Frailty Assessment in an Aging Mouse Model

Abstract

Frailty comprises increased susceptibility to adverse health outcomes associated with old age. Increasing frailty limits independence and can predispose individuals to added co-morbidities and higher mortality. Various tools have been developed to measure frailty in animal models. Phenotyping approaches focus on physical function and include tests for walking speed, strength, endurance, and physical activity. Frailty indices have been proposed based on either invasive or exclusively non-invasive measurements of systemic and functional deficits. The goal of the present protocol is to implement a simple frailty index for non-invasive monitoring of an aging mouse colony adapted from previous methodologies. Based on prior reports and in consultation with institutional veterinary staff, a list of specific deficits of interest was selected and combined into a standardized data collection instrument. The proposed frailty index evaluates integument, musculoskeletal, neuromuscular, sensory, urogenital, and respiratory system deficits and body weight. Each deficit was scored as 0, 0.5, or 1 depending on severity, with the average of all deficits being the frailty index. A naturally aging mouse colony aged 12 to 30 months was observed for approximately 6 months, assessed using the present frailty measurement tool every other week. Frailty index increased over time in both males and females, though males acquired select deficits earlier and faster than females. Certain deficits (e.g., kyphosis, alopecia, poor body condition score) occurred more frequently while most others were less common. Large increases in frailty index often accompany decreases in body weight, particularly in advanced age. The frailty index described in this protocol was easy to use for longitudinal assessments in mice with deficits observable across bodily systems, without the need for invasive measures, specialized equipment, or considerable time or resource commitment. This frailty index does not assess all dimensions of frailty, which should be considered in the context of desired research outcomes.

SUMMARY:

The goal of the present protocol is to provide instructions and guidelines for assessing physical frailty in C57Bl/6x129J laboratory mice. Specific details regarding each measure used in the physical frailty index are explained. Several methods for reporting frailty results are also presented.

INTRODUCTION:

Frailty comprises increased susceptibility to adverse health outcomes typically associated with advanced age^1,2,3. It presents as decreased resilience to stressors and loss of normal physiological function, potentially limiting independence and predisposing older individuals to co-morbidities and mortality^3,4,5. Frailty may help explain the differences in health outcomes commonly seen between those of the same chronological age⁶. In humans, frailty assessments include the frailty phenotype^7,8 and the frailty index^9,10,11, which assess physical performance (e.g., weakness, endurance, or weight loss) and the number of deficits across organ systems, respectively. Animal studies have incorporated a five physical criteria frailty score as well as a 31-item frailty index^12,13,20. Both approaches have their strengths and weaknesses. Frailty phenotyping allows excellent characterization of decreased physical performance, of value for projects interested in sarcopenia, neuromotor dysfunction, or any other aging-related change in physical function. Phenotyping is non-invasive and non-terminal. However, this method of quantifying frailty requires specialized equipment (e.g., rodent treadmills, rotarods, running wheels, and grip meters) and is very time consuming. A frailty phenotype focusing on physical function lacks the robustness of indices that consider frailty across various body systems. Accordingly, frailty indices developed to assess deficits across multiple systems create a more comprehensive rating of frailty. They are also generally non-invasive, easy to administer, and do not require specialized equipment or extensive time. Frailty indices based on deficit accumulation do not readily assess cognitive function and introduce subjectivity in scoring deficits, particularly if multiple evaluators are used, though the effects of this subjectivity can be minimized through discussing and refining assessment techniques²².

The goal of the present method is to quantify frailty using a simple, practical tool, allowing for frailty to be added as a longitudinal outcome to existing and future projects. Multiple considerations were made when developing a frailty measure to fit this goal, one being the reliance on specialized equipment. Several frailty phenotyping methods have been proposed but they require equipment that may be inaccessible to some or impractical for frequent use. For example, open-field monitoring is used in some frailty phenotyping studies to measure exploratory behaviors and voluntary physical activity^14,15. Phenotyping by open-field monitoring requires both specialized camera and computer equipment and considerable time commitment as the mice individually undergo the acclimation and observation periods. Other phenotypes require the use of specialized equipment like rodent treadmills and Rotarods^13,16,17. The demands that a suite of tests including these and other measures places on animal subjects are too great for the frequent, repeat tests needed in a longitudinal study design, not to mention the animal acclimation and investigator training required for each. So, while not aligned with our goal, frailty phenotyping and the common tests used can be applicable to experiments requiring cross-sectional and infrequent longitudinal observation.

