Article Text
PDF
Abstract
Physicians anecdotally report inquiring about incarcerated patients’ crimes and their length of sentence, which has potential implications for the quality of care these patients receive. However, there is minimal research on how a physician’s awareness of their patient’s crimes/length of sentence impacts physician behaviours and attitudes. We performed regression modelling on a 27-question survey to analyse physician attitudes and behaviours towards incarcerated patients. We found that, although most physicians did not usually try to learn of their patients’ crimes, they often became aware of them. We observed associations between awareness of a patient’s crime and poor physician disposition towards their patients and between physicians’ poor dispositions and lower reported quality of care. These associations suggest that awareness of a patient’s crime may reduce quality of care by negatively impacting physicians’ dispositions towards their patients. Future quantitative and qualitative studies, for example, involving physician interviews and direct patient outcome assessments, are needed to confirm these findings and further uncover and address hurdles incarcerated patients face in seeking medical care.
- prisoners
- quality of health care
Data availability statement
Data are available upon request.
Statistics from Altmetric.com
Introduction
Anecdotal evidence indicates that some physicians are curious about the crimes their incarcerated patients committed and even go so far as to ask the patients or others ‘what they did’. Conventional wisdom suggests that this information is irrelevant to the patient’s care, could result in physician bias and could even run afoul of Health Insurance Portability and Accountability Act requirements.1 There is a substantive body of literature on caring for patients in jails or prisons themselves as well as those who have previously been incarcerated; however, there is less written about physician behaviours and attitudes towards incarcerated patients who are cared for outside of a prison or jail infirmary setting (ie, in a public hospital or clinic). Notably, essentially absent from the literature is systematically collected data about physician awareness of an incarcerated patient’s crime or length of sentence when the patient is being cared for outside of a prison or jail infirmary setting (ie, in a public hospital or clinic), normative beliefs about whether these physicians should have that information and how that knowledge affects their behaviours and attitudes. We identified only one study approaching—but not fully addressing—this topic, reporting physician responses to three questions about incarcerated patients in a public healthcare setting: whether the physician asks the prison officer to leave during the visit, whether the physician asks the prison officer if the patient is dangerous and whether the physician examines the patient while shackled.2 A newspaper article published shortly after the Boston Marathon bombings in 2013 discussed the nurses’, but not physicians’, perspective on caring for the individual who committed that crime.3 Another article focused on the physician perspective and highlighted negative views, but in the situational context of assessing patients prior to placement into police custody.4
Biases, both implicit and explicit, exist among healthcare professionals when considering factors like age, race, gender and weight, for example, biases against older adult patients, black patients, female patients and obese patients.5 Further research is critical as physician behaviours and attitudes towards incarcerated patients may reveal bias, which could ultimately affect patient care. Such research also will help healthcare professionals in training and in practice to better care for and communicate with incarcerated patients in light of any explicit or implicit biases. This information may also contribute to the development of codes and guidelines in treating populations of patients involved with the legal system, which some physicians believe is necessary for patients in custody.6 Last, knowing whether there are disparities in physician care due to their biases can help us to explain whether these biases are contributing to the healthcare disparities that are currently plaguing prisoners; for example, prisoners have higher rates of substance use disorders, hypertension, mental illness, cancer and infectious diseases.7 8 Gathering and disseminating this information is important as incarcerated patients are seen outside of the jail system in off-site hospital settings for care.9
This study begins the above process by providing a ‘state of the field’ on physician behaviours and attitudes towards incarcerated patients being cared for outside of a prison or jail (ie, in a public hospital or clinic). This novel survey asked attending physicians who care for incarcerated patients in a public hospital or clinic about their behaviour towards incarcerated patients, such as whether they seek out information about the patient’s crime or sentence length and whether this information ultimately affects their attitudes towards the patient. We expect this will lead to future research exploring implicit and explicit bias and quality of patient care when physicians are aware of the crime an incarcerated patient committed and the length of the patient’s sentence.
