Article Text

Download PDFPDF
Medical student anxiety in caring for dying patients and their family: a cross-sectional study
  1. Geoffrey Wells1,
  2. Carrie Llewellyn2,
  3. Andreas Hiersche1,3,
  4. Ollie Minton3 and
  5. Juliet Wright1
  1. 1Department of Medical Education, Brighton and Sussex Medical School, Brighton, UK
  2. 2Department of Primary Care and Public Health, Brighton and Sussex Medical School, Brighton, Brighton and Hove, UK
  3. 3Department of Palliative Medicine, Brighton and Sussex University Hospitals NHS Trust, Brighton, Brighton and Hove, UK
  1. Correspondence to Dr Geoffrey Wells, Medical Education, Brighton and Sussex Medical School, Brighton BN2 4AT, UK; g.wells{at}


Objectives To investigate the level of medical student anxiety in caring for a dying patient and their family and identify influencing factors.

Methods We conducted a cross-sectional survey in a UK medical school to measure medical student anxiety using a validated Thanatophobia Scale questionnaire.

Results In total, 332 questionnaires were completed. Mean thanatophobia score was 19.5 (SD 7.78, range 7–49). Most respondents were female (67.4%) and did not have a previous undergraduate degree (56%). Median student age was 22 years (IQR 20–24). Year of study influenced anxiety level, with second year students displaying an increase in mean thanatophobia score of 6.088 (95% CI 3.778 to 8.398, p<0.001). No significant differences were observed between final year and first year thanatophobia scores. For each 1-year increase in student age, mean thanatophobia score reduced by −0.282 (95% CI −0.473 to −0.091, p=0.004). Degree status and gender identity did not significantly affect thanatophobia score.

Conclusion A degree of thanatophobia exists among medical students, with no significant improvement observed by completion of training. Recognising this anxiety to care for the dying earlier in undergraduate curricula will give educators the opportunity to address students’ fears and concerns and better prepare our future doctors for their role in caring for our dying patients and their families.

  • education and training
  • end of life care

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Key messages

What was already known?

  • Newly qualified doctors experience anxiety when caring for dying patients.

  • Medical undergraduate thanatophobia is poorly understood.

What are the new findings?

  • Thanatophobia exists among medical undergraduates and does not significantly improve by the end of training.

What is their significance?

  • Recognising and addressing undergraduate thanatophobia may be an important step in reducing thanatophobia among newly qualified doctors


Provision of basic palliative care is the responsibility of every doctor,1 with the most junior of all doctors spending the greatest time with dying patients.2 The General Medical Council recommends all medical students should receive core teaching on care of the dying patient.3 Many Foundation Year 1 (FY1) doctors feel poorly prepared to deliver basic palliative care, reporting high levels of anxiety when doing so.2 4–7

Undergraduate medical education provides variable (and perceived lack of) opportunities in teaching on care of the dying, with student feedback reporting evidence of exclusion from bedside reviews of dying patients.2 This can mean that teaching on the care of dying patients may sometimes be overlooked.2

While the construct of thanatophobia pertains to an individual’s reflections and concerns about their own mortality, it is also possible to measure thanatophobia in relation to one’s own anxiety to care for those who are dying.8 The aim of this study was to assess the extent to which medical undergraduates report anxiety towards caring for a dying person and their family, and explore variables that may influence this anxiety.


Study design

A cross-sectional survey was conducted using the Thanatophobia Scale (TS) questionnaire.8

Participants and setting

This study was conducted with students at Brighton and Sussex Medical School (BSMS). A convenience sampling approach was taken, with any student studying in years 1–5 of the undergraduate medical course eligible to participate.

Data collection and analysis

The validated TS questionnaire contains seven statements measuring attitudes towards caring for a dying patient and their family. Each statement poses a negative attitude to the respondent with each response scored using a Likert scale between 1 (strongly disagree) and 7 (strongly agree). Total scores range between 7 (low degree of thanatophobia) and 49 (high degree of thanatophobia).

Data on student age, gender identity, year of study and possession of a previous degree were collected to identify explanatory variables that may influence TS score.

Questionnaires were distributed to year 1–4 students at compulsory cohort lectures. Due to the geographically dispersed nature of year 5 students, data from this cohort were collected mainly via secure university email. Questionnaires returned by email were printed off prior to deletion to maintain anonymity. Consent was implied for those who returned a completed study questionnaire as per ethical permissions (ref ER/BSMS9BC3/2). Participation was voluntary, anonymous and attracted no financial or academic incentives.

Descriptive statistics were used to summarise continuous (age) and categorical data (gender identity, degree status and year of study). Data distribution was examined for normality through analysis of PP and QQ plots. Means and SD are reported for normally distributed variables, medians and IQRs reported for non-normally distributed variables, with proportions and percentages reported for categorical variables.

Statistical modelling was used to analyse the relationship between the explanatory variables (year of study, age, gender identity and degree status) with the outcome variable of TS score. Multiple linear regression using the entry method enabled all variables to be entered into the model to identify individual effects.9

All data were collected in February 2019 and were analysed using IBM SPSS Statistics for Windows, V.25.0.10


Three hundred and thirty-two TS questionnaires were completed from an eligible student population of 715 (response rate 46.4%).

Mean and median TS scores for all students were 19.5 (SD7.78) and 19.0, respectively.

The greatest number and proportion of questionnaires were returned from year 1 students (n=79) and year 2 students (n=77, 57.9%), respectively. The year 5 cohort returned the least number and proportion of questionnaires (n=45, 28.7%).

Median student age was 22 (range 17–52). All participants identified either as female (67.4%) or male (32.6%), which was representative of our study population. One hundred and forty students (44%) had a previous undergraduate degree.

