Article Text

Original research
Self-compassion letter tool for healthcare worker well-being: a qualitative descriptive analysis
  1. Melissa Powell1,
  2. Bryan Sexton2,
  3. Kathryn C Adair2
  1. 1School of Nursing, Duke University, Durham, North Carolina, USA
  2. 2Duke Center for the Advancement of Well-Being Science, Duke University, Durham, North Carolina, USA
  1. Correspondence to Ms Melissa Powell; map133{at}


Objective This qualitative study aimed to identify categories within therapeutic self-compassion letters written by healthcare workers. Resulting categories were assessed for their relevance to the construct of self-compassion.

Design This was a qualitative descriptive study that used summative content analysis and inductive coding.

Setting A US-based academic healthcare system.

Participants Healthcare workers who attended a self-compassion webinar were recruited.

Intervention The online self-compassion tool asked participants to write a letter to themselves from the perspective of a friend providing support and encouragement.

Results 116 letters were analysed. Five major categories emerged: Looking Forward, Reaffirming Self, Reaffirming Reminders, Hardships and Self-Disparagement. Respondents’ letters were mostly positively framed and forward thinking, including their hopes of improving themselves and their lives in the future. Negative content generally described hardships and often served to provide self-validation or perspective on obstacles that had been overcome.

Conclusion The writing prompt elicited content from the writers that reflected the core elements of self-compassion (ie, self-kindness, common humanity, mindfulness). Continued research to further understand, refine and improve the impact of therapeutic letter writing to enhance well-being is warranted to reduce burnout and promote quality patient care.

  • Nurses

Data availability statement

Data are available upon reasonable request. Data to be available upon request.

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See:

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.


  • The study’s self-compassion letter writing tool indicated that writers are adopting mostly positively framed and forward thinking perspectives when provided with a prompt.

  • A relatively small sample size (N=116) of letters was analysed, potentially limiting the generalisability of self-compassion writing findings.

  • A single self-compassion writing prompt was used and the need for further longitudinal research and diverse prompts may be beneficial for interventions.


As of 2022, there are approximately 65 million healthcare workers (HCWs) in the world according to the WHO’s National Health Workforce Accounts.1 It is estimated that up to one-half of HCWs experience burnout during their careers and the rates are continuing to climb.2 3 Burnout results from chronic workplace stress and is defined as feeling emotionally exhausted, having increased mental distance or negative feelings regarding one’s job, and reduced professional efficacy.4 HCW burnout has led to higher rates of turnover, lower levels of employee engagement, early retirement, reduced productivity, and potential for serious patient safety events and medical errors in healthcare settings.5–8 The consequences to HCWs who experience burnout extend into HCWs’ personal lives, and can include marital problems or increased risk of involvement in a motor vehicle accident.9 10 Burnout has also been linked to health problems in HCWs, including: depression, anxiety, suicidal ideation, substance abuse and decreased immune system function.11–14 The demands and responsibilities of HCWs are increasing (ie, assigned number of patients and patient acuity levels), while resources are decreasing (ie, staffing), resulting in a diminished ability to deliver high-quality patient care.15

Given the rising rates of burnout, the need to understand, research and create interventions aimed at promoting well-being is critical.16 17 Well-being is a positive feature of mental health and includes finding enjoyment, meaning and happiness along with ways to be resilient through coping skills and healthy problem-solving.18 Positive emotions have been linked to feeling greater purpose and well-being, and cultivating positive emotions has been demonstrated to reduce emotional exhaustion and boost well-being in HCWs.19–23 A key resource to enhancing well-being is through the cultivation of self-compassion.