Frailty indices are generally non-invasive, are practical to integrate into projects, and can be time efficient^12,30. Using an index also allows for details to be drawn from existing methodologies that have been shown to be reproducible and reliable^22,31. In collaboration with multiple veterinary staff with experience assessing mouse frailty, an abbreviated list of criteria/deficits was selected from a larger list¹², including criteria of the integument (alopecia, loss of fur color, dermatitis, coat condition), musculoskeletal (kyphosis, tail stiffening, tumors, distended abdomen, body condition score), neuromuscular (gait disorders, tremor), sensory (cataracts, eye discharge/swelling), urogenital (rectal and vaginal/penile prolapse), and respiratory (breathing rate/depth) systems. These deficits were chosen specifically as they cover a comprehensive range of systems and can be evaluated without equipment or specialized examinations beyond open-field and manual examination. The specific list these are drawn from has been clinically validated¹² and confirmed to have high inter-rater reliability as found by multiple laboratory groups^22,31. This was an important consideration given the subjective nature of many of the chosen deficits. Reasons for deficits to be excluded from the abbreviated index included requirement of some form of equipment (grip strength, temperature) specialization beyond the desired scope of this simple index (loss of whiskers, vestibular disturbance, corneal opacity, vision and hearing loss, microphthalmia, nasal discharge, piloerection), and avoidance of institutional humane endpoints (diarrhea, malocclusions, mouse grimace scale). It is important to note that the present protocol describes one such way a larger frailty index can be adapted to fit the goals of a given experiment, serving as a guide for investigators interested in simplified or customized tools for assessing mouse frailty. Body weight was also included in the assessment but not in the estimation of the frailty index as conflicting reports suggest variable impact of both weight gain and loss on frailty^12,18,19,20. The purpose of shortening the list of deficits used in determining the frailty index was to highlight the flexibility and adaptability of such an index while simplify the measurement tool to the selected criteria and retaining the comprehensive assessment of frailty in C57Bl/6×129J mice. This change allows for a complete assessment of frailty without equipment-intensive methods or complicated examination. With some training, investigators can expect this comprehensive assessment of physical frailty to take less than two minutes for a single mouse compared to nearly four minutes for a longer, 31-item index¹².

PROTOCOL:

All mice (C57Bl/6×129J) were bred at colonies maintained at the Mayo Clinic. Mice were group housed with littermates by sex maintained on a 12 h light–dark schedule under specific pathogen-free conditions with ad libitum access to food and water. A convenience sample of mice (n = 46; 36 females) aged 12 to 30 months were evaluated for frailty every 2 weeks for eight months. All protocols and animal care guidelines were approved by the Institutional Animal Care and Use Committee at the Mayo Clinic (protocol A00003349, approved 5 October 2023), in compliance with National Institute of Health Guidelines. Criteria for euthanasia for humane endpoints included weight loss greater than or equal to 20% of body weight, inability to ambulate, inability to reach food and/or water, and tumors including single tumors measuring 2 cm at the longest point of the tumor, multiple tumors with cumulative tumor size (the sum of the longest point of each tumor) equaling 2 cm, or tumor(s) that ulcerate or lose the integrity of the skin covering the tumor for any reason.

1. Setup

NOTE: Ensure a clean and hazard-free workspace for handling mice during the duration of the physical frailty assessment and wear proper personal protection equipment.

1.1. Collect the necessary equipment – Scale, notebook or laptop for recording data, mouse cage.

2. Physical frailty assessment

2.1. Divide the assessment into three parts—open field examination, manual examination, and weighing. Complete assessments at consistent location and time-of-day to minimize anxiety- or ambient-related effect on deficits.

2.1.1. Open field examination: Transfer the mouse of interest into an open environment and examine several physical characteristics, including presence of alopecia, loss of fur color, dermatitis, poor coat condition, kyphosis, tail stiffening, gait disorders, tremor, and abnormal breathing.

2.1.2. Manual examination: Gently scruff the mouse to allow the experimenter to further examine physical characteristics, including presence of alopecia, loss of fur color, dermatitis, tumors, distended abdomen, cataracts, eye discharge, rectal prolapse, vaginal/uterine/penile prolapse, and poor body condition score.

