Exploring elderly patients’ experiences of recovery following complex proximal humerus fracture: A qualitative study

Abstract

This study explores the experiences and perceptions of recovery in elderly patients who had sustained a proximal humerus fracture. In-depth semi-structured interviews were conducted with 15 patients over the age of 65. Thematic analysis identified aspects of care that impacted upon patient experience and quality of life. Seven main patient-reported themes were identified, including pain, sleep, shoulder function, emotional state, social support, relationship with their professional and experience of healthcare institution. These themes offer insight into the experiences of adults receiving care for proximal humerus fracture and highlight that existing quantitative measures of quality of life do not measure domains that are important to patients.

Keywords

patient outcomes proximal humerus fracture qualitative analysis thematic analysis

Introduction

Fractures of the proximal humerus (shoulder) are common musculoskeletal injuries associated with significant morbidity (Palvanen et al., 2006). The effectiveness of management strategies for these injuries has been investigated using quantitative research methods over the last 30 years (Handoll and Brorson, 2015); however, there is currently no published qualitative research examining patient-reported outcomes following the management of this condition. Additional qualitative research in this area would help to uncover the social and psychological impact of living with and recovering from proximal humerus fracture, with the potential to better inform understanding for healthcare professionals and researchers working in this area (Pope and Mays, 1995).

Patient-reported outcome measures (PROMs) are regularly used in the assessment of patients with these injuries throughout their treatment course (Rangan et al., 2015). PROMs such as the Oxford Shoulder Score (OSS; Dawson et al., 1996) are used to evaluate the outcomes of different treatment strategies and aim to guide appropriate management of patients with this injury. Although many PROMs have been developed across orthopaedics, none have been developed for elderly patients who have sustained a proximal humerus fracture. Existing PROMs have been validated for a broad range of shoulder pathology, including proximal humerus fracture, across a wider age demographic and their application in elderly patients with shoulder fractures may not be reliable (Van de Water et al., 2014). Age-related differences in subjective perception of outcome have been suggested to occur because younger patients seek additional functional gains rather than pain relief, compared with elderly patients (Williams et al., 2013).

Generic health utilities to evaluate quality of life are often included as outcome measures in research evaluating management strategies for proximal humerus fractures. Some research indicates that in an elderly population, a decline in functional outcome has little effect on quality of life perception (Torrens et al., 2011). Although the majority of patients in studies reporting on non-operative or operative management describe being satisfied with their outcome, there is no deeper evaluation of what factors contribute to this perception (Valenti et al., 2017). In the absence of a specific measure for older adults with proximal humerus fracture, or qualitative research evaluating patient satisfaction and patient perceptions of recovery, there is a need to better understand the aspects of recovery that patients value the most, so that decisions on treatment can be better informed.

Therefore, the aim of this study was to better understand the experiences and perceptions of elderly patients who had sustained a complex proximal humerus fracture in order to expand the current literature in this area and assess the suitability of existing PROMs.

Methods and materials

Participants and recruitment

Patients were eligible to take part if they were aged 65 and above, had a Neer type 3 or type 4 part fracture. Patients were only approached if at least 1 year had passed since their injury occurred. This time frame was chosen because existing research indicates that patients may experience improvements in outcome up to a year post-injury (Rangan et al., 2015). Patients were unable to participate if they had experienced multiple injuries at the same time as the humerus fracture, were unable to understand verbal or written information in English or were unable to give informed consent.

A purposeful sampling technique was used in order to identify patients who had undergone both conservative and surgical treatments, as well as those that had a variety of pre-morbid states and a range of clinical outcomes. Patients were recruited to the study from a large National Health Service (NHS) teaching hospital in Northwest London. They were identified from their hospital records and were sent information about the study via the post. Patients were then contacted by telephone to see if they were interested and able to take part in the study.

