Impact of 1918 influenza pandemic on mental asylums in India: An analysis of asylum reports.

Introduction: Not much is known about impact of 1918 Influenza on patients confined inside mental asylums of that period. Aim and
Methods: The study aimed to assess impact of Influenza pandemic on mentally ill patients in asylums, based on data extracted from provincial asylum reports (1915-1923) from British India.
Results: The asylum population in most provinces was impacted both in terms of direct mortality (1918) as well as all-cause morbidity and mortality (1918; 1918-1920). Agra and Oudh lost 9.5% of average daily asylum strength directly due to influenza in 1918. All-cause mortality (1918) was highest for mental asylum/s in provinces of Agra and Oudh (16.83%), Punjab (14.83%), and Bombay (14.4%). Isolated outbreaks continued till 1923. Another peculiar finding was markedly reduced sickness and death rates in 1921-1923 compared to 1915-1917, across asylums of all provinces.
Conclusion: Findings point to a significant impact of Influenza pandemic in mental asylums across several provinces, and provide historical insights with implications for the current pandemic. Copyright:

INTRODUCTION

In the historical context, the 1918 influenza pandemic is one of the most devastating pandemics of recent times. India had the largest number of deaths in any country (10–20 million) as well as the highest percentage of excess deaths (4.39%) in the world.[12] The impact of 1918 Influenza pandemic on general population has been described in several papers,[123] but not much is known about its impact on patients in mental asylums of that period.

We conducted a PubMed search using a combination of keywords (influenza OR flu OR pandemic OR viral disease OR Spanish flu OR Bombay fever) AND (mental asylums OR lunatic asylums OR mental hospital OR psychiatric hospital). Surprisingly, it did not yield any published paper or report. Therefore, the study was planned to find out the impact of 1918 Influenza pandemic on mental asylums in India, based on the official sources of information.

METHODS

Search database

The “Medical History of British India” under “India papers” collection at the National Library of Scotland consists of official publications related to disease, public health, and medical research[4] which have been digitized with support from Wellcome trust. The Mental Health collection consists of 45 volumes dating from 1867 to 1948, with reports on lunatic asylums and mental hospitals in British India.

Search strategy and data extraction

The provincial asylum reports were submitted by Inspector General of Civil Hospitals/Surgeon General to the Secretary to Government in a prestructured format. The full text of annual and triennial reports for mental asylums in provinces of British India (1915–1923) was reviewed independently by both authors. Morbidity and mortality statistics were extracted for 1918–1920, in addition to the triennium before (1915–1917) and after (1921–1923), to compare trends.

Supplemental search

In addition to the aforementioned sources, we explored:

National archives of India webportal (Abhilekh Patal) for digitized records, which yielded influenza reports in the general population only

PubMed search/Google scholar search and cross-references of key Influenza papers yielded no relevant results for the impact of influenza on asylums.

RESULTS

Results are presented for nine provinces in British India with a total of 23 mental asylums (1918–1923). Results are provided for direct as well as all-cause morbidity and mortality, as below:

Influenza morbidity and deaths directly attributed to influenza in mental asylums

In 1918, there were 1089 recorded influenza cases in asylums across six provinces for which such data was available. Additional two province reports points to a significant impact of influenza in their asylums but provided only mortality data without morbidity figures.

In 1918, there were at least 247 directly attributed to influenza (DDI) in mental asylums across eight provinces. In terms of proportion of DDI relative to average asylum strength (for the year 1918), Agra and Oudh were affected the most (9.5%), followed by Burma (7.6%), Bombay (4.5%), and Punjab (3.5%). The majority of influenza cases or DDI have been recorded over a span of few weeks to few months during a pandemic.

The year 1919 recorded 205 cases and 27 DDI across asylums in various province reports. Few asylums (e.g., Agra and Oudh, Central province, Madras, Burma) appear to have escaped without any recorded influenza case or DDI. In subsequent years, few isolated outbreaks took place from time to time in one or other asylum, for example, 112 cases in Bombay (1921), 162 cases in Bengal (1921) and 104 cases in Assam (1922). Consequently, there were 19 DDI in the year 1920, 15 DDI in the year 1921, and 2 DDI in the year 1922, as per accessible records.

