Capturing the costs of acute hand and wrist injuries: Lessons learnt from a prospective longitudinal burden of injury study

Abstract

Introduction

Given the high incidence of hand and wrist injuries, they are exceptionally costly to the economy. This prospective, longitudinal study aimed to establish methods for capturing the burden of acute hand and wrist injury from an individual and societal perspective.

Methods

A prospective longitudinal design with baseline measures of injury type and severity, and repeated measures of disability, cost, and activity limitations and participation restrictions at six weeks, three months, and six months was selected. Participants were recruited from two large urban Australian public health care services. We sought to establish methods for capturing the burden of acute hand and wrist injury from an individual and societal perspective and compare survey completion by the method of administration.

Results

A total of 206 patients consented to participate in this study, representing 54% of those invited to participate. The survey completion rates were 18% at six weeks, 2.4% at twelve weeks, and 0.004% at six months following injury. From the limited data collected at six weeks, it was noted that nearly half of the patients reported a decrease in usual financial income, 14% reported absenteeism, and 62% reported presenteeism. Participants who elected to have data collected via phone call had the highest survey completion rate (n = 6/10; 30%) at six-week’s follow-up.

Discussion

The study findings highlight the difficulties of completing longitudinal survey research investigating individual and societal burden with this population. Future research should be carefully designed to encourage participation and retention by considering patient and public involvement in study design, the time burden placed on the participants within and across selected survey time points, providing participants with incentives to participate, and highlighting the relevance and real-world applications of the findings.

Keywords

Hand injury wrist injury cost-of-illness health care cost health expenditure

Introduction

Hands and wrists were the third most commonly injured body part in US workplace injury data,¹ and the largest category of work-related injuries in Australia.² Given their high incidence, these injuries are astonishingly costly to the economy. In the US, cost of injury data collected via the Centers for Disease Control and Prevention WISQARS™ (Web-based Injury Statistics Query and Reporting System) indicated that wrist and hand fractures alone cost the economy more than six billion US dollars in 2010, with work loss costs comprising 78% of total costs.³ Similarly, a Dutch population-based study found that, given the volume of injuries sustained annually, hand and wrist injuries ranked first in the order of most expensive injury group, ahead of knee, hip, and skull–brain injuries.⁴ This was due to impacts on work productivity, especially in males aged between 20 and 65, and health care costs in women aged over 65.

In Australia, evidence on the scale of the economic burden of acute hand and wrist injuries at an individual, community, or societal level is currently limited to work-related injuries.⁵ Therefore, a poor understanding remains of how the specific factors associated with these injuries (such as poorly managed pain, lack of social or workplace support, late presentation to specialist facilities, misdiagnosis by primary health care providers, and sub-optimal treatment and rehabilitation^6–8) and the systems in place outside of the workers’ compensation system to manage these, contribute to or mitigate this burden.

The economic burden of a disease or injury is described as the sum of all costs associated with the condition that would not be incurred if that disease did not exist^9,10 and is calculated by totaling direct, indirect, and intangible costs.^11–14 Direct medical costs relate to diagnostics and the actual treatment provided in response to the injury and can include surgery, inpatient admission, medications, imaging, and postoperative care.^13,14 Direct nonmedical costs are costs and resources used in connection with the health service but are not health sector costs (e.g. transport to and from a medical facility). Indirect costs most commonly relate to productivity losses due to morbidity and mortality and can be borne by the individual, family, society, or the employer.^10,13,14

These costs are due to work absences resulting in foregone productivity (referred to as absenteeism), reduced work capacity due to impairment related to their condition (presenteeism), and unpaid productivity, which involves reduced possibilities of performing usual activities at home such as housework or caring for family members due to illness or disease.^9,10,15–17 Indirect costs are often harder to calculate than direct costs as it is difficult to measure productivity when considering presenteeism or unpaid roles objectively or with certainty. Intangible costs consider burden beyond the monetary costs of goods and services and include other sequelae that reflect decreased enjoyment of life because of illness. Such costs are associated with functional limitations, pain, psychological distress, and decreased social interaction.¹⁸

Information about cost burden and the resources used in the design, implementation, and provision of health care,¹⁹ is of interest to policymakers for resource allocation and cost-minimization. However, such an analysis has not been completed for hand and wrist injuries in Australia. This study focuses on individuals presenting with acute fractures, tendon and nerve injuries of the hand and wrist at two Australian public health services, to provide insight into the direct and intangible costs experienced from an individual patient perspective and the indirect costs experienced from an individual patient and societal perspective. In addition, it seeks to identify the specific drivers of individual and societal economic burden.

This prospective, longitudinal study sought to establish methods for capturing the burden of acute hand and wrist injury from an individual and societal perspective.