NOTE: Because scruffing is inherently stressful, alternative handling methods such as cupping or tube restraint can be used during transfers and some assessments to minimize animal discomfort and better align with institutional policies.

2.1.3. Following open field and manual examinations, weigh the mouse. Zero the scale between each animal and record weights to the nearest 0.1 g. Again, ensure time-of-day consistency of the assessment to minimize natural circadian inconsistencies.

2.2. Following the assessment of each mouse cage, clean the scale and the open field by spraying surfaces with Peroxigard^™ Ready to Use or other appropriate cleaning agent and wiping with paper towel.

3. Examination includes the following 17 characteristics and behaviors that are scored in the assessment of frailty

3.1. Alopecia – signs of hair loss and/or thinning

3.1.1. Assess alopecia during both open field and manual examination. Common areas of hair loss and thinning include the back of the neck/upper back, neck and armpits, stomach. Note that there are natural areas of thinner hair on the underside of mice, especially near joints. Alopecia is only counted if hair loss and thinning is beyond that of natural patterns observed in young, non-frail mice.

3.1.2. Score alopecia as follows: 0 = normal fur density; 0.5 = < 25% of body surface displaying fur loss; 1.0 = > 25% of body surface displaying fur loss

3.2. Loss of fur color – change in fur color, e.g., from black to grey, brown, or white

3.2.1. Assess loss of fur color during both open field and manual examination. Changes in fur color often begin on the underside of the mouse but will progress to whole body changes with progressed frailty.

3.2.2. Score loss of fur color as follows: 0 = normal color; 0.5 = patchy, focal grey, brown, or white changes; 1.0 = grey, brown, or white fur on > 25% of body surface

3.3. Dermatitis – presence of skin lesions

3.3.1. Assess dermatitis during both open field and manual examination. Presence of any skin lesions indicating dermatitis, with most focal lesions found on the neck, flanks, and under the chin.

3.3.2. Score dermatitis as follows: 0 = absent; 0.5 = focal lesions of small size; 1.0 = multifocal or widespread lesions

3.4. Coat condition – signs of poor grooming

3.4.1. Assess coat condition during open field examination. Normal coat condition, usually present in young, non-frail mice, includes a smooth, sleek, shiny coat. Ruffled, matted, and/or ungroomed fur are signs of poor coat condition and often accompany alopecia and changes in fur color.

3.4.2. Score coat condition as follows: 0 = smooth, sleek, shiny coat; 0.5 = slightly ruffled; 1.0 = unkempt and ungroomed, matted appearance

3.5. Kyphosis – presence of curvature of spine or hunched posture

3.5.1. Assess kyphosis during the open field examination. Kyphosis is most often present in the upper back and will present as a clear concave curvature beyond that seen in any young, non-frail mice. To confirm the presence of kyphosis, the experimenter can palpate the upper spine to examine for a kyphotic bend. Permanently hunched posture may follow kyphosis and will be noticeable as a whole-body hunch as opposed to a more local kyphotic bend.

3.5.2. Score kyphosis as follows: 0 = no cervical to thoracic spinal curvature (approximately 180° to 150°); 0.5 = slight cervical to thoracic spinal curvature (approximately 150° to 130°); 1.0 = significant curvature (<130°) and associated hunched posture

3.6. Tail stiffening – lack of curling response to light touch

3.6.1. Assess tail stiffening during open field examination. While the mouse is freely moving, the experimenter will gently stroke the tail with a finger. The tail of a healthy mouse will wrap around the finger when it is touched. This is best observed when the mouse is relaxed.

3.6.2. Score tail stiffening as follows: 0 = none; 0.5 = tail responsive but does not curl; 1.0 = tail unresponsive to touch

3.7. Gait disorders – abnormalities during locomotion

3.7.1. Assess the presence of gait disorders during open field examination. Common changes to the gait include dragging limbs, circling, wobbling, and wide stance. Use climbing ability to assess gait by placing mouse on a large cage lid, tilting the lid and lightly brushing the rear of the mouse to motivate climbing.

3.7.2. Score gait disorders as follows: 0 = none; 0.5 = abnormal gait but mouse can walk; 1.0 = marked gait abnormality that impairs ability to move freely

3.8. Tremor – unprovoked trembling or shaking

3.8.1. Assess tremor during open field observation. The experimenter will monitor mouse during rest and movement for presence of a tremor. This usually presents as paw shaking.

