Original Research Contribution
Provision of Out-of-hospital Analgesia to Older Fallers With Suspected Fractures: Above Par, but Opportunities for Improvement Exist
- The authors have no relevant financial information or potential conflicts of interest to disclose.
Address for correspondence and reprints: Paul M. Simpson; e-mail: email@example.com
Paramedics frequently attend older patients who have fallen and sustained suspected fractures, a population of patients who may be at risk of inadequate analgesic care. This prospective study aimed to describe the rate and effectiveness of analgesia administered by paramedics to older patients with suspected fractures secondary to falls and to identify predictive factors associated with provision of analgesia.
A cohort of older patients aged greater than 65 years with suspected fall-related fractures was extracted from a database of 1,610 cases collected during a prospective, nonconsecutive observational study of older people who had fallen and received an ambulance response from October 1, 2010, through June 30, 2011. Fall-specific data, collected on scene by paramedics using a specially designed data form, were linked to patient clinical records and dispatch information. Descriptive analyses were performed to describe rates and effectiveness of analgesic administration, and multivariate logistic regression was conducted to identify factors associated with provision of analgesia.
Of 1,610 patients in the observational study database, there were 333 patients identified as having suspected fractures, thus forming the study population. The mean (±SD) age was 82 (±8) years, and 75% were female. Suspected fractures of the hip were most common (42%). An initial pain score was recorded in 67% of cases, and the median initial pain severity was 8 of 10 (interquartile range [IQR] = 5 to 9). Overall, 60% received analgesia, and 80% of those received parenteral opiates. Intravenous (IV) morphine was most common (63%), followed by methoxyflurane (39%) and intranasal fentanyl (17%). Administration of oral analgesics was uncommon. Analgesia was considered to be clinically effective (≥30% relative reduction in pain severity) in 62% of cases. Patients with suspected hip fractures had greater odds of receiving analgesia compared to those with suspected fractures at other anatomical sites (odds ratio [OR] = 2.7, 95% confidence interval [CI] = 1.17 to 6.32; p = 0.02). Compared to those with mild pain, the odds of receiving analgesia increased significantly for patients with moderate pain (OR = 6.5, 95% CI = 2.3 to 18.8; p < 0.0001) and severe pain (OR = 31.1, 95% CI = 9.9 to 97.6; p < 0.0001).
In this population of older people who fell and sustained suspected fractures, two-thirds received paramedic-administered analgesia. The majority of patients received clinically effective analgesia, and the presence of a suspected hip fracture increased the likelihood of receiving pain relief.
Administración de Analgesia Extrahospitalaria a los Ancianos con Sospecha de Fractura tras una Caída: Cumplimos (por encima del par), pero Existen Oportunidades para la Mejora
Los paramédicos atienden frecuentemente a los ancianos que se han caído y en los que se sospecha una fractura, los cuales constituyen una población con riesgo de recibir una atención analgésica inadecuada. Este estudio prospectivo tuvo como objetivos describir el porcentaje y la efectividad de la analgesia administrada por paramédicos a los ancianos con sospecha de fracturas secundarias a caídas e identificar los factores predictivos asociados a la administración de la analgesia.
Se extrajo una cohorte de pacientes ancianos mayores de 65 años de edad con sospecha de fracturas relacionadas con caídas de una base de datos de 1.610 casos recogidos durante un estudio observacional prospectivo no consecutivo de ancianos que se habían caído y habían recibido una respuesta de una ambulancia desde el 1 de octubre de 2010 hasta el 30 junio de 2011. Los datos específicos de la caída recogidos en la escena por los paramédicos usando una hoja de datos especialmente diseñada se vincularon a las historias clínicas de los pacientes y a la información del traslado. Se calcularon los porcentajes y la efectividad de la administración de analgésicos, y se llevó a cabo una regresión logística multivariable para identificar los factores asociados con la administración de analgesia.
