Article Text

Download PDFPDF

The test–retest reliability of the Military Physical Loading Questionnaire (MPLQ)
  1. Russell J Coppack1,2,
  2. J L Bilzon2,3,
  3. A K Wills4,
  4. T Papadopoulou5,
  5. R P Cassidy1,5,
  6. A M Nicol5 and
  7. A N Bennett1,6
  1. 1 Academic Department of Military Rehabilitation, Defence Medical Rehabilitation Centre Stanford Hall, Loughborough, UK
  2. 2 Versus Arthritis Centre for Sport, Exercise and Osteoarthritis Research, University of Bath, Bath, UK
  3. 3 Department for Health, University of Bath, Bath, UK
  4. 4 Faculty of Health Sciences, University of Bristol, Bristol, UK
  5. 5 Centre for Lower Limb Rehabilitation, Defence Medical Rehabilitation Centre Stanford Hall, Loughborough, UK
  6. 6 National Heart and Lung Institute, Faculty of Medicine, Imperial College London, London, UK
  1. Correspondence to Russell J Coppack, Academic Department of Military Rehabilitation, Defence Medical Rehabilitation Centre Stanford Hall, Loughborough LE12 5QW, UK; russ.coppack100{at}mod.gov.uk

Abstract

Introduction Despite the high prevalence of musculoskeletal injuries, there is a shortage of data quantifying the risk factors attributable to cumulative occupational demands among UK Military personnel. We developed a new comprehensive questionnaire that examines occupational and operational physical loading during military service. The aim of this study was to examine the test–retest reliability of the Military Physical Loading Questionnaire (MPLQ).

Methods Intraclass correlation coefficients (ICC) were used to evaluate the test–retest reliability (4-week interval) of the MPLQ on 18 occupational and 18 operational items in 50 male (mean age: 36 years; SD ±7.9) UK military personnel. A stratified analysis based on duration of Service (0–10 years, 11–20 years and ≥21 years) was conducted to assess whether stability of task items was dependent on participant length of recall. Internal consistency was assessed by Cronbach’s alpha (α) coefficients.

Results Reliability of individual operational items ranged from fair to almost perfect agreement (ICC range: 0.37–0.89; α range: 0.53–0.94) with most items demonstrating moderate to substantial reliability. Overall scores related to occupational items showed substantial to almost perfect agreement between administrations (ICC range: 0.73–0.94; α range: 0.84–0.96). Stratifying by duration of Service showed similar within group reliability to the entire sample and no pattern of decreasing or increasing reliability with length of recall period was observed.

Conclusions It is essential that data used in planning UK military policy and health services are as accurate as possible. This study provides preliminary support for the MPLQ as a reliable self-report instrument for assessing the cumulative lifelong effects of occupational loading in UK military personnel. Further validation studies using larger and more demographically diverse military populations will support its interpretation in future epidemiological research.

  • occupational & industrial medicine
  • preventive medicine
  • rehabilitation medicine
  • epidemiology

Data availability statement

Data are available upon reasonable request. Due to privacy concerns, some data regarding participants are available only to bona fide researchers working on a related project, subject to completion of a non-disclosure agreement. Access requests for any restricted data should be sent to russ.coppack100@mod.gov.uk.

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Key messages

  • No questionnaire specifically designed to monitor the relationship between occupational physical loading and hip pain/musculoskeletal injury in military populations is available.

  • We report the test–retest reliability of the Military Physical Loading Questionnaire (MPLQ) designed to measure exposure to lifelong occupational physical loading and hip pain risk in military personnel.

  • The study provides evidence supporting the reliability and internal consistency of the MPLQ tested in a convenience sample of UK military personnel.

  • Data used in planning UK military policy and health services must be accurate. The MPLQ may provide a reliable instrument to measure occupational physical workload in military cohorts.

