Advertisement

Gastrostomies Preserve But Do Not Increase Quality of Life for Patients and Caregivers

Published:November 10, 2016DOI:https://doi.org/10.1016/j.cgh.2016.10.032

      Background & Aims

      Gastrostomies are widely used to provide long-term enteral nutrition to patients with neurologic conditions that affect swallowing (eg, following a cerebrovascular accident or for patients with motor neuron disease) or with oropharyngeal malignancies. The benefits derived from this intervention are uncertain for patients and caregivers. We conducted a prospective, multicenter cohort study to determine how gastrostomies affect health-related quality of life (HRQoL) in recipients and caregivers.

      Methods

      We performed a study of 100 patients who received gastrostomies (55% percutaneous endoscopic gastrostomy, 45% radiologically inserted) at 5 centers in the United Kingdom, 100 caregivers, and 200 population control subjects. We used the EuroQol-5D (comprising a questionnaire, index, visual analogue scale) to assess HRQoL for patients and caregivers before the gastrostomy insertion and then 3 months afterward; findings were compared with those from control subjects. Ten patients and 10 caregivers were also interviewed after the procedure to explore quantitative findings. Findings from the EuroQol-5D and semi-structured interviews were integrated using a mixed-methods matrix.

      Results

      Six patients died before the 3-month HRQoL reassessments. We observed no significant longitudinal changes in mean EuroQol-5D index scores for patients (0.70 before vs 0.710 after; P = .83) or caregivers (0.95 before vs 0.95 after; P = .32) following gastrostomy insertion. The semi-structured interviews revealed problems in managing gastrostomy tubes, social isolation, and psychological and emotional consequences that reduced HRQoL.

      Conclusions

      We performed a mixed-methods prospective study of the effects of gastrostomy feeding on HRQoL. HRQoL did not significantly improve after gastrostomy insertion for patients or caregivers. The lack of significant decrease in HRQoL after the procedure indicates that gastrostomies may help maintain HRQoL. Findings have relevance to those involved in gastrostomy insertion decisions and indicate the importance of carefully selecting patients for this intervention, despite the relative ease of insertion.

      Keywords

      Abbreviations used in this paper:

      EQ-5D (EuroQOL-5-Dimensions), HRQoL (health-related quality of life), PEG (percutaneous endoscopic gastrostomy), RIG (radiologically inserted gastrostomies), VAS (visual analogue scale)
      See editorial on page 998.
      Gastrostomies are widely used to provide long-term enteral nutrition when oral intake is inadequate. They are most frequently performed in patients with neurologic conditions affecting swallowing (eg, post cerebrovascular accident, motor neurone disease) and in those with oropharyngeal malignancy.
      • Kurien M.
      • McAlindon M.E.
      • Westaby D.
      • et al.
      Percutaneous endoscopic gastrostomy (PEG) feeding.
      A total of 17,000 gastrostomies are estimated to be inserted annually in the United Kingdom, compared with 125,000 procedures in the United States.
      • Roche V.
      Percutaneous endoscopic gastrostomy. Clinical care of PEG tubes in older adults.
      • Duszak Jr., R.
      • Mabry M.R.
      National trends in gastrointestinal access procedures: an analysis of Medicare services provided by radiologists and other specialists.
      • Day L.W.
      • Nazareth M.
      • Sewell J.L.
      • et al.
      Practice variation in PEG tube placement: trends and predictors among providers in the United States.

      NCEPOD. Scoping our practice: the 2004 report of the National Confidential Enquiry into Patient Outcome and Death. 2005. NCEPOD. London. Available at: http://www.ncepod.org.uk/2004report/04sum.pdf. Accessed December 2, 2016.