De los 1.610 pacientes en la base de datos del estudio observacional, había 333 pacientes identificados con sospecha de fracturas, lo que formaron la población de estudio. La edad media fue de 82 años (DE 8 años), y el 75% fueron mujeres. La sospecha de fractura de cadera fue la más común (42%). Una puntuación de dolor inicial se registró en el 67% de los casos, y la mediana de intensidad de dolor inicial fue de 8 sobre 10 (RIC 5 a 9). Del total, el 60% recibió analgesia y el 80% de ellos recibió un opiáceo parenteral. La morfina intravenosa fue más común (63%), seguida de metoxiflurano (39%) y fentanilo intranasal (17%). La administración de analgésicos orales fue poco frecuente. La analgesia se consideró clínicamente efectiva (≥30% reducción relativa en la intensidad del dolor) en el 62% de los casos. Los pacientes con sospecha de fractura de cadera tuvieron mayor razón de ventaja (odds ratio, OR) de recibir analgesia comparado con otros sitios anatómicos (OR 2,7, IC95% = 1,17 a 6,32; p = 0,02). En comparación con aquéllos con dolor leve, la OR para recibir analgesia se incrementó significativamente para los pacientes con dolor moderado (OR 6,5, IC95% = 2,3 a 18,8; p < 0,0001) y dolor grave (OR 31,1, IC 95% = 9,9 a 97,6; p < 0,0001).
En esta población de ancianos que se cayeron y se sospechó una fractura, dos tercios recibieron analgesia administrada por un paramédico. La mayoría de los pacientes que recibieron analgesia ésta fue clínicamente efectiva, y la presencia de una fractura de cadera incrementó la probabilidad de recibir alivio para el dolor.
Older people who have suffered falls frequently require pain management in the out-of-hospital (OOH) setting. Approximately two-thirds of those who present to paramedics are suspected of having fractures, injuries that almost certainly cause significant pain and discomfort. Paramedics are required not only to provide pain relief to these patients at rest, but provide analgesia in a “procedural” context during difficult extrications such as from confined areas or down flights of stairs and during transport.
Older people with traumatic injuries represent a patient population for whom pain management can be particularly challenging. Changes in physiologic and metabolic processes due to the normal aging process and the presence of comorbidities can result in atypical manifestation and expression of pain, making assessment of pain type and severity difficult for the clinician. Aging also affects the pharmacodynamics and pharmacokinetics of analgesic pharmacology, increasing the risk of unpredictable responses and the frequency of adverse effects arising from standard dosing regimens.[3-5] Consequently, older patients may be at greater risk of inadequate pain relief, or “oligoanalgesia,” due to underdosing by clinicians concerned about the possibility of such problems. While undesirable in patients of any age, oligoanalgesia has been linked with deleterious outcomes for older patients, including exacerbations of existing medical conditions; psychological distress; increased risk of delirium; and onset of chronic, persistent pain.[3, 6]
In most western countries, ambulance services equip their paramedics with several analgesic agents, allowing the emergency medical services (EMS) provider to select a modality of analgesia appropriate to the profile of the patient, the circumstance, and the etiology of pain. Despite this, low rates of OOH analgesia among older patients with suspected or confirmed fractures have been described in several retrospective studies, suggesting that pain management of older patients may be less than optimal.[8-10] The aims of this analytic study were to describe the rate and effectiveness of analgesia administered by paramedics to older patients with suspected fractures secondary to falls and to identify predictive factors associated with provision of analgesia.
This study was a planned subanalysis of a cohort of patients extracted from a larger data set that had been collected during a prospective observational study investigating prehospital management of older people who had fallen and received an emergency response. That epidemiologic study used a prospective, convenience sampling (nonconsecutive enrollment) design to collect fall-specific data over a 9-month period from October 1, 2010, through June 30, 2011. During that period, paramedics who had volunteered to participate in the study, after arriving on scene, prospectively enrolled older people who were confirmed to have fallen and collected fall-specific data in addition to routinely collected clinical record and dispatch data. Ethical approval was granted by the Sydney Local Health District Human Research Ethics Committee (HREC), Royal Prince Alfred Zone (Protocol X10-0152 and HREC/10/RPAH/282).
Study Setting and Population
The study setting was New South Wales (NSW), Australia. NSW is serviced by a single, government-funded, state-based ambulance service that provides EMS coverage to a population of approximately 7.13 million people across a geographical area of 800,000 km2 spanning metropolitan, rural, and remote regions. Fourteen percent of the population (1.02 million) is aged 65 years or more. There are approximately 3,500 paramedics spread across four clinical levels: paramedic trainees (in first year of training), paramedic interns (second and third year of training), qualified paramedics (have completed three years of core training), and paramedic specialists (either intensive care paramedic or extended care paramedic). Training is provided internally using a vocational training model, but with an increasing number of university-educated paramedics.
