How Typical is the Spectrum of COVID-19 in Nursing Home Residents?

To the Editor:

Older persons living in long-term care facilities are underrepresented in studies on the clinical spectrum of Coronavirus Disease 2019 (COVID-19), especially regarding the broad range of COVID-19 signs/symptoms and course over time. In the context of advance care planning and, in the Netherlands, older care physicians providing nursing home care, diagnostics, and optimal supportive care are mostly provided within the nursing home. In this setting, COVID-19 testing is dependent on either signaling signs/symptoms via medical history, observation, and physical examination, or known contact with a confirmed case. However, an atypical disease presentation and course, as signaled by initial studies in hospitalized older persons, may hamper identification of COVID-19 cases.1, 2, 3 Our aim was to gain insight into the broad spectrum of signs/symptoms, disease course, and outcome in nursing home residents with COVID-19.

Methods

We performed a retrospective cohort study among residents with confirmed COVID-19 in the period March to April 2020 at 4 long-term care organizations in the Netherlands (see Supplementary Methods). Electronic health records were searched for demographics; comorbidity; 22 signs/symptoms, including clinical criteria of the World Health Organization (WHO) case definitions from March and August 2020; , dates on first registration; decrease in or full recovery from signs/symptoms; and disease outcome. For registered signs/symptoms, we assessed prevalence at presentation; period prevalence; and time to onset, decrease, and full recovery. We explored differences in characteristics between deceased and recovered residents.

Results

In total, 88 of 94 eligible residents were included (see Supplementary Table 1).

Supplementary Table 1

Characteristics of the Included COVID-19–Positive Nursing Home Residents (n = 88) and Disease Outcome

Characteristic	All	Recovered (Fully or Partially)	Deceased
	(n = 88)	(n = 56)	(n = 32)
Age (y), median (min. – max.)	83.5 (65 – 97)	83.5 (65 – 97)	83.5 (69 – 97)
Sex (male), n (%)∗	24 (27.3)	11 (19.6)	13 (40.6)
BMI, median (min. – max.)	25.1 (17.8 – 43.5)	25.3 (17.8 – 43.5)	24.7 (20.1 – 33.4)
Obesity (BMI ≥ 30 kg/m2), n (%)	12 (19.7)	10 (25.0)	2 (9.1)
Type of stay, n (%)
Long-term stay	74 (84.1)	46 (82.1)	28 (87.5)
Geriatric rehabilitation	14 (15.9)	10 (17.9)	4 (12.5)
Chronic conditions, n (%)
COPD	13 (14.8)	7 (12.5)	6 (18.8)
Asthma	3 (3.4)	2 (3.6)	1 (3.1)
Coronary heart disease	44 (50.0)	24 (42.9)	20 (62.5)
Heart failure	17 (19.3)	12 (21.4)	5 (15.6)
Stroke	20 (22.7)	13 (23.2)	7 (21.9)
Hypertension	51 (58.0)	33 (58.9)	18 (56.3)
Diabetes mellitus	20 (22.7)	13 (23.2)	7 (21.9)
Cancer, excluding nonmelanoma skin cancer	15 (17.0)	12 (21.4)	3 (9.4)
Chronic liver disease	2 (2.3)	2 (3.6)	–
Chronic kidney disease	17 (19.3)	12 (21.4)	5 (15.6)
Dementia∗	50 (56.8)	27 (48.2)	23 (71.9)
Chronic neurological disorder, excluding dementia∗	20 (22.7)	8 (14.3)	12 (37.5)
Current smoking (yes), n (%)†	7 (10.0)	6 (13.0)	1 (4.2)
Mobility before COVID-19, n (%)
Bedridden	1 (1.1)	1 (1.8)	–
In wheelchair	29 (33.0)	17 (30.4)	12 (37.5)
Walking with physical help or supervision	19 (21.6)	10 (17.9)	9 (28.1)
Independent with or without mobility aid	39 (44.3)	28 (50.0)	11 (34.4)

BMI, body mass index; COPD, chronic obstructive pulmonary disease.