Methods
We identified 1120 clinical faculty members at two academic medical centres from departments other than paediatrics and pathology to whom we sent a 27-question, anonymous survey via Qualtrics. A sample size calculation identified that we required 287 responses for a 95% CI with a 5% margin of error. Emails containing an invitation to complete the survey were sent to all 1120 physicians. The survey included the following screening question, ‘As an attending physician, have you ever or do you currently care for incarcerated patients?’ Seventeen physicians replied ‘no’ to the screening question and were deemed ineligible to participate in the survey. The survey was kept open until 290 physicians responded, at which point the survey was closed and data collection completed. Participants were compensated with a $5 electronic gift card for successfully completing the survey. This study was deemed exempt by the university’s Institutional Review Board (IRB).
Faculty were asked to provide demographic information, including specialty/subspecialty, gender and number of years in practice as an attending physician. They were also asked, based on the National Institute of Justice’s definition of ‘violent crime’,10 whether they or a family member had ever been the victim of a ‘violent’ or ‘non-violent’ crime. In addition to demographic questions, the survey explored several areas related to physician attitude and behaviour towards incarcerated patients, in particular what they typically know about the crimes their incarcerated patients committed and the length of their sentences, how respondents learnt this information and how they believed it affected their attitude towards and care for the patient.
Self-reported measures of the quality of care physicians provide, their attitudes and their empathy for their incarcerated patients and incarcerated patients who committed violent crimes were used as outcomes in proportional-odds cumulative logistic regression models. Responses from physicians who reported ‘I never know this information’ (ie, whether the crime is violent or not) were excluded from the analysis. The analyses modelling physician-reported quality of care for incarcerated patients measured associations between negative physician attitudes, less empathy from physicians towards incarcerated patients and how often a physician was knowledgeable of their patient’s crime. The additional analyses modelling physician attitudes and physician empathy towards incarcerated patients measured associations between the number of years a physician spent practicing, whether the physician tried to find out their patient’s crimes, how often a physician was knowledgeable of their patients’ crimes and whether the physician or a family member was a victim of a violent crime. Evaluation of the proportional odds assumption was determined via likelihood ratio tests of the goodness of fit for each variable. The proportional odds assumption was relaxed by fitting a partial proportional odds model where appropriate. Statistical significance was judged at p<0.05.
Results
Respondents were mostly men (n=185, 63.8%) from the Internal Medicine specialty (n=73, 25.2%) with over 15 years of practice (n=111, 38.3%). Nearly two-third of respondents (n=185, 63.8%) had been a victim of a non-violent crime, while a smaller portion, (n=39, 13.4%) had been a victim of a violent crime. About two-third of respondents (n=202, 69.7%) had a family member who was a victim of a non-violent crime, while only slightly over a quarter of respondents (n=85, 29.3%) had a family member who was a victim of a violent crime. About half of respondents (n=138, 47.6%) are never aware of the crime(s) that resulted in their patient’s incarceration, while two-thirds of respondents (n=191, 65.9%) never try to find out the crime(s) that resulted in their patient’s incarceration. Most respondents (n=258, 89.0%) never or rarely asked an incarcerated patient’s correctional officer if their patient is dangerous. A majority (n=240, 82.8%) are never or rarely fearful of their incarcerated patients. Most (n=230, 79.3%) are never or rarely aware of the crime(s) that resulted in their patient’s incarceration and most (n=258, 89.0%) are also never or rarely aware of the length of their incarcerated patient’s sentence. About three quarters of respondents (n=222, 76.6%) never try to find out the length of their incarcerated patient’s sentence. Three quarters of respondents also (n=218, 75.2%) disagree or strongly disagree that they have a more negative attitude towards all their incarcerated patients (as compared with about a quarter (n=84, 29.0%) for incarcerated patients who committed a violent crime). Most (n=240, 82.8%) disagree or strongly disagree that they have less empathy for all their incarcerated patients (as compared with about a third (n=109, 37.6%) for incarcerated patients who committed a violent crime). A vast majority (n=275, 94.8%) disagree or strongly disagree that they provide worse care for all their incarcerated patients (as compared with about two-thirds (n=177, 61.0%) for incarcerated patients who committed a violent crime) (see tables 1 and 2).