The study population consisted of patients aged 65 years or older who had received an emergency ambulance response after experiencing a fall and had been enrolled in the observational study by paramedics. A database consisting of cases collected during the observational study was interrogated to identify patients who had been identified by paramedics at the scene as having suspected fractures to any anatomical regions secondary to falls. A subset of such patients was created, forming the cohort that would constitute the population of interest for this analysis.
Paramedics have several analgesic agents available for treating acute and chronic pain depending on clinical level and training. All paramedics are able to administer oral paracetamol and ibuprofen and inhaled methoxyflurane. Qualified paramedics and paramedic specialists can administer intravenous (IV) or intramuscular morphine and intranasal fentanyl. In addition, extended-care paramedics are authorized to use oral paracetamol with codeine and oral oxycodone. Paramedics administer analgesics under authorization of a medical officer using a protocol-based system. No consultation with physicians is required at the scene prior to administration, and the choice of analgesic from this selection of agents is at the discretion of the paramedic on scene. While maximum doses are mandated for each agent, the amount of analgesia given within that range is at the discretion of the attending paramedic. Pain is measured using the patient-based verbal numeric rating scale of 0–10, with “0” representing “no pain” and “10” representing the “worst pain you can imagine.”
The observational database from which this population of eligible cases were extracted consisted of fall-specific data combined with routinely collected patient clinical records and emergency dispatch data, both of which were completed either during or immediately following each case. The fall-specific data originated from a purpose-designed data collection tool that had been created using a consensus approach by an advisory panel of experts in the areas of falls, geriatrics, and prehospital care (see Data Supplement S1, available as supporting information in the online version of this paper). Prior to the start of data collection, the form was piloted among a small group of patients to ensure ease of use and utility. The form collected data not captured in routinely completed ambulance clinical records, including information about patient demography and profile, specifics relating to the location and nature of the fall, reasons for falling, risk factors, and previous ambulance use for falls. The routinely completed, paper-based patient clinical record contained information about patient care such as whether an analgesic agent was administered, type of analgesic and dose, and contraindications to analgesia if none was administered. Additionally, a short free-text box of the patient clinical record was also searched for information relating to reasons for an analgesic not being administered. As this free-text box is small and constrained to only eight lines, this manual process was undertaken by a single researcher with knowledge and familiarity of the clinical record. Reasons for nonadministration were classified into four categories: “patient denied pain,” “patient refused or declined analgesia,” “analgesia already administered by another health professional or was self-administered prior to ambulance arrival,” and “contraindicated due to patient condition.”
Data were analyzed using SAS 9.2 (SAS Institute, Cary, NC). Descriptive statistics were generated, with continuous variables described as mean with standard deviation (±SD) or median with interquartile range (IQR) for normal and nonnormally distributed data, respectively. Differences in proportions were assessed using chi-square analyses. Statistical significance for all descriptive tests was established when p < 0.05. Assessment of analgesic effectiveness was undertaken for cases in which initial and final pain scores were recorded. “Clinically effective analgesia” was defined as a relative reduction in initial pain score of ≥30%, a reduction that constitutes clinically meaningful pain relief in emergency settings. Univariate logistic regression was performed to identify factors predictive of receiving analgesia. The continuous variable of patient age did not meet the assumptions of linearity, so was categorized into three groups (65–74, 75–84, 85 years and older). Non–English-speaking background status was categorized as yes or no; time of day as night shift (1900–0659) or day shift (0700 to 1859); patient's residential status as community dwelling or residential aged care; anatomical site of fracture as hip or other area; pain score recorded as yes or no; and pain severity as mild (1–3), moderate (4–7), or severe (8–10), using validated cut points previously cited in OOH analgesic literature.[12-14] The dichotomous primary outcome for the logistic regression analyses was administration of any analgesic (yes or no), modeling the odds of receiving analgesia. Variables with significance of p < 0.25 in univariate analysis were included in the base multivariate logistic regression model. A backward elimination approach was used to fit the final model, with the most nonsignificant variable removed at each step until only statistically significant variables (p < 0.05) remained. Interactions between significant variables were then assessed and retained in the regression model only if p < 0.01 and the goodness of fit of any final model assessed and considered acceptable if p > 0.05.