Data collection

Semi-structured face-to-face interviews were conducted on hospital premises by the lead author (S.S.). Prior to the interview, patients gave written consent to participate in the study. The interviews were audio recorded and field notes were taken. Interviews comprised open-ended questions such as tell me what happened to you? and how did you feel in the days and weeks following your injury? There was no time limit on the response to these questions. The interview concluded when the patient confirmed that they had nothing more to say.

A range of interview prompts were identified to ensure a broad range of topics were covered during the interview. While it was anticipated that many of these topics would be described in the responses to the open-ended questions, the prompts were included to ensure that areas relevant to patient recovery were discussed in detail. These prompts were derived from topic areas previously reported on in qualitative research following lower limb traumatic injuries (Trickett et al., 2012), which indicates that sleep, emotional status, mobility and concerns with employment are important aspects of recovery patients with tibial fractures (Table 1).

Table 1.

Interview prompt form.

Open questions – patients expected to talk at length
Tell me what happened to you?
Tell me as much as you can about your experiences during your recovery?
How did you feel in the days and weeks after your injury?
How do you feel you responded to the treatment?
How do you feel your rehabilitation went?
How are you now?
Specific questions – to prompt patients if subject matter not already covered
Tell me about your experience with pain after the injury.
Tell me about your experience with mobility after this injury.
Tell me about your experience with sleep after this injury.
Tell me about your mood after this injury.
Tell me about your concerns after this injury.
Tell me about any problems you had with employment or money after this injury.
Have you needed a lot of help in the last year?
Is there anything specific you want to mention about your injury, treatment or recovery?

Data analysis

Thematic analysis as outlined by Braun and Clarke was used to analyse the data generated in this study (Braun and Clarke, 2006). Thematic analysis is an accessible and theoretically flexible approach to analysing qualitative data and has been used in qualitative analysis of orthopaedic patients in other subspecialities (Bardgett et al., 2016; Moore et al., 2015). In the first instance, audio recordings were transcribed verbatim. In order to become familiar with the data, two authors (S.S. and S.A.) repeatedly read the interview transcripts, and notes were made about potential codes. Following this, two authors (S.S. and S.A.) performed formal coding manually, working through the transcripts and writing notes to support data patterns –these patterns were extracted as codes. Codes were placed in a spreadsheet with the data extracts to support each code. Codes were then drawn together to make a broad set of initial themes. These themes were then reviewed and refined to develop a set of final themes that supported the majority of data. A thematic map was constructed to show that the relationship between themes and examples was taken from the data set in order to illustrate each theme to aid interpretation and understanding. From generation of the initial codes to the point of establishing the final themes, differences between the two authors’ analyses was reviewed and consensus was reached through discussion; in the event that consensus could not be reached between the two authors, a third author (P.R.) would have been included in the discussion, although this was not needed.

Ethical approval

Ethical approval for this study was received from the National Research Ethics Service Committee (Research Ethics Committee reference 14/LO/2191).

Results

Patient demographics

A hospital shoulder fracture database identified 33 eligible patients. These patients were purposefully sampled by the authors to represent the breath of expected patient demographic and patient outcomes within the stated inclusion criteria. The first 20 patients in this list were contacted, of which 15 (75%) were willing or able to participate in the study. In total, 4 male and 11 female patients took part in the semi-structured interviews (Table 2). Patients were aged between 65 and 84 at the time of injury, with the time from injury ranging from 1 to 5 years. Of the patients, 10 had undergone surgical management and 5 had been managed conservatively. Conservative management consisted of 6 weeks of immobilization of the arm in a sling followed by supervised physical therapy for at least 3 months. Seven of the surgically managed patients had been treated with a locking plate, two patients had undergone hemiarthroplasty and one patient had been managed with suture fixation. Several patients had experienced complications following their treatment – information on this is included in Table 2. Interviews lasted between 21 and 35 minutes.

Table 2.

Patient demographics and management strategies employee.