      These estimate values and paucity of contemporary data internationally, highlight the uncertainty that currently exists in the medical literature regarding the frequency of enteral access placements.
      • Day L.W.
      • Nazareth M.
      • Sewell J.L.
      • et al.
      Practice variation in PEG tube placement: trends and predictors among providers in the United States.
      Although these procedures are widely performed, controversies exist regarding the merits of this intervention. These controversies reflect the absence of an evidence base supporting their role in certain patient groups, and the high morbidity and mortality identified in others.
      • Sanders D.S.
      • Carter M.J.
      • D'Silva J.
      • et al.
      Percutaneous endoscopic gastrostomy: a prospective audit of the impact of guidelines in two district general hospitals in the United Kingdom.
      • Katzberg H.D.
      • Benatar M.
      Enteral tube feeding for amyotrophic lateral sclerosis/motor neuron disease.
      • Dennis M.S.
      • Lewis S.C.
      • Warlow C.
      Effect of timing and method of enteral tube feeding for dysphagic stroke patients (FOOD): a multicentre randomised controlled trial.
      • Nugent B.
      • Lewis S.
      • O'Sullivan J.M.
      Enteral feeding methods for nutritional management in patients with head and neck cancers being treated with radiotherapy and/or chemotherapy.
      Quality of life is another important health outcome measure, which has been poorly characterized in this group of patients. Improving an understanding of this outcome measure is pertinent to those involved in the decision-making process regarding gastrostomy insertion, because improvements in survival and nutritional status could be perceived to be of limited gains, if no quality of life improvements are achieved for patients.
      Health-related quality of life (HRQoL) can be defined as the way illness, pain, and motor activity influences daily behavior, social activities, psychological well-being, and other aspects of an individual’s life.
      • Wanden-Berghe C.
      • Nolasco A.
      • Sanz-Valero J.
      • et al.
      Health-related quality of life in patients with home nutritional support.
      Currently, an understanding of HRQoL in gastrostomy patients has been focused on patients with cancer,
      • Senft M.
      • Fietkau R.
      • Iro H.
      • et al.
      The influence of supportive nutritional therapy via percutaneous endoscopically guided gastrostomy on the quality of life of cancer patients.
      • Roberge C.
      • Tran M.
      • Massoud C.
      • et al.
      Quality of life and home enteral tube feeding: a French prospective study in patients with head and neck or oesophageal cancer.
      • Van Bokhorst-de Van der Schuer M.A.
      • Langendoen S.I.
      • Vondeling H.
      • et al.
      Perioperative enteral nutrition and quality of life of severely malnourished head and neck cancer patients: a randomized clinical trial.
      • Rogers S.N.
      • Thomson R.
      • O'Toole P.
      • et al.
      Patients experience with long-term percutaneous endoscopic gastrostomy feeding following primary surgery for oral and oropharyngeal cancer.
      • Oates J.E.
      • Clark J.R.
      • Read J.
      • et al.
      Prospective evaluation of quality of life and nutrition before and after treatment for nasopharyngeal carcinoma.
      • Salas S.
      • Baumstarck-Barrau K.
      • Alfonsi M.
      • et al.
      Impact of the prophylactic gastrostomy for unresectable squamous cell head and neck carcinomas treated with radio-chemotherapy on quality of life: prospective randomized trial.
      • El-Deiry M.W.
      • Futran N.D.
      • McDowell J.A.
      • et al.
      Influences and predictors of long-term quality of life in head and neck cancer survivors.
      • Kwong J.P.
      • Stokes E.J.
      • Posluns E.C.
      • et al.
      The experiences of patients with advanced head and neck cancer with a percutaneous endoscopic gastrostomy tube: a qualitative descriptive study.
      despite most gastrostomy insertions occurring in noncancer individuals.
      • Arora G.
      • Rockey D.
      • Gupta S.
      High In-hospital mortality after percutaneous endoscopic gastrostomy: results of a nationwide population-based study.
      • Stavroulakis T.
      • Baird W.O.
      • Baxter S.K.
      • et al.
      The impact of gastrostomy in motor neurone disease: challenges and benefits from a patient and carer perspective.
      • Zamietra K.
      • Lehman E.B.
      • Felgoise S.H.
      • et al.
      Non-invasive ventilation and gastrostomy may not impact overall quality of life in patients with ALS.
      The influence gastrostomies have on caregivers is also poorly appreciated. A recent British Medical Journal editorial discusses the impact healthcare interventions have on caregivers and raises concerns about the current paucity of knowledge about caregiver outcomes.
      • Al-Janabi H.
      • Nicholls J.
      • Oyebode J.
      The need to “carer proof” healthcare decisions.
      Researchers are now challenged to provide better evidence of the effectiveness and efficiency of healthcare intervention from the perspective of all individuals involved in healthcare decisions (ie, inclusive of caregivers). Through advancing understanding of caregiver outcomes, the novel concept of “carer-proofing” healthcare decisions can be achieved.
      • Al-Janabi H.
      • Nicholls J.
      • Oyebode J.
      The need to “carer proof” healthcare decisions.
      This need to carer/caregiver-proof healthcare decisions is pertinent to gastrostomy insertion decisions.
      Mixed-methods research can be used to assesses HRQoL. Mixed-methods research combines elements from differing research methods (eg, quantitative and qualitative methodologies) to produce converging findings in the context of complex research questions.
      • Lingard L.
      • Albert M.
      • Levinson W.
      Grounded theory, mixed methods, and action research.
      The use of this novel technique has increased over recent years through its ability to enhance understanding, and achieve clinically relevant findings.
      • O'Cathain A.
      • Murphy E.
      • Nicholl J.
      Why, and how, mixed methods research is undertaken in health services research in England: a mixed methods study.
      Another merit of this technique is that views from individual patients and caregivers are expressed. Incorporation of individual’s reported outcomes has the potential of helping inform the clinical decision-making process, pertinent to gastrostomy insertion decisions.
      • Black N.
      Patient reported outcome measures could help transform healthcare.
      This prospective, multicenter study aimed to evaluate how gastrostomies influence HRQoL in patients and caregivers using mixed-methods research.

      Methods

       Setting

      Adult participants (>16 years old) were enrolled from 5 hospitals in South Yorkshire and North East Derbyshire, United Kingdom (Royal Hallamshire Hospital, Northern General Hospital, Chesterfield Royal Hospital, Rotherham Hospital, Doncaster Royal Infirmary). These hospitals in combination serve a population of just over 1.5 million people and perform approximately 350 gastrostomies annually.
      • Kurien M.
      • Westaby D.
      • Romaya C.
      • et al.
      National survey evaluating service provision for percutaneous endoscopic gastrostomy within the UK.
      At all hospitals percutaneous endoscopic gastrostomy (PEG) and radiologically inserted gastrostomies (RIG) were available as potential methods of inserting a gastrostomy. No surgically inserted gastrostomies were performed during the study period. Selection as to which procedure was undertaken was left to the discretion of the referring clinician, with similar outcomes following the 2 techniques (PEG and RIG) reported previously by our group.
      • Leeds J.S.
      • McAlindon M.E.
      • Grant J.
      • et al.
      Survival analysis after gastrostomy: a single-centre, observational study comparing radiological and endoscopic insertion.