There were 1,610 patients in the observational study database. Of these, 333 were documented by paramedics as having had suspected fractures secondary to falls, forming the study population for this analysis. The characteristics of this population are described in Table 1. Three-quarters of patients were female, and 83% were community dwelling. Suspected fractures of the hip were most common. Initial pain scores were documented in 67% of cases, and final pain scores in 52%. The median initial pain score was 8 (IQR = 5 to 9). Of 173 (52%) patients for whom initial and final pain scores were recorded, 107 (62%) received clinically effective analgesia evidenced by reductions in initial pain severity of ≥30%.
Table 1. Characteristics of Patients Who Had Fallen and Presented to Paramedics With Suspected Fractures (n = 333)
|Age, mean (±SD)||82 (±8)|
|Sex, n (%) female||250 (75)|
|Residential status, n (%)|
|Community dwelling||274 (83)|
|Residential aged care||56 (17)|
|Non–English-speaking background, n (%)||55 (18)|
|Anatomical site of suspected fracture, n (%) hip||138 (42)|
|VNRS pain score, median (IQR)|
|Initial severity||8 (5–9)|
|Final severity||3 (2–5)|
A total of 190 patients (60%) with suspected fractures received OOH analgesia. The proportion of those receiving analgesia was higher in patients with suspected hip fracture compared to other fracture sites (67% vs. 55%; χ2 = 4.68, p = 0.03). The use of analgesia was also higher in patients with suspected fractures for whom pain scores of any severity had been recorded (73% vs. 33%; χ2 = 48.13, p < 0.0001). Reasons for not administering analgesia were identified in 72 of 134 cases of nonadministration. In 16 instances the patient refused offers of analgesia despite reporting pain, while 13 denied having pain. Paramedics reported withholding pain relief in eight cases because analgesia had already been administered by another health professional or self-administered by the patient prior to ambulance arrival. Only three cases stated that analgesics were contraindicated due to the patient's condition. Of patients presenting with moderate to severe pain (i.e., ≥4 of 10), 152 of 181 (84%) received analgesia. Among the 29 patients with moderate to severe pain to whom pain relief was not provided, nine refused or declined analgesia, and seven had already been administered pain relief from another health professional or self-administered prior to ambulance arrival, while in one case analgesia was contraindicated. No reason was documented in 13 instances.
Of those to whom analgesia was administered, 80% received parenteral opiates (IV/intramuscular morphine or intranasal fentanyl, either alone or in combination with another agent). Morphine was the most commonly used analgesic, administered to 63% of patients, either alone or in combination with other analgesics (Table 2). Methoxyflurane (39%) and fentanyl (17%) were also frequently used, but oral analgesics were uncommonly administered.
Table 2. Frequencies and Doses of Paramedic-administered Analgesics Provided to Older People With Suspected Fractures Following Falls
|Morphine (parenteral), mg||126 (63)||8.5 (±4.3)|
|Fentanyl (IN), μg||77 (39)||251.0 (±117)|
|Methoxyflurane (INH), mL||34 (17)||3.0 (±0)|
|Paracetamol (O), g||7 (4)||1.0 (±0)|
|Ibuprofen (O), mg||3 (2)||400.0 (±0)|
|Oxycodone (O), mg||1 (1)||7.5 (±2.9)|
|Paracetamol + codeine (O), mg||1 (1)||30.0 (±30)|
The univariate associations of 10 explanatory variables with the outcome of provision of analgesia are outlined in Table 3. There were significant associations for site of fracture, recording of pain score, and pain severity. For the multivariate analysis, six variables were retained for the model based on the significance of the association with the outcome variable being less than 0.25. The final multivariate model included site of suspected fracture and pain severity (Table 4). After pain severity was adjusted for, patients with suspected hip fractures had almost three times greater odds of receiving analgesia compared to those with fractures to other anatomical locations (odds ratio [OR] = 2.7, 95% confidence interval [CI] = 1.17 to 6.32; p = 0.02). When adjusted for site of suspected fracture, and compared to those with mild pain, the odds of receiving analgesia increased significantly for patients with moderate pain (OR = 6.5, 95% CI = 2.3 to 18.8; p < 0.0001) and severe pain (OR = 31.1, 95% CI = 9.9 to 97.6; p < 0.0001). The Hosmer-Lemeshow goodness-of-fit test indicated acceptable fit of the final model (χ2 = 0.66, 3 degrees of freedom, p = 0.88).