ID	Gender	Age at injury (years)	Injury to interview interval (years)	Management method	Complications
1	Male	73	4.5	Conservative management	–
2	Female	65	4.2	Conservative management	–
3	Male	67	1.5	Locking plate	–
4	Female	66	2.4	Locking plate	Axillary nerve injury – ongoing sensory disturbance in cutaneous distribution
5	Female	80	5.1	Locking plate	Developed painful osteoarthritis and avascular necrosis and went on to have a hemiarthroplasty
6	Male	75	1.5	Locking plate	Had to have a revision procedure due to screw penetrating into joint
7	Female	75	2.1	Conservative management	Developed painful osteoarthritis after a year
8	Female	69	2.6	Locking plate	Removal of locking plate after a year – screw penetrating into joint
9	Male	76	2.1	Hemiarthroplasty	–
10	Female	65	1.1	Conservative management	–
11	Female	67	2.2	Locking plate	–
12	Female	84	1.6	Conservative management	–
13	Female	69	1.7	Suture fixation	–
14	Female	72	2.3	Locking plate	–
15	Female	74	1.1	Hemiarthroplasty	–

Thematic analysis

The analysis revealed seven themes associated with recovery after a proximal humerus fracture; these were split into two groups: (1) those that described patient-reported outcomes: pain, sleep, shoulder function and emotional state and (2) those that described moderators of experience: social support, relationship with healthcare professional and experience with healthcare institution. The relationship between these themes is demonstrated in Figure 1 and described in more detail in the theme overviews below.

Figure 1.

Thematic map showing themes associated with proximal humerus fractures.

Patient-reported outcome: pain

Pain was a commonly reported problem both in the acute phase following the injury or surgery, and in the weeks to months that followed. Pain was central to the recovery process and the management of severe pain was important in the early stages of recovery. Subthemes described patient experiences of discomfort and their impact of poorly managed analgesia. When patients were describing their pain, they often compared it to previous experiences to highlight its intensity:

I’ve had every part of my body broken, but I’ve never, ever experienced a pain like this. (Paul, a 77-year-old man, who broke is shoulder 4 years ago and was managed non-operatively)

Furthermore, the relief from acute pain was important:

It’s like when you have a hot shower – it’s wonderful … it’s okay. It’s good as gold. (Alfred, a 68-year-old man, who broke his shoulder a year and a half ago and was managed surgically with a locking plate)

In spite of the relief that pain killers bought, patients reported side effects from medication that were more troublesome than the pain:

I found that really the constipation was worse than the other thing. (Alice, a 67-year-old lady, who broke her shoulder a year ago and was managed non operatively)

Pain was also reported in cases of sequelae of injury such as developing arthritis or avascular necrosis (AVN):

But the pain, it developed later on and I had really quite severe pain. (Betty, an 85-year-old lady, who broke her shoulder 4 years previously and was managed surgically with a locking plate)

Developing AVN was also associated with a sense of disappointment:

I am not sure now because I was very disappointed when I heard that I had the necrosis you know. (Viola, a 68-year-old lady, who broke her shoulder 2 years previously and was managed surgically with a locking plate)

In summary, pain was the outcome most frequently reported by patients recovering from proximal humerus fracture; its presence and absence was seen as a marker of recovery. Pain was an integral part of all of the other pain reported outcomes, with patients’ experiences being moderated by available social support to help with difficult functional tasks (see below).

