Article Text
PDF
Abstract
Background Lack of routine follow-up for trauma patients after hospital discharge likely contributes to high rates of injury-related complications in Cameroon. Mobile phone contact may facilitate timely follow-up and reduce disability for high-risk patients. A previous single-center study showed promising feasibility of mobile health (mHealth) triage, but generalizability remains unknown. We evaluated the feasibility and acceptability of implementing a postdischarge mHealth triage tool at four hospitals in Cameroon.
Methods Trauma patients from four Cameroonian hospitals were contacted at 2 weeks, 1, 3, and 6 months postdischarge. Program feasibility was assessed by calculating the proportion of successful contacts and overall cost. Odds of successful contact were compared using generalized estimating equations across patient socioeconomic status. Acceptability was assessed using a structured patient survey at 2 weeks and 6 months postdischarge.
Results Of 3896 trauma patients, 59% were successfully contacted at 2 weeks postdischarge. Of these, 87% (1370/1587), 86% (1139/1330), and 90% (967/1069) were successfully reached at the 1-month, 3-month, and 6-month timepoints, respectively. The median cost per patient contact was US$3.17 (IQR 2.29–4.29). Higher socioeconomic status was independently associated with successful contact; rural poor patients were the least likely to be reached (adjusted OR 0.11; 95% CI 0.04 to 0.35). Almost all surveyed patients reported phone-based triage to be an acceptable follow-up method.
Conclusion Telephone contact is a feasible and acceptable means to triage postdischarge trauma patients in Cameroon. While scaling an mHealth follow-up program has considerable potential to decrease injury morbidity in this setting, further research is needed to optimize inclusion of socioeconomically marginalized groups.
Level of evidence Level III, prospective observational study.
- global surgery
- Africa
- public health
- wounds and injuries
Data availability statement
Data are available on reasonable request.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
Despite high injury mortality and morbidity rates in Cameroon, routine follow-up after hospital discharge is not currently feasible due to resource and provider constraints.
Early identification of patients who would benefit from further care may facilitate prompt treatment for high-risk patients and mitigate complications, injury-related disability, and death.
WHAT THIS STUDY ADDS
Telephone-based triage is an acceptable and feasible follow-up method among postdischarge trauma patients in Cameroon.
Higher socioeconomic status was independently associated with successful contact with rural poor patients being the least likely to be reached.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
As participation is critical to optimize any programmatic intervention’s impact, these results are promising for broader implementation of triage tool.
Inclusion of socioeconomically marginalized groups must be prioritized in future research.
Background
WHO estimates that 90% of injury-related deaths occur in low-income and middle-income countries (LMICs).1 Moreover, injury-related death rates in LMICs have been rising steadily with increasing industrialization.2–4 In Cameroon, an LMIC in central Africa, injury mortality rates are greater than deaths attributable to malaria and tuberculosis combined.5 Since 2015, a multisite Cameroon National Trauma Registry (CTR) has collected ongoing prospective data on injury to identify targets for trauma process improvements.6 7 However, routine follow-up for trauma patients is not currently feasible in Cameroon due to resource and provider constraints. Consequently, most trauma patients return to hospital care only once complications develop.6 8 Early identification of patients who would benefit from further care may facilitate prompt treatment for high-risk patients and mitigate complications, injury-related disability, and death.
Demographic and health survey and cross-sectional injury surveillance data demonstrate high penetrance of cellular telephone access among Cameroonians (95% in the Southwest Region of Cameroon).9 10 Widespread utilization of mobile telephones makes telephone contact a potentially high-impact target for increasing access to follow-up care in a resource-constrained trauma setting where universal follow-up is not feasible. If successful, telephone triage could help to direct the limited available resources toward the highest risk patients, preventing unnecessary morbidity due to injury. However, implementation of mobile health (mHealth) follow-up programs critically depend on patient participation. A single-center pilot of an mHealth follow-up triage tool demonstrated promising feasibility of an mHealth follow-up triage program, but the generalizability of these results has not yet been established.11 In particular, as cellphone ownership is known to be a marker of socioeconomic status among Cameroonians, program feasibility may vary by socioeconomic group.12 In order to support broader implementation and scaling efforts, rigorous evaluation of the feasibility and acceptability of mHealth follow-up contact is needed. The objective of this study was to determine the feasibility and acceptability of cellular phone-based triage as part of a multisite validation trial. To develop strategies to optimize future implementation, we characterized associations between socioeconomic status, injury characteristics, and feasibility of cellular telephone contact.
