Body composition and nutritional status in nursing home residents during the COVID-19 lockdown: a 15-month follow-up

A detailed description of participants, recruitment, and measurements has been reported in a previous article that included six nursing homes in Zagreb, Croatia (12). The inclusion criteria were the ability to stand/walk independently (with or without aid) and to give informed consent. The exclusion criteria were a history of a serious chronic disease (severe mental/psychiatric disorder, epilepsy, cancer, Parkinson’s disease, stroke, severe dementia, significant cognitive impairment) and health conditions that prevented bioimpedance measurement (e.g., pacemaker, amputation). This study included a total of 24 nursing home residents (22 of whom were women), who were followed for 15 months. Seven (six women) had COVID-19, confirmed by real-time polymerase chain reaction (PCR). All of them recovered without serious consequences.

All measurements were taken and the questionnaires administered in the nursing home in the morning. Body weight and height were measured in indoor clothing without shoes to calculate body mass index (BMI) (kg/m²). Body composition was assessed with a bioelectrical impedance device BIA-ACC^® (BioTekna, Marcon, Venice, Italy) in a supine position with legs slightly spread apart and arms not touching the body. Two electrodes were placed on the right hand (metacarpal and wrist area) and two on the right leg (metatarsal and ankle area). The measured parameters were fat mass (FM; kg; % of body weight); abdominal adipose tissue (AAT; cm²); fat-free mass (FFM; kg; % of body weight); skeletal muscle mass (SM; kg; % of FFM; yielding the S-score), and total bone mass (BM; kg, yielding the T-score). OSA prevalence was calculated based on the criteria used previously (13), including the T-score of ≤-1.0 for the identification of osteopenia/osteoporosis, the S-score of ≤-1.0 for identification of sarcopenia, and total fat mass ≥25 % for men and ≥32 % for women.

All study procedures followed the 1964 Helsinki declaration and its later amendments or comparable ethical standards. The study was approved by the Ethics Committee of the Institute for Medical Research and Occupational Health (approval No. 100-21/18-10), and all participants signed informed consent at the beginning of the study.

The mean age of the participants was 83.3 years at baseline. Eighteen had hypertension, six had chronic heart diseases, and one participant had hypothyroidism. No participants had particular eating habits (dietary requirements). As for the baseline BMI, eight participants had normal weight, seven were overweight, and nine were obese. At the follow-up, seven had normal weight, six were overweight, and 11 were obese.

Table 1 shows mean values of all body composition parameters in both measurements. At baseline, mean values of BMI, FM %, and AAT were higher than normal, while skeletal muscle (SM %, S-score) and bone mass (BM, T-score) were below the reference values. At follow-up, BMI remained unchanged and FM % slightly increased, while skeletal muscle mass decreased (p=0.008 for SM %; p=0.036 for S-score), as did the bone mass (p=0.014 for BM; p=0.005 for T-score). OSA was determined in 14 participants at baseline and in 16 at follow-up (Figure 1).

Figure 1

Prevalence of osteosarcopenic adiposity (OSA) in all participants at baseline and 15 months into the first COVID-19 lockdown

There were no significant differences in body composition between participants who had had COVID-19 and those who had not had it either at the baseline or the follow-up (results not presented). In participants who had had COVID-19 we observed a non-significant decrease in bone, lean, and fat tissue between two measurements. In contrast, participants who had not had COVID-19 experienced a significant increase in FM % (p=0.046), while bone and lean tissue values significantly decreased (p=0.038 for SM %; p=0.027 for BM and p=0.025 for the T-score) (results not presented).

Table 2 shows that, according to the MNA-SF questionnaire, 18 participants had normal nutritional status, far more than those who ran the risk of or suffered malnutrition (p=0.045). Seventeen participants reported no decline in food intake in the previous three months. However, eight reported weight loss (between 1–3 kg or >3 kg) at follow-up, but according to objective body mass measurement, all eight gained weight by that time. This discrepancy is probably the result of false body weight self-perception, possibly due to relatively infrequent weight control. A large majority reported good mobility, being able to go outside the nursing home, having three meals per day and consuming dairy products and meat/fish every day. Most consumed 3–5 cups of fluid per day. We found no significant differences in body composition between participants who had normal nutritional status and those who ran the risk of malnutrition or were malnourished.