Patient-reported outcome: sleep

Sleep was described as an important marker of recovery by all patients. Subthemes described how patients experienced difficulties with sleeping and waking at night. Disturbances in sleep were commonly reported because of difficulty lying on the affected side, or inadvertently rolling onto that side during sleep:

It was just extremely painful for quite a few weeks … I just didn’t know what sleep is. (Alfred, a 68-year-old man, who broke his shoulder a year and a half ago and was managed surgically with a locking plate)

Patients also reported having to change their sleeping position even a year after their injury:

… after about 18 months, I was able to turn over and sleep on my left which was quite nice. (Viola, a 68-year-old lady, who broke her shoulder 2 years previously and was managed surgically with a locking plate)

One patient reported symptoms of anxiety coming on prior to going to sleep:

I don’t know what happened to me, but I, sort of, went into a panic attack. (Julio, a 77-year-old man, who broke his shoulder 2 years ago and was managed surgically with a hemiarthroplasty)

In summary, difficulties with sleeping were an aspect of recovery that placed a significant burden on patients and was closely related to the pain they experienced. Sleep was also negatively affected by the lack of social support of family, friends and carers during the night time (discussed further below). In some accounts, patients described that changes to sleep patterns persisted for several months and impacted on emotional state as they reported being exhausted and anxious.

Patient-reported outcome: shoulder function

Shoulder function was important to all patients, particularly in the first 4–6 weeks following injury. Subthemes focused on self-care, participation in activities and loading the shoulder. Patients also gave descriptions of adaptive or avoidant behaviours. Throughout the recovery period, patients reported struggling with self-care activities such as showering, dressing and feeding. Most of the patients had to rely on friends and family to help:

There was not much that I could do with my arm I could not lift much so my husband did a lot. (Viola, a 68-year-old lady, who broke her shoulder 2 years previously and was managed surgically with a locking plate)

One year following the injury, most patients were able to care for themselves, either through avoiding activities that they found difficult or by adapting their lifestyle:

I can’t do a bra at the back anymore which I don’t wear. I wear the stretchy ones.

(Betty, an 85-year-old lady, who broke her shoulder 4 years previously and was managed surgically with a locking plate)

Over the longer term, patients reported difficulty with heavy lifting on the affected side and became something they learned to avoid:

If I lift something heavy or I push something heavy, I know that then I will have some pain. Even now. (Antoinette, a 73-year-old lady, who broke her shoulder 2 years ago and was managed surgically with a locking plate)

In some cases, patients had to give up activities that they had enjoyed prior to the injury. This was mostly in patients who were relatively active prior to the injury:

I used to be climbing mountains, just as a hobby, cycling, swimming … activities that I used to do, I just no longer can do. (Alfred, a 68-year-old man, who broke his shoulder a year and a half ago, and was managed surgically with a locking plate)

Other patients described how permanent adaptations coupled with acceptance of their disability enabled them to regain and maintain skills that were important for their overall quality of life:

I think I have adapted. I’m driving again, but I’ve got a special attachment on the steering wheel which has made a huge difference because initially it was difficult. (Clara, a 66-year-old lady, who broke her shoulder 2 years ago and was managed surgically with a locking plate)

In summary, shoulder function is synonymous with patient perceptions of recovery; patients acknowledged that shoulder function was one way of assessing the success of their treatment and their progress in recovery process. Where full recovery was not possible, the ability of patients to adapt their behaviour to perform activities of daily living was partly facilitated by the receipt of social support from family, friends and carers.

Patient-reported outcome: emotional state

Patients described that their emotional state was affected in the immediate aftermath of injury and subthemes described patient experiences of disability acceptance, anxiety, fear and altered mood. In many cases, long-term effects on emotional state were related to ongoing pain and a loss of function:

And I lost quite a lot of weight, obviously, because you don’t move and when you’re in pain, you don’t feel like eating and so on. (Alfred, a 68-year-old man, who broke his shoulder a year and a half ago and was managed surgically with a locking plate)

I got quite, you know, despondent about the fact that was it ever going to get better? (Judy, a 69-year-old lady, who broke her shoulder 2 years ago and was managed surgically with a suture fixation)

Some patients reported anxiety related to fear of falling and subsequent modification in behaviour to avoid these falls:

I’m so frightened of falling now. When I go out walking I feel like I’m walking and watching the pavement. (Clara, a 66-year-old lady, who broke her shoulder 2 years ago and was managed surgically with a locking plate)