The study objectives were to:

evaluate methods for collecting data regarding direct and intangible costs experienced at an individual patient perspective in an acute hand/wrist injury population in Australia;

report recruitment and retention rates for all eligible participants;

compare the rate of survey completion by mode of administration;

provide a preliminary estimate of profile of costs incurred by injury type: either fractures; tendon or nerve injuries; and

identify the specific key drivers of economic and individual, family, and social burden.

It was hypothesized that impacts on participation are more pronounced and prolonged for people with tendon injuries than those with fractures or nerve injuries, and that burden of disease is, therefore, higher for this group. This is because tendon injuries typically heal at a slower rate than bones²⁰ and require a longer period of immobilization and protection preventing functional use of the hand.¹⁴

Methods

Design

A longitudinal cohort design with baseline measures of injury type and severity, and repeated measures of disability, cost, and activity limitations and participation restrictions at six weeks, three months, and six months was selected to test study feasibility and allow for a comparison of total burden estimates between three common acute hand or wrist injury groups (fractures, tendon, and nerve injuries).

Target population and subgroups

As we sought to compare the individual and societal costs, level of disability, and the activity limitations and participation restrictions between hand and wrist fractures, tendon and nerve injuries, three subgroups were targeted during participant recruitment. These categories were chosen as they are amongst the most common hand and wrist injury presentations to Australian public hospitals^21,22 and are likely to result in impacts on work, family, and social participation for between six and twelve weeks.

Patients from two major metropolitan hospitals in Melbourne, Australia who met the inclusion criteria of: (i) a clinical diagnosis of an acute hand or wrist fracture, nerve, or tendon injury; (ii) of working age (18–65 years); (iii) able to participate in completing a survey (no serious mental health, cognitive or linguistic impairment that would impact on participation); (iv) not experiencing other serious disorders or injuries that might confound the experience of their acute hand or wrist injury; and (v) identified by their treating therapist during their first outpatient specialist appointment within the study settings.

An invitation to discuss information relating to the study was offered to eligible participants by the treating hand therapist. Written consent was obtained after a full explanation of the nature and scope of the study by the first author (LSR) prior to the commencement of data collection. Data collection occurred between July 2016 and December 2017. Ethical approval was granted by Alfred Health (422/13), Monash Health (LNR/16/MonH/18), and Monash University (CF14/197).

Setting and location

The Alfred Hospital (a campus of Alfred Health) is a 680-bed tertiary referral teaching hospital that is a major provider of specialist state-wide services (e.g. burns, trauma) to residents of Victoria. The Emergency Department (ED) has about 65,000 visits per year.²³ The Dandenong Hospital (a campus of Monash Health) is a 573-bed tertiary referral teaching hospital that is a provider of general and specialist services to the people of Dandenong and surrounding areas. The ED has about 70,000 visits per year.²⁴

Study perspective

We used a prospective, incidence-based approach to costing illness, where we estimated the lifetime costs of a condition from its onset until its disappearance to quantify the economic burden over a determined period.^10,25 Direct and intangible costs were estimated from the perspective of the individual patients, while indirect costs due to lost productivity were estimated from the individual patient and societal perspective.⁹

Data collection

Participants were offered the opportunity to complete online (via email or text-message link), paper-based or telephone surveys at six weeks, three months, and six months following injury. The surveys consisted of:

a demographic form that was specifically designed for this study which included age, gender, marital status, education, employment status prior to injury, current employment status (hours worked in previous week, ANZCO code for type of work performed²⁶) number of children, number of persons dependent on family income, compensation status, and job stability measures (number of full-time jobs in past five years, total years at current job, and total years with current employer);

an injury description form that was specifically designed for this study which included the location of the injury, type of injury, the time from injury to presentation at hospital, and injury mechanism;

the Patient-Rated Wrist and Hand Evaluation (PRWHE),²⁷ which contains 15 items: five of which evaluate pain (intensity and frequency) and 10 evaluate function (specific activities and usual activities). The pain subscale is calculated by summing the five items, while the functional score is calculated by the sum of the ten items divided by two. A score of 100 is reflective of a significant impact, whereas a score of zero reflective of no impact. Information from the PRWHE can be used to determine the magnitude of wrist or hand-related disability at one point in time;

the SF-36, short form of the Health Status measure, which is a comprehensive multidimensional measurement of health status concepts. The scales include a physical functioning scale, two scales that distinguish between role limitations because of “physical health” or “mental health”, a social functioning scale, a mental health scale, and a vitality and general health perception scale;²⁸ and

the Short Form of Health and Labour Questionnaire (SF-H&L), which has three modules covering absence from paid employment, productivity loss without absence from paid employment, and impediments to paid or unpaid employment.²⁹

All instruments were selected based on validity and reliability, sensitivity to detect changes over time, appropriate normative data, time and literacy levels required for completion, self-report, and relevance to workplace productivity.^30–34 Participants who requested to complete their survey via phone call or online (via a text message or email link) were sent two reminders to complete their survey, while a second survey was mailed to participants in the event where they did not return their survey within a week.