Introduction

Musculoskeletal injuries (MSKIs) are a major burden in military populations resulting in a reduction of operational strength and force readiness.1 High incidence rates of MSKI are reported in the literature with military training cited as a common causative factor.2 Two recent UK studies reported 58% of 18103 and 49% of 66084 Army recruits suffered at least one MSKI during training, with overuse lower-limb injury the most common diagnostic category. MSKI was the principle cause in the medical discharge of 4917 British Army personnel (61%) between 2012 and 2016 and accounted for 67% of all medical down gradings.5 Overuse MSKI is also reported as a primary source of disability in non-UK military personnel in training and during combat operations.6

Occupation is an important determinant of cumulative stress and workload and the military population is particularly at risk given the inherent occupational demands.7 However, no studies have investigated cumulative exposure to occupational mechanical loading as a risk factor for developing hip pathology and osteoarthritis (OA) in UK military personnel. Research is required to better understand the root causes of MSKI among UK military cohorts, thereby enabling the development of cost-effective, targeted prevention strategies.

The self-report questionnaire is the preferred instrument for measuring lifetime occupational physical loading of joints in epidemiological studies.8 9 The cumulative, repetitive use and excessive loading of the hip over time has been linked to OA.10 Therefore, it is important to identify the mechanical loads placed on the musculoskeletal system throughout life to accurately assess the occupational risk associated with hip OA. The Military Pre-Training Questionnaire is the only instrument specifically developed to offer a means of assessing important characteristics and injury risk of trainees entering British Army Training.9 To our knowledge, no questionnaire specifically designed to monitor the relationship between lifetime occupational loading and hip injury in military populations is available.

We developed a new comprehensive questionnaire adapted from existing validated instruments used in population-based studies.8 9 11 The Military Physical Loading Questionnaire (MPLQ) examines physical activity levels and occupational mechanical loading prior to and during military service. However, it is not known if UK military personnel can reliably recall information about past occupational exposures. Therefore, the purpose of this study is to report the test–retest reliability of questions examining occupational and operational related mechanical loading in a representative sample of UK military personnel.

Methods

Study design

The study was planned and conducted in accordance with the UK Ministry of Defence (MOD) policy for research using human participants and the Helsinki declaration.12 The study protocol was approved by the MOD research ethics committee (approval code 651/MODREC/15 dated 18 July 2016). A prospective test–retest study design was used to assess the reliability of the MPLQ, completed approximately 4 weeks apart.

Participants

All participants were serving members of the UK Armed Forces employed at the Defence Medical Rehabilitation Centre (DMRC), Headley Court, UK. Potential participants were notified using publicity posters and announcements on the DMRC organisational intranet webpage. Participants who expressed a willingness to participate were provided with a study information sheet detailing the aims and procedures of the study. The inclusion criteria were full-time serving UK military personnel, male, aged 18–55 years. A project investigator provided a verbal brief on questionnaire completion to all participants meeting the study eligibility criteria who provided their signed informed consent.

Sample

A sample of 50 male volunteers were recruited into this study between January 2017 and February 2018. Our sample size was based on the consensus-based standards for the selection of health measurement instruments criteria which states a sample size of 30–49 participants is considered ‘fair’ and 50–99 considered ‘good’ for a validation study.13

Questionnaire development

The MPLQ collects information on various categories of risk factors shown to be associated with MSKI in Military populations14 and hip OA.10 The instrument was designed to assess, in separate sections, pre-entry activity level and exercise, injury history, occupational loading, operational deployment loading, sport and recreation and lifestyle factors. Items were selected from existing questionnaires used in epidemiological research.8 9 11 Questions in the sections pertaining to occupational physical loading and operational deployment loading were made specific to the target military population. The questions in sections surrounding pre-entry activity and exercise, injury history, sport and recreation and lifestyle factors have been shown to be reliable in military populations and young active adults.8 9 11 Therefore, this reliability study focuses only on the questions surrounding occupational (job related) and operational deployment physical loading.

Measurement of occupational physical demands

History of cumulative exposure to occupational (job related) physical demands is measured from the point of enlistment. Participants are asked about each job/posting held for 1 year or longer up to a maximum of eight postings. Job number 1 describes the combined period of phase 1 (recruit) and phase 2 (trade) military training. Participants rate their involvement and exposure to each of 18 physical demand tasks (online supplementary file, MPLQ, section 4). The 18 items comply with the nomenclature used routinely in the UK and North Atlantic Treaty Organisation (NATO) defence forces to categorise high-intensity, moderate-intensity and low-intensity occupational military tasks.15

The frequency of each physical task is rated on a 5-point scale with 0=’never’, 1=’not very often’, 2=’sometimes’, 3=’often’, 4=’very often’. This method of recording occupational physical demands has been used in community-based hip pain studies and its construct validity demonstrated.8

