Length of hospital stay and its predictors among burn patients in a resource limited setting; a multicenter prospective cohort

doi:10.21203/rs.3.rs-2751196/v1

Download PDF

Research Article

Length of hospital stay and its predictors among burn patients in a resource limited setting; a multicenter prospective cohort

https://doi.org/10.21203/rs.3.rs-2751196/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Introduction

Burn injury is a critical and under-recognized public health issue that affects people all over the world. Majority of the burns occur in low and middle income countries like Uganda. There is paucity of data relating to length of hospital stay among burn patients managed in hospitals with no burn units, yet this constitute a big percentage of hospitals in low income countries. This study was aimed at determining the predictors of length of stay at 3 hospitals in resource limited setting.

Methods

This was a prospective cohort multicentre study that enrolled 101 burns patients with moderate and severe burns admitted in 3 tertiary hospitals in western Uganda. Follow up was done till to discharge, length of hospital stay and its predictors determining using logistic regression in SPSS version 26.

Results

Of the 101 participants, the mean age was 21.3 (SD = 16.8) years. Majority were males (69.3%) from a rural setting (61.4%) and were accidentally burnt (95.0%) from home (75.2%). The mean length of hospital stay was 9 days with a standard deviation of 5.7. At multivariate level of analysis, the variables that independently predicted increased length of hospital stay were time to medical attention (AOR = 14.912, CI = 2.743–81.079, P = 0.002), burn severity (AOR = 6.173, CI = 1.392–27.371, P = 0.017) and type of procedure done (AOR = 0.047, CI = 0.009–0.245, P < 0.001).

Conclusion

This study revealed that the length of hospital stay was comparable to that seen in other studies. The factors independently associated with increased length of hospital stay were; time to medical attention, burn severity, and type of procedure done. Measures to ensure timely medical attention can reduce the length of stay among patients with burns even in the absence of a burn unit.

Burns

Length of hospital stay

resource limited setting

Uganda

Tissue damage resulting from hot or corrosive substances, electric current, radiation or friction constitute burn injury (1). Burn injury is a critical and under-recognized public health issue that affects people all over the world (2); However, the world health organization (WHO) states that the vast majority (over 95%) of fire-related burns occur in low and middle income countries like Uganda (3). In low- and middle-income nations, burn injuries are still a substantial source of death, financial loss by long hospital stays and disability (2). In 2018 the WHO gave the three regions with the greatest burden of burn injury: the Eastern Mediterranean Region, the South East Asian Region, and the African Region, with the African Region bearing nearly two-thirds of the total burden (3).

The findings of several studies have demonstrated the importance of different factors on the length of stay of burn injury patients. These factors include: demographics (age, sex), clinical data (admission time, burn size, etiology of burn, depth of burns and existence of inhalation injury), and biological parameters (serum electrolyte level changes), inhalational burns, among others (4–8). Burn injury has early and late outcomes, and Length of hospital stay (LOS) is one of the most important early outcomes. In Gulu, majority of burn patients spent at least 1 week in the hospital (82,6%) according to a study done by Kitara et al. (9). Even though burns have been reported to be associated with prolonged hospital stay, the factors that increase length of hospital stay among burn patients have not been well studied in rural Uganda. This study was aimed at determining the predictors of length of stay at 3 Tertiary hospitals in Uganda with no burn units.

Study design, setting and population

This was a prospective cohort multicenter study carried out at FortPortal regional referral hospital (FRRH), Hoima regional referral hospital (HRRH) and Kampala international university-Teaching hospital (KIU-TH). All the 3 hospitals are located in western Uganda. The hospitals do not have burn units and therefore the patients were managed in the general surgical ward. All patients admitted with burns during the study period were evaluated for eligibility. All patients with moderate to severe burns admitted to the surgical wards of the study centres were included in the study if they consented. For those unable to consent for themselves, the consent was obtained from the next of kin. Patient who were referred were excluded since it was difficult to follow them up. The sample size was determined using the modified Daniel formula. Using findings by Albutt et al. (10), \(n=\frac{\left({{Z}_{\alpha }+{Z}_{\beta })}^{2}\right)p\left(1-p\right)x D}{{e}^{2}}\); p=6.4%; α = 5%; β = 20% at statistical power of 80%; e= 5%; D= 2.0, \(n=375\). On adjusting to a finite population of 120, and adding 10% to cater for loss of follow up, the required sample size was 101.