According to Neff and Germer, finding compassion for others may come fairly easily, particularly compassion for friends and loved ones. In contrast, people are often much harder on themselves when they experience setbacks or make a mistake. Self-compassion specifically aims to turn this process around by accepting one’s own shortcomings and imperfections with self-care and kindness.24 Neff’s definition of self-compassion entails three main components: self-kindness, mindfulness and common humanity.25 Self-kindness is being kind and understanding to one’s self in the midst of suffering or failure; mindfulness is holding painful emotions and thoughts in balanced awareness instead of overidentifying with them; common humanity is viewing one’s own experience as a part of a larger human experience (eg, everyone makes mistakes or faces hardships).25

Research using self-compassion in psychotherapy has demonstrated efficacy in providing relief from disorders like anxiety and depression while reducing stress and self-criticism.24 Self-compassion has been increasingly applied to HCWs in recent years; however, only a few relatively small studies have been published. One pilot study (N=13) of an 8-week mindful self-compassion training programme in nurses found significant reductions in burnout and increases in resilience and compassion satisfaction.26 Similarly, a 1-day self-compassion training for nurses (N=22) found significant decreases in burnout, anxiety and stress, compared with a control group, from pre-intervention to post-intervention, and improvements were sustained at 3 months.27 Finally, an HCW self-compassion training ‘Mindful Self-Compassion Program for Healthcare Communities’ was recently developed by Neff and colleagues.28 The training is comprised of six 1.5-hour long weekly sessions. Compared with a control group, the training group (N=58) reported significant increases in self-compassion and well-being from baseline to post-intervention with improvements for at least 3 months.28 A second study found that participants reported reductions in secondary traumatic stress and burnout. Moreover, improvement in self-compassion appeared to be the explanatory mechanism for these beneficial outcomes.28 Despite the nascent state of self-compassion approaches for HCWs, the growing popularity within healthcare and robust evidence outside of healthcare make it a promising approach for improving HCW well-being.

Self-compassion interventions are delivered in a variety of ways, including meditations (eg, loving kindness), soothing touch and self-compassionate letter writing.29 Brief, one-time writing activities, like writing a gratitude letter, have demonstrated improvements in well-being for HCWs.19 20 Few studies have focused on exploring self-compassion and well-being from the perspective of HCWs, and to our knowledge, none examine the thoughts and emotions expressed by HCWs when using a self-compassion intervention. The purpose of this study was to describe HCWs’ perspectives on well-being and self-compassion through qualitative analysis of a therapeutic letter writing intervention. Understanding the experiences and perspectives expressed within their letters can help inform further development and testing of self-compassion and other well-being interventions.



This study used a qualitative descriptive design to extract rich and in-depth evidence from the HCWs’ therapeutic letters to thoroughly describe the phenomenon of interest, well-being and self-compassion.30 The Standards for Reporting Qualitative Research Checklist was used for reporting the study’s methods and findings and can be found in online supplemental appendix.31

Sample, setting and recruitment

Participants for the study were HCWs who attended a continuing education well-being webinar offered through a large US-based academic healthcare system, from July 2019 to December 2021. Individuals were recruited to attend the training through the healthcare system’s website. The sampling strategy used for recruitment was convenience sampling. Individuals participating in the self-compassion webinars were provided a survey link ( during the webinar and given the opportunity to complete the intervention at their convenience. Completing the survey was not mandatory and no compensation was provided. Inclusion criteria were individuals over the age of 18 years, self-identified as working within the healthcare field as an HCW, and were able to read and comprehend the English language. Exclusionary criteria included: non-HCWs, individuals who did not consent to having their survey responses used for research purposes and individuals who did not use the open-ended text box to elicit their response in the intervention (used here for qualitative analysis).

Patient and public involvement

Patients or the public were not involved in the design, or conduct, or reporting or dissemination plans of our research.

Data collection

The survey link was provided to all participants attending the well-being webinar and all survey responses were collected from July 2019 to December 2021. The demographic variables collected were gender, race, ethnicity and role in healthcare. The self-compassion letter writing tool was inspired by a similar tool in Neff and Germer’s Mindful Self-Compassion Workbook.32 To set up the intervention, participants were asked to respond to the following prompt, prior to engaging in the self-compassion letter:

‘First, identify something about yourself that makes you feel ashamed, insecure or not good enough. It could be something related to your personality, behaviour, abilities, relationships or any other part of your life. Once you identify something, write in the space below how it makes you feel. Sad? Embarrassed? Angry? Try to be as honest as possible, keeping in mind that your data are confidential.’