The province-wise details for asylum morbidity and DDI are provided below:

Punjab province (Lahore asylum): 135 cases (*31 DDI) in 1918; 16 (*3 DDI) in 1919; 64 (*16 DDI) in 1920; cases not specified (*9 DDI) in 1921

United provinces of Agra and Oudh province (Bareilly, Agra, and Benaras asylums): 461 cases (*83 DDI) in 1918

Central provinces and Berar (Nagpur asylum): Data for 1918 and 1920 is not available (NA); no influenza cases or DDI reported in rest of years

Bombay province (Naupada-Thana, Ratnagiri, Central Mental Hospital Yeravda, Harwar, Ahmedabad and Hyderabad asylums): Asylums ‘severely affected’ with unspecified cases (*63 DDI) in 1918; Cases unspecified (*6 DDI) in 1919; Data NA in 1920, 112 (*6 DDI) in 1921

Madras province (Madras, Vizagapatam, and Calicut): Influenza affected asylum, with unspecified cases (1918), no cases or DDI (1919–1923)

Bengal province (Bhawanipur, Behrampur, Dacca and Mental Ward of Albert Victor Leper Asylum at Gobra): 91 cases (*2 DDI) in 1918; 150 (*16 DDI) in 1919; 68 (*3 DDI) in 1920; 162 (*0) in 1921, 62 (*0) in 1922, 33 (*0) in 1923

Bihar and Orissa (Patna and Ranchi asylums): 107 cases (*6 DDI) in 1918; cases unspecified (*2 DDI in 1919; *3 DDI in 1920; *1 DDI in 1922)

Assam province (Tezpur asylum): 97 cases in 1918, 39 cases in 1919, 3 in 1921, 104 cases (*1 DDI) in 1922, and 2 cases in 1923

Burma province (Rangoon and Minbu asylums): 198 (*62 DDI) in 1918, 96 cases (*1 DDI) in 1920.

All-cause sickness and all-cause mortality in mental asylums

Table 1 shows mental asylum data pertaining to all-cause sickness (i.e., daily average sick) and all-cause mortality (i.e., all-cause deaths and % all-cause deaths to daily average strength of an asylum) across all provinces.

Table 1

All-cause sickness and all-cause deaths in mental asylums of nine British India provinces (1915–1923)