Patients also reported a loss of independence following their injury, which in turn affected their mental state. This was occasionally compounded by difficulty with getting help from carers or family members:

My sons stayed with me for two week. And after that, I said they had to go back to their wives and work. (Alfred, a 68-year-old man, who broke his shoulder a year and a half ago and was managed surgically with a locking plate)

Some patients described how accepting their new disability helped them manage their emotional state. Often this was achieved by rationalizing it in relation to their age, or after discussion with healthcare professionals:

Part and parcel of life. Certain things you don’t like, but you’ve got to accept. (Alfred, a 68-year-old man, who broke his shoulder a year and a half ago, and was managed surgically with a locking plate)

I realised that I was treated the proper way and then that helped to accept the pain. (Paul, a 77-year-old man, who broke is shoulder 4 years ago and was managed non-operatively)

Injury and the subsequent treatment had a profound effect on patients’ emotional state, often associated with symptoms of depression and anxiety about falling. The ability to rationalize disability appeared to affect how some patients perceived their final clinical outcome.

Moderator of experience: support networks

Support networks were described by patients as being crucial to their recovery and the subthemes described the importance of the relationships and interactions with family, friends and carers. The presence or absence of relationships with others impacted on experiences of recovery. For example, the first 4–6 weeks following injury were difficult for patients as they often required help with self-caring activities:

A friend of mine cared because she was a nurse, and she helped me, and she helped me with my personal care the first week, but after that I managed. (Clara, a 66-year-old lady, who broke her shoulder 2 years ago and was managed surgically with a locking plate)

Where patients were lacking family support, social care was required, which some patients found embarrassing or intrusive; for many, this had an impact (in most cases temporary) on their emotional state:

So I had a nurse that came in and helped, came in for six weeks to help with my meals. Although I didn’t really want help, but that was an awkward part … I lost a wee bit of my pride. (Mick, a 75-year-old man, who broke his shoulder 2 years ago and was managed surgically with a locking plate)

In cases where social care and family care were lacking, particularly outside working hours, there was an acute sense of loneliness and a lack of support for getting good quality, pain free sleep:

Well at night the main problem was – well nobody was helping me at night because I was alone. (Alice, a 67-year-old lady, who broke her shoulder a year ago and was managed non operatively)

In summary, functional social support was provided by family, and this helped to ameliorate the effects of injury. Often, this was a short-term arrangement and provided motivation for patients to regain some function in order to perform personal care. Where this was unavailable, social care provided some support – while this was not ideal, it was still preferential to no support at all.

Moderator of experience: relationship with health professionals

Patient relationships with healthcare professionals contributed to patient perceptions of recovery. Subthemes focused on healthcare professional appearance, behaviour, support, their ability to set expectations and trustworthiness. Having trust in their healthcare professionals was particularly important for patients in order to alleviate any distress around treatment planning and progress:

Well to be honest with you once I saw Dr X my mind was at rest because I was comfortable and he explained everything to me … I think the way he handled my situation was absolutely perfect for me. (Reema, a 65-year-old lady, who broke her shoulder a year ago and was managed non operatively)

In particular, healthcare professionals were instrumental in setting goals for regaining shoulder function, particularly when patients wanted to regain levels of physical activity consistent with previous lifestyle and level of activity:

The first thing that Mr X asked me was, ‘What was my lifestyle?’ (Suzy, a 67-year-old lady, who broke her shoulder 2 years ago and was managed surgically with a locking plate)

This relationship between patient and healthcare professional was particularly important in cases where patients suffered a complication and needed further surgery. One patient described how he remained positive in spite of the complication:

So I knew something was wrong and that it was going to be rectified. I was happy that it happened whilst I was in hospital as opposed to being at home. I can’t fault the actual treatment I had. (Mick, a 75-year-old man, who broke his shoulder 2 years ago and was managed surgically with a locking plate)

In summary, healthcare professionals were integral to the patient experience from the initial consultation through to recovery. Positive personal and professional attributes of the healthcare professionals increased levels of patient trust. This in turn impacted on patient beliefs and expectations for the recovery of shoulder function and the associated improvement in their emotional state.