       Group 1: Assessment of Health-Related Quality of Life

      An explanatory sequential mixed-methods design was used incorporating 2 differing HRQoL assessment tools.
      • Cresswell J.
      • Plano Clark V.L.
      Designing and conducting mixed methods research.
      The rationale for using a mixed-methods approach was to attain a more complete picture of an individual’s quality of life, using different forms of complementary data. The 2-phase design of this study involved collection and analysis of quantitative data from patients and caregivers, before a second phase when qualitative data were collected. This approach enabled exploration and possible explanation of the quantitative findings and was used to assist in sampling.

       Participants

      Patients referred for a gastrostomy and caregivers of patients with a gastrostomy were eligible for inclusion in this study. Participants were excluded if they were too unwell, unable to communicate, or if English was not understood. Caregivers were defined as an adult family member or friend providing unpaid care for an individual with a gastrostomy. Patients were most frequently recruited alongside their own caregivers, although occasionally patients and caregivers were recruited in isolation if their counterpart was ineligible for the study (eg, unable to communicate).
      Nutrition teams within individual hospitals helped identify potential participants for this study, who were then approached by clinical research nurses if they agreed. Information sheets were provided either by direct contact with the participant in hospital or by writing to them personally at home. In both cases, individuals were contacted a few days after the information sheet was provided, to question if they wanted to meet and discuss the study further with a member of the research team or take part. The research team met those who wanted to meet and following informed consent being obtained, participants were enrolled.

       Phase 1: Quantitative Assessment

      EuroQol-5D (EQ-5D) is the National Institute of Clinical Excellence’s preferred measure of HRQoL in adults.

      NICE. Guide to the methods of technology appraisal. NICE, London 2008.

      It is a quick and easy bedside tool that takes 2–3 minutes to complete consisting of a questionnaire, index, and visual analogue scale (VAS). EQ-5D has previously been evaluated in the area of gastrostomy feeding.
      • Wanden-Berghe C.
      • Nolasco A.
      • Sanz-Valero J.
      • et al.
      Health-related quality of life in patients with home nutritional support.
      Five dimensions (mobility, self-care, usual activity, pain/discomfort, and anxiety/depression) and 3 levels (no problems, some problems, extreme problems) in the EQ-5D questionnaire create 243 unique health states. These states can be then converted into a single index value between -0.59 and 1.00, facilitating the calculation of quality-adjusted life years. The VAS is a millimeter scale, 20 cm in length, ranging from 0 (worst imaginable HRQoL) to 100 (best imaginable HRQoL). The VAS complements the subject’s description of HRQoL within the questionnaire.
      In this study, we use EQ-5D to assess HRQoL in patients and caregivers at baseline (before gastrostomy insertion) and then again at 3 months (postinsertion). The decision to reassess patients and caregivers at 3 months was pragmatic, allowing sufficient time for gastrostomy feeding to have any potential influence on HRQoL, while also considering the high mortality rates identified previously following this intervention. In addition to assessing longitudinal changes in HRQoL in patients and caregivers using EQ-5D, findings at 3 months post gastrostomy insertion were compared with a control group drawn from the general population of Sheffield.

       Phase 2: Qualitative Assessment

      Semi-structured interviews were conducted on a sample of participants who had undergone phase 1. The principle investigator (MK) undertook these interviews, having had no previous involvement in the clinical decision-making process regarding gastrostomy insertion, which could have potentially influenced responses. Purposive sampling was undertaken to ensure assessment of a cross-sectional cohort with differing underlying conditions. Findings and variables assessed in phase 1 study were used to refine interview schedules for the semi-structured interviews, aiming to gain a richer understanding of results obtained using the quantitative assessment. Most questions asked were open ended and undertaken face-to-face within an individual’s own home or if requested in private surroundings within hospitals. Slight differences were made to the interview schedule of the caregivers when compared with the patients, reflecting the impact gastrostomy insertion had on them personally. Most participants were interviewed in isolation; however, 2 patients requested the presence of a relative during the interviews, which was allowed. In patients with dysarthria, written communication was permitted during the course of the interviews. All interviews were undertaken in English, with an average length of 22 minutes. Following informed consent, interviews were tape recorded and transcribed later, with all potentially identifying information excluded or coded. Transcripts were sent to participants to be reviewed afterward, ensuring accuracy of the transcription and allowing a means of quality control.

       Analysis

      The phase 1 quantitative data findings were analyzed using statistical packages for the social sciences (version 20.0; SPSS Inc, Chicago, IL). Baseline and 3 month postinsertion EQ-5D index and VAS scores were compared using a paired Student t test, after Shapiro-Wilk normality testing indicated a good fit (P > .05). An independent Student t test was used to compare findings between differing groups, with P < .05 considered significant. In phase 2 of the study, a thematic interpretive analysis was used, where transcripts were coded in NVivo version 10 (QSR International Pty Ltd, Melbourne, Australia) using a coding structure, based on the interview schedule and findings from phase 1. These codes were subsequently organized by theme, enabling comparisons to be made between differing individuals. Integration of the EQ-5D findings and semi-structured interviews was then done using a mixed-methods matrix, allowing further exploration and understanding of quantitative findings from the qualitative data.
      • O'Cathain A.
      • Murphy E.
      • Nicholl J.
      Three techniques for integrating data in mixed methods studies.

       Ethical Considerations

      This study protocol was approved by the Yorkshire and the Humber Research Ethics committee (REC reference 11/YH/0152), registered with the Sheffield Research and Development department (reference STH15871), and included in the National Institute for Health Research Clinical Research Network Portfolio (Portfolio ID 11090).