Table 3. Univariate Logistic Regression Analysis Modeling the Unadjusted Odds of Receiving Analgesia From Paramedics for Older Patients With Suspected Fractures Following Falls
|All patients (n = 333)||134 (40.2)||199 (59.8)||—||—|
|Patient age group (yr)|
|65–74 (referent)||17 (28.3)||43 (71.7)||1.00||0.08|
|75–84||54 (46.2)||63 (53.9)||0.46 (0.24–0.90)|
|85+||57 (39.6)||87 (60.4)||0.60 (0.31–1.16)|
|Male (referent)||31 (36.9)||53 (63.1)||1.00||0.50|
|Female ||102 (41.1)||146 (58.9)||0.84 (0.50–1.40)|
|Community dwelling (referent)||104 (38)||170 (62)||1.00||0.09|
|Residential aged care||28 (50)||28 (50)||0.61 (0.34–1.09)|
|Yes (referent)||18 (32.7)||37 (67.3)||1.00||0.23|
|No||106 (41.4)||150 (58.6)||0.69 (0.37–1.27)|
|Anatomical site of fracture|
|Other (referent)||88 (45.1)||107 (54.9)||1.00||0.03|
|Hip ||46 (33.3)||92 (66.7)||1.65 (1.05–2.59)|
|Normal (referent)||61 (38.9)||96 (61.2)||1.00||0.61|
|Abnormal||57 (36.1)||101 (63.9)||1.13 (0.71–1.78)|
|Pain score recorded|
|No (referent)||73 (67)||36 (33)||1.00||<0.0001|
|Yes||61 (27.2)||163 (72.8)||5.42 (3.30–8.90)|
|1–3 (mild) (referent)||17 (68)||8 (32)||1.00||<0.0001|
|4–7 (moderate)||20 (29.4)||48 (70.6)||5.10 (1.90–13.7)|
|8–10 (severe)||9 (7.9)||105 (92.1)||24.80 (8.41–73.1)|
|Time of day|
|Day shift (referent)||104 (40.5)||153 (59.5)||1.00||0.88|
|Night shift||30 (39.5)||46 (60.5)||1.04 (0.62–1.76)|
|Treating paramedic clinical level|
|Paramedic trainee (referent)||10 (27.8)||26 (72.2)||1.00||0.47|
|Paramedic intern||12 (44.4)||15 (55.6)||0.48 (0.17–1.38)|
|Paramedic qualified||32 (40.5)||47 (59.5)||0.57 (0.24–1.33)|
|Paramedic specialist||21 (34.4)||40 (65.6)||0.73 (0.30–1.80)|
Table 4. Multivariate Logistic Regression Analysis Modeling the Adjusted Odds of Receiving Analgesia From Paramedics for Older Patients With Suspected Fractures Following Falls
|1–3 (mild) (referent)||1.0||<0.0001|
|4–7 (moderate)||2.70 (1.17–6.32)|
|8–10 (severe)||31.10 (9.9–97.6)|
|Anatomical site of fracture|
|Hip ||2.72 (1.17–6.32)|
The study findings indicate that in this cohort, the majority of older people who had fallen and sustained suspected fractures received OOH analgesia. However, with two in every five patients not receiving pain relief, scope for improvement exists to optimize patient care.
Achieving optimal levels of pain management for patients of any age is challenging, and oligoanalgesia has frequently been reported in EMS and emergency department (ED) settings.[8, 9, 16-20] The analgesia rate in this study of 60% overall and 67% for suspected hip fractures is, however, higher than those previously reported in the emergency care literature for older patients with suspected or confirmed musculoskeletal trauma. In 2002, Vassiliadis et al. retrospectively reviewed medical records of 128 older patients with fractured neck of femur transported by ambulance to a single ED in Sydney, Australia. They reported an EMS analgesic rate of 51% and found that EMS administration of pain relief was associated with a higher triage category and a faster time to ED analgesia. More recently, in a 2010 retrospective audit of Australian ED medical charts involving 646 randomly selected older patients with confirmed diagnoses of hip fractures, Holdgate et al. reported a rate of OOH analgesia of 56.2% (including analgesia self-administrated by the patient). In the United States, McEachin et al. found that only 18% of older patients with predominantly femoral neck fractures received paramedic-initiated analgesia.