	1915	1916	1917	Triennium average 1915-1917	1918	1919	1920	Triennium average 1918-1920	1921	1922	1923	Triennium average 1921-1923
Punjab province (Lahore asylum)
Daily average strength	763.50	831.27	837.03	810.60	876.44	874.99	862.82	871.42	864.58	867.17	899.53	877.09
All-cause deaths	73	89	103	90	130	92	109	127	97	58	92	82.33
Percentage all-cause deaths to daily average strength	9.56	10.71	12.31	10.86	14.83	10.51	12.33	12.56	11.22	6.69	10.23	9.38
Daily average sick (all-cause)	68.49	60.08	66.75	65.11	82.74	90.58	64.79	79.37	36.13	40.61	51.83	42.86
United Provinces of Agra and Oudh province (Bareilly, Benaras and Agra asylums)
Daily average strength	1369.37	1419.47	1487.71	1425.52	1532.30	1404.85	1383.88	1440.34	1364.27	1309.82	1301.25	1325.11
All-cause deaths	140	92	138	123.33	258	123	153	178	191	114	114	139.67
Percentage all-cause deaths to daily average strength	10.22	6.48	9.27	8.99	16.83	8.81	11.65	12.43	14.00	8.70	8.76	10.49
Daily average sick (all-cause)	199.59	175.91	165.10	180.20	208.77	247.57	260.61	238.98	230.07	216.94	214.77	220.59
Central provinces and Berar (Nagpur asylum)
Daily average strength	373.97	364.12	367.17	368.42	376.79	388.85	398.72	388.12	393.30	349.20	342.31	361.60
All-cause deaths	46	40	37	41	32	53	33	39.33	43	20	24	29
Percentage all-cause deaths to daily average strength	12.30	10.99	10.08	11.12	8.50	13.62	8.28	10.13	10.93	5.73	7.01	7.89
Daily average sick (all-cause)	15.32	18.49	8.46	14.09	11.62	34.20	NA	22.91	18.20	22.03	19.83	20.93
Bombay province (Naupada-Thana, Ratnagiri, Yeravda, Dharwar, Ahmedabad and Hyderabad asylums)
Daily average strength	1194.0	1294.6	1335.2	1274.60	1405.4	1438.7	1430.9	1425	1436.2	1443.0	1481.9	1453.70
All-cause deaths	85	106	135	108.67	203	216	176	198.33	157	117	98	124
Percentage all-cause deaths to daily average strength	7.1	8.2	10.1	8.47	14.4	15.0	12.29	13.90	10.9	8.2	6.6	8.57
Daily average sick (all-cause)	39.5	35.3	45.5	40.1	50.7	45.2	52.4	49.43	61.4	58.0	56.9	58.77
Madras province (Madras, Vizagapatam and Calicut)
Daily average strength	799.47	832.68	851.89	828.01	876.88	906.03	894.79	892.57	917.29	967.46	959.07	947.94
All-cause deaths	88	85	77	83.33	119	127	121	122.33	82	82	89	84.33
Percentage all-cause deaths to daily average strength	10.38	10.21	9.04	9.88	13.57	14.02	13.52	13.70	8.94	8.48	9.27	8.90
Daily average sick (all-cause)	103.15	79.49	82.36	88.33	99.31	81.05	70.73	83.70	104.30	125.27	125.47	118.35
Bengal province (Bhawanipur, Dacca, Behrampur and Mental Ward of Albert Victor Leper Asylum at Gobra.)
Daily average strength	1091.40	1076.94	1074.36	1080.90	1,078.97	996.74	966.66	1014.12	957.08	954.55	950.44	954.02
All-cause deaths	85	120	85	96.67	104	100	102	102	66	68	81	71.67
Percentage all-cause deaths to daily average strength	7.79	11.14	7.91	8.95	9.63	10.03	10.55	10.07	6.89	6.60	8.52	7.34
Daily average sick (all-cause)	84.38	90.31	89.09	87.93	94.30	107.78	116.29	106.12	130.33	106.31	116.97	117.87
Bihar and Orissa (Patna asylum) (Ranchi 1918–1920 in brackets*)
Daily average strength	315.18	310.03	299.74	308.32	301.24 (+432.74)	268.81 (+437.97)	252.35 (+416.08)	274.13	268.48	268.56	273.67	270.24
All-cause deaths	41	35	35	37	35 (+46)*	46 (+62)*	24 (+36)*	35	28	21	20	23
Percentage all-cause deaths to daily average strength	13.00	11.28	11.68	11.99	11.62 (10.62)	17.11 (14.15)	7.33 (8.65)	12.02	7.31	7.45	10.43	8.40
Daily average sick (all-cause)	22.61	23.06	27.25	24.31	34.45 (+38.59)	32.67 (+38.72)	44.42 (+52.26)	37.18	57.50	44.18	39.56	47.08
Assam province (Tezpur asylum)
Daily average strength	324.94	360.14	341.44	342.17	376.71	402.65	398.58	392.65	426.62	443.82	444.37	438.27
All-cause deaths	29	52	19	33.33	22	21	14	19	24	27	30	27
Percentage all-cause deaths to daily average strength	8.92	14.44	5.56	9.64	5.81	5.21	3.51	4.84	5.63	6.08	6.75	6.15
Daily average sick (all-cause)	60.46	85.53	78.40	74.80	89.36	95.32	55.07	79.92	55.04	40.09	32.92	42.68
Burma province (Rangoon and Minbu asylums)
Daily average strength	623.10	678.53	752.37	684.67	816.28	828.12	855.90	833.43	859.46	904.98	952.37	905.60
All-cause deaths	53	44	51	49.33	64	80	58	67.33	67	66	80	71
Percentage all-cause deaths to daily average strength	8.51	6.48	6.78	7.26	7.84	9.66	6.77	8.09	7.74	7.29	9.74	8.26
Daily average sick (all-cause)	184.78	287.06	170.67	214.17	174.23	159.07	157.32	163.54	152.33	148.64	122.63	141.20

*Ranchi asylum came into existence in the year 1918 and its data is shown in brackets for 1918–1920 and unavailable for 1920–1923; No other missing data

DISCUSSION

The paper discusses the trends of sickness and mortality among asylum inmates during early 20th century influenza pandemic across various provinces in Colonial India.