      Results

      A total of 100 gastrostomy patients and 100 caregivers were prospectively recruited during the study period, alongside 200 population control subjects. Characteristics of these individuals are demonstrated in Table 1. PEGs accounted for 55% of the gastrostomy insertions, with the remaining insertions being performed radiologically (RIGs). All caregivers completed follow-up assessment at 3 months; however, 6% (6 of 100) of the patients who underwent a gastrostomy died before their 3-month follow-up. Of these 6 individuals, 3 had PEG insertion and 3 had RIG insertions.
      Table 1Characteristics of Patients, Caregivers, and the Population Control Subjects
      PatientsCaregiversControl subjects
      Total number100100200
      Mean age (standard deviation)67 (14.7)65 (12.2)60 (10.1)
      Sex (M/F)56/4446/5489/111
      Ethnicity, %
       White1009996
       Black012
       Asian002
      Patient's underlying diagnosis, %
       Cerebrovascular accident2425
       Oropharyngeal malignancy3330
       Neurodegenerative3435
       Other910

       Phase 1: Quantitative Data (EuroQOL-5-Dimensions)

      Table 2 shows baseline and 3-month follow-up EQ-5D scores for patients and caregivers, alongside EQ-5D scores for the population control subjects. When analyzing longitudinal changes in HRQoL scores, no significant change in mean EQ-5D index scores was noted in either the patients (0.70 vs 0.71; P = .83) or the caregivers (0.95 vs 0.95; P = .32) following gastrostomy. These findings were corroborated by the EQ-VAS, with mean scores unchanged at 3 months in either the patient group (67.2 vs 67.4; P = .65) or in the caregivers (96.5 vs 96.7; P = .18).
      Table 2Number (Percentage) of Respondents Reporting No, Moderate, and Severe Problems in EQ-5D Dimensions and Mean EQ-5D Index and EQ VAS Scores
      EQ-5D DomainsPatients baselinePatients (3 months)Caregivers baselineCaregivers (3 months)General population
      Total n = 100Total n = 94Total n = 100Total n = 100Total n = 200
      n (%)n (%)n (%)n (%)n (%)
      Mobility
       No problems40 (40.0)37 (39.4)86 (86.0)86 (86.0)192 (96.0)
       Moderate problems53 (53.0)50 (53.2)13 (13.0)13 (13.0)8 (4.0)
       Severe problems7 (7.0)7 (7.4)1 (1.0)1 (1.0)0 (0.0)
      Self-care
       No problems52 (52.0)48 (51.1)95 (95.0)97 (97.0)194 (97.0)
       Moderate problems44 (44.0)41 (43.6)4 (4.0)2 (2.0)6 (3.0)
       Severe problems4 (4.0)5 (5.3)1 (1.0)1 (1.0)0 (0.0)
      Usual activities
       No problems35 (35.0)36 (38.3)92 (92.0)93 (93.0)195 (97.5)
       Moderate problems60 (60.0)53 (56.4)7 (7.0)6 (6.0)4 (2.0)
       Severe problems5 (5.0)5 (5.3)1 (1.0)1 (1.0)1 (0.5)
      Pain/discomfort
       No problems62 (62.0)54 (57.4)93 (93.0)95 (95.0)188 (94.0)
       Moderate problems36 (36.0)39 (41.5)7 (7.0)5 (5.0)9 (4.5)
       Severe problems2 (2.0)1 (1.1)0 (0.0)0 (0.0)3 (1.5)
      Anxiety/depression
       No problems81 (81.0)76 (80.9)94 (94.0)92 (92.0)190 (95.0)
       Moderate problems19 (19.0)18 (19.1)5 (5.0)8 (8.0)8 (4.0)
       Severe problems0 (0.0)0 (0.0)1 (1.0)0 (0.0)2 (1.0)
      EQ-5D index mean (± SD)0.70 (0.27)0.71 (0.21)0.95 (0.15)0.95 (0.14)0.93 (0.14)
      EQ VAS mean (± SD)67.2 (14.7)67.4 (14.6)96.5 (9.3)96.7 (8.9)86.2 (11.8)
      SD, standard deviation.
      When compared with population control subjects, caregivers had comparable HRQoL (mean EQ-5D index score, 0.95 vs 0.93; P = .87) following gastrostomy unlike the patients, who had significantly lower HRQoL at 3 months (mean EQ-5D index score, 0.73 vs 0.94; P < .0001). Outcomes in the gastrostomy patients did differ depending on the underlying referral indication, with lowest HRQoL at 3 months being in individuals who had a previous cerebrovascular accident (mean EQ-5D index score, 0.513), followed by progressive neurodegenerative conditions (mean EQ-5D index score, 0.657), and then oropharyngeal malignancy (mean EQ-5D index score, 0.835). In the 9 patients who underwent a gastrostomy for alternative indications the mean EQ-5D index score was 0.756.

       Phase 2: Semi-Structured Interviews

      Ten patients and 10 caregivers agreed to participate in the interviews. Of the patients, 2 had undergone a gastrostomy following a previous cerebrovascular accident, 4 had oropharyngeal cancer, and 4 had progressive neurologic conditions. Six of the caregivers were partners of patients requiring a gastrostomy, 2 were children of patients, and 2 were siblings. Individually, they cared for 6 patients with progressive neurologic disease, 3 with oropharyngeal cancer and 1 who had a previous cerebrovascular accident. The framing of the topics for the semi-structured interviews were informed by the findings from phase 1 where lower HRQoL in patients was identified, alongside differences according to referral indication. Questions devised encouraged discussions as to why HRQoL was lower in patients and exploring factors that may have influenced these outcomes. Five main themes were identified from the data analysis: (1) expectations of gastrostomy feeding, (2) gastrostomy management, (3) physical well-being, (4) psychological and emotional welfare, and (5) social consequences of gastrostomy feeding. Each theme is discussed next incorporating patients and caregivers insights.