Responses to this prompt were used to guide their answer to the targeted self-compassion open-ended question:

‘Next, imagine someone who is unconditionally loving, accepting and supportive of you. The friend accepts and forgives you, embracing you just as you are. Now write a letter to yourself about this aspect of yourself from the perspective of this kind friend. How does this friend encourage and support you?’

Data analysis

A summative content analysis approach was used to help identify and quantify words and content from the text of each letter.33 The coding process included looking for the appearance of particular words or content in the letters while also including a counting and frequency of the appearing content.33 Inductive coding was used to ensure thorough analysis of all the letters to assess for multiple meanings across the content.34 Before analysing the letters, an initial read of all letters was completed to grasp a sense of the whole and gain an understanding of the letters’ contents.34 The coding process began with the first author reviewing of each letter, line by line, reviewing for keywords; keywords were then transformed into individual codes. Individual codes were based on manifest content and were created from individual words and sentences (meaning units) within each letter.35 After codes were created and placed in the codebook for all letters, preliminary categories and subcategories emerged after grouping similar codes together by the first author.34 All coding and emerging categories were reviewed by the third author, who is experienced in self-compassion research, with discrepancies being resolved together. All codes were stored in a codebook for organisation that featured all identified codes, their definitions and quoted exemplars.36 A count of all codes within major categories was calculated as a part of the summative content analysis approach to understand which keywords and findings were appearing often and in any specific pattern to understand the phenomenon of interest.33 Finally, analytical memos were recorded throughout the coding and analysis process for review with peer researchers in order to make coding and grouping decisions. The first, second and third authors collectively reviewed the final categories, subcategories, and counts and were in agreement of study findings.


To enhance rigour and ensure trustworthiness in the study, efforts to ensure credibility, transferability, dependability and confirmability are outlined. Credibility involves ensuring internal validity and seeking to have a study that measures what it is intended.37 To enhance credibility in this study, peer scrutiny and frequent debriefing sessions during the study with colleagues occurred by fellow members of the research team to provide feedback on the study analysis, findings and conclusion. Transferability involves the extent that study findings can be applied to other situations or scenarios.38 To enhance transferability in this study, rich description and detail of the scenario and settings facilitate understanding how these findings can be applied to real-world settings. Dependability involves reliability and the ability to obtain similar results if the same methods were repeated.39 To enhance dependability, detail of the research design and its implementation will be described through its execution. Lastly, confirmability involves concern over objectivity of the study results.38 To enhance confirmability, analytical memos were kept during the analysis and decision-making process that led to the creation of results.


A total of 118 responses included at least one character in the open text field for the open-ended question of interest and met criteria for qualitative analysis. Two letters contained less than three characters with no particular meaning and were excluded from further analysis. The final sample of 116 letters analysed ranged from 5 words to 373 words total, with a mean of 93.3, SD of 70.6 and a median of 78 words. Sample characteristics for the final sample (116) are presented in table 1. Most respondents were female (71.6%), white (71.6%) and non-Hispanic/non-Latinx (66.4%). Registered nurses (including certified registered nurse anaesthetists) (27.6%) made up the largest group of HCW roles followed by those in the other category (detailed in the footnote in table 1) (24%) and attending/staff physicians (16.4%).

Table 1

Sample characteristics (N=116)

A total of 297 unique codes were identified from the letters. There were 998 total occurrences of the unique codes identified at the completion of analysis. From the unique codes, five major categories emerged from reviewing the text and analysing the meaning and structure of keywords chosen by respondents. The five major categories (from largest to smallest) were: Looking Forward, Reaffirming Self, Reaffirming Reminders, Hardships and Self-Disparagement. A description of study findings from each category is detailed below in table 2. Table 2 also details the subcategories within each category. The total occurrence of codes in each category is identified next to the category name. A quotation exemplar is listed for each subcategory.