Data collection procedure

Participants who fulfilled the inclusion criteria were enrolled into the study after signing an informed consent form. However, for very ill patients, consent was obtained from their next of kin. Assessment and resuscitation of the patients was done in accordance to the advanced trauma life support (ATLS) protocols. Participants were consecutively recruited by the principle investigator and trained research assistants until the required number of participants was obtained, blood sample from every consented participant was taken on admission and 72 hours from admission, to determine serum electrolyte levels (sodium, potassium and chloride). Participants were followed up to discharge to determine their length of stay.

Quality Control

The questionnaire was pretested for validity and reliability and necessary adjustments made. The research assistants were trained about the protocol to follow during the research. The principal investigator or his assistant crosschecked the data daily to ensure completeness. An accurate history, physical examination and electrolyte value of the patients was taken on admission and 72 hours from admission.

Data Analysis and Presentation

Date analysis was done using SPSS version 26. The length of stay was presented as a mean with corresponding standard deviation. Using visual binning procedure, the length of hospital stay was transformed into a categorical variable with stay below the median coded 0 and that above the median coded 1. Bivariate and multivariate analysis were done between the independent variables and the length of hospital stay. Factors whose p value was less than 0.2 at bivariate and those with scientific plausibility were re analyzed at multivariate level using backward logistic regression. Variables with p ≤ 0.05 at multivariate were considered significant predictors of length.

This study included 101 patients, the mean age was 21.3 (SD = 16.8) years, with majority in the age bracket of 18–55 (56.4%). Majority were males (69.3%). Majority of the participants were single (68.3%), from a rural setting (61.4%) and were burnt accidentally (95.0%) from home (75.2%) (Table 1).

Table 1

Socio demographic characteristics of study participants
Characteristic	Characteristic
Age (years)	Mean = 21.3, SD = 16.8, Min = 0.9, Max = 81.0
	Frequency	Percentage
Age group
< 18	40	39.6
18.0–55	57	56.4
> 55.0	4	4.0
Sex
Male	70	69.3
Female	31	30.7
Hospital
HRRH	55	54.5
FRRH	41	40.6
KIU-TH	05	4.9
Marital status
Single	69	68.3
Married	26	25.7
Separated	6	5.9
Education level
Never	32	31.7
Primary	32	31.7
Secondary	32	31.7
Tertiary	5	5.0
Residence
Rural	62	61.4
Urban	39	38.6
Place of burn
Home	76	75.2
Work	19	18.8
On road	6	5.9
Circumstance
Accidental	96	95.0
Intentional	5	5.0

SD = Standard deviation, Min = Minimum, Max = Maximum

Length of stay among burn patients.

In this study, the mean length of stay was 9 days with a standard deviation of 5.7 days. The median length of hospital stay was 8 days, with an interquartile range of 5–11 days. Majority of the participants had a length of hospital stay below the median (60.4%). Majority spent more than 1 week (53.5%) (Table 2).