Measurement of deployed operations physical demands

Participants are asked about performance during their time (total summed months) spent on deployed military operations. Information is provided on the average number of hours in a 12-hour day (none, 0–1, 2–4, 5–7 and 8+hours) performing each of the 18 operational tasks (online supplementary file, MPLQ, section 5). These tasks are a variation on the nomclementure used to construct and categorise the occupational physical demand tasks in section 4 of the MPLQ. This section includes questions on tasks specific to the combat environment that may not be otherwise considered routine (eg, flying rotary/fixed wing, armoured convoys, etc). Participation in each specific task is calculated by taking the product of duration (total days on operations) × self-reported length of participation each day (average hours). Output data will yield information used to assess if exposure to physical loading on operational deployments presents an additional risk for developing hip pain compared with other periods during a military career.

A copy of the MPLQ is provided as an online supplementary file.

Study procedures

Participants were asked to complete the self-administered questionnaire (paper version) on two occasions with an interval of approximately 4 weeks between administrations. The 4-week ‘washout’ period was chosen to minimise a ‘learnt’ (recall) response bias to the instrument while avoiding a potential change in the exposure construct being measured.16 Participant feedback confirmed questionnaire completion usually took 25–35 min. The MPLQ employed ‘skip-logic’ allowing participants to avoid negative, irrelevant responses to questions thereby reducing participant burden.11 If questionnaires were not returned within a 3-week delay, one e-mail and single telephone contact was attempted.

Statistical analysis

Statistical calculations were performed using IBM SPSS (V.25.0.0, SPSS, Chicago, IL, USA). Descriptive statistics were performed to characterise the study sample. Differences in scores were calculated for the occupational and operational task questions comparing initial to follow-up scores. Because the number of jobs held for ≥1 year differed across participants, we measured the reliability of aggregated pooled scores for individual questions on each post held (1, 2, 3, etc) for occupational task questions. We also conducted a stratified analysis where participants were classified according to duration of military Service in 10-year intervals (0–10 years, 11–20 years and 21 years+) with the aim of assessing if stability of individual task responses was dependent on participant length of recall.

To examine the test–retest reliability between occupational and operational tasks at baseline and retest, we calculated the intraclass correlation coefficients (ICC1,1) with 95% CIs based on a one-way random-effects analysis of variance model. This ICC1,1 uses test–retest measures to estimate single trial reliability rather than the average of repeated measures. As a guide, strength of agreement ratings between test–retest responses suggested by Landis and Koch17 was used: poor=0–0.2, fair=0.2–0.4, moderate=0.4–0.6, substantial=0.6–0.8 and almost perfect=0.8–1.0. Cronbach’s alpha (α) coefficient was used to measure the internal consistency of the questionnaire. Internal consistency was deemed acceptable if α was >0.7.13

Results

Participant characteristics

Baseline participant characteristics are summarised in Table 1. In all, 50 male participants provided informed consent to participate in the study. All participants were serving UK Military personnel with a mean age of 35.8 years (SD ±7.9). A complete response (ie, MPLQ completed on two occasions) was obtained from 42 respondents (84%). Eight respondents did not complete and return a follow-up questionnaire within the allotted timeframe and could not be included in the data analysis. There was an average of 29 days (SD ±3.6) between each administration of the questionnaire (range: 26–42 days). Most participants were Caucasian (92%) and university educated (68%). The distribution of participants by military branch was 25 (50%) Army, 15 (30%) Royal Air Force, 5 (10%) Royal Navy and 5 (10%) Royal Marines. The patient distribution by rank seniority was 12 (24%) junior ranks, and 19 (38%) for both the senior and officer rank categories. The most common job roles were physical training instructor 12 (24%), physiotherapist 9 (18%), doctor 7 (14%) and logistics specialist 6 (12 %). The mean number of postings for ≥1 year was 4.8 (SD ±2.0) with a cumulative mean 9.1 months (SD ±4.5) served on deployed operations.