Moderator of experience: experience with healthcare institution

Patients’ experiences of the healthcare institution in which they were treated were described in subthemes describing the impact of reduced access to the health service, delays in treatment and perceptions of failure of the system. Negative experiences predominantly focused on problems accessing healthcare facilities and the impact of staff workload on staff availability. This often first came to light in the initial aftermath of injury when accessing fracture clinic appointments:

We had an absolute nightmare both the GP and me trying to phone through to get somebody to give me an appointment. (Alice, a 67-year-old lady, who broke her shoulder a year ago and was managed non operatively)

Furthermore, delays were experienced with delays to surgery, discharge and physiotherapy appointments:

I felt very frustrated because it’s very hard to remember the exercises that you’re given to do, and then you don’t see them for another week or two weeks. (Clara, a 66-year-old lady, who broke her shoulder 2 years ago and was managed surgically with a locking plate)

Similar problems were also experienced throughout the stay in hospital:

In hospital here I found that they are very short of nurses on the ward that I was in. … they have not got the staff to help. (Viola, a 68-year-old lady, who broke her shoulder 2 years previously and was managed surgically with a locking plate)

In summary, healthcare institutions were often associated with negative experiences arising from poor quality care. Patients described that this had a negative impact on their emotional state. Where patients reported better experiences care, it was often linked to their healthcare professional rather than the institution and the associated processes.

Discussion

This study is the first study to explore older adults’ experiences of being treated for and recovering from proximal humerus fracture. The thematic analysis identified seven themes. Patient perceptions of recovery were characterized by pain, lack of sleep and loss of function, particularly in the immediate aftermath of the injury. This left patients feeling exhausted; however, the pain often settled after 6 months. In cases where patients developed complications such as AVN, the recurrence of pain was an important factor. Loss of function was an important theme for patients when it persisted for more than 6 months. Patients’ capacity to manage their new level of disability, either through adaptive behaviours (such as training their other arm to carry out tasks) or through avoiding certain activities such as carrying heavy shopping, was critical to how they viewed their long-term outcome.

Good support networks involving family and friends were vital in helping patients adapt to their new disability, and it was also important as they required assistance following the initial injury. Their relationship with healthcare professionals strongly influenced their outlook and any follow-up surgery that occurred. Some patients struggled with their mood and emotional state following their injury, particularly when they developed avoidant behaviour or felt that their injury led to giving up activities they previously enjoyed. Discontent and anger were experienced when there was difficulty in accessing healthcare institutions and staff during the recovery period.

Currently, there is an absence of patient-related outcome measures that apply to complex proximal humeral fractures in the elderly population. The OSS has been used increasingly in quantitative shoulder research (Boyle et al., 2013; Rangan et al., 2015; Schliemann et al., 2015). This was developed in patients undergoing elective surgery with an average age of 57. The OSS focuses on three themes: pain, sleep and loss of function. Although our study also identified these as being relevant to recovery, the OSS places a heavy emphasis on shoulder function, with this being the focus of 9 out of 12 domains. In the results presented here, patients did discuss their functional outcome, but they also described how their ability to adapt to any residual disability played an important part in how they perceived their functional outcome. Therefore, future iterations of PROMs should consider the moderating or facilitating role of adaptive behaviour during the recovery period, as this may help explain patient’s assessment of functional outcome.

Other fields of medicine have also demonstrated that factors such as adaptive behavioural patterns, family support networks and strong inter-personal relationships with healthcare professionals moderate the psychological effects of disease (Uchino et al., 1996). Moreover, good inter-personal relationships with healthcare professionals are recognized as being important in affecting patient outcomes (AAOS, 2011; Stewart, 1995; Street et al., 2009). Although this relationship has not been studied in detail in an orthopaedic setting in relation to patient outcomes, in our study, patients identified the patient–surgeon relationship as important in terms of their long-term recovery, their ability to come to terms with their loss of function and any complications that they suffered.