During the 15-month lockdown, the nursing home residents in our study lost bone and lean tissue and gained fat tissue. The prevalence of OSA, as the most unfavourable change in body composition, was relatively high at the baseline and slightly increased after 15 months. These changes in body composition occurred even though weight and/or BMI did not change, which is in line with other reports (16, 17). In fact, they are most likely owed to the adverse trajectories of ageing (1). However, that may also have been heightened by the lockdown and restricted living conditions in nursing homes, as indicated by noticeable differences in the pattern of change in body composition between the participants who had had COVID-19 and those who had not had it, since participants with COVID-19 experienced a drop in bone, muscle, and fat tissue. Therefore, our results point to the importance of body composition for overall health and its quick responses, not just to ageing but to other health and living conditions, which were markedly changed in the first year of lockdown, regardless of whether the residents had had COVID-19 or not.

At the follow-up, OSA was found in two new participants, and we believe that in the sample of 24 subjects, this increase is not negligible. To our knowledge, there is no longitudinal follow-up study on OSA, particularly among nursing home residents, to which we can directly compare our results. In one follow-up study conducted in a nursing home in Australia, residents (mean age >80 years) who did not have sarcopenia at baseline had a very slight decrease in skeletal muscle index (kg/m²), and those who had sarcopenia at baseline showed even a slight increase (18). Some other studies in older community-dwelling adults of >65 years determined OSA prevalence of 10–30 % (19, 20). In the most recent epidemiological study (21), the OSA prevalence, determined by bioimpedance measurement in apparently heathy, communitydwelling adults of over 80 years of age was above 76 %, which is much higher than in our participants.

However, when comparing the results, different criteria for the diagnosis of OSA should be considered along with different techniques and instruments. The greatest differences are usually seen in the diagnosis of sarcopenia, which can be based on muscle mass alone, like in our study, but also on muscle mass combined with grip strength (22). These differences in the definition of sarcopenia may produce differences in the diagnosis of OSA, which eventually may result in discrepancies in OSA prevalence between studies.

It is worth noting that mean the S- and T-scores were relatively good in our participants and similar to the results obtained on a larger sample of nursing homes residents (4).

As for nutritional status, it was adequate in most participants. Most consumed enough food rich in proteins and had an adequate intake of fluids. We did not assess the intake of specific nutrients, so it is possible that some participants were deficient in some minerals and vitamins (23). Additionally, nearly all of the participants had good mobility (not bed-bound), and all were able to feed themselves. As we witnessed conducting this study in the nursing home, all necessary health measures were put in place when the pandemic started, which is very commendable. However, the basic organisation and permanent care for residents before the pandemic might be of even greater importance, which seems to have had been in place there as well.

The major limitation of our study is the relatively small number of participants, particularly of those with COVID-19. However, by using appropriate statistical methods, suitable for samples of less than 30 participants, we were able to show that negative changes in all body composition parameters are probably owed to ageing. The study was conducted in one of the ten city or state-owned nursing homes in Zagreb providing uniform care for their residents, including nutrition. Therefore, we believe that our sample, although small in number, is representative of public nursing home users.

Regardless of its limitation, our study has identified a different trend of changes in body composition in participants who had had COVID-19 and those who had not had it, since the participants who had had COVID-19 experienced a drop in all body composition parameters.

As far as we know, this is the only prospective follow-up study on body composition and nutrition in nursing home residents during the 15-month COVID-19 lockdown. The participants, who lived in a nursing home and had adequate nutritional and overall health status, experienced changes in body composition in terms of a decrease in bone and lean tissue and an increase in body fat. These changes can be attributed to the effect of age rather than to changes in the lifestyle during the lockdown, although a larger sample is needed to strengthen this conclusion.

Our findings also confirm the importance of body composition and nutritional status in the health of elderly persons and suggest that healthy body composition and adequate nutritional and health support in nursing homes can contribute to recovery from unfavourable circumstances like the pandemic lockdown and in preventing its consequences. Furthermore, this study provides arguments in favour of regular body composition and nutritional status monitoring to help to preserve the health of nursing home residents, and particularly in cases of possible future lockdowns.