Data analysis

All data were checked for normality and, if skewed, presented as a Median (Mdn) and Interquartile Range [IQR]. Descriptive statistics were used to report demographic data, injury types, direct health care costs (health service use and out of pocket expenses) productivity costs (time off work, as well as time in alternative duties) indirect costs (such as paying others to perform roles/duties usually done by the injured person) and impacts on family and social participation. Multiple linear regression modeling was proposed for prediction of costs and duration of disability/productivity impairment, and latent growth curve analysis was planned to see if changes in one variable result in changes to others over the first six months post-injury. Cost data are presented in Australian dollars (AUD). Data analysis was performed using R studio³⁵ and Microsoft Excel.³⁶

Results

A total of 382 participants were invited to participate in this study with 206 giving written consent to participate (54%) over the 18-month data collection period. Of this group, only 37 participants (18%) completed their six-week survey, and only five participants (2.4%) completed their twelve-week survey, and one participant (0.004%) completed their six-month survey (refer Figure 1). The surveys that were completed at twelve weeks and six months were further complicated by a large portion of missing data in many of the sections meaning comparisons between time points was not feasible.

Figure 1.

Participant recruitment, retention, and survey completion.

This was particularly evident with questions that allow for the valuation of productivity losses in the SF-H&L questionnaire relating to weekly income and household earnings (89% and 73% of responses missing or refused at the six-week data collection respectively). Furthermore, of the five participants who completed their twelve-week’s survey, four did not answer at least one item of the SF-36 subsection that relates to energy and emotion, three did not complete the SF-H&L questionnaire, and two did not answer at least one item on the PRWHE.

The majority of participants who consented to participate in the study requested to complete their surveys online (82%) with a link sent via email (n = 137) or text message (n = 32). The response rate for each mode of survey administration varied with phone calls (n = 6/10; 30%) having highest completion rate for the six-week survey followed by online via email link (n = 25/137; 18.2%), online via text-message link (n = 5/32; 15.5%), and mail (n = 2/27; 7.4%). Completion rate for the twelve-week survey was highest for online access via email link (n = 4/5; 80%).

The demographic and employment details of the initial sample and the 37 participants who completed the six-week’s post-injury survey can be viewed in Tables 1 and 2. The median number of specialist medical and hand therapy appointments attended (as reported by self-recall by the respondents) at six weeks following injury was four [IQR 3–5] and three [IQR 2–4], respectively.

Table 1.

Demographics of participants at entry and on completion of six-week measures.

		Participants who consented to participate		Participants who completed six-week survey
		n		n
		206		37
		n	%	n	%
Gender	Male	159	77%	25	68%
	Female	41	20%	12	32%
	Not stated	6	3%
Age	Median years [IQR]	30	[25–38.25]	32	[26–60]
Location of injury	Finger	80	39%	22	59%
	Hand	103	50%	10	27%
	Wrist	11	5%	5	14%
	Missing	12	6%
Type of injury	Fracture	177	86%	33	89%
	Tendon	11	5%	3	8%
	Nerve	2	1%	1	3%
	Missing	11	5%
Side of injured hand	Left	80	39%	16	43%
	Right	114	55%	20	54%
	Missing	12	6%	1	3%
Injured dominant hand	Yes			21	57%
Number of days between injury and presentation to ED	Median days [IQR]			1	[0–2]
Mechanism	Bite			1	3%
	Crush			4	11%
	Fall / FOOSH			6	16%
	Laceration (knife, blade)			2	5%
	Not stated			2	5%
	Power tool			1	3%
	Punch / altercation			6	16%
	Sport (Inc. bicycle)			15	41%
Marital status	Divorced			1	3%
	In a relationship			1	3%
	Married / long-term cohabitation			23	62%
	Relationship			1	3%
	Single			11	30%
Education	Bachelor’s degree			5	14%
	Certificate III/IV			3	8%
	Diploma / Advanced Diploma			11	30%
	Graduate Diploma / Graduate Certificate			2	5%
	Post graduate degree			5	14%
	Year 10			1	3%
	Year 11 or below (Inc. Cert I/II)			3	8%
	Year 12 / VCE or equivalent			7	19%
Children	Participants with children			14	38%
	Mean number of children (SD)			3	1.3
Number of dependents on income	No dependents			15	41%
	One dependent			12	32%
	Two dependents			3	8%
	Three or more dependents			7	19%

IQR: interquartile range; FOOSH: fall on outstretched hand; Inc.: including; SD: standard deviation; TAC: transport accident commission.