3.8.2. Score tremor as follows: 0 = no tremor; 0.5 = paw shake or tremor during ambulation or high levels of activity; 1.0 = marked tremor with impaired ability to move freely

3.9. Breathing rate/depth – abnormal respiratory effort

3.9.1. Assess breathing rate/depth during open field examination. Any changes from normal, regular breathing at rest as observed in young, non-frail mouse are scored. Usually changes include increased rate and decreased depth of breathing.

3.9.2. Score breathing rate/depth as follows: 0 = normal; 0.5 = modest change in rate/depth; 1.0 = marked change in rate/depth, or gasping.

3.10. Tumors – presence of abnormal masses

3.10.1 Assess the presence of tumors during manual examination. While scruffing the mouse, the experimenter will examine the mouse for visible tumors and run finger along the underside of the mouse to check for palpable tumors. Visible asymmetry is a sign of potential tumors. The presence of tumors can be confirmed post-mortem if possible.

3.10.2. Score tumors as follows: 0 = none; 0.5 = one tumor of less than 1 cm³ in volume; 1.0 = single tumor greater than 1 cm³ in volume or multiple smaller tumors

3.11. Distended abdomen – presence of excess bulge below rib cage

3.11.1. Assess distended abdomen during manual examination. While scruffing the mouse, the experimenter will assess the presence of excess bulging or abdominal fluid. Presence of a slight bulge below the rib cage to a pronounced W-shaped bulge may be visible in the abdomen and palpable during examination.

3.11.2. Score distended abdomen as follows: 0 = none; 0.5 = slight bulge; 1.0 = clear distension, W- shaped bulge below ribs

3.12. Cataracts – presence of cloudy eyes/white spots

3.12.1. Assess cataracts during manual examination. While scruffed, the examiner will evaluate the mouse’s eyes for clouding of the cornea. Complete clouding of at least one eye is counted as 1.0.

3.12.2. Score cataracts as follows: 0 = none; 0.5 = small opaque spot/changes to one or both corneas; 1.0 = significant clouding/spotting of one of both corneas

3.13. Eye discharge – buildup of bodily fluid around the eyes

3.13.1. Assess eye discharge during manual examination. While scruffed, the examiner will evaluate the mouse’s eyes for evidence of secretions, crusting or discoloration around the edges of the eyes. Discharge is usually dried, associated with swelling, and will result in changes to the normal sleek, black eye color seen on young, non-frail mice.

3.13.2. Score eye discharge as follows: 0 = normal; 0.5 = slight swelling of and/or secretions around one or other eyes; 1.0 = significant swelling of and/or secretions around one or both eyes

3.14. Rectal prolapse – abnormal extrusion of the rectum

3.14.1. Assess rectal prolapse during manual examination. While scruffed, examine for signs of prolapsed tissue in the absence of evident straining. Any rectal tissue prolapsed out of anus is scored.

3.14.2. Score rectal prolapse as follows: 0 = no prolapse; 0.5 = some prolapsed tissue, healthy looking; 1.0 = significant prolapsed tissue, unhealthy tissue

3.15. Vaginal/uterine/penile prolapse – abnormal extrusion of genitalia

3.15.1. Assess vaginal/uterine/penile prolapse during manual examination. While scruffed, examine for signs of prolapsed tissue in the absence of evident straining. Any genitalia prolapse is scored.

3.15.2. Score vaginal/uterine/penile prolapse as follows: 0 = no prolapse; 0.5 some prolapsed tissue, healthy looking; 1.0 = significant prolapsed tissue, unhealthy tissue

NOTE: Prolapsed can occur due to distress from handling, so if a deficit is suspected, place the mouse back onto the open field, lifting the tail to confirm the presence of prolapsed tissue.

3.16. Body condition score – presence of excess or minimal fat and evidence of loss of muscle mass

3.16.1. Assess body condition score during manual examination. While scruffing the mouse, take note of the amount of fat covering the lower back and pubic bone. Frailty will usually present a loss of fat in this region. Moderate loss of fat resulting in more prominent bones than young, non-frail mice will be scored as 0.5. Once the bones become very prominent, even sharp, or there is evident loss of muscle mass, score will be 1.0.