Table 2

Qualitative study findings

Of the five major categories, 80% (798) of the total occurrences fell within the Looking Forward category and the Reaffirming Self category, which each represented 42% (423) and 38% (375) of the total occurrences, respectively. The next categories, Reaffirming Reminders and Hardships, were less frequent but also similar in size, representing 8% (79) and 7% (73) of the total occurrences. The Self-Disparagement category made up less than 5% (48) of the total occurrences. No letters contained only codes from the Self-Disparagement category.

Looking Forward

The largest category of codes is the Looking Forward category, which includes future oriented reflections. Participants discussed both goals for themselves and their lives moving forward in this category; much of the writing from participants included specific behaviours or circumstances they wanted to change in the future that would be seen as an improvement from their perspective (see table 2 for statement examples across all of the categories). The Looking Forward category includes four subcategories: Relationships with Others, Doing this for Yourself, Intentional Pausing and Behaviours to Abandon. The subcategories Relationships with Others and Doing this for Yourself were largely focused on supportive behaviours, such as ‘surround yourself with support’ (Relationships with Others) and ‘take time to care for yourself’ (Doing this for Yourself). Participants were often yearning for self-reflection, health, peace and rest in the Intentional Pausing subcategory. Participants frequently desired to let go of guilt, extreme thinking and superficiality in the Behaviours to Abandon subcategory.

Reaffirming Self

The Reaffirming Self category includes codes and subcategories where participants were using supportive language to describe themselves. The language was self-validating and mostly positively framed regarding their qualities and accomplishments they had achieved. Of the five categories, the Reaffirming Self category has the second highest count of codes. Five subcategories were uncovered during analysis within Reaffirming Self: Self-Acceptance, Achievements and Success, Personal Qualities, Skills and Talents, Strength and Growth, and Positive Influence on Others. Across these subcategories, participants expressed being proud of their imperfections, personalities, worthy contributions and success in life. Participants wrote of their diverse attributes, including: beauty, value, uniqueness, empathy, happiness and resilience.

Reaffirming Reminders

Many participants used reaffirming language to write short reminders to themselves. These positively framed reminders make up the Reaffirming Reminders category. The Reaffirming Reminders category has five subcategories, including: Be Easier on Yourself, Optimism, Impact of Faith, Reminders for Life Ahead and Common Humanity. This category is closely related to codes in Looking Forward and Reaffirming Self, as many of the respondents included reminders to themselves that were reaffirming their perspective. Respondents’ reminders to themselves were often simple, yet powerful words, such as ‘no judgement, acceptance, understanding’.


The Hardships category includes codes regarding challenges participants have endured or are currently enduring. Codes within this category did not specifically use negatively framed language, but the majority of codes were factual statements regarding the realities they live in. For example, ‘Sometimes you have scars or stiff parts that remind you of the war you have been through. Part of the scars of being a human’ highlights past hardship with the ability to recognise that they are only human. Codes within the Hardships category were often paired with codes from the Reaffirming Self, Looking Forward or Reaffirming Reminders categories. Within the Hardships category, there are four subcategories: Relationships with Others, Psychological, Physical and Emotional Challenges, Exhaustion and Overload, and Unexpected Hurdles. The majority of codes within the four subcategories of the Hardships category were specific to personal challenges. For example, health challenges were categorised within the Psychological, Physical and Emotional Challenges category. A respondent wrote, “you’ve been diagnosed with hypothyroidism, so I know that it can make you tired and sluggish.” Unexpected hurdles were primarily related to the COVID-19 pandemic due to the timing of the letter writing, with one respondent expressing, “Being a parent is one of the hardest jobs there is and during this time of COVID with the kids being home from school it is at its absolute hardest.” However, few letters (N=3 of 116) included content specifically related to the COVID-19 pandemic. In addition to parenting challenges, these letters also included unique codes in the Unexpected Hurdles subcategory about chaos, and worries about uncertainty regarding the unknown.