Table 2

Length of stay among burn patients admitted at FRRH, HRRH and KIU-TH.
Statistic	Value
Mean	9.0 days
Standard deviation	5.7 days
Median	8 days
Interquartile range	5–11 days
Minimum	3 days
Maximum	28 days
LOS ≤ Median (8 days)	N = 61 (60.4%)
LOS > Median (8 day)	N = 40 (39.6%)
At most 1 week	47(46.5%)
More than 1 week	54(53.5%)

LOS = Length of hospital stay

Predictors of length of hospital stay among burn patients

At multivariate level of analysis, the variables that were independently associated with length of hospital stay and therefore predictors of having a length of stay above the median (8 days) were time to medical attention, burn severity, and type of procedure done. It was observed that a patient who received medical attention after 12 hours was 14.912 times (CI = 2.743–81.079, P = 0.002) more likely to stay longer compared to one who received care in less than 3 hours. A patient with a major burn was 6.173 times (CI = 1.392–27.371, P = 0.017) more likely to stay longer compared to the one with a moderate burn. A patient who underwent dressing alone was 0.047 times (CI = 0.009–0.245, P < 0.001) less likely to stay longer than a patient who had debridement. This means that the patient who had debridement was 21.277 (1/0.047) times more likely to stay longer compared to one that had dressing alone (Table 3). The results of bivariate analysis are supplied as supplementary file 1.

Table 3

Multivariable analysis of factors associated with length of hospital stay among burn patients in FRRH, HRRH and KIU-TH.
Characteristic	Bivariate analysis			Multivariate analysis
Characteristic	cOR	95% CI	P value	AOR	95% CI	P value
Time to medical attention
< 2.9 hrs	Ref			Ref
3-<5.9 hrs	0.485	0.166–1.416	0.185	0.111	0.022–0.563	0.008
6-11.9 hrs	2.667	0.407–17.477	0.307	3.714	0.333–41.400	0.286
> 12 hrs	4.622	1.417–15.073	0.011	14.912	2.743–81.079	0.002
Burn severity
Moderate	Ref			Ref
Major	3.692	1.544–8.827	0.003	6.173	1.392–27.371	0.017
Procedure
Debridement	Ref			Ref
Fasciotomy	1.600	0.512–4.996	0.418	0.213	0.035–1.314	0.096
Dressing alone	0.185	0.065–0.527	0.002	0.047	0.009–0.245	< 0.001

AOR = Adjusted odds ratio, Ref = Reference category, cOR = Crude odds ratio, CI = Confidence interval.

In this study, the mean length of stay was 9 days with a standard deviation of 5.7 days. The median length of hospital stay was 8 days, with an interquartile range of 5–11 days. The minimum LOS was 3 days while the maximum was 28. Majority of the participants had a length of hospital stay below the median (60.4%). Majority spent more than 7 days (53.5%).

The mean length of hospital stay (LOS) in this study was slightly lower than that reported by Knowlin et al. (11) where the average LOS was 12.7 ± 24.7 days. The range in our study was within the limits reported by Chukamei et al. (12) in Iran that included 899 patients with burns and reported a range of 1 to 47 days, with an average of 3.22 days, which average was much lower than the 9.0 seen in our study. The length of stay in this study was comparable to that reported by Kitara et al. (9) in Gulu-Uganda where 52.1% of the study participants spent at most 7 days in hospital, yet in our study, 46.5% spent a maximum of 7 days.

The length of hospital stay in this study was much lower than that reported in Mulago where the average length of stay for patients was 24.3 days (± 22.1 days) (13). This big difference is most likely because this study included patients admitted to the burn centre of the national referral hospital, which centre is more likely to admit patients with very severe burns. This could also explain the difference seen in Togo where the mean hospital stay was 21 (range 1–99) days (14) since this study only included patients admitted in the intensive care unit. Other differences noted in our length of stay in comparison to that seen in other studies is possibly because of the difference in the characteristic of the study participants since gender, age, size, depth of burn, and other burn specific characteristics have been shown to affect length of hospital stay (15).

At multivariate level of analysis, the variables that were independent associated with length of hospital stay and therefore predictors of having a length of stay above median (8 days) were time to medical attention, burn severity, and type of procedure done. Our findings are in agreement with what was reported in Mulago where the degree/ depth of burns (OR = 44.22, 95% CI = 10.86- 180.08, P = 0.000) was the single most significant predictor of length of stay of patients with burns at multivariate analysis level (13).