Table 2.

Employment demographics of participants who completed six-week measures (n = 37).

Employment status (prior to injury)	Employed, casual	6	16%
	Employed, full-time	25	68%
	Employed, part-time	3	8%
	Not in labor force	2	5%
	Unemployed, looking for full time work	1	3%
Employment status (six-weeks post injury)	Employed, casual	6	16%
	Employed, full-time	21	57%
	Employed, part-time	1	3%
	Employed, temporarily away from work	5	14%
	Not in labor force	2	5%
	Unemployed, looking for full time work	2	5%
Occupation	Clerical and administrative workers	3	8%
	Community and personal service workers	2	5%
	Managers	9	24%
	Not in the labor force	4	11%
	Professionals	10	27%
	Sales workers	3	8%
	Technicians and trade workers	6	16%
Number of full-time positions in the last five years	No full-time position	6	16%
	One full-time position	15	41%
	Two full-time positions	10	27%
	Three full-time positions	4	11%
	More than four full-time positions	2	5%
Number of years with current employer (n = 33)	Median [IQR]	2	[1–7.75]

IQR: interquartile range.

Despite the majority of patients identifying as being covered by either universal health care (73%) or insurance-based compensation systems (8%), both of which cover medical and therapy costs, all participants reported out-of-pocket medical costs associated with their injury at a median cost of $75 [IQR $40–$200]. Prior to their injury, 92% of participants reported that they were working either full-time, part-time or casually (see Table 2). At six weeks following injury, 14% of participants who were working prior to their injury reported they were forced to take a temporary absence from work, with one participant losing their employment role held before injury. Of the participants who reported loss of income from lost penalty rates, commissions or other forms of additional income (n = 17) the median financial deficit six weeks following injury was $1000 [IQR $100–$2500]. Also, nearly half of participants (n = 14) reported needing to accommodate for expenditure on tasks that were usually performed independently before injury (e.g. home maintenance, lawn mowing) at a median cost of $150 [IQR $100–$275] when this occurred.

Patient-rated wrist and hand evaluation scores

At six weeks, the median PRWHE score for all participants was 39.5 [IQR 25.5–53.5]. Although not specified in the assessment manual, this could be interpreted as having a mild to moderate impact as a score of 100 represents maximum impact of pain or functional disability while a score of zero indicates no impact.²⁷ The median scores of the pain and function subscales (both out of a possible score of 50) were 15 [IQR 10–26] and 21 [IQR 14–29.5], respectively. No comparison between six weeks, twelve weeks, and six months was possible owning to missing responses.

SF-36 scores

The mean scores of the subscales of the SF-36 following the scoring guidelines proposed by Ware and Sherbourne²⁸ are presented in Table 3. It was observed that role limitations due to physical problems (mean: 16.2; SD 28.4) was the most noted impact on health status, which when compared to the norm for healthy Australians (both genders) (mean: 79.7; SD 35.1), indicates that our sample observed significant impact on the ability to complete their normal roles.³⁷ The mean summary measures of physical health (physical components summary) and mental health (mental component summary), calculated as proposed by Ware and Gandek,³⁴ were 62.1 (SD 10.6) and 57.2 (SD 16.8), respectively. This indicates that these injuries appear to have had a slightly greater impact on mental health than physical health for those who completed the survey.

Table 3.

Mean scores and standard deviations of SF-36 subscales.

Subscale	Items	Mean	SD
Physical functioning	10	77.7	10.9
Role functioning/physical	4	16.2	28.4
Role functioning/emotional	3	41.4	40.3
Energy/fatigue	4	53.1	17.9
Emotional well-being	5	62.9	19.9
Social functioning	2	74.8	20.2
Pain	2	62.7	22.9
General health	5	67.4	18.8

SD: standard deviation.

A higher score indicates a better health state.

Short form of health and labor questionnaire findings

Responses to the SF-H&L questionnaire revealed that of participants who identified as remaining in a paid employment role six weeks following injury (n = 32), six participants were unable to perform any work in the two-week period before completing the survey. Of the participants who were able to work in the two-week period prior to completing the survey (n = 26), 62% reported that their work was hindered as a direct result of their hand or wrist injury. The most common experiences were needing to work at a slower pace (n = 8 often/always), putting off some of their regular work tasks (n = 6 often/always), and needing other people to complete routine work tasks (n = 6 often/always). Details relating to annual or weekly income were largely absent in the completed survey responses indicating a preference of participants not to disclose such information.