In comparison to the rate reported by Vassiliadis et al., the higher proportion of patients receiving analgesia in this study could be the result of significant enhancements to analgesic capability of paramedics (both studies were conducted in the same jurisdiction). At the time of the study by Vassiliadis et al., paramedics were equipped primarily with methoxyflurane and nitrous oxide. Morphine was available, but restricted to a subset of more highly trained paramedics, resulting in only 5% of patients receiving morphine in that cohort. Currently in NSW, all qualified paramedics are authorized to use parenteral morphine, intranasal fentanyl, and inhaled methoxyflurane. When those agents are not appropriate, simple analgesics such as paracetamol and the nonsteroidal anti-inflammatory ibuprofen are available. This range of analgesics enables paramedics to select the most appropriate analgesic for each patient encounter, theoretically increasing the number of patients to whom some type of analgesia can be administered. For example, a patient reporting an allergy to morphine could be given methoxyflurane, and an anticoagulated patient who had been warned by his or her physician to avoid ibuprofen could be given fentanyl. This liberal, multiagent approach to prehospital pain management is representative of the aggressive philosophy toward pain management that is prevalent among most Australasian ambulance services. This approach may be a key contributor to the comparatively high overall rates of administration that have been reported in Australia; Jennings et al. reported an overall administration rate of 51% for patients presenting with pain of any severity or etiology, while Lord et al. found 55% of patients received analgesia. These rates are comparatively higher than those reported in North America overall (17%), and more specifically for limb injuries, for which rates range from 2% to 29%.[10, 16, 24]
The most commonly used analgesic for this cohort of patients was morphine, followed by fentanyl. This is consistent with an earlier EMS study, in which older patients had twice the odds of receiving an opiate for acute pain compared to younger patients and more broadly with analgesic practice in the ED.[8, 10, 21] Administration of parenteral opioids to older patients is not without potential for complications. There is a greater risk of adverse effects and side effects when used in geriatric patients due to changes in pharmacokinetics and pharmacodynamics arising from the normal process of aging. Underdosing of opiate analgesia by clinicians concerned about the possibility of such problems has been associated with an increased risk of delirium in older patients with acute pain secondary to hip fracture, while oligoanalgesia more broadly in this population has been associated with increased risk of inducement of chronic pain states of long duration. Gupta and Avram argue that opioids can be safely used in older patients if rational adjustments for age are made when determining the dose and that fear of adverse effects should not outweigh the benefits arising from provision of effective analgesia.
Increasing the use of orally administered simple analgesics with or without codeine could reduce the reliance on parenteral opiates and increase the proportion of older people with suspected fractures who receive pain relief. The apparent underuse of oral analgesics in this study population might be explained by their recent addition (introduced in 2010) to the scope of practice for general pain management by paramedics, so these drugs may have not become accepted yet as a viable alternative to parenteral opioids. However, as most patients presented with moderate to severe pain in this cohort, it is also possible that paramedics may have deemed parenteral opiates to be more appropriate to provide relief of pain in a timely manner. Codeine-supplemented oral analgesics were only introduced to paramedic practice in 2009 and are limited for use by a very small group of specialist extended care paramedics, few of whom were involved in this study. Currently there is no literature describing the administration of these medications for acute pain by paramedics, so research investigating their utility and efficacy in the OOH setting is warranted.
Another novel alternative for paramedics to reduce the reliance on opioids could be the introduction of regional anesthesia for hip and lower limb fractures. Nerve blocks are well described in the emergency medicine literature, but appear to be underused in the ED[25-28] and rarely used in the OOH context. A recently completed, but as-yet-unpublished clinical trial (n = 25) investigated the use of paramedic-performed fascia–iliaca compartment block for femoral trauma. Enrolled patients had a median age of 81 years (IQR = 70 to 85 years) and predominantly had fractures of the hip. When compared to IV morphine alone, the blocks provided significantly greater pain relief and resulted in no adverse events when performed in the field by intensive care paramedics. This approach to analgesia shows great promise when performed by paramedics in the OOH setting and could provide meaningful clinical benefits to older patients, although more research in this area is required.
