on data and/or information compiled and provided by CIHI. However, the analyses, conclusions, opinions and statements expressed in the material are those of the authors, and not necessarily those of CIHI. Thank you to Jun Guan, Zhiyin Li, and staff at ICES for their contributions to this paper.
Author contributions
All authors contributed to the development of this paper, its study design and interpretation of results. AS conceptualized the study, analyzed the data, interpreted results, and wrote the manuscript. SBJ guided the data analyses and development of manuscript. SBJ, KT, MP, AMM, and BCC helped with the overall interpretation of findings, as well as reviewed and edited the manuscript.
Data availability
The dataset from this study is held securely in coded form at ICES. While legal data sharing agreements between ICES and data providers prohibit ICES from making the dataset publicly available, access may be granted to those who meet pre-specified criteria for confidential access, available at www.ices.on.ca/DAS (email: das@ices.on.ca). The full dataset creation plan and underlying analytic code are available from the authors upon request, understanding that the computer programs may rely upon coding templates or macros that are unique to ICES and are therefore either inaccessible or may require modification.
Competing interests
The authors declare no competing interests.
Ethical approval
The use of data in this project was authorized under section 45 of Ontario’s Personal Health Information Protection Act, which does not require further review by a Research Ethics Board.
Footnotes
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Suda KJ, Patel UC, Sabzwari R, Cao L, Ramanathan S, Hill JN, et al. Bacterial susceptibility patterns in patients with spinal cord injury and disorder (SCI/D): an opportunity for customized stewardship tools. Spinal Cord. 2016;54:1001–9. doi: 10.1038/sc.2016.38. [DOI] [PubMed] [Google Scholar]
- 2.Salameh A, Mohajer MA, Daroucihe RO. Prevention of urinary tract infections in patients with spinal cord injury. Can Med Assoc J. 2015;187:807–11. doi: 10.1503/cmaj.141044. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Abbasi F, Korooni S Infectious Complications after Spinal Cord Injury. In: Dionyssiotis Y (ed). Essentials of Spinal Cord Injury Medicine. InTech, 2018 10.5772/intechopen.72783.
- 4.Evans C, Weaver F, Rogers T, Rapacki L, Miskevics S, Hahm B, et al. Guideline-recommended management of community-acquired pneumonia in veterans with spinal cord injury. Top Spinal Cord Inj Rehabil. 2012;18:300–5. doi: 10.1310/sci1804-300. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Heym B, Rimareix F, Lortat-Jacob A, Nicolas-Chanoine M-H. Bacteriological investigation of infected pressure ulcers in spinal cord-injured patients and impact on antibiotic therapy. Spinal Cord. 2004;42:230–4. doi: 10.1038/sj.sc.3101568. [DOI] [PubMed] [Google Scholar]
- 6.Fitzpatrick MA, Suda KJ, Safdar N, Burns SP, Jones MM, Poggensee L, et al. Changes in bacterial epidemiology and antibiotic resistance among veterans with spinal cord injury/disorder over the past 9 years. J Spinal Cord Med. 2018;41:199–207. doi: 10.1080/10790268.2017.1281373. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Evans CT, Fitzpatrick MA, Jones MM, Burns SP, Poggensee L, Ramanathan S, et al. Prevalence and Factors Associated With Multidrug-Resistant Gram-Negative Organisms in Patients With Spinal Cord Injury. Infect Control Hosp Epidemiol. 2017;38:1464–71. doi: 10.1017/ice.2017.238. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Yoon SB, Lee BS, Lee KD, Hwang SI, Lee HJ, Han ZA. Comparison of bacterial strains and antibiotic susceptibilities in urinary isolates of spinal cord injury patients from the community and hospital. Spinal Cord. 2014;52:298–301. doi: 10.1038/sc.2014.10. [DOI] [PubMed] [Google Scholar]
- 9.Garcia-Arguello LY, O’Horo JC, Farrell A, Blakney R, Sohail MR, Evans CT, et al. Infections in the spinal cord-injured population: a systematic review. Spinal Cord. 2017;55:526–34. doi: 10.1038/sc.2016.173. [DOI] [PubMed] [Google Scholar]