Table 1

Baseline descriptive characteristics of study participants (n=50)

Test–retest reliability

Operational loading items

Table 2 summarises the results of the test–retest reliability for 18 operational loading items of the MPLQ. A significant number of missing items were recorded at baseline and re-test by 15 (38%) of participants. This reflected responses from participants with no operational exposure during their career. Including ‘none’ response options from this subgroup in the analysis could introduce a degree of bias that over-estimates the stability of these MPLQ items. Therefore, only data from participants with a minimum 6 months exposure on deployed/combat operations (n=27) was used for analysis purposes.

Table 2

Test–retest reliability of MPLQ operational loading task items

The highest reliability coefficients were obtained for the items flying (fixed-wing fast jet), ICC1,1, 0.89 (95% CI: 0.78–0.95), operating heavy tools and/or weapon systems ICC1,1, 0.89 (95% CI: 0.77–0.95) and driving over ‘rough’ terrain, ICC1,1, 0.80 (95% CI: 0.61–0.90) all demonstrating substantial to almost perfect agreement. The lowest reliability was found for items related to crawling, ICC1,1, 0.37 (95% CI: 0.01–0.65) and climbing/scaling walls, ICC1,1, 0.38 (95% CI: 0.78–0.95) showing fair strength of agreement. Reliability of all other occupational loading items ranged from moderate to substantial (ICC1,1 range: 0.44–0.74) with a majority of items showing moderate agreement between administrations. Internal consistency determined by Cronbach’s alpha coefficient was ≥0.7 for 13 of the 18 occupational loading items (range: 0.70–0.94); crawling had the lowest internal consistency (α=0.53).

Occupational loading items

Within the entire sample, the occupational loading items showed substantial to almost perfect agreement across all summary measures (Table 3). Reliability coefficients for questions relating to lifting and moving weights showed the highest ICC1,1 values (range: 0.91–0.94). The item on frequency of climbing ladders showed the lowest reliability coefficient in this section ICC1,1, 0.73 (95% CI: 0.66–0.80). All occupational loading items showed Cronbach’s alpha (α) values greater than 0.70 (range: 0.84–0.96) suggesting high internal consistency and homogeneity for these items.

Table 3

Test–retest reliability of MPLQ occupational loading task items

Stratifying by duration of Service 0–10 years (n=15), 11–20 years (n=16) and >21 years (n=11) showed similar within group reliability to the entire sample. The majority of items demonstrated substantial to almost perfect agreement in each subgroup (table 4). The item on road driving for a least 4-hours had the lowest reliability, ICC1,1, 0.53 (95% CI: 0.33–0.69) in the >21 years subgroup. However, a pattern of decreasing or increasing reliability with length of recall period was not observed and internal consistency (α) were comparable regardless of duration of service. In general, better reliability was observed for occupational loading items than operational items.

Table 4

Test-retest reliability of MPLQ occupational loading task items by duration of Service (0–10 years, 11–20 years, >21 years)

Discussion

This study reports the 4-week test–retest reliability and internal consistency of created occupational and operational exposure items of the MPLQ. Results showed moderate to almost perfect agreement for operational items (ICC1,1, range: 0.37–0.89), and substantial to almost perfect agreement for all occupational items (ICC1,1, range: 0.73–0.94). Length of recall period did not influence reliability scores and acceptable to good internal consistency was shown for the majority of all task items. The reliability of occupational task items was generally higher than operational task items. These results are important as they provide preliminary support for the MPLQ as a reliable measure of occupational physical workload and MSKI risk in UK military personnel.

Reliability responses

For items concerning operational tasks, the highest repeatability was found for ‘operating heavy tools/weapon systems’, ‘flying (fixed wing fast-jet)’ and ‘driving over rough terrain causing your body to shake’. Higher reliability in response to questions concerning occupational ‘vibrations’ and working postures involving the whole body have previously been reported.18 Furthermore, heavy load activity is consistently recalled more reliably than less intense activity.19 Activities of mild activity are more common, less memorable and less likely to be accurately captured by self-report.20 Lower test–retest reliability estimates were found for the operational tasks ‘crawling’ (ICC1,1, 0.38) and ‘climbing/scaling walls’ (ICC1,1, 0.37). It is possible the lower reliability for these tasks may be a result of reduced precision attributed to crawling and climbing activities occupying little time and therefore difficult to memorise in self-report.21