Statistically significant difference between deceased residents and fully or partially recovered residents in χ2-test (P < .05).

Values based on the number of residents with available data: data are missing on body mass index for 26 residents and on smoking for 18 residents.

Spectrum of Signs/Symptoms

Fever and respiratory symptoms, especially cough and shortness of breath, were the most frequently registered signs/symptoms, at presentation as well as over the disease course. Nevertheless, only up to 63 residents [71.6%, 95% confidence interval (CI) 62.6%–81.0%] fulfilled the clinical criteria of the initial WHO case definition (March 2020). Using the current definition (August 2020), up to 84 residents (95.5%, 95% CI 91.1%–99.8%) would have fulfilled the criteria (Figure 1 ). Frequently reported signs/symptoms from this list are of a more general nature (up to 83.0% malaise/fatigue; 72.7% loss of appetite/decreased intake) or related to the gastrointestinal tract (46.6% diarrhea; 36.4% nausea/vomiting) or altered mental status (20.5% confusion/delirium; 25.0% behavioral change). Behavioral change includes agitation/wandering (13.6%), mood changes/anxiety (5.7%), and apathy (4.5%). In addition, 61.4% of the residents experienced reduced mobility (20.5% unstable walking and/or falling; 40.9% becoming bedridden).

Fig. 1

Frequency of reported symptoms on the first day and over the course of COVID-19, that is, the period prevalence. The error bars represent the 95% confidence intervals.

Clinical Course

Cough and malaise/fatigue were mostly present at presentation, that is, after median 0 days [interquartile range (IQR) 0–4] (see Supplementary Table 2). Fever developed mostly within 1 day (IQR 0–4), peaked with median 38.7°C (IQR 38.4–39.2°C, maximum 41.6°C) after 1 day (IQR 0–5) and disappeared after 3 days (IQR 1–9). Other respiratory and general symptoms were observed after median 2 to 3 days and mostly disappeared after 2 to 3 weeks. Gastrointestinal complaints appeared to occur later, after median 4 to 5 days, and to disappear within a week.

Supplementary Table 2

The Course of the 12 Most Frequently (≥20%) Reported Signs and Symptoms, Per Tract∗

Sign/symptom	Day Start		Duration Until Decrease (d)		Duration Until Full Recovery (d)
	n	Median (IQR)	n	Median (IQR)	n	Median (IQR)
Respiratory
Cough	68	0 (0–4)	38	9 (2–16)	34	18 (7–26)
Shortness of breath	52	3 (1–5)	26	7 (1–12)	18	15 (5–21)
Rhinorrhea/nasal congestion	33	2 (0–9)	8	5 (1–8)	8	6 (1–15)
Sore throat	18	2 (0–7)	3	3 (–)†	4	12 (–)†
General
Malaise/fatigue	73	0 (0–4)	34	12 (6–20)	28	17 (10–25)
Fever	67	1 (0–4)	–	Not applicable	51	3 (1–9)
Loss of appetite/decreased intake	64	3 (1–7)	28	8 (3–13)	27	13 (7–19)
Decreased mobility	54	3 (1–6)	17	12 (4–17)	11	17 (3–22)
Gastrointestinal
Diarrhea	41	4 (1–7)	24	3 (1–13)	22	7 (1–19)
Nausea/vomiting	32	5 (2–8)	21	1 (1–12)	19	1 (1–16)
Neurological/behavior
Change in behavior	22	3 (1–5)	10	5 (3–8)	9	8 (4–11)
Confusion/delirium	18	8 (4–10)	8	5 (3–19)	6	11 (3–24)

The durations until decrease and full recovery of signs and symptoms are based on a smaller group of residents, because a group of residents died before decrease and/or full recovery.

No IQR available because of the small number of observations.

Oxygen therapy was started in 49 (55.7%) residents after median 5 days (IQR 3–8), for a period of 9 days (IQR 1–24) and with maximum supplementation after 1.5 days (IQR 0–5). Lowest oxygen saturation during supplementation was median 89% (IQR 84–93).