Methods
Design, setting, and population
We evaluated implementation of a mobile telephone follow-up program in a cohort of Cameroonian trauma patients hospitalized for injury at four study sites between June 2019 and December 2021. The previously described CTR collects ongoing prospective data on all injured patients admitted for injury at Limbe Regional Hospital, Laquintinie Hospital of Douala, Edea Regional Hospital, and Catholic Hospital Pouma. All patients enrolled in the CTR who were discharged after hospital admission for injury were eligible for inclusion in the mobile follow-up program.
All participants were contacted postdischarge via mobile phone by research assistants at four timepoints: 2 weeks, 1 month, 3 months, and 6 months. CTR patients who were unable to provide a cellular phone number or those with missing date of death or availability at 2 weeks were excluded.
At each timepoint, participants were contacted at least three times by calling and once by text message over a 1-week interval. At the 2-week and 6-month timepoints, a subset of participants was also asked to complete an acceptability survey. This 5 minute survey included six yes/no questions assessing perceived convenience, acceptability, and utility of mobile telephone follow-up.
Data were documented on paper forms and entered into Research Electronic Data Capture by research assistants hosted on University of California Los Angeles server.13
Variables
The primary outcome of feasibility was the proportion of enrolled patients successfully contacted at each timepoint, defined as answering the phone and being available on at least one of three attempts. Secondary feasibility outcomes of interest included number of call attempts, cost per contact, and patient-reported acceptability.
CTR data were extracted for enrolled patients and tested for associations with telephone contact feasibility. Variables extracted included sex, age, injury characteristics, household demographics, occupation, education level, alcohol use, and discharge status. Patients were categorized into socioeconomic clusters using the previously validated Cameroon EconomicCluster model. Specifically, patient household demographic information across five dimensions (urban or rural location, home ownership status, cellphone ownership, agricultural land ownership, and type of cooking fuel used) was used to identify each patient as belonging to rural poor, rural wealthy, urban middle-class homeowners, urban middle-class tenants, or urban wealthy household socio-economic status (SES) clusters.12 14
Resource utilization was defined as personnel time and cost per patient contact. Research assistants documented time to administer survey per patient. Program costs for the study period were retrieved from each hospital, which accounted for personnel, patient travel and call time reimbursement, and research infrastructure. Cost per patient contact were calculated by dividing the total cost by the number of patients per hospital over the study duration. Acceptability survey responses were summarized as proportions of enrolled patients. The Strengthening the Reporting of Observational Studies in Epidemiology guidelines was used to ensure proper reporting of methods, results, and discussion (online supplemental file 1).
Supplemental material
Statistical analysis
Categorical variables are reported as frequencies and percentages of enrolled patients. Continuous variables are reported as means with SD for normally distributed variable or medians with IQR for non-parametric variables. Differences between groups were tested with χ2 test for categorical variables and Wald test for continuous variables.
Since each participant was contacted at multiple timepoints, we assessed risk-adjusted feasibility using a repeated-measures analysis. We used a generalized estimating equation (GEE) model with binary distribution and logit link function to calculate ORs with 95% CIs for correlated observations.15 16 For all analyses, we set α at 0.05. Data were analyzed using Stata (V.16.1, StataCorp, College Station, Texas, USA).17
Results
Of 4511 CTR patients admitted during the study period and consented for cellular telephone contact, 3896 were eligible for postdischarge follow-up (figure 1). The remaining 615 patients were excluded for in-hospital mortality (n=247, 5.5%), missing date of death (n=82, 1.8%), or missing availability at the first timepoint (n=286, 6.3%). The telephone follow-up cohort was representative of a standard young, male predominant trauma population. Patient demographics, injury characteristics, and clinical course of the telephone follow-up cohort are reported in table 1. Of note, only 9.4% of participants lived in a rural setting and 19.8% of patients reported being unable to receive recommended medical care due to prohibitive healthcare costs (table 1). Among the cohort, 81.0% had access to cellphones while 1.9% did not and the remaining 17.1% had an unknown status.
Consolidated Standards of Reporting Trials diagram for mobile health (mHealth) follow-up cohort. CTR, Cameroon National Trauma Registry.