- 10.Wong S, Santullo P, O’Driscoll J, Jamous A, Hirani SP, Saif M. Use of antibiotic and prevalence of antibiotic-associated diarrhoea in-patients with spinal cord injuries: a UK national spinal injury centre experience. Spinal Cord. 2017;55:583–7. doi: 10.1038/sc.2016.193. [DOI] [PubMed] [Google Scholar]
- 11.Bayoumi I, Dolovich L, Hutchison B, Holbrook A. Medication-related emergency department visits and hospitalizations among older adults. Can Fam Physician. 2014;60:e217–22. [PMC free article] [PubMed] [Google Scholar]
- 12.McColl MA, Aiken A, McColl A, Sakakibara B, Smith K. Primary care of people with spinal cord injury: scoping review. Can Fam Physician. 2012;58:1207–16. [PMC free article] [PubMed] [Google Scholar]
- 13.Donnelly C, McColl MA, Charlifue S, Glass C, O’Brien P, Savic G, et al. Utilization, access and satisfaction with primary care among people with spinal cord injuries: a comparison of three countries. Spinal Cord. 2007;45:25–36. doi: 10.1038/sj.sc.3101933. [DOI] [PubMed] [Google Scholar]
- 14.Lofters A, Chaudhry M, Slater M, Schuler A, Milligan J, Lee J, et al. Preventive care among primary care patients living with spinal cord injury. J Spinal Cord Med. 2019;42:702–8. doi: 10.1080/10790268.2018.1432308. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Milligan J, Lee J, Hillier LM, Slonim K, Craven C. Improving primary care for persons with spinal cord injury: development of a toolkit to guide care. J Spinal Cord Med. 2020;43:364–73. doi: 10.1080/10790268.2018.1468584. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Munce SEP, Guilcher SJT, Couris CM, Fung K, Craven BC, Verrier M, et al. Physician utilization among adults with traumatic spinal cord injury in Ontario: a population-based study. Spinal Cord. 2009;47:470–6. doi: 10.1038/sc.2008.173. [DOI] [PubMed] [Google Scholar]
- 17.Guilcher SJT, Craven BC, Calzavara A, McColl MA, Jaglal SB. Is the emergency department an appropriate substitute for primary care for persons with traumatic spinal cord injury? Spinal Cord. 2013;51:202–8. doi: 10.1038/sc.2012.123. [DOI] [PubMed] [Google Scholar]
- 18.Evans CT, Rogers TJ, Chin A, Johnson S, Smith B, Weaver FM, et al. Antibiotic prescribing trends in the emergency department for veterans with spinal cord injury and disorder 2002-2007. J Spinal Cord Med. 2013;36:492–8. doi: 10.1179/2045772312Y.0000000076. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Fan VS, Burman M, McDonell MB, Fihn SD. Continuity of care and other determinants of patient satisfaction with primary care. J Gen Intern Med. 2005;20:226–33. doi: 10.1111/j.1525-1497.2005.40135.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Pereira Gray DJ, Sidaway-Lee K, White E, Thorne A, Evans PH. Continuity of care with doctors—a matter of life and death? A systematic review of continuity of care and mortality. BMJ Open. 2018;8:e021161. doi: 10.1136/bmjopen-2017-021161. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.McIsaac WJ, Fuller-Thomson CE, Talbot Y. Does having regular care by a family physician improve preventive care? Can Fam Phys.47:70–76. [PMC free article] [PubMed]
- 22.Saultz JW. Interpersonal continuity of care and care outcomes: a critical review. Ann Fam Med. 2005;3:159–66. doi: 10.1370/afm.285. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Baker R, Freeman GK, Haggerty JL, Bankart MJ, Nockels KH. Primary medical care continuity and patient mortality: a systematic review. Br J Gen Pr. 2020;70:e600–e611. doi: 10.3399/bjgp20X712289. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Van Walraven C, Oake N, Jennings A, Forster AJ. The association between continuity of care and outcomes: a systematic and critical review: Association between continuity of care and outcomes. J Eval Clin Pr. 2010;16:947–56. doi: 10.1111/j.1365-2753.2009.01235.x. [DOI] [PubMed] [Google Scholar]
- 25.Senthinathan A, Craven BC, Morris AM, Penner M, Tu K, Jaglal SB. Examining antibiotic prescribing and urine culture testing for urinary tract infections (UTIs) in a primary care spinal cord injury (SCI) cohort. Spinal Cord. 2023. 10.1038/s41393-023-00899-x. [DOI] [PubMed]
- 26.ICES Data Dictionary online.