All participants who reported that they were unable to work (n = 6) did not provide income data, and therefore, an estimate of indirect costs was not feasible. The median number of hours spent in a one-week period on household work (e.g. preparing meals, cleaning the house, washing clothes), shopping (e.g. daily groceries), odd jobs (e.g. house repairs, gardening), specific activities for or with own children (e.g. providing care, providing transport), and providing care for a person other than a child (e.g. spouse, parent, or other) and frequency of reported impact is presented in Table 4. The frequency of participants seeking assistance with household work, the source of assistance and the median number of weekly hours of assistance are presented in Table 5.

Table 4.

Hours spent in weekly tasks and reported impact from injury (Short Form of Health and Labour Questionnaire) at six weeks.

	No. of responses	Median hours (IQR) spent in previous week (no. of responses reporting time spent)	% reported impact from injury
Household work	37	10 (IQR 4–20) (37)	70
Going shopping	36	2.5 (IQR 2–45) (36)	56
Odd jobs	37	2 (IQR 0–2) (37)	65
Specific activities with children	37	20 (IQR 10–35.25) (10)	30
Carer role	37	5 (IQR 2.25–8) (11)	24

IQR: interquartile range.

Table 5.

Source and hours of assistance in household work (Short Form of Health and Labor Questionnaire) at six weeks.

	No. of responses who reported assistance	Median hours (IQR) of assistance
Family	16	6.5 (IQR 2–10.25)
Neighbor/volunteer	1	18 (IQR 18–18)
Paid home help	4	2.75 (IQR 1.59–12.62)
Other	0

IQR: interquartile range.

Discussion

In this study, we aimed to establish methods for capturing the burden of acute hand and wrist injury from an individual and societal perspective as well as providing data on cost of injury. As we achieved a less than desirable retention and survey completion in those who had consented to participate in this study, the reported cost burden experienced by individuals presented within this investigation should be considered as preliminary data. This was further confounded by missing responses on completed surveys.

The findings do, however, suggest that the individual burden experienced from these injuries could be substantial and has the potential to impact the individual in terms of loss of income (as a result of lost penalty rates, overtime, and commissions), out-of-pocket expenses (including medical costs despite universal health care and insurance-based compensation systems), and impaired role performance, health and well-being. Based on these limited findings, the role of a hand therapist in providing occupation-based interventions that consider the unique individual’s roles and responsibilities^38,39 undertaken in the workplace, community and family home, could be viewed as one method that may reduce the individual burden experienced. However, a well-designed economic evaluation that considers the cost-benefit is required to validate this proposal.

The findings of PRWHE total scores in this investigation at six-weeks (mean 41.7; SD 20.5) are comparable to a previously published study where the PRWHE was administered for various hand and wrist pathologies at the same timepoint (mean 43.28; SD 23.61).⁴⁰ This suggests that the burden experienced by the limited participants in this study six weeks following their injury is common and that this could be an indication of the cost burden that may be experiencing. A large-scale study that investigates the correlations between PRWHE scores and cost burden experiences is required to test this hypothesis.

We also found that these injuries have the potential to lead to significant negative impacts and productivity costs for employers at a societal level as the majority of respondents who were at work four to six weeks after their injury were not performing tasks at their normal capacity as the result of their injury. This supports the notion that these injuries contribute to societal burden via absenteeism (i.e. absence from the work setting) and presenteeism (i.e. when the individual is present at the work setting, but performance is likely impacted owing to a health condition⁴¹). However, the true scale of this cost burden in Australia and many other countries remains largely unknown owing to a paucity in published literature.

Although this investigation provides some insight into the individual and societal cost burden experienced at six weeks following a hand or wrist injury by a small group of participants, there are several limitations. First, data collection was restricted to two Australian public health care services in Victoria meaning it is possible that the recruitment, retention, and cost burden observed in this study may not be generalizable to the broader Australian population.

While we demonstrated an acceptable level of recruitment of those invited to participate, retention of those who had consented in the days following injury and completion rates of the six-week, the twelve-week, and six-month post-injury surveys were extremely low. This outcome was observed despite providing recruited participants with a choice of how they would like to complete the surveys (e.g. online, paper-based, telephone) and method of reminder notices (e.g. email, text messages, telephone messages).