Support from family, friends and social services is important during a patient’s recovery; this is particularly relevant in the first 6 weeks of a procedure and has been shown in lower limb elective surgery (Theiss et al., 2011) but not in upper limb trauma. While patients interviewed in our study were often discharged from the emergency department and then seen at an outpatient follow-up appointment because they were ambulatory, they were anxious and reliant on family during this period as well as during the immediate post-operative period, for those that underwent surgery. Family support also affected the outcomes in the later stages of recovery, as those with support had developed better adaptive behaviours and had a more positive outlook.

Access to healthcare was a notable cause of frustration in our study group. It is known that better access to healthcare improves clinical outcomes (Arraras et al., 2013; Raivio et al., 2014). Initially, patients complained of problems with arranging fracture clinic follow-up and waiting for confirmation of operation dates. Later on in their recovery, limited time with their physiotherapist was a common source of frustration. The effect of poor access to health professionals and institutions was an important finding and underlines the need for further research on how these factors influences patients’ perceptions of their clinical outcomes. Such work may have important policy implications especially in the context of clinical pathway development to optimize patient outcomes.

One of the key implications of this study is that the breadth of themes emerging from this qualitative analysis extend beyond the domains reported in quantitative measurement tools (Dawson et al., 1996). If health outcomes are to truly reflect the values of the target population with this specific injury, there is a need to develop a new appropriately validated PROM that accurately reflects the experiences of older adults recovering from proximal humeral fractures. Existing measures are heavily weighted towards patient function; however, as demonstrated by the findings of this study, there is far more to evaluate when looking at aspects of recovery that patients perceive to be important. Furthermore, while randomized trials within this field may adopt a myriad of health measures to look for differences in treatment options, the provision of a single validated PROM for this condition and population of patients offers the potential to limit the burden of exhaustive assessment with the various measurement tools used on patients enrolled in these trials.

There are three main limitations to this study. First, some patients struggled to recall the events following their injury as at least a year had passed. Recall bias could be reduced in future research by conducting a prospective qualitative analysis during recovery. Second, this study does not differentiate between the patient experiences of different management strategies such as hemiarthroplasty, internal fixation or non-operative management. The recruitment strategy and study design were designed to develop themes common to all patients, and future work will need a larger group of patients undergoing the different treatments to conceptualise why patients may have better quantitative outcomes after specific treatments. Third, this study was carried out in a single centre, therefore the external validity may be in question; however, these patients are not polytrauma patients and the selection criteria adopted makes the target population representative of the majority of elderly patients who would sustain complex proximal humerus fracture.

Conclusion

Seven themes were identified in this qualitative analysis of patient recovery following injury. Pain, sleep and function were themes identified in this study that have been used in existing quantitative assessment. However, patients valued the improvement in their sleep and pain more than their function. This is at odds with existing quantitative measures where a heavier emphasis is placed on function. Other themes that were found but had not previously been described in this field included emotional state, the role of social support networks, relationships with healthcare professionals and the healthcare institutions. These findings justify further work into qualitative assessment of patient recovery so that a more representative PROM can be developed.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship and/or publication of this article.

ORCID iDs

Sanjeeve Sabharwal

Stephanie Archer

References

AAOS (2011) Patient physician communication. Available at: https://www.aaos.org/uploadedFiles/PreProduction/About/Opinion_Statements/position/1182%20Shared%20Physician%20Patient%20Communication.pdf

Arraras

Illarramendi

Viudez

, et al. (2013) Determinants of patient satisfaction with care in a Spanish oncology day hospital and its relationship with quality of life. Psychooncology 22: 2454–2461.