All the other factors above except for timing of medical attention have been associated with injury severity including the procedure done since it is those patients that have severe burns that are more likely to undergo debridement as opposed to dressing alone (16–18). Therefore it can be explained that the association with length of hospital stay is as a result of the severity. The time to medical attention can be explained by the fact that early medical care has been shown to improve outcome (19). It therefore makes sense that early medical care will reduce length of hospital stay and delay in medical attention, increase length of hospital stay.

Contrally to what was reported in japan where burns from assault were independently associated with fewer hospital-free days, coefficient = − 0.6 [− 1.0 to − 0.1] days; p = 0.009) regardless of the severity of burn injuries (20), we observed no significant association between the circumstances of injury and length of hospital stay. This was possibly because the number of patients who had intentional injury was only 5, which was too small to make meaning full comparisons.

Study Limitations

This study was done in hospitals that had no burn units. Though the findings are representative of most settings in Uganda, but may not be representative of the facilities with burn units.

The length of hospital stay was comparable to that seen in other studies. The predictors of length of stay were time to medical attention, burn severity, and type of procedure done.

Recommendations

Patients with burns should be encouraged to report to hospital early and medical care should be availed as soon as possible since this was the modifiable factor associated with length of hospital stay, and therefore could reduce length of stay even in hospitals with no burn units..

WHO=World health organization, HRRH=Hoima regional referral hospital, FRRH=Fortportal regional referral hospital, KIU-TH=Kampala international university teaching hospital

Ethics approval and consent to participate: All methods were carried out in accordance with relevant guidelines and regulations. Ethical approval was obtained from the Research and Ethics Committee of Kampala International University Western Campus REC number KIU-2022-147. Informed consent was obtained from all subjects and their legal guardian(s).

Consent for publication: Not applicable.

Availability of data and materials: Data is available upon request. Requests should be sent to [email protected] (SMB)

Competing interests: The authors declare that they have no conflict of interest

Funding: This study did not receive any specific grant from funding agencies in public, commercial, or not for profit sectors.

Authors’ contributions: SMB was the principle investigator, conceived and designed the study, collected data, analysed data and wrote the draft of the manuscript. JM participated in data analysis, discussion of results and revised the manuscript. IE, GVZ, SFM and LK participated in discussion of results and revised the manuscript. AK and XFO supervised the work, revised the manuscript and all authors approved the final paper.

Acknowledgements: We acknowledge all patients that accepted to participate in this study.