In this study, administration of analgesia was not influenced by patient age, sex, residential status, English-speaking status, time of day, or clinical level of paramedic. Patients with hip fractures were more likely to receive pain relief compared to those with fractures of other anatomical sites, a finding that is in contrast to a U.S. study of predominantly older patients with suspected fractures of the lower extremities. Pain severity was strongly predictive of provision of analgesia. In those for whom pain scores were recorded, most had moderate to severe pain. Pain scores were not recorded for one-third of patients, a common phenomenon in emergency care settings.[14, 31-34] In this cohort, analgesia was still administered to almost one-third of patients with suspected fractures but for whom no pain scores were recorded. This suggests that assessment of pain by paramedics is multifaceted and supports previous qualitative research indicating that quantifying pain severity using a score or scale may be a minor component of a more holistic process culminating in the decision to administer or withhold analgesia.[35, 36] Determining pain severity using pain scores can be a challenging, and at times a frustrating task, particularly for older patients with cognitive impairment such as dementia. Educating paramedics in assessment of pain severity using alternative measurement tools might serve to increase the proportion of older patients in whom pain severity is quantified and analgesia subsequently provided; however, the utility and validity of pain assessment alternatives in the OOH setting remains unclear.[2, 11] The influence of pain scores on the decision of whether to provide analgesia by paramedics is worthy of further investigation that could focus on attitudes and perceptions towards patient-reported pain measurement methods.
There are several limitations that should be kept in mind when considering the results from this study. The observational study from which this population was extracted used a convenience sampling approach. A consecutive patient enrollment was not logistically feasible due to the statewide service in which the study was conducted and reliance on paramedics to elect to participate. The nonconsecutive nature of the enrollment in the original study could introduce a selection bias that has the potential to confound the results of this study, as it is possible that participating paramedics could have purposely chosen to include or exclude certain patients in a nonuniform manner. The patients in this study were suspected of having fractures by paramedics, but radiographic confirmation of fractures was not established, as it was beyond the scope of the project. Paramedics can only suspect the presence of fractures, and our focus pertained to establishing analgesic practice based on the clinical information paramedics had available to them at the point of care. Further to this, we selected patients with suspected fractures under an assumption that these are serious conditions that would in most instances produce discomfort and pain. This was a surrogate measure for pain, given that previous research suggested pain scores were frequently not documented in this service. We were also interested in describing the proportion of older patients for whom pain scores were recorded and the association of pain scores with provision of analgesia, so restricting the population to only those with documented pain would have removed our ability to investigate that area of practice. Finally, our findings relating to effectiveness of the analgesia provided are limited by the large amount of cases for which both initial and final pain scores was recorded (48%), and by the small numbers of administrations for four of the seven analgesics available, which prevented exploring effectiveness of individual agents. The effectiveness of the three main agents (morphine, fentanyl, and methoxyflurane) has, however, previously been described for adults in the same ambulance service. It is also conceivable that higher rate of analgesic provision seen in this study could be the result of altered clinician behavior secondary to participating in an observational study (the “Hawthorne effect”). The potential for the Hawthorne effect to bias observational studies has been widely described and has been shown to produce changes in paramedic behavior in the prehospital context. An attempt was made to limit the effect of this effect by deliberately keeping the participating paramedics naive to the specific research questions that would later constitute specific, planned subanalyses. Therefore, paramedics were not aware that analgesic provision was a key focus of interest until after completion of the data collection period. While we could not compare analgesic behavior in this study with a retrospective cohort, a comparison of the transport rate (a key area of paramedic practice in this context) in this study to that found in an earlier, population-based retrospective study in the same service revealed a very similar pattern, suggesting minimal effect on paramedic behavior arising from the observational study design.
In this cohort of older people who had fallen and were suspected by paramedics of having fractures, two-thirds received some type of analgesia, a rate substantially higher than previously reported in emergency settings. Parenteral opioids, in particular morphine, were most commonly administered. Hip injuries and severe pain were associated with greater likelihood of receiving analgesia, although pain scores were not documented in one-third of patients.