For occupational task items, the present results were consistent with previous studies reporting higher reliability responses for questions concerning repetitive lifting of manual loads.22 The ICC values in our study (0.91–0.94) for ‘lifting & moving weights’ were generally higher than previously reported. Military personnel routinely plan and perform weight carriage activity with specified loads. Our finding that load lifting activity showed the highest test–retest reliability may reflect the routine nature of this activity and explain why military personnel display accurate recall of weight carriage task categories.23

A main finding in the present study was the higher reliability and consistency found for occupational task items compared with operational task questions. Occupational histories are easier to recall than events occurring irregularly as they rely on generic knowledge rather than specific memories.24 The 18 occupational items in the MPLQ centred around patterns of activity during specified time periods (job’s/postings held) where generic memory may be more important than the specific, episodic recall of operational experiences. For military personnel working life comprises a significant span of time and posting’s that potentially facilitates recall of occupational activities.24 However, the smaller sample used for the operational tasks subgroup analyses may have resulted in recruitment bias and a misclassification of occupational exposure, thereby diluting a potential relationship between exposure and response compared with occupational task scores.18

We did not find any significant group differences when reliability scores were stratified by duration of military service. Earlier research has shown self-report accuracy decreases with an increase in time from a given event.25 Our findings suggest using individual jobs/postings of over 1 year was effective in increasing the reliability of recall for specific time periods during the respondents’ military career.24 The internal consistency of occupational task questions was very high with a Cronbach’s alpha range of 0.84–0.97 across the 18 items. While this could support the notion the MPLQ is a stable measure of military occupational exposure, a Cronbach’s alpha score over 0.90 indicates redundancy rather than a desirable level of internal consistency.26

Methodological considerations and study limitations

The study has some methodological limitations that should be noted. We aimed to assess the reliability of operational and occupational questions in the military population in which the MPLQ will be used. Study participants in sedentary or light-to-moderate activity occupations were over represented (eg, administration, medical, logistics). This can lead to a disproportionate concentration of responses for low exposures on the numeric ordinal scale affecting resultant ICC scores.24 Furthermore, the majority of participants in our sample were Caucasian, university educated and male only. Education level may influence the reliability of responses as higher educational attainment is associated with greater consistency of recall.18 Therefore, the reliability of MPLQ items requires further evaluation using military participants from high, medium and low loading exposure occupations and a more representative mix of educational level, ethnic background and gender. Our test–retest sample for operational items was limited to 27 participants with exposure to deployed operations and some imprecision in ICC estimates is possible in this small subsample. Future studies need to validate the MPLQ in a larger sample of military personnel.

Conclusions

The availability of reliable physical loading data is essential for epidemiological investigations of MSKIs, particularly in military populations. We have developed a self-administered screening questionnaire designed to measure lifelong exposure to occupational physical loading as a risk factor for hip pain in military personnel. Results provide initial support for the test–retest reliability of the MPLQ occupational and operational items. With a re-design of existing questions, the MPLQ could potentially be used to measure the association between cumulative physical workload and injury risk for other musculoskeletal disorders. Further studies are encouraged with larger, demographically diverse military populations to further validate this tool.

Data availability statement

Data are available upon reasonable request. Due to privacy concerns, some data regarding participants are available only to bona fide researchers working on a related project, subject to completion of a non-disclosure agreement. Access requests for any restricted data should be sent to russ.coppack100@mod.gov.uk.

Ethics statements

Patient consent for publication

Ethics approval

This study was approved by the UK Ministry of Defence Research Ethics Committee, approval code 651/MODREC/15 dated 18 Jul 2016.

Acknowledgments

The authors wish to thank those military members of permanent staff at DMRC Headley Court for their time and contribution to participation in this study.

References

Footnotes

  • Contributors RJC designed the study, conducted the initial analysis, drafted the initial manuscript and approved the final manuscript as submitted. All authors analysed and interpreted the findings. JLB, AKW and ANB supervised the conduct of the study, assisted with data analysis, reviewed and revised the manuscript and approved the final manuscript as submitted. RPC assisted with data collection and participant recruitment, drafted the initial manuscript with RJC and critically reviewed the final manuscript. TP and AMN reviewed and revised the manuscript, and approved the final manuscript as submitted.

  • Funding This study is funded by the Versus Arthritis Centre for Sport, Exercise & Osteoarthritis Research (Grant Reference 20194).

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally 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.