       Expectations of Gastrostomy Feeding

      Patients and caregivers at the commencement of gastrostomy feeding conveyed high expectations, with aspirations that it would help improve nutrition and help survival. Many believed that information about gastrostomy feeding was well communicated by healthcare professionals at the outset, but some wished to have had more information, particularly pertaining to some of the long-term issues and practicalities. Most agreed that after 3 months, the gastrostomy had met expectations with regards to nutritional benefits (eg, weight gain, halting previously identified weight loss); however 2 caregivers questioned this benefit because the patient’s underlying medical condition had not significantly improved.

       Gastrostomy Management

      Once at home, caregivers had an integral part in the management of patients’ gastrostomy tubes. Many expressed early anxiety and lack of education as barriers to initial success, which were overcome with further experience and support from differing healthcare professionals. Some caregivers expressed a dependence on themselves to provide the feeding regimes to their relatives, which influenced their own daily routines. Although frequently expressed as being time consuming and sometimes messy, many patients and caregivers were accepting of the need for gastrostomy feeding, and had adopted strategies to maintain daily lifestyles and social interactions.

       Physical Well-Being

      A number of patients noted improvement in fatigue levels following gastrostomy insertion; however, this did not necessarily reflect improvements in actual physical function, with underlying diseases usually determining capabilities. Two patients commented on improvements in reflux and vomiting symptoms postgastrostomy. No direct influence of gastrostomy feeding was identified on the physical well-being of caregivers.

       Psychological and Emotional Welfare

      In both patients and caregivers, perceptions of psychological and emotional well-being were diverse. Patients frequently expressed long-term health as a concern and the impact this may have on their friends and family. One patient expressed a view that the gastrostomy feeding tube was a burden to them and their partner but that they had no alternative if they wanted to survive. The impact of not being able to eat was also explored in all patients and many missed the sensation of being able to eat and taste in the normal manner, when compared with feeding via a gastrostomy. With regards to the caregivers, concerns most often expressed by them were regarding their loved ones who had undergone gastrostomy placement and their future health. Financial concerns were also of anxiety to some, having previously been managed by their partners.

       Social Consequences of Gastrostomy Feeding

      Three patients reported that they felt socially excluded because they could not eat meals in a normal manner following gastrostomy. This made it difficult for them to watch other people eat and enjoy food, so they actively avoided mealtimes and often going out. Others felt that exclusion of mealtimes was beneficial to them following gastrostomy, because the pressure and struggle to eat no longer existed. With regard to getting out of the house, some patients and caregivers expressed that the gastrostomy was more discrete than other nutritional support methods, enabling them to do their normal daily activities in public without the perception of others of being unwell.

       Integration of Phase 1 and 2

      Following identification of differing mean EQ-5D scores in the different gastrostomy referral subgroups a mixed-methods matrix was created to explore findings (Table 3). This helped provide several potential explanations as to why gastrostomy patients have lower HRQoL when compared with the general population. Interestingly, although some of the responses from the interviews were disease specific, most perceptions provided by participants were relevant to all groups with social isolation, complications, and management issues of tubes all being pertinent to gastrostomy patients’ and caregivers’ HRQoL.
      Table 3Mixed Methods Matrix Exploring Potential Causes for HRQoL Differences Between Referral Indication Subgroups
      Gastrostomy indicationMean EQ-5D index score of patientsDefining statements made by patients during interviewsDefining statements made by caregivers during interviews
      Cerebrovascular accident0.513“I've had lots of problems with my gastrostomy, particularly with leakage” (B)“I don't like the look of feeding tubes” (D)
      “The feed is less pleasurable than eating” (E)“The PEG has helped him put on weight” (C)
      “The accumulation of saliva in my mouth is a big problem. I constantly need to spit it out” (E)
      Progressive neurodegenerative conditions0.657“I miss eating the things I like” (D)“I mainly look after the feeding tube and give the feeds” (B)
      “Feeding via the tube is time consuming” (E)“It took me a while to get used to the tube but once I got the hang of it, it was ok” (B)
      “I wish I had the tube gastrostomy tube inserted earlier” (A)“The sickness is a lot better and the weight has come back, which is what wanted” (C)
      Oropharyngeal malignancy0.835“The PEG helped as I had pain at the back of my throat following the radiotherapy” (C)“I'm not concerned about the gastrostomy tube, it’s the not knowing of cancer itself” (D)
      “The support teams were really important to both of us” (B)“Without family support, the last few months would have been very difficult” (D)
      “I was really anxious about having the gastrostomy inserted” (D)
      NOTE. Themes from where the defining statements were derived: (A) expectations of gastrostomy feeding, (B) gastrostomy management, (C) physical well-being, (D) psychological and emotional welfare, and (E) social consequences of gastrostomy feeding.