Demographics, injury characteristics, and hospital course for patients enrolled in telephone follow-up program
Feasibility
Overall, 2304 (59%) of 3896 enrolled patients were successfully contacted at 2 weeks postdischarge. Of those successfully contacted at subsequent 1-month, 3-month, and 6-month timepoints, 87% (1370/1587), 86% (1139/1330), and 90% (967/1069) of patients were successfully reached, respectively. Among the 1592 patients who were not reached at the 2-week timepoint, only 26 (1.6%) were successfully contacted at subsequent timepoints.
At each timepoint, it took a median of 1 call attempt (IQR 1–1) to successfully contact participants. Patients who were successfully contacted at the 2-week timepoint were more frequently female (58.8% vs 41.2%; p=0.007), had completed more years of education, and were less commonly members of the rural poor SES group (table 2; all p<0.05). Rates of successful contact did not differ by age or injury location (table 2).
Frequency of successful contact at each timepoint (2 weeks, 1 month, 3 months and 6 months) by patient demographic and hospital variables
Call times were recorded for 1070 patients who completed the triage survey. Median call time was 57 seconds (IQR 43–92). The median cost per patient contact was US$3.17 (IQR 2.29–4.29). Cost varied by hospital site: median cost per contact was highest at Edea Regional Hospital at US$4.43 (IQR US$3.32–US$5.71) compared with 3.94 (US$2.38–US$5.62), 2.60 (US$1.88–US$3.76), and 1.72 (US$1.58–US$2.08) at Catholic Hospital Pouma, Limbe Regional Hospital, and Laquintinie Hospital of Douala, respectively (p<0.001).
Respondent distribution
At all timepoints the most common survey respondent was the patient themselves, followed by a surrogate. The proportion of patients responding increased overtime from 50.0% at the 2-week timepoint to 58.9% at the 6-month timepoint.
Adjusted association with feasibility
The GEE model revealed SES was independently associated with feasibility, with rural poor patients least likely to be successfully contacted (adjusted OR (AOR) 0.11; 95% CI 0.04 to 0.35) compared with urban wealthy patients. With male as reference, female sex was linked to increased odds of contact (AOR 1.46; 95% CI 1.00 to 2.12). Compared with discharge home, transfer to another hospital was associated with reduced odds of successful contact (AOR 0.55; 95% CI 0.31 to 0.98). Compared with Laquintinie Hospital of Douala, Catholic Hospital Pouma (AOR 2.10; 95% CI 1.09 to 4.04) and Edea Regional Hospital (AOR 5.19; 95% CI 2.53 to 10.66) were associated with increased odds of successfully contacting patients. Limbe Regional Hospital had the inverse relationship (AOR 0.38; 95% CI 0.26 to 0.55).
Acceptability
Among a subset of discharged patients, 1372 were contacted to complete the acceptability survey 2 weeks postdischarge while 570 were contacted after 6 months. Overall, respondents at both 2 weeks and 6 months found telephone contact to be highly acceptable. Approximately 98.6% of participants believed mobile phone to be a convenient way to be contacted with >90% of surveyed patients responding positively to all acceptability questions (table 3). The proportion of patients willing to go to a clinic or hospital if recommended during the phone call increased with time (91.4% at 2 weeks vs 97.5% at 6 months, p<0.01). Similarly, more patients at 6 months reported that they would be willing to send a picture to a healthcare provider to aid in evaluation (93.5% vs 95.6%; p<0.001).
Patient acceptability survey results at T1 (2 weeks) and T2 (6 months) after discharge
Discussion
In this prospective multisite implementation study, we present the first longitudinal evaluation of an mHealth trauma follow-up program in Cameroon. We found mobile phone contact to be both a feasible and highly acceptable mechanism to triage injured patients for follow-up care after hospital discharge, with nearly all surveyed patients reporting that telephone contact was both convenient and acceptable for follow-up. As participation is critical to optimize program impact, these results are promising for broader implementation of such a triage tool.
Importantly, the majority of successfully contacted patients were reached at the earliest call timepoint, suggesting that additional contacts may be of limited utility for triage, with nearly all surveyed patients reporting that telephone contact was both convenient and acceptable. These findings bolster feasibility findings from prior single-site data11 and provide the critical knowledge base needed to support scaling and optimization efforts for broader adaptation.