3.16.2. Score body condition score as follows: 0 = bones palpable but not prominent; 0.5 = bones prominent and/or barely felt; 1.0 = bones very prominent and/or not felt

3.17. Weight – assessment at consistent time of day (e.g., within 2 to 4 hours of beginning of inactive period) following the open-field and manual examination.

3.17.1. Assess weight following frailty measures.

4. Implementing the Physical Frailty Assessment

NOTE: Much of the implementation of this assessment should be customized to the needs of the experimenter and their experimental goals. Details provided here can serve as an example of how one might implement the frailty assessment and provide some helpful guidance for general purposes.

4.1. Frequency - Once a cohort or colony of mice to be assessed is identified, make physical frailty assessments every two weeks. Keep the time and day of the assessments consistent during the evaluation period.

4.2. Data Recording – During the evaluation of each mouse, assess a score (0, 0.5, or 1.0) to each of the frailty measures and record a weight. An average of the scores of all sixteen deficits is used as a final frailty score. Take any relevant notes regarding the health and behavior of the mice, including changes to any of the measures that will require increased attention during subsequent evaluations and abnormal posthandling behaviors. It is vital to ensure the correct data is being ascribed to the correct mouse over the course of the longitudinal evaluation period. Perform no rescoring for prior scores.

REPRESENTATIVE RESULTS:

A convenience sample of naturally aging C57Bl/6×129J mice were assessed for frailty using the index described here. Each mouse (n = 46; 36 females) over 12 months of age was assessed every two weeks, with the evaluation period lasting approximately 6 months. A cross-sectional cohort at four ages of interest (12, 18, 24, and 30 months) was selected to assess frailty index scoring across age groups representing 100%, 95%, 80% and 30% survival, respectively²¹. There was a consistent upward trend in frailty as age progressed (Figure 1). The 18- and 24-month groups were also used to examine the frequency of deficits across age and sex (Figure 2). Notably, only ten of the sixteen deficits assessed were present in the specific mice at the time points of interest, with the six not present being dermatitis, tail stiffening, breathing rate/depth, tumors, cataracts, and vaginal/uterine/penile prolapse. Alopecia, kyphosis, and loss of fur color were the most common deficits at 18 months, with body condition score and poor coat condition deficits becoming similarly common at 24 months. Integumentary deficits such as loss of fur color and poor coat condition occurred frequently in male mice while not seeming to affect females. Body condition score was one deficit that increased frequency with age in both males and females, highlighted well when its progression is independently tracked within the colony (Figure 3A). Alopecia is another common deficit but with a different progression across sexes: female mice show some progression towards increased alopecia with age, but not as striking as that seen with body condition score, while males show early evidence of alopecia but minimal progression once it is acquired (Figure 3B).

Figure 1: Frailty index across age.

Single frailty assessments of animals at ages 12, 18, 24, and 30 were selected from the larger colony, with no animals included in multiple age groups. Only older aged data were included for animals present in multiple age groups. Quartile box plots are shown, with medians marked by clear line and maximum and minimum values shown with whiskers.

Figure 2: Deficit frequency according to age and sex.

The frequency of all frailty index deficits was compared for animals in 18- and 24-month age groups. Deficits not shown in this figure were assessed but were not present at the measurement time point for these animals. This included dermatitis, tail stiffening, breathing rate/depth, tumors, cataracts, and vaginal/uterine/penile prolapse. (18 months: male=4, female=7; 24 months: male=6, female=8)

Figure 3: Changes in body condition score deficit and alopecia with age and sex.

(A) Presence of Body Condition Score deficit increases with age. Each deficit can be separated from the overall frailty index to be individually examined. Body Condition Score, for example, shows progression from mostly 0 at adult ages to 0.5 in old adults and 1.0 in advanced age in both male and female groups. (B) Presence and progression of Alopecia varies with sex. When separated from the overall frailty index, differences in the acquisition and progression of alopecia can be observed between sexes. While females tend to acquire mild alopecia later in life, several show signs of hair loss much earlier while others reach advanced age with no such signs. Males have mostly developed alopecia prior to reaching 12 months of age and show no progression to severe alopecia.