      Discussion

      This is the first mixed-methods study performed in gastrostomy feeding, demonstrating that HRQoL does not significantly improve in either patient recipients or caregivers following gastrostomy insertion. Although improvements were not identified, HRQoL did not significantly change over time in either group. Given that gastrostomy recipients frequently have chronic progressive illnesses where deterioration in HRQoL may be expected, the demonstration that HRQoL may be preserved following gastrostomy insertion is important and could be used to inform gastrostomy insertion decisions.
      In the first part of this study, quantitative assessment demonstrated that gastrostomy patients had significantly lower HRQoL compared with population control subjects, which was further explored using semi-structured interviews. This mixed-methods approach enabled a better understanding of HRQoL in these individuals, and also provided insights into the impact this had on their caregivers. Findings from this study could be used to support future clinical decisions regarding gastrostomy insertions and potentially “carer-proof” them, by better informing patients and their caregivers about the merits of this intervention.
      • Brotherton A.
      • Abbott J.
      Clinical decision making and the provision of information in PEG feeding: an exploration of patients and their carers' perceptions.
      • Carey T.S.
      • Hanson L.
      • Garrett J.M.
      • et al.
      Expectations and outcomes of gastric feeding tubes.
      This research could also provide an opportunity to coproduce with patients a resource to aid shared decision-making.
      • Oshima Lee E.
      • Emanuel E.J.
      Shared decision making to improve care and reduce costs.
      • Stiggelbout A.M.
      • Van der Weijden T.
      • De Wit M.P.
      • et al.
      Shared decision making: really putting patients at the centre of healthcare.
      The importance of patient selection and the need to appropriately counsel patients and their caregivers before gastrostomy insertion is emphasized by this research. Although gastrostomies are relatively easy to perform, challenges are frequently encountered regarding gastrostomy insertion decisions in certain patients. Factors considered on an individual basis include survival, healing of pressure sores, reduction of aspiration, and the possibility of discharge to a nursing home.
      • Wilcox C.M.
      • McClave S.A.
      To PEG or not to PEG.
      Ethical and medicolegal issues add to the complexity of this decision making, alongside assessments of risk (procedural and patient related).
      • DeLegge M.H.
      • McClave S.A.
      • DiSario J.A.
      • et al.
      Ethical and medicolegal aspects of PEG-tube placement and provision of artificial nutritional therapy.
      Our research attempts to better characterize HRQoL. Although this is only 1 facet involved in the complex decision-making process, this outcome has advantages over other measures in being a patient- and a caregiver-reported outcome. Given the importance of patient autonomy in decision making, we believe that the true merits of our work is in helping endoscopists and healthcare teams provide more accurate and factual, but sensitive, education regarding gastrostomy benefits to patients and their family.
      • Wilcox C.M.
      • McClave S.A.
      To PEG or not to PEG.
      • DeLegge M.H.
      • McClave S.A.
      • DiSario J.A.
      • et al.
      Ethical and medicolegal aspects of PEG-tube placement and provision of artificial nutritional therapy.
      Given the absence of prospective randomized controlled trials evaluating the alternatives to gastrostomy feeding (eg, long-term nasogastric feeding, optimizing oral feeding techniques, or confining nutritional intake to a patient’s own volition), this work could help better inform patients and their caregivers.
      Although this study has strengths in being prospective in design, multicenter, and novel in using mixed methods, there are potential limitations to this work. First, patients who were too unwell or could not participate because of cognitive or communicative problems were not included in the HRQoL assessment. This could have resulted in a selection bias. This may also explain why the mortality was relatively low in this group of patients at 6% at 3 months. In addition, in assessing longitudinal change in HRQoL a longer follow-up of individuals beyond 3 months could have demonstrated different outcomes and potential improvements in HRQoL. The high baseline HRQoL in caregivers (comparable with the general population) may have also limited the opportunity for an improvement in HRQoL over time.
      Another limitation of this work is in the analysis of EQ-5D findings. Patients and caregiver groups showed no statistically significant change in mean index or VAS scores between baseline and 3 months. Although this may not represent statistical significance, slight differences in index scores could be deemed to be clinically relevant. Although there is a paucity of work evaluating the smallest change in EQ-5D index score needed to be important, 1 study found that the minimally important difference in EQ-5D index scores was 0.07 in 11 different patient groups (not including gastrostomy patients), which was not achieved in either our patient or caregiver groups at 3 months.
      • Walters S.J.
      • Brazier J.E.
      Comparison of the minimally important difference for two health state utility measures: EQ-5D and SF-6D.
      The validity of our comparator group to assess HRQoL is another potential limitation to our work. In this study we used population control subjects to compare HRQoL with both patients and caregivers. An alternative and possibly better comparator group could have been patients with similar diagnoses who refused or who were deferred a gastrostomy. Although this issue was considered, our previous study evaluating deferred gastrostomy patients showed that often these individuals were more unwell, limiting their potential to complete our HRQoL assessments, recruitment into the study, and potentially raising questions as to their suitability as appropriate comparators.
      • Kurien M.
      • Leeds J.S.
      • Delegge M.H.
      • et al.
      Mortality among patients who receive or defer gastrostomies.
      Although we recognize this limitation, we believe our study design was pragmatic and addresses a common perception that gastrostomies lead to improvement in outcomes.
      Previous published data assessing HRQoL following gastrostomy insertion are conflicting, with some studies suggesting HRQoL benefit following gastrostomy insertions.
      • Roberge C.
      • Tran M.
      • Massoud C.
      • et al.
      Quality of life and home enteral tube feeding: a French prospective study in patients with head and neck or oesophageal cancer.
      • Bannerman E.
      • Pendlebury J.
      • Phillips F.
      • et al.
      A cross-sectional and longitudinal study of health-related quality of life after percutaneous gastrostomy.
      • Schneider S.M.
      • Pouget I.
      • Staccini P.
      • et al.
      Quality of life in long-term home enteral nutrition patients.
      This study adds to the literature by being the largest prospective study evaluating a cross-sectional cohort of gastrostomy patients, and provides longitudinal outcomes in both gastrostomy patients and their caregivers. What our findings do support from previous work is that HRQoL does vary in gastrostomy patients dependent on the referral indication.
      • Wanden-Berghe C.
      • Nolasco A.
      • Sanz-Valero J.
      • et al.
      Health-related quality of life in patients with home nutritional support.
      • Schneider S.M.
      • Pouget I.
      • Staccini P.
      • et al.
      Quality of life in long-term home enteral nutrition patients.
      When our findings are considered in the context of our previous work evaluating mortality and other recently published prospective mortality studies, it seems that patients with oropharyngeal cancer have the best outcomes, both with regards to mortality and HRQoL.
      • Kurien M.
      • Leeds J.S.
      • Delegge M.H.
      • et al.
      Mortality among patients who receive or defer gastrostomies.
      • Blomberg J.
      • Lagergren P.
      • Martin L.
      • et al.
      Albumin and C-reactive protein levels predict short-term mortality after percutaneous endoscopic gastrostomy in a prospective cohort study.
      Our study also provides insights into why patients with gastrostomy feeding tubes have lower HRQoL when compared with population control subjects. Although underlying conditions did influence HRQoL, the burden associated with managing gastrostomy feeding tubes, the feelings of social isolation, and the impact this had on psychological and emotional well-being were cross cutting themes pertinent to most gastrostomy patients and caregivers.
      In conclusion, HRQoL does not significantly improve in either patient recipients or caregivers following gastrostomy insertion, although the absence of a significant fall suggests gastrostomies may still confer benefits by maintaining HRQoL. This work emphasizes the importance of patient selection and the need to appropriately counsel patients and their caregivers before gastrostomy insertion.