Interestingly, we found that administering the survey via a phone call had the highest successful completion rate (80%) but the lowest requested administration mode (n = 10). While it is impossible to explain the exact reasons why this occurred in this study, previous consumer survey research has shown that telephone versions of a survey tend to have a higher response rate than online surveys and they tend to be more representative of a sample even following the weighting of demographic variables.⁴²

A further limitation is that we recruited participants in the days following their injury, generally during their first outpatient appointment. It is possible that recruiting participants closer to the first survey time point (i.e. six weeks after injury) or contact between data collection time points may have led to a higher retention rate due to the recency of discussing the research project, its aims and its real real-world applications.⁴³ Alternatively, arranging a data collection appointment to coincide with medical or therapy appointments may encourage a higher level of study participation and retention.⁴⁴

Increasingly lower participation rates in epidemiological studies have been observed over the last 40 years,⁴⁵ with even steeper declines observed in more recent years.⁴⁶ Such observations have been noted and documented by government agencies,⁴⁷ academic researchers,⁴⁸ and for-profit companies.⁴⁹ Commonly reported reasons for this decline in participation include the “over surveyed” effect, where there is increasing number of requests to participate in various research and consumer surveys;⁴⁵ the decrease in societal volunteerism, which has been found to parallel participation in survey research;⁵⁰ the growing disillusionment with science and scientific studies;⁵¹ and the increasingly complex nature and demands of research studies and their perceived time burden.⁴⁵ It is, therefore, plausible to suggest that some or all of these factors could have contributed to the results observed.

When considering why people choose to participate in survey research, three main factors have been suggested. These being altruism (e.g. furthers some purpose important to the respondent or fulfilling a social obligation), survey-related reasons (e.g. respondent is interested in the survey topic or find the interviewer appealing), and egoistic reasons (e.g. respondents “like it” or are motivated by money).⁵² Groves et al.⁵³ proposed a leverage-salience theory to describe the decision to participate in survey research. The theory suggests that a decision is made resulting from multiple considerations, some of which are survey specific (e.g. topic and sponsorship), some person-specific (e.g. privacy concerns), and others being specific to the respondent’s social and physical environment.⁵⁴ It is postulated that a combination of these factors may have resulted in the low retention and completion rates observed in this study.

As participants received no compensation for their time to complete the surveys, one potential consideration for future researchers is to have some form of incentive available for participants to encourage retention.^55,56 Further, as participants were asked to provide potentially sensitive information relating to income, despite assurances of confidentiality, some privacy concerns may have led to non-participation and/or omission of these requested details. It is also suggested that the retention rates observed may be the result of the first data collection occurring at six-weeks following a hand or wrist injury, a time that is well into recovery for most uncomplicated injuries.

It is likely that the cost estimates presented in the study could be either under- or over-estimated as is typical when surveying participants about the cost burden encountered as a result of injury or disease. Further, it is recognized that the use of self-reported questionnaires is prone to recall-bias. Participants were asked about their health status, number of medical and hand therapy appointments, and accrued financial costs over the period between the previous questionnaire and the current questionnaire which can impact on the accuracy of the data provided.

Finally, this investigation did not consider patient and public involvement in design of data collection methods. By having patients and public partners act as consultants in the design and review of the questionnaire relating to burden, an increased response rate is likely to be observed.⁵⁷

Directions for future research

Future research should be carefully designed to encourage participation and retention by considering the time burden placed on the participants within and across selected survey time points, providing participants with incentives to participate, and highlighting the relevance and real-world applications of the findings at the point of recruitment and potentially again at the time of survey completion. Also, researchers should consider patient and public involvement in the research design process, use hospital costing databases to calculate direct costs alongside indirect costs to capture a comprehensive picture of burden of injury^21,22,58 and explore patient experiences of burden using a qualitative design.

It was anticipated that the impacts on participation would be more pronounced and prolonged for people with tendon injuries than those with fractures or nerve injuries, and that burden of disease would be higher for this group. Owing to the extremely low response and retention rate of participants who consented to participate in conjunction with low numbers of participants outside the fracture subgroup, this assumption could not be formally tested.

Conclusions

The study highlights the challenges of completing longitudinal survey research investigating individual and societal burden with this population. While limited findings were achieved in this investigation, it does provide some preliminary insights into the possible individual and societal burden resulting from fractures, tendon and nerve injuries of the hand or wrist that warrants further investigation.

Footnotes

Acknowledgements

The authors would like to thank Ms Vincci Lee for her assistance with participant recruitment. The authors would like to thank the hand therapy departments at the Alfred and Dandenong Hospitals for their assistance with participant recruitment.

Declaration of Conflicting Interests

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

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported in part by an Australian Postgraduate Award (APA) scholarship administered by Monash University and a research scholarship from the Australian Hand Therapy Association (AHTA).

Ethical approval

Ethical approval was granted by Alfred Health, Monash Health and Monash University.

Informed consent

All procedures were followed as per the ethics application that led to approval. This included all participants providing written consent after they were informed of the aims, scope and methods of the study.

Guarantor

LSR.

Contributorship

LO’B and LSR conceived the study and designed the methodology. LSR collected and analyzed the data. LO’B and TB supervised the conduct of the study, data collection, and analysis. LSR drafted the article, LO’B and TB contributed substantially to its revision. All authors read and approved the final version.