Whole colony longitudinal frailty index data clearly illustrates the progression of frailty with age. Within our colony, males had both a higher age-matched frailty score and faster frailty progression with time. A lack of males over 120 weeks of age also suggested a lower survival rate in males within the colony (Figure 4). Two-week weight change was also assessed longitudinally, with most mice never experiencing a weight change of ± 5%. The mice that experienced the largest changes in weight tended to lose weight at advanced age and displayed higher frailty index (>0.150) (Figure 5A). Also of note were the changes in frailty index as they related to weight change. For most mice, two-week frailty change stayed between ± 0.05, which may represent some level of noise within the measure as certain deficits show a lack of consistent severity progression. As with weight change, the largest increases in frailty were more evident with advanced age and commonly occurred simultaneously with large decreases in weight (Figure 5B).

Figure 4: Whole Colony Frailty Index Progression.

Mice between the ages of 12 and 30 months were assessed for frailty once every two weeks. Shown together, whole colony data suggests a sex difference in the frailty scores and accumulation of new or worsened deficits.

Figure 5: Changes in animal weight across age and frailty score change.

(A) Animal weight changes across age and frailty score. For each frailty score assessed to mice in the colony, weight change across the prior two weeks was calculated. Changes in weight remain minimal throughout most of the colony’s lifespan, with the largest changes occurring in older, frailer mice. Colored circles indicate weight gain, while open circles indicate weight loss. (B) Animal weight changes across age and frailty score change. For each frailty score assessed to mice in the colony, frailty and weight change across the prior two weeks was calculated. Changes in both frailty and weight remain minimal throughout most of the colony’s lifespan, with large changes in frailty and weight often occurring simultaneously. Colored circles indicate weight gain, while open circles indicate weight loss.

Individual frailty index and weight progression were also examined for all mice that received at least 4 assessments during the evaluation period (n = 40; 30 females). Four distinctive phenotypes of frailty weight change emerged and were exemplified with representative animal data (Figure 6). Animals that underwent an increase in frailty and no change in weight were the most common phenotype (46%), with those having unchanged frailty and weight being the next most common (30%). The combinations of increased frailty with either increased or decreased body weight occurred with similar frequency (12%).

Figure 6: Distinctive patterns of frailty and weight change.

When examining the frailty and weight changes of individual animals, four distinctive patterns (F↑, W↑; F↑, W→; F↑, W↓; F→, W→) were present and are shown here using representative animals.

DISCUSSION:

Critical to the success of this frailty index is consistency when evaluating the many subjective criteria that may be present with deficits. Several steps can be taken to ensure the quality of results and overall reliability of this index. The same investigator can be used during observation, ensuring that the same judgement of deficit score will be given across animals and time points. If experiment design or logistics prevents this consistency, each investigator should independently familiarize themselves with the index before comparing scores and notes to arrive at an agreement for consistently judging deficits similarly. Reliability across investigators using similar frailty indices based on deficit accumulation is well established. Multiple laboratory groups have confirmed inter-rater reliability of the frailty index developed by Whitehead and colleagues¹², from which the deficits used in this simplified index were drawn^22,31. Vital to establishing reliability is refined score descriptions of each criteria²², which have been provided for the 16 criteria chosen for the present index. Accordingly, the assessment of the deficit severity as presented in this manuscript reflects the existing data in support of using subjective measures in calculating frailty indices. Based on the desired features of frailty for specific investigations it is thus possible to customize frailty indices that simplify the evaluation in both cross-sectional and longitudinal aging studies. In the present work, we aim to present the utility of a simplified tool that does not require equipment-intensive methods or complicated examinations and can be accomplished in a very short amount of time. It is also crucial for the investigator(s) to have a healthy, non-frail baseline for each deficit. In other words, to ensure the accuracy of each score given, one must have a good grasp on what a score of 0 looks or feels like. While expected to be different for each colony and strain, mice from ages 9 to 12 months were used to calibrate the investigator’s baseline for the use of this index periodically as mice aged into this group. Similar steps are suggested for those implementing a similar index of deficits into their own projects to ensure a clear understanding of healthy criteria during the evaluation period.