      Acknowledgments

      The authors thank the NIHR Clinical Research Network for their help and support in performing this study.

      References

        • Kurien M.
        • McAlindon M.E.
        • Westaby D.
        • et al.
        Percutaneous endoscopic gastrostomy (PEG) feeding.
        BMJ. 2010; 340: c2414
        • Roche V.
        Percutaneous endoscopic gastrostomy. Clinical care of PEG tubes in older adults.
        Geriatrics. 2003; 58 (28–29): 22-26
        • Duszak Jr., R.
        • Mabry M.R.
        National trends in gastrointestinal access procedures: an analysis of Medicare services provided by radiologists and other specialists.
        J Vasc Interv Radiol. 2003; 14: 1031-1036
        • Day L.W.
        • Nazareth M.
        • Sewell J.L.
        • et al.
        Practice variation in PEG tube placement: trends and predictors among providers in the United States.
        Gastrointest Endosc. 2015; 82: 37-45
      1. NCEPOD. Scoping our practice: the 2004 report of the National Confidential Enquiry into Patient Outcome and Death. 2005. NCEPOD. London. Available at: http://www.ncepod.org.uk/2004report/04sum.pdf. Accessed December 2, 2016.

        • Sanders D.S.
        • Carter M.J.
        • D'Silva J.
        • et al.
        Percutaneous endoscopic gastrostomy: a prospective audit of the impact of guidelines in two district general hospitals in the United Kingdom.
        Am J Gastroenterol. 2002; 97: 2239-2245
        • Katzberg H.D.
        • Benatar M.
        Enteral tube feeding for amyotrophic lateral sclerosis/motor neuron disease.
        Cochrane Database Syst Rev. 2011; : CD004030
        • Dennis M.S.
        • Lewis S.C.
        • Warlow C.
        Effect of timing and method of enteral tube feeding for dysphagic stroke patients (FOOD): a multicentre randomised controlled trial.
        Lancet. 2005; 365: 764-772
        • Nugent B.
        • Lewis S.
        • O'Sullivan J.M.
        Enteral feeding methods for nutritional management in patients with head and neck cancers being treated with radiotherapy and/or chemotherapy.
        Cochrane Database Syst Rev. 2013; 1: CD007904
        • Wanden-Berghe C.
        • Nolasco A.
        • Sanz-Valero J.
        • et al.
        Health-related quality of life in patients with home nutritional support.
        J Hum Nutr Diet. 2009; 22: 219-225
        • Senft M.
        • Fietkau R.
        • Iro H.
        • et al.
        The influence of supportive nutritional therapy via percutaneous endoscopically guided gastrostomy on the quality of life of cancer patients.
        Support Care Cancer. 1993; 1: 272-275
        • Roberge C.
        • Tran M.
        • Massoud C.
        • et al.
        Quality of life and home enteral tube feeding: a French prospective study in patients with head and neck or oesophageal cancer.
        Br J Cancer. 2000; 82: 263-269
        • Van Bokhorst-de Van der Schuer M.A.
        • Langendoen S.I.
        • Vondeling H.
        • et al.
        Perioperative enteral nutrition and quality of life of severely malnourished head and neck cancer patients: a randomized clinical trial.
        Clin Nutr. 2000; 19: 437-444
        • Rogers S.N.
        • Thomson R.
        • O'Toole P.
        • et al.
        Patients experience with long-term percutaneous endoscopic gastrostomy feeding following primary surgery for oral and oropharyngeal cancer.
        Oral Oncol. 2007; 43: 499-507
        • Oates J.E.
        • Clark J.R.
        • Read J.
        • et al.
        Prospective evaluation of quality of life and nutrition before and after treatment for nasopharyngeal carcinoma.
        Arch Otolaryngol Head Neck Surg. 2007; 133: 533-540
        • Salas S.
        • Baumstarck-Barrau K.
        • Alfonsi M.
        • et al.
        Impact of the prophylactic gastrostomy for unresectable squamous cell head and neck carcinomas treated with radio-chemotherapy on quality of life: prospective randomized trial.
        Radiother Oncol. 2009; 93: 503-509
        • El-Deiry M.W.
        • Futran N.D.
        • McDowell J.A.
        • et al.
        Influences and predictors of long-term quality of life in head and neck cancer survivors.
        Arch Otolaryngol Head Neck Surg. 2009; 135: 380-384
        • Kwong J.P.
        • Stokes E.J.
        • Posluns E.C.
        • et al.
        The experiences of patients with advanced head and neck cancer with a percutaneous endoscopic gastrostomy tube: a qualitative descriptive study.
        Nutr Clin Pract. 2014; 29: 526-533