Disease Outcome

At data collection 32 residents (36.4%, 95% CI 26.3%–46.4%) had died at median 10.5 days (range 6–96) after first signs/symptoms; 30 died of COVID-19 after 6 to 23 days, and 2 from general health decline afterward. Full recovery was registered in 47 (53.4%) residents after median 26 days (range 4–47) and partial recovery in 9 (10.2%) after 50 days (range 39–128) follow-up.

Several signs/symptoms were associated with death, including higher fever and lower oxygen saturations during supplementation: median 38.9°C (IQR 38.4–39.6°C) versus 38.5°C (38.2–38.8°C) and 85% (79%–88%) versus 91% (88%–93%) (see Supplementary Table 3). Male residents and residents with dementia or another chronic neurological disorder were at increased risk for 4-week mortality, with the following hazard ratios in multivariable Cox regression analysis: 2.02 (95% CI 0.93–4.40), 2.22 (95% CI 0.97–5.08), and 1.97 (95% CI 0.89–4.39).

Supplementary Table 3

Comparison of the 12 Most Frequently Reported Symptoms and Related Signs During the Course of COVID-19 Between Recovered and Deceased Residents

Symptoms and Signs∗	Recovered (Fully or Partially)	Deceased	P Value†
	(n = 56)	(n = 32)
Clinical criteria suspected COVID-19 case (WHO)	52 (92.9)	32 (100)	.12
Respiratory symptoms and signs
Cough	47 (83.9)	21 (65.6)	.05
Shortness of breath‡	29 (51.8)	23 (71.9)	.07
Highest respiratory rate (/min), median (IQR)	28 (24–31)	33 (24–46)	.04
Lowest oxygen saturation on room air (%), median (IQR)	89 (85–93)	85 (80–89)	.01
Supplemental oxygen therapy	20 (69.0)	17 (73.9)	.70
Lowest oxygen saturation on supplemental oxygen (%), median (IQR)	91 (88–93)	85 (79–88)	.02
Rhinorrhea/nasal congestion	21 (37.5)	12 (37.5)	1.00
Sore throat	13 (23.2)	5 (15.6)	.40
General symptoms and signs
Malaise/fatigue	43 (76.8)	30 (93.8)	.04
Fever‡	39 (69.6)	28 (87.5)	.06
Highest body temperature (°C), median (IQR)	38.5 (38.2–38.8)	38.9 (38.4–39.6)	<.01
Loss of appetite/decreased intake	32 (57.1)	32 (100.0)	<.01
Decreased mobility	24 (42.9)	30 (93.8)	<.01
Gastrointestinal symptoms
Diarrhea	26 (46.4)	15 (46.9)	.97
Nausea/vomiting	16 (28.6)	16 (50.0)	.04
Neurological/behavioral symptoms
Change in behavior	12 (21.4)	10 (31.3)	.31
Confusion/delirium	8 (14.3)	10 (31.3)	.06

Variables are presented as frequencies, n (%), unless indicated otherwise.

χ2-test for categorical variables and Mann-Whitney U test for continuous variables.

Related variables are presented for the group of individuals with the sign/symptom concerned, that is, shortness of breath or fever.

Conclusions

Our findings underline the importance of awareness of the broad spectrum of signs/symptoms to identify nursing home residents with COVID-19. A large proportion of these residents (28.4%) did not develop fever with ≥1 respiratory symptoms, thereby not fulfilling the initial WHO case definition. The observed spectrum of signs/symptoms includes atypical symptoms and geriatric syndromes (eg, gastrointestinal symptoms, confusion/delirium, behavioral change, and decreased mobility). Therefore, the current, extended WHO case definition covers our cases better (95.5%). Because we actively searched for this broad range of signs/symptoms (at presentation as well as over the disease course), our results complement studies on symptomatology in nursing home residents.6, 7, 8, 9

The observed mortality rate (36.4%) and the increased 4-week mortality risk in male residents and residents with dementia are confirmed by recently published studies.7, 8, 9 These findings may serve as point of departure for future studies on prognostic factors.