- 27.Singh J, Dahrouge S, Green ME. The impact of the adoption of a patient rostering model on primary care access and continuity of care in urban family practices in Ontario, Canada. BMC Fam Pract. 2019;20:52. doi: 10.1186/s12875-019-0942-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Kralji B. Measuring Rurality – RIO2008BASIC: Methodology and Results. [Toronto: Ontario Medical Association; 2008.]. https://www.oma.org/uploadedfiles/oma/media/pagetree/expert--advice/2008rio-fulltechnicalpaper.pdf. Accessed 20 Jan 2024. Toronto: OMA Department of Economics; 2009.
- 29.Senthinathan A, Tu K, Penner M, Morris AM, Craven BC, Jaglal SB Identifying Patterns of Antibiotic Prescribing for a Primary Care Spinal Cord Injury (SCI) Cohort using an Electronic Medical Records (EMR) Database [Manuscript submitted for publication]. Inst Health Policy Manag Eval Univ Tor 2022.
- 30.Senthinathan A Using Clinical Vignettes and a Modified Expert Delphi Panel to Determine Parameters for Identifying Non-Traumatic Spinal Cord Injury in Health Administrative and Electronic Medical Record Databases. 10.1016/j.apmr.2022.08.002. [DOI] [PubMed]
- 31.Schwartz KL, Wilton AS, Langford BJ, Brown KA, Daneman N, Garber G, et al. Comparing prescribing and dispensing databases to study antibiotic use: a validation study of the Electronic Medical Record Administrative data Linked Database (EMRALD) J Antimicrob Chemother. 2019;74:2091–7. doi: 10.1093/jac/dkz033. [DOI] [PubMed] [Google Scholar]
- 32.O’Malley AS, Cunningham PJ. Patient experiences with coordination of care: the benefit of continuity and primary care physician as referral source. J Gen Intern Med. 2009;24:170–7. doi: 10.1007/s11606-008-0885-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Anderson LH, Flottemesch TJ, Fontaine P, Solberg LI, Asche SE. Patient medical group continuity and healthcare utilization. Am J Manag Care. 2012;18:450–7. https://www.ajmc.com/view/patient-medical-group-continuity-and-healthcare-utilization. [PubMed]
- 34.Haggerty JL, Roberge D, Lévesque J-F, Gauthier J, Loignon C. An exploration of rural–urban differences in healthcare-seeking trajectories: Implications for measures of accessibility. Health Place. 2014;28:92–98. doi: 10.1016/j.healthplace.2014.03.005. [DOI] [PubMed] [Google Scholar]
- 35.Khoong EC, Gibbert WS, Garbutt JM, Sumner W, Brownson RC. Rural, suburban, and urban differences in factors that impact physician adherence to clinical preventive service guidelines: rural-urban impact on guideline adherence. J Rural Health. 2014;30:7–16. doi: 10.1111/jrh.12025. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Peterson LE, Dodoo M, Bennett KJ, Bazemore A, Phillips RL. Nonemergency medicine-trained physician coverage in rural emergency departments. J Rural Health. 2008;24:183–8. doi: 10.1111/j.1748-0361.2008.00156.x. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The dataset from this study is held securely in coded form at ICES. While legal data sharing agreements between ICES and data providers prohibit ICES from making the dataset publicly available, access may be granted to those who meet pre-specified criteria for confidential access, available at www.ices.on.ca/DAS (email: das@ices.on.ca). The full dataset creation plan and underlying analytic code are available from the authors upon request, understanding that the computer programs may rely upon coding templates or macros that are unique to ICES and are therefore either inaccessible or may require modification.