ORCID iDs

Luke Steven Robinson

Lisa O’Brien

References

U.S. Bureau of Labor Statistics. Nonfatal occupational injuries and illnesses requiring days away from work. Survey of Occupational Injuries and Illnesses. Washington, DC, 2014.

Safe Work Australia. Work-related injuries resulting in hospitalisation, July 2006 to June 2009. Canberra, Australia, 2013.

Centers for Disease Control Prevention. Web-based injury statistics query and reporting system (WISQARS), 2010.

de Putter

Selles

Polinder

, et al. Economic impact of hand and wrist injuries: health-care costs and productivity costs in a population-based study. J Bone Joint Surg 2012; 94: e56–57.

Driscoll

Flood

Harrison

Work related hand and wrist injuries in Australia. Canberra: Commonwealth of Australia, 2008.

MacDermid

Donner

Richards

, et al. Patient versus injury factors as predictors of pain and disability six months after a distal radius fracture. J Clin Epidemiol 2002; 55: 849–854.

Matsuzaki

Narisawa

Miwa

, et al. Predicting functional recovery and return to work after mutilating hand injuries: usefulness of Campbell’s Hand Injury Severity Score. J Hand Surg 2009; 34: 880–885.

Guly

Diagnostic errors in an accident and emergency department. Emerg Med J 2001; 18: 263–269.

Dagenais

Caro

Haldeman

A systematic review of low back pain cost of illness studies in the United States and internationally. Spine J 2008; 8: 8–20.

10.

Cost-of-illness studies: concepts, scopes, and methods. Clin Mol Hepatol 2014; 20: 327.

11.

World Health Organization. WHO guide to identifying the economic consequences of disease and injury. Geneva: WHO, 2009.

12.

de Putter

Selles

Polinder

, et al. Economic impact of hand and wrist injuries: health- care costs and productivity costs in a population- based study. J Bone Joint Surg 2012; 94: e56–51.

13.

Sahin

Akca

Akkaya

, et al. Cost analysis and related factors in patients with traumatic hand injury. J Hand Surg Eur 2013; 38: 673–679.

14.

Dias

Garcia-Elias

Hand injury costs. Injury 2006; 37: 1071–1077.

15.

Drummond

McGuire

Economic evaluation in health care: merging theory with practice. Oxford: Oxford University Press, 2001.

16.

Guinness

Wiseman

Introduction to health economics. New York: McGraw-Hill Education (UK), 2011.

17.

Wonderling

Introduction to health economics. New York: McGraw-Hill Education (UK), 2011.

18.

De Putter

Selles

Haagsma

, et al. Health-related quality of life after upper extremity injuries and predictors for suboptimal outcome. Injury 2014; 45: 1752–1758.

19.

Drummond

Sculpher

Claxton

, et al. Methods for the economic evaluation of health care programmes. Oxford: Oxford University Press, 2015.

20.

Sharma

Maffulli

Biology of tendon injury: healing, modelling and remodelling. J Musculoskelet Neuronal Interact 2006; 6: 181.

21.

Robinson

O’Brien

Description and cost-analysis of Emergency Department attendances for hand and wrist injuries. Emerg Med Australas 2019; 31: 772–779.

22.

Robinson

Brown

O'Brien

, Cost, profile, and postoperative resource use for surgically managed acute hand and wrist injuries with emergency department presentation. J Hand Ther 2020.

23.

State Government of Victoria. Emergency department total attendances. The Alfred, Victorian Health Services Performance - Department of Health, Victoria, Australia, http://performance.health.vic.gov.au/Home/Report.aspx?ReportKey=157&HospitalKey=3, 2016.

24.

Victorian Agency for Health Information. Dandenong Hospital - Emergency care [online], https://performance.vahi.vic.gov.au/emergency-care/patients-treated-urgency-category?hid=7326&pid=3 (2019, accessed 10 October 2019).

25.

Luppa

Heinrich

Angermeyer

, et al. Cost-of-illness studies of depression: a systematic review. J Affect Disord 2007; 98: 29–43.

26.

Australian Bureau of Statistics. 1221.0 Information Paper: ANZSCO – Australian and New Zealand Standard Classification of Occupations. New Zealand: Australian Bureau of Statistics and Statistics, 2005.

27.

MacDermid

Tottenham

Responsiveness of the disability of the arm, shoulder, and hand (DASH) and patient-rated wrist/hand evaluation (PRWHE) in evaluating change after hand therapy. J Hand Ther 2004; 17: 18–23.

28.

Ware

Sherbourne

The MOS 36-item short-form health survey (SF-36) 1. Conceptual framework and item selection. Med Care 1992; 30: 473–483.

29.