        • Arora G.
        • Rockey D.
        • Gupta S.
        High In-hospital mortality after percutaneous endoscopic gastrostomy: results of a nationwide population-based study.
        Clin Gastroenterol Hepatol. 2013; 11: 1437-1444
        • Stavroulakis T.
        • Baird W.O.
        • Baxter S.K.
        • et al.
        The impact of gastrostomy in motor neurone disease: challenges and benefits from a patient and carer perspective.
        BMJ Support Palliat Care. 2016; 6: 52-59
        • Zamietra K.
        • Lehman E.B.
        • Felgoise S.H.
        • et al.
        Non-invasive ventilation and gastrostomy may not impact overall quality of life in patients with ALS.
        Amyotroph Lateral Scler. 2012; 13: 55-58
        • Al-Janabi H.
        • Nicholls J.
        • Oyebode J.
        The need to “carer proof” healthcare decisions.
        BMJ. 2016; 352: i1651
        • Lingard L.
        • Albert M.
        • Levinson W.
        Grounded theory, mixed methods, and action research.
        BMJ. 2008; 337: a567
        • O'Cathain A.
        • Murphy E.
        • Nicholl J.
        Why, and how, mixed methods research is undertaken in health services research in England: a mixed methods study.
        BMC Health Serv Res. 2007; 7: 85
        • Black N.
        Patient reported outcome measures could help transform healthcare.
        BMJ. 2013; 346: f167
        • Kurien M.
        • Westaby D.
        • Romaya C.
        • et al.
        National survey evaluating service provision for percutaneous endoscopic gastrostomy within the UK.
        Scand J Gastroenterol. 2011; 46: 1519-1524
        • Leeds J.S.
        • McAlindon M.E.
        • Grant J.
        • et al.
        Survival analysis after gastrostomy: a single-centre, observational study comparing radiological and endoscopic insertion.
        Eur J Gastroenterol Hepatol. 2010; 22: 591-596
        • Cresswell J.
        • Plano Clark V.L.
        Designing and conducting mixed methods research.
        2nd ed. Sage, Thousand Oaks, CA2011
      2. NICE. Guide to the methods of technology appraisal. NICE, London 2008.

        • O'Cathain A.
        • Murphy E.
        • Nicholl J.
        Three techniques for integrating data in mixed methods studies.
        BMJ. 2010; 341: c4587
        • Brotherton A.
        • Abbott J.
        Clinical decision making and the provision of information in PEG feeding: an exploration of patients and their carers' perceptions.
        J Hum Nutr Diet. 2009; 22: 302-309
        • Carey T.S.
        • Hanson L.
        • Garrett J.M.
        • et al.
        Expectations and outcomes of gastric feeding tubes.
        Am J Med. 2006; 119: 527
        • Oshima Lee E.
        • Emanuel E.J.
        Shared decision making to improve care and reduce costs.
        N Engl J Med. 2013; 368: 6-8
        • Stiggelbout A.M.
        • Van der Weijden T.
        • De Wit M.P.
        • et al.
        Shared decision making: really putting patients at the centre of healthcare.
        BMJ. 2012; 344: e256
        • Wilcox C.M.
        • McClave S.A.
        To PEG or not to PEG.
        Clin Gastroenterol Hepatol. 2013; 11: 1451-1452
        • DeLegge M.H.
        • McClave S.A.
        • DiSario J.A.
        • et al.
        Ethical and medicolegal aspects of PEG-tube placement and provision of artificial nutritional therapy.
        Gastrointest Endosc. 2005; 62: 952-959
        • Walters S.J.
        • Brazier J.E.
        Comparison of the minimally important difference for two health state utility measures: EQ-5D and SF-6D.
        Qual Life Res. 2005; 14: 1523-1532
        • Kurien M.
        • Leeds J.S.
        • Delegge M.H.
        • et al.
        Mortality among patients who receive or defer gastrostomies.
        Clin Gastroenterol Hepatol. 2013; 11: 1445-1450
        • Bannerman E.
        • Pendlebury J.
        • Phillips F.
        • et al.
        A cross-sectional and longitudinal study of health-related quality of life after percutaneous gastrostomy.
        Eur J Gastroenterol Hepatol. 2000; 12: 1101-1109
        • Schneider S.M.
        • Pouget I.
        • Staccini P.
        • et al.
        Quality of life in long-term home enteral nutrition patients.
        Clin Nutr. 2000; 19: 23-28
        • Blomberg J.
        • Lagergren P.
        • Martin L.
        • et al.
        Albumin and C-reactive protein levels predict short-term mortality after percutaneous endoscopic gastrostomy in a prospective cohort study.
        Gastrointest Endosc. 2011; 73: 29-36