Bardgett

Lally

Malviya

, et al. (2016) Patient-reported factors influencing return to work after joint replacement. Occupational Medicine (London) 66: 215–221.

Boyle

Youn

Frampton

, et al. (2013) Functional outcomes of reverse shoulder arthroplasty compared with hemiarthroplasty for acute proximal humeral fractures. Journal of Shoulder and Elbow Surgery 22: 32–37.

Braun

Clarke

(2006) Using thematic analysis in psychology. Qualitative Research in Psychology 32: 77–101.

Dawson

Fitzpatrick

Carr

(1996) Questionnaire on the perceptions of patients about shoulder surgery. The Journal of Bone and Joint Surgery 78: 593–600.

Handoll

Brorson

(2015) Interventions for treating proximal humeral fractures in adults. Cochrane Database of Systematic Reviews 12: CD000434.

Moore

Blom

Whitehouse

, et al. (2015) Deep prosthetic joint infection: A qualitative study of the impact on patients and their experiences of revision surgery. BMJ Open 5: e009495.

Palvanen

Kannus

Niemi

, et al. (2006) Update in the epidemiology of proximal humeral fractures. Clinical Orthopaedics and Related Research 442: 87–92.

10.

Pope

Mays

(1995) Reaching the parts other methods cannot reach: An introduction to qualitative methods in health and health services research. British Medical Journal 311: 42–45.

11.

Raivio

Jääskeläinen

Holmberg-Marttila

, et al. (2014) Decreasing trends in patient satisfaction, accessibility and continuity of care in Finnish primary health care – A 14-year follow-up questionnaire study. BMC Family Practice 15: 98.

12.

Rangan

Handoll

Brealey

, et al. (2015) Surgical vs nonsurgical treatment of adults with displaced fractures of the proximal humerus: The PROFHER randomized clinical trial. JAMA 313: 1037–1047.

13.

Schliemann

Hartensuer

Koch

, et al. (2015) Treatment of proximal humerus fractures with a CFR-PEEK plate: 2-year results of a prospective study and comparison to fixation with a conventional locking plate. Journal of Shoulder and Elbow Surgery 24: 1282–1288.

14.

Stewart

(1995) Effective physician-patient communication and health outcomes: A review. Canadian Medical Association Journal 152: 1423–1433.

15.

Street

Makoul

Arora

, et al. (2009) How does communication heal? Pathways linking clinician-patient communication to health outcomes. Patient Education and Counseling 74: 295–301.

16.

Theiss

Ellison

Tea

, et al. (2011) The connection between strong social support and joint replacement outcomes. Orthopedics 34: 357.

17.

Torrens

Corrales

Vila

, et al. (2011) Functional and quality-of-life results of displaced and nondisplaced proximal humeral fractures treated conservatively. Journal of Orthopaedic Trauma 25: 581–587.

18.

Trickett

Mudge

Price

, et al. (2012) A qualitative approach to recovery after open tibial fracture: The road to a novel, patient-derived recovery scale. Injury 43: 1071–1078.

19.

Uchino

Cacioppo

Kiecolt-Glaser

(1996) The relationship between social support and physiological processes: A review with emphasis on underlying mechanisms and implications for health. Psychological Bulletin 119: 488–531.

20.

Valenti

Aliani

Maroun

, et al. (2017) Shoulder hemiarthroplasty for proximal humeral fractures: Analysis of clinical and radiographic outcomes at midterm follow-up: A series of 51 patients. European Journal of Orthopaedic Surgery & Traumatology 27: 309–315.

21.

Van de Water

Shields

Davidson

, et al. (2014) Reliability and validity of shoulder function outcome measures in people with a proximal humeral fracture. Disability and Rehabilitation 36: 1072–1079.

22.

Williams

Price

Beard

, et al. (2013) The effects of age on patient-reported outcome measures in total knee replacements. The Bone & Joint Journal 95: 38–44.