Hakkaart-van Roijen

Short form-health and labour questionnaire. Rotterdam: University Medical Centre, 2007.

30.

Packham

MacDermid

JC.

Measurement properties of the Patient-Rated Wrist and Hand Evaluation: Rasch analysis of responses from a traumatic hand injury population. J Hand Ther 2013; 26: 216–223.

31.

Dacombe

Amirfeyz

Davis

Patient-reported outcome measures for hand and wrist trauma: is there sufficient evidence of reliability, validity, and responsiveness?.

Hand (N Y) 2016; 11: 11–21.

32.

Jenkinson

Wright

Coulter

Criterion validity and reliability of the SF-36 in a population sample. Qual Life Res 1994; 3: 7–12.

33.

Kopjar

The SF-36 health survey: a valid measure of changes in health status after injury. Inj Prev 1996; 2: 135–139.

34.

Ware

Jr , and Gandek

Overview of the SF-36 health survey and the international quality of life assessment (IQOLA) project. J Clin Epidemiol 1998; 51: 903–912.

35.

Team R. RStudio: integrated development for R. Boston, MA: RStudio, Inc., 2015. www.rstudio.com.

36.

Microsoft. Microsoft Excel [computer software]. Redmon, WA: Microsoft, 2016.

37.

Stevenson

SF-36: interim norms for Australian data. Australian Institute of Health and Welfare, 1996.

38.

Bhavnani

Toward occupation-based practice in hand rehabilitation. Phys Disabil Special Interest Sect Quart 2000; 23: 1–2.

39.

Robinson

Brown

O’Brien

Embracing an occupational perspective: occupation-based interventions in hand therapy practice. Aust Occup Ther J 2016; 63: 293–296.

40.

Wah

Wang

Ping

CL.

Construct validity of the Chinese version of the Patient-rated Wrist Evaluation Questionnaire (PRWE-Hong Kong Version). J Hand Ther 2006; 19: 18–26.

41.

Gosselin

Lemyre

Corneil

Presenteeism and absenteeism: differentiated understanding of related phenomena. J Occup Health Psychol 2013; 18: 71–76.

42.

Szolnoki

Hoffmann

Online, face-to-face and telephone surveys—comparing different sampling methods in wine consumer research. Wine Econ Policy 2013; 2: 57–66.

43.

Davis

Broome

Cox

RP.

Maximizing retention in community‐based clinical trials. J Nurs Scholarsh 2002; 34: 47–53.

44.

Shatenstein

Kergoat

M-J

Reid

Issues in recruitment, retention, and data collection in a longitudinal nutrition study of community-dwelling older adults with early-stage Alzheimer’s dementia. J Appl Gerontol 2008; 27: 267–285.

45.

Galea

Tracy

Participation rates in epidemiologic studies. Ann Epidemiol 2007; 17: 643–653.

46.

Curtin

Presser

Singer

Changes in telephone survey nonresponse over the past quarter century. Public Opin Quart 2005; 69: 87–98.

47.

Plewes

Tourangeau

Nonresponse in social science surveys: a research agenda. Washington, DC: National Academies Press, 2013.

48.

Nohr

Frydenberg

Henriksen

, et al. Does low participation in cohort studies induce bias? Epidemiology 2006; 17: 413–418.

49.

Tortora

RD.

Response trends in a national random digit dial survey. Metodoloski Zvezki 2004; 1: 21.

50.

Groves

Couper

MP.

Nonresponse in household interview surveys. New York: John Wiley & Sons, 2012.

51.

Ross

Grant

Counsell

, et al. Barriers to participation in randomised controlled trials: a systematic review. J Clin Epidemiol 1999; 52: 1143–1156.

52.

Singer

Bossarte

RM.

Incentives for survey participation: when are they “coercive”?

Am J Prev Med 2006; 31: 411–418.

53.

Groves

Singer

Corning

Leverage-saliency theory of survey participation: description and an illustration. Public Opin Quart 2000; 64: 299–308.

54.

Groves

RM.

Nonresponse rates and nonresponse bias in household surveys. Public Opinion Quart 2006; 70: 646–675.

55.

Larson

Chow

Total cost/response rate trade-offs in mail survey research: impact of follow-up mailings and monetary incentives. Ind Market Manage 2003; 32: 533–537.

56.

Larson

Poist

RF.

Improving response rates to mail surveys: a research note. Transport J 2004; 43: 67–74.

57.

Evans

Coad

Cottrell

, et al. Public involvement in research: assessing impact through a realist evaluation. Health Services and Delivery Research, No. 2.36, 2014.

58.

Robinson

Brown

O’Brien

Profile and cost of sport and exercise-related hand and wrist injuries with Emergency Department presentation. J Sci Med Sport 2020; 23: 683–689.