Opposing the Reaffirming Self category is the Self-Disparagement category. Codes and subcategories in this category include: Questioning Self, Self-Criticism and Possessing Negative Qualities. Codes in this category included language from the participants that challenged or doubted their current choices, behaviours and/or qualities. Of note, this category included the fewest count of codes compared with the other four major categories. During analysis of the participant letters, no letters presented with Self-Disparagement codes in isolation; codes in this category were often paired with positively framed statements from either the Reaffirming Self or Looking Forward category. For example, a participant wrote, “Why are you worrying about this (Self-Disparagement). You have so many strengths to overcome the challenges that you are facing (Reaffirming Self).” Despite participants’ questioning themselves due to their current struggles, participants were able to rationalise their own thinking to focus on their positive attributes and abilities.


In this content analysis study of a self-compassion letter intervention for HCWs, five major categories were identified: Looking Forward, Reaffirming Self, Reaffirming Reminders, Hardships and Self-Disparagement. This self-compassion intervention resulted in extensive positively framed language from participants that presented itself in the Looking Forward, Reaffirming Self and Reaffirming Reminders categories regarding the individual’s perspective on themselves, their life and their future to manifest positive changes for their lives ahead. These three categories represented 88% of the total occurrences of codes identified across all five categories. Less frequently, participants wrote about Hardships (7% of total occurrences) and engaged in Self-Disparagement (5% of the total occurrences). However, the content in these categories often explained internal or external factors that impeded resiliency and served to validate the extent of the difficulties they encountered.

Participants in the current study appear to have tapped into the three components of self-compassion (ie, self-kindness, mindfulness and common humanity), as these elements emerged within numerous codes.23 Self-kindness was frequently expressed in the context of writing about personal attributes, accomplishments and life situations. Notions of mindfulness emerged in several codes within the Hardships category, where participants often expressed negative emotions or situations, but with balance and greater perspective. Recognising one’s difficulties without becoming overidentified with them (a key aspect of mindfulness) emerged in other areas as well, including the Self-Acceptance and Strength and Growth subcategories within the Reaffirming Self category. Finally, Common Humanity emerged as its own subcategory within the Reaffirming Reminders category, as many participants recognised that other people also struggle and that they are not alone. These examples highlight the power of this writing exercise to help individuals practise self-compassionate thinking about their current and future lives.

The majority of letters included positively framed and future focused language to express self-compassion; however, negatively framed language was used to describe challenges in life and views of self, which emerged in the Hardships and the Self-Disparagement categories. Our analysis found that even within these categories, the writing exercise naturally guided respondents to reframe their thinking about their difficulties and remember that humans make mistakes. These results indicate that the writing prompt helped participants examine their thoughts and perspectives in ways that lead to processing challenging life experiences and enhancing self-compassion.

Prior studies have found self-compassion to be a protective factor against depression,40 anxiety41 and burnout, which have all reached concerning levels in HCWs in recent years.3 42 Fortunately, research has also demonstrated that self-compassion can be increased through participation in self-compassion-based courses, typically lasting between 1 day and 8 weeks.26–28 Participants who completed an 8-week course had reductions in burnout and secondary trauma, and increases in self-compassion, mindfulness, resilience and compassion satisfaction at the end of the course.26 Participants who completed 1-day training had significant decreases in burnout, anxiety and stress with sustained benefits 3 months later.27

In comparison, our study’s findings indicate that a simple and brief (typically less than 7 min), one-time letter writing exercise helps participants tap into greater self-compassion. For overwhelmed HCWs who may not have the capacity to participate in a course, this tool shows promise as a bite-sized method to increase well-being. Prior cohort and randomised controlled designs of similar bite-sized well-being tools have be shown efficacious.19–23 Even a small ‘dose’ or opportunity to reflect through self-compassion writing may improve resilience and coping with the possibility of reaching a larger sample of HCWs. Moreover, engaging in a bite-sized tool might spark greater interest for HCWs to participate in a self-compassion course.