Guarantor: SMB

Haagsma JA, Graetz N, Bolliger I, Naghavi M, Higashi H, Mullany EC, et al. The global burden of injury: Incidence, mortality, disability-adjusted life years and time trends from the global burden of disease study 2013. Inj Prev. 2016;22(1):3–18.
Price K, Lee KC, Woolley KE, Falk H, Peck M, Lilford R et al. Burn injury prevention in low-and middle-income countries: Scoping systematic review.Burn Trauma. 2021;9.
World Health Organization. Burns [Internet]. Fact sheet. 2018 [cited 2023 Feb 26]. p. 1. Available from: https://www.who.int/newa-room/fact-sheets/detail/burns
Setoodehzadeh F, Ansari H, Zarei T, Surodi Z, Arbabi Z. Factors affecting mortality in burn patients in the poorest region of Iran; a case control study. 2019;9(3):66–72.
Mulugeta T, Alemayehu H, Gerema U. Clinical profiles and the outcomes of burn patients admitted to the burn unit of jimma medical center. Clin Cosmet Investig Dermatol. 2021;14:859–66.
Williams FN, Strassle PD, Knowlin L, Napravnik S, van Duin D, Charles A, et al. Sex-Based Differences in Inpatient Burn Mortality. World J Surg. 2019;43(12):3035–43.
Jeschke MG, Baar ME, Choudhry MA, Chung KK, Gibran NS, Logsetty S. Burn injury. Nat Rev Dis Prim [Internet]. Available from: http://dx.doi.org/10.1038/s41572-020-0145-5
Outwater AH, Ismail H, Mgalilwa L, Temu MJ, Mbembati NA. Burns in Tanzania: morbidity and mortality, causes and risk factors. a review. 2013;3(1):18–29.
Kitara D, Aloyo J, Obol J, Anywar DA. Epidemiology of burn injuries: A basis for prevention in a post-conflict, Gulu, northern Uganda: A cross- sectional descriptive study design. undefined. 2011
Albutt K, Tungotyo M, Drevin G, Ngonzi J, Firth P, Nehra D. 83.02 Burns: Epidemiology, Regional Referral Hospital In Treatment, and Outcomes at a Uganda. Acad Surg Congr. 2018;1(1):1–2.
Knowlin L, Stanford L, Moore D, Cairns B, Charles A. The measured effect magnitude of co-morbidities on burn injury mortality. Burns. 2016;42(7):1433–8.
Ghaed Chukamei Z, Mobayen M, Bagheri Toolaroud P, Ghalandari M, Delavari S. The length of stay and cost of burn patients and the affecting factors. Int J Burns Trauma [Internet]. 2021;11(5):397–405. Available from: http://www.ncbi.nlm.nih.gov/pubmed/34858720%0Ahttp://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=PMC8610821
Odoch R, Atuhairwe C, Amogin D. Predictors of Length of Hospital Stay among Burns Patients in Mulago National Referral Hospital, Kampala- Uganda. J Heal Med Nurs. 2016;29:32–9.
Amouzou KS, Kouevi-Koko TE, Egbohou P, Tchetike P, Hounmenou AKF. Epidemiology and outcomes of in-hospital acute burn-patients in Togo: A retrospective analysis. Burn Open [Internet]. 2019;3(4):141–6. Available from: https://doi.org/10.1016/j.burnso.2019.07.002
Vuola J. 18th European Burns Association Congress. 2020;3–190.
ABA. Advanced Burn Life Support Provider Manual. Am Burn Asoc. 2011;20(312):25–7.
Hahn B, Roh SA, Price C, Fu W, Dibello J, Berwald N, et al. Estimates of Total Burned Surface Area by Emergency Department Clinicians and Burn Specialists. Cureus. 2020;12(7):e9362.
Phillip L, Rice DPO. Assessment and classification of burn injury - UpToDate [Internet]. Up todate. 2020. p. 1–11. Available from: https://www.uptodate.com/contents/assessment-and-classification-of-burn-injury#H2%0Ahttps://www.uptodate.com/contents/assessment-and-classification-of-burn-injury%0Ahttps://www-uptodate-com.ezproxy.utas.edu.au/contents/assessment-and-classification-of-bur
Alyahya T, Ali Al Jabr F, Hussain Alrashid A, Khalifah Almulhim F, Tayi Almutairi MM, Abdulwahab Buhalim R et al. First Aid for Burns and Burn-Related Nutrition among 2437 Inhabitants: A Nationwide Survey in Saudi Arabia.World J Plast Surg. 2020;0–0.
Yamamoto R, Toyosaki M, Kurihara T, Sasaki J. Length of hospital stay and mortality associated with burns from assault: a retrospective study with inverse probability weighting analysis. Burn Trauma. 2020;8:1–8.

No competing interests reported.

Samplimentaryfile1Bivariateanalysis114400.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

Length of hospital stay and its predictors among burn patients in a resource limited setting; a multicenter prospective cohort

Status:

Version 1

Abstract

Introduction

Methods

Results

Conclusion

Background

Study Methods

Results

Length of stay among burn patients.

Predictors of length of hospital stay among burn patients

Discussion

Study Limitations

Conclusion

Recommendations

Abbreviations

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1