First major finding is that Influenza was directly responsible for death of 9.5% of average asylum strength in Agra and Oudh, 7.5% in Burma, 4.5% in Bombay, and 3.5% in Punjab. The officially estimated mortality rates in general population across British India were 2.2% for Agra and Oudh, 0.6% for Burma, 4.5% for Bombay, and 4.2% for Punjab province (up to November 30, 1918).[5] It appears that the mortality rate in confined asylum population was relatively higher than population rates. Infection is likely to spread faster in confined spaces and can be more fatal in mentally ill patients in view of preexisting vulnerabilities such as higher medical morbidities, malnutrition, and other risk factors.[678] Certain provinces such as Assam and Madras asylums had relatively less influenza deaths, which is broadly consistent with low mortality of 1918 influenza pandemic in eastern and southern parts of India, compared to Northern and Western parts of India.[5]

A second major finding from this study is a clear jump in all-cause death rates and all-cause sickness in year 1918 and triennium 1918–1920 compared to preceding triennium (1915–1917) in most provinces. Apart from influenza, there was no other documented epidemic which could explain this observation across geographically far-placed provinces. Influenza pandemic is likely to be a major contributor as also indicated in Agra and Oudh report (1918–1920) “Influenza epidemic accounted for 54 deaths and of the remaining deaths, many were due to the debility ensuing among those who recovered from influenza, rendering them a ready prey to other diseases.”[9] A general lowering of health of asylum patients as a sequel of the very severe influenza epidemic is a common mention in many reports. There are obvious limitations in sole reliance on directly attributable influenza morbidity and mortality. The causative agent for influenza was not isolated till a decade later and atypical presentations of influenza may have been missed.[10] Even in COVID-19 pandemic, estimates suggest that official death figures under-represent the actual mortality.[11] Consequently, all-cause mortality and morbidity data can provide additional useful insights.

Third, another striking observation from the study is that the all-cause sickness and all-cause death rates (as percentage of daily average strength) were discernably lower in the triennium after pandemic (1921–1923) compared to the preinfluenza triennium (1915–1917). This phenomenon was observed consistently across most of the provincial asylum/s. A reduction in all-cause sickness/death rates (1921–1923) could be due to selective and premature deaths in those with respiratory diseases, immunosuppression, old age, or other vulnerabilities, leading to decline in mortality figures in ensuing years. Such phenomenon has been referred to as “harvesting effect.”[1213] We attempted to see the corresponding figures in Census of India for those decades[141516] to see if it is a part of a widespread trend. The Census for decade 1901–1911 had annual death rate (per thousand persons) of 42.6, which jumped to 48.6 during 1911–1921, largely due to influenza pandemic, followed by a significant decline by 25% in 1921–1931 (36 per thousand persons).[141516]

The study has certain limitations. We focused on data which was accessible through web portal and some data was missing as stated in results. Data are presented for province as a whole, which does not cover inter-asylum and intra-asylum variances. For example, 1918 Influenza was virulent at the Hyderabad, Dhárwár and Ratnágiri Asylums but not in other asylums of Bombay.[17] In Punjab, male wing was selectively affected in 1918–1919 and female wing in 1920 outbreak.[18] Finally, there may be potential, unknown confounders to influence data across trienniums.

Findings have some implications for mental health facilities in the current pandemic. The vulnerable populations, including patients with serious mental illness, are impacted at disproportionately higher rates. Mental health facilities need to foresee challenges and re-organize their mental health services for a rigorous infection control and prevention.

To conclude, this paper points to a significant impact of 1918 Influenza pandemic on mental asylums across several provinces. These historical insights emphasize on the need for proactive measures for hospitalized, mentally-ill population amidst current pandemic.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.