Descriptive breakdown of sample: demographics (n=290)
Descriptive breakdown of sample: descriptive and attitude assessments (n=290)
The results of the regression analyses are displayed in tables 3 and 4. A more negative attitude towards incarcerated patients was associated with increasing odds of physicians reporting they provide worse care towards their incarcerated patients overall (OR=1.84; 95% CI 1.18 to 2.87); however, no association was observed towards their incarcerated patients who committed a violent crime (OR=1.15; 95% CI 0.71 to 1.84). The same pattern was observed for empathy, where lower empathy for incarcerated patients was associated with increasing odds of physicians reporting they provide worse care for their incarcerated patients overall (OR=2.10; 95% CI 1.33 to 3.33) and patients who committed a violent crime (OR=1.99; 95% CI 1.21 to 3.30). No associations were observed between how often physicians were knowledgeable of their patients’ crimes and the quality of care they claim to provide. See Table 3.
Odds ratios for associations with physicians who report providing worse care to prisoners (n=290)
Odds ratios for associations with physician-reported negative attitudes and less empathy toward prisoners (n=290)
Increasing negative attitudes (OR=1.75; 95% CI 1.30 to 2.36) and decreasing empathy (OR=1.55; 95% CI 1.15 to 2.09) for incarcerated patients were associated with physicians who were more frequently aware of their patients’ crimes overall. No associations between decreasing empathy towards patients who committed a violent crime were observed for any of the variables under investigation; however, more negative attitudes towards incarcerated patients who committed a violent crime and awareness of patient’s crimes were associated (OR=1.41; 95% CI 1.02 to 1.94). No associations between negative attitudes toward and less empathy for incarcerated patients were observed for years practicing, trying to find out the patient’s crime, being a victim of violent crime, or having a family member who was a victim of violent crime. See table 4.
Discussion
Our cohort represents various ranges in number of practicing years, with a significant number of experienced physicians. It also represents a large number of physicians who were victims to non-violent and even violent crimes. Many also had family members who fell victim to such crimes.
We anticipated that many physicians would report asking questions about their patients’ status as a prisoner. While many did, the majority of respondents reported rarely or never trying to learn the crime their incarcerated patients committed. However, given that in our survey more physicians reported learning of their patients’ crimes than those who reported seeking out that information, we anticipate some physicians will likely become aware of patients’ crimes despite what may be their best efforts to avoid this information. Notably, the survey results revealed no associations between physician attitudes and whether they reported trying to learn the crime their incarcerated patients committed.
A study by Haider et al showed that there was no association between clinical decision making and biases towards social classes and races when physicians were faced with hypothetical situations. However, the authors concluded that physicians overall have an implicit bias towards white patients and higher-social class patients.11 Therefore, we hypothesised that these biases would be similar when applied to prisoners, that is, the information physicians learned would not impact the quality of care they provide but may affect their attitude. In our survey, physicians were more likely to report providing worse care when they reported a more negative attitude or less empathy towards incarcerated patients overall. Physicians who tend to be aware of their patients’ crimes generally did not report providing worse care for their incarcerated patients. However, both negative attitudes and less empathy were associated with the tendency to be aware of patients’ crimes. Future research would be needed to show whether this pattern of associations could reveal an indirect impact on quality of care when a physician is aware of a patient’s crime, by affecting the physician’s disposition towards their incarcerated patient and in turn impacting quality of care.