Missing data

Seventeen questionnaires (5.1%) contained missing data. Fifteen (4.5%) were missing some or all demographic data; therefore, questionnaire scores were only analysed for those variables completed. Two (0.6%) were missing a single thanatophobia score; missing scores were statistically imputed using the mean score obtained from the remaining six questions in each questionnaire.

Significant findings

The most significant influence on TS score was related to year of study (table 1). Each year was compared with reference year 1. Results demonstrated that being a second-year student increased mean TS score by 6.088 (95% CI 3.778 to 8.398, p<0.001) meaning that second year students possess a higher degree of thanatophobia than their peers in any of the other year groups. Age exerted a small but significant influence on TS score. For an increase in age of 1 year, the TS score changed by −0.282 (95% CI −0.473 to −0.091, p=0.004), meaning that for every 3.5-year increase in age, the TS score fell by one point (2.5%), corresponding to a lower level of thanatophobia.

Table 1

Multivariate analysis of Thanatophobia Scale (TS) score


Results of this study demonstrate that anxiety in relation to caring for a dying patient and their family is greatest among younger students. While second-year students demonstrate a statistically significant increase in thanatophobia compared with any other year group, overall medical student anxiety by the final year of training does not significantly differ from that seen at the start of training. Degree status and gender identity exert no significant effect on student thanatophobia.

Thanatophobia is not uncommon among medical students. Studies have demonstrated that a greater degree of fear and anxiety is seen within medical students compared with that of other healthcare professionals (student nurses, nurses and qualified senior doctors)11 and may be the result of teaching methods that generally follow a biomedical models of disease within medical education, with focus on the curative nature of medicine. By contrast, it is argued that nursing education focuses more on individual patients’ problems as caused by the underlying disease without necessarily bestowing a curative expectation.8

Findings in relation to age echo those found in studies conducted both in the UK and USA, which have demonstrated lower levels of thanatophobia among older medical students, with exposure to different life experiences cited as a potential explanatory factor.8 12 A more recent (2019) Korean study exploring attitudes towards end-of-life care among medical students found that while differences as a result of gender identity were inconsistent, older students (>26 years) and graduates were statistically more interested in end-of-life care in general,13 which may be influencing TS score.

Results of studies exploring medical student attitudes on death have shown that clinical placements positively influence students’ fear of the death of patients.14 This suggests that exposure to dying patients has the potential to positively change attitudes towards death and dying and may provide an explanation as to why the second-year group in this study returned higher thanatophobia scores. This cohort will have experienced 2 years of training with limited patient contact. Anticipation of increased patient encounters as they approach third-year study may lead to feelings of increased anxiety and may explain why TS scores were higher in year 2 compared with those of year 1.

There are few studies documenting thanatophobia among medical students. Results of this study will add a greater understanding of thanatophobia within medical students, particularly in relation to variables of age, gender identity, degree status and year of study. A further strength of this study relates to the decision to use the TS questionnaire. It is validated for use in medical undergraduates and takes just 5 minutes to complete, which meant little disruption to compulsory lectures.

Studies suggest that care of the dying is a skill most likely to be developed in the postgraduate setting.4 15 This needs to change, because simply accepting this to be the case will create barriers to dealing with junior doctor anxiety and will provide little comfort and support to the newly qualified FY1 doctor or the dying patient they are caring for. It is important for medical educators to recognise concerns around provision of care to the dying within their undergraduate student population and seek opportunities to reform their curricula so that these anxieties can be open to further exploration and management.

A next step may be to ascertain whether addressing thanatophobia at undergraduate level has a positive impact on FY1 anxiety to care for the dying and whether this translates into improved patient care.


By the end of medical school training, levels of thanatophobia experienced by students do not significantly differ from that experienced in the first year of study. While it may be reasonable to expect some anxiety when thinking about caring for someone who is dying, medical schools could do more to identify and explore thanatophobia within their students. Normalisation and demystification of death and dying by way of increased clinical exposure may afford students the ability to understand why they experience thanatophobia. Once thanatophobia is recognised, undergraduates should be encouraged to seek support if they believe the degree of anxiety they experience may be negatively impacting their learning in end of life care. Addressing high levels of thanatophobia in a timely and appropriate should be considered essential prior to qualification and commencement of FY1 duties.

Ethics statements

Patient consent for publication

Ethics approval

Permission to undertake this study was granted by the Brighton and Sussex Medical School Research Governance and Ethics Committee on 28 January 2019 (ref ER/BSMS9BC3/2).


The authorship would like to thank Dr Elaney Youssef for undertaking independent measurements of study questionnaires to ensure data accuracy. The authorship would also like to thank Dr Chris Jones for his support with data analysis.



  • Contributors GW is the corresponding author and primary researcher. GW has been involved in every aspect of this study from initial conception, data collection, analysis and development of this paper. This study was undertaken as part of a Medical Doctorate. CL is one of the primary supervisors of the corresponding author. She has advised on all aspects of data collection, analysis and interpretation. She has reviewed and edited this paper prior to submission. AH is one of the primary supervisors of the corresponding author. He has advised on aspects of data collection using the Thanatophobia Scale questionnaire, provided his expertise as a consultant in palliative medicine and has reviewed and edited this paper prior to submission. OM is one of the primary supervisors of the corresponding author. He has provided his expertise as a consultant in palliative medicine and has reviewed and edited this paper prior to submission. JW is one of the primary supervisors of the corresponding author. She has overseen this study, providing the resources to allow the study to be undertaken. She has advised the primary researcher with respect to obtaining ethical approval and has supported the study at every stage. She has also reviewed and edited this paper prior to submission.

  • Funding This study has been undertaken as part of a fully funded medical doctorate being undertaken by the primary researcher at Brighton and Sussex Medical School.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; internally peer reviewed.