One key objective of this study was to understand how call response varied by contact point to maximize program feasibility and minimize redundancy. We found feasibility at the 2-week timepoint was consistent with findings from a prior single-center pilot assessment of the telephone triage tool (62% at 2 weeks)11 but notably higher than previously reported estimates for telephone contact of trauma populations at 6 weeks after discharge in other LMICs. Specifically, Ethiopian trauma patients contacted via telephone follow-up at 6 weeks postdischarge were reached only 47% of the time.18 Similarly, a study by Rapp et al reported 51% of patients were compliant with a telephone follow-up 6 weeks after discharge from a urological procedure.19 It is not possible to fully delineate the likely multifactorial causes underlying the higher contact feasibility found in Cameroon, which may include difference in telephone penetrance over time and in different locations. However, our findings suggest that earlier contact postdischarge may be more feasible than later contact which has the secondary benefit that earlier identification of off-trajectory patients is more likely to result in improved outcomes. Ding et al demonstrated that 90% of postdischarge trauma deaths occurred within the first 2 weeks of leaving the hospital,20 suggesting that contacting patients early after injury is critical to try to mitigate preventable morbidity and mortality. Conversely, patients were unlikely to be successfully contacted at subsequent timepoints if they were not reached at the 2-week timepoint (<1%) indicating that there is relatively little utility in pursuing later contact and these timepoints can be removed to improve overall program cost effectiveness.
Importantly, we identified significant differences in triage feasibility based on patient demographics. In both high-income and LMIC settings, patients from rural environments have been found to have worse trauma outcomes than their urban counterparts, making increasing access among rural populations a particular priority.21 Unfortunately, using the current triage protocol we identified reduced feasibility among the rural poor population. As such, there is a critical need to understand limitations in a telephone follow-up tool in different contexts to optimize uptake among marginalized populations. Notably, contact feasibility differed by admission hospital, with smaller regional and private hospitals demonstrating higher odds of successfully contacting patients compared with a larger referral tertiary hospital. With 81% of our cohort reporting cellphone access, future studies to assess mitigation strategies such as lending phones to high-risk participants who do not have access to telephones may be required to ensure equitable follow-up care among trauma patients. Notably, only 10% of the enrolled cohort identified as being from a rural community which may reflect failure to present to the hospital in this population.
The present study has several limitations. First, acceptability surveys were collected using convenience sampling and participants who completed surveys may be more likely to report favorable acceptability compared with those not willing to participate. As described above, patients identifying as rural are under-represented among the contacted cohort compared with the general population of Cameroon. Additionally, this study only targeted patients who already had engaged formal care and does not address access among patients who are not treated in the formal care system and who may be particularly vulnerable. The cost estimates were also based off total site costs, which included research personnel and patient reimbursements for call time and travel and thus maybe an overestimate for telephone triage. Conversely, there may be future costs required for program implementation that are not captured in these data that may limit the accuracy of our findings. Finally, while we present data demonstrating multisite feasibility from Cameroon, given the diversity of LMIC trauma systems, generalizability of our findings to other LMIC clinical settings may be variable.
In conclusion, we found telephone contact to be a feasible and acceptable means to triage postdischarge trauma patients in Cameroon. Contact at 2 weeks after injury appears to be feasible and is early enough to facilitate prompt repatriation into care. While scaling a telephone follow-up program has considerable potential to decrease injury morbidity in this setting, further research is needed to optimize inclusion of socioeconomically marginalized groups.
Data availability statement
Data are available on reasonable request.
Ethics statements
Patient consent for publication
Ethics approval
The study was approved by the institutional review boards of University of California Los Angeles (IRB#19-000086) and University of Buea (2019/868-11/UB/SG/IRB/FHS). All participants provided informed consent before inclusion in the study.
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Presented at American College of Surgeons Clinical Congress in San Diego, California on October 18, 2022.
Contributors SR, RO, MC, AC-M, CJ, and SAC were involved in the conceptualization and design of the study. SR and RO completed data analysis and interpretation under the guidance of CJ and SAC. SR, RO, and MC conducted literature search and writing. MAM, FY, FE, and GEM were involved in data acquisition. All authors critically reviewed the manuscript. SAC is the guarantor and accepts full responsibility for the finished work and/or the conduct of the study, had access to the data, and controlled the decision to publish.
Funding Research reported in this publication was supported by the Fogarty International Center of the National Institutes of Health under Award Number R21TW010956.
Disclaimer The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Competing interests None declared.
Provenance and peer review Not commissioned; internally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.