During early implementation of this index and as the investigator(s) familiarity with each criterion is improving, there may be some variation in scores between subsequent ratings that do not reflect true changes. It is important to deduce which criterion are driving this variation and address them promptly as inaccurate ratings stand to jeopardize data collected using this frailty index. One example encountered during our implementation was with rectal prolapse. Early in the evaluation period, this deficit was scored much too liberally as any change from the normal anus color was considered a 0.5. However, given the external appearance of the anus can change merely due to holding the mouse in a scruffed position or recent defecation, scores fluctuated between 0 and 0.5 week-to-week. To address this concern, a set of more accurate scoring directions was made following consultation with murine-specific resources^23,24. Moving forward, deficits related to prolapse would require both changes in tissue appearance and extent to prolapse at rest (e.g. not under clear distress while being scruffed). Healthy looking, minimally prolapsed tissue was scored as 0.5, with inflamed tissue accounting for a 1.0 score. Taking these steps eliminated the fluctuations in rectal prolapse scoring, and the same troubleshooting process was followed for other ratings when fluctuations were explainable by investigator error. Once deficit scoring is clarified, it is vital that it remains consistent longitudinally to avoid potential drift in scoring. We do acknowledge that we did see some fluctuations in the severity of some deficits during the longitudinal evaluation period. One example of this was eye discharge, which occasionally spontaneously recovered without investigator-related explanation. Recovery did not always occur, however, and the presence of the deficit itself was more common at advanced age (26+ months).

Body weight was included for this index, although its place in assessing frailty varies across literature. It is understood that unintentional weight loss is a dimension of frailty in humans^7,25,26. With the recent rise of obesity²⁷, however, a simple check for weight loss may not be as useful to establish frailty clinically. For example, patients with sarcopenic obesity would have the functional hallmarks of frailty (e.g., strength loss, quicker exhaustion) while not presenting with weight loss. It may be the case, then, that human frailty assessments considering only weight loss as a criterion of frailty may miss the presence of frailty because of obesity. While some animal models mirror this treatment of body weight by considering weight loss alone as a component of a frailty phenotype^4,18, others consider weight change with more nuance^12,19. This includes assessing a deficit score to weight based on deviation from young adult animal positively or negatively or considering only significant weight gain to be a sign of frailty²⁰. Due to the lack of consistent handling of body weight as a deficit of interest when calculating frailty, it is not directly involved in the frailty score of this protocol. Because the two were kept independent, we were able to assess frailty and weight change separately, finding there was no consistent trend of weight change associated with increases in frailty, supporting our decision to exclude weight change as a frailty index criterion. We did observe that the largest decreases in weight were often associated with old age and high frailty. Having body weight separate from the frailty index allows investigators to establish the relationship between weight and frailty within their own cohorts or colonies.

Frailty indices are a uniquely flexibly method for assessing frailty. The present frailty index can be modified in accordance with the goals of the investigation, as it includes criteria across six body systems, allowing a broad assessment of physical frailty but it does not comprise all dimensions of frailty. This limited scope leads to a lack of consideration of physical function beyond gait and breathing, cognitive ability, social behavior, and physiological changes (e.g., blood pressure, body fat percentage). Should criteria like these be of interest, investigators can easily tailor this index to include additional body systems or physical functions as needed. There are several other tools for measuring frailty that aim to be comprehensive and can be used to draw from in creating an index with the desired level of specificity and complexity^19,28,29. It should also be noted that this index was used in C57Bl/6×129J mice specifically, and thus the criteria described here may not be relevant when measuring frailty in other mice strains (e.g., with light-colored fur). Investigators can modify a frailty index to best suit the known signs of aging in particular mouse stains, including using deficits of the same system in the place of those not scorable in certain strains.

The goal of this protocol is not to provide a direct comparison to existing methods for assessing frailty but instead suggest a more accessible alternative while providing guidance for investigators interested in implementing a simplified frailty index based on deficit accumulation. As such, the individual scores themselves presented as results here are not directly comparable to those from other frailty assessments as the quantity and type of deficits examined differ from existing methods. These differences are important to note, however, as the simplicity and accessibility of this frailty index separates it from many of the existing methods. Assessments can be made quickly without any special equipment and without excess stress placed on the animals. Once proficient, investigators can expect to have a full assessment of each animal to take no more than two minutes, allowing for widespread implementation. This frailty assessment tool allows investigators to focus experiments on changes in frailty and develop interventions to slow or reverse age-related increases in frailty. Many of the current preclinical methods to assess frailty have requirements that make them difficult to easily implement, because of the need for specialized equipment, invasive measures, or considerable time commitments. The index discussed here gives investigators a simple, time efficient alternative to assess frailty.