The findings on quality of care and physician attitudes/awareness of crimes for patients who committed violent crimes specifically differed from the results on incarcerated patients overall. An association between less empathy towards incarcerated patients who committed violent crimes and worse care was found, but no clear association was found between reporting worse care and negative attitudes. In contrast, awareness of patient crimes was associated with stronger negative attitudes towards incarcerated patients, but it was not found to impact physician empathy. These findings could point towards important differences in how physicians perceive and treat their patients who committed a violent crime compared with incarcerated patients in general. However, it is also possible that the analysis on violent crimes was underpowered to detect these associations after excluding responses from physicians who are never aware of the nature of their incarcerated patients’ crimes. Further research on this topic is necessary to determine the true associations between physician attitudes/empathy and quality of care for incarcerated patients who committed a violent crime.
Because this study relied on a convenience sample from two academic medical centres, we anticipate that our sample was not representative of all physicians practicing in academic medicine. Nevertheless, along some dimensions, our sample characteristics are similar to national-level characteristics. For example, physicians working in academic medicine in the USA are predominantly men (58.3%),12 similar to our sample (63.8%). Our sample is also consistent with specialties in academic medicine, with 25% of our sample specialising in internal medicine, compared with 27.4% in the USA.12 Although associations between physician quality of care and attitudes/empathy were observed, it is important to remember that responses to the survey are all self-reported. A physician’s self-assessment of their own care represents a potential source of bias. It is possible these physicians may perceive they provide worse care to their incarcerated patients, but, in practice, their care may not be worse at all. It is also possible that social desirability bias limited the number of physicians who were willing to acknowledge providing worse care to their incarcerated patients.13 This bias would create a tendency to report better care than is actually being delivered. The associations between negative attitudes/less empathy and worse care would be attenuated with respect to their true values, as only those physicians who are willing to acknowledge they provide worse to care to a subset of their patients would answer honestly. A different methodology which directly measures the care physicians provide could better illustrate whether incarcerated patients actually receive worse care. An experimental design similar to Haider et al, where physicians considered hypothetical scenarios and were asked to make clinical decisions, would be one possible approach.11 Other quantitative and qualitative methods assessing physician beliefs and patient outcomes are appropriate as well. Though we hypothesise that a physician may report providing worse care due to bias, there is a chance a physician may report providing worse care due to healthcare systematic issues rather than negative bias. For example, they may report providing worse care because of the challenges—as recognised by the WHO—in maintaining continuity of care for prisoners.14 Last, it may serve as a limitation that we did not ask survey participants to consider intimate partners or close friends when inquiring about family members who were victims of violent crimes.
Conclusion
This paper provides a novel look at physician attitudes and behaviours towards incarcerated patients. In light of the results, the potential impact of physician attitudes and behaviour on care of these patients, especially if/when physicians learn of the crimes their patients committed, is an area ripe for future research. Findings from such studies could represent an important step in understanding health disparities currently prevalent among prisoners.7 Furthermore, in the USA, the rate of incarceration is significantly higher among African-American men and women, who already experience greater health disparities.7 Thus, a more in-depth understanding of the findings from this study is warranted to help promote equality and equitability in healthcare.
Data availability statement
Data are available upon request.
Ethics statements
Ethics approval
This study was deemed exempt by the university’s Institutional Review Board.
Footnotes
Contributors All authors have significantly contributed to this project. KP and LBS performed the initial literature review and drafted the research proposal. LBS, KP and KPR identified the subject population, drafted the IRB-02 proposal and identified and recruited subjects. KPR and BJR performed statistical the statistical analyses. All authors were involved in interpreting statistical analyses and drafting the manuscript.
Funding This study was funded by the Department of Community Health & Family Medicine in the University of Florida College of Medicine. The authors also acknowledge the partial support for this project provided by a medical student scholarship from the Goodman Trust.
Disclaimer The manuscript has not been published in another journal, is not under consideration by any other journal, and the final manuscript has been seen and approved by all authors.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.