In addition to its brevity, the self-compassion tool is low risk. Although participants reflected on a negative aspect of themselves, they are doing so from the perspective of a friend. This perspective drastically shifts how participants frame the negative topic, typically through a significantly greater perspective, context, understanding and kindness. Participants are not asked to share what is written with anyone and they are reminded that their data are confidential.

Limitations of this study include a small sample (N=116) of a specific population (HCWs interested in well-being tools) which may limit generalisability. It is also possible that social desirability could have biased participants’ responses. Letters varied in length, with some containing a few words and others containing numerous paragraphs. In some cases, shorter letters limited our ability to fully grasp the author’s complete meaning. Additionally, a portion of data collection of participants’ letters occurred during the COVID-19 pandemic, which led to widespread challenges related to well-being for HCWs. Despite this timing, letters were not largely centred around the pandemic and only 3 of 116 letters explicitly mentioned ‘COVID-19’ or ‘pandemic’. However, the COVID-19 pandemic must be considered as a possible contributing factor in participants’ lives that might have affected the topics written about, such as hardships.

Future research may benefit from looking at specific groups of HCWs (eg, at risk of burnout, and those currently experiencing low, medium or high levels of emotional exhaustion) to understand their specific perspectives and challenges, as well as the extent to which self-compassion letter writing may impact emotional exhaustion. Future research could also focus on the utilisation of therapeutic letter writing activities implemented on more than one occasion to better understand the impact of ‘dosage’ of these activities on well-being over time. Additionally, investigations that assess the most efficacious writing prompts for improving well-being would help HCWs get the greatest benefit in the shortest amount of time. Future research should also collect clinical and operational outcome data to examine whether gains in well-being over time from self-compassion activities are linked to metrics such as intentions to leave, patient engagement, safety culture, sense of belonging, disruptive behaviours and patient safety assessments (e.g., medication errors, infection rates, and risk-adjusted mortality).

This study contributes new and rich understanding of how self-compassion writing may positively benefit HCWs seeking well-being resources. Content analysis findings suggest that participants were frequently expressing self-kindness (ie, Reaffirming Self and Reaffirming Reminders categories) and that they were optimistic and future oriented (ie, Looking Forward category). Participants were aware of the hurdles they were encountering (ie, Hardships category), and in many instances, this awareness served as validation of the difficulties being experienced and how they planned to overcome them now and in the future. Self-disparaging statements (ie, Self-Disparagement category), which were less frequent, often explained why the participant was being hard on themselves or internal characteristics that made self-compassion more difficult to experience.

With HCWs nationally battling burnout, understanding brief and low/no-cost interventions, like self-compassion letter writing, is pivotal to promote well-being of HCWs. Continued research on the use of self-compassion and other well-being interventions in HCWs is necessary, to support their well-being, and ultimately, to increase patient outcomes and quality of care.

Data availability statement

Data are available upon reasonable request. Data to be available upon request.

Ethics statements

Patient consent for publication

Ethics approval

The study proposal was reviewed and approved through the healthcare system’s Institutional Review Board (IRB Pro00063703) to ensure the protection of human subjects. All participants consented to their data being used for research purposes prior to beginning the online survey. Each survey respondent was assigned a participant ID for tracking and confidentiality purposes. All survey data and data analysis documents have been stored on secure hard drives and the data analysis of letters was performed with participant information deidentified.


Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.


  • Contributors MP, BS and KCA contributed to the design of the study. BS and KCA contributed to the data acquisition. MP and KCA contributed to the data analysis and interpretation and initial draft of the paper. MP, BS and KCA contributed to the review and multiple drafts of the manuscript. All authors, with MP as guarantor, accept responsibility for the work and/or conduct related to the study and had full access to the data and decision to publish the findings.

  • Funding This analysis was funded by the Health Resources and Services Administration (HRSA; grant no. U3NHP45396‐01‐00).

  • Competing interests None declared.

  • Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

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

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.